Datasets:

ChicagoHAI

/

CaseSumm

like 10

Follow

ChicagoHAI 19

571.US.320

Title 21 U. S. C. §853(e)(1) empowers courts to enter pre-trial restraining orders to “preserve the availability of [forfeitable] property” while criminal proceedings are pending. Such pre-trial asset restraints are constitutionally permissible whenever probable cause exists to think that a defendant has committed an offense permitting forfeiture and that the assets in dispute are traceable or otherwise sufficiently related to the crime charged. United States v. Monsanto, 491 U.S. 600. After a grand jury indicted petitioners, Kerri and Brian Kaley, for reselling stolen medical devices and laundering the proceeds, the Government obtained a §853(e)(1) restraining order against their assets. The Kaleys moved to vacate the order, intending to use a portion of the disputed assets for their legal fees. The District Court allowed them to challenge the assets’ traceability to the offenses in question but not the facts supporting the underlying indictment. The Eleventh Circuit affirmed. Held: When challenging the legality of a §853(e)(1) pre-trial asset seizure, a criminal defendant who has been indicted is not constitutionally entitled to contest a grand jury’s determination of probable cause to believe the defendant committed the crimes charged. Pp. 5–21. (a) In Monsanto, this Court held that the Government may seize assets before trial that a defendant intends to use to pay an attorney, so long as probable cause exists “to believe that the property will ultimately be proved forfeitable.” 491 U. S., at 615. The question whether indicted defendants like the Kaleys are constitutionally entitled to a judicial re-determination of the grand jury’s probable cause conclusion in a hearing to lift an asset restraint has a ready answer in the fundamental and historic commitment of the criminal justice system to entrust probable cause findings to a grand jury. A probable cause finding sufficient to initiate a prosecution for a serious crime is “conclusive[e],” Gerstein v. Pugh, 420 U.S. 103, 117, n. 19, and, as a general matter, “a challenge to the reliability or competence of the evidence” supporting that finding “will not be heard,” United States v. Williams, 504 U.S. 36, 54. A grand jury’s probable cause finding may, on its own, effect a pre-trial restraint on a person’s liberty. Gerstein, 420 U. S., at 117, n. 19. The same result follows when it works to restrain a defendant’s property. The Kaleys’ alternative rule would have strange and destructive consequences. Allowing a judge to decide anew what the grand jury has already determined could result in two inconsistent findings governing different aspects of one criminal proceeding, with the same judge who found probable cause lacking presiding over a trial premised on its existence. That legal dissonance could not but undermine the criminal justice system’s integrity, especially the grand jury’s constitutional role. Pp. 5–12. (b) The balancing test of Mathews v. Eldridge, 424 U. S. 319—which requires a court to weigh (1) the burdens that a requested procedure would impose on the government against (2) the private interest at stake, as viewed alongside (3) “the risk of an erroneous deprivation” of that interest without the procedure and “the probable value, if any, of [the] additional . . . procedural safeguar[d],” id., at 335—if applicable here, tips against the Kaleys. Because the Government’s interest in freezing potentially forfeitable assets without an adversarial hearing about the probable cause underlying criminal charges and the Kaleys’ interest in retaining counsel of their own choosing are both substantial, the test’s third prong is critical. It boils down to the “probable value, if any,” of a judicial hearing in uncovering mistaken grand jury probable cause findings. But when the legal standard is merely probable cause and the grand jury has already made that finding, a full-dress hearing will provide little benefit. See Florida v. Harris, 568 U. S. ___, ___. A finding of probable cause to think that a person committed a crime “can be [made] reliably without an adversary hearing,” Gerstein, 420 U. S., at 120, and the value of requiring additional “formalities and safeguards” would “[i]n most cases . . . be too slight,” id., at 121–122. The experience of several Circuits corroborates this view. Neither the Kaleys nor their amici point to a single case in two decades where courts, holding hearings of the kind they seek, have found the absence of probable cause to believe that an indicted defendant committed the crime charged. Pp. 12–20. 677 F.3d 1316, affirmed and remanded. Kagan, J., delivered the opinion of the Court, in which Scalia, Kennedy, Thomas, Ginsburg, and Alito, JJ., joined. Roberts, C. J., filed a dissenting opinion, in which Breyer and Sotomayor, JJ., joined.

A federal statute, 21 U. S. C. §853(e), authorizes a court to freeze an indicted defendant’s assets prior to trial if they would be subject to forfeiture upon conviction. In United States v. Monsanto, 491 U. S. 600, 615 (1989) , we approved the constitutionality of such an order so long as it is “based on a finding of probable cause to believe that the property will ultimately be proved forfeitable.” And we held that standard to apply even when a defendant seeks to use the disputed property to pay for a lawyer. In this case, two indicted defendants wishing to hire an attorney challenged a pre-trial restraint on their property. The trial court convened a hearing to consider the seizure’s legality under Monsanto. The question presented is whether criminal defendants are constitutionally entitled at such a hearing to contest a grand jury’s prior determination of probable cause to believe they committed the crimes charged. We hold that they have no right to relitigate that finding. I A Criminal forfeitures are imposed upon conviction to confiscate assets used in or gained from certain serious crimes. See 21 U. S. C. §853(a). Forfeitures help to ensure that crime does not pay: They at once punish wrongdoing, deter future illegality, and “lessen the economic power” of criminal enterprises. Caplin & Drysdale, Chartered v. United States, 491 U. S. 617, 630 (1989) ; see id., at 634 (“Forfeiture provisions are powerful weapons in the war on crime”). The Government also uses forfeited property to recompense victims of crime, improve conditions in crime-damaged communities, and support law enforcement activities like police training. See id., at 629–630.[1] Accordingly, “there is a strong governmental interest in obtaining full recovery of all forfeitable assets.” Id., at 631. In line with that interest, §853(e)(1) empowers courts to enter pre-trial restraining orders or injunctions to “preserve the availability of [forfeitable] property” while criminal proceedings are pending. Such an order, issued “[u]pon application of the United States,” prevents a defendant from spending or transferring specified property, including to pay an attorney for legal services. Ibid. In Monsanto, our principal case involving this procedure, we held a pre-trial asset restraint constitutionally permissible whenever there is probable cause to believe that the property is forfeitable. See 491 U. S., at 615. That determinationhas two parts, reflecting the requirements for forfeit-ure under federal law: There must be probable cause to think (1) that the defendant has committed an offense permitting forfeiture, and (2) that the property at issue has the requisite connection to that crime. See §853(a). The Monsanto Court, however, declined to consider “whether the Due Process Clause requires a hearing” to establish either or both of those aspects of forfeitability. Id., at 615, n. 10.[2] Since Monsanto, the lower courts have generally pro-vided a hearing to any indicted defendant seeking to lift an asset restraint to pay for a lawyer. In that hearing, they have uniformly allowed the defendant to litigate the second issue stated above: whether probable cause exists to believe that the assets in dispute are traceable or otherwise sufficiently related to the crime charged in the indictment.[3] But the courts have divided over extending the hearing to the first issue. Some have considered, while others have barred, a defendant’s attempt to challenge the probable cause underlying a criminal charge.[4] This case raises the question whether an indicted defendant has a constitutional right to contest the grand jury’s prior determination of that matter. B The grand jury’s indictment in this case charges a scheme to steal prescription medical devices and resell them for profit. The indictment accused petitioner Kerri Kaley, a sales representative for a subsidiary of Johnson & Johnson, and petitioner Brian Kaley, her husband, with transporting stolen medical devices across state lines and laundering the proceeds of that activity.[5] The Kaleys have contested those allegations throughout this litigation, arguing that the medical devices at issue were unwanted, excess hospital inventory, which they could lawfully take and market to others. Immediately after obtaining the indictment, the Government sought a restraining order under §853(e)(1) to prevent the Kaleys from transferring any assets traceable to or involved in the alleged offenses. Included among those assets is a $500,000 certificate of deposit that the Kaleys intended to use for legal fees. The District Court entered the requested order. Later, in response to the Kaleys’ motion to vacate the asset restraint, the court denied a request for an evidentiary hearing and confirmed the order, except as to $63,000 that it found (based on the parties’ written submissions) was not connected to the alleged offenses. On interlocutory appeal, the Eleventh Circuit reversed and remanded for further consideration of whether some kind of evidentiary hearing was warranted. See 579 F. 3d 1246 (2009). The District Court then concluded that it should hold a hearing, but only as to “whether the restrained assets are traceable to or involved in the alleged criminal conduct.” App. to Pet. for Cert. 43, n. 5. The Kaleys informed the court that they no longer disputed that issue; they wished to show only that the “case against them is ‘baseless.’ ” Id., at 39; see App. 107 (“We are not contesting that the assets restrained were . . . traceable to the conduct. Our quarrel is whether that conduct constitutes a crime”). Accordingly, the District Court affirmed the restraining order, and the Kaleys took another appeal. The Eleventh Circuit this time affirmed, holding that the Kaleys were not entitled at a hearing on the asset freeze “to challenge the factual foundation supporting the grand jury’s probable cause determination[ ]”—that is, “the very validity of the underlying indictment.” 677 F. 3d 1316, 1317 (2012). We granted certiorari in light of the Circuit split on the question presented, 568 U. S. ___ (2013), and we now affirm the Eleventh Circuit. II This Court has twice considered claims, similar to the Kaleys’, that the Fifth Amendment’s right to due process and the Sixth Amendment’s right to counsel constrain the way the federal forfeiture statute applies to assets needed to retain an attorney. See Caplin & Drysdale, 491 U. S. 617 ; Monsanto, 491 U. S. 600 . We begin with those rulings not as mere background, but as something much more. On the single day the Court decided both those cases, it cast the die on this one too. In Caplin & Drysdale, we considered whether the Fifth and Sixth Amendments exempt from forfeiture money that a convicted defendant has agreed to pay his attorney. See 491 U. S., at 623–635. We conceded a factual premise of the constitutional claim made in the case: Sometimes “a defendant will be unable to retain the attorney of his choice,” if he cannot use forfeitable assets. Id., at 625. Still, we held, the defendant’s claim was “untenable.” Id., at 626. “A defendant has no Sixth Amendment right to spend another person’s money” for legal fees—even if that is the only way to hire a preferred lawyer. Ibid. Consider, we submitted, the example of a “robbery suspect” who wishes to “use funds he has stolen from a bank to retain an attorney to defend him if he is apprehended.” Ibid. That money is “not rightfully his.” Ibid. Accordingly, we concluded, the Government does not violate the Constitution if, pursuant to the forfeiture statute, “it seizes the robbery proceeds and refuses to permit the defendant to use them” to pay for his lawyer. Ibid. And then, we confirmed in Monsanto what our “robbery suspect” hypothetical indicated: Even prior to conviction (or trial)—when the presumption of innocence still applies—the Government could constitutionally use §853(e) to freeze assets of an indicted defendant “based on a find-ing of probable cause to believe that the property will ultimately be proved forfeitable.” 491 U. S., at 615. In Monsanto, too, the defendant wanted to use the property at issue to pay a lawyer, and maintained that the Fifth and Sixth Amendments entitled him to do so. We dis-agreed. We first noted that the Government may sometimes “restrain persons where there is a finding of probable cause to believe that the accused has committed a serious offense.” Id., at 615–616. Given that power, we could find “no constitutional infirmity in §853(e)’s authorization of a similar restraint on [the defendant’s] property” in order to protect “the community’s interest” in recovering “ill-gotten gains.” Id., at 616. Nor did the defendant’s interest in retaining a lawyer with the disputed assets change the equation. Relying on Caplin & Drysdale, we reasoned: “[I]f the Government may, post-trial, forbid the use of forfeited assets to pay an attorney, then surely no constitutional violation occurs when, after probable cause is adequately established, the Government obtains an order barring a defendant from frustrating that end by dissipating his assets prior to trial.” Ibid. So again: With probable cause, a freeze is valid. The Kaleys little dispute that proposition; their argument is instead about who should have the last word as to probable cause. A grand jury has already found probable cause to think that the Kaleys committed the offenses charged; that is why an indictment issued. No one doubts that those crimes are serious enough to trigger forfeiture. Similarly, no one contests that the assets in question derive from, or were used in committing, the offenses. See supra, at 5. The only question is whether the Kaleys are constitutionally entitled to a judicial re-determination of the conclusion the grand jury already reached: that probable cause supports this criminal prosecution (or alternatively put, that the prosecution is not “baseless,” as the Kaleys believe, supra, at 5). And that question, we think, has a ready answer, because a fundamental and historic commitment of our criminal justice system is to entrust those probable cause findings to grand juries. This Court has often recognized the grand jury’s singular role in finding the probable cause necessary to initiate a prosecution for a serious crime. See, e.g., Costello v. United States, 350 U. S. 359, 362 (1956) . “[A]n indictment ‘fair upon its face,’ and returned by a ‘properly constituted grand jury,’ ” we have explained, “conclusively determines the existence of probable cause” to believe the defendant perpetrated the offense alleged. Gerstein v. Pugh, 420 U. S. 103 , n. 19 (1975) (quoting Ex parte United States, 287 U. S. 241, 250 (1932) ). And “conclusively” has meant, case in and case out, just that. We have found no “authority for looking into and revising the judgment of the grand jury upon the evidence, for the purpose of determining whether or not the finding was founded upon sufficient proof.” Costello, 350 U. S., at 362–363 (quoting United States v. Reed, 27 F. Cas. 727, 738 (No. 16,134) (CC NDNY 1852) (Nelson, J.)). To the contrary, “the whole history of the grand jury institution” demonstrates that “a challenge to the reliability or competence of the evidence” supporting a grand jury’s finding of probable cause “will not be heard.” United States v. Williams, 504 U. S. 36, 54 (1992) (quoting Costello, 350 U. S., at 364, and Bank of Nova Scotia v. United States, 487 U. S. 250, 261 (1988) ). The grand jury gets to say—without any review, oversight, or second-guessing—whether probable cause exists to think that a person committed a crime. And that inviolable grand jury finding, we have decided, may do more than commence a criminal proceeding (with all the economic, reputational, and personal harm that entails); the determination may also serve the purpose of immediately depriving the accused of her freedom. If the person charged is not yet in custody, an indictment triggers “issuance of an arrest warrant without further inquiry” into the case’s strength. Gerstein, 420 U. S., at 117, n. 19; see Kalina v. Fletcher, 522 U. S. 118, 129 (1997) . Alternatively, if the person was arrested without a warrant, an indictment eliminates her Fourth Amendment right to a prompt judicial assessment of probable cause to support any detention. See Gerstein, 420 U. S., at 114, 117, n. 19. In either situation, this Court—relying on the grand jury’s “historical role of protecting individuals from unjust persecution”—has “let [that body’s] judgment substitute for that of a neutral and detached magistrate.” Ibid. The grand jury, all on its own, may effect a pre-trial restraint on a person’s liberty by finding probable cause to support a criminal charge.[6] The same result follows when, as here, an infringement on the defendant’s property depends on a showing of probable cause that she committed a crime. If judicial review of the grand jury’s probable cause determination is not warranted (as we have so often held) to put a defendant on trial or place her in custody, then neither is it needed to freeze her property. The grand jury that is good enough—reliable enough, protective enough—to inflict those other grave consequences through its probable cause findings must needs be adequate to impose this one too. Indeed, Monsanto already noted the absence of any reason to hold property seizures to different rules: As described earlier, the Court partly based its adoption of the probable cause standard on the incongruity of subjecting an asset freeze to any stricter requirements than apply to an arrest or ensuing detention. See supra, at 6; 491 U. S., at 615 (“[I]t would be odd to conclude that the Government may not restrain property” on the showing often sufficient to “restrain persons”). By similar token, the probable cause standard, once selected, should work no differently for the single purpose of freezing assets than for all others.[7] So the longstanding, unvarying rule of criminal procedure we have just described applies here as well: The grand jury’s determination is conclusive. And indeed, the alternative rule the Kaleys seek would have strange and destructive consequences. The Kaleys here demand a do-over, except with a different referee. They wish a judge to decide anew the exact question the grand jury has already answered—whether there is probable cause to think the Kaleys committed the crimes charged. But suppose the judge performed that task and came to the opposite conclusion. Two inconsistent findings would then govern different aspects of one criminal proceeding: Probable cause would exist to bring the Kaleys to trial (and, if otherwise appropriate, hold them in prison), but not to restrain their property. And assuming the prosecutor continued to press the charges,[8] the same judge who found probable cause lacking would preside over a trial premised on its presence. That legal dissonance, if sustainable at all, could not but undermine the criminal justice system’s integrity—and especially the grand jury’s integral, constitutionally prescribed role. For in this new world, every prosecution involving a pre-trial asset freeze would potentially pit the judge against the grand jury as to the case’s foundational issue.[9] The Kaleys counter (as does the dissent, post, at 7) that apparently inconsistent findings are not really so, because the prosecutor could have presented scantier evidence to the judge than he previously offered the grand jury. Suppose, for example, that at the judicial hearing the prosecutor put on only “one witness instead of all five”; then, the Kaleys maintain, the judge’s decision of no probable cause would mean only that “the Government did not satisfy its burden[ ] on that one day in time.” Tr. of Oral Arg. 12, 18; see Reply Brief 11–12. But we do not think that hypothetical solves the problem. As an initial matter, it does not foreclose a different fact pattern: A judge could hear the exact same evidence as the grand jury, yet respond to it differently, thus rendering what even the Kaleys must concede is a contradictory finding. And when the Kaleys’ hypothetical is true, just what does it show? Consider that the prosecutor in their example has left home some of the witnesses he took to the grand jury—presumably because, as we later discuss, he does not yet wish to reveal their identities or likely testimony. See infra, at 14–15. The judge’s ruling of no probable cause therefore would not mean that the grand jury was wrong: As the Kaleys concede, the grand jury could have heard more than enough evidence to find probable cause that they committed the crimes charged. The Kaleys would win at the later hearing despite, not because of, the case’s true merits. And we would then see still less reason for a judge to topple the grand jury’s (better supported) finding of probable cause.[10] Our reasoning so far is straightforward. We held in Monsanto that the probable cause standard governs the pre-trial seizure of forfeitable assets, even when they are needed to hire a lawyer. And we have repeatedly affirmed a corollary of that standard: A defendant has no right to judicial review of a grand jury's determination of probable cause to think a defendant committed a crime. In combination, those settled propositions signal defeat for the Kaleys because, in contesting the seizure of their property, they seek only to relitigate such a grand jury finding. III The Kaleys would have us undertake a different analysis, which they contend would lead to a different conclusion. They urge us to apply the balancing test of Mathews v. Eldridge, 424 U. S. 319 (1976) , to assess whether they have received a constitutionally sufficient opportunity to challenge the seizure of their assets. See Brief for Petitioners 32–64. Under that three-pronged test (reordered here for expositional purposes), a court must weigh (1) the burdens that a requested procedure would impose on the Government against (2) the private interest at stake, as viewed alongside (3) “the risk of an erroneous deprivation” of that interest without the procedure and “the probable value, if any, of [the] additional . . . procedural safeguard[ ].” Mathews, 424 U. S., at 335. Stressing the importance of their interest in retaining chosen counsel, the Kaleys argue that the Mathews balance tilts hardin their favor. It thus overrides—or so the Kaleys claim—all we have previously held about the finality of grand jury findings, entitling them to an evidentiary hearing be-fore a judge to contest the probable cause underlying the indictment. The Government battles with the Kaleys over whether Mathews has any application to this case. This Court devised the test, the Government notes, in an administrative setting—to decide whether a Social Security recipient was entitled to a hearing before her benefits were terminated. And although the Court has since employed the approach in other contexts, the Government reads Medina v. California, 505 U. S. 437 (1992) , as foreclosing its use here. In that case, we held that “the Mathews balancing test does not provide the appropriate framework for assessing the validity of state procedural rules which . . . are part of the criminal process,” reasoning that because the “Bill of Rights speaks in explicit terms to many aspects of criminal procedure,” the Due Process Clause “has limited operation” in the field. Id., at 443. That settles that, asserts the Government. See Brief for United States 18. But the Kaleys argue that Medina addressed a State’s procedural rule and relied on federalism principles not implicated here. Further, they claim that Medina concerned a criminal proceeding proper, not a collateral action seizing property. See Reply Brief 1–5. As to that sort of action, the Kaleys contend, Mathews should govern. We decline to address those arguments, or to define the respective reach of Mathews and Medina, because we need not do so. Even if Mathews applied here—even if, that is, its balancing inquiry were capable of trumping this Court’s repeated admonitions that the grand jury’s word is conclusive—the Kaleys still would not be entitled to the hearing they seek. That is because the Mathews test tips against them, and so only reinforces what we have already said. As we will explain, the problem for the Kaleys comes from Mathews’ prescribed inquiry into the requested procedure’s usefulness in correcting erroneous deprivations of their private interest. In light of Monsanto’s holding that a seizure of the Kaleys’ property is erroneous only if unsupported by probable cause, the added procedure demanded here is not sufficiently likely to make any difference. To begin the Mathews analysis, the Government has a substantial interest in freezing potentially forfeitable assets without an evidentiary hearing about the probable cause underlying criminal charges. At the least, such an adversarial proceeding—think of it as a pre-trial mini-trial (or maybe a pre-trial not-so-mini-trial)—could consume significant prosecutorial time and resources. The hearing presumably would rehearse the case’s merits, including the Government’s theory and supporting evidence. And the Government also might have to litigate a range of ancillary questions relating to the conduct of the hearing itself (for example, could the Kaleys subpoena witnesses or exclude certain evidence?). Still more seriously, requiring a proceeding of that kind could undermine the Government’s ability either to obtain a conviction or to preserve forfeitable property. To ensure a favorable result at the hearing, the Government could choose to disclose all its witnesses and other evidence. But that would give the defendant knowledge of the Government’s case and strategy well before the rules of criminal procedure—or principles of due process, see, e.g., Brady v. Maryland, 373 U. S. 83 (1963) —would otherwise require. See Fed. Rules Crim. Proc. 26.2(a), 16(a)(2); Weatherford v. Bursey, 429 U. S. 545 –561 (1977) (“There is no general constitutional right to discovery in a criminal case”). And sometimes (particularly in organized crime and drug trafficking prosecutions, in which forfeit-ure questions often arise), that sneak preview might not just aid the defendant’s preparations but also facilitate witness tampering or jeopardize witness safety. Alternatively, to ensure the success of its prosecution, the Government could hold back some of its evidence at the hearing or give up on the pre-trial seizure entirely. But if the Government took that tack, it would diminish the likelihood of ultimately recovering stolen assets to which the public is entitled.[11] So any defense counsel worth his salt—whatever the merits of his case—would put the prosecutor to a choice: “Protect your forfeiture by providing discovery” or “protect your conviction by surrendering the assets.”[12] It is small wonder that the Government wants to avoid that lose-lose dilemma. For their part, however, defendants like the Kaleys have a vital interest at stake: the constitutional right to retain counsel of their own choosing. See Wheat v. United States, 486 U. S. 153, 159 (1988) (describing the scope of, and various limits on, that right). This Court has recently described that right, separate and apart from the guarantee to effective representation, as “the root meaning” of the Sixth Amendment. United States v. Gonzalez-Lopez, 548 U. S. 140 –148 (2006); cf. Powell v. Alabama, 287 U. S. 45, 53 (1932) (“It is hardly necessary to say that, the right to counsel being conceded, a defendant should be afforded a fair opportunity to secure counsel of his own choice”).[13] Indeed, we have held that the wrongful deprivation of choice of counsel is “structural error,” immune from review for harmlessness, because it “pervades the entire trial.” Gonzalez-Lopez, 548 U. S., at 150. Different lawyers do all kinds of things differently, sometimes “affect[ing] whether and on what terms the defendant . . . plea bargains, or decides instead to go to trial”—and if the latter, possibly affecting whether she gets convicted or what sentence she receives. Ibid. So for defendants like the Kaleys, having the ability to retain the “counsel [they] believe[ ] to be best”—and who might in fact be superior to any existing alternatives—matters profoundly. Id., at 146. And yet Monsanto held, crucially for the last part of our Mathews analysis, that an asset freeze depriving a defend-ant of that interest is erroneous only when unsupportedby a finding of probable cause. Recall that Monsanto considered a case just like this one, where the defendant wanted to use his property to pay his preferred lawyer. He urged the Court to hold that the Government could seize assets needed for that purpose only after conviction. But we instead decided that the Government could act “after probable cause [that the assets are forfeitable] is adequately established.” 491 U. S., at 616. And that means in a case like this one—where the assets’ connection to the allegedly illegal conduct is not in dispute, see supra, at 5—that a pre-trial seizure is wrongful only when there is no probable cause to believe the defendants committed the crimes charged. Or to put the same point differently, such a freeze is erroneous—notwithstanding the weighty burden it imposes on the defendants’ ability to hire a chosen lawyer—only when the grand jury should never have issued the indictment. The Mathews test’s remaining prong—critical when the governmental and private interests both have weight—thus boils down to the “probable value, if any,” of a judicial hearing in uncovering mistaken grand jury findings of probable cause. 424 U. S., at 335. The Kaleys (and the dissent) contend that such proceedings will serve an important remedial function because grand juries hear only a “one-sided presentation[ ]” of evidence. Brief for Petitioners 57; see post, at 16. And that argument rests on a generally sound premise: that the adversarial process leads to better, more accurate decision-making. But in this context—when the legal standard is merely probable cause and the grand jury has already made that finding—both our precedents and other courts’ experience indicate that a full-dress hearing will provide little benefit. This Court has repeatedly declined to require the use of adversarial procedures to make probable cause determinations. Probable cause, we have often told litigants, is not a high bar: It requires only the “kind of ‘fair probability’ on which ‘reasonable and prudent [people,] not legal technicians, act.’ ” Florida v. Harris, 568 U. S. __, __ (2013) (slip op., at 5) (quoting Illinois v. Gates, 462 U. S. 213, 231, 238 (1983) ); see Gerstein, 420 U. S., at 121 (contrasting probable cause to reasonable-doubt and preponderance standards). That is why a grand jury’s finding of probable cause to think that a person committed a crime “can be [made] reliably without an adversary hearing,” id., at 120; it is and “has always been thought sufficient to hear only the prosecutor’s side,” United States v. Williams, 504 U. S. 36, 51 (1992) . So, for example, we have held the “confrontation and cross-examination” of witnesses unnecessary in a grand jury proceeding. Gerstein, 420 U. S., at 121–122. Similarly, we have declined to require the presentation of exculpatory evidence, see Williams, 504 U. S., at 51, and we have allowed the introduction of hearsay alone, see Costello, 350 U. S., at 362–364. On each occasion, we relied on the same reasoning, stemming from our recognition that probable cause served only a gateway function: Given the relatively undemanding “nature of the determination,” the value of requiring any additional “formalities and safeguards” would “[i]n most cases . . . be too slight.” Gerstein, 420 U. S., at 121–122. We can come out no differently here. The probable cause determinations the Kaleys contest are simply those underlying the charges in the indictment. No doubt the Kaleys could seek to poke holes in the evidence the Government offered the grand jury to support those allegations. No doubt, too, the Kaleys could present evidence of their own, which might cast the Government’s in a different light. (Presumably, the Kaleys would try in those two ways to show that they did not steal, but instead lawfully obtained the medical devices they later resold. See supra, at 4.) Our criminal justice system of course relies on such contestation at trial when the question becomes whether a defendant is guilty beyond peradventure. But as we have held before, an adversarial process is far less useful to the threshold finding of probable cause, which determines only whether adequate grounds exist to proceed to trial and reach that question. The probable cause decision, by its nature, is hard to undermine, and still harder to reverse. So the likelihood that a judge holding an evidentiary hearing will repudiate the grand jury’s decision strikes us, once more, as “too slight” to support a constitutional requirement. Gerstein, 420 U. S., at 122. The evidence from other courts corroborates that view, over and over and over again. In the past two decades, the courts in several Circuits have routinely held the kind of hearing the Kaleys seek. See supra, at 3, and n. 4. Yet neither the Kaleys nor their amici (mostly lawyers’ associations) have found a single case in which a judge found an absence of probable cause to believe that an indicted defendant committed the crime charged. One amicus cites 25 reported cases involving pre-trial hearings on asset freezes. See Brief for New York Council of Defense Lawyers 4, n. 2. In 24 of those, the defendant lost outright. The last involved a not-yet-indicted defendant (so no grand jury finding); there, the District Court’s ruling for him was reversed on appeal. See Tr. of Oral Arg. 15, 36. To be sure, a kind of selection bias might affect those statistics: Perhaps a prosecutor with a very weak case would choose to abandon an asset freeze rather than face a difficult hearing. See id., at 16, 37. But the Kaleys and their amici have also failed to offer any anecdotes of that kind; and we suspect that the far more common reason a prosecutor relinquishes a freeze is just to avoid premature discovery. See supra, at 14–15. So experience, as far as anyone has discerned it, cuts against the Kaleys: It confirms that even under Mathews, they have no right to revisit the grand jury’s finding.[14] IV When we decided Monsanto, we effectively resolved this case too. If the question in a pre-trial forfeiture case is whether there is probable cause to think the defendant committed the crime alleged, then the answer is: whatever the grand jury decides. And even if we test that proposition by applying Mathews, we arrive at the same place: In considering such findings of probable cause, we have never thought the value of enhanced evidentiary procedures worth their costs. Congress of course may strike its own balance and give defendants like the Kaleys the kind of hearing they want. Indeed, Congress could disapprove of Monsanto itself and hold pre-trial seizures of property to a higher standard than probable cause. But the Due Process Clause, even when combined with a defendant’s Sixth Amendment interests, does not command those results. Accordingly, the Kaleys cannot challenge the grand jury’s conclusion that probable cause supports the charges against them. The grand jury gets the final word. We therefore affirm the judgment of the Eleventh Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 Between January 2012 and April 2013, for example, the Department of Justice returned over $1.5 billion in forfeited assets to more than 400,000 crime victims. See Dept. of Justice, Justice Department Returned $1.5 Billion to Victims of Crime Since January 2012 (Apr. 26, 2013), online at http://www.justice.gov/opa/pr/2013/April/13-crm-480.html (as visited Feb. 21, 2014 and available in the Clerk of the Court’s case file). 2 The forfeiture statute itself requires a hearing when the Government seeks to restrain the assets of someone who has not yet been indicted. See . That statutory provision is not at issue in this case, which involves a pair of indicted defendants. 3 At oral argument, the Government agreed that a defendant has a constitutional right to a hearing on that question. See Tr. of Oral Arg. 45. We do not opine on the matter here. 4 Compare v. , 521 F. 3d 411 (CADC 2008) (holding that a defendant is entitled to raise such a challenge); v. , 37 Fed. Appx. 870, 873 (CA9 2002) (same); v. , 39 F. 3d 684, 700 (CA7 1994) (same); v. , 924 F. 2d 1186 (CA2 1991) (en banc) (same), with v. , 427 F. 3d 394, 406–407 (CA6 2005) (prohibiting a defendant from raising such a challenge); v. , 274 F. 3d 800, 803–806 (CA4 2001) (same); v. , 160 F. 3d 641, 648–649 (CA10 1998) (same). 5 An earlier version of the indictment did not include the money laundering charge. In its superseding indictment, the Government also accused Jennifer Gruenstrass, another sales representative, of transporting stolen property and money laundering. Her case went to trial, and she was acquitted. Several other sales representatives participating in the Kaleys’ activity entered guilty pleas (each to a charge of shipping stolen goods) during the Government’s investigation. 6 The grand jury’s unreviewed finding similarly may play a significant role in determining a defendant’s eligibility for release before trial under the Bail Reform Act of 1984, That statute creates a rebuttable presumption that a defendant is ineligible for bail if “there is probable cause to believe” she committed certain serious crimes. §§3142(e)(2)–(3), (f). The Courts of Appeal have uniformly held that presumption to operate whenever an indictment charges those offenses. Relying on our instruction that an indictment returned by a proper grand jury “conclusively determines the existence of probable cause,” the courts have denied defendants’ calls for any judicial reconsideration of that issue. v. , 776 F. 2d 51, 54 (CA2 1985) (quoting v. , , n. 19 (1975)); see, v. , 799 F. 2d 115, 117–119 (CA3 1986); v. , 804 F. 2d 157, 162–163 (CA1 1986) (); v. , 779 F. 2d 1467, 1477–1479 (CA11 1985). 7 Contrary to the dissent’s characterization, see at 11–12, nothing in our reasoning depends on viewing one consequence of a probable cause determination (say, detention) as “greater” than another (say, the asset freeze here). (We suspect that would vary from case to case, with some defendants seeing the loss of liberty as the more significant deprivation and others the loss of a chosen lawyer.) We simply see no reason to treat a grand jury’s probable cause determination as conclusive for all other purposes (including, in some circumstances, locking up the defendant), but not for the one at issue here. 8 A prosecutor, of course, might drop the case because of the court’s ruling, especially if he thought that decision would bring into play an ethical standard barring any charge “that the prosecutor knows is not supported by probable cause.” ABA Model Rule of Professional Conduct 3.8(a) (2013). But then the court would have effectively done what we have long held it cannot: overrule the grand jury on whether to bring a defendant to trial. See , at 7–8. 9 The dissent argues that the same is true when a judge hears evidence on whether frozen assets are traceable to a crime, because that allegation also appears in the indictment. See , at 6–7;, at 3, and n. 3.But the tracing of assets is a technical matter far removed from the grand jury’s core competence and traditional function—to determine whether there is probable cause to think the defendant committed a crime. And a judge’s finding that assets are not traceable to the crime charged in no way casts doubt on the prosecution itself. So that determination does not similarly undermine the grand jury or create internal contradictions within the criminal justice system. 10 The dissent claims as well that the hearing the Kaleys seek “would not be mere relitigation” of the grand jury’s decision because they could now “tell their side of the story.” , at 8. But the same could be said of an adversarial hearing on an indictment’s validity, which everyone agrees is impermissible because it “look[s] into and revise[s]” the grand jury’s judgment. See (quoting v. ). The lesson of our precedents, as described above, is that a grand jury’s finding is “conclusive”—and thus precludes subsequent proceedings on the same matter—even though not arising from adversarial testing. See at 7–8; see also , at 17–18. 11 The dissent says not to worry—the Government can obtain the assets after conviction by using ’s “relation-back” provision. See at 15. That provision is intended to aid the Government in recovering funds transferred to a third party—here, the Kaleys’ lawyer—subsequent to the crime. But forfeiture applies only to specific assets, so in the likely event that the third party has spent the money, the Government must resort to a State’s equitable remedies—which may or may not even be available—to force him to disgorge an equivalent amount. See Tr. of Oral Arg. 48–49. And indeed, if the Government easily recover such monies, then few lawyers would agree to represent defendants like the Kaleys, and the dissent’s proposed holding would be for naught. 12 Compare Cassella, Criminal Forfeiture Procedure, 32 Am. J. Crim. L. 55, 63 (2004) (explaining that “defendants tend to demand the hearing . . . to afford defense counsel an early opportunity to discover the nature of the Government’s criminal case and to cross-examine some of the Government’s witnesses”) with May, Attorney Fees and Government Forfeiture, 34 Champion 20, 23 (Apr. 2010) (advising that “[e]ven if defense counsel cannot prevail on the facts or the law, he may be able to prevail anyway” because “[s]ometimes the government will decide to give up its restraint on a piece of property rather than engage in litigation that will result in early discovery”). 13 Still, a restraint on assets could not deprive the Kaleys of representation sufficient to ensure fair proceedings. The would require the appointment of effective counsel if the Kaleys were unable to hire a lawyer. See v. , ; v. , . The vast majority of criminal defendants proceed with appointed counsel. And the Court has never thought, as the dissent suggests today, that doing so risks the “fundamental fairness of the actual trial.” , at 12; see , at 17–18. If it does, the right way to start correcting the problem is not by adopting the dissent’s position, but by ensuring that the right to effective counsel is fully vindicated. 14 As against all this—all we have formerly held and all other courts have actually found—the dissent cites nothing: not a single decision of ours suggesting, nor a single decision of a lower court demonstrating, that formal, adversarial procedures are at all likely to correct any grand jury errors. The dissent argues only that a hearing will have “probable value” for the Kaleys because “the deprivation of [their] right” to chosen counsel, once accomplished, is “effectively permanent.” , at 16. But that argument confuses two different parts of the inquiry. The dissent’s point well underscores the importance of the Kaleys’ interest: As we have readily acknowledged, if the grand jury made a mistake, the Kaleys have suffered a serious injury, which cannot later be corrected. See , at 16–17. (We note, though, that the dissent, in asserting that injury’s uniqueness, understates the losses that always attend a mistaken indictment, which no ultimate verdict can erase.) But the dissent’s argument about what is at stake for the Kaleys says nothing about the crucial, last prong of , which asks whether and to what extent the adversarial procedures they request will in fact correct any grand jury errors. That part of the analysis is what requires our decision, and the dissent’s view that the Government overreached in this particular case cannot overcome it.

571.US.87

Shortly after respondent Cheever was charged with capital murder, the Kansas Supreme Court found the State’s death penalty scheme unconstitutional. State prosecutors then dismissed their charges to allow federal authorities to prosecute him. When Cheever filed notice that he intended to introduce expert evidence that methamphetamine intoxication negated his ability to form specific intent, the Federal District Court ordered Cheever to submit to a psychiatric evaluation. The federal case was eventually dismissed without prejudice. Meanwhile, this Court held the State’s death penalty scheme constitutional, see Kansas v. Marsh, 548 U.S. 163. The State then brought a second prosecution. At trial, Cheever raised a voluntary intoxication defense, offering expert testimony regarding his methamphetamine use. In rebuttal, the State sought to present testimony from the expert who had examined Cheever by the Federal District Court order. Defense counsel objected, arguing that since Cheever had not agreed to the examination, introduction of the testimony would violate the Fifth Amendment proscription against compelling an accused to testify against himself. The trial court allowed the testimony, and the jury found Cheever guilty and voted to impose a death sentence. The Kansas Supreme Court vacated the conviction and sentence, relying on Estelle v. Smith, 451 U.S. 454, in which this Court held that a court-ordered psychiatric examination violated a defendant’s Fifth Amendment rights when the defendant neither initiated the examination nor put his mental capacity in dispute. The court distinguished the holding of Buchanan v. Kentucky, 483 U.S. 402, that a State may introduce the results of such an examination for the limited purpose of rebutting a mental-status defense, on the basis that voluntary intoxication is not a mental disease or defect under Kansas law. Held: The rule of Buchanan, reaffirmed here, applies in this case to permit the prosecution to offer the rebuttal evidence at issue. Pp. 4–10. (a) In Buchanan, the prosecution presented evidence from a court-ordered evaluation to rebut the defendant’s affirmative defense of extreme emotional disturbance. This Court concluded that this rebuttal testimony did not offend the Fifth Amendment, holding that when a defense expert who has examined the defendant testifies that the defendant lacked the requisite mental state to commit an offense, the prosecution may present psychiatric evidence in rebuttal. Buchanan’s reasoning was not limited to the circumstance that the evaluation was requested jointly by the defense and the government. Nor did the case turn on whether state law referred to extreme emotional disturbance as an affirmative defense. Pp. 4–6. (b) The admission of rebuttal testimony under the rule of Buchanan harmonizes with the principle that when a defendant chooses to testify in a criminal case, the Fifth Amendment does not allow him to refuse to answer related questions on cross-examination. See Fitzpatrick v. United States, 178 U.S. 304, 315. Here, the prosecution elicited testimony from its expert only after Cheever offered expert testimony about his inability to form the requisite mens rea. Excluding this testimony would have undermined Buchanan and the core truth-seeking function of trial. Pp. 6–7. (c) This Court is not persuaded by the Kansas Supreme Court’s reasoning that Cheever did not waive his Fifth Amendment privilege because voluntary intoxication is not a mental disease or defect as a matter of state law. “Mental disease or defect” is not the salient phrase under this Court’s precedents, which use the much broader phrase “mental status,” Buchanan, 483 U. S., at 423. Mental-status defenses include those based on psychological expert evidence as to a defendant’s mens rea, mental capacity to commit the crime, or ability to premeditate. To the extent that the Kansas Supreme Court declined to apply Buchanan because Cheever’s intoxication was “temporary,” this Court’s precedents are again not so narrowly circumscribed, as evidenced by the fact that the courts where Buchanan was tried treated his extreme emotional disturbance as a “temporary” condition. Pp. 7–8. (d) This Court declines to address in the first instance Cheever’s contention that the prosecution’s use of the court-ordered psychiatric examination exceeded the rebuttal-purpose limit established by Buchanan, see 483 U. S., at 424. Pp. 9–10. 295 Kan. 229, 284 P.3d 1007, vacated and remanded. Sotomayor, J., delivered the opinion for a unanimous Court.

The Fifth Amendment to the United States Constitution provides that “[n]o person . . . shall be compelled in any criminal case to be a witness against himself . . . .” The question here is whether the Fifth Amendment prohibits the government from introducing evidence from a court-ordered mental evaluation of a criminal defendant to rebut that defendant’s presentation of expert testimony in support of a defense of voluntary intoxication. We hold that it does not. I On the morning of January 19, 2005, Scott Cheever shot and killed Matthew Samuels, a sheriff of Greenwood County, Kansas, and shot at other local law enforcement officers. In the hours before the shooting, Cheever and his friends had cooked and smoked methamphetamine at a home near Hilltop, Kansas. Samuels and multiple deputies drove there to arrest Cheever on an unrelated outstanding warrant. When one of Cheever’s friends warned him that officers were en route, Cheever rushed outside and tried to drive away, but his car had a flat tire. He returned inside and hid with a friend in an upstairs bedroom, holding a loaded .44 caliber revolver. Cheever then heard footsteps on the stairs leading up to the room, and he stepped out and shot Samuels, who was climbing the stairs. After briefly returning to the bedroom, Cheever walked back to the staircase and shot Samuels again. He also shot at a deputy and a detective, as well as members of a local SWAT (special weapons and tactics) team that had since arrived. Only Samuels was hit. The State charged Cheever with capital murder. But shortly thereafter, in an unrelated case, the Kansas Supreme Court found the State’s death penalty scheme unconstitutional. State v. Marsh, 278 Kan. 520, 102 P. 3d 445 (2004). Rather than continuing to prosecute Cheever without any chance of a death sentence, state prosecutors dismissed their charges and allowed federal authorities to prosecute Cheever under the Federal Death Penalty Act of 1994, 18 U. S. C. §3591 et seq. In the federal case, Cheever filed notice that he “intend[ed] to introduce expert evidence relating to his intoxication by methamphetamine at the time of the events on January 19, 2005, which negated his ability to form spe-cific intent, e.g., malice aforethought, premeditation and deliberation.” App. to Pet. for Cert. 69–70. Pursuant to Federal Rule of Criminal Procedure 12.2(b), the District Court ordered Cheever to submit to a psychiatric evaluation by Michael Welner, a forensic psychiatrist, to assess how methamphetamine use had affected him when he shot Samuels. Welner interviewed Cheever for roughly five and a half hours. The federal case proceeded to trial. Seven days into jury selection, however, defense counsel became unable to continue; the court suspended the proceedings and later dismissed the case without prejudice. Meanwhile, this Court had reversed the Kansas Supreme Court and held that the Kansas death penalty statute was constitutional. Kansas v. Marsh, 548 U. S. 163, 167 (2006) . A second federal prosecution never commenced. Kansas then brought a second state prosecution. At the state trial, Cheever presented a voluntary-intoxication defense, arguing that his methamphetamine use had ren-dered him incapable of premeditation. In support of this argument, Cheever offered testimony from Roswell Lee Evans, a specialist in psychiatric pharmacy and dean of the Auburn University School of Pharmacy. Evans opined that Cheever’s long-term methamphetamine use had damaged his brain. [ 1 ] Evans also testified that on the morning of the shooting, Cheever was acutely intoxicated. According to Evans, Cheever’s actions were “very much influenced by” his use of methamphetamine. After the defense rested, the State sought to present rebuttal testimony from Welner, the expert who had examined Cheever by order of the federal court. Defense counsel objected, arguing that because Welner’s opinions were based in part on an examination to which Cheever had not voluntarily agreed, his testimony would violate the Fifth Amendment proscription against compelling an accused to testify against himself. The State countered that the testimony was necessary to rebut Cheever’s voluntary-intoxication defense. The trial court agreed with the State. The court was persuaded, in part, by the fact that the defense expert had himself relied on Welner’s examination report: “I think that fact alone probably allows the State to call [Welner] to give his own point of view.” App. 92. The court allowed Welner’s testimony for the purpose of showing that Cheever shot Samuels “because of his antisocial personal-ity, not because his brain was impaired by methamphetamine.” Id., at 94. The jury found Cheever guilty of murder and attempted murder. At the penalty phase, it unanimously voted to impose a sentence of death, and the trial court accepted that verdict. On appeal to the Kansas Supreme Court, Cheever argued that the State had violated his Fifth Amendment rights when it introduced, through Welner’s testimony, statements that he had made during the federal court-ordered mental examination. The court agreed, relying primarily on Estelle v. Smith, 451 U. S. 454 (1981) , in which we held that a court-ordered psychiatric exami-nation violated the defendant’s Fifth Amendment rights when the defendant neither initiated the examination nor put his mental capacity in dispute at trial. 295 Kan. 229, 243–244, 284 P. 3d 1007, 1019–1020 (2012) (per curiam). The court acknowledged, id., at 244–245, 284 P. 3d, at 1020, our holding that a State may introduce the results of a court-ordered mental examination for the limited purpose of rebutting a mental-status defense. Buchanan v. Kentucky, 483 U. S. 402 –424 (1987). But it distinguished Buchanan on the basis that under Kansas law, voluntary intoxication is not a “mental disease or defect.” 295 Kan., at 250, 284 P. 3d, at 1023. Consequently, it vacated Cheever’s conviction and sentence, holding that Cheever had not waived his Fifth Amendment privilege and that his federal court-ordered examination should not have been used against him at the state-court trial. Ibid. We granted certiorari, 568 U. S. ___ (2013), and now reverse. II The Fifth Amendment guarantees that “[n]o person . . . shall be compelled in any criminal case to be a witness against himself . . . .” We held in Estelle that under the Fifth Amendment, when a criminal defendant “neither initiates a psychiatric evaluation nor attempts to introduce any psychiatric evidence,” his compelled statements to a psychiatrist cannot be used against him. 451 U. S., at 468. In that case, a judge ordered a psychiatric exam-ination to determine the defendant’s competency to stand trial. Id., at 456–457. The prosecution then used statements from that examination during the sentencing phase of the trial as evidence of the defendant’s future dangerousness. Id., at 458–460. Emphasizing that the defendant had neither “introduced” any “psychiatric evidence,” nor even “indicated that he might do so,” id., at 466, we concluded that the Fifth Amendment did not permit the State to use the defendant’s statements in this manner. In Buchanan, we addressed the admissibility of evidence from a court-ordered evaluation where—unlike in Estelle—a defendant had introduced psychiatric evidence related to his mental-status defense. We held that the Fifth Amendment allowed the prosecution to present evidence from the evaluation to rebut the defendant’s affirmative defense of extreme emotional disturbance. And while, as Cheever notes, the mental evaluation in Buchanan was requested jointly by the defense and the government, our holding was not limited to that circumstance. Moreover, contrary to Cheever’s suggestion, the case did not turn on whether state law referred to extreme emotional disturbance as an “affirmative defense.” Buchanan, 483 U. S., at 408, 422 (holding that the prosecution’s use of rebuttal expert testimony is permissible where a defendant “presents psychiatric evidence”). The rule of Buchanan, which we reaffirm today, is that where a defense expert who has examined the defendant testifies that the defendant lacked the requisite mental state to commit an offense, the prosecution may present psychiatric evidence in rebuttal. Ibid. Any other rule would undermine the adversarial process, allowing a defendant to provide the jury, through an expert operating as proxy, with a one-sided and potentially inaccurate view of his mental state at the time of the alleged crime. The admission of this rebuttal testimony harmonizes with the principle that when a defendant chooses to testify in a criminal case, the Fifth Amendment does not allow him to refuse to answer related questions on cross-examination. A defendant “has no right to set forth to the jury all the facts which tend in his favor without laying himself open to a cross-examination upon those facts.” Fitzpatrick v. United States, 178 U. S. 304, 315 (1900) . We explained in Brown v. United States, 356 U. S. 148 (1958) , which involved a witness’s refusal to answer questions in a civil case, that where a party provides testimony and then refuses to answer potentially incriminating questions, “[t]he interests of the other party and regard for the function of courts of justice to ascertain the truth become relevant, and prevail in the balance of considerations determining the scope and limits of the privilege against self-incrimination.” Id., at 156. When a defendant presents evidence through a psychological expert who has examined him, the government likewise is permitted to use the only effective means of challenging that evidence: testimony from an expert who has also examined him. See United States v. Byers, 740 F. 2d 1104, 1113 (CADC 1984) (en banc) (holding that the Government could present rebuttal expert testimony in part because it is perhaps “the most trustworthy means of attempting to meet” the burden of proof (internal quotation marks omitted)). [ 2 ] The prosecution here elicited testimony from its expert only after Cheever offered expert testimony about his in-ability to form the requisite mens rea. The testimony of the government expert rebutted that of Cheever’s expert. See id. at 1114 (“Ordinarily the only effective rebuttal of psychiatric opinion testimony is contradictory opinion tes-timony; and for that purpose . . . the basic tool of psy-chiatric study remains the personal interview, which requires rapport between the interviewer and the subject” (internal quotation marks omitted)); State v. Druke, 143 Ariz. 314, 318, 693 P. 2d 969, 973 (App. 1984) (“[A]n in-ference would arise that the evidence presented by the [defendant] as to his mental condition is true because un-contradicted”). The trial court therefore did not violate the Fifth Amendment when it allowed Welner to testify that Cheever “made a choice to shoot,” App. 131, because the State permissibly followed where the defense led. Excluding this testimony would have undermined Buchanan and the core truth-seeking function of the trial. III Neither the Kansas Supreme Court’s reasoning, nor Cheever’s arguments, persuade us not to apply the settled rule of Buchanan. A Although the Kansas Supreme Court acknowledged that the State may present evidence obtained from a compelled psychiatric examination when “the defendant presents evidence at trial that he or she lacked the requisite criminal intent due to mental disease or defect,” 295 Kan., at 249, 284 P. 3d, at 1023, it reasoned that voluntary intoxication is not a “mental disease or defect” as a matter of state law. Id., at 250, 284 P. 3d, at 1023–1024 (citing State v. Kleypas, 272 Kan. 894, 40 P. 3d 139 (2001)). The court therefore concluded that “Cheever did not waive his Fifth Amendment privilege and thus permit his court-ordered examination by Dr. Welner to be used against him at trial.” 295 Kan., at 251, 284 P. 3d, at 1024. This reasoning misconstrues our precedents. Although Kansas law defines “mental disease or defect” narrowly, to exclude voluntary intoxication, that phrase is actually not the salient one under our precedents. In Buchanan, we permitted rebuttal testimony where the defendant presented evidence of “the ‘mental status’ defense of extreme emotional disturbance.” 483 U. S., at 423. And “mental status” is a broader term than “mental disease or defect,” at least to the extent that Kansas law excludes voluntary intoxication from that definition. Mental-status defenses include those based on psychological expert evidence as to a defendant’s mens rea, mental capacity to commit the crime, or ability to premeditate. Defendants need not as-sert a “mental disease or defect” in order to assert a defense based on “mental status.” To the extent that the Kansas Supreme Court declined to apply Buchanan because Cheever’s intoxication was “temporary,” our precedents are again not so narrowly circumscribed. Like voluntary intoxication, extreme emotional disturbance is a “temporary” condition, at least according to the Kentucky state courts where Buchanan was tried. See McClellan v. Commonwealth, 715 S. W. 2d 464, 468–469 (Ky. 1986) (defining extreme emotional disturbance as “a temporary state of mind so enraged, inflamed, or disturbed as to overcome one’s judgment, and to cause one to act uncontrollably from [an] impelling force of [an] extreme emotional disturbance rather than from evil or malicious purposes”). We nonetheless held in Buchanan that the defense of extreme emotional disturbance, when supported by expert testimony, may be re-butted with expert testimony. The same is true here. Cheever’s psychiatric evidence concerned his mental status because he used it to argue that he lacked the requisite mental capacity to premeditate. The Fifth Amendment therefore did not bar the State from using Welner’s examination to rebut Cheever’s voluntary-intoxication defense. B Cheever further contends that the Fifth Amendment imposes limits on the State’s ability to introduce rebuttal evidence regarding a defendant’s mental status. According to Cheever, Welner’s testimony exceeded these limits by describing the shooting from Cheever’s perspective; [ 3 ] by insinuating that he had a personality disorder; and by discussing his alleged infatuation with criminals. We have held that testimony based on a court-ordered psychiatric evaluation is admissible only for a “limited rebuttal purpose.” Buchanan, 483 U. S., at 424. In Buchanan, for example, although the prosecution had used a psychiatric report to rebut the defendant’s evidence of extreme emotional disturbance, we noted that the trial court had redacted the report so as to avoid exposing the jury to “the very different issue” of the defendant’s competency to stand trial. Id., at 423, n. 20. Two years later, we explained in dictum that “[n]othing” in our precedents “suggests that a defendant opens the door to the admission of psychiatric evidence on future dangerousness by raising an insanity defense at the guilt stage of the trial.” Powell v. Texas, 492 U. S. 680 –686, n. 3 (1989) ( per curiam). Here, however, the Kansas Supreme Court did not address whether Welner’s testimony exceeded the scope of rebuttal testimony permitted by the Fifth Amendment or by the State’s evidentiary rules. We accordingly decline to address this issue in the first instance. [ 4 ] * * * We hold that where a defense expert who has examined the defendant testifies that the defendant lacked the requisite mental state to commit a crime, the prosecution may offer evidence from a court-ordered psychological examination for the limited purpose of rebutting the defendant’s evidence. The judgment of the Kansas Supreme Court is therefore vacated, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Notes 1 Evans described this damage as “neurotoxicity,” which is “the qual-ity of exerting a destructive or poisonous effect upon the nerve tissue.” The Sloane-Dorland Annotated Medical-Legal Dictionary 498 (1987). 2 For that reason, we reject Cheever’s suggestion that the State could effectively have rebutted the testimony of his expert by introducing testimony from experts who had not personally examined him. 3 In an extended soliloquy, Dr. Welner narrated the crime from Cheever’s perspective, in part as follows: “I don’t jump out of the window the way my confederate later does. And when I do shoot,I don’t shoot before Matthew Samuels walks through the curtain in such a way that I might scare him, the way my later shots frightened the deputies that came to pull him away, but I shoot him at a point in which he is very much within my range, has passed through that curtain, and I know that he is coming upstairs, and that is when I shoot.” App. 130–131. 4 Kansas contends that reaching a federal constitutional question may not be necessary because Cheever argued in opposing certiorari that the scope of Welner’s testimony violated state evidentiary rules. Reply Brief 4–5. We agree with the State that the impact of Kansas evidentiary rules is a matter best left to the state courts to decide on remand. We do observe, however, that while our holding today suggests a constitutional ceiling on the scope of expert testimony that the prosecution may introduce in rebuttal, States (and Congress) remain free to impose additional limitations on the scope of such rebuttal evidence in state and federal trials.

573.US.228

As Director of Community Intensive Training for Youth (CITY), a program for underprivileged youth operated by Central Alabama Community College (CACC), petitioner Edward Lane conducted an audit of the program’s expenses and discovered that Suzanne Schmitz, an Alabama State Representative on CITY’s payroll, had not been reporting for work. Lane eventually terminated Schmitz’ employment. Shortly thereafter, federal authorities indicted Schmitz on charges of mail fraud and theft concerning a program receiving federal funds. Lane testified, under subpoena, regarding the events that led to his terminating Schmitz. Schmitz was convicted and sentenced to 30 months in prison. Meanwhile, CITY was experiencing significant budget shortfalls. Respondent Franks, then CACC’s president, terminated Lane along with 28 other employees in a claimed effort to address the financial difficulties. A few days later, however, Franks rescinded all but 2 of the 29 terminations—those of Lane and one other employee. Lane sued Franks in his individual and official capacities under 42 U. S. C. §1983, alleging that Franks had violated the First Amendment by firing him in retaliation for testifying against Schmitz. The District Court granted Franks’ motion for summary judgment, holding that the individual-capacity claims were barred by qualified immunity and the official-capacity claims were barred by the Eleventh Amendment. The Eleventh Circuit affirmed, holding that Lane’s testimony was not entitled to First Amendment protection. It reasoned that Lane spoke as an employee and not as a citizen because he acted pursuant to his official duties when he investigated and terminated Schmitz’ employment. Held: 1. Lane’s sworn testimony outside the scope of his ordinary job duties is entitled to First Amendment protection. Pp. 6–13. (a) Pickering v. Board of Ed. of Township High School Dist. 205, Will Cty., 391 U.S. 563, 568, requires balancing “the interests of the [employee], as a citizen, in commenting upon matters of public concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.” Under the first step of the Pickering analysis, if the speech is made pursuant to the employee’s ordinary job duties, then the employee is not speaking as a citizen for First Amendment purposes, and the inquiry ends. Garcetti v. Ceballos, 547 U.S. 410, 421. But if the “employee spoke as a citizen on a matter of public concern,” the inquiry turns to “whether the relevant government entity had an adequate justification for treating the employee differently from any other member of the general public.” Id., at 418. Pp. 6–8. (b) Lane’s testimony is speech as a citizen on a matter of public concern. Pp. 8–12. (1) Sworn testimony in judicial proceedings is a quintessential example of citizen speech for the simple reason that anyone who testifies in court bears an obligation, to the court and society at large, to tell the truth. That obligation is distinct and independent from any separate obligations a testifying public employee might have to his employer. The Eleventh Circuit read Garcetti far too broadly in holding that Lane did not speak as a citizen when he testified simply because he learned of the subject matter of that testimony in the course of his employment. Garcetti said nothing about speech that relates to public employment or concerns information learned in the course of that employment. The critical question under Garcetti is whether the speech at issue is itself ordinarily within the scope of an employee’s duties, not whether it merely concerns those duties. Indeed, speech by public employees on subject matter related to their employment holds special value precisely because those employees gain knowledge of matters of public concern through their employment. Pp. 9–11. (2) Whether speech is a matter of public concern turns on the “content, form, and context” of the speech. Connick v. Myers, 461 U.S. 138, 147–148. Here, corruption in a public program and misuse of state funds obviously involve matters of significant public concern. See Garcetti, 547 U. S., at 425. And the form and context of the speech—sworn testimony in a judicial proceeding—fortify that conclusion. See United States v. Alvarez, 567 U. S. ___, ___. Pp. 11–12. (c) Turning to Pickering’s second step, the employer’s side of the scale is entirely empty. Respondents do not assert, and cannot demonstrate, any government interest that tips the balance in their favor—for instance, evidence that Lane’s testimony was false or erro-neous or that Lane unnecessarily disclosed sensitive, confidential, or privileged information while testifying. Pp. 12–13. 2. Franks is entitled to qualified immunity for the claims against him in his individual capacity. The question here is whether Franks reasonably could have believed that, when he fired Lane, a government employer could fire an employee because of testimony the employee gave, under oath and outside the scope of his ordinary job responsibilities. See Ashcroft v. al-Kidd, 563 U. S. ___, ___. At the relevant time, Eleventh Circuit precedent did not preclude Franks from holding that belief, and no decision of this Court was sufficiently clear to cast doubt on controlling Circuit precedent. Any discrepancies in Eleventh Circuit precedent only serve to highlight the dispositive point that the question was not beyond debate at the time Franks acted. Pp. 13–17. 3. The Eleventh Circuit declined to consider the District Court’s dismissal of the claims against respondent Burrow in her official capacity as CACC’s acting president, and the parties have not asked this Court to consider them here. The judgment of the Eleventh Circuit as to those claims is reversed, and the case is remanded for further proceedings. P. 17. 523 Fed. Appx. 709, affirmed in part, reversed in part, and remanded. Sotomayor, J., delivered the opinion for a unanimous Court. Thomas, J., filed a concurring opinion, in which Scalia and Alito, JJ., joined.

Almost 50 years ago, this Court declared that citizens do not surrender their First Amendment rights by accepting public employment. Rather, the First Amendment protection of a public employee’s speech depends on a careful balance “between the interests of the [employee], as a citizen, in commenting upon matters of public concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.” Pickering v. Board of Ed. of Township High School Dist. 205, Will Cty., 391 U. S. 563, 568 (1968) . In Pickering, the Court struck the balance in favor of the public employee, extending First Amendment protection to a teacher who was fired after writing a letter to the editor of a local newspaper criticizing the school board that employed him. Today, we consider whether the First Amendment similarly protects a public employee who provided truthful sworn testimony, compelled by sub-poena, outside the course of his ordinary job responsibilities. We hold that it does. I In 2006, Central Alabama Community College (CACC) hired petitioner Edward Lane to be the Director of Community Intensive Training for Youth (CITY), a statewide program for underprivileged youth. CACC hired Lane on a probationary basis. In his capacity as Director, Lane was responsible for overseeing CITY’s day-to-day operations, hiring and firing employees, and making decisions with respect to the program’s finances. At the time of Lane’s appointment, CITY faced significant financial difficulties. That prompted Lane to conduct a comprehensive audit of the program’s expenses. The audit revealed that Suzanne Schmitz, an Alabama State Representative on CITY’s payroll, had not been reporting to her CITY office. After unfruitful discussions with Schmitz, Lane shared his finding with CACC’s president and its attorney. They warned him that firing Schmitz could have negative repercussions for him and CACC. Lane nonetheless contacted Schmitz again and in-structed her to show up to the Huntsville office to serveas a counselor. Schmitz refused; she responded that shewished to “ ‘continue to serve the CITY program in the same manner as [she had] in the past.’ ” Lane v. Central Ala. Community College, 523 Fed. Appx. 709, 710 (CA11 2013) (per curiam). Lane fired her shortly thereafter. Schmitz told another CITY employee, Charles Foley, that she intended to “ ‘get [Lane] back’ ” for firing her. 2012 WL 5289412, *1 (ND Ala., Oct. 18, 2012). She also said that if Lane ever requested money from the state legislature for the program, she would tell him, “ ‘[y]ou’re fired.’ ” Ibid. Schmitz’ termination drew the attention of many, including agents of the Federal Bureau of Investigation, which initiated an investigation into Schmitz’ employment with CITY. In November 2006, Lane testified before a federal grand jury about his reasons for firing Schmitz. In January 2008, the grand jury indicted Schmitz on four counts of mail fraud and four counts of theft concerning a program receiving federal funds. See United States v. Schmitz, 634 F. 3d 1247, 1256–1257 (CA11 2011). The indictment alleged that Schmitz had collected $177,251.82 in federal funds even though she performed “ ‘virtually no services,’ ” “ ‘generated virtually no work product,’ ” and “ ‘rarely even appeared for work at the CITY Program offices.’ ” Id., at 1260. It further alleged that Schmitz had submitted false statements concerning the hours she worked and the nature of the services she performed. Id., at 1257. Schmitz’ trial, which garnered extensive press coverage,[1] commenced in August 2008. Lane testified, under subpoena, regarding the events that led to his terminating Schmitz. The jury failed to reach a verdict. Roughly six months later, federal prosecutors retried Schmitz, and Lane testified once again. This time, the jury convicted Schmitz on three counts of mail fraud and four countsof theft concerning a program receiving federal funds. The District Court sentenced her to 30 months in prison and ordered her to pay $177,251.82 in restitution and forfeiture. Meanwhile, CITY continued to experience considerable budget shortfalls. In November 2008, Lane began reporting to respondent Steve Franks, who had become president of CACC in January 2008. Lane recommended that Franks consider layoffs to address the financial difficulties. In January 2009, Franks decided to terminate 29 probationary CITY employees, including Lane. Shortly thereafter, however, Franks rescinded all but 2 of the 29 terminations—those of Lane and one other employee— because of an “ambiguity in [those other employees’] probationary service.” Brief for Respondent Franks 11. Franks claims that he “did not rescind Lane’s termination . . . because he believed that Lane was in a fundamentally different category than the other employees: he was the director of the entire CITY program, and not simply an employee.” Ibid. In September 2009, CACC eliminated the CITY program and terminated the program’s remaining employees. Franks later retired, and respondent Susan Burrow, the current Acting President of CACC, replaced him while this case was pending before the Eleventh Circuit. In January 2011, Lane sued Franks in his individual and official capacities under Rev. Stat. §1979, 42 U. S. C. §1983, alleging that Franks had violated the First Amendment by firing him in retaliation for his testimony against Schmitz.[2] Lane sought damages from Franks in his individual capacity and sought equitable relief, including reinstatement, from Franks in his official capacity.[3] The District Court granted Franks’ motion for summary judgment. Although the court concluded that the record raised “genuine issues of material fact . . . concerning [Franks’] true motivation for terminating [Lane’s] employment,” 2012 WL 5289412, *6, it held that Franks was entitled to qualified immunity as to the damages claims because “a reasonable government official in [Franks’] position would not have had reason to believe that the Constitution protected [Lane’s] testimony,” id., *12. The District Court relied on Garcetti v. Ceballos, 547 U. S. 410 (2006) , which held that “ ‘when public employees make statements pursuant to their official duties, the employees are not speaking as citizens for First Amendment pur-poses.’ ” 2012 WL 5289412, *10 (quoting Garcetti, 547 U. S.,at 421). The court found no violation of clearly established law because Lane had “learned of the information that he testified about while working as Director at [CITY],” such that his “speech [could] still be considered as part of his official job duties and not made as a citizen on a matter of public concern.” 2012 WL 5289412, *10. The Eleventh Circuit affirmed. 523 Fed. Appx., at 710. Like the District Court, it relied extensively on Garcetti. It reasoned that, “[e]ven if an employee was not required to make the speech as part of his official duties, he enjoys no First Amendment protection if his speech ‘owes its existence to [the] employee’s professional responsibilities’ and is ‘a product that the “employer himself has commissioned or created.” ’ ” Id., at 711 (quoting Abdur-Rahman v. Walker, 567 F. 3d 1278, 1283 (CA11 2009)). The court concluded that Lane spoke as an employee and not as a citizen because he was acting pursuant to his official duties when he investigated Schmitz’ employment, spoke with Schmitz and CACC officials regarding the issue, and terminated Schmitz. 523 Fed. Appx., at 712. “That Lane testified about his official activities pursuant to a sub-poena and in the litigation context,” the court continued,“does not bring Lane’s speech within the protection of the First Amendment.” Ibid. The Eleventh Circuit also concluded that, “even if . . . a constitutional violation of Lane’s First Amendment rights occurred in these circumstances, Franks would be entitled to qualified immunity in his personal capacity” because the right at issue had not been clearly established. Id., at 711, n. 2. We granted certiorari, 571 U. S. __ (2014), to resolve discord among the Courts of Appeals as to whether public employees may be fired—or suffer other adverse employment consequences—for providing truthful subpoenaed testimony outside the course of their ordinary job responsibilities. Compare 523 Fed. Appx., at 712 (case below), with, e.g., Reilly v. Atlantic City, 532 F. 3d 216, 231 (CA3 2008). II Speech by citizens on matters of public concern lies at the heart of the First Amendment, which “was fashioned to assure unfettered interchange of ideas for the bringing about of political and social changes desired by the people,” Roth v. United States, 354 U. S. 476, 484 (1957) . This remains true when speech concerns information related to or learned through public employment. After all, public employees do not renounce their citizenship when they accept employment, and this Court has cautioned time and again that public employers may not condition employment on the relinquishment of constitutional rights. See, e.g., Keyishian v. Board of Regents of Univ. of State of N. Y., 385 U. S. 589, 605 (1967) ; Pickering, 391 U. S., at 568; Connick v. Myers, 461 U. S. 138, 142 (1983) . There is considerable value, moreover, in encouraging, rather than inhibiting, speech by public employees. For “[g]overnment employees are often in the best position to know what ails the agencies for which they work.” Waters v. Churchill, 511 U. S. 661, 674 (1994) (plurality opinion). “The interest at stake is as much the public’s interest in receiving informed opinion as it is the employee’s own right to disseminate it.” San Diego v. Roe, 543 U. S. 77, 82 (2004) (per curiam). Our precedents have also acknowledged the government’s countervailing interest in controlling the operation of its workplaces. See, e.g., Pickering, 391 U. S., at 568. “Government employers, like private employers, need a significant degree of control over their employees’ words and actions; without it, there would be little chance for the efficient provision of public services.” Garcetti, 547 U. S., at 418. Pickering provides the framework for analyzing whether the employee’s interest or the government’s interest should prevail in cases where the government seeks to curtail the speech of its employees. It requires “balanc[ing] . . . the interests of the [public employee], as a citizen, in commenting upon matters of public concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.” 391 U. S., at 568. In Pickering, the Court held that a teacher’s letter to the editor of a local news-paper concerning a school budget constituted speech on amatter of public concern. Id., at 571. And in balancing the employee’s interest in such speech against the government’s efficiency interest, the Court held that the publication of the letter did not “imped[e] the teacher’s proper performance of his daily duties in the classroom” or “interfer[e] with the regular operation of the schools generally.” Id., at 572–573. The Court therefore held that the teacher’s speech could not serve as the basis for his dismissal. Id., at 574. In Garcetti, we described a two-step inquiry into whether a public employee’s speech is entitled to protection: “The first requires determining whether the employee spoke as a citizen on a matter of public concern. If the answer is no, the employee has no First Amendment cause of action based on his or her employer’s reaction to the speech. If the answer is yes, then the possibility of a First Amendment claim arises. The question becomes whether the relevant government entity had an adequate justification for treating the employee differently from any other member of the general public.” 547 U. S., at 418 (citations omitted). In describing the first step in this inquiry, Garcetti distinguished between employee speech and citizen speech. Whereas speech as a citizen may trigger protection, the Court held that “when public employees make statements pursuant to their official duties, the employees are not speaking as citizens for First Amendment pur-poses, and the Constitution does not insulate their communi-cations from employer discipline.” Id., at 421. Applying that rule to the facts before it, the Court found that an internal memorandum prepared by a prosecutor in the course of his ordinary job responsibilities constituted unprotected employee speech. Id., at 424. III Against this backdrop, we turn to the question pre-sented: whether the First Amendment protects a public employee who provides truthful sworn testimony, compelledby subpoena, outside the scope of his ordinary job responsibilities.[4] We hold that it does. A The first inquiry is whether the speech in question—Lane’s testimony at Schmitz’ trials—is speech as a citizen on a matter of public concern. It clearly is. 1 Truthful testimony under oath by a public employee outside the scope of his ordinary job duties is speech as a citizen for First Amendment purposes. That is so even when the testimony relates to his public employment or concerns information learned during that employment. In rejecting Lane’s argument that his testimony was speech as a citizen, the Eleventh Circuit gave short shrift to the nature of sworn judicial statements and ignored the obligation borne by all witnesses testifying under oath. See 523 Fed. Appx., at 712 (finding immaterial the fact that Lane spoke “pursuant to a subpoena and in the litigation context”). Sworn testimony in judicial proceedings is a quintessential example of speech as a citizen for a simple reason: Anyone who testifies in court bears an obligation, to the court and society at large, to tell the truth. See, e.g., 18 U. S. C. §1623 (criminalizing false statements under oath in judicial proceedings); United States v. Mandujano, 425 U. S. 564, 576 (1976) (plurality opinion) (“Perjured testimony is an obvious and flagrant affront to the basic concept of judicial proceedings”). When the person testifying is a public employee, he may bear separate obligations to his employer—for example, an obligation not to show up to court dressed in an unprofessional manner. But any such obligations as an employee are distinct and independent from the obligation, as a citizen, to speak the truth. That independent obligation renders sworn testimony speech as a citizen and sets it apart from speech made purely in the capacity of an employee. In holding that Lane did not speak as a citizen whenhe testified, the Eleventh Circuit read Garcetti far too broadly. It reasoned that, because Lane learned of the sub-ject matter of his testimony in the course of his employment with CITY, Garcetti requires that his testimony be treated as the speech of an employee rather than that of a citizen. See 523 Fed. Appx., at 712. It does not. The sworn testimony in this case is far removed from the speech at issue in Garcetti—an internal memorandum prepared by a deputy district attorney for his supervisors recommending dismissal of a particular prosecution. The Garcetti Court held that such speech was made pursuant to the employee’s “official responsibilities” because “[w]hen [the employee] went to work and performed the tasks he was paid to perform, [he] acted as a government employee. The fact that his duties sometimes required him to speak or write does not mean that his supervisors were prohib-ited from evaluating his performance.” 547 U. S., at 422, 424. But Garcetti said nothing about speech that simply relates to public employment or concerns information learned in the course of public employment. The Garcetti Court made explicit that its holding did not turn on the fact that the memo at issue “concerned the subject matter of [the prosecutor’s] employment,” because “[t]he First Amendment protects some expressions related to the speaker’s job.” Id., at 421. In other words, the mere fact that a citizen’s speech concerns information acquired by virtue of his public employment does not transform that speech into employee—rather than citizen—speech. The critical question under Garcetti is whether the speech at issue is itself ordinarily within the scope of an employee’s duties, not whether it merely concerns those duties. It bears emphasis that our precedents dating back to Pickering have recognized that speech by public employees on subject matter related to their employment holdsspecial value precisely because those employees gainknowledge of matters of public concern through their employment. In Pickering, for example, the Court observed that “[t]eachers are . . . the members of a commu-nity most likely to have informed and definite opinions as to how funds allotted to the operation of the schools should be spent. Accordingly, it is essential that they be able to speak out freely on such questions without fear of retaliatory dismissal.” 391 U. S., at 572; see also Garcetti, 547 U. S., at 421 (recognizing that “[t]he same is true of many other categories of public employees”). Most recently, in San Diego v. Roe, 543 U. S., at 80, the Court again observed that public employees “are uniquely qualified to comment” on “matters concerning government policies that are of interest to the public at large.” The importance of public employee speech is especially evident in the context of this case: a public corruption scandal. The United States, for example, represents that because “[t]he more than 1000 prosecutions for federal corruption offenses that are brought in a typical year . . . often depend on evidence about activities that government officials undertook while in office,” those prosecutions often “require testimony from other government employees.” Brief for United States as Amicus Curiae 20. It would be antithetical to our jurisprudence to conclude that the very kind of speech necessary to prosecute corruption by public officials—speech by public employees regarding information learned through their employment—may never form the basis for a First Amendment retaliation claim. Such a rule would place public employees who witness corruption in an impossible position, torn between the obligation to testify truthfully and the desire to avoid retaliation and keep their jobs. Applying these principles, it is clear that Lane’s sworn testimony is speech as a citizen. 2 Lane’s testimony is also speech on a matter of public concern. Speech involves matters of public concern “when it can ‘be fairly considered as relating to any matter of political, social, or other concern to the community,’ or when it ‘is a subject of legitimate news interest; that is, a subject of general interest and of value and concern to the public.’ ” Snyder v. Phelps, 562 U. S. ___, ___ (2011) (slip op., at 6–7) (citation omitted). The inquiry turns on the “content, form, and context” of the speech. Connick, 461 U. S., at 147–148. The content of Lane’s testimony—corruption in a public program and misuse of state funds—obviously involves a matter of significant public concern. See, e.g., Garcetti, 547 U. S., at 425 (“Exposing governmental inefficiency and misconduct is a matter of considerable significance”). And the form and context of the speech—sworn testimony in a judicial proceeding—fortify that conclusion. “Unlike speech in other contexts, testimony under oath has the formality and gravity necessary to remind the witness that his or her statements will be the basis for official governmental action, action that often affects the rights and liberties of others.” United States v. Alvarez, 567 U. S. ___, ___ (2012) (slip op., at 8–9) (plurality opinion). * * * We hold, then, that Lane’s truthful sworn testimony at Schmitz’ criminal trials is speech as a citizen on a matter of public concern. B This does not settle the matter, however. A public employee’s sworn testimony is not categorically entitled to First Amendment protection simply because it is speech as a citizen on a matter of public concern. Under Pickering, if an employee speaks as a citizen on a matter of public concern, the next question is whether the government had “an adequate justification for treating the employee differently from any other member of the public” based on the government’s needs as an employer. Garcetti, 547 U. S., at 418. As discussed previously, we have recognized that government employers often have legitimate “interest[s] in the effective and efficient fulfillment of [their] responsibilities to the public,” including “ ‘promot[ing] efficiency and integrity in the discharge of official duties,’ ” and “ ‘maintain[ing] proper discipline in public service.’ ” Connick, 461 U. S., at 150–151. We have also cautioned, however, that “a stronger showing [of government interests] may be necessary if the employee’s speech more substantially involve[s] matters of public concern.” Id., at 152. Here, the employer’s side of the Pickering scale is entirely empty: Respondents do not assert, and cannot demonstrate, any government interest that tips the balance in their favor. There is no evidence, for example, that Lane’s testimony at Schmitz’ trials was false or erroneous or that Lane unnecessarily disclosed any sensitive, confidential, or privileged information while testifying.[5] In these circumstances, we conclude that Lane’s speech is entitled to protection under the First Amendment. The Eleventh Circuit erred in holding otherwise and dismissing Lane’s claim of retaliation on that basis. IV Respondent Franks argues that even if Lane’s testimony is protected under the First Amendment, the claims against him in his individual capacity should be dismissed on the basis of qualified immunity. We agree. Qualified immunity “gives government officials breathing room to make reasonable but mistaken judgments about open legal questions.” Ashcroft v. al-Kidd, 563 U. S. ___, ___ (2011) (slip op., at 12). Under this doctrine, courts may not award damages against a government official in his personal capacity unless “the official violated a statutory or constitutional right,” and “the right was ‘clearly established’ at the time of the challenged conduct.” Id., at ___ (slip op., at 3). The relevant question for qualified immunity purposes is this: Could Franks reasonably have believed, at the time he fired Lane, that a government employer could fire an employee on account of testimony the employee gave, under oath and outside the scope of his ordinary job responsibilities? Eleventh Circuit precedent did not preclude Franks from reasonably holding that belief. And no decision of this Court was sufficiently clear to cast doubt on the controlling Eleventh Circuit precedent. In dismissing Lane’s claim, the Eleventh Circuit relied on its 1998 decision in Morris v. Crow, 142 F. 3d 1379 (per curiam). There, a deputy sheriff sued the sheriff and two other officials, alleging that he had been fired in retaliation for statements he made in an accident report and later giving deposition testimony about his investigation of a fatal car crash between another officer and a citizen. Id., at 1381. In his accident report, the plaintiff noted that the officer was driving more than 130 mph in a 50 mph zone, without using his emergency blue warning light. See ibid. The plaintiff later testified to these facts at a deposition in a wrongful death suit against the sheriff’s office. Ibid. His superiors later fired him. Ibid. The Eleventh Circuit, in a pre-Garcetti decision, concluded that the plaintiff’s deposition testimony was unprotected. It held that a public employee’s speech is protected only when it is “ ‘made primarily in the employee’s role as citizen,’ ” rather than “ ‘primarily in the role of employee.’ ” Morris, 142 F. 3d, at 1382. And it found the plaintiff’s deposition testimony to be speech as an em-ployee because it “reiterated the conclusions regardinghis observations of the accident” that he “generated in thenormal course of [his] duties.” Ibid. Critically, the court acknowledged—and was unmoved by—the fact that al-though the plaintiff had investigated the accident andprepared the report pursuant to his official duties, there was no “evidence that [he] gave deposition testimony for any reason other than in compliance with a subpoena to testify truthfully in the civil suit regarding the . . . accident.” Ibid. The court further reasoned that the speech could not “be characterized as an attempt to make public comment on sheriff’s office policies and procedures, the internal workings of the department, the quality of its employees or upon any issue at all.” Ibid. Lane argues that two other Eleventh Circuit precedents put Franks on notice that his conduct violated the First Amendment: Martinez v. Opa-Locka, 971 F. 2d 708 (1992) (per curiam), and Tindal v. Montgomery Cty. Comm’n, 32 F. 3d 1535 (1994). Martinez involved a public employee’s subpoenaed testimony before the Opa-Locka City Commission regarding her employer’s procurement practices. 971 F. 2d, at 710. The Eleventh Circuit held that her speech was protected, reasoning that it addressed a matter of public concern and that her interest in speaking freely was not outweighed by her employer’s interest in providing government services. Id., at 712. It held, further, that the relevant constitutional rules were so clearly established at the time that qualified immunity did not apply. Id., at 713. Tindal, decided two years after Martinez, involved a public employee’s subpoenaed testimony in her co-worker’s sexual harassment lawsuit. 32 F. 3d, at 1537–1538. The court again ruled in favor of the em-ployee. It held that the employee’s speech touched upona public concern and that her employer had not offered any evidence that the speech hindered operations. Id., at 1539–1540. Morris, Martinez, and Tindal represent the landscape of Eleventh Circuit precedent the parties rely on for qualified immunity purposes. If Martinez and Tindal were controlling in the Eleventh Circuit in 2009, we would agree with Lane that Franks could not reasonably have believed that it was lawful to fire Lane in retaliation for his testimony. But both cases must be read together with Morris, which reasoned—in declining to afford First Amendment protection—that the plaintiff’s decision to testify was motivated solely by his desire to comply with a subpoena. The same could be said of Lane’s decision to testify. Franks was thus entitled to rely on Morris when he fired Lane.[6] Lane argues that Morris is inapplicable because it distinguished Martinez, suggesting that Martinez survived Morris. See Morris, 142 F. 3d, at 1382–1383. But this debate over whether Martinez or Morris applies to Lane’s claim only highlights the dispositive point: At the time of Lane’s termination, Eleventh Circuit precedent did not provide clear notice that subpoenaed testimony concerning information acquired through public employment is speech of a citizen entitled to First Amendment protection. At best, Lane can demonstrate only a discrepancy in Eleventh Circuit precedent, which is insufficient to defeat the defense of qualified immunity. Finally, Lane argues that decisions of the Third and Seventh Circuits put Franks on notice that his firing of Lane was unconstitutional. See Reilly, 532 F. 3d, at 231 (CA3) (truthful testimony in court is citizen speech protected by the First Amendment); Morales v. Jones, 494 F. 3d 590, 598 (CA7 2007) (similar). But, as the court below acknowledged, those precedents were in direct conflict with Eleventh Circuit precedent. See 523 Fed. Appx., at 712, n. 3. There is no doubt that the Eleventh Circuit incorrectly concluded that Lane’s testimony was not entitled to First Amendment protection. But because the question was not “beyond debate” at the time Franks acted, al-Kidd, 563 U. S., at ___ (slip op., at 9), Franks is entitled to qualified immunity. V Lane’s speech is entitled to First Amendment protection, but because respondent Franks is entitled to qualified immunity, we affirm the judgment of the Eleventh Circuit as to the claims against Franks in his individual capacity. Our decision does not resolve, however, the claims against Burrow—initially brought against Franks when he served as President of CACC—in her official capacity. Although the District Court dismissed those claims for prospective relief as barred by the Eleventh Amendment, the Eleventh Circuit declined to consider that question on appeal, see 523 Fed. Appx., at 711 (“Because Lane has failed to establish a prima facie case of retaliation, we do not decide about Franks’ defense of sovereign immunity”), and the parties have not asked us to consider it now. We therefore reverse the judgment of the Eleventh Circuit as to those claims and remand for further proceedings. * * * For the foregoing reasons, the judgment of the United States Court of Appeals for the Eleventh Circuit is affirmed in part and reversed in part, and the case is remandedfor further proceedings consistent with this opinion. It is so ordered.Notes 1 See, Lawmaker Faces Fraud Charge in June, Montgomery Advertiser, May 6, 2008, p. 1B; Johnson, State Lawmaker’s Fraud Trial Starts Today, Montgomery Advertiser, Aug. 18, 2008, p. 1B; Faulk, Schmitz Testifies in Her Defense: Says State Job was Legitimate, Birmingham News, Feb. 20, 2009, p. 1A; Faulk, Schmitz Convicted, Loses her State Seat, Birmingham News, Feb. 25, 2009, p. 1A. 2 Lane also brought claims against CACC, as well as claims under a state whistleblower statute, Ala. Code §36–26A–3 (2013), and . Those claims are not at issue here. 3 Because Burrow replaced Franks as President of CACC during the pendency of this lawsuit, the claims originally filed against Franks in his official capacity are now against Burrow. 4 It is undisputed that Lane’s ordinary job responsibilities did not include testifying in court proceedings. See v. , 523 Fed. Appx. 709, 712 (CA11 2013). For that reason, Lane asked the Court to decide only whether truthful sworn testimony that is not a part of an employee’s ordinary job responsibilities is citizen speech on a matter of public concern. Pet. for Cert. i. We accordingly need not address in this case whether truthful sworn testimony would constitute citizen speech under when given as part of a public employee’s ordinary job duties, and express no opinion on the matter today. 5 Of course, quite apart from balancing, wrongdoing that an employee admits to while testifying may be a valid basis for termination or other discipline. 6 There is another reason undermines and . In and , the Eleventh Circuit asked only whether the speech at issue addressed a matter of public concern. , which appeared to anticipate , asked both whether the speech at issue was speech of an employee (and not a citizen) and whether it touched upon a matter of public concern. In this respect, one could read as cabining and .

571.US.415

Petitioner Law filed for Chapter 7 bankruptcy. He valued his California home at $363,348, claiming that $75,000 of that value was covered by California’s homestead exemption and thus was exempt from the bankruptcy estate. See 11 U. S. C. §522(b)(3)(A). He also claimed that the sum of two voluntary liens—one of which was in favor of “Lin’s Mortgage & Associates”—exceeded the home’s nonexempt value, leaving no equity recoverable for his other creditors. Respondent Siegel, the bankruptcy estate trustee, challenged the “Lin” lien in an adversary proceeding, but protracted and expensive litigation ensued when a supposed “Lili Lin” in China claimed to be the beneficiary of Law’s deed of trust. Ultimately, the Bankruptcy Court concluded that the loan was a fiction created by Law to preserve his equity in the house. It thus granted Siegel’s motion to “surcharge” Law’s $75,000 homestead exemption, making those funds available to defray attorney’s fees incurred by Siegel in overcoming Law’s fraudulent misrepresentations. The Ninth Circuit Bankruptcy Appellate Panel and the Ninth Circuit affirmed. Held: The Bankruptcy Court exceeded the limits of its authority when it ordered that the $75,000 protected by Law’s homestead exemption be made available to pay Siegel’s attorney’s fees. Pp. 5–12. (a) A bankruptcy court may not exercise its authority to “carry out” the provisions of the Code, 11 U. S. C. §105(a), or its “inherent power . . . to sanction ‘abusive litigation practices,’ ” Marrama v. Citizens Bank of Mass., 549 U.S. 365, 375–376, by taking action prohibited elsewhere in the Code. Here, the Bankruptcy Court’s “surcharge” contravened §522, which (by reference to California law) entitled Law to exempt $75,000 of equity in his home from the bankruptcy estate, §522(b)(3)(A), and which made that $75,000 “not liable for payment of any administrative expense,” §522(k), including attorney’s fees, see §503(b)(2). The surcharge thus exceeded the limits of both the court’s authority under §105(a) and its inherent powers. Pp. 5–7. (b) Siegel argues that an equitable power to deny an exemption by “surcharging” exempt property in response to a debtor’s misconduct can coexist with §522. But insofar as that argument equates the surcharge with an outright denial of Law’s homestead exemption, it founders on this case’s procedural history. The Bankruptcy Appellate Panel recognized that because no one timely objected to the homestead exemption, it became final before the surcharge was imposed. And a trustee who fails to make a timely objection cannot challenge an exemption. Taylor v. Freeland & Kronz, 503 U.S. 638, 643–644. Assuming the Bankruptcy Court could have revisited Law’s entitlement to the exemption, §522 specifies the criteria that render property exempt, and a court may not refuse to honor a debtor’s invocation of an exemption without a valid statutory basis. Federal courts may apply state law to disallow state-created exemptions, but federal law itself provides no authority for bankruptcy courts to deny an exemption on a ground not specified in the Code. Pp. 7–10. (c) Neither the holding of Marrama v. Citizens Bank nor its dictum points toward a different result. There, the debtor’s bad faith kept him from converting his bankruptcy from a Chapter 7 liquidation to a Chapter 13 reorganization as permitted by §706(a). But that was because his conduct prevented him from qualifying under Chapter 13, and thus he could not satisfy §706(d), which expressly conditions conversion on the debtor’s ability to qualify under Chapter 13. Pp. 10–11. (d) This ruling forces Siegel to shoulder a heavy financial burden due to Law’s egregious misconduct and may produce inequitable results for other trustees and creditors, but it is not for courts to alter the balance that Congress struck in crafting §522. Cf. Guidry v. Sheet Metal Workers National Pension Fund, 493 U.S. 365, 376–377. P. 11. (e) Ample authority remains to address debtor misconduct, including denial of discharge, see §727(a)(2)–(6); sanctions for bad-faith litigation conduct under the Bankruptcy Rules, §105(a), or a bankruptcy court’s inherent powers; enforcement of monetary sanctions through the normal procedures for collecting money judgments, see §727(b); or possible prosecution under 18 U. S. C. §152. Pp. 11–12. 435 Fed. Appx. 697, reversed and remanded. Scalia, J., delivered the opinion for a unanimous Court.

The Bankruptcy Code provides that a debtor may exempt certain assets from the bankruptcy estate. It further provides that exempt assets generally are not liable for any expenses associated with administering the estate. In this case, we consider whether a bankruptcy court nonetheless may order that a debtor’s exempt assets be used to pay administrative expenses incurred as a result of the debtor’s misconduct. I. Background A Chapter 7 of the Bankruptcy Code gives an insolvent debtor the opportunity to discharge his debts by liquidating his assets to pay his creditors. 11 U. S. C. §§704(a)(1), 726, 727. The filing of a bankruptcy petition under Chapter 7 creates a bankruptcy “estate” generally comprising all of the debtor’s property. §541(a)(1). The estate is placed under the control of a trustee, who is responsible for managing liquidation of the estate’s assets and distribution of the proceeds. §704(a)(1). The Code authorizes the debtor to “exempt,” however, certain kinds of property from the estate, enabling him to retain those assets postbankruptcy. §522(b)(1). Except in particular situations specified in the Code, exempt property “is not liable” for the payment of “any [prepetition] debt” or “any administrative expense.” §522(c), (k). Section 522(d) of the Code provides a number of exemptions unless they are specifically prohibited by state law. §522(b)(2), (d). One, commonly known as the “homestead exemption,” protects up to $22,975 in equity in the debtor’s residence. §522(d)(1) and note following §522; see Owen v. Owen, 500 U. S. 305, 310 (1991) . The debtor may elect, however, to forgo the §522(d) exemptions and instead claim whatever exemptions are available under applicable state or local law. §522(b)(3)(A). Some States provide homestead exemptions that are more generous than the federal exemption; some provide less generous versions; but nearly every State provides some type of homestead exemption. See López, State Homestead Exemptions and Bankruptcy Law: Is It Time for Congress To Close the Loophole? 7 Rutgers Bus. L. J. 143, 149–165 (2010) (listing state exemptions). B Petitioner, Stephen Law, filed for Chapter 7 bankruptcy in 2004, and respondent, Alfred H. Siegel, was appointed to serve as trustee. The estate’s only significant asset was Law’s house in Hacienda Heights, California. On a schedule filed with the Bankruptcy Court, Law valued the house at $363,348 and claimed that $75,000 of its value was covered by California’s homestead exemption. See Cal. Civ. Proc. Code Ann. §704.730(a)(1) (West Supp. 2014). He also reported that the house was subject to two voluntary liens: a note and deed of trust for $147,156.52 in favor of Washington Mutual Bank, and a second note and deed of trust for $156,929.04 in favor of “Lin’s Mortgage & Associates.” Law thus represented that there was no equity in the house that could be recovered for his other creditors, because the sum of the two liens exceeded the house’s nonexempt value. If Law’s representations had been accurate, he presumably would have been able to retain the house, since Siegel would have had no reason to pursue its sale. Instead, a few months after Law’s petition was filed, Siegel initiated an adversary proceeding alleging that the lien in favor of “Lin’s Mortgage & Associates” was fraudulent. The deed of trust supporting that lien had been recorded by Law in 1999 and reflected a debt to someone named “Lili Lin.” Not one but two individuals claiming to be Lili Lin ultimately responded to Siegel’s complaint. One, Lili Lin of Artesia, California, was a former acquaintance of Law’s who denied ever having loaned him money and described his repeated efforts to involve her in various sham transactions relating to the disputed deed of trust. That Lili Lin promptly entered into a stipulated judgment disclaiming any interest in the house. But that was not the end of the matter, because the second “Lili Lin” claimed to be the true beneficiary of the disputed deed of trust. Over the next five years, this “Lili Lin” managed—despite supposedly living in China and speaking no English—to engage in extensive and costly litigation, including several appeals, contesting the avoidance of the deed of trust and Siegel’s subsequent sale of the house. Finally, in 2009, the Bankruptcy Court entered an order concluding that “no person named Lili Lin ever made a loan to [Law] in exchange for the disputed deed of trust.” In re Law, 401 B. R. 447, 453 (Bkrtcy. Ct. CD Cal.). The court found that “the loan was a fiction, meant to preserve [Law’s] equity in his residence beyond what he was entitled to exempt” by perpetrating “a fraud on his creditors and the court.” Ibid. With regard to the second “Lili Lin,” the court declared itself “unpersuaded that Lili Lin of China signed or approved any declaration or pleading purporting to come from her.” Ibid. Rather, it said, the “most plausible conclusion” was that Law himself had “authored, signed, and filed some or all of these papers.” Ibid. It also found that Law had submitted false evidence “in an effort to persuade the court that Lili Lin of China—rather than Lili Lin of Artesia—was the true holder of the lien on his residence.” Id., at 452. The court determined that Siegel had incurred more than $500,000 in attorney’s fees overcoming Law’s fraudulent misrepresentations. It therefore granted Siegel’s motion to “surcharge” the entirety of Law’s $75,000 homestead exemption, making those funds available to defray Siegel’s attorney’s fees. The Ninth Circuit Bankruptcy Appellate Panel affirmed. BAP No. CC–09–1077–PaMkH, 2009 WL 7751415 (Oct. 22, 2009) ( per curiam). It held that the Bankruptcy Court’s factual findings regarding Law’s fraud were not clearly erroneous and that the court had not abused its discretion by surcharging Law’s exempt assets. It explained that in Latman v. Burdette, 366 F. 3d 774 (2004), the Ninth Circuit had recognized a bankruptcy court’s power to “equitably surcharge a debtor’s statutory ex-emptions” in exceptional circumstances, such as “when a debtor engages in inequitable or fraudulent conduct.” 2009 WL 7751415, *5, *7. The Bankruptcy Appellate Panel acknowledged that the Tenth Circuit had disagreed with Latman, see In re Scrivner, 535 F. 3d 1258, 1263–1265 (2008), but the panel affirmed that Latman was correct. 2009 WL 7751415, *7, n. 10. Judge Markell filed a concurring opinion agreeing with the panel’s application of Latman but questioning “whether Latman remains good policy.” 2009 WL 7751415, *10. The Ninth Circuit affirmed. In re Law, 435 Fed. Appx. 697 (2011) ( per curiam). It held that the surcharge was proper because it was “calculated to compensate the estate for the actual monetary costs imposed by the debtor’s misconduct, and was warranted to protect the integrity of the bankruptcy process.” Id., at 698. We granted certiorari. 570 U. S. ___ (2013). II. Analysis A A bankruptcy court has statutory authority to “issue any order, process, or judgment that is necessary or appropriate to carry out the provisions of” the Bankruptcy Code. 11 U. S. C. §105(a). And it may also possess “inherent power . . . to sanction ‘abusive litigation practices.’ ” Marrama v. Citizens Bank of Mass., 549 U. S. 365 –376 (2007). But in exercising those statutory and inherent powers, a bankruptcy court may not contravene specific statutory provisions. It is hornbook law that §105(a) “does not allow the bankruptcy court to override explicit mandates of other sections of the Bankruptcy Code.” 2 Collier on Bankruptcy ¶105.01[2], p. 105–6 (16th ed. 2013). Section 105(a) confers authority to “carry out” the provisions of the Code, but it is quite impossible to do that by taking action that the Code prohibits. That is simply an application of the axiom that a statute’s general permission to take actions of a certain type must yield to a specific prohibition found elsewhere. See Morton v. Mancari, 417 U. S. 535 –551 (1974); D. Ginsberg & Sons, Inc. v. Popkin, 285 U. S. 204 –208 (1932).[1] Courts’ inherent sanctioning powers are likewise subordinate to valid statutory directives and prohibitions. Degen v. United States, 517 U. S. 820, 823 (1996) ; Chambers v. NASCO, Inc., 501 U. S. 32, 47 (1991) . We have long held that “whatever equitable powers remain in the bankruptcy courts must and can only be exercised within the confines of” the Bankruptcy Code. Norwest Bank Worthington v. Ahlers, 485 U. S. 197, 206 (1988) ; see, e.g., Raleigh v. Illinois Dept. of Revenue, 530 U. S. 15 –25 (2000); United States v. Noland, 517 U. S. 535, 543 (1996) ; SEC v. United States Realty & Improvement Co., 310 U. S. 434, 455 (1940) . Thus, the Bankruptcy Court’s “surcharge” was unauthorized if it contravened a specific provision of the Code. We conclude that it did. Section 522 (by reference to California law) entitled Law to exempt $75,000 of equity in his home from the bankruptcy estate. §522(b)(3)(A). And it made that $75,000 “not liable for payment of any administrative expense.” §522(k).[2] The reasonable attorney’s fees Siegel incurred defeating the “Lili Lin” lien were indubitably an administrative expense, as a short march through a few statutory cross-references makes plain: Section 503(b)(2) provides that administrative expenses include “compensation . . . awarded under” §330(a); §330(a)(1) authorizes “reasonable compensation for actual, necessary services rendered” by a “professional person employed under” §327; and §327(a) authorizes the trustee to “employ one or more attorneys . . . to represent or assist the trustee in carrying out the trustee’s duties under this title.” Siegel argues that even though attorney’s fees incurred responding to a debtor’s fraud qualify as “administrative expenses” for purposes of determining the trustee’s right to reimbursement under §503(b), they do not so qualify for purposes of §522(k); but he gives us no reason to depart from the “ ‘normal rule of statutory construction’ ” that words repeated in different parts of the same statute generally have the same meaning. See Department of Revenue of Ore. v. ACF Industries, Inc., 510 U. S. 332, 342 (1994) (quoting Sorenson v. Secretary of Treasury, 475 U. S. 851, 860 (1986) ). The Bankruptcy Court thus violated §522’s express terms when it ordered that the $75,000 protected by Law’s homestead exemption be made available to pay Siegel’s attorney’s fees, an administrative expense. In doing so, the court exceeded the limits of its authority under §105(a) and its inherent powers. B Siegel does not dispute the premise that a bankruptcy court’s §105(a) and inherent powers may not be exercised in contravention of the Code. Instead, his main argument is that the Bankruptcy Court’s surcharge did not contravene §522. That statute, Siegel contends, “establish[es] the procedure by which a debtor may seek to claim exemptions” but “contains no directive requiring [courts] to allow [an exemption] regardless of the circumstances.” Brief for Respondent 35. Thus, he says, recognition of an equitable power in the Bankruptcy Court to deny an exemption by “surcharging” the exempt property in response to the debtor’s misconduct can coexist comfortably with §522. The United States, appearing in support of Siegel, agrees, arguing that §522 “neither gives debtors an absolute right to retain exempt property nor limits a court’s authority to impose an equitable surcharge on such property.” Brief for United States as Amicus Curiae 23. Insofar as Siegel and the United States equate the Bankruptcy Court’s surcharge with an outright denial of Law’s homestead exemption, their arguments founder upon this case’s procedural history. The Bankruptcy Appellate Panel stated that because no one “timely oppose[d] [Law]’s homestead exemption claim,” the exemption “became final” before the Bankruptcy Court imposed the surcharge. 2009 WL 7751415, at *2. We have held that a trustee’s failure to make a timely objection prevents him from challenging an exemption. Taylor v. Freeland & Kronz, 503 U. S. 638 –644 (1992). But even assuming the Bankruptcy Court could have revisited Law’s entitlement to the exemption, §522 does not give courts discretion to grant or withhold exemptions based on whatever considerations they deem appropriate. Rather, the statute exhaustively specifies the criteria that will render property exempt. See §522(b), (d). Siegel insists that because §522(b) says that the debtor “may exempt” certain property, rather than that he “shall be entitled” to do so, the court retains discretion to grant or deny exemptions even when the statutory criteria are met. But the subject of “may exempt” in §522(b) is the debtor, not the court, so it is the debtor in whom the statute vests discretion. A debtor need not invoke an exemption to which the statute entitles him; but if he does, the court may not refuse to honor the exemption absent a valid statutory basis for doing so. Moreover, §522 sets forth a number of carefully calibrated exceptions and limitations, some of which relate to the debtor’s misconduct. For example, §522(c) makes exempt property liable for certain kinds of prepetition debts, including debts arising from tax fraud, fraud in connection with student loans, and other specified types of wrongdoing. Section 522(o) prevents a debtor from claiming a homestead exemption to the extent he acquired the homestead with nonexempt property in the previous 10 years “with the intent to hinder, delay, or defraud a creditor.” And §522(q) caps a debtor’s homestead exemption at approximately $150,000 (but does not eliminate it en-tirely) where the debtor has been convicted of a felony that shows “that the filing of the case was an abuse of the provisions of” the Code, or where the debtor owes a debt arising from specified wrongful acts—such as securities fraud, civil violations of the Racketeer Influenced and Corrupt Organizations Act, or “any criminal act, intentional tort, or willful or reckless misconduct that caused serious physical injury or death to another individualin the preceding 5 years.” §522(q) and note following§522. The Code’s meticulous—not to say mind-numbingly detailed—enumeration of exemptions and exceptions to those exemptions confirms that courts are not authorized to create additional exceptions. See Hillman v. Maretta, 569 U. S. ___, ___ (2013) (slip op., at 12); TRW Inc. v. Andrews, 534 U. S. 19 –29 (2001). Siegel points out that a handful of courts have claimed authority to disallow an exemption (or to bar a debtor from amending his schedules to claim an exemption, which is much the same thing) based on the debtor’s fraudulent concealment of the asset alleged to be exempt. See, e.g., In re Yonikus, 996 F. 2d 866, 872–873 (CA7 1993); In re Doan, 672 F. 2d 831, 833 (CA11 1982) ( per curiam); Stewart v. Ganey, 116 F. 2d 1010, 1011 (CA5 1940). He suggests that those decisions reflect a general, equitable power in bankruptcy courts to deny exemptions based on a debtor’s bad-faith conduct. For the reasons we have given, the Bankruptcy Code admits no such power. It is of course true that when a debtor claims a state-created exemption, the exemption’s scope is determined by state law, which may provide that certain types of debtor misconduct warrant denial of the exemption. E.g., In re Sholdan, 217 F. 3d 1006, 1008 (CA8 2000); see 4 Collier on Bankruptcy ¶522.08[1]–[2], at 522–45 to 522–47. Some of the early decisions on which Siegel relies, and which the Fifth Circuit cited in Stewart, are instances in which federal courts applied state law to disallow state-created exemptions. See In re Denson, 195 F. 857, 858 (ND Ala. 1912); Cowan v. Burchfield, 180 F. 614, 619 (ND Ala. 1910); In re Ansley Bros., 153 F. 983, 984 (EDNC 1907). But federal law provides no authority for bankruptcy courts to deny an exemption on a ground not specified in the Code. C Our decision in Marrama v. Citizens Bank, on which Siegel and the United States heavily rely, does not point toward a different result. The question there was whether a debtor’s bad-faith conduct was a valid basis for a bankruptcy court to refuse to convert the debtor’s bankruptcy from a liquidation under Chapter 7 to a reorganization under Chapter 13. Although §706(a) of the Code gave the debtor a right to convert the case, §706(d) “expressly conditioned” that right on the debtor’s “ability to qualify as a ‘debtor’ under Chapter 13.” 549 U. S., at 372. And §1307(c) provided that a proceeding under Chapter 13 could be dismissed or converted to a Chapter 7 proceeding “for cause,” which the Court interpreted to authorize dismissal or conversion for bad-faith conduct. In light of §1307(c), the Court held that the debtor’s bad faith could stop him from qualifying as a debtor under Chapter 13, thus preventing him from satisfying §706(d)’s express condition on conversion. Id., at 372–373. That holding has no relevance here, since no one suggests that Law failed to satisfy any express statutory condition on his claiming of the homestead exemption. True, the Court in Marrama also opined that the Bankruptcy Court’s refusal to convert the case was authorized under §105(a) and might have been authorized under the court’s inherent powers. Id., at 375–376. But even that dictum does not support Siegel’s position. In Marrama, the Court reasoned that if the case had been converted to Chapter 13, §1307(c) would have required it to be either dismissed or reconverted to Chapter 7 in light of the debtor’s bad faith. Therefore, the Court suggested, even if the Bankruptcy Court’s refusal to convert the case had not been expressly authorized by §706(d), that action could have been justified as a way of providing a “prompt, rather than a delayed, ruling on [the debtor’s] unmeritorious at-tempt to qualify” under §1307(c). Id., at 376. At most, Marrama’s dictum suggests that in some circumstances a bankruptcy court may be authorized to dispense with futile procedural niceties in order to reach more expeditiously an end result required by the Code. Marrama most certainly did not endorse, even in dictum, the view that equitable considerations permit a bankruptcy court to contravene express provisions of the Code. D We acknowledge that our ruling forces Siegel to shoulder a heavy financial burden resulting from Law’s egregious misconduct, and that it may produce inequitable results for trustees and creditors in other cases. We have recognized, however, that in crafting the provisions of §522, “Congress balanced the difficult choices that exemption limits impose on debtors with the economic harm that exemptions visit on creditors.” Schwab v. Reilly, 560 U. S. 770, 791 (2010) . The same can be said of the limits imposed on recovery of administrative expenses by trustees. For the reasons we have explained, it is not for courts to alter the balance struck by the statute. Cf. Guidry v. Sheet Metal Workers Nat. Pension Fund, 493 U. S. 365 –377 (1990). * * * Our decision today does not denude bankruptcy courts of the essential “authority to respond to debtor misconduct with meaningful sanctions.” Brief for United States as Amicus Curiae 17. There is ample authority to deny the dishonest debtor a discharge. See §727(a)(2)–(6). (That sanction lacks bite here, since by reason of a postpetition settlement between Siegel and Law’s major creditor, Law has no debts left to discharge; but that will not often be the case.) In addition, Federal Rule of Bankruptcy Pro-cedure 9011—bankruptcy’s analogue to Civil Rule 11—authorizes the court to impose sanctions for bad-faith litigation conduct, which may include “an order directing payment. . . of some or all of the reasonable attorneys’ fees and other expenses incurred as a direct result of the violation.” Fed. Rule Bkrtcy. Proc. 9011(c)(2). The court may also possess further sanctioning authority under either §105(a) or its inherent powers. Cf. Chambers, 501 U. S., at 45–49. And because it arises postpetition, a bankruptcy court’s monetary sanction survives the bankruptcy case and is thereafter enforceable through the normal procedures for collecting money judgments. See §727(b). Fraudulent conduct in a bankruptcy case may also subject a debtor to criminal prosecution under 18 U. S. C. §152, which carries a maximum penalty of five years’ imprisonment. But whatever other sanctions a bankruptcy court may impose on a dishonest debtor, it may not contravene express provisions of the Bankruptcy Code by ordering that the debtor’s exempt property be used to pay debts and expenses for which that property is not liable under the Code. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The second sentence of §105(a) adds little to the analysis. It states: “No provision of this title providing for the raising of an issue by a party in interest shall be construed to preclude the court from, sua sponte, taking any action or making any determination necessary or appropriate to enforce or implement court orders or rules, or to prevent an abuse of process.” Even if the “abuse of process” language were deemed to confer additional authority beyond that conferred by the first sentence (which is doubtful), that general authority would also be limited by more specific provisions of the Code. 2 The statute’s general rule that exempt assets are not liable foradministrative expenses is subject to two narrow exceptions, both per-taining to the use of exempt assets to pay expenses associated with the avoidance of certain voidable transfers of exempt property. §522(k)(1)–(2). Neither of those exceptions is relevant here.

571.US.429

To safeguard investors in public companies and restore trust in the financial markets following the collapse of Enron Corporation, Congress passed the Sarbanes-Oxley Act of 2002. One of the Act’s provisions protects whistleblowers; at the time relevant here, that provision instructed: “No [public] company . . ., or any . . . contractor [or] subcontractor . . . of such company, may discharge, demote, suspend, threaten, harass, or . . . discriminate against an employee in the terms and conditions of employment because of [whistleblowing activity].” 18 U. S. C. §1514A(a). Plaintiffs below, petitioners here, are former employees of respondents (collectively FMR), private companies that contract to advise or manage mutual funds. As is common in the industry, the mutual funds served by FMR are public companies with no employees. Both plaintiffs allege that they blew the whistle on putative fraud relating to the mutual funds and, as a consequence, suffered retaliation by FMR. Each commenced suit in federal court. Moving to dismiss the suits, FMR argued that the plaintiffs could state no claim under §1514A, for that provision protects only employees of public companies, and not employees of private companies that contract with public companies. On interlocutory appeal from the District Court’s denial of FMR’s motion to dismiss, the First Circuit reversed, concluding that the term “an employee” in §1514A(a) refers only to employees of public companies. Held: The judgment is reversed and the case is remanded. 670 F.3d 61, reversed and remanded. Justice Ginsburg delivered the opinion of the Court, concluding that §1514A’s whistleblower protection includes employees of a public company’s private contractors and subcontractors. Pp. 9–29. (a) This reading of §1514A is supported by the provision’s text. Pp. 9–16. (1) The Court looks first to the ordinary meaning of the provision’s language. See Moskal v. United States, 498 U.S. 103, 108. As relevant here, §1514A(a) provides that “no . . . contractor . . . may discharge . . . an employee.” The ordinary meaning of “an employee” in this proscription is the contractor’s own employee. FMR’s “narrower construction” requires inserting “of a public company” after “an employee,” but where Congress meant “an employee of a public company,” it said so. The provision as a whole supports this reading. The prohibited retaliatory measures enumerated in §1514A(a)—discharge, demotion, suspension, threats, harassment, or discrimination in employment terms and conditions—are actions an employer takes against its own employees. Contractors are not ordinarily positioned to take adverse actions against employees of the public company with whom they contract. FMR’s interpretation of §1514A, therefore, would shrink to insignificance the provision’s ban on retaliation by contractors. The protected activity covered by §1514A, and the provision’s enforcement procedures and remedies, also indicate that Congress presumed an employer-employee relationship between the retaliator and the whistleblowing employee. Pp. 9–14. (2) FMR’s textual arguments are unpersuasive. It urges that “an employee” must be read to refer exclusively to public company employees to avoid the absurd result of extending protection to the personal employees of company officers and employees, e.g., their housekeepers or gardeners. This concern appears more theoretical than real and, in any event, is outweighed by the compelling arguments opposing FMR’s reading of §1514A. FMR also urges that its reading is supported by the provision’s statutory headings, but those headings are “not meant to take the place of the detailed provisions of the text.” Trainmen v. Baltimore & Ohio R. Co., 331 U.S. 519, 528. Pp. 14–16. (b) Other considerations support the Court’s textual analysis. Pp. 16–27. (1) The Court’s reading fits §1514A’s aim to ward off another Enron debacle. The legislative record shows Congress’ understanding that outside professionals bear significant responsibility for reporting fraud by the public companies with whom they contract, and that fear of retaliation was the primary deterrent to such reporting by the employees of Enron’s contractors. Sarbanes-Oxley contains numerous provisions designed to control the conduct of accountants, auditors, and lawyers who work with public companies, but only §1514A affords such employees protection from retaliation by their employers for complying with the Act’s reporting requirements. Pp. 16–20. (2) This Court’s reading of §1514A avoids insulating the entire mutual fund industry from §1514A. Virtually all mutual funds are structured so that they have no employees of their own; they are managed, instead, by independent investment advisors. Accordingly, the “narrower construction” endorsed by FMR would leave §1514A with no application to mutual funds. The Court’s reading of §1514A, in contrast, protects the employees of investment advisors, who are often the only firsthand witnesses to shareholder fraud involving mutual funds. Pp. 20–22. (3) There is scant evidence that today’s decision will open any floodgates for whistleblowing suits outside §1514A’s purposes. The Department of Labor’s regulations have interpreted §1514A as protecting contractor employees for almost a decade, yet FMR is unable to identify a single case in which the employee of a private contractor has asserted a §1514A claim based on allegations unrelated to shareholder fraud. Plaintiffs and the Solicitor General suggest various limiting principles to dispel any overbreadth problems. This Court need not determine §1514A’s bounds here, however, because, if plaintiffs’ allegations prove true, plaintiffs are precisely the “firsthand witnesses to [the shareholder] fraud” Congress anticipated §1514A would protect. S. Rep. No. 107–146, p. 10. Pp. 22–24. (4) The 2010 Dodd-Frank Wall Street Reform and Consumer Protection Act does not affect this Court’s task of determining whether Congress in 2002 afforded protection to whistleblowing contractor employees. Pp. 24–27. (c) AIR 21’s whistleblower protection provision has been read to cover, in addition to employees of air carriers, employees of contractors and subcontractors of the carriers. Given the parallel statutory texts and whistleblower protective aims, the Court reads the words “an employee” in AIR 21 and in §1514A to have similar import. Pp. 27–29. Justice Scalia, joined by Justice Thomas, relying only on 18 U. S. C. §1514A’s text and broader context, agreed that §1514A protects employees of private contractors from retaliation when they report covered forms of fraud. Pp. 1–3. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Breyer and Kagan, JJ., joined, and in which Scalia and Thomas, JJ., joined in principal part. Scalia, J., filed an opinion concurring in principal part and concurring in the judgment, in which Thomas, J., joined. Sotomayor, J., filed a dissenting opinion, in which Kennedy and Alito, JJ., joined.

To safeguard investors in public companies and restore trust in the financial markets following the collapse of Enron Corporation, Congress enacted the Sarbanes-Oxley Act of 2002, 116Stat. 745. See S. Rep. No. 107–146, pp. 2–11 (2002). A provision of the Act, 18 U. S. C. §1514A, protects whistleblowers. Section 1514A, at the time here relevant, instructed: “No [public] company . . . , or any officer, employee, contractor, subcontractor, or agent of such company, may discharge, demote, suspend, threaten, harass,or in any other manner discriminate against an employee in the terms and conditions of employment because of [whistleblowing or other protected activity].” §1514A(a) (2006 ed.). This case concerns the definition of the protected class: Does §1514A shield only those employed by the public company itself, or does it shield as well employees of privately held contractors and subcontractors—for example, investment advisers, law firms, accounting enterprises—who perform work for the public company? We hold, based on the text of §1514A, the mischief to which Congress was responding, and earlier legislation Congress drew upon, that the provision shelters employees of private contractors and subcontractors, just as it shelters employees of the public company served by the contractors and subcontractors. We first summarize our principal reasons, then describe this controversy and explain our decision more comprehensively. Plaintiffs below, petitioners here, are former employees of private companies that contract to advise or manage mutual funds. The mutual funds themselves are public companies that have no employees. Hence, if the whistle is to be blown on fraud detrimental to mutual fund investors, the whistleblowing employee must be on another company’s payroll, most likely, the payroll of the mutual fund’s investment adviser or manager. Taking the allegations of the complaint as true, both plaintiffs blew the whistle on putative fraud relating to the mutual funds and, as a consequence, suffered adverse action by their employers. Plaintiffs read §1514A to convey that “[n]o . . . contractor . . . may . . . discriminate against [its own] employee [for whistleblowing].” We find that reading consistent with the text of the statute and with common sense. Contractors are in control of their own employees, but are not ordinarily positioned to control someone else’s workers. Moreover, we resist attributing to Congress a purpose to stop a contractor from retaliating against whistleblowers employed by the public company the contractor serves, while leaving the contractor free to retaliate against its own employees when they reveal corporate fraud. In the Enron scandal that prompted the Sarbanes-Oxley Act, contractors and subcontractors, including the accounting firm Arthur Andersen, participated in Enron’s fraud and its coverup. When employees of those contractors attempted to bring misconduct to light, they encountered retaliation by their employers. The Sarbanes-Oxley Act contains numerous provisions aimed at controlling the conduct of accountants, auditors, and lawyers who work with public companies. See, e.g., 116Stat. 750–765, 773–774, 784, §§101–107, 203–206, 307. Given Congress’ concern about contractor conduct of the kind that contributed to Enron’s collapse, we regard with suspicion construction of §1514A to protect whistleblowers only when they are employed by a public company, and not when they work for the public company’s contractor. Congress borrowed §1514A’s prohibition against retaliation from the wording of the 2000 Wendell H. Ford Aviation Investment and Reform Act for the 21st Century (AIR 21), 49 U. S. C. §42121. That Act provides: “No air carrier or contractor or subcontractor of an air carrier may discharge an employee or otherwise discriminate against an employee with respect to compensation, terms, conditions, or privileges of employment” when the employee provides information regarding violations “relating to air carrier safety” to his or her employer or federal authorities. §42121(a)(1). AIR 21 has been read to cover, in addition to employees of air carriers, employees of contractors and subcontractors of the carriers. Given the parallel statu-tory texts and whistleblower protective aims, we readthe words “an employee” in AIR 21 and in §1514A to have similar import. I A The Sarbanes-Oxley Act of 2002 (Sarbanes-Oxley or Act) aims to “prevent and punish corporate and criminal fraud, protect the victims of such fraud, preserve evidence of such fraud, and hold wrongdoers accountable for their actions.” S. Rep. No. 107–146, p. 2 (2002) (hereinafter S. Rep.).[1] Of particular concern to Congress was abundant evidence that Enron had succeeded in perpetuating its massive shareholder fraud in large part due to a “corporate code of silence”; that code, Congress found, “discourage[d] employees from reporting fraudulent behavior not only to the proper authorities, such as the FBI and the SEC, but even internally.” Id., at 4–5 (internal quotation marks omitted). When employees of Enron and its accounting firm, Arthur Andersen, attempted to report corporate misconduct, Congress learned, they faced retaliation, including discharge. As outside counsel advised company officials at the time, Enron’s efforts to “quiet” whistleblowers generally were not proscribed under then-existing law. Id., at 5, 10. Congress identified the lack of whistleblower protection as “a significant deficiency” in the law, for in complex securities fraud investigations, employees “are [often] the only firsthand witnesses to the fraud.” Id., at 10. Section 806 of Sarbanes-Oxley addresses this concern. Titled “Protection for Employees of Publicly Traded Companies Who Provide Evidence of Fraud,” §806 added a new provision to Title 18 of the United States Code, 18 U. S. C. §1514A, which reads in relevant part: “Civil action to protect against retaliation in fraud cases “(a) Whistleblower Protection for Employees of Publicly Traded Companies.—No company with a class of securities registered under section 12 ofthe Securities Exchange Act of 1934 ( 15 U. S. C. §78l), or that is required to file reports under section 15(d)of the Securities Exchange Act of 1934 (15 U. S. C. §78o(d)), or any officer, employee, contractor, subcontractor, or agent of such company, may discharge, demote, suspend, threaten, harass, or in any othermanner discriminate against an employee in the terms and conditions of employment because of any lawful act done by the employee— “(1) to provide information, cause information to be provided, or otherwise assist in an investigation regarding any conduct which the employee reasonably believes constitutes a violation of section 1341 [mail fraud], 1343 [wire fraud], 1344 [bank fraud], or 1348 [securities or commodities fraud], any rule or regulation of the Securities and Exchange Commission, or any provision of Federal law relating to fraud against shareholders, when the information or assistance is provided to or the investigation is conducted by [a federal agency, Congress, or supervisor] . . . .” §806, 116Stat. 802.[2] Congress has assigned whistleblower protection largely to the Department of Labor (DOL), which administers some 20 United States Code incorporated whistleblower protection provisions. See 78 Fed. Reg. 3918 (2013). The Secretary has delegated investigatory and initial adju-dicatory responsibility over claims under a number of these provisions, including §1514A, to DOL’s Occupational Safety and Health Administration (OSHA). Ibid. OSHA’s order may be appealed to an administrative law judge, and then to DOL’s Administrative Review Board (ARB). 29 CFR §§1980.104 to 1980.110 (2011). In common with other whistleblower protection provisions enforced by DOL, see 77 Fed. Reg. 3912 (2012), the ARB’s determination on a §1514A claim constitutes the agency’s final decision and is reviewable in federal court under the standards stated in the Administrative Procedure Act, 5 U. S. C. §706. If, however, the ARB does not issue a final decision within 180 days of the filing of the complaint, and the delay is not due to bad faith on the claimant’s part, the claimant may proceed to federal district court for de novo review. 18 U. S. C. §1514A(b). An employee prevailing in a proceeding under §1514A is entitled to “all relief necessary to make the employee whole,” including “reinstatement with the same seniority status that the employee would have had, but for the discrimination,” backpay with interest, and compensation for litigation costs. §1514A(c). Congress modeled §1514A on the anti-retaliation provision of the Wendell H. Ford Aviation Investment and Reform Act for the 21st Century (AIR 21), 49 U. S. C. §42121, a measure enacted two years earlier. See S. Rep., at 30 (corporate whistleblower protections “track [AIR 21’s] protections as closely as possible”). Section 1514A incorporates by cross-reference AIR 21’s administrative enforcement procedures. 18 U. S. C. §1514A(b)(2). B Petitioners Jackie Hosang Lawson and Jonathan M. Zang (plaintiffs) separately initiated proceedings under §1514A against their former employers, privately held companies that provide advisory and management services to the Fidelity family of mutual funds. The Fidelity funds are not parties to either case; as is common in the mutual fund industry, the Fidelity funds themselves have no employees. Instead, they contract with investment advisers like respondents to handle their day-to-day operations, which include making investment decisions, preparing reports for shareholders, and filing reports with the Securities and Exchange Commission (SEC). Lawson was employed by Fidelity Brokerage Services, LLC, a subsidiary of FMR Corp., which was succeeded by FMR LLC. Zang was employed by a different FMR LLC subsidiary, Fidelity Management & Research Co., and later by one of that company’s subsidiaries, FMR Co., Inc. For convenience, we refer to respondents collectively as FMR. Lawson worked for FMR for 14 years, eventually serving as a Senior Director of Finance. She alleges that, after she raised concerns about certain cost accounting methodologies, believing that they overstated expenses associated with operating the mutual funds, she suffered a seriesof adverse actions, ultimately amounting to constructive discharge. Zang was employed by FMR for eight years, most recently as a portfolio manager for several of the funds. He alleges that he was fired in retaliation for raising concerns about inaccuracies in a draft SEC reg-istration statement concerning certain Fidelity funds. Lawson and Zang separately filed administrative complaints alleging retaliation proscribed by §1514A. After expiration of the 180-day period specified in §1514A(b)(1), Lawson and Zang each filed suit in the U. S. District Court for the District of Massachusetts. FMR moved to dismiss the suits, arguing, as relevant, that neither plaintiff has a claim for relief under §1514A. FMR is privately held, and maintained that §1514A protects only employees of public companies—i.e., companies that either have “a class of securities registered under section 12 of the Securities Exchange Act of 1934,” or that are “required to file reports under section 15(d)” of that Act. §1514A(a).[3] In a joint order, the District Court rejected FMR’s interpretation of §1514A and denied the dismissal motions in both suits. 724 F. Supp. 2d 141 (Mass. 2010). On interlocutory appeal, a divided panel of the First Circuit reversed. 670 F. 3d 61 (2012). The Court of Appeals majority acknowledged that FMR is a “contractor”[4] within the meaning of §1514A(a), and thus among the actors prohibited from retaliating against “an employee” who engages in protected activity. The majority agreed with FMR, however, that “an employee” refers only to employees of public companies and does not cover a contractor’s own employees. Id., at 68–80. Judge Thompson dissented. In her view, the majority had “impose[d] an unwarranted restriction on the intentionally broad language of the Sarbanes-Oxley Act” and “bar[red] a significant class of potential securities-fraud whistleblowers from any legal protection.” Id., at 83. Several months later, the ARB issued a decision in an unrelated case, Spinner v. David Landau & Assoc., LLC, No. 10–111 etc., ALJ No. 2010–SOX–029 (May 31, 2012),[5] disagreeing with the Court of Appeals’ interpretation of §1514A. In a comprehensive opinion, the ARB explained its position that §1514A affords whistleblower protection to employees of privately held contractors that render services to public companies. Ibid.[6] We granted certiorari, 569 U. S. ___ (2013), to resolve the division of opinion on whether §1514A extends whistleblower protection to employees of privately held contractors who perform work for public companies. II A In determining the meaning of a statutory provision, “we look first to its language, giving the words used their ordinary meaning.” Moskal v. United States, 498 U. S. 103, 108 (1990) (citation and internal quotation marks omitted). As Judge Thompson observed in her dissent from the Court of Appeals’ judgment, “boiling [§1514A(a)] down to its relevant syntactic elements, it provides that ‘no . . . contractor . . . may discharge . . . an employee.’ ” 670 F. 3d, at 84 (quoting §1514A(a)). The ordinary meaning of “an employee” in this proscription is the contractor’s own employee. FMR’s interpretation of the text requires insertion of“of a public company” after “an employee.” But where Con-gress meant “an employee of a public company,” it said so: With respect to the actors governed by §1514A, the provision’s interdictions run to the officers, employees, contractors, subcontractors, and agents “of such company,” i.e., a public company. §1514A(a). Another anti-retaliation pro-vision in Sarbanes-Oxley provides: “[A] broker or dealer and persons employed by a broker or dealer who areinvolved with investment banking activities may not, directly or indirectly, retaliate against or threaten to retaliate against any securities analyst employed by that broker or dealer or its affiliates . . . .” 15 U. S. C. §78o–6(a)(1)(C) (emphasis added). In contrast, nothing in §1514A’s language confines the class of employees protected to those of a designated employer. Absent any textual qualification, we presume the operative language means what it appears to mean: A contractor may not retaliate against its own employee for engaging in protected whistle-blowing activity.[7] Section 1514A’s application to contractor employeesis confirmed when we enlarge our view from the term“an employee” to the provision as a whole. The prohib-ited retaliatory measures enumerated in §1514A(a)—discharge, demotion, suspension, threats, harassment, or dis-crimination in the terms and conditions of employment—are commonly actions an employer takes against itsown employees. Contractors are not ordinarily posi-tioned to take adverse actions against employees of the public company with whom they contract. FMR’s interpretation of §1514A, therefore, would shrink to insignificance the provision’s ban on retaliation by contractors. The dissent embraces FMR’s “narrower” construction. See post, at 2, 3, 4, 7. FMR urges that Congress included contractors in §1514A’s list of governed actors simply to prevent public companies from avoiding liability by employing contractors to effectuate retaliatory discharges. FMR describes such a contractor as an “ax-wielding specialist,” illustrated by George Clooney’s character in the movie Up in the Air.[8] Brief for Respondents 24–25 (internal quotation marks omitted). As portrayed by Clooney, an ax-wielding specialist is a contractor engaged only as the bearer of the bad news that the employee has been fired; he plays no role in deciding who to terminate. If the company employing the ax-wielder chose the recipients of the bad tidings for retaliatory reasons, the §1514A claim would properly be directed at the company. Hiring the ax-wielder would not insulate the company from liability. Moreover, we see no indication that retaliatory ax-wielding specialists are the real-world problem that prompted Congress to add contractors to §1514A.[9] Moving further through §1514A to the protected activity described in subsection (a)(1), we find further reason to believe that Congress presumed an employer-employee relationship between the retaliator and the whistleblower. Employees gain protection for furnishing information to a federal agency, Congress, or “a person with supervisory authority over the employee (or such other person working for the employer who has the authority to investigate, discover, or terminate misconduct).” §1514A(a)(1) (emphasis added). And under §1514A(a)(2), employees are protected from retaliation for assisting “in a proceeding filed or about to be filed (with any knowledge of the employer) relating to an alleged violation” of any of the enumerated fraud provisions, securities regulations, or other federal law relating to shareholder fraud. §1514A(a)(2) (emphasis added). The reference to employer knowledge is an additional indicator of Congress’ expectation that the retaliator typically will be the employee’s employer, not another entity less likely to know of whistleblower complaints filed or about to be filed. Section 1514A’s enforcement procedures and remedies similarly contemplate that the whistleblower is an employee of the retaliator. As earlier noted, see supra, at 6, §1514A(b)(2)(A) provides that a claim under §1514A “shall be governed under the rules and procedures set forth in section 42121(b) of title 49,” i.e., AIR 21’s anti-retaliation provision. Throughout §42121(b), the respondent is referred to as “the employer.” See 49 U. S. C. §42121(b)(2)(B)(ii) (The Secretary shall not conduct an investigation into a retaliation claim “if the employer demonstrates,by clear and convincing evidence, that the employer would have taken the same unfavorable personnel action inthe absence of that behavior.”); §42121(b)(2)(B)(iv)(“Relief may not be ordered . . . if the employer demonstrates by clear and convincing evidence that the employer would have taken the same unfavorable personnel action in the absence of that behavior.”). Regarding remedies, §1514A(c)(2) states that a successful claimant shall be entitled to “reinstatement with the same seniority status that the employee would have had, but for the discrimination,” as well as “the amount of back pay, with interest.” As the Solicitor General, for the United States as amicus curiae, observed, “It is difficult, if not impossible, to see how a contractor or subcontractor could provide those remedies to an employee of a public company.” Brief for United States as Amicus Curiae 15. The most sensible reading of §1514A’s numerous references to an employer-employee relationship between the respondent and the claimant is that the provision’s protections run between contractors and their own employees. Remarkably, the dissent attributes to Congress a strange design. Under the dissent’s “narrower” construction, post, at 2, 3, 4, 7, a public company’s contractor may not retaliate against a public company’s employees, academic here because the public company has no employees. According to the dissent, this coverage is necessary to prevent “a gaping hole” that would allow public companies to “evade §1514A simply by hiring a contractor to engage in the very retaliatory acts that an officer or employee could not.” Post, at 10. This cannot be right—even if Congress had omitted any reference to contractors, subcontractors, or agents in §1514A, the remaining language surely would prohibit a public company from directing someone else to engage in retaliatory conduct against the public company’s employees; hiring an ax-wielder to announce an employee’s demotion does not change the fact that the public company is the entity commanding the demotion. Under the dissent’s reading of §1514A, the inclusion of contractors as covered employers does no more than make the contractor secondarily liable for complying with such marching orders—hardly a hole at all.[10] There would be a huge hole, on the other hand, were the dissent’s view of §1514A’s reach to prevail: Contractors’ employees would be disarmed; they would be vulnerableto retaliation by their employers for blowing the whistleon a scheme to defraud the public company’s investors, even a scheme engineered entirely by the contractor. Not only would mutual fund advisers and managers escape §1514A’s control. Legions of accountants and lawyers would be denied §1514A’s protections. See infra, at 19–22. Instead of indulging in fanciful visions of whistleblowing babysitters and the like, post, at 1–2, 6, 12–13, 20, the dissent might pause to consider whether a Congress, prompted by the Enron debacle, would exclude from whistleblower protection countless professionals equipped to bring fraud on investors to a halt. B We turn next to two textual arguments made by FMR. First, FMR urges that “an employee” must be read to refer exclusively to public company employees to avoid the absurd result of extending protection to the personal employees of company officers and employees, e.g., their housekeepers or gardeners. See Brief for Respondents 19–20; post, at 1–2, 6, 12–13, 20. Plaintiffs and the Solicitor General do not defend §1514A’s application to personal employees. They argue, instead, that the prohibition against an “officer” or “employee” retaliating against “an employee” may be read as imposing personal liability only on officers and employees who retaliate against other public company employees. Brief for Petitioners 12; Brief for United States as Amicus Curiae 16.[11] FMR calls this reading “bizarre,” for it would ascribe to the words “an employee” in §1514A(a) “one meaning if the respondent is an ‘officer’ and a different meaning if the respondent is a ‘contractor.’ ” Brief for Respondents 20–21. We agree with FMR that plaintiffs and the Solicitor General offer an interpretation at odds with the text Congress enacted. If, as we hold, “an employee” includes employees of contractors, then grammatically, the term also includes employees of public company officers and employees. Nothing suggests Congress’ attention was drawn to the curiosity its drafting produced. The issue, however, is likely more theoretical than real. Few housekeepers or gardeners, we suspect, are likely to come upon and comprehend evidence of their employer’s complicity in fraud. In any event, FMR’s point is outweighed by the compelling arguments opposing FMR’s contention that “an employee” refers simply and only to public company employees. See supra, at 9–14. See also infra, at 23–24 (limiting principles may serve as check against overbroad applications). Second, FMR argues that the statutory headings support the exclusion of contractor employees from §1514A’s protections. Although §1514A’s own heading is broad (“Civil action to protect against retaliation in fraud cases”), subsection (a) is captioned “Whistleblower Protection for Employees of Publicly Traded Companies.” Similarly, the relevant public law section, §806 of Sarbanes-Oxley, is captioned “Protection for Employees of Publicly Traded Companies Who Provide Evidence of Fraud.” 116 Stat. 802. The Court of Appeals described the latter two headings as “explicit guides” limiting protection under §1514A to employees of public companies. 670 F. 3d, at 69. This Court has placed less weight on captions. In Trainmen v. Baltimore & Ohio R. Co., 331 U. S. 519 (1947) , we explained that where, as here, “the [statutory] text is complicated and prolific, headings and titles can do no more than indicate the provisions in a most general manner.” Id., at 528. The under-inclusiveness of the two headings relied on by the Court of Appeals is apparent. The provision indisputably extends protection to employees of companies that file reports with the SEC pursuant to §15(d) of the 1934 Act, even when such companies are not “publicly traded.” And the activity protected under §1514A is not limited to “provid[ing] evidence of fraud”; it also includes reporting violations of SEC rules or regulations. §1514A(a)(1). As in Trainmen, the headings here are “but a short-hand reference to the general subject matter” of the provision, “not meant to take the place of the detailed provisions of the text.” 331 U. S., at 528. Section 1514A is attended by numerous indicators that the statute’s prohibitions govern the relationship between a contractor and its own employees; we do not read the headings to “undo or limit” those signals. Id., at 529.[12] III A Our textual analysis of §1514A fits the provision’s purpose. It is common ground that Congress installed whistleblower protection in the Sarbanes-Oxley Act as one means to ward off another Enron debacle. S. Rep., at 2–11. And, as the ARB observed in Spinner, “Congress plainly recognized that outside professionals—ac-countants, law firms, contractors, agents, and the like—were complicit in, if not integral to, the shareholder fraud and subsequent cover-up [Enron] officers . . . perpetrated.” ALJ No. 2010–SOX–029, pp. 12–13. Indeed, the Senate Report demonstrates that Congress was as focused on the role of Enron’s outside contractors in facilitating the fraud as it was on the actions of Enron’s own officers. See, e.g., S. Rep., at 3 (fraud “occurred with extensive participation and structuring advice from Arthur Andersen . . . which was simultaneously serving as both consultant and independent auditor for Enron” (internal quotation marks and brackets omitted)); id., at 4 (“professionals from accounting firms, law firms and business consulting firms, who were paid millions to advise Enron on these practices, assured others that Enron was a solid investment”); id., at 4–5 (team of Andersen employees were tasked with destroying “physical evidence and documents” relating to Enron’s fraud); id., at 5 (“Enron and Andersen were taking advantage of a system that allowed them to behave in an apparently fraudulent manner”); id., at 11 (Enron’s fraud partly attributable to “the well-paid professionals who helped create, carry out, and cover up the complicated corporate ruse when they should have been raising concerns”); id., at 20–21 (“Enron’s accountants and lawyers brought all their skills and knowledge to bear in assisting the fraud to succeed and then in covering it up.”). Also clear from the legislative record is Congress’ understanding that outside professionals bear significant responsibility for reporting fraud by the public companies with whom they contract, and that fear of retaliation was the primary deterrent to such reporting by the employ-ees of Enron’s contractors. Congressional investigators discovered ample evidence of contractors demoting or dis-charging employees they have engaged who jeopardized the contractor’s business relationship with Enron by objecting to Enron’s financial practices. See, e.g., Oppel, Merrill Replaced Research Analyst Who Upset Enron, N. Y. Times, July 30, 2002, p. A1 (“In the summer of 1998, when it was eager to win more investment banking business from Enron, Merrill Lynch replaced a research analyst who had angered Enron executives by rating the company’s stock ‘neutral’ with an analyst who soon upgraded the rating, according to Congressional investi-gators.”); Yost, Andersen Whistleblower Was Removed, Associated Press (Apr. 3, 2002) (Congressional investigation reveals that Andersen removed one of its partners from its Enron team after Enron officials expressed unhappiness with the partner’s questioning of certain accounting practices); Oppel, The Man Who Paid the Price for Sizing up Enron, N. Y. Times, Mar. 27, 2002, p. C1 (“Enron executives pressed UBS PaineWebber to take action against a broker who advised some Enron employees to sell their shares in August and was fired by the brokerage firm within hours of the complaint, accordingto e-mail messages released today by Congressionalinvestigators.”). In the same vein, two of the four examples of whistleblower retaliation recounted in the Senate Report involved outside professionals retaliated against by their own employers. S. Rep., at 5 (on Andersen and UBS Paine-Webber employees); see also id., at 4–5 (Andersen employees who “attempted to report or ‘blow the whistle’ on [Enron’s] fraud . . . were discouraged at nearly every turn”). Emphasizing the importance of outside professionals as “gatekeepers who detect and deter fraud,” the Senate Report concludes: “Congress must reconsider the incentive system that has been set up that encourages accountants and lawyers who come across fraud in their work to remain silent.” Id., at 20–21. From this legislative history, one can safely conclude that Congress enacted §1514A aiming to encourage whistleblowing by contractor employees who suspect fraud involving the public companies with whom they work.[13] FMR argues that Congress addressed its concerns about the role of outside accountants and lawyers in facilitating Enron’s wrongdoing, not in §1514A, but exclusively in other provisions of Sarbanes-Oxley “directly regulat[ing] accountants and lawyers.” Brief for Respondents 40. In particular, FMR points to sections of the Act requiring accountants and lawyers for public companies to investigate and report misconduct, or risk being banned from further practice before the SEC. Id., at 41 (citing 15 U. S. C. §§7215(c)(4), 7245). These requirements, however, indicate why Congress would have wanted to extend §1514A’s coverage to the many lawyers and accountants who perform outside work for public companies. Although lawyers and accountants are subject to extensive regulations and sanctions throughout Sarbanes-Oxley, no provision of the Act other than §1514A affords them protection from retaliation by their employers for complying with the Act’s reporting requirements.[14] In short, we cannot countenance the position advanced by FMR and the dissent, see post, at 14–16, that Congress intended to leave these professionals vulnerable to discharge or other retaliatory action for complying with the law. B Our reading of §1514A avoids insulating the entire mutual fund industry from §1514A, as FMR’s and the dissent’s “narrower construction” would do. As companies “required to file reports under section 15(d) of the Securities Exchange Act of 1934,” 18 U. S. C. §1514A(a), mutual funds unquestionably are governed by §1514A. Because mutual funds figure prominently among such report-filing companies, Congress presumably had them in mind when it added to “publicly traded companies” the discrete category of companies “required to file reports under section 15(d).” Virtually all mutual funds are structured so that they have no employees of their own; they are managed, instead, by independent investment advisers. See S. Rep. No. 91–184, p. 5 (1969) (accompanying the 1970 amendments to the Investment Company Act of 1940). The United States investment advising industry manages $14.7 trillion on behalf of nearly 94 million investors. See 2013 Investment Company Fact Book 7 (53d ed.), available at http://www.icifactbook.org/pdf/2013_factbook.pdf (as visited Feb. 20, 2014, and available in Clerk of Court’s case file). These investment advisers, under our reading of §1514A, are contractors prohibited from retaliating against their own employees for engaging in whistleblowing activity. This construction protects the “insiders [who] are the only firsthand witnesses to the [shareholder] fraud.” S. Rep., at 10. Under FMR’s and the dissent’s reading, in contrast, §1514A has no application to mutual funds, for all of the potential whistleblowers are employed by the privately held investment management companies, not by the mutual funds themselves. See Brief for Respondents 45 (describing this glaring gap as “merely a consequence of the corporate structure” of mutual funds). The Court of Appeals found exclusion of the mutual fund industry from §1514A tenable because mutual funds and their investment advisers are separately regulated under the Investment Company Act of 1940, 15 U. S. C. §80a–1 et seq., the Investment Advisers Act of 1940, 15 U. S. C. §80b–1 et seq., and elsewhere in Sarbanes-Oxley. 670 F. 3d, at 72–73. See also post, at 16–17, n. 10. But this separate regulation does not remove the problem, for nowhere else in these legislative measures are investment management employees afforded whistleblower protection. Section 1514A alone shields them from retaliation for bringing fraud to light. Indeed, affording whistleblower protection to mutual fund investment advisers is crucial to Sarbanes-Oxley’s endeavor to “protect investors by improving the accuracy and reliability of corporate disclosures made pursuant to the securities laws.” 116Stat. 745. As plaintiffs observe, these disclosures are written, not by anyone at the mutual funds themselves, but by employees of the investment advisers. “Under FMR’s [and the dissent’s] proposed in-terpretation of section 1514A, FMR could dismiss any FMR employee who disclosed to the directors of or lawyers for the Fidelity funds that there were material falsehoods in the documents being filed by FMR with the SEC in the name of those funds.” Reply Brief 13. It is implausible that Congress intended to leave such an employee remediless. See id., at 14. C Unable credibly to contest the glaring under-inclusiveness of the “narrower reading” FMR urges, the dissent emphasizes instead FMR’s claim that the reading of §1514A we adopt is all too inclusive. See post, at 1–2, 6, 12–13, 20–21. FMR’s amici also press this point, observing that the activity protected under §1514A(a)(1) encompasses reporting not only securities fraud ( 18 U. S. C. §1348), but also mail, wire, and bank fraud (§§1341, 1343, 1344). Including contractor employees in the protected class, they therefore assert, could “cas[t] a wide net over employees who have no exposure to investor-related activities and thus could not possibly assist in detecting investor fraud.” Brief for Chamber of Commerce of the United States of America as Amicus Curiae 3. See also Brief for Securities Industry and Financial Markets Association as Amicus Curiae 7–16. There is scant evidence, however, that these floodgate-opening concerns are more than hypothetical. DOL’s regulations have interpreted §1514A as protecting contractor employees for almost a decade.[15] See 69 Fed. Reg. 52105–52106 (2004). Yet no “narrower construction” advocate has identified even a single case in which the employee of a private contractor has asserted a §1514A claim based on allegations unrelated to shareholder fraud. FMR’s parade of horribles rests solely on Lockheed Martin Corp. v. ARB, 717 F. 3d 1121 (CA10 2013), a case involving mail and wire fraud claims asserted by an employee of a public company—i.e., claims in no way affected by today’s decision. The dissent’s fears that household employees and others, on learning of today’s decision, will be prompted to pursue retaliation claims, post, at 13, and that OSHA will find them meritorious under §1514A, seem to us unwarranted. If we are wrong, however, Congress can easily fix the problem by amending §1514A explicitly to remove personal employees of public company officers and employees from the provision’s reach. But it would thwart Congress’ dominant aim if contractors were taken off the hook for retaliating against their whistleblowing employees, just to avoid the unlikely prospect that babysitters, nannies, gardeners, and the like will flood OSHA with §1514A complaints. Plaintiffs and the Solicitor General observe that overbreadth problems may be resolved by various limit-ing principles. They point specifically to the word“contractor.” Plaintiffs note that in “common parlance,” “contractor” does not extend to every fleeting business relationship. Instead, the word “refers to a party whose performance of a contract will take place over a significant period of time.” Reply Brief 16. See also Fleszar v. United States Dept. of Labor, 598 F. 3d 912, 915 (CA7 2010) (“Nothing in §1514A implies that, if [a privately held business] buys a box of rubber bands from Wal-Mart, a company with traded securities, the [business] becomes covered by §1514A.”). The Solicitor General further maintains that §1514A protects contractor employees only to the extent that their whistleblowing relates to “the contractor . . . fulfilling its role as a contractor for the public company, not the contractor in some other capacity.” Tr. of Oral Arg. 18–19 (Government counsel). See also id., at 23 (“[I]t has to be a person who is in a position to detect and report the types of fraud and securities violations that are included in the statute. . . . [W]e think that ‘the contractor of such com-pany’ refers to the contractor in that role, working for the public company.’ ”). Finally, the Solicitor General suggests that we need not determine the bounds of §1514A today, because plaintiffs seek only a “mainstream application” of the provision’s protections. Id., at 20 (Government counsel). We agree. Plaintiffs’ allegations fall squarely within Congress’ aim in enacting §1514A. Lawson alleges that she was constructively discharged for reporting accounting practices that overstated expenses associated with managing certain Fidelity mutual funds. This alleged fraud directly implicates the funds’ shareholders: “By inflating its expenses, and thus understating its profits, [FMR] could potentially increase the fees it would earn from the mutual funds, fees ultimately paid by the shareholders of those funds.” Brief for Petitioners 3. Zang alleges that he was fired for expressing concerns about inaccuracies in a draft registration statement FMR prepared for the SEC on behalf of certain Fidelity funds. The potential impact on shareholders of false or misleading registration statements needs no elaboration. If Lawson and Zang’s allegations prove true, these plaintiffs would indeed be “firsthand witnesses to [the shareholder] fraud” Congress anticipated §1514A would protect. S. Rep., at 10. D FMR urges that legislative events subsequent toSarbanes-Oxley’s enactment show that Congress did not intend to extend §1514A’s protections to contractor employees.[16] In particular, FMR calls our attention to the 2010 Dodd-Frank Wall Street Reform and Consumer Protection Act, 124Stat. 1376 (Dodd-Frank). Dodd-Frank amended §1514A(a) to read: “No company with a class of securities registered under section 12 of the Securities Exchange Act of 1934 (15 U. S. C. 78l), or that is required to file reports under [section 12] of the [1934 Act] (15 U. S. C. 78o(d)) including any subsidiary or affiliate whose financial information is included in the consolidated financial statements of such company, or nationally recognized statistical rating organization (as defined in section 3(a) of the [1934 Act] (15 U. S. C. 78c), or any officer, employee, contractor, subcontractor, or agent of such company or nationally recognized statistical rating organization, may discharge, demote, suspend, threaten, harass, or in any other manner discriminateagainst an employee in the terms and conditions of employment because of any [protected activity].” 18 U. S. C. §1514A(a) (2012 ed.) (emphasis added; footnote omitted.) The amended provision extends §1514A’s protection to employees of public company subsidiaries and nationally recognized statistical rating organizations (NRSROs). FMR asserts that Congress’ decision to add NRSROs to §1514A shows that the provision did not previously cover contractor employees: “If [§1514A] already covered every private company contracting with a public company, there would have been no need for Congress to extend [§1514A] to certain private companies.” Brief for Respondents 35–36. This argument fails at the starting gate, for FMR concedes that not all NRSROs are privately held, and not all NRSROs contract with public companies. Id., at 36. We see nothing useful to our inquiry in Congress’ decision to amend §1514A to include public company sub-sidiaries and NRSROs. More telling, at the time of the Dodd-Frank amendments, DOL regulations provided that §1514A protects contractor employees. See 29 CFR §1980.101 (2009). Congress included in its alterations no language gainsaying that protection. As Judge Thompson’s dissent from the First Circuit’s judgment observes, “Congress had a miles-wide opening to nip [DOL’s] regulation in the bud if it had wished to do so. It did not.” 670 F. 3d, at 88. Dodd-Frank also establishes a corporate whistleblowing reward program, accompanied by a new provision pro-hibiting any employer from retaliating against “a whistleblower” for providing information to the SEC, participating in an SEC proceeding, or making disclosures requiredor protected under Sarbanes-Oxley and certain other securities laws. 15 U. S. C. §78u–6(a)(6), (b)(1), (h). FMR urges that, as this provision covers employees of all companies, public or private, “[t]here is no justification” for reading §1514A to cover employees of contractors: “Any ‘gap’ that might, arguendo, have existed for employees of private entities between 2002 and 2010 has now been closed.” Brief for Respondents 44.[17] FMR, we note, somewhat overstates Dodd-Frank’s cov-erage. Section 1514A’s protections include employeeswho provide information to any “person with supervisory authority over the employee.” §1514A(a)(1)(C). Dodd-Frank’s whistleblower provision, however, focuses primarily on reporting to federal authorities. See Brief for United States as Amicus Curiae 30 (“[I]f employees of contrac-tors of public companies are not protected under Section 1514A, they are not protected for making internal complaints under . . . the Dodd-Frank Act.”). In any event, our task is not to determine whether including contractor employees in the class protected by §1514A remains necessary in 2014. It is, instead, to determine whether Congress afforded protection to contractor employees when it enacted §1514A in 2002. If anything relevant to our inquiry can be gleaned from Dodd-Frank,it is that Congress apparently does not share FMR’sconcerns about extending protection comprehensivelyto corporate whistleblowers.[18] IV We end by returning to AIR 21’s whistleblower protection provision, 49 U. S. C. §42121, enacted two years before Sarbanes-Oxley. Congress designed §1514A to “track . . . as closely as possible” the protections afforded by §42121. S. Rep., at 30. To this end, §1514A incorporates by cross-reference §42121’s administrative enforcement regime, see 18 U. S. C. §1514A(b)(2), and contains parallel statutory text. Compare §1514A(a) (“No [public] company . . . or any officer, employee, contractor, subcontractor, or agent of such company, may discharge, demote, suspend, threaten, harass, or in any other manner discriminate against an employee in the terms and conditions of employment” for engaging in protected activity) with 49 U. S. C. §42121(a) (“No air carrier or contractor or subcontractor of an air carrier may discharge an employee or otherwise discriminate against an employee with respect to compensation, terms, conditions, or privileges of employment” for engaging in protected activity).[19] Section 42121 has been read to protect employees of contractors covered by the provision. The ARB has consistently construed AIR 21 to cover contractor employees. E.g., Evans v. Miami Valley Hospital, ARB No. 07–118 etc., ALJ No. 2006–AIR–022, pp. 9–11 (June 30, 2009); Peck v. Safe Air Int’l, Inc., ARB No. 02–028, ALJ No. 2001–AIR–3, p. 13 (Jan. 30, 2004).[20] And DOL’s regulations adopting this interpretation of §42121 date back to April 1, 2002, before §1514A was enacted. 67 Fed. Reg. 15454, 15457–15458 (2002). The Senate Report for AIR 21 supports this reading, explaining that the Act “provide[s] employees of airlines, and employees of airline contractors and subcontractors, with statutory whistleblower protection.” S. Rep. No. 105–278, p. 22 (1998).[21] The Court of Appeals recognized that Congress modeled §1514A on §42121, and that §42121 has been understood to protect contractor employees. 670 F. 3d, at 73–74. It nonetheless declined to interpret §1514A the same way, because, in its view, “important differences” separate the two provisions. First, unlike §1514A, §42121 contains a definition of “contractor”: “a company that performs safety-sensitive functions by contract for an air carrier.” 49 U. S. C. §42121(e). Second, unlike §1514A, §42121 does not include “officers” or “employees” among governed actors. 670 F. 3d, at 74. These distinctions, the Court of Appeals reasoned, render §1514A less amenable to an inclusive construction of the protected class. Ibid.[22] We do not find these textual differences overwhelming. True, Congress strayed from §42121’s pattern in failing to define “contractor” for purposes of §1514A, and in adding “officers” and “employees” to §1514A’s list of governed actors. And we agree that §1514A covers a far wider range than §42121 does. But in our view, neither difference warrants the determination that §1514A omits employees of contractors while §42121 includes them. The provisions’ parallel text and purposes counsel in favor of interpreting the two provisions consistently. And we have already canvassed the many reasons why §1514A is most sensibly read to protect employees of contractors. See supra, at 9–22. * * * For the reasons stated, we hold that 18 U. S. C. §1514A whistleblower protection extends to employees of contractors and subcontractors. The judgment of the U. S. Court of Appeals for the First Circuit is therefore reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Title VIII of the Act, which contains the whistleblower protection provision at issue in this case, was authored by Senators Leahy and Grassley and originally constituted a discrete bill, S. 2010. We thus look to the Senate Report for S. 2010, S. Rep. No. 107–146, as the Senate Report relevant here. See 148 Cong. Rec. S7418 (daily ed. July 26, 2002) (statement of Sen. Leahy) (“unanimous consent” to “includ[e] in the as part of the official legislative history” of Sarbanes-Oxley that Title VIII’s “terms track almost exactly the provisions of S. 2010, introduced by Senator Leahy and reported unanimously from the Committee on the Judiciary”). 2 As discussed , at 24–26, Congress amended §1514A in 2010 to extend whistleblower coverage to employees of public companies’ subsidiaries and nationally recognized statistical ratings organizations. . Plaintiffs do not fall in either category and, in any event, their claims are governed by the prior version of §1514A. Unless otherwise noted, all citations to §1514A are to the original text in the 2006 edition of the United States Code. 3 Here, as just noted, the public company has no employees. See, at 2. 4 As §1514A treats contractors and subcontractors identically, we generally refer simply to “contractors” without distinguishing between the two. 5 The whistleblower in was an employee of an accounting firm that provided auditing, consulting, and Sarbanes-Oxley compliance services to a public company. 6 The dissent maintains that the ARB’s interpretation of §1514A is not entitled to deference because, “if any agency has the authority to resolve ambiguities in §1514A with the force of law, it is the SEC, not the Department of Labor.” at 18. Because we agree with the ARB’s conclusion that §1514A affords protection to a contractor’s employees, we need not decide what weight that conclusion should carry. We note, however, that the SEC apparently does not share the dissent’s view that it, rather than DOL, has interpretive authority over §1514A. To the contrary, the SEC is a signatory to the Government’s brief in this case, which takes the position that Congress has charged the Secretary of Labor with interpreting §1514A. Brief for United States as 9–11, 31–34. That view is hardly surprising given the lead role played by DOL in administering whistleblower statutes. See at 5. The dissent observes that the SEC “hasnot issued a regulation applying §1514A whistleblower protection to employees of public company contractors,” , at 18, but omits to inform that the SEC has not promulgated regulations interpreting §1514A, consistent with its view that Congress delegated that responsibility to DOL. 7 We need not decide in this case whether §1514A also prohibits a contractor from retaliating against an employee of one of the other actors governed by the provision. 8 This hypothetical originates in a Seventh Circuit opinion, v., 598 F. 3d 912, 915 (2010), and is mentioned in a footnote in the First Circuit’s opinion in this case, 670 F. 3d 61, 69, n. 11 (2012). 9 When asked during oral argument for an example of actual circumstances in which a contractor would have employment decisionmak-ing authority over public company employees, FMR’s counsel cited v., No. 05–139 etc., ALJ No. 2004–SOX–056 (Feb. 27, 2009). Tr. of Oral Arg. 33. That case involved a bankrupt public company that hired a private company to handle its dissolution. The ARB found the private company liable under §1514A because it acted as a “contractor, subcontractor, ” of the public company in discharging the claimant. ALJ No. 2004–SOX–056, at 10 (emphasis added). Neither FMR nor its have pointed us to any actual situation in which a public company employee would be vulnerable to retaliatory conduct by a contractor not already covered as an “agent” under §1514A. Notably, even in v., 644 F. 3d 809 (CA9 2011), the case cited by the dissent for the proposition that contractors may possess “managerial authority” over public company employees, , at 10, the alleged retaliation was by the public company itself. 10 The dissent suggests that we “fai[l] to recognize” that its construction also makes contractors primarily liable for retaliating of their own volition against employees of public companies. , at 10, n. 6. As explained at 11–12, n. 9, however, FMR and its supporters have identified not even one real-world instance of a public company employee asserting a §1514A claim alleging retaliatory conduct by a contractor. Again, no “gaping hole,” practically no hole at all. 11 The ARB endorsed this view in v., No. 10–111 etc., ALJ No. 2010–SOX–029, p. 8 (May 31, 2012). We have no occasion to determine whether the ARB would be entitled to deference in this regard, for, as explained in text, we find that the statutory text unambiguously affords protection to personal employees of public company officers and employees. §1514A(a). 12 AIR 21’s anti-retaliation provision, on which §1514A is based, includes a similarly composed heading, “Discrimination against airline employees.” . Nevertheless, that provision has been read to cover employees of companies rendering contract services to airlines. See, at 27–29. 13 FMR urges that the Senate Report’s references to “employees of publicly traded companies” demonstrate that Congress wanted to limit whistleblower protection to such employees. Brief for Respondents 30–31. This argument fails for the same reason that FMR’s reliance on the statutory section headings fails: “employees of publicly traded companies” must be understood as shorthand not designed to capture every employee covered by §1514A. See at 15–16. Senator Sarbanes’ statement, cited in the concurring opinion, , at 2, is similarly imprecise. The Act indisputably covers private accounting firms and law firms that provide services to public companies. See, ., 15 U. S. C. §§7215, 7245. Indeed, Senator Sarbanes acknowledged this point in his very next sentence. See 148 Cong. Rec. 14440 (2002) (remarks of Sen. Sarbanes) (“This legislation prohibits accounting firms from providing certain specified consulting services if they are also the auditors of the company.”). 14 The dissent suggests that the Public Company Accounting Oversight Board’s and the SEC’s authority to sanction unprofessional conduct by accountants and lawyers, respectively, “could well provide” a disincentive to retaliate against other accountants and lawyers. See , at 15. The possibility of such sanctions, however, is cold comfort to the accountant or lawyer who loses her job in retaliation for her efforts to comply with the Act’s requirements if, as the dissent would have it, §1514A does not enable her to seek reinstatement or backpay. 15 Although the dissent suggests that the ARB had not provided “definitive clarification” on the issue prior to , at 14, the ARB “repeatedly interpreted [§1514A] as affording whistleblower protection to employees of [private] contractors” before . See , No. 10–111 etc., ALJ No. 2010–SOX–029, p. 5 (citing prior decisions). 16 We can easily dismiss FMR’s invocation of a failed bill from 2004, the Mutual Fund Reform Act, S. 2059, 108th Cong., 2d Sess., §116(b), which would have amended §1514A explicitly to cover employees of investment advisers and affiliates. Brief for Respondents 34–35. “[F]ailed legislative proposals are a particularly dangerous ground on which to rest an interpretation of a prior statute.” v., (internal quotation marks omitted). Where, as here, the proposed amendment amounted to six lines in a 51-page bill that died without any committee action, its failure is scarcely relevant to Congress’ intentions regarding a different bill enacted two years earlier. 17 FMR acknowledges that plaintiffs’ claims could have proceeded under Dodd-Frank, but for the date of enactment. Brief for Respondents 43. 18 Section 1107 of the Act is of similar breadth, declaring it a criminal offense to “tak[e] any action harmful to any person, including interference with the lawful employment or livelihood of any person, for providing to a law enforcement officer any truthful information relating to the commission or possible commission of any Federal offense.” . 19 For other provisions borrowing from AIR 21, see , governing rail carriers, which incorporates AIR 21’s enforcement procedures, and §31105, governing motor carriers, which incorporates AIR 21’s proof burdens. 20 The ARB has also interpreted similarly worded whistleblower protection provisions in the Pipeline Safety Improvement Act of 2002, , and the Energy Reorganization Act of 1974, , as protecting employees of contractors. See v.., ARB No. 07–112, ALJ No. 2006–PSI–001 etc., p. 2 (June 25, 2009); v.., ARB No. 10–013, ALJ No. 2006–ERA–031, pp. 8–9 (Mar. 28, 2012). 21 FMR protests that there is no court of appeals precedent on point, Brief for Respondents 24, n. 6, but the courts of appeals are not, of course, the only lodestar for determining whether a proposition of law is plainly established. 22 The dissent suggests the provisions’ headings are also distinguishable because §42121’s title—“Protection of employees providing air safety information”—“comfortably encompasses the employees of contractors.” , at 8. The dissent omits, however, the subsection heading directly following the title: “Discrimination against airline employees.” §42121(a).

572.US.118

Petitioner Lexmark sells the only style of toner cartridges that work with the company’s laser printers, but “remanufacturers” acquire and refurbish used Lexmark cartridges to sell in competition with Lexmark’s own new and refurbished ones. Lexmark’s “Prebate” program gives customers a discount on new cartridges if they agree to return empty cartridges to the company. Each Prebate cartridge has a microchip that disables the empty cartridge unless Lexmark replaces the chip. Respondent Static Control, a maker and seller of components for the remanufacture of Lexmark cartridges, developed a microchip that mimicked Lexmark’s. Lexmark sued for copyright infringement, but Static Control counterclaimed, alleging that Lexmark engaged in false or misleading advertising in violation of §43(a) of the Lanham Act, 15 U. S. C. §1125(a), and that its misrepresentations had caused Static Control lost sales and damage to its business reputation. The District Court held that Static Control lacked “prudential standing” to bring the Lanham Act claim, applying a multifactor balancing test the court attributed to Associated Gen. Contractors of Cal., Inc. v. Carpenters, 459 U.S. 519. In reversing, the Sixth Circuit relied on the Second Circuit’s “reasonable interest” test. Held: Static Control has adequately pleaded the elements of a Lanham Act cause of action for false advertising. Pp. 6–22. (a) The question here is whether Static Control falls within the class of plaintiffs that Congress authorized to sue under §1125(a). To decide that question, this Court must determine the provision’s meaning, using traditional principles of statutory interpretation. It is misleading to label this a “prudential standing” question. Lexmark bases its “prudential standing” arguments on Associated General Contractors, but that case rested on statutory considerations: The Court sought to “ascertain,” as a statutory-interpretation matter, the “scope of the private remedy created by” Congress in §4 of the Clayton Act, and the “class of persons who [could] maintain a private damages action under” that legislatively conferred cause of action, 459 U. S., at 529, 532. And while this Court may have placed the “zone of interests” test that Static Control relies on under the “prudential” rubric in the past, see, e.g., Elk Grove Unified School Dist. v. Newdow, 542 U.S. 1, 12, it does not belong there any more than Associated General Contractors does. Rather, whether a plaintiff comes within the zone of interests requires the Court to determine, using traditional statutory-interpretation tools, whether a legislatively conferred cause of action encompasses a particular plaintiff’s claim. See, e. g., Steel Co. v. Citizens for Better Environment, 523 U.S. 83, 97, and n. 2. Pp. 6–9. (b) The §1125(a) cause of action extends to plaintiffs who fall within the zone of interests protected by that statute and whose injury was proximately caused by a violation of that statute. Pp. 10–18. (1) A statutory cause of action is presumed to extend only to plaintiffs whose interests “fall within the zone of interests protected by the law invoked.” Allen v. Wright, 468 U.S. 737, 751. “[T]he breadth of [that] zone . . . varies according to the provisions of law at issue.” Bennett v. Spear, 520 U.S. 154, 163. The Lanham Act includes a detailed statement of its purposes, including, as relevant here, “protect[ing] persons engaged in [commerce within the control of Congress] against unfair competition,” 15 U. S. C. §1127; and “unfair competition” was understood at common law to be concerned with injuries to business reputation and present and future sales. Thus, to come within the zone of interests in a §1125(a) false-advertising suit, a plaintiff must allege an injury to a commercial interest in reputation or sales. Pp. 10–13. (2) A statutory cause of action is also presumed to be limited to plaintiffs whose injuries are proximately caused by violations of the statute. See, e.g., Holmes v. Securities Investor Protection Corporation, 503 U.S. 258, 268–270. This requirement generally bars suits for alleged harm that is “too remote” from the defendant’s unlawful conduct, such as when the harm is purely derivative of “misfortunes visited upon a third person by the defendant’s acts.” Id., at 268–269. In a sense, all commercial injuries from false advertising are derivative of those suffered by consumers deceived by the advertising. But since the Lanham Act authorizes suit only for commercial injuries, the intervening consumer-deception step is not fatal to the proximate-cause showing the statute requires. Cf. Bridge v. Phoenix Bond & Indemnity Co., 553 U.S. 639, 656. Thus, a plaintiff suing under §1125(a) ordinarily must show that its economic or reputational injury flows directly from the deception wrought by the defendant’s advertising; and that occurs when deception of consumers causes them to withhold trade from the plaintiff. Pp. 13–15. (3) Direct application of the zone-of-interests test and the proximate-cause requirement supplies the relevant limits on who may sue under §1125(a). These principles provide better guidance than the multifactor balancing test urged by Lexmark, the direct-competitor test, or the reasonable-interest test applied by the Sixth Circuit. Pp. 15–18. (c) Under these principles, Static Control comes within the class of plaintiffs authorized to sue under §1125(a). Its alleged injuries—lost sales and damage to its business reputation—fall within the zone of interests protected by the Act, and Static Control sufficiently alleged that its injuries were proximately caused by Lexmark’s misrepresentations. Pp. 18–22. 697 F.3d 387, affirmed. Scalia, J., delivered the opinion for a unanimous Court.

This case requires us to decide whether respondent, Static Control Components, Inc., may sue petitioner, Lex- mark International, Inc., for false advertising under the Lanham Act, 15 U. S. C. §1125(a). I. Background Lexmark manufactures and sells laser printers. It also sells toner cartridges for those printers (toner being the powdery ink that laser printers use to create images on paper). Lexmark designs its printers to work only with its own style of cartridges, and it therefore dominates the market for cartridges compatible with its printers. That market, however, is not devoid of competitors. Other businesses, called “remanufacturers,” acquire used Lex- mark toner cartridges, refurbish them, and sell them in competition with new and refurbished cartridges sold by Lexmark. Lexmark would prefer that its customers return their empty cartridges to it for refurbishment and resale, rather than sell those cartridges to a remanufacturer. So Lexmark introduced what it called a “Prebate” program, which enabled customers to purchase new toner cartridges at a 20-percent discount if they would agree to return the cartridge to Lexmark once it was empty. Those terms were communicated to consumers through notices printed on the toner-cartridge boxes, which advised the consumer that opening the box would indicate assent to the terms—a practice commonly known as “shrinkwrap licensing,” see, e.g., ProCD, Inc. v. Zeidenberg, 86 F. 3d 1447, 1449 (CA7 1996). To enforce the Prebate terms, Lexmark included a microchip in each Prebate cartridge that would disable the cartridge after it ran out of toner; for the cartridge to be used again, the microchip would have to be replaced by Lexmark. Static Control is not itself a manufacturer or remanufacturer of toner cartridges. It is, rather, “the market leader [in] making and selling the components necessary to remanufacture Lexmark cartridges.” 697 F. 3d 387, 396 (CA6 2012) (case below). In addition to supplying remanufacturers with toner and various replacement parts, Static Control developed a microchip that could mimic the microchip in Lexmark’s Prebate cartridges. By purchasing Static Control’s microchips and using them to replace the Lexmark microchip, remanufacturers were able to refurbish and resell used Prebate cartridges. Lexmark did not take kindly to that development. In 2002, it sued Static Control, alleging that Static Control’s microchips violated both the Copyright Act of 1976, 17 U. S. C. §101 et seq., and the Digital Millennium Copyright Act, 17 U. S. C. §1201 et seq. Static Control counterclaimed, alleging, among other things, violations of §43(a) of the Lanham Act, 60Stat. 441, codified at 15 U. S. C. §1125(a). Section 1125(a) provides: “(1) Any person who, on or in connection with any goods or services, or any container for goods, uses in commerce any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact, which— “(A) is likely to cause confusion, or to cause mistake, or to deceive as to the affiliation, connection, or association of such person with another person, or as to the origin, sponsorship, or approval of his or her goods, services, or commercial activities by another person, or “(B) in commercial advertising or promotion, misrepresents the nature, characteristics, qualities, or geo- graphic origin of his or her or another person’s goods, services, or commercial activities, “shall be liable in a civil action by any person who believes that he or she is or is likely to be damaged by such act.” Section 1125(a) thus creates two distinct bases of liability: false association, §1125(a)(1)(A), and false advertising, §1125(a)(1)(B). See Waits v. Frito-Lay, Inc., 978 F. 2d 1093, 1108 (CA9 1992). Static Control alleged only false advertising. As relevant to its Lanham Act claim, Static Control alleged two types of false or misleading conduct by Lexmark. First, it alleged that through its Prebate program Lexmark “purposefully misleads end-users” to believe that they are legally bound by the Prebate terms and are thus required to return the Prebate-labeled cartridge to Lexmark after a single use. App. 31, ¶39. Second, it alleged that upon introducing the Prebate program, Lexmark “sent letters to most of the companies in the toner cartridge remanufacturing business” falsely advising those companies that it was illegal to sell refurbished Prebate cartridges and, in particular, that it was illegal to use Static Control’s products to refurbish those cartridges. Id., at 29, ¶35. Static Control asserted that by those statements, Lexmark had materially misrepresented “the nature, characteristics, and qualities” of both its own products and Static Control’s products. Id., at 43–44, ¶85. It further maintained that Lexmark’s misrepresentations had “proximately caused and [we]re likely to cause injury to [Static Control] by diverting sales from [Static Control] to Lexmark,” and had “substantially injured [its] business reputation” by “leading consumers and others in the trade to believe that [Static Control] is engaged in illegal conduct.” Id., at 44, ¶88. Static Control sought treble damages, attorney’s fees and costs, and injunctive relief. [ 1 ] The District Court granted Lexmark’s motion to dismiss Static Control’s Lanham Act claim. It held that Static Control lacked “prudential standing” to bring that claim, App. to Pet. for Cert. 83, relying on a multifactor balancing test it attributed to Associated Gen. Contractors of Cal., Inc. v. Carpenters, 459 U. S. 519 (1983) . The court emphasized that there were “more direct plaintiffs in the form of remanufacturers of Lexmark’s cartridges”; that Static Control’s injury was “remot[e]” because it was a mere “byproduct of the supposed manipulation of consumers’ relationships with remanufacturers”; and that Lexmark’s “alleged intent [was] to dry up spent cartridge supplies at the remanufacturing level, rather than at [Static Control]’s supply level, making remanufacturers Lexmark’s alleged intended target.” App. to Pet. for Cert. 83. The Sixth Circuit reversed the dismissal of Static Control’s Lanham Act claim. 697 F. 3d, at 423. Taking the lay of the land, it identified three competing approaches to determining whether a plaintiff has standing to sue under the Lanham Act. It observed that the Third, Fifth, Eighth, and Eleventh Circuits all refer to “antitrust standing or the [Associated General Contractors] factors in deciding Lanham Act standing,” as the District Court had done. Id., at 410 (citing Conte Bros. Automotive, Inc. v. Quaker State-Slick 50, Inc., 165 F. 3d 221, 233–234 (CA3 1998); Procter & Gamble Co. v. Amway Corp., 242 F. 3d 539, 562–563 (CA5 2001); Gilbert/Robinson, Inc. v. Carrie Beverage-Missouri, Inc., 989 F. 2d 985, 990–991 (CA8 1993); Phoenix of Broward, Inc. v. McDonald’s Corp., 489 F. 3d 1156, 1162–1164 (CA11 2007)). By contrast, “[t]he Seventh, Ninth, and Tenth [Circuits] use a categorical test, permitting Lanham Act suits only by an actual competitor.” 697 F. 3d, at 410 (citing L. S. Heath & Son, Inc. v. AT&T Information Systems, Inc., 9 F. 3d 561, 575 (CA7 1993); Waits, supra, at 1108–1109; Stanfield v. Osborne Industries, Inc., 52 F. 3d 867, 873 (CA10 1995)). And the Second Circuit applies a “ ‘reasonable interest’ approach,” under which a Lanham Act plaintiff “has standing if the claimant can demonstrate ‘(1) a reasonable interest to be protected against the alleged false advertising and (2) a reasonable basis for believing that the interest is likely to be damaged by the alleged false advertising.’ ” 697 F. 3d, at 410 (quoting Famous Horse, Inc. v. 5th Avenue Photo Inc., 624 F. 3d 106, 113 (CA2 2010)). The Sixth Circuit applied the Second Circuit’s reasonable-interest test and concluded that Static Control had standing because it “alleged a cognizable interest in its business reputation and sales to remanufacturers and sufficiently alleged that th[o]se interests were harmed by Lexmark’s statements to the remanufacturers that Static Control was engaging in illegal conduct.” 697 F. 3d, at 411. We granted certiorari to decide “the appropriate ana- lytical framework for determining a party’s standing to maintain an action for false advertising under the Lanham Act.” Pet. for Cert. i; 569 U. S. ____ (2013). [ 2 ] II. “Prudential Standing” The parties’ briefs treat the question on which we granted certiorari as one of “prudential standing.” Because we think that label misleading, we begin by clarifying the nature of the question at issue in this case. From Article III’s limitation of the judicial power to resolving “Cases” and “Controversies,” and the separation-of-powers principles underlying that limitation, we have deduced a set of requirements that together make up the “irreducible constitutional minimum of standing.” Lujan v. Defenders of Wildlife, 504 U. S. 555, 560 (1992) . The plaintiff must have suffered or be imminently threatened with a concrete and particularized “injury in fact” that is fairly traceable to the challenged action of the defendant and likely to be redressed by a favorable judicial decision. Ibid. Lexmark does not deny that Static Control’s alle- gations of lost sales and damage to its business reputa- tion give it standing under Article III to press its false-advertising claim, and we are satisfied that they do. Although Static Control’s claim thus presents a case or controversy that is properly within federal courts’ Article III jurisdiction, Lexmark urges that we should decline to adjudicate Static Control’s claim on grounds that are “prudential,” rather than constitutional. That request is in some tension with our recent reaffirmation of the principle that “a federal court’s ‘obligation’ to hear and decide” cases within its jurisdiction “is ‘virtually unflagging.’ ” Sprint Communications, Inc. v. Jacobs, 571 U. S. ___, ___ (2013) (slip op., at 6) (quoting Colorado River Water Con-servation Dist. v. United States, 424 U. S. 800, 817 (1976) ). In recent decades, however, we have adverted to a “prudential” branch of standing, a doctrine not derived from Article III and “not exhaustively defined” but encompassing (we have said) at least three broad principles: “ ‘the general prohibition on a litigant’s raising another person’s legal rights, the rule barring adjudication of generalized grievances more appropriately addressed in the representative branches, and the requirement that a plaintiff’s complaint fall within the zone of interests protected by the law invoked.’ ” Elk Grove Unified School Dist. v. Newdow, 542 U. S. 1, 12 (2004) (quoting Allen v. Wright, 468 U. S. 737, 751 (1984) ). Lexmark bases its “prudential standing” arguments chiefly on Associated General Contractors, but we did not describe our analysis in that case in those terms. Rather, we sought to “ascertain,” as a matter of statutory interpretation, the “scope of the private remedy created by” Congress in §4 of the Clayton Act, and the “class of persons who [could] maintain a private damages action under” that legislatively conferred cause of action. 459 U. S., at 529, 532. We held that the statute limited the class to plaintiffs whose injuries were proximately caused by a defendant’s antitrust violations. Id., at 532–533. Later decisions confirm that Associated General Contractors rested on statutory, not “prudential,” considerations. See, e.g., Holmes v. Securities Investor Protection Corporation, 503 U. S. 258 –268 (1992) (relying on Associated General Contractors in finding a proximate-cause requirement in the cause of action created by the Racketeer Influenced and Corrupt Organizations Act (RICO), 18 U. S. C. §1964(c)); Anza v. Ideal Steel Supply Corp., 547 U. S. 451, 456 (2006) (affirming that Holmes “relied on a careful interpretation of §1964(c)”). Lexmark’s arguments thus do not deserve the “prudential” label. Static Control, on the other hand, argues that we should measure its “prudential standing” by using the zone-of-interests test. Although we admittedly have placed that test under the “prudential” rubric in the past, see, e.g., Elk Grove, supra, at 12, it does not belong there any more than Associated General Contractors does. Whether a plain- tiff comes within “the ‘zone of interests’ ” is an issue that requires us to determine, using traditional tools of statutory interpretation, whether a legislatively conferred cause of action encompasses a particular plaintiff’s claim. See Steel Co. v. Citizens for Better Environment, 523 U. S. 83 , and n. 2 (1998); Clarke v. Securities Industry Assn., 479 U. S. 388 –395 (1987); Holmes, supra, at 288 (Scalia, J., concurring in judgment). As Judge Silberman of the D. C. Circuit recently observed, “ ‘prudential standing’ is a misnomer” as applied to the zone-of-interests analysis, which asks whether “this particular class of persons ha[s] a right to sue under this substantive statute.” Association of Battery Recyclers, Inc. v. EPA, 716 F. 3d 667, 675–676 (2013) (concurring opinion). [ 3 ] In sum, the question this case presents is whether Static Control falls within the class of plaintiffs whom Congress has authorized to sue under §1125(a). In other words, we ask whether Static Control has a cause of action under the statute. [ 4 ] That question requires us to determine the meaning of the congressionally enacted provision creating a cause of action. In doing so, we apply traditional principles of statutory interpretation. We do not ask whether in our judgment Congress should have authorized Static Control’s suit, but whether Congress in fact did so. Just as a court cannot apply its independent policy judgment to recognize a cause of action that Congress has denied, see Alexander v. Sandoval, 532 U. S. 275 –287 (2001), it cannot limit a cause of action that Congress has created merely because “prudence” dictates. III. Static Control’s Right To Sue Under §1125(a) Thus, this case presents a straightforward question of statutory interpretation: Does the cause of action in §1125(a) extend to plaintiffs like Static Control? The statute authorizes suit by “any person who believes that he or she is likely to be damaged” by a defendant’s false advertising. §1125(a)(1). Read literally, that broad language might suggest that an action is available to anyone who can satisfy the minimum requirements of Article III. No party makes that argument, however, and the “unlikelihood that Congress meant to allow all factually injured plaintiffs to recover persuades us that [§1125(a)] should not get such an expansive reading.” Holmes, 503 U. S., at 266 (footnote omitted). We reach that conclusion in light of two relevant background principles already mentioned: zone of interests and proximate causality. A. Zone of Interests First, we presume that a statutory cause of action extends only to plaintiffs whose interests “fall within the zone of interests protected by the law invoked.” Allen, 468 U. S., at 751. The modern “zone of interests” formulation originated in Association of Data Processing Service Organizations, Inc. v. Camp, 397 U. S. 150 (1970) , as a limitation on the cause of action for judicial review conferred by the Administrative Procedure Act (APA). We have since made clear, however, that it applies to all statutorily created causes of action; that it is a “requirement of general application”; and that Congress is presumed to “legislat[e] against the background of” the zone-of-interests limitation, “which applies unless it is expressly negated.” Bennett v. Spear, 520 U. S. 154, 163 (1997) ; see also Holmes, supra, at 287–288 (Scalia, J., concurring in judgment). It is “perhaps more accurat[e],” though not very different as a practical matter, to say that the limitation always applies and is never negated, but that our analysis of certain statutes will show that they protect a more-than-usually “expan[sive]” range of interests. Bennett, supra, at 164. The zone-of-interests test is therefore an appropriate tool for determining who may invoke the cause of action in §1125(a). [ 5 ] We have said, in the APA context, that the test is not “ ‘especially demanding,’ ” Match-E-Be-Nash-She-Wish Band of Pottawatomi Indians v. Patchak, 567 U. S. ___, ___ (2012) (slip op., at 15). In that context we have often “conspicuously included the word ‘arguably’ in the test to indicate that the benefit of any doubt goes to the plaintiff,” and have said that the test “forecloses suit only when a plaintiff’s ‘interests are so marginally related to or inconsistent with the purposes implicit in the statute that it cannot reasonably be assumed that’ ” Congress authorized that plaintiff to sue. Id., at ___ (slip op., at 15–16). That lenient approach is an appropriate means of preserving the flexibility of the APA’s omnibus judicial-review provision, which permits suit for violations of numerous statutes of varying character that do not themselves include causes of action for judicial review. “We have made clear, however, that the breadth of the zone of interests varies according to the provisions of law at issue, so that what comes within the zone of interests of a statute for purposes of obtaining judicial review of administrative action under the ‘ “generous review provisions” ’ of the APA may not do so for other purposes.” Bennett, supra, at 163 (quot- ing Clarke, 479 U. S., at 400, n. 16, in turn quoting Data Processing, supra, at 156). Identifying the interests protected by the Lanham Act, however, requires no guesswork, since the Act includes an “unusual, and extraordinarily helpful,” detailed statement of the statute’s purposes. H. B. Halicki Productions v. United Artists Communications, Inc., 812 F. 2d 1213, 1214 (CA9 1987). Section 45 of the Act, codified at 15 U. S. C. §1127, provides: “The intent of this chapter is to regulate commerce within the control of Congress by making actionable the deceptive and misleading use of marks in such commerce; to protect registered marks used in such commerce from interference by State, or territorial legislation; to protect persons engaged in such commerce against unfair competition; to prevent fraud and deception in such commerce by the use of reproductions, copies, counterfeits, or colorable imitations of registered marks; and to provide rights and remedies stipulated by treaties and conventions respect- ing trademarks, trade names, and unfair competition entered into between the United States and foreign nations.” Most of the enumerated purposes are relevant to false-association cases; a typical false-advertising case will implicate only the Act’s goal of “protect[ing] persons engaged in [commerce within the control of Congress] against unfair competition.” Although “unfair competition” was a “plastic” concept at common law, Ely-Norris Safe Co. v. Mosler Safe Co., 7 F. 2d 603, 604 (CA2 1925) (L. Hand, J.), it was understood to be concerned with injuries to business reputation and present and future sales. See Rogers, Book Review, 39 Yale L. J. 297, 299 (1929); see generally 3 Restatement of Torts, ch. 35, Introductory Note, pp. 536–537 (1938). We thus hold that to come within the zone of interests in a suit for false advertising under §1125(a), a plaintiff must allege an injury to a commercial interest in reputation or sales. A consumer who is hoodwinked into purchasing a disappointing product may well have an injury-in-fact cognizable under Article III, but he cannot invoke the protection of the Lanham Act—a conclusion reached by every Circuit to consider the question. See Colligan v. Activities Club of N. Y., Ltd., 442 F. 2d 686, 691–692 (CA2 1971); Serbin v. Ziebart Int’l Corp., 11 F. 3d 1163, 1177 (CA3 1993); Made in the USA Foundation v. Phillips Foods, Inc., 365 F. 3d 278, 281 (CA4 2004); Procter & Gamble Co., 242 F. 3d, at 563–564; Barrus v. Sylvania, 55 F. 3d 468, 470 (CA9 1995); Phoenix of Broward, 489 F. 3d, at 1170. Even a business misled by a supplier into purchasing an inferior product is, like consumers generally, not under the Act’s aegis. B. Proximate Cause Second, we generally presume that a statutory cause of action is limited to plaintiffs whose injuries are proxi- mately caused by violations of the statute. For centuries, it has been “a well established principle of [the common] law, that in all cases of loss, we are to attribute it to the proximate cause, and not to any remote cause.” Waters v. Merchants’ Louisville Ins. Co., 11 Pet. 213, 223 (1837); see Holmes, 503 U. S., at 287 (Scalia, J., concurring in judgment). That venerable principle reflects the reality that “the judicial remedy cannot encompass every conceivable harm that can be traced to alleged wrongdoing.” Associ- ated Gen. Contractors, 459 U. S., at 536. Congress, we assume, is familiar with the common-law rule and does not mean to displace it sub silentio. We have thus construed federal causes of action in a variety of contexts to incorporate a requirement of proximate causation. See, e.g., Dura Pharmaceuticals, Inc. v. Broudo, 544 U. S. 336, 346 (2005) (securities fraud); Holmes, supra, at 268–270 (RICO); Associated Gen. Contractors, supra, at 529–535 (Clayton Act). No party disputes that it is proper to read §1125(a) as containing such a requirement, its broad language notwithstanding. The proximate-cause inquiry is not easy to define, and over the years it has taken various forms; but courts have a great deal of experience applying it, and there is a wealth of precedent for them to draw upon in doing so. See Exxon Co., U. S. A. v. Sofec, Inc., 517 U. S. 830 –839 (1996); Pacific Operators Offshore, LLP v. Valladolid, 565 U. S. ___, ___ (2012) (Scalia, J., concurring in part and concurring in judgment) (slip op., at 3). Proximate-cause analysis is controlled by the nature of the statutory cause of action. The question it presents is whether the harm alleged has a sufficiently close connection to the conduct the statute prohibits. Put differently, the proximate-cause requirement generally bars suits for alleged harm that is “too remote” from the defendant’s unlawful conduct. That is ordinarily the case if the harm is purely derivative of “misfortunes visited upon a third person by the defendant’s acts.” Holmes, supra, at 268–269; see, e.g., Hemi Group, LLC v. City of New York, 559 U. S. 1 –11 (2010). In a sense, of course, all commercial injuries from false advertising are derivative of those suffered by consumers who are deceived by the advertising; but since the Lanham Act authorizes suit only for commercial injuries, the intervening step of consumer deception is not fatal to the showing of proximate causation required by the statute. See Harold H. Huggins Realty, Inc. v. FNC, Inc., 634 F. 3d 787, 800–801 (CA5 2011). That is consistent with our recognition that under common-law principles, a plaintiff can be directly injured by a misrepresentation even where “a third party, and not the plaintiff, . . . relied on” it. Bridge v. Phoenix Bond & Indemnity Co., 553 U. S. 639, 656 (2008) . We thus hold that a plaintiff suing under §1125(a) ordinarily must show economic or reputational injury flowing directly from the deception wrought by the defendant’s advertising; and that that occurs when deception of consumers causes them to withhold trade from the plaintiff. That showing is generally not made when the deception produces injuries to a fellow commercial actor that in turn affect the plaintiff. For example, while a competitor who is forced out of business by a defendant’s false advertising generally will be able to sue for its losses, the same is not true of the competitor’s landlord, its electric company, and other commercial parties who suffer merely as a result of the competitor’s “inability to meet [its] financial obligations.” Anza, 547 U. S., at 458. [ 6 ] C. Proposed Tests At oral argument, Lexmark agreed that the zone of in- terests and proximate causation supply the relevant background limitations on suit under §1125(a). See Tr. of Oral Arg. 4–5, 11–12, 17–18. But it urges us to adopt, as the optimal formulation of those principles, a multifactor balancing test derived from Associated General Contrac-tors. In the alternative, it asks that we adopt a categorical test permitting only direct competitors to sue for false advertising. And although neither party urges adoption of the “reasonable interest” test applied below, several amici do so. While none of those tests is wholly without merit, we decline to adopt any of them. We hold instead that a direct application of the zone-of-interests test and the proximate-cause requirement supplies the relevant limits on who may sue. The balancing test Lexmark advocates was first articulated by the Third Circuit in Conte Bros. and later adopted by several other Circuits. Conte Bros. identified five relevant considerations: “(1) The nature of the plaintiff’s alleged injury: Is the injury of a type that Congress sought to redress in providing a private remedy for violations of the [Lanham Act]? “(2) The directness or indirectness of the asserted injury. “(3) The proximity or remoteness of the party to the alleged injurious conduct. “(4) The speculativeness of the damages claim. “(5) The risk of duplicative damages or complexity in apportioning damages.” 165 F. 3d, at 233 (citations and internal quotation marks omitted). This approach reflects a commendable effort to give content to an otherwise nebulous inquiry, but we think it slightly off the mark. The first factor can be read as requiring that the plaintiff’s injury be within the relevant zone of interests and the second and third as requiring (somewhat redundantly) proximate causation; but it is not correct to treat those requirements, which must be met in every case, as mere factors to be weighed in a balance. And the fourth and fifth factors are themselves problem- atic. “[T]he difficulty that can arise when a court attempts to ascertain the damages caused by some remote action” is a “motivating principle” behind the proximate-cause requirement, Anza, supra, at 457–458; but potential diffi- culty in ascertaining and apportioning damages is not, as Conte Bros. might suggest, an independent basis for denying standing where it is adequately alleged that a defendant’s conduct has proximately injured an interest of the plaintiff’s that the statute protects. Even when a plaintiff cannot quantify its losses with sufficient certainty to re- cover damages, it may still be entitled to injunctive re- lief under §1116(a) (assuming it can prove a likelihood of future injury) or disgorgement of the defendant’s ill-gotten profits under §1117(a). See TrafficSchool.com, Inc. v. Edriver Inc., 653 F. 3d 820, 831 (CA9 2011); Johnson & Johnson v. Carter-Wallace, Inc., 631 F. 2d 186, 190 (CA2 1980). Finally, experience has shown that the Conte Bros. approach, like other open-ended balancing tests, can yield unpredictable and at times arbitrary results. See, e.g., Tushnet, Running the Gamut from A to B: Federal Trademark and False Advertising Law, 159 U. Pa. L. Rev. 1305, 1376–1379 (2011). In contrast to the multifactor balancing approach, the direct-competitor test provides a bright-line rule; but it does so at the expense of distorting the statutory language. To be sure, a plaintiff who does not compete with the defendant will often have a harder time establishing proximate causation. But a rule categorically prohibiting all suits by noncompetitors would read too much into the Act’s reference to “unfair competition” in §1127. By the time the Lanham Act was adopted, the common-law tort of unfair competition was understood not to be limited to actions between competitors. One leading authority in the field wrote that “there need be no competition in unfair competition,” just as “[t]here is no soda in soda water, no grapes in grape fruit, no bread in bread fruit, and a clothes horse is not a horse but is good enough to hang things on.” Rogers, 39 Yale L. J., at 299; accord, Vogue Co. v. Thompson-Hudson Co., 300 F. 509, 512 (CA6 1924); 1 H. Nims, The Law of Unfair Competition and Trade-Marks, p. vi (4th ed. 1947); 2 id., at 1194–1205. It is thus a mistake to infer that because the Lanham Act treats false advertising as a form of unfair competition, it can protect only the false-advertiser’s direct competitors. Finally, there is the “reasonable interest” test applied by the Sixth Circuit in this case. As typically formulated, it requires a commercial plaintiff to “demonstrate ‘(1) a reasonable interest to be protected against the alleged false advertising and (2) a reasonable basis for believing that the interest is likely to be damaged by the alleged false advertising.’ ” 697 F. 3d, at 410 (quoting Famous Horse, 624 F. 3d, at 113). A purely practical objection to the test is that it lends itself to widely divergent appli- cation. Indeed, its vague language can be understood as requiring only the bare minimum of Article III standing. The popularity of the multifactor balancing test reflects its appeal to courts tired of “grappl[ing] with defining” the “ ‘reasonable interest’ ” test “with greater precision.” Conte Bros., 165 F. 3d, at 231. The theoretical difficulties with the test are even more substantial: The relevant question is not whether the plaintiff’s interest is “reasonable,” but whether it is one the Lanham Act protects; and not whether there is a “reasonable basis” for the plaintiff’s claim of harm, but whether the harm alleged is proximately tied to the defendant’s conduct. In short, we think the principles set forth above will provide clearer and more accurate guidance than the “reasonable interest” test. IV. Application Applying those principles to Static Control’s false-advertising claim, we conclude that Static Control comes within the class of plaintiffs whom Congress authorized to sue under §1125(a). To begin, Static Control’s alleged injuries—lost sales and damage to its business reputation—are injuries to precisely the sorts of commercial interests the Act protects. Static Control is suing not as a deceived consumer, but as a “perso[n] engaged in” “commerce within the control of Congress” whose position in the marketplace has been damaged by Lexmark’s false advertising. §1127. There is no doubt that it is within the zone of interests protected by the statute. Static Control also sufficiently alleged that its injuries were proximately caused by Lexmark’s misrepresentations. This case, it is true, does not present the “classic Lanham Act false-advertising claim” in which “ ‘one competito[r] directly injur[es] another by making false statements about his own goods [or the competitor’s goods] and thus inducing customers to switch.’ ” Harold H. Huggins Realty, 634 F. 3d, at 799, n. 24. But although diversion of sales to a direct competitor may be the paradigmatic direct injury from false advertising, it is not the only type of injury cognizable under §1125(a). For at least two reasons, Static Control’s allegations satisfy the requirement of proximate causation. First, Static Control alleged that Lexmark disparaged its business and products by asserting that Static Control’s business was illegal. See 697 F. 3d, at 411, n. 10 (noting allegation that Lexmark “directly target[ed] Static Control” when it “falsely advertised that Static Control infringed Lexmark’s patents”). When a defendant harms a plaintiff’s reputation by casting aspersions on its business, the plaintiff’s injury flows directly from the audience’s belief in the disparaging statements. Courts have therefore afforded relief under §1125(a) not only where a defendant denigrates a plaintiff’s product by name, see, e.g., McNeilab, Inc. v. American Home Prods. Corp., 848 F. 2d 34, 38 (CA2 1988), but also where the defendant damages the product’s reputation by, for example, equat-ing it with an inferior product, see, e.g., Camel Hair and Cashmere Inst. of Am., Inc. v. Associated Dry Goods Corp., 799 F. 2d 6, 7–8, 11–12 (CA1 1986); PPX Enterprises, Inc. v. Audiofidelity, Inc., 746 F. 2d 120, 122, 125 (CA2 1984). Traditional proximate-causation principles support those results: As we have observed, a defendant who “ ‘seeks to promote his own interests by telling a known falsehood to or about the plaintiff or his product’ ” may be said to have proximately caused the plaintiff’s harm. Bridge, 553 U. S., at 657 (quoting Restatement (Second) of Torts §870, Comment h (1977); emphasis added in Bridge). The District Court emphasized that Lexmark and Static Control are not direct competitors. But when a party claims reputational injury from disparagement, competition is not required for proximate cause; and that is true even if the defendant’s aim was to harm its immediate competitors, and the plaintiff merely suffered collateral damage. Consider two rival carmakers who purchase airbags for their cars from different third-party manufacturers. If the first carmaker, hoping to divert sales from the second, falsely proclaims that the airbags used by the second carmaker are defective, both the second carmaker and its airbag supplier may suffer reputational injury, and their sales may decline as a result. In those circumstances, there is no reason to regard either party’s injury as de- rivative of the other’s; each is directly and independently harmed by the attack on its merchandise. In addition, Static Control adequately alleged proximate causation by alleging that it designed, manufactured, and sold microchips that both (1) were necessary for, and (2) had no other use than, refurbishing Lexmark toner cartridges. See App. 13, ¶31; id., at 37, ¶54. [ 7 ] It follows from that allegation that any false advertising that reduced the remanufacturers’ business necessarily injured Static Control as well. Taking Static Control’s assertions at face value, there is likely to be something very close to a 1:1 relationship between the number of refurbished Prebate cartridges sold (or not sold) by the remanufacturers and the number of Prebate microchips sold (or not sold) by Static Control. “Where the injury alleged is so integral an aspect of the [violation] alleged, there can be no question” that proximate cause is satisfied. Blue Shield of Va. v. McCready, 457 U. S. 465, 479 (1982) . To be sure, on this view, the causal chain linking Static Control’s injuries to consumer confusion is not direct, but includes the intervening link of injury to the remanufacturers. Static Control’s allegations therefore might not support standing under a strict application of the “ ‘ “general tendency” ’ ” not to stretch proximate causation “ ‘ “beyond the first step.” ’ ” Holmes, 503 U. S., at 271. But the reason for that general tendency is that there ordinarily is a “discontinuity” between the injury to the direct victim and the injury to the indirect victim, so that the latter is not surely attributable to the former (and thus also to the defendant’s conduct), but might instead have resulted from “any number of [other] reasons.” Anza, 547 U. S., at 458–459. That is not the case here. Static Control’s allegations suggest that if the remanufacturers sold 10,000 fewer refurbished cartridges because of Lexmark’s false advertising, then it would follow more or less automatically that Static Control sold 10,000 fewer microchips for the same reason, without the need for any “speculative . . . proceedings” or “intricate, uncertain inquiries.” Id., at 459–460. In these relatively unique circumstances, the remanufacturers are not “more immediate victim[s]” than Static Control. Bridge, supra, at 658. Although we conclude that Static Control has alleged an adequate basis to proceed under §1125(a), it cannot obtain relief without evidence of injury proximately caused by Lexmark’s alleged misrepresentations. We hold only that Static Control is entitled to a chance to prove its case. * * * To invoke the Lanham Act’s cause of action for false advertising, a plaintiff must plead (and ultimately prove) an injury to a commercial interest in sales or business reputation proximately caused by the defendant’s mis- representations. Static Control has adequately pleaded both elements. The judgment of the Court of Appeals is affirmed. It is so ordered. Notes 1 Lexmark contends that Static Control’s allegations failed to describe “commercial advertising or promotion” within the meaning of . That question is not before us, and we express no view on it. We assume without deciding that the communica-tions alleged by Static Control qualify as commercial advertising or promotion. 2 Other aspects of the parties’ sprawling litigation, including Lexmark’s claims under federal copyright and patent law and Static Control’s claims under federal antitrust and North Carolina unfair-competition law, are not before us. Our review pertains only to Static Control’s Lanham Act claim. 3 The zone-of-interests test is not the only concept that we have previously classified as an aspect of “prudential standing” but for which, upon closer inspection, we have found that label inapt. Take, for example, our reluctance to entertain generalized grievances—i.e., suits “claiming only harm to [the plaintiff’s] and every citizen’s interest in proper application of the Constitution and laws, and seeking relief that no more directly and tangibly benefits him than it does the public at large.” Lujan v. Defenders of Wildlife, –574 (1992). While we have at times grounded our reluctance to entertain such suits in the “counsels of prudence” (albeit counsels “close[ly] relat[ed] to the policies reflected in” Article III), Valley Forge Christian College v. Americans United for Separation of Church and State, Inc., , we have since held that such suits do not present constitutional “cases” or “controversies.” See, e.g., Lance v. Coffman, (per curiam); DaimlerChrysler Corp. v. Cuno, –346 (2006); Defenders of Wildlife, supra, at 573–574. They are barred for constitutional reasons, not “prudential” ones. The limitations on third-party standing are harder to classify; we have observed that third-party standing is “ ‘closely related to the question whether a person in the litigant’s position will have a right of action on the claim,’ ” Department of Labor v. Triplett, , n. ** (1990) (quoting Warth v. Seldin, , n. 12 (1975)), but most of our cases have not framed the inquiry in that way. See, e.g., Kowalski v. Tesmer, –129 (2004) (suggesting it is an element of “prudential standing”). This case does not present any issue of third-party standing, and consideration of that doctrine’s proper place in the standing firmament can await another day. 4 We have on occasion referred to this inquiry as “statutory standing” and treated it as effectively jurisdictional. See, e.g., Steel Co. v. Citizens for Better Environment, , and n. 2 (1998); cases citedid., at 114–117 (Stevens, J., concurring in judgment). That label is an improvement over the language of “prudential standing,” since it correctly places the focus on the statute. But it, too, is misleading, since “the absence of a valid (as opposed to arguable) cause of action does not implicate subject-matter jurisdiction, i.e., the court’s statutory or constitutional power to adjudicate the case.’ ” Verizon Md. Inc. v. Public Serv. Comm’n of Md., –643 (2002) (quoting Steel Co., supra, at 89); see also Grocery Mfrs. Assn. v. EPA, 693 F. 3d 169, 183–185 (Kavanaugh, J., dissenting), and cases cited therein; Pathak, Statutory Standing and the Tyranny of Labels, 62 Okla. L. Rev. 89, 106 (2009). 5 Although we announced the modern zone-of-interests test in 1971, its roots lie in the common-law rule that a plaintiff may not recover under the law of negligence for injuries caused by violation of a statute unless the statute “is interpreted as designed to protect the class of persons in which the plaintiff is included, against the risk of the type of harm which has in fact occurred as a result of its violation.” W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts §36, pp. 229–230 (5th ed. 1984); see cases cited id., at 222–227; Gorris v. Scott, [1874] 9 L. R. Exch. 125 (Eng.). Statutory causes of action are regularly interpreted to incorporate standard common-law limitations on civil liability—the zone-of-interests test no less than the requirement of proximate causation, see Part III–B, infra. 6 Proximate causation is not a requirement of Article III standing, which requires only that the plaintiff’s injury be fairly traceable to the defendant’s conduct. Like the zone-of-interests test, see supra, at 8–9, and nn. 3–4, it is an element of the cause of action under the statute, and so is subject to the rule that “the absence of a valid (as opposed to arguable) cause of action does not implicate subject-matter jurisdiction.” Steel Co., 523 U. S., at 89. But like any other element of a cause of action, it must be adequately alleged at the pleading stage in order for the case to proceed. See Ashcroft v. Iqbal, –679 (2009). If a plaintiff’s allegations, taken as true, are insufficient to establish proximate causation, then the complaint must be dismissed; if they are sufficient, then the plaintiff is entitled to an opportunity to prove them. 7 We understand this to be the thrust of both sides’ allegations concerning Static Control’s design and sale of specialized microchips for the specific purpose of enabling the remanufacture of Lexmark’s Prebate cartridges.

572.US.915

Akamai Technologies, Inc., a respondent here, is the exclusive licensee of a patent that claims a method of delivering electronic data using a content delivery network (CDN). Petitioner, Limelight Networks, Inc., also operates a CDN and carries out several of the steps claimed in the patent, but its customers, rather than Limelight itself, perform a step of the patent known as “tagging.” Under Federal Circuit case law, liability for direct infringement under 35 U. S. C. §271(a) requires performance of all steps of a method patent to be attributable to a single party. This position was most recently refined in Muniauction, Inc. v. Thomson Corp., 532 F.3d 1318. The District Court concluded that Limelight could not have directly infringed the patent at issue because performance of the tagging step could not be attributed to it. The en banc Federal Circuit reversed, holding that a defendant who performed some steps of a method patent and encouraged others to perform the rest could be liable for inducement of infringement even if no one was liable for direct infringement. The en banc court concluded that the evidence could support liability for Limelight on an inducement theory and remanded for further proceedings. Held: A defendant is not liable for inducing infringement under §271(b) when no one has directly infringed under §271(a) or any other statutory provision. Pp. 4–11. (a) Liability for inducement must be predicated on direct infringement. Aro Mfg. Co. v. Convertible Top Replacement Co., 365 U.S. 336, 341. Assuming that Muniauction’s holding is correct, respondents’ method has not been infringed because the performance of all of its steps is not attributable to any one person. Since direct infringe-ment has not occurred, there can be no inducement of infringement under §271(b). The Federal Circuit’s contrary view would deprive §271(b) of ascertainable standards and require the courts to develop two parallel bodies of infringement law. This Court’s reading of §271(b) is reinforced by §271(f)(1), which illustrates that Congress knows how to impose inducement liability predicated on non-infringing conduct when it wishes to do so. The notion that conduct which would be infringing in altered circumstances can form the basis for contributory infringement has been rejected, see Deepsouth Packing Co. v. Laitram Corp., 406 U.S. 518, 526–527, and there is no reason to apply a different rule for inducement. Pp. 4–7. (b) Respondents claim that principles from tort law and criminal aiding and abetting doctrine, as well as patent law principles in existence before the 1952 Patent Act, support the Federal Circuit’s reading of the statute, but their arguments are unpersuasive. Though a would-be infringer could evade liability by dividing performance of a method patent’s steps with another whose conduct cannot be attributed to the defendant, this is merely a result of the Federal Circuit’s interpretation of §271(a), and a desire to avoid this consequence does not justify fundamentally altering the rules of inducement liability clearly required by the Patent Act’s text and structure. Pp. 8–10. (c) Because the question presented here is clearly focused on §271(b) and presupposes that Limelight has not committed direct infringement under §271(a), the Court declines to address whether the Federal Circuit’s decision in Muniauction is correct. P. 10. 692 F.3d 1301, reversed and remanded. Alito, J., delivered the opinion for a unanimous Court.

This case presents the question whether a defendant may be liable for inducing infringement of a patent under 35 U. S. C. §271(b) when no one has directly infringed the patent under §271(a) or any other statutory provision. The statutory text and structure and our prior case law require that we answer this question in the negative. We accordingly reverse the Federal Circuit, which reached the opposite conclusion. I A Respondent the Massachusetts Institute of Technology is the assignee of U. S. Patent No. 6,108,703 (’703 patent), which claims a method of delivering electronic data using a “content delivery network,” or “CDN.” Respondent Akamai Technologies, Inc., is the exclusive licensee. Akamai maintains many servers distributed in various locations. Proprietors of Web sites, known as “content providers,” contract with Akamai to deliver their Web sites’ content to individual Internet users. The ’703 patent provides for the designation of certain components of a content provider’s Web site (often large files, such as video or music files) to be stored on Akamai’s servers and accessed from those servers by Internet users. The process of designating components to be stored on Akamai’s servers is known as “tagging.” By “aggregat[ing] the data demands of multiple content providers with differing peak usage patterns and serv[ing] that content from multiple servers in multiple locations,” 614 F. Supp. 2d 90, 96 (Mass. 2009), as well as by delivering content from servers located in the same geographic area as the users who are attempting to access it, Akamai is able to increase the speed with which Internet users access the content of its customers’ Web sites. Petitioner Limelight Networks, Inc., also operates a CDN and carries out several of the steps claimed in the ’703 patent. But instead of tagging those components of its customers’ Web sites that it intends to store on its servers (a step included in the ’703 patent), Limelight requires its customers to do their own tagging.[1] Respondents claim that Limelight “provides instructions and offers technical assistance” to its customers regarding how to tag, 629 F. 3d 1311, 1321 (CA Fed. 2010), but the record is undisputed that Limelight does not tag the components to be stored on its servers. B In 2006, respondents sued Limelight in the United States District Court for the District of Massachusetts, claiming patent infringement. The case was tried to a jury, which found that Limelight had committed infringement and awarded more than $40 million in damages. Respondents’ victory was short-lived, however. After the jury returned its verdict, the Federal Circuit decided Muniauction, Inc. v. Thomson Corp., 532 F. 3d 1318 (2008). In that case the Court of Appeals rejected a claim that the defendant’s method, involving bidding on financial instruments using a computer system, directly infringed the plaintiff’s patent. The defendant performed some of the steps of the patented method, and its customers, to whom the defendant gave access to its system along with instructions on the use of the system, performed the remaining steps. The court started from “the proposition that direct infringement requires a single party to perform every step of a claimed method.” Id., at 1329. This requirement is satisfied even though the steps are actually undertaken by multiple parties, the court explained, if a single defendant “exercises ‘control or direction’ over the entire process such that every step is attributable to the controlling party.” Ibid. The court held that the defendant in Muniauction was not liable for direct infringement because it did not exercise control or direction over its customers’ performance of those steps of the patent that the defendant itself did not perform. Id., at 1330. In light of Muniauction, Limelight moved for reconsideration of its earlier motion for judgment as a matter of law, which the District Court had denied. The District Court granted the motion, concluding that Muniauction precluded a finding of direct infringement under §271(a) because infringement of the ’703 patent required tagging and Limelight does not control or direct its customers’ tagging. A panel of the Federal Circuit affirmed, explaining that a defendant that does not itself undertake all of a patent’s steps can be liable for direct infringement only “when there is an agency relationship between the parties who perform the method steps or when one party is contractually obligated to the other to perform the steps.” 629 F. 3d, at 1320. Since neither of these conditions was met in the present case, the Federal Circuit panel held that Limelight could not be held liable for direct infringement.[2] Ibid. The Federal Circuit granted en banc review and reversed. The en banc court found it unnecessary to revisit its §271(a) direct infringement case law. Instead, it concluded that the “evidence could support a judgment in [respondents’] favor on a theory of induced infringement” under §271(b). 692 F. 3d 1301, 1319 (2012) (per curiam). This was true, the court explained, because §271(b) liability arises when a defendant carries out some steps constituting a method patent and encourages others to carry out the remaining steps—even if no one would be liable as a direct infringer in such circumstances, because those who performed the remaining steps did not act as agents of, or under the direction or control of, the defendant. The Court of Appeals did not dispute that “there can be no indirect infringement without direct infringement,” id., at 1308, but it explained that “[r]equiring proof that there has been direct infringement . . . is not the same as requiring proof that a single party would be liable as a direct infringer,” id., at 1308–1309 (emphasis deleted). Judge Newman and Judge Linn both dissented (with the latter joined by Judges Dyk, Prost, and O’Malley). Limelight sought certiorari, which we granted. 571 U. S. ___ (2014). II A Neither the Federal Circuit, see 692 F. 3d, at 1308, nor respondents, see Tr. of Oral Arg. 44, dispute the proposition that liability for inducement must be predicated on direct infringement. This is for good reason, as our case law leaves no doubt that inducement liability may arise “if, but only if, [there is] . . . direct infringement.” Aro Mfg. Co. v. Convertible Top Replacement Co., 365 U. S 336, 341 (1961) (emphasis deleted).[3] One might think that this simple truth is enough to dispose of this appeal. But the Federal Circuit reasoned that a defendant can be liable for inducing infringement under §271(b) even if no one has committed direct infringement within the terms of §271(a) (or any other provision of the patent laws), because direct infringement can exist independently of a violation of these statutory provisions. See 692 F. 3d, at 1314. The Federal Circuit’s analysis fundamentally misunderstands what it means to infringe a method patent. A method patent claims a number of steps; under this Court’s case law, the patent is not infringed unless all the steps are carried out. See, e.g., Aro, supra, at 344 (a “pat-ent covers only the totality of the elements in the claim and . . . no element, separately viewed, is within the grant”). This principle follows ineluctably from what a patent is: the conferral of rights in a particular claimed set of elements. “Each element contained in a patent claim is deemed material to defining the scope of the patented invention,” Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U. S. 17, 29 (1997) , and a patentee’s rights extend only to the claimed combination of elements, and no further. The Federal Circuit held in Muniauction that a method’s steps have not all been performed as claimed by the patent unless they are all attributable to the same de-fendant, either because the defendant actually performed those steps or because he directed or controlled others who performed them. See 532 F. 3d, at 1329–1330. Assuming without deciding that the Federal Circuit’s holding in Muniauction is correct, there has simply been no infringement of the method in which respondents have staked out an interest, because the performance of all the patent’s steps is not attributable to any one person. And, as both the Federal Circuit and respondents admit, where there has been no direct infringement, there can be no inducement of infringement under §271(b). The Federal Circuit’s contrary view would deprive §271(b) of ascertainable standards. If a defendant can be held liable under §271(b) for inducing conduct that does not constitute infringement, then how can a court assess when a patent holder’s rights have been invaded? What if a defendant pays another to perform just one step of a 12-step process, and no one performs the other steps, but that one step can be viewed as the most important step in the process? In that case the defendant has not encouraged infringement, but no principled reason prevents him from being held liable for inducement under the Federal Circuit’s reasoning, which permits inducement liability when fewer than all of a method’s steps have been performed within the meaning of the patent. The decision below would require the courts to develop two parallel bodies of infringement law: one for liability for direct infringement, and one for liability for inducement. Section 271(f)(1) reinforces our reading of §271(b). That subsection imposes liability on a party who “supplies or causes to be supplied in or from the United States all or a substantial portion of the components of a patented invention . . . in such manner as to actively induce the combination of such components outside of the United States in a manner that would infringe the patent if such combination occurred within the United States” (emphasis added). As this provision illustrates, when Congress wishes to impose liability for inducing activity that does not itself constitute direct infringement, it knows precisely how to do so. The courts should not create liability for inducement of non-infringing conduct where Congress has elected not to extend that concept. The Federal Circuit seems to have adopted the view that Limelight induced infringement on the theory that the steps that Limelight and its customers perform would infringe the ’703 patent if all the steps were performed by the same person. But we have already rejected the notion that conduct which would be infringing in altered circumstances can form the basis for contributory infringement, and we see no reason to apply a different rule for inducement. In Deepsouth Packing Co. v. Laitram Corp., 406 U. S. 518 (1972) , a manufacturer produced components of a patented machine and then exported those components overseas to be assembled by its foreign customers.[4] (The assembly by the foreign customers did not violate U. S. patent laws.) In both Deepsouth and this case, the conduct that the defendant induced or contributed to would have been infringing if committed in altered circumstances: in Deepsouth if the machines had been assembled in the United States, see id., at 526, and in this case if performance of all of the claimed steps had been attributable to the same person. In Deepsouth, we rejected the possibility of contributory infringement because the machines had not been assembled in the United States, and direct infringement had consequently never occurred. See id., at 526–527. Similarly, in this case, performance of all the claimed steps cannot be attributed to a single person, so direct infringement never occurred. Limelight cannot be liable for inducing infringement that never came to pass. B Respondents’ arguments in support of the Federal Circuit’s reading of the statute are unpersuasive. First, respondents note that tort law imposes liability on a defendant who harms another through a third party, even if that third party would not himself be liable, and respondents contend that, given the background tort principles against which the Patent Act of 1952 was enacted, it should not matter that no one is liable for direct infringement in this case. But the reason Limelight could not have induced infringement under §271(b) is not that no third party is liable for direct infringement; the problem, instead, is that no direct infringement was committed. Muniauction (which, again, we assume to be correct) instructs that a method patent is not directly infringed—and the patentee’s interest is thus not violated—unless a single actor can be held responsible for the performance of all steps of the patent. Because Limelight did not undertake all steps of the ’703 patent and cannot otherwise be held responsible for all those steps, respondents’ rights have not been violated. Unsurprisingly, respondents point us to no tort case in which liability was imposed because a defendant caused an innocent third party to undertake action that did not violate the plaintiff’s legal rights. In a related argument, respondents contend that, at tort, liability sometimes attaches where two or more defendants inflict injury, even if each defendant’s conduct, standing alone, would not be actionable. See W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Torts §52, p. 354 (5th ed. 1984) (multiple defendants who each add negligible impurities to stream liable if aggregate impurities cause harm). But the rationale for imposing liability in these circumstances is that the defendants collectively invaded the plaintiff’s protected interests. See ibid. By contrast, under the Muniauction rule, respondents’ interests in the ’713 patent have not been invaded. Second, respondents seek to analogize §271(b) to the federal aiding and abetting statute, 18 U. S. C. §2, and they argue that two parties who divide all the necessary elements of a crime between them are both guilty under §2. The analogy does not hold up. The aiding and abetting statute must be read “against its common-law background,” Standefer v. United States, 447 U. S. 10, 19 (1980) , and at common law two or more defendants, each of whom committed an element of a crime, were liable as principals. See, e.g., 1 J. Bishop, Commentaries on the Criminal Law §649, p. 392 (7th ed. 1882). While we have drawn on criminal law concepts in the past in interpreting §271(b), see Global-Tech Appliances, Inc. v. SEB S. A., 563 U. S. ___, ___ (2011) (slip op., at 10–12), we think it unlikely that Congress had this particular doctrine in mind when it enacted the Patent Act of 1952, given the doctrine’s inconsistency with the Act’s cornerstone principle that patentees have a right only to the set of elements claimed in their patents and nothing further. Third, respondents contend that patent law principles established before the enactment of the Patent Act demonstrate that a defendant that performs some steps of a patent with the purpose of having its customers perform the remaining steps is liable for inducing infringement. But here, too, the nature of the rights created by the Pat-ent Act defeats the notion that Congress could haveintended to permit inducement liability where there is no underlying direct infringement. According to respondents, their understanding of the pre-1952 doctrine casts doubt on the Muniauction rule for direct infringement under §271(a), on the ground that that rule has the indirect effect of preventing inducement liability where Congress would have wanted it. But the possibility that the Federal Circuit erred by too narrowly circumscribing the scope of §271(a) is no reason for this Court to err a second time by misconstruing §271(b) to impose liability for inducing infringement where no infringement has occurred. Finally, respondents, like the Federal Circuit, criticize our interpretation of §271(b) as permitting a would-be infringer to evade liability by dividing performance of a method patent’s steps with another whom the defendant neither directs nor controls. We acknowledge this concern. Any such anomaly, however, would result from the Fed-eral Circuit’s interpretation of §271(a) in Muniauction. A desire to avoid Muniauction’s natural consequences does not justify fundamentally altering the rules of inducement liability that the text and structure of the Patent Act clearly require—an alteration that would result in its own serious and problematic consequences, namely, creating for §271(b) purposes some free-floating concept of “infringement” both untethered to the statutory text and difficult for the lower courts to apply consistently. III Respondents ask us to review the merits of the Federal Circuit’s Muniauction rule for direct infringement under §271(a). We decline to do so today. In the first place, the question presented is clearly focused on §271(b), not §271(a). We granted certiorari on the following question: “Whether the Federal Circuit erred in holding that a defendant may be held liable for inducing patent infringement under 35 U. S. C. §271(b) even though no one has committed direct infringement under §271(a).” Pet. for Cert. i. The question presupposes that Limelight has not committed direct infringement under §271(a). And since the question on which we granted certiorari did not involve §271(a), petitioner did not address that important issue in its opening brief. Our decision on the §271(b) question necessitates a remand to the Federal Circuit, and on remand, the Federal Circuit will have the opportunity to revisit the §271(a) question if it so chooses. IV The judgment below is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 In its brief, Limelight disputes whether its customers actually “tag” within the meaning of the patent. Brief for Petitioner 7, n. 4. We assume that Limelight’s customers do in fact “tag” within the patent’s meaning. 2 The panel noted that Limelight’s contracts instruct its customers to tag the components they wish to be stored on Limelight’s CDN, but concluded that, because these contracts did not give Limelight control over its customers, the customers’ tagging could not be attributed to Limelight. See629 F. 3d, at 1321. 3 addressed contributory infringement under §271(c), rather than inducement of infringement under §271(b), but we see no basis to distinguish for these purposes between the two, which after all spring from common stock. Seev. , 563 U. S. ___, ___ (2011) (slip op., at 8). 4 Section 271(f) now prohibits the exporter’s conduct at issue in .

573.US.351

A part of the federal bank fraud statute, 18 U. S. C. §1344(2), makes it a crime to “knowingly execut[e] a scheme . . . to obtain” property owned by, or under the custody of, a bank “by means of false or fraudulent pretenses.” Petitioner Kevin Loughrin was charged with bank fraud after he was caught forging stolen checks, using them to buy goods at a Target store, and then returning the goods for cash. The District Court declined to give Loughrin’s proposed jury instruction that a conviction under §1344(2) required proof of “intent to defraud a financial institution.” The jury convicted Loughrin, and the Tenth Circuit affirmed. Held: Section 1344(2) does not require the Government to prove that a defendant intended to defraud a financial institution. Pp. 4–15. (a) Section 1344(2) requires only that the defendant intend to obtain bank property and that this end is accomplished “by means of” a false statement. No additional requirement of intent to defraud a bank appears in the statute’s text. And imposing that requirement would prevent §1344(2) from applying to cases falling within the statute’s clear terms, such as frauds directed against a third-party custodian of bank-owned property. Loughrin’s construction would also make §1344(2) a mere subset of §1344(1), which prohibits any scheme “to defraud a financial institution.” That view is untenable because those clauses are separated by the disjunctive “or,” signaling that each is intended to have separate meaning. And to read clause (1) as fully encompassing clause (2) contravenes two related interpretive canons: that different language signals different meaning, and that no part of a statute should be superfluous. Pp. 4–6. (b) Loughrin claims that his view is supported by similar language in the federal mail fraud statute and by federalism principles, but his arguments are unpersuasive. Pp. 7–15. (1) In McNally v. United States, 483 U.S. 350, this Court interpreted similar language in the mail fraud statute, §1341—which served as a model for §1344—to set forth just one offense, despite the use of the word “or.” But the two statutes have notable textual differences. The mail fraud law contains two phrases strung together in a single, unbroken sentence, whereas §1344’s two clauses have separate numbering, line breaks, and equivalent indentation—all indications of separate meaning. Moreover, Congress likely did not intend to adopt McNally’s interpretation when it enacted §1344, because at that time (three years before McNally) every Court of Appeals had interpreted the word “or” in the mail fraud statute in its usual, disjunctive sense. And while McNally found that unique features of the mail fraud statute’s history supported its view, the legislative history surrounding the adoption of §1344 points the other way. Pp. 7–9. (2) Loughrin also contends that without an element of intent to defraud a bank, §1344(2) would apply to every minor fraud in which the victim happens to pay by check. This, he says, would unduly expand the reach of federal criminal law into an area traditionally left to the States. But this argument ignores a significant textual limit on §1344(2)’s reach: The criminal must acquire (or attempt to acquire) the bank property “by means of” the misrepresentation. That language limits §1344(2)’s application to cases (like this one) in which the misrepresentation has some real connection to a federally insured bank, and thus to the pertinent federal interest. Pp. 9–15. 710 F.3d 1111, affirmed. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined, and in which Scalia and Thomas, JJ., joined as to Parts I and II, Part III–A except the last paragraph, and the last footnote of Part III–B. Scalia, J., filed an opinion concurring in part and concurring in the judgment, in which Thomas, J., joined. Alito, J., filed an opinion concurring in part and concurring in the judgment.

A provision of the federal bank fraud statute, 18 U. S. C. §1344(2), makes criminal a knowing scheme to obtain property owned by, or in the custody of, a bank “by means of false or fraudulent pretenses, representations, or promises.” The question presented is whether the Government must prove that a defendant charged with violating that provision intended to defraud a bank. We hold that the Government need not make that showing. I Petitioner Kevin Loughrin executed a scheme to convert altered or forged checks into cash. Pretending to be a Mormon missionary going door-to-door in a neighborhood in Salt Lake City, he rifled through residential mailboxes and stole any checks he found. Sometimes, he washed, bleached, ironed, and dried the checks to remove the existing writing, and then filled them out as he wanted; other times, he did nothing more than cross out the name of the original payee and add another. And when he was lucky enough to stumble upon a blank check, he completed it and forged the accountholder’s signature. Over several months, Loughrin made out six of these checks to the retailer Target, for amounts of up to $250. His modus operandi was to go to a local store and, posing as the accountholder, present an altered check to a cashier to purchase merchandise. After the cashier accepted the check (which, remarkably enough, happened time after time), Loughrin would leave the store, then turn around and walk back inside to return the goods for cash. Each of the six checks that Loughrin presented to Target was drawn on an account at a federally insured bank, including Bank of America and Wells Fargo. Employees in Target’s back office identified three of the checks as fraudulent, and so declined to submit them for payment. Target deposited the other three checks. The bank refused payment on one, after the accountholder notified the bank that she had seen a man steal her mail. Target appears to have received payment for the other two checks, though the record does not conclusively establish that fact. See Brief for United States 6, 7, n. 3. The Federal Government eventually caught up with Loughrin and charged him with six counts of committing bank fraud—one for each of the altered checks presented to Target. The federal bank fraud statute, 18 U. S. C. §1344, provides as follows: “Whoever knowingly executes, or attempts to execute, a scheme or artifice— (1) to defraud a financial institution; or (2) to obtain any of the moneys, funds, credits, assets, securities, or other property owned by, or under the custody or control of, a financial institution, by means of false or fraudulent pretenses, representations, or promises; shall be fined not more than $1,000,000 or imprisoned not more than 30 years, or both.”[1] Ruling (for a reason not material here) that Circuit precedent precluded convicting Loughrin under the statute’s first clause, §1344(1), the District Court allowed the case to go to the jury on the statute’s second, §1344(2). The court instructed the jury that it could convict Loughrin under that clause if, in offering the fraudulent checks to Target, he had “knowingly executed or at-tempted to execute a scheme or artifice to obtain money or property from the [banks on which the checks were drawn] by means of false or fraudulent pretenses, representations, or promises.” App. 7. Loughrin asked as well for another instruction: The jury, he argued, must also find that he acted with “intent to defraud a financial institution.” App. to Pet. for Cert. 43a. The court, however, declined to give that charge, and the jury convicted Loughrin on all six counts. The United States Court of Appeals for the Tenth Circuit affirmed. See 710 F. 3d 1111 (2013). As relevant here, it rejected Loughrin’s argument that “a conviction under §1344(2) requires proof that he intended to defraud the banks on which the [altered] checks had been drawn.” Id., at 1115. That intent, the court reasoned, is necessary only under the bank fraud law’s first clause. The court acknowledged that under its interpretation, §1344(2) “cast[s] a wide net for bank fraud liability,” but concluded that such a result is “dictated by the plain language of the statute.” Id., at 1117. We granted certiorari, 571 U. S. ___ (2013), to resolve a Circuit split on whether §1344(2) requires the Government to show that a defendant intended to defraud a federally insured bank or other financial institution.[2] We now affirm the Tenth Circuit’s decision. II We begin with common ground. All parties agree, as do we and the Courts of Appeals, that §1344(2) requires that a defendant “knowingly execute[ ], or attempt[ ] to execute, a scheme or artifice” with at least two elements. First, the clause requires that the defendant intend “to obtain any of the moneys . . . or other property owned by, or under the custody or control of, a financial institution.” (We refer to that element, more briefly, as intent “to obtain bank property.”) Brief for United States 11, 17, 20, 22, 32; Brief for Petitioner 30–31. And second, the clause requires that the envisioned result—i.e., the obtaining of bank property—occur “by means of false or fraudulent pretenses, representations, or promises.” See Brief for United States 21–22; Reply Brief 18–19. Loughrin does not contest the jury instructions on either of those two elements. Nor does he properly challenge the sufficiency of the evidence supporting them here.[3] The single question presented is whether the Government must prove yet another element: that the defendant intended to defraud a bank. As Loughrin describes it, that element would compel the Government to show not just that a defendant intended to obtain bank property (as the jury here found), but also that he specifically intended to deceive a bank. See Reply Brief 17. And that difference, Loughrin claims, would have mattered in this case, because his intent to deceive ran only to Target, and not to any of the banks on which his altered checks were drawn. But the text of §1344(2) precludes Loughrin’s argument. That clause focuses, first, on the scheme’s goal (obtaining bank property) and, second, on the scheme’s means (a false representation). We will later address how the “means” component of §1344(2) imposes certain inherent limits on its reach. See infra, at 11–14. But nothing in the clause additionally demands that a defendant have a specific intent to deceive a bank. And indeed, imposing that requirement would prevent §1344(2) from applying to a host of cases falling within its clear terms. In particular, the clause covers property “owned by” the bank but in someone else’s custody and control (say, a home that the bank entrusted to a real estate company after foreclosure); thus, a person violates §1344(2)’s plain text by deceiving a non-bank custodian into giving up bank property that it holds. Yet under Loughrin’s view, the clause would not apply to such a case except in the (presumably rare) circumstance in which the fraudster’s intent to deceive extended beyond the custodian to the bank itself. His proposed inquiry would thus function as an extra-textual limit on the clause’s compass. And Loughrin’s construction of §1344(2) becomes yet more untenable in light of the rest of the bank fraud statute. That is because the first clause of §1344, as all agree, includes the requirement that a defendant intend to “defraud a financial institution”; indeed, that is §1344(1)’s whole sum and substance. See Brief for United States 18; Brief for Petitioner 8. To read the next clause, following the word “or,” as somehow repeating that requirement, even while using different words, is to disregard what “or” customarily means. As we have recognized, that term’s “ordinary use is almost always disjunctive, that is, the words it connects are to be given separate meanings.” United States v. Woods, 571 U. S. ___, ___ (2013) (slip op., at 14). Yet Loughrin would have us construe the two entirely distinct statutory phrases that the word “or” joins as containing an identical element. And in doing so, his interpretation would make §1344’s second clause a mere subset of its first: If, that is, §1344(2) implicitly required intent to defraud a bank, it would apply only to conduct already falling within §1344(1). Loughrin’s construction thus effectively reads “or” to mean “including”—a definition foreign to any dictionary we know of. As that account suggests, Loughrin’s view collides as well with more general canons of statutory interpretation. We have often noted that when “Congress includes particular language in one section of a statute but omits it in another”—let alone in the very next provision—this Court “presume[s]” that Congress intended a difference in meaning. Russello v. United States, 464 U. S. 16, 23 (1983) (citation omitted). And here, as just stated, overriding that presumption would render §1344’s second clause superfluous. Loughrin’s view thus runs afoul of the “cardinal principle” of interpretation that courts “must give effect, if possible, to every clause and word of a statute.” Williams v. Taylor, 529 U. S. 362, 404 (2000) (citation omitted).[4] III Loughrin makes two principal arguments to avoid the import of the statute’s plain text. First, he relies on this Court’s construction of comparable language in the federal mail fraud statute to assert that Congress intended §1344(2) merely to explicate the scope of §1344(1)’s prohibition on scheming to defraud a bank, rather than to cover any additional conduct. And second, he contends that unless we read the second clause in that duplicative way, its coverage would extend to a vast range of fraudulent schemes, thus intruding on the historic criminal jurisdiction of the States. Neither argument is without force, but in the end, neither carries the day. A “[D]espite appearances,” Loughrin avers, §1344(2) has no independent meaning: It merely specifies part of what §1344(1) already encompasses. Brief for Petitioner 8. To support that concededly counterintuitive argument, Loughrin invokes our decision in McNally v. United States, 483 U. S. 350 (1987) , interpreting similar language in the mail fraud statute, 18 U. S. C. §1341. That law, which served as a model for §1344, see Neder v. United States, 527 U. S. 1 –21 (1999), prohibits using the mail to further “any scheme or artifice to defraud, or for obtaining money or property by means of false or fraudulent pretenses, representations, or promises.” Loughrin rightly explains that, despite the word “or,” McNally understood that provision as setting forth just one offense—using the mails to advance a scheme to defraud. The provision’s back half, we held, merely codified a prior judicial decision applying the front half: In other words, the back clarified that the front included certain conduct, rather than doing independent work. 483 U. S., at 358–359. According to Loughrin, we should read the bank fraud statute in the same way. But the two statutes, as an initial matter, have notable textual differences. The mail fraud law contains two phrases strung together in a single, unbroken sentence. By contrast, §1344’s two clauses have separate numbers, line breaks before, between, and after them, and equivalent indentation—thus placing the clauses visually on an equal footing and indicating that they have separate meanings. The legislative structure thus reinforces the usual (even if not McNally’s) understanding of the word “or” as meaning . . . well, “or”—rather than, as Loughrin would have it, “including.” Moreover, Loughrin’s reliance on McNally encounters a serious chronological problem. Congress passed the bank fraud statute in 1984, three years before we decided that case. And at that time, every Court of Appeals to have addressed the issue had concluded that the two relevant phrases of the mail fraud law must be read “in the disjunctive” and “construed independently.” 483 U. S., at 358 (citing, e.g., United States v. Clapps, 732 F. 2d 1148, 1152 (CA3 1984); United States v. States, 488 F. 2d 761, 764 (CA8 1973)). McNally disagreed, eschewing the most natural reading of the text in favor of evidence it found in the drafting history of the statute’s money-or-property clause. But the Congress that passed the bank fraud statute could hardly have predicted that McNally would overturn the lower courts’ uniform reading. We thus see no reason to doubt that in enacting §1344, Congress said what it meant and meant what it said, see Connecticut Nat. Bank v. Germain, 503 U. S. 249, 254 (1992) —i.e., that it both said “or” and meant “or” in the usual sense. And a peek at history, of the kind McNally found decisive, only cuts against Loughrin’s reading of the bank fraud statute. According to McNally, Congress added the mail fraud statute’s second, money-or-property clause merely to affirm a decision of ours interpreting the ban on schemes “to defraud”: The second clause, McNally reasoned, thus worked no substantive change in the law. See 483 U. S., at 356–359 (discussing Congress’s codification of Durland v. United States, 161 U. S. 306 (1896) ). By contrast, Congress passed the bank fraud statute to disapprove prior judicial rulings and thereby expand federal criminal law’s scope—and indeed, partly to cover cases like Loughrin’s. One of the decisions prompting enactment of the bank fraud law, United States v. Maze, 414 U. S. 395 (1974) , involved a defendant who used a stolen credit card to obtain food and lodging. (Substitute a check for a credit card and Maze becomes Loughrin.) The Government brought charges of mail fraud, relying on post-purchase mailings between the merchants and issuing bank to satisfy the statute’s mailing element. But the Court held those mailings insufficiently integral to the fraudulent scheme to support the conviction. See id., at 402. Hence, Maze created a “serious gap[ ] . . . in Federal jurisdiction over frauds against banks.” S. Rep. No. 98–225, p. 377 (1983). Congress passed §1344 to fill that gap, enabling the Federal Government to prosecute fraudsters like Maze and Loughrin. We will not deprive that enactment of its full effect because McNally relied on different history to adopt a counter-textual reading of a similar provision. B Loughrin also appeals to principles of federalism to support his proffered construction. Unless we read §1344(2) as requiring intent to defraud a bank, Loughrin contends, the provision will extend to every fraud, no matter how prosaic, happening to involve payment with a check—even when that check is perfectly valid. Consider, for example, a garden-variety con: A fraudster sells something to a customer, misrepresenting its value. There are countless variations, but let’s say the fraudster passes off a cheap knock-off as a Louis Vuitton handbag. The victim pays for the bag with a good check, which the criminal cashes. Voila!, Loughrin says, bank fraud has just happened—unless we adopt his narrowing construction. After all, the criminal has intended to “obtain . . . property . . . under the custody or control of” the bank (the money in the victim’s checking account), and has made “false or fraudulent . . . representations” (the lies to the victim about the handbag).[5] But if the bank fraud statute were to encompass all such schemes, Loughrin continues, it would interfere with matters “squarely within the traditional criminal jurisdiction of the state courts.” Brief for Petitioner 29. We should avoid such a “sweeping expansion of federal criminal” law, he concludes, by reading §1344(2), just like §1344(1), as requiring intent to defraud a bank. Reply Brief 3 (quoting Cleveland v. United States, 531 U. S. 12, 24 (2000) ). We agree with this much of what Loughrin argues: Unless the text requires us to do so, we should not construe §1344(2) as a plenary ban on fraud, contingent only on use of a check (rather than cash). As we have often (and recently) repeated, “we will not be quick to assume that Congress has meant to effect a significant change in the sensitive relation between federal and state criminal jurisdiction.” Bond v. United States, 572 U. S. ___, ___ (2014) (slip op., at 13) (quoting United States v. Bass, 404 U. S. 336, 349 (1971) ); see Cleveland, 531 U. S., at 24 (“We resist the Government’s reading . . . because it invites us to approve a sweeping expansion of federal criminal jurisdiction in the absence of a clear statement by Congress”); Jones v. United States, 529 U. S. 848, 858 (2000) (similar). Just such a rebalancing of criminal jurisdiction would follow from interpreting §1344(2) to cover every pedestrian swindle happening to involve payment by check, but in no other way affecting financial institutions. Indeed, even the Government expresses some mild discomfort with “federalizing frauds that are only tangentially related to the banking system.” Brief for United States 41. But in claiming that we must therefore recognize an invisible element, Loughrin fails to take account of a significant textual limitation on §1344(2)’s reach. Under that clause, it is not enough that a fraudster scheme to obtain money from a bank and that he make a false statement. The provision as well includes a relational component: The criminal must acquire (or attempt to acquire) bank property “by means of” the misrepresentation. That phrase typically indicates that the given result (the “end”) is achieved, at least in part, through the specified action, instrument, or method (the “means”), such that the connection between the two is something more than oblique, indirect, and incidental. See, e.g., Webster’s Third New International Dictionary 1399 (2002) (defining “by means of” as “through the instrumentality of: by the use of as a means”); 9 Oxford English Dictionary 516 (2d ed. 1989) (defining “means” as “[a]n instrument, agency, method, or course of action, by the employment of which some object is or may be attained, or which is concerned in bringing about some result”). In other words, not every but-for cause will do. If, to pick an example out of a hat, Jane traded in her car for money to take a bike trip cross-country, no one would say she “crossed the Rockies by means of a car,” even though her sale of the car somehow figured in the trip she took. The relation between those things would be (as the Government puts it) too “tangential[ ]” to make use of the phrase at all appropriate. Brief for United States 41. Section 1344(2)’s “by means of” language is satisfied when, as here, the defendant’s false statement is the mechanism naturally inducing a bank (or custodian of bank property) to part with money in its control. That occurs, most clearly, when a defendant makes a misrepresentation to the bank itself—say, when he attempts to cash, at the teller’s window, a forged or altered check. In that event, the defendant seeks to obtain bank property by means of presenting the forgery directly to a bank em-ployee. But no less is the counterfeit check the “means” of obtaining bank funds when a defendant like Loughrin offers it as payment to a third party like Target.[6] After all, a merchant accepts a check only to pass it along to a bank for payment; and upon receipt from the merchant, that check triggers the disbursement of bank funds just as if presented by the fraudster himself. So in either case, the forged or altered check—i.e., the false statement—serves in the ordinary course as the means (or to use other words, the mechanism or instrumentality) of obtaining bank property. To be sure, a merchant might detect the fraud (as Target sometimes did) and decline to submit the forged or altered check to the bank. But that is to say only that the defendant’s scheme to obtain bank property by means of a false statement may not succeed. And we have long made clear that such failure is irrelevant in a bank fraud case, because §1344 punishes not “completed frauds,” but instead fraudulent “scheme[s].” Neder, 527 U. S., at 25. By contrast, the cases Loughrin hopes will unnerve us—exemplified by the handbag swindle—do not satisfy §1344(2)’s “means” requirement.[7] Recall that in such a case the check is perfectly valid; so the check itself is not (as it was here) a false or fraudulent means of obtaining bank money. And the false pretense that has led, say, the handbag buyer to give a check to the fraudster has nothing to do with the bank that will cash it: No one would dream of passing on to the bank (as Target would forward a forged check) the lie that a knock-off is a Louis Vuitton. The bank’s involvement in the scheme is, indeed, wholly fortuitous—a function of the victim’s paying the fraudster by (valid) check rather than cash. Of course, the bank would not have disbursed funds had the misrepresentation never occurred, and in that sense, the lie counts as a but-for cause of the bank’s payment. But as we have said, §1344(2)’s “by means of” language requires more, see supra, at 11–12: It demands that the defendant’s false statement is the mechanism naturally inducing a bank (or custodian) to part with its money. And in cases like the handbag swindle, where no false statement will ever go to a financial institution, the fraud is not the means of obtaining bank property.[8] The premise of Loughrin’s federalism argument thus collapses. He claims that we must import an unstated element into §1344(2) to avoid covering run-of-the-mill frauds, properly of concern only to States. But in fact, the text of §1344(2) already limits its scope to deceptions that have some real connection to a federally insured bank, and thus implicate the pertinent federal interest. See S. Rep. No. 98–225, at 378 (noting that federal “jurisdiction is based on the fact that the victim of the offense is a federally controlled or insured institution”). And Loughrin’s own crime, as we have explained, is one such scheme, because he made false statements, in the form of forged and altered checks, that a merchant would, in the ordinary course of business, forward to a bank for payment. See supra, at 12–13. We therefore reject Loughrin’s reading of §1344(2) and his challenge to his conviction.[9] For the reasons stated, we affirm the judgment of the Tenth Circuit. It is so ordered.Notes 1 A “financial institution,” as defined in , includes a federally insured bank of the kind involved here. 2 Compare 710 F. 3d 1111, 1116 (CA10 2013) (case below) (§1344(2) does not require intent to defraud a bank); v, 270 F. 3d 986, 991 (CA6 2001) (same), with v. , 315 F. 3d 190, 197 (CA3 2002) (§1344(2) requires such intent); v. , 221 F. 3d 19, 29 (CA1 2000) (same); v. , 117 F. 3d 82, 92–93 (CA2 1997) (same). 3 Loughrin argued to the jury that the evidence failed to show that he intended to obtain bank property: He claimed that once he “obtained cash from Target, . . . he was indifferent to whether Target ever submitted the check to a bank or whether a bank ever made payment on it.” Brief for Petitioner 32; see Tr. 233–235; App. to Pet. for Cert. 46a. The jury rejected that contention, as did the District Court on a motion for judgment of acquittal. See Record 168.In his appeal, Loughrin waived the argument by conceding that if the District Court correctly instructed the jury on §1344(2)’s elements, “then there was sufficient evidence to convict.” Appellant’s Opening Brief in No. 11–4158 (CA10), p. 34. And although Loughrin’s briefs to this Court attempt to cast doubt on the jury’s finding that he intended to obtain bank property, see Brief for Petitioner 30–32, that issue is not “fairly included” in the question his certiorari petition presented, Sup. Ct. R. 14.1(a). 4 Loughrin responds that our interpretation of the statute creates a converse problem of superfluity: Clause (2), he says, would emerge so broad as to wholly swallow Clause (1). See Reply Brief 7. But that is not right. The Courts of Appeals, for example, have unanimously agreed that the Government can prosecute check kiting (,writing checks against an account with insufficient funds in a way designed to keep them from bouncing) only under Clause (1), because such schemes do not involve any false representations. See Tr. of Oral Arg. 46–47; see, v. , 969 F. 2d 425, 427–428 (CA7 1992) (citing v, –285 (1982)). No doubt, the overlap between the two clauses is substantial on our reading, but that is not uncommon in criminal statutes. See, ,v, , n. 14 (1995). 5 One might think the Federal Government would never use the bank fraud statute to prosecute such ordinary frauds just because they happen to involve payment by check rather than cash. But in fact, the Government has brought a number of cases alleging violations of §1344(2) on that theory (so far, it appears, unsuccessfully). See, , 315 F. 3d 190 (a home health care worker got a valid check from a patient to buy groceries, but then cashed the check and pocketed the money); v. , 140 F. 3d 163 (CA2 1998) (an employee filed fake invoices with her employer, causing the company to issue valid checks to her friend for services never rendered). 6 The Government in such a case may, of course, face the separate claim that the defendant did not intend to obtain bank property at all: As noted earlier, Loughrin argued this point to the jury, contending (unsuccessfully) that he merely wanted to get cash from Target. See n. 3, . All we say here, for the reasons next stated, is that when the defendant has the requisite intent to acquire bank property, his presentation of a forged or altered check to a third party satisfies §1344(2)’s “means” requirement. 7 Even the Government, we note, acknowledges that §1344(2) is reasonably read to exclude such cases from its coverage. See Brief for United States 40–44; Tr. of Oral Arg. 43–47. 8 takes issue with our limitation of §1344(2), contending first that the fraudster’s “indifferen[ce] to the victim’s method of payment” does not “cause what is a means not to be a means.” ,at 2–3 (opinion concurring in part and concurring in judgment) (emphasis deleted). To illustrate the point, he offers an example: Someone “obtain[s] 7-Eleven coffee by means of [his] two dollars” even if he went to 7-Eleven rather than Sheetz only because it happened to be the closest. at 3. But that objection is based on a misunderstanding of our opinion. The “by means of” phrase calls for an inquiry into the directness of the relationship between means and ends, not the fraudster’s subjective intent. (We take it agrees; he recognizes that “not every but-for cause of an act is a cause ‘by means of’ which the act has occurred.” ,at 2.) And we concur with the bottom line of sexample: There, the means (the two dollars) is the thing that achieves the specified end (getting the cup of 7-Eleven coffee). By contrast, for the reasons elaborated above, the misstatement in our handbag hypothetical is not the mechanism by which the fraudster obtains property, given that the lie will never reach the bank. 9 As a last-gasp argument, Loughrin briefly asserts that §1344(2) at least requires the Government to prove that the defendant’s scheme created a risk of financial loss to the bank. See Brief for Petitioner 36–40. But once again, nothing like that element appears in the clause's text. Indeed, the broad language in §1344(2) describing the property at issue—“property owned by or under the custody or control of” a bank—appears calculated to avoid entangling courts in technical issues of banking law about whether the financial institution or, alternatively, a depositor would suffer the loss from a successful fraud. See v. , 699 F. 3d 743, 754 (CA2 2012) (Lynch, J., concurring in part and concurring in judgment in part). And Loughrin’s argument fits poorly with our prior holding that the gravamen of §1344 is the “scheme,” rather than “the completed fraud,” and that the offense therefore does not require “damage” or “reliance.” v. , ; see at 13.

572.US.1

When one parent abducts a child and flees to another country, the other parent may file a petition in that country for the return of the child pursuant to the Hague Convention on the Civil Aspects of International Child Abduction (Hague Convention or Convention). If the parent files a petition within one year of the child’s removal, a court “shall order the return of the child forthwith.” But when the petition is filed after the 1-year period expires, the court “shall . . . order the return of the child, unless it is demonstrated that the child is now settled in its new environment.” Respondent Montoya Alvarez and petitioner Lozano resided with their daughter in London until November 2008, when Montoya Alvarez left with the child for a women’s shelter. In July 2009, Montoya Alvarez and the child left the United Kingdom and ultimately settled in New York. Lozano did not locate Montoya Alvarez and the child until November 2010, more than 16 months after Montoya Alvarez and the child had left the United Kingdom. At that point, Lozano filed a Petition for Return of Child pursuant to the Hague Convention in the Southern District of New York. Finding that the petition was filed more than one year after removal, the court denied the petition on the basis that the child was now settled in New York. It also held that the 1-year period could not be extended by equitable tolling. The Second Circuit affirmed. Held: Article 12’s 1-year period is not subject to equitable tolling. Pp. 7–16. (a) The doctrine of equitable tolling, as applied to federal statutes of limitations, extends an otherwise discrete limitations period set by Congress. Thus, whether tolling is available is fundamentally a question of statutory intent. Because Congress “legislate[s] against a background of common-law adjudicatory principles,” Astoria Fed. Sav. & Loan Assn. v. Solimino, 501 U.S. 104, 108, including equitable tolling, see Holmberg v. Armbrecht, 327 U.S. 392, 397, equitable tolling is presumed to apply if the period in question is a statute of limitations and if tolling is consistent with the statute, Young v. United States, 535 U.S. 43, 49–50. Pp. 7–8. (b) In assessing whether equitable tolling applies to treaties, which are “ ‘compact[s] between independent nations,’ ” Medellín v. Texas, 552 U.S. 491, 505, this Court’s “duty [i]s to ascertain the intent of the parties” by looking to the document’s text and context, United States v. Choctaw Nation, 179 U.S. 494, 535. The parties to the Hague Convention did not intend equitable tolling to apply to Article 12’s 1-year period. Pp. 8–16. (1) There is no general presumption that equitable tolling applies to treaties. Though part of the established backdrop of American law, equitable tolling has no proper role in the interpretation of treaties unless that principle is shared by the parties to the “agreement among sovereign powers,” Zicherman v. Korean Air Lines Co., 516 U.S. 217, 226. Lozano has identified no such shared principle among the Convention signatories, and the courts of several signatories have explicitly rejected equitable tolling of the Convention. Thus, the American presumption does not apply to this multilateral treaty. The International Child Abduction Remedies Act, 42 U. S. C. §§11601–11610, which Congress enacted to implement the Convention, neither addresses the availability of equitable tolling nor purports to alter the Convention, and therefore does not affect this conclusion. Pp. 9–11. (2) Even if the Convention were subject to a presumption that statutes of limitations may be tolled, Article 12’s 1-year period is not a statute of limitations. Statutes of limitations embody a “policy of repose, designed to protect defendants,” Burnett v. New York Central R. Co., 380 U.S. 424, 428, and foster the “elimination of stale claims, and certainty about a plaintiff’s opportunity for recovery and a defendant’s potential liabilities,” Rotella v. Wood, 528 U.S. 549, 555. Here, the remedy the Convention affords the left-behind parent—return of the child—continues to be available after one year, thus preserving the possibility of relief for that parent and preventing repose for the abducting parent. The period’s expiration also does not establish certainty about the parties’ respective rights. Instead, it opens the door to consideration of a third party’s interests, i.e., the child’s interest in settlement. Because that is not the sort of interest addressed by a statute of limitations, the 1-year period should not be treated as a statute of limitations. Young, supra, at 47, distinguished. Pp. 11–13. (3) Without a presumption of equitable tolling, the Convention does not support extending the 1-year period during concealment. Article 12 explicitly provides for the period to commence on “the date of the wrongful removal or retention” and makes no provision for an extension. Because the drafters did not choose to delay the period’s commencement until discovery of the child’s location—the obvious alternative to the date of wrongful removal—the natural implication is that they did not intend to commence the period on that later date. Lozano contends that equitable tolling is nonetheless consistent with the Convention’s goal of deterring child abductions, but the Convention does not pursue that goal at any cost, having recognized that the return remedy may be overcome by, e.g., the child’s interest in settlement. And the abducting parent does not necessarily profit by running out the clock, since both American courts and other Convention signatories have considered concealment as a factor in determining whether a child is settled. Equitable tolling is therefore neither required by the Convention nor the only available means to advance its objectives. Pp. 13–15. (4) Lozano contends that there is room for United States courts to apply equitable tolling because the Convention recognizes that other sources of law may permit signatory states to return abducted children even when return is not available or required by the Convention. But this contention mistakes the nature of equitable tolling, which may be applied to the Hague Convention only if the treaty drafters so intended. For the foregoing reason, they did not. Pp. 15–16. 697 F.3d 41, affirmed. Thomas, J., delivered the opinion for a unanimous Court. Alito, J., filed a concurring opinion, in which Breyer and Sotomayor, JJ., joined.

When a parent abducts a child and flees to another country, the Hague Convention on the Civil Aspects of International Child Abduction generally requires that country to return the child immediately if the other parent requests return within one year. The question in this case is whether that 1-year period is subject to equitable tolling when the abducting parent conceals the child’s location from the other parent. We hold that equitable tolling is not available. I To address “the problem of international child abductions during domestic disputes,” Abbott v. Abbott, 560 U. S. 1, 8 (2010) , in 1980 the Hague Conference on Private International Law adopted the Convention on the Civil Aspects of International Child Abduction (Hague Convention or Convention), T. I. A. S. No. 11670, S. Treaty Doc. No. 99–11 (Treaty Doc.). The Convention states two primary objectives: “to secure the prompt return of children wrongfully removed to or retained in any Contracting State,” and “to ensure that rights of custody and of access under the law of one Contracting State are effectively respected in the other Contracting States.” Art. 1, id.,at 7. To those ends, the Convention’s “central operating feature” is the return of the child. Abbott, 560 U. S., at 9. That remedy, in effect, lays venue for the ultimate custody determination in the child’s country of habitual residence rather than the country to which the child is abducted. See id., at 20 (“The Convention is based on the principle that the best interests of the child are well served when de-cisions regarding custody rights are made in the countryof habitual residence”). The return remedy is not absolute. Article 13 excuses return where, for example, the left-behind parent was not “actually exercising” custody rights when the abducting parent removed the child, or where there is a “grave risk” that return would “place the child in an intolerable situation.” Hague Convention, Arts. 13(a)–(b), Treaty Doc., at 10. A state may also refuse to return the child if doingso would contravene “fundamental principles . . . relating to the protection of human rights and fundamental freedoms.” Art. 20, id., at 11. This case concerns another exception to the return remedy. Article 12 of the Convention states the general rule that when a court receives a petition for return within one year after the child’s wrongful removal, the court “shall order the return of the child forthwith.” Id., at 9. Article 12 further provides that the court, “where the proceedings have been commenced after the expiration of the period of one year [from the date of the wrongful removal], shall also order the return of the child, unless it is demonstrated that the child is now settled in its new environment.” Ibid. Thus, at least in some cases, failure to file a petitionfor return within one year renders the return remedy unavailable. The United States ratified the Hague Convention in 1988, and Congress implemented the Convention that same year through the International Child Abduc-tion Remedies Act (ICARA). 102Stat. 437, 42 U. S. C. §§11601–11610. That statute instructs courts to “decide the case in accordance with the Convention.” §11603(d). Echoing the Convention, ICARA further provides that “[c]hildren who are wrongfully removed . . . are to be promptly returned unless one of the narrow exceptions set forth in the Convention applies.” §11601(a)(4). Finally, ICARA requires the abducting parent to establish by a preponderance of the evidence that Article 12’s exception to return applies. §11603(e)(2)(B). II Diana Lucia Montoya Alvarez and Manuel Jose Lozano are the parents of the girl at the center of this dispute.[1] Montoya Alvarez and Lozano met and began dating in London in early 2004. Montoya Alvarez gave birth to a daughter in October 2005. Montoya Alvarez and Lozano describe their relationship in starkly different terms. Lozano stated that they were “ ‘very happy together,’ ” albeit with “normal couple problems.” In re Lozano, 809 F. Supp. 2d 197, 204 (SDNY 2011). Montoya Alvarez described a pattern of physical and emotional abuse that included multiple incidents of rape and battery. The District Court found insufficient evidence to make specific findings about domestic violence but determined that Lozano’s claim that he never mistreated Montoya Alvarez was “not credible.” Id., at 206. The parties also differ as to the child’s well-being during the first three years of her life. Lozano stated that he and the child had a very good relationship, and that the child was generally happy. Montoya Alvarez believed otherwise. In October 2008, Montoya Alvarez reported to the child’s doctor that she refused to speak at the nursery she attended, cried often, and wet the bed. Montoya Alvarez also stated that the child refused to speak when Lozano was present. The child’s nursery manager wrote that the girl was “ ‘very withdrawn,’ ” and noted that the home “ ‘environment obviously had a negative effect’ ” on her. Id., at 207. The District Court found insufficient evidence that Lozano had physically abused the child, but did conclude that the child had seen and heard her parents arguing at home. In November 2008, when the child was just over three years old, Montoya Alvarez went to New York to visit her sister Maria. During that time, the child remained in London with Lozano and his visiting mother. When Montoya Alvarez returned on November 18, she became acutely concerned about the child’s fearful behavior around Lo-zano. The next day, Montoya Alvarez left with the child and never returned. Montoya Alvarez and the child lived at a women’s shelter for the next seven months. After Montoya Alvarez was unable to find suitable long-term accommodations in the United Kingdom, she and the child left for France on July 3, 2009, and then for the United States, arriving five days later. Since their arrival, Montoya Alvarez and the child have lived with Montoya Alvarez’ sister Maria and her family in New York. When they arrived in New York, Montoya Alvarez and the child began seeing a therapist at a family medical clinic. The therapist testified that, at first, the child was withdrawn and would wet herself. The therapist diagnosed her with posttraumatic stress disorder. Within six months, however, the therapist described her as “ ‘a completely different child,’ ” who had stopped wetting herself, was excited to play with friends, and was able to speak freely about her emotions. Id., at 212. When Montoya Alvarez and the child returned to the therapist after Lo-zano filed a petition for the child’s return, the therapist noted that the child was doing well but did not wish to see her father. In the meantime, Lozano attempted to find Montoya Alvarez and the child. Shortly after Montoya Alvarez left in November 2008, he called her sister Gloria in London, but eventually received legal advice not to speak with Montoya Alvarez’ family. A mediation service also sent several letters to Montoya Alvarez on Lozano’s behalf without receiving a response. In July 2009, Lozano filed an application for a court order in the United Kingdom “ ‘to ensure that he obtains regular contact with his [child] and plays an active role in [her] life.’ ” Id., at 210. He also sought court orders to compel Montoya Alvarez’ sisters and legal counsel, the child’s doctor and nursery, and various government offices in London to disclose the child’s whereabouts. On March 15, 2010, after determining that the child was not in the United Kingdom (and suspecting that the child was in New York), Lozano filed a form with the Hague Convention Central Authority for England and Wales seeking to have the child returned.[2] The United States Central Authority—the Office of Children’s Issues in the Department of State, see 22 CFR §94.2 (2013)—received the application on March 23, 2010. After the Office of Children’s Issues confirmed that Montoya Alvarez had entered the United States, Lozano located Montoya Alvarez’ address in New York. On November 10, 2010, more than 16 months after Montoya Alvarez and the child left the United Kingdom, Lozano filed a Petition for Return of Child pursuant to the Hague Convention and ICARA, 42 U. S. C. §11603, in the United States District Court for the Southern District of New York. After a 2-day evidentiary hearing, the District Court denied Lozano’s petition. 809 F. Supp. 2d 197. The District Court concluded that Lozano had stated a prima facie case of wrongful removal under the Hague Convention. Id., at 219–220. Prior to her removal, the child was a habitual resident of the United Kingdom, see Hague Convention, Art. 4, and Lozano had custody rights that hewas actually exercising at the time of removal, see Arts. 3(a)–(b). Because the petition was filed more than one year after the child’s wrongful removal, however, the District Court denied the petition on the basis that the child was now settled in New York. Id., at 230, 234. “Viewing the total-ity of the circumstances,” the court found sufficient indicia of “stability in her family, educational, social, and most importantly, home life,” id., at 233, to conclude that the child was settled in her current environment and that repatriation would be “extremely disruptive,” id., at 234. Lozano argued that the child should be returned forthwith because the 1-year period in Article 12 should be equitably tolled during the period that Montoya Alvarez concealed the child. The court rejected that argument, holding that the 1-year period could not be extended by equitable tolling.[3] Id., at 228–229. On appeal, the Second Circuit affirmed. 697 F. 3d 41 (2012). The Court of Appeals agreed that the 1-year per-iod in Article 12 is not subject to equitable tolling. According to the court, unlike a statute of limitations that would prohibit the filing of a return petition after one year, the1-year period in Article 12 merely permits courts, after that period has run, to consider the interests of the child in settlement. Id., at 52. The Second Circuit concluded that allowing equitable tolling to delay consideration of the child’s interests would undermine the purpose of the Hague Convention. Id., at 54. We granted certiorari to decide whether Article 12’s1-year period is subject to equitable tolling. 570 U. S. ___ (2013). Compare 697 F. 3d, at 50–55 (equitable tolling not available); and Yaman v. Yaman, 730 F. 3d 1, 12–16 (CA1 2013) (same), with Duarte v. Bardales, 526 F. 3d 563, 568–570 (CA9 2008) (equitable tolling available); and Furnes v. Reeves, 362 F. 3d 702, 723–724 (CA11 2004) (same). We hold that equitable tolling is not available, and therefore affirm. III Although this case concerns the application of equitable tolling to a treaty, we begin with a more familiar context: equitable tolling of federal statutes of limitations. As a general matter, equitable tolling pauses the running of, or “tolls,” a statute of limitations when a litigant has pursued his rights diligently but some extraordinary circumstance prevents him from bringing a timely action. See, e.g., Pace v. DiGuglielmo, 544 U. S. 408, 418 (2005) . Because the doctrine effectively extends an otherwise discrete limitations period set by Congress, whether equitable tolling is available is fundamentally a question of statutory intent. See, e.g., Irwin v. Department of Veterans Affairs, 498 U. S. 89, 95 (1990) ; Bowen v. City of New York, 476 U. S. 467 –480 (1986); Honda v. Clark, 386 U. S. 484, 501 (1967) . As applied to federal statutes of limitations, the inquiry begins with the understanding that Congress “legislate[s] against a background of common-law adjudicatory principles.” Astoria Fed. Sav. & Loan Assn. v. Solimino, 501 U. S. 104, 108 (1991) . Equitable tolling, a long-established feature of American jurisprudence derived from “the old chancery rule,” Holmberg v. Armbrecht, 327 U. S. 392, 397 (1946) , is just such a principle. See Young v. United States, 535 U. S. 43 –50 (2002) (“Congress must be presumed to draft limitations periods in light of this background principle”); Bailey v. Glover, 21 Wall. 342, 349–350 (1875). We therefore presume that equitable tolling applies if the period in question is a statute of limitations and if tolling is consistent with the statute. Young, supra, at 49–50 (“It is hornbook law that limitations periods are ‘customarily subject to “equitable tolling,” ’ unless tolling would be ‘inconsistent with the text of the relevant statute’ ” (citation omitted)). IV The Hague Convention, of course, is a treaty, not a federal statute. For treaties, which are primarily “ ‘compact[s] between independent nations,’ ” Medellín v. Texas, 552 U. S. 491, 505 (2008) , our “duty [i]s to ascertain the intent of the parties” by looking to the document’s text and context, United States v. Choctaw Nation, 179 U. S. 494, 535 (1900) ; see also BG Group plc v. Republic of Argen-tina, post, at 10. We conclude that the parties to the Hague Convention did not intend equitable tolling to apply to the 1-year period in Article 12. Unlike federal statutes of limitations, the Convention was not adopted against a shared background of equitable tolling. Even if the Convention were subject to a presumption that statutes of limitations may be tolled, the 1-year period in Article 12 is not a statute of limitations. And absent a presumption in favor of equitable tolling, nothing in the Convention warrants tolling the 1-year period. A First, there is no general presumption that equitable tolling applies to treaties. Congress is presumed to incorporate equitable tolling into federal statutes of limitations because equitable tolling is part of the established backdrop of American law. Rotella v. Wood, 528 U. S. 549, 560 (2000) (“[F]ederal statutes of limitations are generally subject to equitable principles of tolling”). It does not follow, however, that we can export such background principles of United States law to contexts outside their jurisprudential home. It is particularly inappropriate to deploy this background principle of American law automatically when interpreting a treaty. “A treaty is in its nature a contract between . . . nations, not a legislative act.” Foster v. Neilson, 2 Pet. 253, 314 (1829) (Marshall, C. J., for the Court); see also 2 Debates on the Federal Constitution 506 (J. Elliot 2d ed. 1863) (James Wilson) (“[I]n their nature treaties originate differently from laws. They are made by equal parties, and each side has half of the bargain to make . . . ”). That distinction has been reflected in the way we interpret treaties. It is our “responsibility to read the treaty in a manner ‘consistent with the shared expectations of the contracting parties.’ ” Olympic Airways v. Husain, 540 U. S. 644, 650 (2004) (quoting Air France v. Saks, 470 U. S. 392, 399 (1985) ; emphasis added). Even if a background principle is relevant to the interpretation of federal statutes, it has no proper role in the interpretation of treaties unless that principle is shared by the parties to “an agreement among sovereign powers,” Zicherman v. Korean Air Lines Co., 516 U. S. 217, 226 (1996) . Lozano has not identified a background principle of equitable tolling that is shared by the signatories to the Hague Convention. To the contrary, Lozano concedes that in the context of the Convention, “foreign courts have failed to adopt equitable tolling . . . because they lac[k] the presumption that we [have].” Tr. of Oral Arg. 19–20. While no signatory state’s court of last resort has resolved the question, intermediate courts of appeals in several states have rejected equitable tolling. See Cannon v. Cannon, [2004] EWCA (Civ) 1330, [2005] 1 W. L. R. 32, ¶51 (Eng.), (rejecting the “tolling rule” as “too crude an approach” for the Convention); Kubera v. Kubera, 3 B. C. L. R. (5th) 121, ¶64, 317 D. L. R. (4th) 307, ¶64 (2010) (Can.) (equitable tolling “has not been adopted in other jurisdictions, including Canada”); see also HJ v. Secretary for Justice, [2006] NZFLR 1005, ¶53 (CA), appeal dism’d on other grounds, [2007] 2 NZLR 289; A. C. v. P. C., [2005] HKEC 839, 2005 WL 836263, ¶55, (Hong Kong Ct. 1st Instance).[4] The American presumption that federal statutes of limitations can be equitably tolled therefore does not apply to this multilateral treaty. Cf. Eastern Airlines, Inc. v. Floyd, 499 U. S. 530 –545, and n. 10 (1991) (declining to adopt liability for psychic injury under the Warsaw Convention because “the unavailability of compensation for purely psychic injury in many common and civil law countries at the time of the Warsaw Conference persuades us that the signatories had no specific intentto include such a remedy in the Convention” (footnote omitted)). It does not matter to this conclusion that Congress enacted a statute to implement the Hague Convention. See ICARA, 42 U. S. C. §§11601–11610. ICARA does not address the availability of equitable tolling. Nor does it purport to alter the Convention. See §11601(b)(2) (“The provisions of [ICARA] are in addition to and not in lieu of the provisions of the Convention”). In fact, Congress explicitly recognized “the need for uniform international interpretation of the Convention.” §11601(b)(3)(B). Congress’ mere enactment of implementing legislation did not somehow import background principles of American law into the treaty interpretation process, thereby altering our understanding of the treaty itself. B Even if the presumption in favor of equitable tolling had force outside of domestic law, we have only applied that presumption to statutes of limitations. See Hallstrom v. Tillamook County, 493 U. S. 20, 27 (1989) (no equitable tolling of a 60-day presuit notice requirement that does not operate as a statute of limitations). The 1-year period in Article 12 is not a statute of limitations. As a general matter, “[s]tatutes of limitations establish the period of time within which a claimant must bring an action.” Heimeshoff v. Hartford Life & Accident Ins. Co., 571 U. S. ___, ___ (2013) (slip op., at 4). They characteristically embody a “policy of repose, designed to protect defendants.” Burnett v. New York Central R. Co., 380 U. S. 424, 428 (1965) . And they foster the “elimination of stale claims, and certainty about a plaintiff’s opportunity for recovery and a defendant’s potential liabilities.” Rotella, supra, at 555. In Young, 535 U. S. 43 , we evaluated whether those characteristics of statutes of limitations were present in the “three-year lookback period” for tax liabilities in bankruptcy proceedings. The Bankruptcy Code favors tax claims less than three years old in two respects: Such claims cannot be discharged, and they have priority over certain others in bankruptcy proceedings. See 11 U. S. C. §§507(a)(8)(A)(i), 523(a)(1)(A). If the Internal Revenue Service “sleeps on its rights” by failing to prosecute those claims within three years, however, then those mechanisms for enforcing claims against bankrupt taxpayers are eliminated. Young, 535 U. S., at 47. We concluded that the lookback period “serves the same ‘basic policies [furthered by] all limitations provisions,’ ” ibid. (quoting Ro-tella, 528 U. S., at 555), i.e., certainty and repose. We accordingly held that it was a limitations periodpresumptively subject to equitable tolling. 535 U. S.,at 47. Unlike the 3-year lookback period in Young, expiration of the 1-year period in Article 12 does not eliminate the remedy the Convention affords the left-behind parent—namely, the return of the child. Before one year has elapsed, Article 12 provides that the court “shall order the return of the child forthwith.” Treaty Doc., at 9. But even after that period has expired, the court “shall also order the return of the child, unless it is demonstrated that the child is now settled.” Ibid. The continued availability of the return remedy after one year preserves the possibility of relief for the left-behind parent and prevents reposefor the abducting parent.[5] Rather than establishing any certainty about the respective rights of the parties, the expiration of the 1-year period opens the door to consideration of a third party’s interests, i.e., the child’s interest in settlement. Because that is not the sort of interest addressed by a statute of limitations, we decline to treat the 1-year period as a statute of limitations.[6] C Without a presumption of equitable tolling, the Convention does not support extending the 1-year period during concealment. Article 12 explicitly provides that the 1-year period commences on “the date of the wrongful removal or retention,” and makes no provision for an extension of that period. Id., at 9. Further, the practical effect of the tolling that Lozano requests would be to delay the commencement of the 1-year period until the left-behind parent discovers the child’s location. Commencing the 1-year period upon discovery is the obvious alternative to the commencement date the drafters actually adopted because the subject of the Hague Convention, child abduction, is naturally associated with the sort of concealment that might justify equitable tolling under other circumstances. See 697 F. 3d, at 51, n. 8 (“It would have been a simple matter, if the state parties to the Convention wished to take account of the possibility that an abducting parent might make it difficult for the petitioning parent to discover the child’s whereabouts, to run the period ‘from the date that the petitioning parent learned [or, could reasonably have learned] of the child’s whereabouts’ ” (alterations in original)). Given that the drafters did not adopt that alternative, the natural implication is that they did not intend the 1-year period to commence on that later date. Cf. Sebelius v. Auburn Regional Medical Center, 568 U. S. ___, ___ (2013) (slip op., at 10–11). We cannot revisit that choice. Lozano contends that equitable tolling is nevertheless consistent with the purpose of the Hague Convention because it is necessary to deter child abductions. In his view, “absent equitable tolling, concealment ‘probably will’ result in non-return,” which will in turn encourage abduction. Reply Brief 14–15; see also Duarte, 526 F. 3d, at 570. We agree, of course, that the Convention reflects a design to discourage child abduction. But the Convention does not pursue that goal at any cost. The child’s interest in choosing to remain, Art. 13, or in avoiding physical or psychological harm, Art. 13(b), may overcome the return remedy. The same is true of the child’s interest in settlement. See supra, at 2; see also In re M, [2008] 1 A. C. 1288, 1310 (Eng. 2007) (opinion of Baroness Hale of Richmond) (“These children should not be made to suffer for the sake of general deterrence of the evil of child abduction world wide”). We are unwilling to apply equitable tolling principles that would, in practice, rewrite the treaty. See Chan v. Korean Air Lines, Ltd., 490 U. S. 122 –135 (1989) (“ ‘[T]o alter, amend, or add to any treaty by inserting any clause, whether small or great, importantor trivial, would be . . . to make, and not to construe a treaty’ ” (quoting The Amiable Isabella, 6 Wheat. 1, 71 (1821) (Story, J., for the Court))). Nor is it true that an abducting parent who conceals a child’s whereabouts will necessarily profit by running out the clock on the 1-year period. American courts have found as a factual matter that steps taken to promote concealment can also prevent the stable attachments that make a child “settled.” See, e.g., Mendez Lynch v. Mendez Lynch, 220 F. Supp. 2d 1347, 1363–1364 (MD Fla. 2002) (children not settled when they “lived in seven different locations” in 18 months); Wigley v. Hares, 82 So. 3d 932, 942 (Fla. App. 2011) (“The mother purposely kept him out of all community activities, sports, and even church to avoid detection by the father”); In re Coffield, 96 Ohio App. 3d 52, 58, 644 N. E. 2d 662, 666 (1994) (child not settled when the abducting parent “was attempting to hide [child’s] identity” by withholding child from school and other organized activities). Other signatories to the Hague Convention have likewise recognized that concealment may be taken into account in the factual determination whether the child is settled. See, e.g., Cannon, [2005] 1 W. L. R., ¶¶52–61. See also Kubera, 3 B. C. L. R. (5th), ¶47, 317 D. L. R. (4th), ¶47; A. C. v. P. C., [2005] HKEC 839, ¶39, 2005 WL 836263, ¶39. Equitable tolling is therefore neither required by the Convention nor the only available means to advance its objectives. D Finally, Lozano contends that the Hague Convention leaves room for United States courts to apply their own “common law doctrine of equitable tolling” to the 1-year period in Article 12 without regard to whether the drafters of the Convention intended equitable tolling to apply. Brief for Petitioner 25. Specifically, Lozano contends that the Convention recognizes additional sources of law that permit signatory states to return abducted children even when return is not available or required pursuant to the Convention. Article 34 of the Convention provides that “for the purpos[e] of obtaining the return of a child,” the Convention “shall not restrict the application of an international instrument in force between the State of origin and the State addressed” or the application of “other law of the State addressed.” Treaty Doc., at 13; see also Art. 18, id., at 11 (“The provisions of this Chapter do not limit the power of a judicial or administrative authority to order the return of the child at any time”). In Lozano’s view, equitable tolling principles constitute “other law” that should apply here. That contention mistakes the nature of equitable tolling as this Court has applied it. We do not apply equitable tolling as a matter of some independent authority to reconsider the fairness of legislative judgments balancing the needs for relief and repose. See supra, at 7–8. To the contrary, we may apply equitable tolling to the Hague Convention only if we determine that the treaty drafters so intended. See Choctaw Nation, 179 U. S., at 535. For the foregoing reasons, we conclude that they did not. V The Court of Appeals correctly concluded that the 1-year period in Article 12 of the Hague Convention is not subject to equitable tolling. We therefore affirm that court’s judgment. It is so ordered. Notes 1 Except where otherwise noted, the facts are taken from the District Court’s findings. Like the courts below, we refer to Montoya Alvarez and Lozano’s daughter as “the child” to protect her identity. 2 Article 6 of the Hague Convention requires each Contracting State to “designate a Central Authority to discharge the duties which are imposed by the Convention upon such authorities.” Treaty Doc., at 8. 3 The District Court held in the alternative that even if equitable tolling could apply, it would not be warranted in this case because Lozano had contact information for Montoya Alvarez’ sister Maria in New York. Lozano’s solicitors did not attempt to contact Maria to determine if Montoya Alvarez and the child were there. 809 F. Supp. 2d, at 229–230. 4 Lozano contends that a single-judge decision by an English family court adopted equitable tolling without referring to it by name. See ,[2000] 2 F. L. R. 51, [2000] 3 F. C. R. 404 (Eng.). It is unclear whether the logic of that decision survived the decision of the Court of Appeals for England and Wales in . 5 In the State Department’s view, the Hague Convention confers equitable discretion on courts to order the return of a child even if the court determines that the child is “settled” within the meaning of Article 12. See Brief for United States as 19–25. If accurate, that interpretation would reinforce that Article 12 is not meant to provide repose to the abducting parent, and it would underscore that the 1-year period is not a statute of limitations. But we do not decide whether, and under what circumstances, a court may exercise discretion to order return notwithstanding the child’s subsequent settlement. In the Court of Appeals, Lozano failed to challenge the District Court’s decision not to exercise its discretion to order the return of the settled child, see n. 3, and that issue is beyond the scope of the question presented before this Court. 6 Lozano argues that the United States delegation referred to the1-year period as a “statute of limitations” at various points during and after the drafting process. Brief for Petitioner 27–28. Because the determination whether the 1-year period is a statute of limitations depends on its functional characteristics, it is not significant that the delegation used that label. In any event, we doubt that the remarks of a single delegation are sufficient under these circumstances to establish the “ ‘shared expectations of the contracting parties.’ ” v. , (quoting v. , ).

573.US.464

In 2007, Massachusetts amended its Reproductive Health Care Facilities Act, which had been enacted in 2000 to address clashes between abortion opponents and advocates of abortion rights outside clinics where abortions were performed. The amended version of the Act makes it a crime to knowingly stand on a “public way or sidewalk” within 35 feet of an entrance or driveway to any “reproductive health care facility,” defined as “a place, other than within or upon the grounds of a hospital, where abortions are offered or performed.” Mass. Gen. Laws, ch. 266, §§120E½(a), (b). The Act exempts from this prohibition four classes of individuals, including “employees or agents of such facility acting within the scope of their employment.” §120E½(b)(2). Another provision of the Act proscribes the knowing obstruction of access to an abortion clinic. §120E½(e). McCullen and the other petitioners are individuals who attempt to engage women approaching Massachusetts abortion clinics in “sidewalk counseling,” which involves offering information about alternatives to abortion and help pursuing those options. They claim that the 35-foot buffer zones have displaced them from their previous positions outside the clinics, considerably hampering their counseling efforts. Their attempts to communicate with patients are further thwarted, they claim, by clinic “escorts,” who accompany arriving patients through the buffer zones to the clinic entrances. Petitioners sued Attorney General Coakley and other Commonwealth officials, seeking to enjoin the Act’s enforcement on the ground that it violates the First and Fourteenth Amendments, both on its face and as applied to them. The District Court denied both challenges, and the First Circuit affirmed. With regard to petitioners’ facial challenge, the First Circuit held that the Act was a reasonable “time, place, and manner” regulation under the test set forth in Ward v. Rock Against Racism, 491 U.S. 781. Held: The Massachusetts Act violates the First Amendment. Pp. 8–30. (a) By its very terms, the Act restricts access to “public way[s]” and “sidewalk[s],” places that have traditionally been open for speech activities and that the Court has accordingly labeled “traditional public fora,” Pleasant Grove City v. Summum, 555 U.S. 460, 469. The government’s ability to regulate speech in such locations is “very limited.” United States v. Grace, 461 U.S. 171, 177. “[E]ven in a public forum,” however, “the government may impose reasonable restrictions on the time, place, or manner of protected speech, provided the restrictions ‘are justified without reference to the content of the regulated speech, that they are narrowly tailored to serve a significant governmental interest, and that they leave open ample alternative channels for communication of the information,’ ” Ward, supra, at 791. Pp. 8–10. (b) Because the Act is neither content nor viewpoint based, it need not be analyzed under strict scrutiny. Pp. 10–18. (1) The Act is not content based simply because it establishes buffer zones only at abortion clinics, as opposed to other kinds of facilities. First, the Act does not draw content-based distinctions on its face. Whether petitioners violate the Act “depends” not “on what they say,” Holder v. Humanitarian Law Project, 561 U.S. 1, 27, but on where they say it. Second, even if a facially neutral law disproportionately affects speech on certain topics, it remains content neutral so long as it is “ ‘justified without reference to the content of the regulated speech.’ ” Renton v. Playtime Theatres, Inc., 475 U.S. 41, 48. The Act’s purposes include protecting public safety, patient access to healthcare, and unobstructed use of public sidewalks and streets. The Court has previously deemed all these concerns to be content neutral. See Boos v. Barry, 485 U.S. 312, 321. An intent to single out for regulation speech about abortion cannot be inferred from the Act’s limited scope. “States adopt laws to address the problems that confront them.” Burson v. Freeman, 504 U.S. 191, 207. There was a record of crowding, obstruction, and even violence outside Massachusetts abortion clinics but not at other kinds of facilities in the Commonwealth. Pp. 11–15. (2) The Act’s exemption for clinic employees and agents acting within the scope of their employment does not appear to be an attempt to favor one viewpoint about abortion over the other. City of Ladue v. Gilleo, 512 U.S. 43, 51, distinguished. Given that some kind of exemption was necessary to allow individuals who work at the clinics to enter or remain within the buffer zones, the “scope of employment” qualification simply ensures that the exemption is limited to its purpose of allowing the employees to do their jobs. Even assuming that some clinic escorts have expressed their views on abortion inside the zones, the record does not suggest that such speech was within the scope of the escorts’ employment. If it turned out that a particular clinic authorized its employees to speak about abortion in the buffer zones, that would support an as-applied challenge to the zones at that clinic. Pp. 15–18. (c) Although the Act is content neutral, it is not “narrowly tailored” because it “burden[s] substantially more speech than is necessary to further the government’s legitimate interests.” Ward, 491 U. S., at 799. Pp. 18–29. (1) The buffer zones serve the Commonwealth’s legitimate interests in maintaining public safety on streets and sidewalks and in preserving access to adjacent reproductive healthcare facilities. See Schenck v. Pro-Choice Network of Western N. Y., 519 U.S. 357, 376. At the same time, however, they impose serious burdens on petitioners’ speech, depriving them of their two primary methods of communicating with arriving patients: close, personal conversations and distribution of literature. Those forms of expression have historically been closely associated with the transmission of ideas. While the Act may allow petitioners to “protest” outside the buffer zones, petitioners are not protestors; they seek not merely to express their opposition to abortion, but to engage in personal, caring, consensual conversations with women about various alternatives. It is thus no answer to say that petitioners can still be seen and heard by women within the buffer zones. If all that the women can see and hear are vociferous opponents of abortion, then the buffer zones have effectively stifled petitioners’ message. Pp. 19–23. (2) The buffer zones burden substantially more speech than necessary to achieve the Commonwealth’s asserted interests. Subsection (e) of the Act already prohibits deliberate obstruction of clinic entrances. Massachusetts could also enact legislation similar to the federal Freedom of Access to Clinic Entrances Act of 1994, 18 U. S. C. §248(a)(1), which imposes criminal and civil sanctions for obstructing, intimidating, or interfering with persons obtaining or providing reproductive health services. Obstruction of clinic driveways can readily be addressed through existing local traffic ordinances. While the Commonwealth contends that individuals can inadvertently obstruct access to clinics simply by gathering in large numbers, that problem could be addressed through a law requiring crowds blocking a clinic entrance to disperse for a limited period when ordered to do so by the police. In any event, crowding appears to be a problem only at the Boston clinic, and even there, only on Saturday mornings. The Commonwealth has not shown that it seriously undertook to address these various problems with the less intrusive tools readily available to it. It identifies not a single prosecution or injunction against individuals outside abortion clinics since the 1990s. The Commonwealth responds that the problems are too widespread for individual prosecutions and injunctions to be effective. But again, the record indicates that the problems are limited principally to the Boston clinic on Saturday mornings, and the police there appear perfectly capable of singling out lawbreakers. The Commonwealth also claims that it would be difficult to prove intentional or deliberate obstruction or intimidation and that the buffer zones accordingly make the police’s job easier. To meet the narrow tailoring requirement, however, the government must demonstrate that alternative measures that burden substantially less speech would fail to achieve the government’s interests, not simply that the chosen route is easier. In any event, to determine whether someone intends to block access to a clinic, a police officer need only order him to move; if he refuses, then there is no question that his continued conduct is knowing or intentional. For similar reasons, the Commonwealth’s reliance on Burson v. Freeman, 504 U.S. 191, is misplaced. There, the Court upheld a law establishing buffer zones outside polling places on the ground that less restrictive measures were inadequate. But whereas “[v]oter intimidation and election fraud” are “difficult to detect,” id., at 208, obstruction and harassment at abortion clinics are anything but subtle. And while the police “generally are barred from the vicinity of the polls to avoid any appearance of coercion in the electoral process,” id., at 207, they maintain a significant presence outside Massachusetts abortion clinics. In short, given the vital First Amendment interests at stake, it is not enough for Massachusetts simply to say that other approaches have not worked. Pp. 23–29. 708 F.3d 1, reversed and remanded. Roberts, C. J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Scalia, J., filed an opinion concurring in the judgment, in which Kennedy and Thomas, JJ., joined. Alito, J., filed an opinion concurring in the judgment.

A Massachusetts statute makes it a crime to knowingly stand on a “public way or sidewalk” within 35 feet of an entrance or driveway to any place, other than a hospital, where abortions are performed. Mass. Gen. Laws, ch. 266, §§120E½(a), (b) (West 2012). Petitioners are individuals who approach and talk to women outside such facilities, attempting to dissuade them from having abortions. The statute prevents petitioners from doing so near the facilities’ entrances. The question presented is whether the statute violates the First Amendment. I A In 2000, the Massachusetts Legislature enacted the Massachusetts Reproductive Health Care Facilities Act, Mass. Gen. Laws, ch. 266, §120E½ (West 2000). The law was designed to address clashes between abortion opponents and advocates of abortion rights that were occurring outside clinics where abortions were performed. The Act established a defined area with an 18-foot radius around the entrances and driveways of such facilities. §120E½(b). Anyone could enter that area, but once within it, no one (other than certain exempt individuals) could knowingly approach within six feet of another person—unless that person consented—“for the purpose of passing a leaflet or handbill to, displaying a sign to, or engaging in oral protest, education, or counseling with such other person.” Ibid. A separate provision subjected to criminal punishment anyone who “knowingly obstructs, detains, hinders, impedes or blocks another person’s entry to or exit from a reproductive health care facility.” §120E½(e). The statute was modeled on a similar Colorado law that this Court had upheld in Hill v. Colorado, 530 U. S. 703 (2000) . Relying on Hill, the United States Court of Appeals for the First Circuit sustained the Massachusetts statute against a First Amendment challenge. McGuire v. Reilly, 386 F. 3d 45 (2004) (McGuire II), cert. denied, 544 U. S. 974 (2005) ; McGuire v. Reilly, 260 F. 3d 36 (2001) (McGuire I). By 2007, some Massachusetts legislators and law enforcement officials had come to regard the 2000 statute as inadequate. At legislative hearings, multiple witnesses recounted apparent violations of the law. Massachusetts Attorney General Martha Coakley, for example, testified that protestors violated the statute “on a routine basis.” App. 78. To illustrate this claim, she played a video depicting protestors approaching patients and clinic staff within the buffer zones, ostensibly without the latter individuals’ consent. Clinic employees and volunteers also testified that protestors congregated near the doors and in the driveways of the clinics, with the result that prospective patients occasionally retreated from the clinics rather than try to make their way to the clinic entrances or parking lots. Captain William B. Evans of the Boston Police Department, however, testified that his officers had made “no more than five or so arrests” at the Planned Parenthood clinic in Boston and that what few prosecutions had been brought were unsuccessful. Id., at 68–69. Witnesses attributed the dearth of enforcement to the difficulty of policing the six-foot no-approach zones. Captain Evans testified that the 18-foot zones were so crowded with protestors that they resembled “a goalie’s crease,” making it hard to determine whether a protestor had deliberately approached a patient or, if so, whether the patient had consented. Id., at 69–71. For similar reasons, Attorney General Coakley concluded that the six-foot no-approach zones were “unenforceable.” Id., at 79. What the police needed, she said, was a fixed buffer zone around clinics that protestors could not enter. Id., at 74, 76. Captain Evans agreed, explaining that such a zone would “make our job so much easier.” Id., at 68. To address these concerns, the Massachusetts Legislature amended the statute in 2007, replacing the six-foot no-approach zones (within the 18-foot area) with a 35-foot fixed buffer zone from which individuals are categorically excluded. The statute now provides: “No person shall knowingly enter or remain on a public way or sidewalk adjacent to a reproductive health care facility within a radius of 35 feet of any portion of an entrance, exit or driveway of a reproductive health care facility or within the area within a rectangle created by extending the outside boundaries of any entrance, exit or driveway of a reproductive health care facility in straight lines to the point where such lines intersect the sideline of the street in front of such entrance, exit or driveway.” Mass. Gen. Laws, ch. 266, §120E½(b) (West 2012). A “reproductive health care facility,” in turn, is defined as “a place, other than within or upon the grounds of a hospital, where abortions are offered or performed.” §120E½(a). The 35-foot buffer zone applies only “during a facility’s business hours,” and the area must be “clearly marked and posted.” §120E½(c). In practice, facilities typically mark the zones with painted arcs and posted signs on adjacent sidewalks and streets. A first violation of the statute is punishable by a fine of up to $500, up to three months in prison, or both, while a subsequent offense is punishable by a fine of between $500 and $5,000, up to two and a half years in prison, or both. §120E½(d). The Act exempts four classes of individuals: (1) “persons entering or leaving such facility”; (2) “employees or agents of such facility acting within the scope of their employment”; (3) “law enforcement, ambulance, firefighting, construction, utilities, public works and other municipal agents acting within the scope of their employment”; and (4) “persons using the public sidewalk or street right-of-way adjacent to such facility solely for the purposeof reaching a destination other than such facility.” §120E½(b)(1)–(4). The legislature also retained the separate provision from the 2000 version that proscribes the knowing obstruction of access to a facility. §120E½(e). B Some of the individuals who stand outside Massachusetts abortion clinics are fairly described as protestors, who express their moral or religious opposition to abortion through signs and chants or, in some cases, more aggressive methods such as face-to-face confrontation. Petitioners take a different tack. They attempt to engage women approaching the clinics in what they call “sidewalk counseling,” which involves offering information about alternatives to abortion and help pursuing those options. Petitioner Eleanor McCullen, for instance, will typically initiate a conversation this way: “Good morning, may I give you my literature? Is there anything I can do for you? I’m available if you have any questions.” App. 138. If the woman seems receptive, McCullen will provide additional information. McCullen and the other petitioners consider it essential to maintain a caring demeanor, a calm tone of voice, and direct eye contact during these exchanges. Such interactions, petitioners believe, are a much more effective means of dissuading women from having abortions than confrontational methods such as shouting or brandishing signs, which in petitioners’ view tend only to antagonize their intended audience. In unrefuted testimony, petitioners say they have collectively persuaded hundreds of women to forgo abortions. The buffer zones have displaced petitioners from their previous positions outside the clinics. McCullen offers counseling outside a Planned Parenthood clinic in Boston, as do petitioners Jean Zarrella and Eric Cadin. Petitioner Gregory Smith prays the rosary there. The clinic occupies its own building on a street corner. Its main door is recessed into an open foyer, approximately 12 feet back from the public sidewalk. Before the Act was amended to create the buffer zones, petitioners stood near the entryway to the foyer. Now a buffer zone—marked by a painted arc and a sign—surrounds the entrance. This zone extends 23 feet down the sidewalk in one direction, 26 feet in the other, and outward just one foot short of the curb. The clinic’s entrance adds another seven feet to the width of the zone. Id., at 293–295. The upshot is that petitioners are effectively excluded from a 56-foot-wide expanse of the public sidewalk in front of the clinic.[1] Petitioners Mark Bashour and Nancy Clark offer counseling and information outside a Planned Parenthood clinic in Worcester. Unlike the Boston clinic, the Worcester clinic sits well back from the public street and sidewalks. Patients enter the clinic in one of two ways. Those arriving on foot turn off the public sidewalk and walk down a nearly 54-foot-long private walkway to the main entrance. More than 85% of patients, however, arrive by car, turning onto the clinic’s driveway from the street, parking in a private lot, and walking to the main entrance on a private walkway. Bashour and Clark would like to stand where the private walkway or driveway intersects the sidewalk and offer leaflets to patients as they walk or drive by. But a painted arc extends from the private walkway 35 feet down the sidewalk in either direction and outward nearly to the curb on the opposite side of the street. Another arc surrounds the driveway’s entrance, covering more than 93 feet of the sidewalk (including the width of the driveway) and extending across the street and nearly six feet onto the sidewalk on the opposite side. Id., at 295–297. Bashour and Clark must now stand either some distance down the sidewalk from the private walkway and driveway or across the street. Petitioner Cyril Shea stands outside a Planned Parenthood clinic in Springfield, which, like the Worcester clinic, is set back from the public streets. Approximately 90% of patients arrive by car and park in the private lots surrounding the clinic. Shea used to position himself at an entrance to one of the five driveways leading to the parking lots. Painted arcs now surround the entrances, each spanning approximately 100 feet of the sidewalk parallel to the street (again, including the width of the driveways) and extending outward well into the street. Id., at 297–299. Like petitioners at the Worcester clinic, Shea now stands far down the sidewalk from the driveway entrances. Petitioners at all three clinics claim that the buffer zones have considerably hampered their counseling efforts. Although they have managed to conduct some counseling and to distribute some literature outside the buffer zones—particularly at the Boston clinic—they say they have had many fewer conversations and distributed many fewer leaflets since the zones went into effect. Id., at 136–137, 180, 200. The second statutory exemption allows clinic employees and agents acting within the scope of their employment to enter the buffer zones. Relying on this exemption, the Boston clinic uses “escorts” to greet women as they approach the clinic, accompanying them through the zones to the clinic entrance. Petitioners claim that the escorts sometimes thwart petitioners’ attempts to communicate with patients by blocking petitioners from handing literature to patients, telling patients not to “pay any attention” or “listen to” petitioners, and disparaging petitioners as “crazy.” Id., at 165, 178. C In January 2008, petitioners sued Attorney General Coakley and other Commonwealth officials. They sought to enjoin enforcement of the Act, alleging that it violates the First and Fourteenth Amendments, both on its face and as applied to them. The District Court denied petitioners’ facial challenge after a bench trial based on a stipulated record. 573 F. Supp. 2d 382 (Mass. 2008). The Court of Appeals for the First Circuit affirmed. 571 F. 3d 167 (2009). Relying extensively on its previous decisions upholding the 2000 version of the Act, see McGuire II, 386 F. 3d 45; McGuire I, 260 F. 3d 36, the court upheld the 2007 version as a reasonable “time, place, and manner” regulation under the test set forth in Ward v. Rock Against Racism, 491 U. S. 781 (1989) . 571 F. 3d, at 174–181. It also rejected petitioners’ arguments that the Act was substantially overbroad, void for vagueness, and an impermissible prior restraint. Id., at 181–184. The case then returned to the District Court, which held that the First Circuit’s decision foreclosed all but one of petitioners’ as-applied challenges. 759 F. Supp. 2d 133 (2010). After another bench trial, it denied the remain-ing as-applied challenge, finding that the Act left petitioners ample alternative channels of communication. 844 F. Supp. 2d 206 (2012). The Court of Appeals once again affirmed. 708 F. 3d 1 (2013). We granted certiorari. 570 U. S. ___ (2013). II By its very terms, the Massachusetts Act regulates access to “public way[s]” and “sidewalk[s].” Mass. Gen. Laws, ch. 266, §120E½(b) (Supp. 2007). Such areas occupy a “special position in terms of First Amendment protection” because of their historic role as sites for discussion and debate. United States v. Grace, 461 U. S. 171, 180 (1983) . These places—which we have labeled “traditional public fora”—“ ‘have immemorially been held in trust for the use of the public and, time out of mind, have been used for purposes of assembly, communicating thoughts between citizens, and discussing public questions.’ ” Pleasant Grove City v. Summum, 555 U. S. 460, 469 (2009) (quoting Perry Ed. Assn. v. Perry Local Educators’ Assn., 460 U. S. 37, 45 (1983) ). It is no accident that public streets and sidewalks have developed as venues for the exchange of ideas. Even today, they remain one of the few places where a speaker can be confident that he is not simply preaching to the choir. With respect to other means of communication, an individual confronted with an uncomfortable message can always turn the page, change the channel, or leave the Web site. Not so on public streets and sidewalks. There, a listener often encounters speech he might otherwise tune out. In light of the First Amendment’s purpose “to preserve an uninhibited marketplace of ideas in which truth will ultimately prevail,” FCC v. League of Women Voters of Cal., 468 U. S. 364, 377 (1984) (internal quotation marks omitted), this aspect of traditional public fora is a virtue, not a vice. In short, traditional public fora are areas that have historically been open to the public for speech activities. Thus, even though the Act says nothing about speech on its face, there is no doubt—and respondents do not dispute—that it restricts access to traditional public fora and is therefore subject to First Amendment scrutiny. See Brief for Respondents 26 (although “[b]y its terms, the Act regulates only conduct,” it “incidentally regulates the place and time of protected speech”). Consistent with the traditionally open character of public streets and sidewalks, we have held that the government’s ability to restrict speech in such locations is “very limited.” Grace, supra, at 177. In particular, the guiding First Amendment principle that the “government has no power to restrict expression because of its message, its ideas, its subject matter, or its content” applies with full force in a traditional public forum. Police Dept. of Chicago v. Mosley, 408 U. S. 92, 95 (1972) . As a general rule, in such a forum the government may not “selectively . . . shield the public from some kinds of speech on the ground that they are more offensive than others.” Erznoznik v. Jacksonville, 422 U. S. 205, 209 (1975) . We have, however, afforded the government somewhat wider leeway to regulate features of speech unrelated to its content. “[E]ven in a public forum the government may impose reasonable restrictions on the time, place, or manner of protected speech, provided the restrictions ‘are justified without reference to the content of the regulated speech, that they are narrowly tailored to serve a significant governmental interest, and that they leave open ample alternative channels for communication of the information.’ ” Ward, 491 U. S., at 791 (quoting Clark v. Community for Creative Non-Violence, 468 U. S. 288, 293 (1984) ).[2] While the parties agree that this test supplies theproper framework for assessing the constitutionality of the Massachusetts Act, they disagree about whether the Act satisfies the test’s three requirements. III Petitioners contend that the Act is not content neutral for two independent reasons: First, they argue that it discriminates against abortion-related speech because it establishes buffer zones only at clinics that perform abortions. Second, petitioners contend that the Act, by exempting clinic employees and agents, favors one viewpoint about abortion over the other. If either of these arguments is correct, then the Act must satisfy strict scrutiny—that is, it must be the least restrictive means of achieving a compelling state interest. See United States v. Playboy Entertainment Group, Inc., 529 U. S. 803, 813 (2000) . Respondents do not argue that the Act can survive this exacting standard. Justice Scalia objects to our decision to consider whether the statute is content based and thus subject to strict scrutiny, given that we ultimately conclude that it is not narrowly tailored. Post, at 2 (opinion concurring in judgment). But we think it unexceptional to perform the first part of a multipart constitutional analysis first. The content-neutrality prong of the Ward test is logically antecedent to the narrow-tailoring prong, because it determines the appropriate level of scrutiny. It is not unusual for the Court to proceed sequentially in applying a constitutional test, even when the preliminary steps turn out not to be dispositive. See, e.g., Bartnicki v. Vopper, 532 U. S. 514 –527 (2001); Holder v. Humanitarian Law Project, 561 U. S. 1 –28 (2010) (concluding that a law was content based even though it ultimately survived strict scrutiny). The Court does sometimes assume, without deciding, that a law is subject to a less stringent level of scrutiny, as we did earlier this Term in McCutcheon v. Federal Election Commission, 572 U. S. ___, ___ (2014) (plurality opinion) (slip op., at 10). But the distinction between that case and this one seems clear: Applying any standard of review other than intermediate scrutiny in McCutcheon—the standard that was assumed to apply—would have required overruling a precedent. There is no similar reason to forgo the ordinary order of operations in this case. At the same time, there is good reason to address content neutrality. In discussing whether the Act is narrowly tailored, see Part IV, infra, we identify a number of less-restrictive alternative measures that the Massachusetts Legislature might have adopted. Some apply only at abortion clinics, which raises the question whether those provisions are content neutral. See infra, at 12–15. While we need not (and do not) endorse any of those measures, it would be odd to consider them as possible alternatives if they were presumptively unconstitutional because they were content based and thus subject to strict scrutiny. A The Act applies only at a “reproductive health care facility,” defined as “a place, other than within or upon the grounds of a hospital, where abortions are offered or performed.” Mass. Gen. Laws, ch. 266, §120E½(a). Given this definition, petitioners argue, “virtually all speech affected by the Act is speech concerning abortion,” thus rendering the Act content based. Brief for Petitioners 23. We disagree. To begin, the Act does not draw content-based distinctions on its face. Contrast Boos v. Barry, 485 U. S. 312, 315 (1988) (ordinance prohibiting the display within 500 feet of a foreign embassy of any sign that tends to bring the foreign government into “ ‘public odium’ ” or “ ‘public disrepute’ ”); Carey v. Brown, 447 U. S. 455, 465 (1980) (statute prohibiting all residential picketing except “peaceful labor picketing”). The Act would be content based if it required “enforcement authorities” to “examine the content of the message that is conveyed to determine whether” a violation has occurred. League of Women Voters of Cal., supra, at 383. But it does not. Whether petitioners violate the Act “depends” not “on what they say,” Humanitarian Law Project, supra, at 27, but simply on where they say it. Indeed, petitioners can violate the Act merely by standing in a buffer zone, without displaying a sign or uttering a word. It is true, of course, that by limiting the buffer zones to abortion clinics, the Act has the “inevitable effect” of restricting abortion-related speech more than speech on other subjects. Brief for Petitioners 24 (quoting United States v. O’Brien, 391 U. S. 367, 384 (1968) ). But a facially neutral law does not become content based simply be-cause it may disproportionately affect speech on certain topics. On the contrary, “[a] regulation that serves purposes unrelated to the content of expression is deemed neutral, even if it has an incidental effect on some speakers or messages but not others.” Ward, supra, at 791. The question in such a case is whether the law is “ ‘justified without reference to the content of the regulated speech.’ ” Renton v. Playtime Theatres, Inc., 475 U. S. 41, 48 (1986) (quoting Virginia Pharmacy Board v. Virginia Citizens Consumer Council, Inc., 425 U. S. 748, 771 (1976) ; emphasis deleted). The Massachusetts Act is. Its stated purpose is to “increase forthwith public safety at reproductive health care facilities.” 2007 Mass. Acts p. 660. Respondents have articulated similar purposes before this Court—namely, “public safety, patient access to healthcare, and the unobstructed use of public sidewalks and roadways.” Brief for Respondents 27; see, e.g., App. 51 (testimony of Attorney General Coakley); id., at 67–70 (testimony of Captain William B. Evans of the Boston Police); id., at 79–80 (testimony of Mary Beth Heffernan, Undersecretary for Criminal Justice); id., at 122–124 (affidavit of Captain Evans). It is not the case that “[e]very objective indication shows that the provision’s primary purpose is to restrict speech that opposes abortion.” Post, at 7. We have previously deemed the foregoing concerns to be content neutral. See Boos, 485 U. S., at 321 (identifying “congestion,” “interference with ingress or egress,” and “the need to protect . . . security” as content-neutral concerns). Obstructed access and congested sidewalks are problems no matter what caused them. A group of individuals can obstruct clinic access and clog sidewalks just as much when they loiter as when they protest abortion or counsel patients. To be clear, the Act would not be content neutral if it were concerned with undesirable effects that arise from “the direct impact of speech on its audience” or “[l]isteners’ reactions to speech.” Ibid. If, for example, the speech outside Massachusetts abortion clinics caused offense or made listeners uncomfortable, such offense or discomfort would not give the Commonwealth a content-neutral justification to restrict the speech. All of the problems identified by the Commonwealth here, however, arise irrespective of any listener’s reactions. Whether or not a single person reacts to abortion protestors’ chants or petitioners’ counseling, large crowds outside abortion clinics can still compromise public safety, impede access, and obstruct sidewalks. Petitioners do not really dispute that the Commonwealth’s interests in ensuring safety and preventing obstruction are, as a general matter, content neutral. But petitioners note that these interests “apply outside every building in the State that hosts any activity that might occasion protest or comment,” not just abortion clinics. Brief for Petitioners 24. By choosing to pursue these interests only at abortion clinics, petitioners argue, the Massachusetts Legislature evinced a purpose to “single[ ] out for regulation speech about one particular topic: abortion.” Reply Brief 9. We cannot infer such a purpose from the Act’s limited scope. The broad reach of a statute can help confirm that it was not enacted to burden a narrower category of disfavored speech. See Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 451–452 (1996). At the same time, however, “States adopt laws to address the problems that confront them. The First Amendment does not require States to regulate for problems that do not exist.” Burson v. Freeman, 504 U. S. 191, 207 (1992) (plurality opinion). The Massachusetts Legislature amended the Act in 2007 in response to a problem that was, in its experience, limited to abortion clinics. There was a record of crowding, obstruction, and even violence outside such clinics. There were apparently no similar recurring problems associated with other kinds of healthcare facilities, let alone with “every building in the State that hosts any activity that might occasion protest or comment.” Brief for Petitioners 24. In light of the limited nature of the problem, it was reasonable for the Massachusetts Legislature to enact a limited solution. When selecting among various options for combating a particular problem, legislatures should be encouraged to choose the one that restricts less speech, not more. Justice Scalia objects that the statute does restrict more speech than necessary, because “only one [Massachusetts abortion clinic] is known to have been beset by the problems that the statute supposedly addresses.” Post, at 7. But there are no grounds for inferring content-based discrimination here simply because the legislature acted with respect to abortion facilities generally rather than proceeding on a facility-by-facility basis. On these facts, the poor fit noted by Justice Scalia goes to the question of narrow tailoring, which we consider below. See infra, at 26–28. B Petitioners also argue that the Act is content based because it exempts four classes of individuals, Mass. Gen. Laws, ch. 266, §§120E½(b)(1)–(4), one of which comprises “employees or agents of [a reproductive healthcare] facil-ity acting within the scope of their employment.” §120E½(b)(2). This exemption, petitioners say, favors one side in the abortion debate and thus constitutes viewpoint discrimination—an “egregious form of content discrimination,” Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819, 829 (1995) . In particular, petitioners argue that the exemption allows clinic employees and agents—including the volunteers who “escort” patients arriving at the Boston clinic—to speak inside the buffer zones. It is of course true that “an exemption from an otherwise permissible regulation of speech may represent a governmental ‘attempt to give one side of a debatable public question an advantage in expressing its views to the people.’ ” City of Ladue v. Gilleo, 512 U. S. 43, 51 (1994) (quoting First Nat. Bank of Boston v. Bellotti, 435 U. S. 765 –786 (1978)). At least on the record before us, however, the statutory exemption for clinic employees and agents acting within the scope of their employment does not appear to be such an attempt. There is nothing inherently suspect about providing some kind of exemption to allow individuals who work at the clinics to enter or remain within the buffer zones. In particular, the exemption cannot be regarded as simply a carve-out for the clinic escorts; it also covers employees such as the maintenance worker shoveling a snowy sidewalk or the security guard patrolling a clinic entrance, see App. 95 (affidavit of Michael T. Baniukiewicz). Given the need for an exemption for clinic employees, the “scope of their employment” qualification simply ensures that the exemption is limited to its purpose of allowing the employees to do their jobs. It performs the same function as the identical “scope of their employment” restriction on the exemption for “law enforcement, ambulance, fire-fighting, construction, utilities, public works and other municipal agents.” §120E½(b)(3). Contrary to the suggestion of Justice Scalia, post, at 11–12, there is little reason to suppose that the Massachusetts Legislature intended to incorporate a common law doctrine developed for determining vicarious liability in tort when it used the phrase “scope of their employment” for the wholly different purpose of defining the scope of an exemption to a criminal statute. The limitation instead makes clear—with respect to both clinic employees and municipal agents—that exempted individuals are allowed inside the zones only to perform those acts authorized by their employers. There is no suggestion in the record that any of the clinics authorize their employees to speak about abortion in the buffer zones. The “scope of their employment” limitation thus seems designed to protect against exactly the sort of conduct that petitioners and Justice Scalia fear. Petitioners did testify in this litigation about instances in which escorts at the Boston clinic had expressed views about abortion to the women they were accompanying, thwarted petitioners’ attempts to speak and hand literature to the women, and disparaged petitioners in various ways. See App. 165, 168–169, 177–178, 189–190. It is unclear from petitioners’ testimony whether these alleged incidents occurred within the buffer zones. There is no viewpoint discrimination problem if the incidents occurred outside the zones because petitioners are equally free to say whatever they would like in that area. Even assuming the incidents occurred inside the zones, the record does not suggest that they involved speech within the scope of the escorts’ employment. If the speech was beyond the scope of their employment, then each of the alleged incidents would violate the Act’s express terms. Petitioners’ complaint would then be that the police were failing to enforce the Act equally against clinic escorts. Cf. Hoye v. City of Oakland, 653 F. 3d 835, 849–852 (CA9 2011) (finding selective enforcement of a similar ordinance in Oakland, California). While such allegations might state a claim of official viewpoint discrimination, that would not go to the validity of the Act. In any event, petitioners nowhere allege selective enforcement. It would be a very different question if it turned out that a clinic authorized escorts to speak about abortion inside the buffer zones. See post, at 1–2 (Alito, J., concurring in judgment). In that case, the escorts would not seem to be violating the Act because the speech would be within the scope of their employment.[3] The Act’s exemption for clinic employees would then facilitate speech on only one side of the abortion debate—a clear form of viewpoint discrimination that would support an as-applied challenge to the buffer zone at that clinic. But the record before us contains insufficient evidence to show that the exemption operates in this way at any of the clinics, perhaps because the clinics do not want to doom the Act by allowing their employees to speak about abortion within the buffer zones.[4] We thus conclude that the Act is neither content nor viewpoint based and therefore need not be analyzed under strict scrutiny. IV Even though the Act is content neutral, it still must be “narrowly tailored to serve a significant governmental interest.” Ward, 491 U. S., at 796 (internal quotation marks omitted). The tailoring requirement does not sim-ply guard against an impermissible desire to censor. The government may attempt to suppress speech not only because it disagrees with the message being expressed, but also for mere convenience. Where certain speech is associated with particular problems, silencing the speech is sometimes the path of least resistance. But by demanding a close fit between ends and means, the tailoring requirement prevents the government from too readily “sacrific[ing] speech for efficiency.” Riley v. National Federation of Blind of N. C., Inc., 487 U. S. 781, 795 (1988) . For a content-neutral time, place, or manner regulation to be narrowly tailored, it must not “burden substantially more speech than is necessary to further the government’s legitimate interests.” Ward, 491 U. S., at 799. Such a regulation, unlike a content-based restriction of speech, “need not be the least restrictive or least intrusive means of” serving the government’s interests. Id., at 798. But the government still “may not regulate expression in such a manner that a substantial portion of the burden on speech does not serve to advance its goals.” Id., at 799. A As noted, respondents claim that the Act promotes “public safety, patient access to healthcare, and the unobstructed use of public sidewalks and roadways.” Brief for Respondents 27. Petitioners do not dispute the significance of these interests. We have, moreover, previously recognized the legitimacy of the government’s interests in “ensuring public safety and order, promoting the free flow of traffic on streets and sidewalks, protecting property rights, and protecting a woman’s freedom to seek pregnancy-related services.” Schenck v. Pro-Choice Network ofWestern N. Y., 519 U. S. 357, 376 (1997) . See also Madsen v. Women’s Health Center, Inc., 512 U. S. 753 –768 (1994). The buffer zones clearly serve these interests. At the same time, the buffer zones impose serious burdens on petitioners’ speech. At each of the three Planned Parenthood clinics where petitioners attempt to counsel patients, the zones carve out a significant portion of the adjacent public sidewalks, pushing petitioners well back from the clinics’ entrances and driveways. The zones thereby compromise petitioners’ ability to initiate the close, personal conversations that they view as essential to “sidewalk counseling.” For example, in uncontradicted testimony, McCullen explained that she often cannot distinguish patients from passersby outside the Boston clinic in time to initiate a conversation before they enter the buffer zone. App. 135. And even when she does manage to begin a discussion outside the zone, she must stop abruptly at its painted border, which she believes causes her to appear “untrustworthy” or “suspicious.” Id., at 135, 152. Given these limitations, McCullen is often reduced to raising her voice at patients from outside the zone—a mode of communication sharply at odds with the compassionate message she wishes to convey. Id., at 133, 152–153. Clark gave similar testimony about her experience at the Worcester clinic. Id., at 243–244. These burdens on petitioners’ speech have clearly taken their toll. Although McCullen claims that she has persuaded about 80 women not to terminate their pregnancies since the 2007 amendment, App. to Pet. for Cert. 42a, she also says that she reaches “far fewer people” than she did before the amendment, App. 137. Zarrella reports an even more precipitous decline in her success rate: She estimated having about 100 successful interactions over the years before the 2007 amendment, but not a single one since. Id., at 180. And as for the Worcester clinic, Clark testified that “only one woman out of 100 will make the effort to walk across [the street] to speak with [her].” Id., at 217. The buffer zones have also made it substantially more difficult for petitioners to distribute literature to arriving patients. As explained, because petitioners in Boston cannot readily identify patients before they enter the zone, they often cannot approach them in time to place literature near their hands—the most effective means of getting the patients to accept it. Id., at 179. In Worcester and Springfield, the zones have pushed petitioners so far back from the clinics’ driveways that they can no longer even attempt to offer literature as drivers turn into the parking lots. Id., at 213, 218, 252–253. In short, the Act operates to deprive petitioners of their two primary methods of communicating with patients. The Court of Appeals and respondents are wrong to downplay these burdens on petitioners’ speech. As the Court of Appeals saw it, the Constitution does not accord “special protection” to close conversations or “handbilling.” 571 F. 3d, at 180. But while the First Amendment does not guarantee a speaker the right to any particular form of expression, some forms—such as normal conversation and leafletting on a public sidewalk—have historically been more closely associated with the transmission of ideas than others. In the context of petition campaigns, we have observed that “one-on-one communication” is “the most effective, fundamental, and perhaps economical avenue of political discourse.” Meyer v. Grant, 486 U. S. 414, 424 (1988) . See also Schenck, supra, at 377 (invalidating a “floating” buffer zone around people entering an abortion clinic partly on the ground that it prevented protestors “from communicating a message from a normal conversational distance or handing leaflets to people entering or leaving the clinics who are walking on the public sidewalks”). And “handing out leaflets in the advocacy of a politically controversial viewpoint . . . is the essence of First Amendment expression”; “[n]o form of speech is entitled to greater constitutional protection.” McIntyre v. Ohio Elections Comm’n, 514 U. S. 334, 347 (1995) . See also Schenck, supra, at 377 (“Leafletting and commenting on matters of public concern are classic forms of speech that lie at the heart of the First Amendment”). When the government makes it more difficult to engage in these modes of communication, it imposes an especially significant First Amendment burden.[5] Respondents also emphasize that the Act does not prevent petitioners from engaging in various forms of “protest”—such as chanting slogans and displaying signs—outside the buffer zones. Brief for Respondents 50–54. That misses the point. Petitioners are not protestors. They seek not merely to express their opposition to abortion, but to inform women of various alternatives and to provide help in pursuing them. Petitioners believe that they can accomplish this objective only through personal, caring, consensual conversations. And for good reason: It is easier to ignore a strained voice or a waving hand than a direct greeting or an outstretched arm. While the record indicates that petitioners have been able to have a number of quiet conversations outside the buffer zones, respondents have not refuted petitioners’ testimony that the conversations have been far less frequent and far less successful since the buffer zones were instituted. It is thus no answer to say that petitioners can still be “seen and heard” by women within the buffer zones. Id., at 51–53. If all that the women can see and hear are vociferous opponents of abortion, then the buffer zones have effectively stifled petitioners’ message. Finally, respondents suggest that, at the Worcester and Springfield clinics, petitioners are prevented from communicating with patients not by the buffer zones but by the fact that most patients arrive by car and park in the clinics’ private lots. Id., at 52. It is true that the layout of the two clinics would prevent petitioners from approaching the clinics’ doorways, even without the buffer zones. But petitioners do not claim a right to trespass on the clinics’ property. They instead claim a right to stand on the public sidewalks by the driveway as cars turn into the parking lot. Before the buffer zones, they could do so. Now they must stand a substantial distance away. The Act alone is responsible for that restriction on their ability to convey their message. B 1 The buffer zones burden substantially more speech than necessary to achieve the Commonwealth’s asserted interests. At the outset, we note that the Act is truly exceptional: Respondents and their amici identify no other State with a law that creates fixed buffer zones around abortion clinics.[6] That of course does not mean that the law is invalid. It does, however, raise concern that the Commonwealth has too readily forgone options that could serve its interests just as well, without substantially burdening the kind of speech in which petitioners wish to engage. That is the case here. The Commonwealth’s interests include ensuring public safety outside abortion clinics, preventing harassment and intimidation of patients and clinic staff, and combating deliberate obstruction of clinic entrances. The Act itself contains a separate provision, subsection (e)—unchallenged by petitioners—that prohibits much of this conduct. That provision subjects to criminal punishment “[a]ny person who knowingly obstructs, detains, hinders, impedes or blocks another person’s entry to or exit from a reproductive health care facility.” Mass. Gen. Laws, ch. 266, §120E½(e).[7] If Massachusetts determines that broader prohibitions along the same lines are necessary, it could enact legislation similar to the federal Freedom of Access to Clinic Entrances Act of 1994 (FACE Act), 18 U. S. C. §248(a)(1), which subjects to both criminal and civil penalties anyone who “by force or threat of force or by physical obstruction, intentionally injures, intimidates or interferes with or attempts to injure, intimidate or interfere with any person because that person is or has been, or in order to intimidate such person or any other person or any class of persons from, obtaining or providing reproductive health services.” Some dozen other States have done so. See Brief for State of New York et al. as Amici Curiae 13, and n. 6. If the Commonwealth is particularly concerned about harassment, it could also consider an ordinance such as the one adopted in New York City that not only prohibits obstructing access to a clinic, but also makes it a crime “to follow and harass another person within 15 feet of the premises of a reproductive health care facility.” N. Y. C. Admin. Code §8–803(a)(3) (2014).[8] The Commonwealth points to a substantial public safety risk created when protestors obstruct driveways leading to the clinics. See App. 18, 41, 51, 88–89, 99, 118–119. That is, however, an example of its failure to look to less intrusive means of addressing its concerns. Any such obstruction can readily be addressed through existing local ordinances. See, e.g., Worcester, Mass., Revised Ordinances of 2008, ch. 12, §25(b) (“No person shall stand, or place any obstruction of any kind, upon any street, sidewalk or crosswalk in such a manner as to obstruct a free passage for travelers thereon”); Boston, Mass., Municipal Code, ch. 16–41.2(d) (2013) (“No person shall solicit while walking on, standing on or going into any street or highway used for motor vehicle travel, or any area appurtenant thereto (including medians, shoulder areas, bicycle lanes, ramps and exit ramps)”). All of the foregoing measures are, of course, in addition to available generic criminal statutes forbidding assault, breach of the peace, trespass, vandalism, and the like. In addition, subsection (e) of the Act, the FACE Act, and the New York City anti-harassment ordinance are all enforceable not only through criminal prosecutions but also through public and private civil actions for injunctions and other equitable relief. See Mass. Gen. Laws §120E½(f); 18 U. S. C. §248(c)(1); N. Y. C. Admin. Code §§8–804, 8–805. We have previously noted the First Amendment virtues of targeted injunctions as alternatives to broad, prophylactic measures. Such an injunction “regulates the activities, and perhaps the speech, of a group,” but only “because of the group’s past actions in the context of a specific dispute between real parties.” Madsen, 512 U. S., at 762 (emphasis added). Moreover, given the equitable nature of injunctive relief, courts can tailor a remedy to ensure that it restricts no more speech than necessary. See, e.g., id., at 770; Schenck, 519 U. S., at 380–381. In short, injunctive relief focuses on the precise individuals and the precise conduct causing a particular problem. The Act, by contrast, categorically excludes non-exempt individuals from the buffer zones, unnecessarily sweeping in innocent individuals and their speech. The Commonwealth also asserts an interest in preventing congestion in front of abortion clinics. According to respondents, even when individuals do not deliberately obstruct access to clinics, they can inadvertently do so simply by gathering in large numbers. But the Commonwealth could address that problem through more targeted means. Some localities, for example, have ordinances that require crowds blocking a clinic entrance to disperse when ordered to do so by the police, and that forbid the individuals to reassemble within a certain distance of the clinic for a certain period. See Brief for State of New York et al. as Amici Curiae 14–15, and n. 10. We upheld a similar law forbidding three or more people “ ‘to congregate within 500 feet of [a foreign embassy], and refuse to disperse after having been ordered so to do by the police,’ ” Boos, 485 U. S., at 316 (quoting D. C. Code §22–1115 (1938))—an order the police could give only when they “ ‘reasonably believe[d] that a threat to the security or peace of the embassy [was] present,’ ” 485 U. S., at 330 (quoting Finzer v. Barry, 798 F. 2d 1450, 1471 (CADC 1986)). And to the extent the Commonwealth argues that even these types of laws are ineffective, it has another problem. The portions of the record that respondents cite to support the anticongestion interest pertain mainly to one place at one time: the Boston Planned Parenthood clinic on Saturday mornings. App. 69–71, 88–89, 96, 123. Respondents point us to no evidence that individuals regularly gather at other clinics, or at other times in Boston, in sufficiently large groups to obstruct access. For a problem shown to arise only once a week in one city at one clinic, creating 35-foot buffer zones at every clinic across the Commonwealth is hardly a narrowly tailored solution. The point is not that Massachusetts must enact all or even any of the proposed measures discussed above. The point is instead that the Commonwealth has available to it a variety of approaches that appear capable of serving its interests, without excluding individuals from areas historically open for speech and debate. 2 Respondents have but one reply: “We have tried other approaches, but they do not work.” Respondents emphasize the history in Massachusetts of obstruction at abortion clinics, and the Commonwealth’s allegedly failed attempts to combat such obstruction with injunctions and individual prosecutions. They also point to the Commonwealth’s experience under the 2000 version of the Act, during which the police found it difficult to enforce the six-foot no-approach zones given the “frenetic” activity in front of clinic entrances. Brief for Respondents 43. According to respondents, this history shows that Massachusetts has tried less restrictive alternatives to the buffer zones, to no avail. We cannot accept that contention. Although respondents claim that Massachusetts “tried other laws already on the books,” id., at 41, they identify not a single prosecution brought under those laws within at least the last 17 years. And while they also claim that the Commonwealth “tried injunctions,” ibid., the last injunctions they cite date to the 1990s, see id., at 42 (citing Planned Parenthood League of Mass., Inc. v. Bell, 424 Mass. 573, 677 N. E. 2d 204 (1997); Planned Parenthood League of Mass., Inc. v. Operation Rescue, 406 Mass. 701, 550 N. E. 2d 1361 (1990)). In short, the Commonwealth has not shown that it seriously undertook to address the problem with less intrusive tools readily available to it. Nor has it shown that it considered different methods that other jurisdictions have found effective. Respondents contend that the alternatives we have discussed suffer from two defects: First, given the “widespread” nature of the problem, it is simply not “practicable” to rely on individual prosecutions and injunctions. Brief for Respondents 45. But far from being “widespread,” the problem appears from the record to be limited principally to the Boston clinic on Saturday mornings. Moreover, by their own account, the police appear per-fectly capable of singling out lawbreakers. The legislative testimony preceding the 2007 Act revealed substantial police and video monitoring at the clinics, especially when large gatherings were anticipated. Captain Evans testified that his officers are so familiar with the scene outside the Boston clinic that they “know all the players down there.” App. 69. And Attorney General Coakley relied on video surveillance to show legislators conduct she thought was “clearly against the law.” Id., at 78. If Commonwealth officials can compile an extensive record of obstruction and harassment to support their preferred legislation, we do not see why they cannot do the same to support injunctions and prosecutions against those who might deliberately flout the law. The second supposed defect in the alternatives we have identified is that laws like subsection (e) of the Act and the federal FACE Act require a showing of intentional or deliberate obstruction, intimidation, or harassment, which is often difficult to prove. Brief for Respondents 45–47. As Captain Evans predicted in his legislative testimony, fixed buffer zones would “make our job so much easier.” App. 68. Of course they would. But that is not enough to satisfy the First Amendment. To meet the requirement of narrow tailoring, the government must demonstrate that alternative measures that burden substantially less speech would fail to achieve the government’s interests, not simply that the chosen route is easier. A painted line on the sidewalk is easy to enforce, but the prime objective of the First Amendment is not efficiency. In any case, we do not think that showing intentional obstruction is nearly so difficult in this context as respondents suggest. To determine whether a protestor intends to block access to a clinic, a police officer need only order him to move. If he refuses, then there is no question that his continued conduct is knowing or intentional. For similar reasons, respondents’ reliance on our decision in Burson v. Freeman is misplaced. There, we upheld a state statute that established 100-foot buffer zones outside polling places on election day within which no one could display or distribute campaign materials or solicit votes. 504 U. S., at 193–194. We approved the buffer zones as a valid prophylactic measure, noting that existing “[i]ntimidation and interference laws fall short of serving a State’s compelling interests because they ‘deal with only the most blatant and specific attempts’ to impede elections.” Id., at 206–207 (quoting Buckley v. Valeo, 424 U. S. 1, 28 (1976) (per curiam)). Such laws were insufficient because “[v]oter intimidation and election fraud are . . . difficult to detect.” Burson, 504 U. S., at 208. Obstruction of abortion clinics and harassment of patients, by contrast, are anything but subtle. We also noted in Burson that under state law, “law enforcement officers generally are barred from the vicinity of the polls to avoid any appearance of coercion in the electoral process,” with the result that “many acts of interference would go undetected.” Id., at 207. Not so here. Again, the police maintain a significant presence outside Massachusetts abortion clinics. The buffer zones in Burson were justified because less restrictive measures were inadequate. Respondents have not shown that to be the case here. Given the vital First Amendment interests at stake, it is not enough for Massachusetts simply to say that other approaches have not worked.[9] * * * Petitioners wish to converse with their fellow citizens about an important subject on the public streets and sidewalks—sites that have hosted discussions about the issues of the day throughout history. Respondents assert undeniably significant interests in maintaining public safety on those same streets and sidewalks, as well as in preserving access to adjacent healthcare facilities. But here the Commonwealth has pursued those interests by the extreme step of closing a substantial portion of a traditional public forum to all speakers. It has done so without seriously addressing the problem through alternatives that leave the forum open for its time-honored purposes. The Commonwealth may not do that consistent with the First Amendment. The judgment of the Court of Appeals for the First Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The zone could have extended an additional 21 feet in width under the Act. Only the smaller area was marked off, however, so only that area has legal effect. See Mass. Gen. Laws, ch. 266, §120E½(c). 2 A different analysis would of course be required if the government property at issue were not a traditional public forum but instead “a forum that is limited to use by certain groups or dedicated solely to the discussion of certain subjects.” v. , . 3 Less than two weeks after the instant litigation was initiated, the Massachusetts Attorney General’s Office issued a guidance letter clarifying the application of the four exemptions. The letter interpreted the exemptions as not permitting clinic employees or agents, municipal employees or agents, or individuals passing by clinics “to express their views about abortion or to engage in any other partisan speech within the buffer zone.” App. 93, 93–94. While this interpretation supports our conclusion that the employee exemption does not render the Act viewpoint based, we do not consider it in our analysis because it appears to the scope of the Act—a criminal statute—rather than to adopt a “ ‘limiting construction.’ ” v. , (quoting v. , , n. 5 (1982)). 4 Of course we do not hold that “[s]peech restrictions favoring one viewpoint over another are not content based unless it can be shown that the favored viewpoint has actually been expressed.” ,at 13. We instead apply an uncontroversial principle of constitutional adjudication: that a plaintiff generally cannot prevail on an challenge without showing that the law has in fact been (or is sufficiently likely to be) unconstitutionally to him. Specifically, when someone challenges a law as viewpoint discriminatory but it is not clear from the face of the law which speakers will be allowed to speak, he must show that he was prevented from speaking while someone espousing another viewpoint was permitted to do so. can decry this analysis as “astonishing” only by quoting a sentence that is explicitly limited to as-applied challenges and treating it as relevant to facial challenges. 5 As a leading historian has noted: 6 do identify five localities with laws similar to the Act here. Brief for State of New York et al. as 14, n. 7. 7 Massachusetts also has a separate law prohibiting similar kinds of conduct at any “medical facility,” though that law, unlike the Act, requires explicit notice before any penalty may be imposed. Mass. Gen. Laws, ch. 266, §120E. 8 We do not “give [our] approval” to this or any of the other alternatives we discuss. ,at 4. We merely suggest that a law like the New York City ordinance could in principle constitute a permissible alternative. Whether such a law would pass constitutional muster would depend on a number of other factors, such as whether the term “harassment” had been authoritatively construed to avoid vagueness and overbreadth problems of the sort noted by . 9 Because we find that the Act is not narrowly tailored, we need not consider whether the Act leaves open ample alternative channels of communication. Nor need we consider petitioners’ overbreadth challenge.

572.US.185

The right to participate in democracy through political contributions is protected by the First Amendment, but that right is not absolute. Congress may regulate campaign contributions to protect against corruption or the appearance of corruption. See, e.g., Buckley v. Valeo, 424 U.S. 1, 26–27. It may not, however, regulate contributions simply to reduce the amount of money in politics, or to restrict the political participation of some in order to enhance the relative influence of others. See, e.g., Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett, 564 U. S. ___, ___. The Federal Election Campaign Act of 1971 (FECA), as amended by the Bipartisan Campaign Reform Act of 2002 (BCRA), imposes two types of limits on campaign contributions. Base limits restrict how much money a donor may contribute to a particular candidate or committee while aggregate limits restrict how much money a donor may contribute in total to all candidates or committees. 2 U. S. C. §441a. In the 2011–2012 election cycle, appellant McCutcheon contributed to 16 different federal candidates, complying with the base limits applicable to each. He alleges that the aggregate limits prevented him from contributing to 12 additional candidates and to a number of noncandidate political committees. He also alleges that he wishes to make similar contributions in the future, all within the base limits. McCutcheon and appellant Republican National Committee filed a complaint before a three-judge District Court, asserting that the aggregate limits were unconstitutional under the First Amendment. The District Court denied their motion for a preliminary injunction and granted the Government’s motion to dismiss. Assuming that the base limits appropriately served the Government’s anticorruption interest, the District Court concluded that the aggregate limits survived First Amendment scrutiny because they prevented evasion of the base limits. Held: The judgment is reversed, and the case is remanded. 893 F. Supp. 2d 133, reversed and remanded. Chief Justice Roberts, joined by Justice Scalia, Justice Kennedy, and Justice Alito, concluded that the aggregate limits are invalid under the First Amendment. Pp. 7–40. (a) Appellants’ substantial First Amendment challenge to the current system of aggregate limits merits plenary consideration. Pp. 7–14. (1) In Buckley, this Court evaluated the constitutionality of the original contribution and expenditure limits in FECA. Buckley distinguished the two types of limits based on the degree to which each encroaches upon protected First Amendment interests. It subjected expenditure limits to “the exacting scrutiny applicable to limitations on core First Amendment rights of political expression.” 424 U. S., at 44–45. But it concluded that contribution limits impose a lesser restraint on political speech and thus applied a lesser but still “rigorous standard of review,” id., at 29, under which such limits “may be sustained if the State demonstrates a sufficiently important interest and employs means closely drawn to avoid unnecessary abridgement of associational freedoms,” id., at 25. Because the Court found that the primary purpose of FECA—preventing quid pro quo corruption and its appearance—was a “sufficiently important” governmental interest, id., at 26–27, it upheld the base limit under the “closely drawn” test, id., at 29. After doing so, the Court devoted only one paragraph of its 139-page opinion to the aggregate limit then in place under FECA, noting that the provision “ha[d] not been separately addressed at length by the parties.” Id., at 38. It concluded that the aggregate limit served to prevent circumvention of the base limit and was “no more than a corollary” of that limit. Id., at 38. Pp. 7–9. (2) There is no need in this case to revisit Buckley’s distinction between contributions and expenditures and the corresponding distinction in standards of review. Regardless whether strict scrutiny or the “closely drawn” test applies, the analysis turns on the fit between the stated governmental objective and the means selected to achieve that objective. Here, given the substantial mismatch between the Government’s stated objective and the means selected to achieve it, the aggregate limits fail even under the “closely drawn” test. Buckley’s ultimate conclusion about the constitutionality of the aggregate limit in place under FECA does not control here. Buckley spent just three sentences analyzing that limit, which had not been separately addressed by the parties. Appellants here, by contrast, have directly challenged the aggregate limits in place under BCRA, a different statutory regime whose limits operate against a distinct legal backdrop. Most notably, statutory safeguards against circumvention have been considerably strengthened since Buckley. The 1976 FECA Amendments added another layer of base limits—capping contributions from individuals to political committees—and an antiproliferation rule prohibiting donors from creating or controlling multiple affiliated political committees. Since Buckley, the Federal Election Commission has also enacted an intricate regulatory scheme that further limits the opportunities for circumvention of the base limits through “unearmarked contributions to political committees likely to contribute” to a particular candidate. 424 U. S., at 38. In addition to accounting for such statutory and regulatory changes, appellants raise distinct legal arguments not considered in Buckley, including an overbreadth challenge to the aggregate limit. Pp. 10–14. (b) Significant First Amendment interests are implicated here. Contributing money to a candidate is an exercise of an individual’s right to participate in the electoral process through both political expression and political association. A restriction on how many candidates and committees an individual may support is hardly a “modest restraint” on those rights. The Government may no more restrict how many candidates or causes a donor may support than it may tell a newspaper how many candidates it may endorse. In its simplest terms, the aggregate limits prohibit an individual from fully contributing to the primary and general election campaigns of ten or more candidates, even if all contributions fall within the base limits. And it is no response to say that the individual can simply contribute less than the base limits permit: To require one person to contribute at lower levels because he wants to support more candidates or causes is to penalize that individual for “robustly exercis[ing]” his First Amendment rights. Davis v. Federal Election Comm’n, 554 U.S. 724, 739. In assessing the First Amendment interests at stake, the proper focus is on an individual’s right to engage in political speech, not a collective conception of the public good. The whole point of the First Amendment is to protect individual speech that the majority might prefer to restrict, or that legislators or judges might not view as useful to the democratic process. Pp. 14–18. (c) The aggregate limits do not further the permissible governmental interest in preventing quid pro quo corruption or its appearance. Pp. 18–36. (1) This Court has identified only one legitimate governmental interest for restricting campaign finances: preventing corruption or the appearance of corruption. See Davis, supra, at 741. Moreover, the only type of corruption that Congress may target is quid pro quo corruption. Spending large sums of money in connection with elections, but not in connection with an effort to control the exercise of an officeholder’s official duties, does not give rise to quid pro quo corruption. Nor does the possibility that an individual who spends large sums may garner “influence over or access to” elected officials or political parties. Citizens United v. Federal Election Comm’n, 558 U.S. 310, 359. The line between quid pro quo corruption and general influence must be respected in order to safeguard basic First Amendment rights, and the Court must “err on the side of protecting political speech rather than suppressing it.” Federal Election Comm’n v. Wisconsin Right to Life, 551 U.S. 449, 457 (opinion of Roberts, C. J.). Pp. 18–21. (2) The Government argues that the aggregate limits further the permissible objective of preventing quid pro quo corruption. The difficulty is that once the aggregate limits kick in, they ban all contributions of any amount, even though Congress’s selection of a base limit indicates its belief that contributions beneath that amount do not create a cognizable risk of corruption. The Government must thus defend the aggregate limits by demonstrating that they prevent circumvention of the base limits, a function they do not serve in any meaningful way. Given the statutes and regulations currently in effect, Buckley’s fear that an individual might “contribute massive amounts of money to a particular candidate through . . . unearmarked contributions” to entities likely to support the candidate, 424 U. S., at 38, is far too speculative. Even accepting Buckley’s circumvention theory, it is hard to see how a candidate today could receive “massive amounts of money” that could be traced back to a particular donor uninhibited by the aggregate limits. The Government’s scenarios offered in support of that possibility are either illegal under current campaign finance laws or implausible. Pp. 21–30. (3) The aggregate limits also violate the First Amendment because they are not “closely drawn to avoid unnecessary abridgment of associational freedoms.” Buckley, supra, at 25. The Government argues that the aggregate limits prevent an individual from giving to too many initial recipients who might then recontribute a donation, but experience suggests that the vast majority of contributions are retained and spent by their recipients. And the Government has provided no reason to believe that candidates or party committees would dramatically shift their priorities if the aggregate limits were lifted. The indiscriminate ban on all contributions above the aggregate limits is thus disproportionate to the Government’s interest in preventing circumvention. Importantly, there are multiple alternatives available to Congress that would serve the Government’s interest in preventing circumvention while avoiding “unnecessary abridgment” of First Amendment rights. Buckley, supra, at 25. Such alternatives might include targeted restrictions on transfers among candidates and political committees, or tighter earmarking rules. Transfers, after all, are the key to the Government’s concern about circumvention, but they can be addressed without such a direct and broad interference with First Amendment rights. Pp. 30–35. (4) Disclosure of contributions also reduces the potential for abuse of the campaign finance system. Disclosure requirements, which are justified by “a governmental interest in ‘provid[ing] the electorate with information’ about the sources of election-related spending,” Citizens United, supra, at 367, may deter corruption “by exposing large contributions and expenditures to the light of publicity,” Buckley, supra at 67. Disclosure requirements may burden speech, but they often represent a less restrictive alternative to flat bans on certain types or quantities of speech. Particularly with modern technology, disclosure now offers more robust protections against corruption than it did when Buckley was decided. Pp. 35–36. (d) The Government offers an additional rationale for the aggregate limits, arguing that the opportunity for corruption exists whenever a legislator is given a large check, even if the check consists of contributions within the base limits to be divided among numerous candidates or committees. That rationale dangerously broadens the circumscribed definition of quid pro quo corruption articulated in prior cases. Buckley confined its analysis to the possibility that “massive amounts of money” could be funneled to a particular candidate in excess of the base limits. 424 U. S., at 38. Recasting as corruption a donor’s widely distributed support for a political party would dramatically expand government regulation of the political process. And though the Government suggests that solicitation of large contributions poses the corruption danger, the aggregate limits are not limited to any direct solicitation by an officeholder or candidate. Pp. 36–39. Justice Thomas agreed that the aggregate limits are invalid under the First Amendment, but would overrule Buckley v. Valeo, 424 U.S. 1, and subject BCRA’s aggregate limits to strict scrutiny, which they would surely fail. Buckley’s “analytic foundation . . . was tenuous from the very beginning and has only continued to erode in the intervening years.” Nixon v. Shrink Missouri Government PAC, 528 U.S. 377, 412 (Thomas, J., dissenting). Contributions and expenditures are simply “two sides of the same First Amendment coin,” and this Court’s efforts to distinguish the two have produced mere “word games” rather than any cognizable constitutional law principle. Buckley, supra, at 241, 244 (Burger, C. J., concurring in part and dissenting in part). Pp. 1–5. Roberts, C. J., announced the judgment of the Court and delivered an opinion, in which Scalia, Kennedy, and Alito, JJ., joined. Thomas, J., filed an opinion concurring in the judgment. Breyer, J., filed a dissenting opinion, in which Ginsburg, Sotomayor, and Kagan, JJ., joined.

. At the same time, we have made clear that Congress may not regulate contributions simply to reduce the amount of money in politics, or to restrict the political participation of some in order to enhance the relative influence of others. See, e.g., Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett, 564 U. S. ___, ___ (2011) (slip op., at 24–25). Many people might find those latter objectives attractive: They would be delighted to see fewer television commercials touting a candidate’s accomplishments or disparaging an opponent’s character. Money in politics may at times seem repugnant to some, but so too does much of what the First Amendment vigorously protects. If the First Amendment protects flag burning, funeral protests, and Nazi parades—despite the profound offense such spectacles cause—it surely protects political campaign speech despite popular opposition. See Texas v. Johnson, 491 U. S. 397 (1989) ; Snyder v. Phelps, 562 U. S. ___ (2011); National Socialist Party of America v. Skokie, 432 U. S. 43 (1977) (per curiam). Indeed, as we have emphasized, the First Amendment “has its fullest and most urgent application precisely to the conduct of campaigns for political office.” Monitor Patriot Co. v. Roy, 401 U. S. 265, 272 (1971) . In a series of cases over the past 40 years, we have spelled out how to draw the constitutional line between the permissible goal of avoiding corruption in the political process and the impermissible desire simply to limit political speech. We have said that government regulation may not target the general gratitude a candidate may feel toward those who support him or his allies, or the political access such support may afford. “Ingratiation and access . . . are not corruption.” Citizens United v. Federal Election Comm’n, 558 U. S. 310, 360 (2010) . They embody a central feature of democracy—that constituents support candidates who share their beliefs and interests, and candidates who are elected can be expected to be responsive to those concerns. Any regulation must instead target what we have called “quid pro quo” corruption or its appearance. See id., at 359. That Latin phrase captures the notion of a direct exchange of an official act for money. See McCormick v. United States, 500 U. S. 257, 266 (1991) . “The hallmark of corruption is the financial quid pro quo: dollars for po- litical favors.” Federal Election Comm’n v. National Conservative Political Action Comm., 470 U. S. 480, 497 (1985) . Campaign finance restrictions that pursue other objectives, we have explained, impermissibly inject the Government “into the debate over who should govern.” Bennett, supra, at ___ (slip op., at 25). And those who govern should be the last people to help decide who should govern. The statute at issue in this case imposes two types of limits on campaign contributions. The first, called base limits, restricts how much money a donor may contribute to a particular candidate or committee. 2 U. S. C. §441a(a)(1). The second, called aggregate limits, restricts how much money a donor may contribute in total to all candidates or committees. §441a(a)(3). This case does not involve any challenge to the base limits, which we have previously upheld as serving the permissible objective of combatting corruption. The Government contends that the aggregate limits also serve that objective, by preventing circumvention of the base limits. We conclude, however, that the aggregate limits do little, if anything, to address that concern, while seriously restricting participation in the democratic process. The aggregate limits are therefore invalid under the First Amendment. I A For the 2013–2014 election cycle, the base limits in the Federal Election Campaign Act of 1971 (FECA), as amended by the Bipartisan Campaign Reform Act of 2002 (BCRA), permit an individual to contribute up to $2,600 per election to a candidate ($5,200 total for the primary and general elections); $32,400 per year to a national party committee; [ 1 ] $10,000 per year to a state or local party committee; and $5,000 per year to a political action committee, or “PAC.” 2 U. S. C. §441a(a)(1); 78 Fed. Reg. 8532 (2013). [ 2 ] A national committee, state or local party committee, or multicandidate PAC may in turn contribute up to $5,000 per election to a candidate. §441a(a)(2). [ 3 ] The base limits apply with equal force to contributions that are “in any way earmarked or otherwise directed through an intermediary or conduit” to a candidate. §441a(a)(8). If, for example, a donor gives money to a party committee but directs the party committee to pass the contribution along to a particular candidate, then the transaction is treated as a contribution from the original donor to the specified candidate. For the 2013–2014 election cycle, the aggregate limits in BCRA permit an individual to contribute a total of $48,600 to federal candidates and a total of $74,600 to other political committees. Of that $74,600, only $48,600 may be contributed to state or local party committees and PACs, as opposed to national party committees. §441a(a)(3); 78 Fed. Reg. 8532. All told, an individual may contribute up to $123,200 to candidate and noncandidate committees during each two-year election cycle. The base limits thus restrict how much money a donor may contribute to any particular candidate or committee; the aggregate limits have the effect of restricting how many candidates or committees the donor may support, to the extent permitted by the base limits. B In the 2011–2012 election cycle, appellant Shaun McCutcheon contributed a total of $33,088 to 16 different federal candidates, in compliance with the base limits applicable to each. He alleges that he wished to contribute $1,776 to each of 12 additional candidates but was prevented from doing so by the aggregate limit on contributions to candidates. McCutcheon also contributed a total of $27,328 to several noncandidate political committees, in compliance with the base limits applicable to each. He alleges that he wished to contribute to various other political committees, including $25,000 to each of the three Republican national party committees, but was prevented from doing so by the aggregate limit on contributions to political committees. McCutcheon further alleges that he plans to make similar contributions in the future. In the 2013–2014 election cycle, he again wishes to contribute at least $60,000 to various candidates and $75,000 to non-candidate political committees. Brief for Appellant McCutcheon 11–12. Appellant Republican National Committee is a national political party committee charged with the general management of the Republican Party. The RNC wishes to receive the contributions that McCutcheon and similarly situated individuals would like to make—contributions otherwise permissible under the base limits for national party committees but foreclosed by the aggregate limit on contributions to political committees. In June 2012, McCutcheon and the RNC filed a complaint before a three-judge panel of the U. S. District Court for the District of Columbia. See BCRA §403(a), 116Stat. 113–114. McCutcheon and the RNC asserted that the aggregate limits on contributions to candidates and to noncandidate political committees were unconstitutional under the First Amendment. They moved for a preliminary injunction against enforcement of the challenged provisions, and the Government moved to dismiss the case. The three-judge District Court denied appellants’ motion for a preliminary injunction and granted the Government’s motion to dismiss. Assuming that the base limits appropriately served the Government’s anticorruption interest, the District Court concluded that the aggregate limits survived First Amendment scrutiny because they prevented evasion of the base limits. 893 F. Supp. 2d 133, 140 (2012). In particular, the District Court imagined a hypothetical scenario that might occur in a world without aggregate limits. A single donor might contribute the maximum amount under the base limits to nearly 50 separate committees, each of which might then transfer the money to the same single committee. Ibid. That committee, in turn, might use all the transferred money for coordinated expenditures on behalf of a particular candidate, allowing the single donor to circumvent the base limit on the amount he may contribute to that candidate. Ibid. The District Court acknowledged that “it may seem unlikely that so many separate entities would willingly serve as conduits” for the single donor’s interests, but it concluded that such a scenario “is not hard to imagine.” Ibid. It thus rejected a constitutional challenge to the aggregate limits, characterizing the base limits and the aggregate limits “as a coherent system rather than merely a collection of individual limits stacking prophylaxis upon prophylaxis.” Ibid. McCutcheon and the RNC appealed directly to this Court, as authorized by law. 28 U. S. C. §1253. In such a case, “we ha[ve] no discretion to refuse adjudication of the case on its merits,” Hicks v. Miranda, 422 U. S. 332, 344 (1975) , and accordingly we noted probable jurisdiction. 568 U. S. ___ (2013). II A Buckley v. Valeo, 424 U. S. 1 , presented this Court with its first opportunity to evaluate the constitutionality of the original contribution and expenditure limits set forth in FECA. FECA imposed a $1,000 per election base limit on contributions from an individual to a federal candidate. It also imposed a $25,000 per year aggregate limit on all contributions from an individual to candidates or political committees. 18 U. S. C. §§608(b)(1), 608(b)(3) (1970 ed., Supp. IV). On the expenditures side, FECA imposed limits on both independent expenditures and candidates’ overall campaign expenditures. §§608(e)(1), 608(c). Buckley recognized that “contribution and expenditure limitations operate in an area of the most fundamental First Amendment activities.” 424 U. S., at 14. But it distinguished expenditure limits from contribution limits based on the degree to which each encroaches upon protected First Amendment interests. Expenditure limits, the Court explained, “necessarily reduce[ ] the quantity of expression by restricting the number of issues discussed, the depth of their exploration, and the size of the audience reached.” Id., at 19. The Court thus subjected expenditure limits to “the exacting scrutiny applicable to lim- itations on core First Amendment rights of political expression.” Id., at 44–45. Under exacting scrutiny, the Government may regulate protected speech only if such regulation promotes a compelling interest and is the least restrictive means to further the articulated interest. See Sable Communications of Cal., Inc. v. FCC, 492 U. S. 115, 126 (1989) . By contrast, the Court concluded that contribution limits impose a lesser restraint on political speech because they “permit[ ] the symbolic expression of support evidenced by a contribution but do[ ] not in any way infringe the contributor’s freedom to discuss candidates and issues.” Buckley, 424 U. S., at 21. As a result, the Court focused on the effect of the contribution limits on the freedom of political association and applied a lesser but still “rigorous standard of review.” Id., at 29. Under that standard, “[e]ven a ‘ “significant interference” with protected rights of political association’ may be sustained if the State demonstrates a sufficiently important interest and employs means closely drawn to avoid unnecessary abridgement of associational freedoms.” Id., at 25 (quoting Cousins v. Wigoda, 419 U. S. 477, 488 (1975) ). The primary purpose of FECA was to limit quid pro quo corruption and its appearance; that purpose satisfied the requirement of a “sufficiently important” governmental interest. 424 U. S., at 26–27. As for the “closely drawn” component, Buckley concluded that the $1,000 base limit “focuses precisely on the problem of large campaign contributions . . . while leaving persons free to engage in independent political expression, to associate actively through volunteering their services, and to assist to a limited but nonetheless substantial extent in supporting candidates and committees with financial resources.” Id., at 28. The Court therefore upheld the $1,000 base limit under the “closely drawn” test. Id., at 29. The Court next separately considered an overbreadth challenge to the base limit. See id., at 29–30. The challengers argued that the base limit was fatally overbroad because most large donors do not seek improper influence over legislators’ actions. Although the Court accepted that premise, it nevertheless rejected the overbreadth challenge for two reasons: First, it was too “difficult to isolate suspect contributions” based on a contributor’s subjective intent. Id., at 30. Second, “Congress was justified in concluding that the interest in safeguarding against the appearance of impropriety requires that the opportunity for abuse inherent in the process of raising large monetary contributions be eliminated.” Ibid. Finally, in one paragraph of its 139-page opinion, the Court turned to the $25,000 aggregate limit under FECA. As a preliminary matter, it noted that the constitution- ality of the aggregate limit “ha[d] not been separately addressed at length by the parties.” Id., at 38. Then, in three sentences, the Court disposed of any constitutional objections to the aggregate limit that the challengers might have had: “The overall $25,000 ceiling does impose an ultimate restriction upon the number of candidates and committees with which an individual may associate himself by means of financial support. But this quite modest restraint upon protected political activity serves to prevent evasion of the $1,000 contribution limitation by a person who might otherwise contribute massive amounts of money to a particular candidate through the use of unearmarked contributions to political committees likely to contribute to that candidate, or huge contributions to the candidate’s political party. The limited, additional restriction on associational freedom imposed by the overall ceiling is thus no more than a corollary of the basic individual contribution limitation that we have found to be constitutionally valid.” Ibid. B 1 The parties and amici curiae spend significant energy debating whether the line that Buckley drew between contributions and expenditures should remain the law. Notwithstanding the robust debate, we see no need in this case to revisit Buckley’s distinction between contributions and expenditures and the corollary distinction in the applicable standards of review. Buckley held that the Government’s interest in preventing quid pro quo corruption or its appearance was “sufficiently important,” id., at 26–27; we have elsewhere stated that the same interest may properly be labeled “compelling,” see National Conservative Political Action Comm., 470 U. S., at 496–497, so that the interest would satisfy even strict scrutiny. Moreover, regardless whether we apply strict scrutiny or Buckley’s “closely drawn” test, we must assess the fit between the stated governmental objective and the means selected to achieve that objective. See, e.g., National Conservative Political Action Comm., supra, at 496–501; Randall v. Sorrell, 548 U. S. 230 –262 (2006) (opinion of Breyer, J.). Or to put it another way, if a law that restricts political speech does not “avoid unnecessary abridgement” of First Amendment rights, Buckley, 424 U. S., at 25, it cannot survive “rigorous” review. Because we find a substantial mismatch between the Government’s stated objective and the means selected to achieve it, the aggregate limits fail even under the “closely drawn” test. We therefore need not parse the differences between the two standards in this case. 2 Buckley treated the constitutionality of the $25,000 aggregate limit as contingent upon that limit’s ability to prevent circumvention of the $1,000 base limit, describing the aggregate limit as “no more than a corollary” of the base limit. Id., at 38. The Court determined that circumvention could occur when an individual legally contributes “massive amounts of money to a particular candidate through the use of unearmarked contributions” to entities that are themselves likely to contribute to the candidate. Ibid. For that reason, the Court upheld the $25,000 aggregate limit. Although Buckley provides some guidance, we think that its ultimate conclusion about the constitutionality of the aggregate limit in place under FECA does not control here. Buckley spent a total of three sentences analyzing that limit; in fact, the opinion pointed out that the constitutionality of the aggregate limit “ha[d] not been separately addressed at length by the parties.” Ibid. We are now asked to address appellants’ direct challenge to the aggregate limits in place under BCRA. BCRA is a different statutory regime, and the aggregate limits it imposes operate against a distinct legal backdrop. Most notably, statutory safeguards against circumvention have been considerably strengthened since Buckley was decided, through both statutory additions and the introduction of a comprehensive regulatory scheme. With more targeted anticircumvention measures in place today, the indiscriminate aggregate limits under BCRA appear particularly heavy-handed. The 1976 FECA Amendments, for example, added another layer of base contribution limits. The 1974 version of FECA had already capped contributions from political committees to candidates, but the 1976 version added limits on contributions to political committees. This change was enacted at least “in part to prevent circumvention of the very limitations on contributions that this Court upheld in Buckley.” California Medical Assn. v. Federal Election Comm’n, 453 U. S. 182 –198 (1981) (plurality opinion); see also id., at 203 (Blackmun, J., concurring in part and concurring in judgment). Because a donor’s contributions to a political committee are now limited, a donor cannot flood the committee with “huge” amounts of money so that each contribution the committee makes is perceived as a contribution from him. Buckley, supra, at 38. Rather, the donor may contribute only $5,000 to the committee, which hardly raises the specter of abuse that concerned the Court in Buckley. Limits on contributions to political committees consequently create an additional hurdle for a donor who seeks both to channel a large amount of money to a particular candidate and to ensure that he gets the credit for doing so. The 1976 Amendments also added an antiprolifera- tion rule prohibiting donors from creating or controlling multiple affiliated political committees. See 2 U. S. C. §441a(a)(5); 11 CFR §100.5(g)(4). The Government ac- knowledges that this antiproliferation rule “forecloses what would otherwise be a particularly easy and effective means of circumventing the limits on contributions to any particular political committee.” Brief for Appellee 46. In effect, the rule eliminates a donor’s ability to create and use his own political committees to direct funds in excess of the individual base limits. It thus blocks a straightforward method of achieving the circumvention that was the underlying concern in Buckley. The intricate regulatory scheme that the Federal Election Commission has enacted since Buckley further limits the opportunities for circumvention of the base limits via “unearmarked contributions to political committees likely to contribute” to a particular candidate. 424 U. S., at 38. Although the earmarking provision, 2 U. S. C. §441a(a)(8), was in place when Buckley was decided, the FEC has since added regulations that define earmarking broadly. For example, the regulations construe earmarking to include any designation, “whether direct or indirect, express or implied, oral or written.” 11 CFR §110.6(b)(1). The regulations specify that an individual who has contributed to a particular candidate may not also contribute to a single-candidate committee for that candidate. §110.1(h)(1). Nor may an individual who has contributed to a candidate also contribute to a political committee that has supported or anticipates supporting the same candidate, if the individual knows that “a substantial portion [of his contribution] will be contributed to, or expended on behalf of,” that candidate. §110.1(h)(2). In addition to accounting for statutory and regulatory changes in the campaign finance arena, appellants’ challenge raises distinct legal arguments that Buckley did not consider. For example, presumably because of its cursory treatment of the $25,000 aggregate limit, Buckley did not separately address an overbreadth challenge with respect to that provision. The Court rejected such a challenge to the base limits because of the difficulty of isolating suspect contributions. The propriety of large contributions to in- dividual candidates turned on the subjective intent of donors, and the Court concluded that there was no way to tell which donors sought improper influence over legislators’ actions. See 424 U. S., at 30. The aggregate limit, on the other hand, was upheld as an anticircumvention measure, without considering whether it was possible to discern which donations might be used to circumvent the base limits. See id., at 38. The Court never addressed overbreadth in the specific context of aggregate limits, where such an argument has far more force. Given the foregoing, this case cannot be resolved merely by pointing to three sentences in Buckley that were written without the benefit of full briefing or argument on the issue. See Toucey v. New York Life Ins. Co., 314 U. S. 118 –140 (1941) (departing from “[l]oose language and a sporadic, ill-considered decision” when asked to resolve a question “with our eyes wide open and in the light of full consideration”); Hohn v. United States, 524 U. S. 236, 251 (1998) (departing from a prior decision where it “was rendered without full briefing or argument”). We are confronted with a different statute and different legal arguments, at a different point in the development of campaign finance regulation. Appellants’ sub- stantial First Amendment challenge to the system of aggregate limits currently in place thus merits our plenary consideration. [ 4 ] III The First Amendment “is designed and intended to remove governmental restraints from the arena of public discussion, putting the decision as to what views shall be voiced largely into the hands of each of us, . . . in the belief that no other approach would comport with the premise of individual dignity and choice upon which our political system rests.” Cohen v. California, 403 U. S. 15, 24 (1971) . As relevant here, the First Amendment safeguards an individual’s right to participate in the public debate through political expression and political association. See Buckley, 424 U. S., at 15. When an individual contributes money to a candidate, he exercises both of those rights: The contribution “serves as a general expression of support for the candidate and his views” and “serves to affiliate a person with a candidate.” Id., at 21–22. Those First Amendment rights are important regardless whether the individual is, on the one hand, a “lone pamphleteer[ ] or street corner orator[ ] in the Tom Paine mold,” or is, on the other, someone who spends “substan-tial amounts of money in order to communicate [his] political ideas through sophisticated” means. National Conservative Political Action Comm., 470 U. S., at 493. Either way, he is participating in an electoral debate that we have recognized is “integral to the operation of the system of government established by our Constitution.” Buckley, supra, at 14. Buckley acknowledged that aggregate limits at least diminish an individual’s right of political association. As the Court explained, the “overall $25,000 ceiling does impose an ultimate restriction upon the number of candidates and committees with which an individual may associate himself by means of financial support.” 424 U. S., at 38. But the Court characterized that restriction as a “quite modest restraint upon protected political activity.” Ibid. We cannot agree with that characterization. An aggregate limit on how many candidates and committees an individual may support through contributions is not a “modest restraint” at all. The Government may no more restrict how many candidates or causes a donor may support than it may tell a newspaper how many candidates it may endorse. To put it in the simplest terms, the aggregate limits prohibit an individual from fully contributing to the primary and general election campaigns of ten or more candidates, even if all contributions fall within the base limits Congress views as adequate to protect against corruption. The individual may give up to $5,200 each to nine candidates, but the aggregate limits constitute an outright ban on further contributions to any other candidate (beyond the additional $1,800 that may be spent before reaching the $48,600 aggregate limit). At that point, the limits deny the individual all ability to exercise his expressive and associational rights by contributing to someone who will advocate for his policy preferences. A donor must limit the number of candidates he supports, and may have to choose which of several policy concerns he will advance—clear First Amendment harms that the dissent never acknowledges. It is no answer to say that the individual can simply contribute less money to more people. To require one person to contribute at lower levels than others because he wants to support more candidates or causes is to impose a special burden on broader participation in the democratic process. And as we have recently admonished, the Government may not penalize an individual for “robustly exercis[ing]” his First Amendment rights. Davis v. Federal Election Comm’n, 554 U. S. 724, 739 (2008) . The First Amendment burden is especially great for individuals who do not have ready access to alternative avenues for supporting their preferred politicians and policies. In the context of base contribution limits, Buckley observed that a supporter could vindicate his associational interests by personally volunteering his time and energy on behalf of a candidate. See 424 U. S., at 22, 28. Such personal volunteering is not a realistic alternative for those who wish to support a wide variety of candidates or causes. Other effective methods of supporting preferred candidates or causes without contributing money are reserved for a select few, such as entertainers capable of raising hundreds of thousands of dollars in a single evening. Cf. Davis, supra, at 742. [ 5 ] The dissent faults this focus on “the individual’s right to engage in political speech,” saying that it fails to take into account “the public’s interest” in “collective speech.” Post, at 6 (opinion of Breyer, J). This “collective” interest is said to promote “a government where laws reflect the very thoughts, views, ideas, and sentiments, the expression of which the First Amendment protects.” Post, at 7. But there are compelling reasons not to define the boundaries of the First Amendment by reference to such a generalized conception of the public good. First, the dissent’s “collective speech” reflected in laws is of course the will of the majority, and plainly can include laws that restrict free speech. The whole point of the First Amendment is to afford individuals protection against such infringements. The First Amendment does not protect the government, even when the government purports to act through legislation reflecting “collective speech.” Cf. United States v. Alvarez, 567 U. S. ___ (2012); Wooley v. Maynard, 430 U. S. 705 (1977) ; West Virginia Bd. of Ed. v. Barnette, 319 U. S. 624 (1943) . Second, the degree to which speech is protected cannot turn on a legislative or judicial determination that particular speech is useful to the democratic process. The First Amendment does not contemplate such “ad hoc balancing of relative social costs and benefits.” United States v. Stevens, 559 U. S. 460, 470 (2010) ; see also United States v. Playboy Entertainment Group, Inc., 529 U. S. 803, 818 (2000) (“What the Constitution says is that” value judgments “are for the individual to make, not for the Government to decree, even with the mandate or approval of a majority”). Third, our established First Amendment analysis already takes account of any “collective” interest that may justify restrictions on individual speech. Under that accepted analysis, such restrictions are measured against the asserted public interest (usually framed as an important or compelling governmental interest). As explained below, we do not doubt the compelling nature of the “collective” interest in preventing corruption in the electoral process. But we permit Congress to pursue that interest only so long as it does not unnecessarily infringe an individual’s right to freedom of speech; we do not truncate this tailoring test at the outset. IV A With the significant First Amendment costs for individual citizens in mind, we turn to the governmental interests asserted in this case. This Court has identified only one legitimate governmental interest for restricting campaign finances: preventing corruption or the appearance of corruption. See Davis, supra, at 741; National Conservative Political Action Comm., 470 U. S., at 496–497. We have consistently rejected attempts to suppress campaign speech based on other legislative objectives. No matter how desirable it may seem, it is not an acceptable governmental objective to “level the playing field,” or to “level electoral opportunities,” or to “equaliz[e] the financial resources of candidates.” Bennett, 564 U. S., at ___ (slip op., at 22–23); Davis, supra, at 741–742; Buckley, supra, at 56. The First Amendment prohibits such legislative attempts to “fine-tun[e]” the electoral process, no matter how well intentioned. Bennett, supra, at ___ (slip op., at 21). As we framed the relevant principle in Buckley, “the concept that government may restrict the speech of some elements of our society in order to enhance the relative voice of others is wholly foreign to the First Amendment.” 424 U. S., at 48–49. The dissent’s suggestion that Buckley supports the opposite proposition, see post, at 6, simply ignores what Buckley actually said on the matter. See also Citizens Against Rent Control/Coalition for Fair Housing v. Berkeley, 454 U. S. 290, 295 (1981) (“Buckley . . . made clear that contributors cannot be protected from the possibility that others will make larger contributions”). Moreover, while preventing corruption or its appearance is a legitimate objective, Congress may target only a specific type of corruption—“quid pro quo” corruption. As Buckley explained, Congress may permissibly seek to rein in “large contributions [that] are given to secure a political quid pro quo from current and potential office holders.” 424 U. S., at 26. In addition to “actual quid pro quo arrangements,” Congress may permissibly limit “the ap- pearance of corruption stemming from public awareness of the opportunities for abuse inherent in a regime of large individual financial contributions” to particular candidates. Id., at 27; see also Citizens United, 558 U. S., at 359 (“When Buckley identified a sufficiently important governmental interest in preventing corruption or the appearance of corruption, that interest was limited to quid pro quo corruption”). Spending large sums of money in connection with elections, but not in connection with an effort to control the exercise of an officeholder’s official duties, does not give rise to such quid pro quo corruption. Nor does the possibility that an individual who spends large sums may garner “influence over or access to” elected officials or political parties. Id., at 359; see McConnell v. Federal Election Comm’n, 540 U. S. 93, 297 (2003) (Kennedy, J., concurring in judgment in part and dissenting in part). And because the Government’s interest in preventing the appearance of corruption is equally confined to the appearance of quid pro quo corruption, the Government may not seek to limit the appearance of mere influence or access. See Citizens United, 558 U. S., at 360. The dissent advocates a broader conception of corruption, and would apply the label to any individual contributions above limits deemed necessary to protect “collective speech.” Thus, under the dissent’s view, it is perfectly fine to contribute $5,200 to nine candidates but somehow corrupt to give the same amount to a tenth. It is fair to say, as Justice Stevens has, “that we have not always spoken about corruption in a clear or consistent voice.” Id., at 447 (opinion concurring in part and dissenting in part). The definition of corruption that we apply today, however, has firm roots in Buckley itself. The Court in that case upheld base contribution limits because they targeted “the danger of actual quid pro quo arrangements” and “the impact of the appearance of corruption stemming from public awareness” of such a system of unchecked direct contributions. 424 U. S., at 27. Buckley simultaneously rejected limits on spending that was less likely to “be given as a quid pro quo for improper commitments from the candidate.” Id., at 47. In any event, this case is not the first in which the debate over the proper breadth of the Government’s anticorruption interest has been engaged. Compare Citizens United, 558 U. S., at 356–361 (majority opinion), with id., at 447–460 (opinion of Stevens, J.). The line between quid pro quo corruption and general influence may seem vague at times, but the distinction must be respected in order to safeguard basic First Amendment rights. In addition, “[i]n drawing that line, the First Amendment requires us to err on the side of protecting political speech rather than suppressing it.” Federal Election Comm’n v. Wisconsin Right to Life, 551 U. S. 449, 457 (2007) (opinion of Roberts, C. J.). The dissent laments that our opinion leaves only remnants of FECA and BCRA that are inadequate to combat corruption. See post, at 2. Such rhetoric ignores the fact that we leave the base limits undisturbed. [ 6 ] Those base limits remain the primary means of regulating campaign contributions—the obvious explanation for why the aggregate limits received a scant few sentences of attention in Buckley. [ 7 ] B “When the Government restricts speech, the Government bears the burden of proving the constitutionality of its actions.” United States v. Playboy Entertainment Group, Inc., 529 U. S., at 816. Here, the Government seeks to carry that burden by arguing that the aggregate limits further the permissible objective of preventing quid pro quo corruption. The difficulty is that once the aggregate limits kick in, they ban all contributions of any amount. But Congress’s selection of a $5,200 base limit indicates its belief that contributions of that amount or less do not create a cognizable risk of corruption. If there is no corruption concern in giving nine candidates up to $5,200 each, it is difficult to understand how a tenth candidate can be regarded as corruptible if given $1,801, and all others corruptible if given a dime. And if there is no risk that additional candidates will be corrupted by donations of up to $5,200, then the Government must defend the aggregate limits by demonstrating that they prevent circumvention of the base limits. The problem is that they do not serve that function in any meaningful way. In light of the various statutes and regulations currently in effect, Buckley’s fear that an individual might “contribute massive amounts of money to a particular candidate through the use of unearmarked contributions” to entities likely to support the candi- date, 424 U. S., at 38, is far too speculative. And—importantly—we “have never accepted mere conjecture as adequate to carry a First Amendment burden.” Nixon v. Shrink Missouri Government PAC, 528 U. S. 377, 392 (2000) . As an initial matter, there is not the same risk of quid pro quo corruption or its appearance when money flows through independent actors to a candidate, as when a donor contributes to a candidate directly. When an individual contributes to a candidate, a party committee, or a PAC, the individual must by law cede control over the funds. See 2 U. S. C. §441a(a)(8); 11 CFR §110.6. The Government admits that if the funds are subsequently re-routed to a particular candidate, such action occurs at the initial recipient’s discretion—not the donor’s. See Brief for Appellee 37. As a consequence, the chain of attribution grows longer, and any credit must be shared among the various actors along the way. For those reasons, the risk of quid pro quo corruption is generally applicable only to “the narrow category of money gifts that are directed, in some manner, to a candidate or officeholder.” McConnell, 540 U. S., at 310 (opinion of Kennedy, J.). Buckley nonetheless focused on the possibility that “unearmarked contributions” could eventually find their way to a candidate’s coffers. 424 U. S., at 38. Even ac-cepting the validity of Buckley’s circumvention theory, it is hard to see how a candidate today could receive a “massive amount[ ] of money” that could be traced back to a particular contributor uninhibited by the aggregate limits. Ibid. The Government offers a series of scenarios in support of that possibility. But each is sufficiently implausible that the Government has not carried its burden of demonstrating that the aggregate limits further its anticircumvention interest. The primary example of circumvention, in one form or another, envisions an individual donor who contributes the maximum amount under the base limits to a particular candidate, say, Representative Smith. Then the donor also channels “massive amounts of money” to Smith through a series of contributions to PACs that have stated their intention to support Smith. See, e.g., Brief for Appellee 35–37; Tr. of Oral Arg. 4, 6. Various earmarking and antiproliferation rules disarm this example. Importantly, the donor may not contribute to the most obvious PACs: those that support only Smith. See 11 CFR §110.1(h)(1); see also §102.14(a). Nor may the donor contribute to the slightly less obvious PACs that he knows will route “a substantial portion” of his contribution to Smith. §110.1(h)(2). The donor must instead turn to other PACs that are likely to give to Smith. When he does so, however, he discovers that his contribution will be significantly diluted by all the contributions from others to the same PACs. After all, the donor cannot give more than $5,000 to a PAC and so cannot dominate the PAC’s total receipts, as he could when Buckley was decided. 2 U. S. C. §441a(a)(1)(C). He cannot retain control over his contribution, 11 CFR §110.1(h)(3), direct his money “in any way” to Smith, 2 U. S. C. §441a(a)(8), or even imply that he would like his money to be recontributed to Smith, 11 CFR §110.6(b)(1). His salience as a Smith supporter has been diminished, and with it the potential for corruption. It is not clear how many candidates a PAC must support before our dedicated donor can avoid being tagged with the impermissible knowledge that “a substantial portion” of his contribution will go to Smith. But imagine that the donor is one of ten equal donors to a PAC that gives the highest possible contribution to Smith. [ 8 ] The PAC may give no more than $2,600 per election to Smith. Of that sum, just $260 will be attributable to the donor intent on circumventing the base limits. Thus far he has hardly succeeded in funneling “massive amounts of money” to Smith. Buckley, supra, at 38. But what if this donor does the same thing via, say, 100 different PACs? His $260 contribution will balloon to $26,000, ten times what he may contribute directly to Smith in any given election. This 100-PAC scenario is highly implausible. In the first instance, it is not true that the individual donor will necessarily have access to a sufficient number of PACs to effectuate such a scheme. There are many PACs, but they are not limitless. For the 2012 election cycle, the FEC reported about 2,700 nonconnected PACs (excluding PACs that finance independent expenditures only). And not every PAC that supports Smith will work in this scheme: For our donor’s pro rata share of a PAC’s contribution to Smith to remain meaningful, the PAC must be funded by only a small handful of donors. The antiproliferation rules, which were not in effect when Buckley was decided, prohibit our donor from creating 100 pro-Smith PACs of his own, or collaborating with the nine other donors to do so. See 2 U. S. C. §441a(a)(5) (“all contributions made by political committees established or financed or maintained or controlled by . . . any other person, or by any group of such persons, shall be considered to have been made by a single political committee”). Moreover, if 100 PACs were to contribute to Smith and few other candidates, and if specific individuals like our ardent Smith supporter were to contribute to each, the FEC could weigh those “circumstantial factors” to determine whether to deem the PACs affiliated. 11 CFR §100.5(g)(4)(ii). The FEC’s analysis could take account of a “common or overlapping membership” and “similar patterns of contributions or contributors,” among other considerations. §§100.5(g)(4)(ii)(D), (J). The FEC has in the past initiated enforcement proceedings against contributors with such suspicious patterns of PAC donations. See, e.g., Conciliation Agreement, In re Riley, Matters Under Review 4568, 4633, 4634, 4736 (FEC, Dec. 19, 2001). On a more basic level, it is hard to believe that a rational actor would engage in such machinations. In the example described, a dedicated donor spent $500,000—donating the full $5,000 to 100 different PACs—to add just $26,000 to Smith’s campaign coffers. That same donor, meanwhile, could have spent unlimited funds on independent expenditures on behalf of Smith. See Buckley, 424 U. S., at 44–51. Indeed, he could have spent his entire $500,000 advocating for Smith, without the risk that his selected PACs would choose not to give to Smith, or that he would have to share credit with other contributors to the PACs. We have said in the context of independent expenditures that “ ‘[t]he absence of prearrangement and coordination of an expenditure with the candidate or his agent . . . undermines the value of the expenditure to the candidate.’ ” Citizens United, 558 U. S., at 357 (quoting Buckley, supra, at 47). But probably not by 95 percent. And at least from the donor’s point of view, it strikes us as far more likely that he will want to see his full $500,000 spent on behalf of his favored candidate—even if it must be spent independently—rather than see it diluted to a small fraction so that it can be contributed directly by someone else. [ 9 ] Another circumvention example is the one that apparently motivated the District Court. As the District Court crafted the example, a donor gives a $500,000 check to a joint fundraising committee composed of a candidate, a national party committee, and “most of the party’s state party committees” (actually, 47 of the 50). 893 F. Supp. 2d, at 140. The committees divide up the money so that each one receives the maximum contribution permissible under the base limits, but then each transfers its allocated portion to the same single committee. That committee uses the money for coordinated expenditures on behalf of a particular candidate. If that scenario “seem[s] unlikely,” the District Court thought so, too. Ibid. But because the District Court could “imagine” that chain of events, it held that the example substantiated the Government’s circumvention concerns. Ibid. One problem, however, is that the District Court’s speculation relies on illegal earmarking. Lest there be any confusion, a joint fundraising committee is simply a mechanism for individual committees to raise funds collectively, not to circumvent base limits or earmarking rules. See 11 CFR §102.17(c)(5). Under no circumstances may a contribution to a joint fundraising committee result in an allocation that exceeds the contribution limits applicable to its constituent parts; the committee is in fact required to return any excess funds to the contributor. See §102.17(c)(6)(i). The District Court assumed compliance with the specific allocation rules governing joint fundraising committees, but it expressly based its example on the premise that the donor would telegraph his desire to support one candidate and that “many separate entities would willingly serve as conduits for a single contributor’s interests.” 893 F. Supp. 2d, at 140. Regardless whether so many distinct entities would cooperate as a practical matter, the earmarking provision prohibits an individual from directing funds “through an intermediary or conduit” to a particular candidate. 2 U. S. C. §441a(8). Even the “implicit[ ]” agreement imagined by the District Court, 893 F. Supp. 2d, at 140, would trigger the earmarking provision. See 11 CFR §110.6(b)(1). So this circumvention scenario could not succeed without assuming that nearly 50 separate party committees would engage in a transparent violation of the earmarking rules (and that they would not be caught if they did). Moreover, the District Court failed to acknowledge that its $500,000 example cannot apply to most candidates. It crafted the example around a presidential candidate, for whom donations in the thousands of dollars may not seem remarkable—especially in comparison to the nearly $1.4 billion spent by the 2012 presidential candidates. The same example cannot, however, be extrapolated to most House and Senate candidates. Like contributions, coordinated expenditures are limited by statute, with different limits based on the State and the office. See 2 U. S. C. §441a(d)(3). The 2013 coordinated expenditure limit for most House races is $46,600, well below the $500,000 in coordinated expenditures envisioned by the District Court. The limit for Senate races varies significantly based on state population. See 78 Fed. Reg. 8531 (2013). A scheme of the magnitude imagined by the District Court would be possible even in theory for no House candidates and the Senate candidates from just the 12 most populous States. Ibid. Further, to the extent that the law does not foreclose the scenario described by the District Court, experience and common sense do. The Government provides no reason to believe that many state parties would willingly participate in a scheme to funnel money to another State’s candidates. A review of FEC data of Republican and Democratic state party committees for the 2012 election cycle reveals just 12 total instances in which a state party committee contributed to a House or Senate candidate in another State. No surprise there. The Iowa Democratic Party, for example, has little reason to transfer money to the California Democratic Party, especially when the Iowa Democratic Party would be barred for the remainder of the election cycle from receiving another contribution for its own activities from the particular donor. These scenarios, along with others that have been suggested, are either illegal under current campaign finance laws or divorced from reality. The three examples posed by the dissent are no exception. The dissent does not explain how the large sums it postulates can be legally rerouted to a particular candidate, why most state committees would participate in a plan to redirect their donations to a candidate in another State, or how a donor or group of donors can avoid regulations prohibiting con- tributions to a committee “with the knowledge that a substantial portion” of the contribution will support a candidate to whom the donor has already contributed, 11 CFR §110.1(h)(2). The dissent argues that such knowledge may be difficult to prove, pointing to eight FEC cases that did not proceed because of insufficient evidence of a donor’s incriminating knowledge. See post, at 24–25. It might be that such guilty knowledge could not be shown because the donors were not guilty—a possibility that the dissent does not entertain. In any event, the donors described in those eight cases were typically alleged to have exceeded the base limits by $5,000 or less. The FEC’s failure to find the requisite knowledge in those cases hardly means that the agency will be equally powerless to prevent a scheme in which a donor routes millions of dollars in excess of the base limits to a particular candidate, as in the dissent’s “Example Two.” And if an FEC official cannot establish knowledge of circumvention (or establish affiliation) when the same ten donors contribute $10,000 each to 200 newly created PACs, and each PAC writes a $10,000 check to the same ten candidates—the dissent’s “Example Three”—then that official has not a heart but a head of stone. See post, at 19–20, 25. The dissent concludes by citing three briefs for the proposition that, even with the aggregate limits in place, individuals “have transferred large sums of money to specific candidates” in excess of the base limits. Post, at 26. But the cited sources do not provide any real-world examples of circumvention of the base limits along the lines of the various hypotheticals. The dearth of FEC prosecutions, according to the dissent, proves only that people are getting away with it. And the violations that surely must be out there elude detection “because in the real world, the methods of achieving circumvention are more subtle and more complex” than the hypothetical examples. Ibid. This sort of speculation, however, cannot justify the substantial intrusion on First Amendment rights at issue in this case. Buckley upheld aggregate limits only on the ground that they prevented channeling money to candidates beyond the base limits. The absence of such a prospect today belies the Government’s asserted objective of preventing corruption or its appearance. The improbability of circumvention indicates that the aggregate limits instead further the impermissible objective of simply limiting the amount of money in political campaigns. C Quite apart from the foregoing, the aggregate limits violate the First Amendment because they are not “closely drawn to avoid unnecessary abridgment of associational freedoms.” Buckley, 424 U. S., at 25. In the First Amendment context, fit matters. Even when the Court is not applying strict scrutiny, we still require “a fit that is not necessarily perfect, but reasonable; that represents not necessarily the single best disposition but one whose scope is ‘in proportion to the interest served,’ . . . that employs not necessarily the least restrictive means but . . . a means narrowly tailored to achieve the desired objective.” Board of Trustees of State Univ. of N. Y. v. Fox, 492 U. S. 469, 480 (1989) (quoting In re R. M. J., 455 U. S. 191, 203 (1982) ). Here, because the statute is poorly tailored to the Government’s interest in preventing circumvention of the base limits, it impermissibly restricts participation in the political process. 1 The Government argues that the aggregate limits are justified because they prevent an individual from giving to too many initial recipients who might subsequently recontribute a donation. After all, only recontributed funds can conceivably give rise to circumvention of the base limits. Yet all indications are that many types of recipients have scant interest in regifting donations they receive. Some figures might be useful to put the risk of circumvention in perspective. We recognize that no data can be marshaled to capture perfectly the counterfactual world in which aggregate limits do not exist. But, as we have noted elsewhere, we can nonetheless ask “whether experience under the present law confirms a serious threat of abuse.” Federal Election Comm’n v. Colorado Republican Federal Campaign Comm., 533 U. S. 431, 457 (2001) . It does not. Experience suggests that the vast majority of contri- butions made in excess of the aggregate limits are likely to be retained and spent by their recipients rather than rerouted to candidates. In the 2012 election cycle, federal candidates, political parties, and PACs spent a total of $7 billion, according to the FEC. In particular, each national political party’s spending ran in the hundreds of millions of dollars. The National Republican Senatorial Committee (NRSC), National Republican Congressional Committee (NRCC), Democratic Senatorial Campaign Committee (DSCC), and Democratic Congressional Campaign Committee (DCCC), however, spent less than $1 million each on direct candidate contributions and less than $10 million each on coordinated expenditures. Brief for NRSC et al. as Amici Curiae 23, 25 (NRSC Brief). Including both coordinated expenditures and direct candidate contributions, the NRSC and DSCC spent just 7% of their total funds on contributions to candidates and the NRCC and DCCC spent just 3%. Likewise, as explained previously, state parties rarely contribute to candidates in other States. In the 2012 election cycle, the Republican and Democratic state party committees in all 50 States (and the District of Columbia) contributed a paltry $17,750 to House and Senate candidates in other States. The state party committees spent over half a billion dollars over the same time period, of which the $17,750 in contributions to other States’ candidates constituted just 0.003%. As with national and state party committees, candidates contribute only a small fraction of their campaign funds to other candidates. Authorized candidate committees may support other candidates up to a $2,000 base limit. 2 U. S. C. §432(e)(3)(B). In the 2012 election, House candidates spent a total of $1.1 billion. Candidate-to-candidate contributions among House candidates totaled $3.65 million, making up just 0.3% of candidates’ overall spending. NRSC Brief 29. The most that any one individual candidate received from all other candidates was around $100,000. Brief for Appellee 39. The fact is that candidates who receive campaign contributions spend most of the money on themselves, rather than passing along donations to other candidates. In this arena at least, charity begins at home. [ 10 ] Based on what we can discern from experience, the indiscriminate ban on all contributions above the aggregate limits is disproportionate to the Government’s interest in preventing circumvention. The Government has not given us any reason to believe that parties or candidates would dramatically shift their priorities if the aggregate limits were lifted. Absent such a showing, we cannot conclude that the sweeping aggregate limits are appropriately tailored to guard against any contributions that might implicate the Government’s anticircumvention interest. A final point: It is worth keeping in mind that the base limits themselves are a prophylactic measure. As we have explained, “restrictions on direct contributions are preventative, because few if any contributions to candidates will involve quid pro quo arrangements.” Citizens United, 558 U. S., at 357. The aggregate limits are then layered on top, ostensibly to prevent circumvention of the base limits. This “prophylaxis-upon-prophylaxis approach” requires that we be particularly diligent in scrutinizing the law’s fit. Wisconsin Right to Life, 551 U. S., at 479 (opinion of Roberts, C. J.); see McConnell, 540 U. S., at 268–269 (opinion of Thomas, J.). 2 Importantly, there are multiple alternatives available to Congress that would serve the Government’s anticircumvention interest, while avoiding “unnecessary abridgment” of First Amendment rights. Buckley, 424 U. S., at 25. The most obvious might involve targeted restrictions on transfers among candidates and political committees. There are currently no such limits on transfers among party committees and from candidates to party committees. See 2 U. S. C. §441a(a)(4); 11 CFR §113.2(c). Perhaps for that reason, a central concern of the District Court, the Government, multiple amici curiae, and the dissent has been the ability of party committees to transfer money freely. If Congress agrees that this is problematic, it might tighten its permissive transfer rules. Doing so would impose a lesser burden on First Amendment rights, as compared to aggregate limits that flatly ban contributions beyond certain levels. And while the Government has not conceded that transfer restrictions would be a perfect substitute for the aggregate limits, it has recognized that they would mitigate the risk of circumvention. See Tr. of Oral Arg. 29. One possible option for restricting transfers would be to require contributions above the current aggregate limits to be deposited into segregated, nontransferable accounts and spent only by their recipients. Such a solution would address the same circumvention possibilities as the current aggregate limits, while not completely barring contributions beyond the aggregate levels. In addition (or as an alternative), if Congress believes that circumvention is especially likely to occur through creation of a joint fundraising committee, it could require that funds received through those committees be spent by their recipients (or perhaps it could simply limit the size of joint fundraising committees). Such alternatives to the aggregate limits properly refocus the inquiry on the delinquent actor: the recipient of a contribution within the base limits, who then routes the money in a manner that undermines those limits. See Citizens United, supra, at 360–361; cf. Bartnicki v. Vopper, 532 U. S. 514 –530 (2001). Indeed, Congress has adopted transfer restrictions, and the Court has upheld them, in the context of state party spending. See 2 U. S. C. §441i(b). So-called “Levin funds” are donations permissible under state law that may be spent on certain federal election activity—namely, voter registration and identification, get-out-the-vote efforts, or generic campaign activities. Levin funds are raised directly by the state or local party committee that ultimately spends them. §441i(b)(2)(B)(iv). That means that other party committees may not transfer Levin funds, solicit Levin funds on behalf of the particular state or local committee, or engage in joint fundraising of Levin funds. See McConnell, 540 U. S., at 171–173. McConnell upheld those transfer restrictions as “justifiable anticircumvention measures,” though it acknowledged that they posed some associational burdens. Id., at 171. Here, a narrow transfer restriction on contributions that could otherwise be recontributed in excess of the base limits could rely on a similar justification. Other alternatives might focus on earmarking. Many of the scenarios that the Government and the dissent hy-pothesize involve at least implicit agreements to circumvent the base limits—agreements that are already prohibited by the earmarking rules. See 11 CFR §110.6. The FEC might strengthen those rules further by, for exam- ple, defining how many candidates a PAC must support in order to ensure that “a substantial portion” of a do- nor’s contribution is not rerouted to a certain candidate. §110.1(h)(2). Congress might also consider a modified version of the aggregate limits, such as one that prohibits donors who have contributed the current maximum sums from further contributing to political committees that have indicated they will support candidates to whom the donor has already contributed. To be sure, the existing earmarking provision does not define “the outer limit of accept- able tailoring.” Colorado Republican Federal Campaign Comm., 533 U. S., at 462. But tighter rules could have a significant effect, especially when adopted in concert with other measures. We do not mean to opine on the validity of any particular proposal. The point is that there are numerous al- ternative approaches available to Congress to prevent circumvention of the base limits. D Finally, disclosure of contributions minimizes the potential for abuse of the campaign finance system. Disclosure requirements are in part “justified based on a governmental interest in ‘provid[ing] the electorate with information’ about the sources of election-related spending.” Citizens United, 558 U. S., at 367 (quoting Buckley, supra, at 66). They may also “deter actual corruption and avoid the appearance of corruption by exposing large contributions and expenditures to the light of publicity.” Id., at 67. Disclosure requirements burden speech, but—unlike the aggregate limits—they do not impose a ceiling on speech. Citizens United, supra, at 366; but see McConnell, supra, at 275–277 (opinion of Thomas, J.). For that reason, disclosure often represents a less restrictive alternative to flat bans on certain types or quantities of speech. See, e.g., Federal Election Comm’n v. Massachusetts Citizens for Life, Inc., 479 U. S. 238, 262 (1986) . With modern technology, disclosure now offers a particularly effective means of arming the voting public with information. In 1976, the Court observed that Congress could regard disclosure as “only a partial measure.” Buckley, 424 U. S., at 28. That perception was understandable in a world in which information about campaign contributions was filed at FEC offices and was therefore virtually inaccessible to the average member of the public. See Brief for Cause of Action Institute as Amicus Curiae 15–16. Today, given the Internet, disclosure offers much more robust protections against corruption. See Citizens United, supra, at 370–371. Reports and databases are availa- ble on the FEC’s Web site almost immediately after they are filed, supplemented by private entities such as OpenSecrets.org and FollowTheMoney.org. Because massive quantities of information can be accessed at the click of a mouse, disclosure is effective to a degree not possible at the time Buckley, or even McConnell, was decided. The existing aggregate limits may in fact encourage the movement of money away from entities subject to dis-closure. Because individuals’ direct contributions are limited, would-be donors may turn to other avenues for political speech. See Citizens United, supra, at 364. Individuals can, for example, contribute unlimited amounts to 501(c) organizations, which are not required to publicly disclose their donors. See 26 U. S. C. §6104(d)(3). Such organizations spent some $300 million on independent expenditures in the 2012 election cycle. V At oral argument, the Government shifted its focus from Buckley’s anticircumvention rationale to an argument that the aggregate limits deter corruption regardless of their ability to prevent circumvention of the base limits. See Tr. of Oral Arg. 29–30, 50–52. The Government argued that there is an opportunity for corruption whenever a large check is given to a legislator, even if the check consists of contributions within the base limits to be appropriately divided among numerous candidates and committees. The aggregate limits, the argument goes, ensure that the check amount does not become too large. That new rationale for the aggregate limits—embraced by the dissent, see post, at 15–17—does not wash. It dangerously broadens the circumscribed definition of quid pro quo corruption articu- lated in our prior cases, and targets as corruption the general, broad-based support of a political party. In analyzing the base limits, Buckley made clear that the risk of corruption arises when an individual makes large contributions to the candidate or officeholder himself. See 424 U. S., at 26–27. Buckley’s analysis of the aggregate limit under FECA was similarly confined. The Court noted that the aggregate limit guarded against an individual’s funneling—through circumvention—“massive amounts of money to a particular candidate.” Id., at 38 (emphasis added). We have reiterated that understanding several times. See, e.g., National Conservative Political Action Comm., 470 U. S., at 497 (quid pro quo corruption occurs when “[e]lected officials are influenced to act contrary to their obligations of office by the prospect of financial gain to themselves or infusions of money into their campaigns” (emphasis added)); Citizens Against Rent Control/Coalition for Fair Housing v. Berkeley, 454 U. S. 290, 297 (1981) (Buckley’s holding that contribution limits are permissible “relates to the perception of undue influence of large contributors to a candidate”); McConnell, 540 U. S., at 296 (opinion of Kennedy, J.) (quid pro quo corruption in Buckley involved “contributions that flowed to a particular candidate’s benefit” (emphasis added)). Of course a candidate would be pleased with a donor who contributed not only to the candidate himself, but also to other candidates from the same party, to party committees, and to PACs supporting the party. But there is a clear, administrable line between money beyond the base limits funneled in an identifiable way to a candidate—for which the candidate feels obligated—and money within the base limits given widely to a candidate’s party—for which the candidate, like all other members of the party, feels grateful. When donors furnish widely distributed support within all applicable base limits, all members of the party or supporters of the cause may benefit, and the leaders of the party or cause may feel particular gratitude. That gratitude stems from the basic nature of the party system, in which party members join together to further common political beliefs, and citizens can choose to support a party because they share some, most, or all of those beliefs. See Tashjian v. Republican Party of Conn., 479 U. S. 208 –216 (1986). To recast such shared interest, standing alone, as an opportunity for quid pro quo corruption would dramatically expand government regulation of the political process. Cf. California Democratic Party v. Jones, 530 U. S. 567 –573 (2000) (recognizing the Government’s “role to play in structuring and monitoring the election process,” but rejecting “the proposition that party affairs are public affairs, free of First Amendment protections”). The Government suggests that it is the solicitation of large contributions that poses the danger of corruption, see Tr. of Oral Arg. 29–30, 38–39, 50–51; see also post, at 15–16, 20, but the aggregate limits are not limited to any direct solicitation by an officeholder or candidate. Cf. McConnell, supra, at 298–299, 308 (opinion of Kennedy, J.) (rejecting a ban on “soft money” contributions to national parties, but approving a ban on the solicitation of such contributions as “a direct and necessary regulation of federal candidates’ and officeholders’ receipt of quids”). We have no occasion to consider a law that would specifically ban candidates from soliciting donations—within the base limits—that would go to many other candidates, and would add up to a large sum. For our purposes here, it is enough that the aggregate limits at issue are not directed specifically to candidate behavior. * * * For the past 40 years, our campaign finance jurisprudence has focused on the need to preserve authority for the Government to combat corruption, without at the same time compromising the political responsiveness at the heart of the democratic process, or allowing the Government to favor some participants in that process over others. As Edmund Burke explained in his famous speech to the electors of Bristol, a representative owes constituents the exercise of his “mature judgment,” but judgment informed by “the strictest union, the closest correspondence, and the most unreserved communication with his constituents.” The Speeches of the Right Hon. Edmund Burke 129–130 (J. Burke ed. 1867). Constituents have the right to support candidates who share their views and concerns. Representatives are not to follow constituent orders, but can be expected to be cognizant of and responsive to those concerns. Such responsiveness is key to the very concept of self-governance through elected officials. The Government has a strong interest, no less critical to our democratic system, in combatting corruption and its appearance. We have, however, held that this interest must be limited to a specific kind of corruption—quid pro quo corruption—in order to ensure that the Government’s efforts do not have the effect of restricting the First Amendment right of citizens to choose who shall govern them. For the reasons set forth, we conclude that the aggregate limits on contributions do not further the only governmental interest this Court accepted as legitimate in Buckley. They instead intrude without justification on a citizen’s ability to exercise “the most fundamental First Amendment activities.” Buckley, 424 U. S., at 14. The judgment of the District Court is reversed, and the case is remanded for further proceedings. It is so ordered. Notes 1 There are six authorized national party committees: the Republican National Committee, the Democratic National Committee, the National Republican Senatorial Committee, the Democratic Senatorial Campaign Committee, the National Republican Congressional Committee, and the Democratic Congressional Campaign Committee. See . 2 A PAC is a business, labor, or interest group that raises or spends money in connection with a federal election, in some cases by contributing to candidates. A so-called “Super PAC” is a PAC that makes only independent expenditures and cannot contribute to candidates. The base and aggregate limits govern contributions to traditional PACs, but not to independent expenditure PACs. See SpeechNow.org v. Federal Election Comm’n, 599 F. 3d 686, 695–696 (CADC 2010) (en banc). 3 A multicandidate PAC is a PAC with more than 50 contributors that has been registered for at least six months and has made contributions to five or more candidates for federal office. 11 CFR §100.5(e)(3) (2012). PACs that do not qualify as multicandidate PACs must abide by the base limit applicable to individual contributions. 4 The dissent contends that we should remand for development of an evidentiary record before answering the question with which we were presented. See post, at 28–30 (opinion of Breyer, J). But the parties have treated the question as a purely legal one, and the Government has insisted that the aggregate limits can be upheld under the existing record alone. See Tr. of Oral Arg. 43, 55–56. We take the case as it comes to us. 5 See, e.g., Felsenthal, Obama Attends Fundraiser Hosted by Jay-Z, Beyonce, Reuters, Sept. 18, 2012; Coleman, Kid Rock Supports Paul Ryan at Campaign Fundraiser, Rolling Stone, Aug. 25, 2012; Mason, Robert Duvall to Host Romney Fundraiser, L. A. Times, July 25, 2012; Piazza, Hillary Lands 2.5M with Rocket Man, N. Y. Daily News, Apr. 10, 2008, p. 2. 6 The fact that this opinion does not address the base limits also belies the dissent’s concern that we have silently overruled the Court’s holding in McConnell v. Federal Election Comm’n, . See post, at 12–13. At issue in McConnell was BCRA’s extension of the base limits to so-called “soft money”—previously unregulated contributions to national party committees. See 540 U. S., at 142; see also post, at 31–38 (appendix A to opinion of Breyer, J.) (excerpts from McConnell record discussing unregulated “soft money”). Our holding about the constitutionality of the aggregate limits clearly does not overrule McConnell’s holding about “soft money.” 7 It would be especially odd to regard aggregate limits as essential to enforce base limits when state campaign finance schemes typically include base limits but not aggregate limits. Just eight of the 38 States that have imposed base limits on contributions from individuals to candidates have also imposed aggregate limits (excluding restrictions on a specific subset of donors). See Conn. Gen. Stat. §9–611(c) (2013); Me. Rev. Stat. Ann., Tit. 21–A, §1015(3) (Supp. 2013); Md. Elec. Law Code Ann. §13–226(b) (Lexis Supp. 2013); Mass. Gen. Laws, ch. 55, §7A(a)(5) (West 2012); N. Y. Elec. Law Ann. §14–114(8) (West Supp. 2013); R. I. Gen. Laws §17–25–10.1(a)(1) (Lexis 2013); Wis. Stat. §11.26(4) (2007–2008); Wyo. Stat. Ann. §22–25–102(c)(ii) (2013). The Government presents no evidence concerning the circumvention of base limits from the 30 States with base limits but no aggregate limits. 8 Even those premises are generous because they assume that the donor contributes to non-multicandidate PACs, which are relatively rare. Multicandidate PACs, by contrast, must have more than 50 contributors. 11 CFR §100.5(e)(3). The more contributors, of course, the more the donor’s share in any eventual contribution to Smith is diluted. 9 The Justice Department agrees. As Acting Assistant Attorney General Mythili Raman recently testified before Congress: “We anticipate seeing fewer cases of conduit contributions directly to campaign committees or parties, because individuals or corporations who wish to influence elections or officials will no longer need to attempt to do so through conduit contribution schemes that can be criminally prosecuted. Instead, they are likely to simply make unlimited contributions to Super PACs or 501(c)s.” Hearing on Current Issues in Campaign Finance Law Enforcement before the Subcommittee on Crime and Terrorism of the Senate Committee on the Judiciary, 113th Cong., 1st Sess., 3 (2013). 10 In addition, the percentage of contributions above the aggregate limits that even could be used for circumvention is limited by the fact that many of the modes of potential circumvention can be used only once each election. For example, if one donor gives $2,600 to 100 candidates with safe House seats in the hopes that each candidate will reroute $2,000 to Representative Smith, a candidate in a contested district, no other donor can do the same, because the candidates in the safe seats will have exhausted their permissible contributions to Smith. So there is no risk that the circumvention scheme will repeat itself with multiple other would-be donors to Smith.

571.US.191

Petitioner Medtronic, Inc., designs, makes, and sells medical devices. Respondent Mirowski Family Ventures, LLC, owns patents relating to implantable heart stimulators. They have a licensing agreement that permits Medtronic to practice certain Mirowski patents in exchange for royalty payments, and that specifies procedures to identify products covered by the license and to resolve disputes between the parties. Pursuant to those procedures, Mirowski notified Medtronic of its belief that several of Medtronic’s products infringed the licensed patents, and Medtronic then challenged that assertion of infringement in a declaratory judgment action, while accumulating disputed royalties in an escrow account for distribution to the prevailing party. The District Court concluded that Mirowski, as the party asserting infringement, had the burden of proving infringement and that Mirowski had not met that burden. The Federal Circuit disagreed. It acknowledged that a patentee normally bears the burden of proof, but concluded that where the patentee is a declaratory judgment defendant and, like Mirowski, is foreclosed from asserting an infringement counterclaim by the continued existence of a licensing agreement, the party seeking the declaratory judgment, namely Medtronic, bears the burden of persuasion. Held: 1. The Federal Circuit did not lack subject-matter jurisdiction in this case. Title 28 U. S. C. §1338(a) gives federal district courts exclusive jurisdiction over “any civil action arising under any Act of Congress relating to patents,” and §1295(a)(1) gives the Federal Circuit appellate jurisdiction over any case where jurisdiction in the district court “was based, in whole or in part, on section 1338.” The Declaratory Judgment Act does not “extend” the federal courts’ “jurisdiction,” Skelly Oil Co. v. Phillips Petroleum Co., 339 U.S. 667, 671; and federal courts determining declaratory judgment jurisdiction often look to the “character” of the declaratory judgment defendant’s “threatened action,” Public Serv. Comm’n of Utah v. Wycoff Co., 344 U.S. 237, 248, i.e., whether the defendant’s hypothetical “coercive action” “would necessarily present a federal question,” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U.S. 1, 19. Here, if Medtronic had acted consistent with the understanding of its rights that it seeks to establish through the declaratory judgment suit (by ceasing to pay royalties), Mirowski could terminate the license and bring a suit for infringement. That suit would arise under federal patent law because “patent law creates the cause of action.” Christianson v. Colt Industries Operating Corp., 486 U.S. 800, 809. Thus, this declaratory judgment action, which avoids that hypothetical threatened action, also “arises under” federal patent law. See, e.g., Security-First Nat. Bank of Los Angeles v. Franchise Tax Bd. of Cal., 368 U.S. 3, 19. Pp. 4–6. 2. When a licensee seeks a declaratory judgment against a patentee that its products do not infringe the licensed patent, the patentee bears the burden of persuasion on the issue of infringement. Pp. 6–11. (a) This conclusion is strongly supported by three settled legal propositions: First, a patentee ordinarily bears the burden of proving infringement, see, e.g., Agawam Co. v. Jordan, 7 Wall. 583, 609; second, the “operation of the Declaratory Judgment Act” is only “procedural,” Aetna Life Ins. Co. v. Haworth, 300 U.S. 227, 240, leaving “substantive rights unchanged,” Beacon Theatres, Inc. v. Westover, 359 U.S. 500, 509; and third, “the burden of proof” is a “ ‘substantive’ aspect of a claim,” Raleigh v. Illinois Dept. of Revenue, 530 U.S. 15, 20–21. Practical considerations lead to the same conclusion. Shifting the burden based on the form of the action could create postlitigation uncertainty about a patent’s scope. It may also create unnecessary complexity by compelling a licensee to prove a negative. Finally, burden shifting is difficult to reconcile with the Declaratory Judgment Act’s purpose of ameliorating the “dilemma” posed by “putting” one challenging a patent’s scope “to the choice between abandoning his rights or risking” suit, MedImmune, Inc. v. Genentech, Inc., 549 U.S. 118, 129. To the extent that the Federal Circuit’s burden shifting rule makes the declaratory judgment procedure disadvantageous, that rule recreates the dilemma that the Declaratory Judgment Act sought to avoid. Pp. 6–9. (b) Several arguments to the contrary are unconvincing. First, Schaffer v. Weast, 546 U.S. 49, which noted the “ordinary default rule” that “plaintiffs” have the “risk of failing to prove their claims,” does not support the Federal Circuit’s conclusion. Schaffer was not a declaratory judgment case, and it described exceptions to its basic burden of proof rule. For reasons explained in this case, declaratory judgment suits like this one are an exception to Schaffer’s default rule. Second, the fact that the Federal Circuit limited its holding to the circumstance where a license forecloses an infringement counterclaim by a patentee cannot, by itself, show that the holding is legally justified. Third, contrary to one amicus’ concern that this Court’s holding will permit licensees to force patent holders into full-blown infringement litigation, such litigation can occur only when there is a genuine and sufficiently “immedia[te]” dispute about a patent’s validity or application, MedImmune, supra, at 127. Here, Mirowski set this dispute in motion by accusing Medtronic of infringement, and there is no convincing reason why burden of proof law should favor the patentee. General considerations relating to the public interest in maintaining a well-functioning patent system are, at most, in balance, and do not favor changing the ordinary burden of proof rule. Pp. 9–11. 695 F.3d 1266, reversed and remanded. Breyer, J., delivered the opinion for a unanimous Court.

A patentee ordinarily bears the burden of proving infringement. Agawam Co. v. Jordan, 7 Wall. 583, 609 (1869). This case asks us to decide whether the burden of proof shifts when the patentee is a defendant in a declaratory judgment action, and the plaintiff (the potential infringer) seeks a judgment that he does not infringe the patent. We hold that, when a licensee seeks a declaratory judgment against a patentee to establish that there is no infringement, the burden of proving infringement remains with the patentee. We reverse the Federal Circuit’s determination to the contrary. I A We set forth a simplified version of the facts. The parties are Medtronic, Inc., a firm that (among other things) designs, makes, and sells medical devices, and Mirowski Family Ventures, LLC, a firm that owns patents relating to implantable heart stimulators. In 1991 Medtronic and Mirowski entered into an agreement permitting Medtronic to practice certain Mirowski patents in exchange for royalty payments. In less simplified form: Mirowski entered into a license agreement with Eli Lilly & Co., which then sublicensed the Mirowski patents to Medtronic. Guidant Corp. is Eli Lilly’s successor in interest. For present purposes we shall ignore Eli Lilly, Guidant, and other parties on Mirowski’s side, using “Mirowski” to refer to any and all of them. The 1991 agreement also provided that, if Mirowski gave notice to Medtronic that a new Medtronic product “infringe[d]” a Mirowski patent, Medtronic had a choice. App. 13. Medtronic could simply “cure the nonpayment of royalties.” Ibid. Or it could pay royalties and, at the same time, “challenge” the “assertion of infringement of any of the Mirowski patents through a Declaratory Judgment action.” Ibid. Medtronic, of course, might just ignore the agreement and decide not to pay royalties at all, in which case Mirowski would have “the right to terminate the [l]icense,” ibid., and, if it wished, bring an infringement action. In 2006 the parties entered into a further agreement that slightly modified the procedure for resolving disputes. If Medtronic, having received “timely written notice of infringement,” chose to pursue a declaratory judgment action “challenging infringement,” it could “accumulate disputed royalties” in an escrow account. Id., at 24, 27. The prevailing party in the declaratory judgment action would receive the royalties. Id., at 28. In 2007 the parties found themselves in the midst of an “infringement” dispute. Mirowski gave Medtronic notice that it believed seven new Medtronic products violated various claims contained in two of its patents (related to devices that cause the heart’s ventricles to contract simultaneously as the heart beats). Medtronic thought that its products did not infringe Mirowski’s patents, either because the products fell outside the scope of the patent claims or because the patents were invalid. B In 2007 Medtronic brought this declaratory judgment action in Federal District Court in Delaware. It sought a declaration that its products did not infringe Mirowski’s patents and that the patents were invalid. But, as its agreement with Mirowski provided, Medtronic paid all the relevant royalties into an escrow account. The District Court recognized that Mirowski was the defendant in the action. But it nonetheless believed that Mirowski, “[a]s the part[y] asserting infringement,” bore the burden of proving infringement. Medtronic, Inc. v. Boston Scientific Corp., 777 F. Supp. 2d 750, 766 (Del. 2011); see Under Sea Industries, Inc. v. Dacor Corp., 833 F. 2d 1551, 1557 (CA Fed. 1987) (“The burden always is on the patentee to show infringement”). After a bench trial, the court found that Mirowski had not proved infringement, either directly or under the doctrine of equivalents. And since Mirowski, the patentee, bore the burden of proof, it lost. 777 F. Supp. 2d, at 767–770. The Court of Appeals for the Federal Circuit considered the burden of proof question, and it came to the opposite conclusion. It held that Medtronic, the declaratory judgment plaintiff, bore the burden. It acknowledged that normally the patentee, not the accused infringer, bears the burden of proving infringement, and that the burden normally will not “shift” even when the patentee is “a counterclaiming defendant in a declaratory judgment action.” 695 F. 3d 1266, 1272 (2012). Nonetheless, the Court of Appeals believed that a different rule applies where that patentee is a declaratory judgment defendant and, like Mirowski, that patentee/defendant is “foreclosed” from asserting an “infringement counterclaim” by the “continued existence of a license.” Id., at 1274. In that case, the Court of Appeals held, the party “seeking a declaratory judgment of noninfringement,” namely Medtronic, “bears the burden of persuasion.” Ibid. Medtronic sought certiorari, asking us to review the Federal Circuit’s burden of proof rule. In light of the importance of burdens of proof in patent litigation, we granted the petition. II We begin with a jurisdictional matter. An amicus claims that we must vacate the Federal Circuit’s decision because that court lacked subject-matter jurisdiction. Amicus agrees with the parties that 28 U. S. C. §1338(a) gives federal district courts exclusive jurisdiction over “any civil action arising under any Act of Congress relating to patents” (emphasis added). Moreover, the version of §1295(a)(1) governing this appeal gives the Federal Circuit exclusive appellate jurisdiction over any case where jurisdiction in the district court “was based, in whole or in part, on section 1338.” But, amicus says, in determining whether this case is a “civil action arising under” an “Act of Congress relating to patents,” we must look to the nature of the action that the declaratory judgment defendant, namely the patentee, Mirowski, could have brought in the absence of a declaratory judgment. And that action, amicus adds (in its most significant argument against jurisdiction), would not be a patent infringement action but, rather, an action for damages for breach of contract, namely an action for breach of the Mirowski-Medtronic licensing contract, in which patent infringement is the central issue. See Brief for Tessera Technologies, Inc., as Amicus Curiae 2–3. We agree with amicus that the Declaratory Judgment Act does not “extend” the “jurisdiction” of the federal courts. Skelly Oil Co. v. Phillips Petroleum Co., 339 U. S. 667, 671 (1950) . We also agree that federal courts, when determining declaratory judgment jurisdiction, often look to the “character of the threatened action.” Public Serv. Comm’n of Utah v. Wycoff Co., 344 U. S. 237, 248 (1952) . That is to say, they ask whether “a coercive action” brought by “the declaratory judgment defendant” (here Mirowski) “would necessarily present a federal question.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 19 (1983) . But we do not agree with amicus’ characterization of the “threatened” or “coercive” action that Mirowski might have brought. The patent licensing agreement specifies that, if Medtronic stops paying royalties, Mirowski can terminate the contract and bring an ordinary patent infringement action. Such an action would arise under federal patent law because “federal patent law creates the cause of action.” Christianson v. Colt Industries Operating Corp., 486 U. S. 800, 809 (1988) . Amicus says that an infringement suit would be un-likely. But that is not the relevant question. The relevant question concerns the nature of the threatened action in the absence of the declaratory judgment suit. Medtronic believes—and seeks to establish in this declaratory judgment suit—that it does not owe royalties because its products are noninfringing. If Medtronic were to act on that belief (by not paying royalties and not bringing a declaratory judgment action), Mirowski could terminate the license and bring an ordinary federal patent law action for infringement. See Brief for Respondent 48 (acknowledging that if Medtronic had “chosen not to pay the royalties . . . it would have subjected itself to a suit for infringement”). Consequently this declaratory judgment action, which avoids that threatened action, also “arises under” federal patent law. See Franchise Tax Bd., supra, at 19; Wycoff Co., supra, at 248. See also MedImmune, Inc. v. Genentech, Inc., 549 U. S. 118, 128 (2007) (concluding that Article III’s case-or-controversy requirement was satis-fied where a patent licensee faced the threat of suit if it ceased making payments under a license agreement, notwithstanding that the licensee’s continued royalty payments rendered the prospect of such a suit “remote, if not nonexistent”). For this reason we believe that the hypothetical threatened action is properly characterized as an action “arising under an Act of Congress relating to patents.” 28 U. S. C. §1338(a). III We now turn to the question presented. A patent li-censee paying royalties into an escrow account under a pat-ent licensing agreement seeks a declaratory judgment that some of its products are not covered by or do not infringe the patent, and that it therefore does not owe royalties for those products. In that suit, who bears the burden of proof, or, to be more precise, the burden of persuasion? Must the patentee prove infringement or must the licensee prove noninfringement? In our view, the burden of persuasion is with the patentee, just as it would be had the patentee brought an infringement suit. A Simple legal logic, resting upon settled case law, strongly supports our conclusion. It is well established that the burden of proving infringement generally rests upon the patentee. See, e.g., Imhaeuser v. Buerk, 101 U. S. 647, 662 (1880) (“[T]he burden to prove infringement never shifts [to the alleged infringer] if the charge is denied in the plea or answer”); Agawam Co., 7 Wall., at 609 (“Infringement is an affirmative allegation made by the complainant, and the burden of proving it is upon him . . .”). See also Under Sea Industries, 833 F. 2d, at 1557 (“The burden always is on the patentee to show infringement”); 5B Chisum §18.06[1][a], at 18–1180 (2007) (“[T]he burden of proof on factual issues relating to infringement rests upon the patent owner”). We have long considered “the operation of the Declaratory Judgment Act” to be only “procedural,” Aetna Life Ins. Co. v. Haworth, 300 U. S. 227, 240 (1937) , leaving “substantive rights unchanged,” Beacon Theatres, Inc. v. Westover, 359 U. S. 500, 509 (1959) . See also Vaden v. Discover Bank, 556 U. S. 49 , n. 19 (2009); Skelly Oil Co., 339 U. S., at 674 (noting the “limited procedural purpose of the Declaratory Judgment Act”). And we have held that “the burden of proof ” is a “ ‘substantive’ aspect of a claim.” Raleigh v. Illinois Dept. of Revenue, 530 U. S. 15 –21 (2000); Director, Office of Workers’ Compensation Programs v. Greenwich Collieries, 512 U. S. 267, 271 (1994) (“[T]he assignment of the burden of proof is a rule of substantive law . . .”); Garrett v. Moore-McCormack Co., 317 U. S. 239, 249 (1942) (“[T]he burden of proof . . . [is] part of the very substance of [the plaintiff ’s] claim and cannot be considered a mere incident of a form of procedure”). Taken together these three legal propositions indicate that, in a licensee’s declaratory judgment action, the burden of proving infringement should remain with the patentee. Several practical considerations lead to the same conclusion. To shift the burden depending upon the form of the action could create postlitigation uncertainty about the scope of the patent. Suppose the evidence is inconclusive, and an alleged infringer loses his declaratory judgment action because he failed to prove noninfringement. The alleged infringer, or others, might continue to engage in the same allegedly infringing behavior, leaving it to the patentee to bring an infringement action. If the burden shifts, the patentee might lose that action because, the evidence being inconclusive, he failed to prove infringement. So, both sides might lose as to infringement, leaving the infringement question undecided, creating uncertainty among the parties and others who seek to know just what products and processes they are free to use. The example is not fanciful. The Restatement (Second) of Judgments says that relitigation of an issue (say, infringement) decided in one suit “is not precluded” in a subsequent suit where the burden of persuasion “has shifted” from the “party against whom preclusion is sought . . . to his adversary.” Restatement (Second) of Judgments §28(4) (1980). Rather, the “[f ]ailure of one party to carry the burden of persuasion on an issue should not establish the issue in favor of an adversary who otherwise would have the burden of persuasion on that issue in later litigation.” 18 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure §4422, p. 592 (2d ed. 2002). Thus the declaratory judgment suit in the example above would have failed to achieve its object: to provide “an immediate and definitive determination of the legal rights of the parties.” Aetna, supra, at 241. Moreover, to shift the burden can, at least on occasion, create unnecessary complexity by making it difficult for the licensee to understand upon just what theory the patentee’s infringement claim rests. A complex patent can contain many pages of claims and limitations. A patent holder is in a better position than an alleged infringer to know, and to be able to point out, just where, how, and why a product (or process) infringes a claim of that patent. Until he does so, however, the alleged infringer may have to work in the dark, seeking, in his declaratory judg-ment complaint, to negate every conceivable infringement theory. Finally burden shifting here is difficult to reconcile with a basic purpose of the Declaratory Judgment Act. In MedImmune, Inc. v. Genentech, Inc., a case that similarly concerned a patent licensee that brought a declaratory judgment action after the patentee accused it of infringement, we wrote that the “ ‘very purpose’ ” of that Act is to “ ‘ameliorate’ ” the “dilemma” posed by “putting” one who challenges a patent’s scope “to the choice between abandoning his rights or risking” suit. 549 U. S., at 129 (quoting Abbott Laboratories v. Gardner, 387 U. S. 136, 152 (1967) ). In the absence of the declaratory judgment procedure, Medtronic would face the precise dilemma that MedImmune describes. Either Medtronic would have to abandon its right to challenge the scope of Mirowski’s patents, or it would have to stop paying royalties, risk losing an ordinary patent infringement lawsuit, and thereby risk liability for treble damages and attorney’s fees as well as injunctive relief. See 35 U. S. C. §§283–285 (providing for injunctive relief, treble damages, and—in “exceptional cases”—attorney’s fees as remedies for patent infringement). As in MedImmune, the declaratory judgment action rescues Medtronic from this dilemma. The Federal Circuit’s burden shifting rule does not deprive Medtronic of the right to seek a declaratory judgment. But it does create a significant obstacle to use of that action. It makes the declaratory judgment procedure—compared to, say, just refusing to pay royalties—dis-advantageous. To that extent it recreates the dilemma that the Declaratory Judgment Act sought to avoid. As we have made clear (and as we explain below), we are un-aware of any strong reason for creating that obstacle. B We are not convinced by the arguments raised to the contrary. First, the Federal Circuit thought it had found support in a recent case of this Court, Schaffer v. Weast, 546 U. S. 49 (2005) . In that case we referred to the “ordinary default rule” as placing upon the “plaintiffs” the “risk of failing to prove their claims.” Id., at 56. We added that that is because the plaintiffs are normally the parties “seeking relief.” Id., at 58. And Medtronic, not Mirowski, is the declaratory judgment “plaintif [f ]” here. Schaffer, however, was not a declaratory judgment case. And Schaffer described exceptions to its basic burden of proof rule. E.g., id., at 57 (when an element of a claim “can fairly be characterized as [an] affirmative defens[e],” the burden of proof “may be shifted to [the] defendants”); id., at 60 (“ ‘[T]he ordinary rule, based on considerations of fairness, does not place the burden upon a litigant of establishing facts peculiarly within the knowledge of his adversary’ ” (quoting United States v. New York, N. H. & H. R. Co., 355 U. S. 253 , n. 5 (1957)). For the reasons we have set forth in Part III–A, supra, declaratory judgment suits like the one at issue here constitute a further exception to the basic rule Schaffer described. Second, the Federal Circuit emphasized that its holding applied only in “the limited circumstance when an infringement counterclaim by a patentee is foreclosed by the continued existence of a license.” 695 F. 3d, at 1274. The fact that the Federal Circuit’s opinion is limited, how-ever, does not support its conclusion. The “limited cir-cumstance” it described is often present when a patent licensee faces an ordinary but disputed claim of infringement. And that “circumstance” is virtually identical to MedImmune, where we found a declaratory judgment action constitutionally permissible. In any event, the fact that a rule’s scope is limited cannot, by itself, show that the rule is legally justified. Third, an amicus supporting Mirowski fears that our holding, unlike the Federal Circuit’s rule, will “burden . . . patent owners” by permitting “a licensee . . . —at its sole discretion—[to] force the patentee into full-blown patent-infringement litigation.” Brief for Intellectual Property Owners Association as Amicus Curiae 9. The short answer to this argument, however, is that litigation can occur only in the presence of a genuine dispute, “ ‘of sufficient immediacy and reality,’ ” about the patent’s validity or its application. MedImmune, supra, at 127 (quoting Maryland Casualty Co. v. Pacific Coal & Oil Co., 312 U. S. 270, 273 (1941) ). Indeed, it was Mirowski that set the present dispute in motion by accusing Medtronic of infringement. And in such an instance, we see no convincing reason why burden of proof law should favor the patentee. The public interest, of course, favors the maintenance of a well-functioning patent system. But the “public” also has a “paramount interest in seeing that patent monopolies . . . are kept within their legitimate scope.” Precision Instrument Mfg. Co. v. Automotive Maintenance Machinery Co., 324 U. S. 806, 816 (1945) . A patentee “should not be . . . allowed to exact royalties for the use of an idea . . . that is beyond the scope of the patent monopoly granted.” Blonder-Tongue Laboratories, Inc. v. University of Ill. Foundation, 402 U. S. 313 –350 (1971). And “[l]icensees may often be the only individuals with enough economic incentive” to litigate questions of a patent’s scope. Lear, Inc. v. Adkins, 395 U. S. 653, 670 (1969) . The general public interest considerations are, at most, in balance. They do not favor a change in the ordinary rule imposing the burden of proving infringement upon the patentee. For these reasons the judgment of the Federal Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

572.US.782

The State of Michigan, petitioner, entered into a compact with respondent Bay Mills Indian Community pursuant to the Indian Gaming Regulatory Act (IGRA). See 25 U. S. C. §2710(d)(1)(C). The compact authorizes Bay Mills to conduct class III gaming activities (i.e., to operate a casino) on Indian lands located within the State’s borders, but prohibits it from doing so outside that territory. Bay Mills later opened a second casino on land it had purchased through a congressionally established land trust. The Tribe claimed it could operate a casino there because the property qualified as Indian land. Michigan disagreed and sued the Tribe under §2710(d)(7)(A)(ii), which allows a State to enjoin “class III gaming activity located on Indian lands and conducted in violation of any Tribal-State compact.” The District Court granted the injunction, but the Sixth Circuit vacated. It held that tribal sovereign immunity barred the suit unless Congress provided otherwise, and that §2710(d)(7)(A)(ii) only authorized suits to enjoin gaming activity located “on Indian lands,” whereas Michigan’s complaint alleged the casino was outside such territory. Held: Michigan’s suit against Bay Mills is barred by tribal sovereign immunity. Pp. 4–21. (a) As “ ‘domestic dependent nations,’ ” Indian tribes exercise “inherent sovereign authority” that is subject to plenary control by Congress. Oklahoma Tax Comm’n v. Citizen Band Potawatomi Tribe of Okla., 498 U.S. 505, 509. Unless and “until Congress acts, the tribes retain” their historic sovereign authority. United States v. Wheeler, 435 U.S. 313, 323. Among the core aspects of sovereignty that tribes possess—subject to congressional action—is the “common-law immunity from suit traditionally enjoyed by sovereign powers.” Santa Clara Pueblo v. Martinez, 436 U.S. 49, 58. That immunity applies whether a suit is brought by a State, see, e.g., Puyallup Tribe, Inc. v. Department of Game of Wash., 433 U.S. 165, or arises from a tribe’s commercial activities off Indian lands, see Kiowa Tribe of Okla. v. Manufacturing Technologies, Inc., 523 U.S. 751. Therefore, unless Congress has “unequivocally” authorized Michigan’s suit, C & L Enterprises, Inc. v. Citizen Band Potawatomi Tribe of Okla., 532 U.S. 411, 418, it must be dismissed. Pp. 4–8. (b) IGRA’s plain terms do not authorize this suit. Section 2710(d)(7)(A)(ii) partially abrogates tribal immunity with respect to class III gaming located “on Indian lands,” but the very premise of Michigan’s suit is that Bay Mills’ casino is unlawful because it is outside Indian lands. Michigan argues that the casino is authorized, licensed, and operated from within the reservation, and that such administrative action constitutes “class III gaming activity.” However, numerous other IGRA provisions make clear that “class III gaming activity” refers to the gambling that goes on in a casino, not the off-site licensing of such games. See, e.g., §§2710(d)(3)(C)(i), (d)(9). IGRA’s history and design also explain why Congress would have authorized a State to enjoin illegal tribal gaming on Indian lands but not on lands subject to the State’s own sovereign jurisdiction. Congress adopted IGRA in response to California v. Cabazon Band of Mission Indians, 480 U.S. 202, 221–222, which held that States lacked regulatory authority over gaming on Indian lands but left intact States’ regulatory power over tribal gaming outside Indian territory. A State therefore has many tools to enforce its law on state land that it does not possess in Indian territory, including, e.g., bringing a civil or criminal action against tribal officials rather than the tribe itself for conducting illegal gaming. A State can also use its leverage in negotiating an IGRA compact to bargain for a waiver of the tribe’s immunity. Pp. 8–14. (c) Michigan urges the Court to overrule Kiowa and hold that tribal immunity does not apply to commercial activity outside Indian territory. However, “any departure” from precedent “demands special justification,” Arizona v. Rumsey, 467 U.S. 203, 212, and Michigan offers nothing more than arguments already rejected in Kiowa. Kiowa rejected these arguments because it is fundamentally Congress’s job to determine whether or how to limit tribal immunity; Congress had restricted tribal immunity “in limited circumstances” like §2710(d)(7)(A)(ii), while “in other statutes” declaring an “intention not to alter it.” 523 U. S., at 758. Kiowa therefore chose to “defer to the role Congress may wish to exercise in this important judgment.” Ibid. Congress has since reflected on Kiowa and decided to retain tribal immunity in a case like this. Having held that the issue is up to Congress, the Court cannot reverse itself now simply because some may think Congress’s conclusion wrong. Pp. 14–21. 695 F.3d 406, affirmed and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, and Sotomayor, JJ., joined. Sotomayor, J., filed a concurring opinion. Scalia, J., filed a dissenting opinion. Thomas, J., filed a dissenting opinion, in which Scalia, Ginsburg, and Alito, JJ., joined. Ginsburg, J., filed a dissenting opinion.

The question in this case is whether tribal sovereign immunity bars Michigan’s suit against the Bay Mills Indian Community for opening a casino outside Indian lands. We hold that immunity protects Bay Mills from this legal action. Congress has not abrogated tribal sovereign immunity from a State’s suit to enjoin gaming off a reservation or other Indian lands. And we decline to revisit our prior decisions holding that, absent such an abrogation (or a waiver), Indian tribes have immunity even when a suit arises from off-reservation commercial activity. Michigan must therefore resort to other mechanisms, including legal actions against the responsible individuals, to resolve this dispute. I The Indian Gaming Regulatory Act (IGRA or Act), 102Stat. 2467, 25 U. S. C. §2701 et seq., creates a framework for regulating gaming activity on Indian lands.[1] See §2702(3) (describing the statute’s purpose as establishing “regulatory authority . . . [and] standards for gaming on Indian lands”). The Act divides gaming into three classes. Class III gaming, the most closely regulated and the kind involved here, includes casino games, slot machines, and horse racing. See §2703(8). A tribe may conduct such gaming on Indian lands only pursuant to, and in compliance with, a compact it has negotiated with the sur-rounding State. See §2710(d)(1)(C). A compact typically prescribes rules for operating gaming, allocates law enforcement authority between the tribe and State, and provides remedies for breach of the agreement’s terms. See §§2710(d)(3)(C)(ii), (v). Notable here, IGRA itself authorizes a State to bring suit against a tribe for certain conduct violating a compact: Specifically, §2710(d)(7)(A)(ii) allows a State to sue in federal court to “enjoin a class III gaming activity located on Indian lands and conducted in violation of any Tribal-State compact . . . that is in effect.” Pursuant to the Act, Michigan and Bay Mills, a federally recognized Indian Tribe, entered into a compact in 1993. See App. to Pet. for Cert. 73a–96a. The compact empowers Bay Mills to conduct class III gaming on “Indian lands”; conversely, it prohibits the Tribe from doing so outside that territory. Id., at 78a, 83a; see n. 1, supra. The compact also contains a dispute resolution mechanism, which sends to arbitration any contractual differences the parties cannot settle on their own. See App. to Pet. for Cert. 89a–90a. A provision within that arbitration section states that “[n]othing in this Compact shall be deemed a waiver” of either the Tribe’s or the State’s sovereign immunity. Id., at 90a. Since entering into the com-pact, Bay Mills has operated class III gaming, as authorized, on its reservation in Michigan’s Upper Peninsula. In 2010, Bay Mills opened another class III gaming facility in Vanderbilt, a small village in Michigan’s Lower Peninsula about 125 miles from the Tribe’s reservation. Bay Mills had bought the Vanderbilt property with accrued interest from a federal appropriation, which Congress had made to compensate the Tribe for 19th-century takings of its ancestral lands. See Michigan Indian Land Claims Settlement Act, 111Stat. 2652. Congress had directed that a portion of the appropriated funds go into a “Land Trust” whose earnings the Tribe was to use to improve or purchase property. According to the legislation, any land so acquired “shall be held as Indian lands are held.” §107(a)(3), id., at 2658. Citing that provision, Bay Mills contended that the Vanderbilt property was “Indian land” under IGRA and the compact; and the Tribe thus claimed authority to operate a casino there. Michigan disagreed: The State sued Bay Mills in federal court to enjoin operation of the new casino, alleging that the facility violated IGRA and the compact because it was located outside Indian lands. The same day Michigan filed suit, the federal Department of the Interior issued an opinion concluding (as the State’s complaint said) that the Tribe’s use of Land Trust earnings to purchase the Vanderbilt property did not convert it into Indian territory. See App. 69–101. The District Court entered a preliminary injunction against Bay Mills, which promptly shut down the new casino and took an interlocutory appeal. While that appeal was pending, Michigan amended its complaint to join various tribal officials as defendants, as well as to add state law and federal common law claims. The Court of Appeals for the Sixth Circuit then vacated the injunction, holding (among other things) that tribal sovereign immunity barred Michigan’s suit against Bay Mills unless Congress provided otherwise, and that §2710(d)(7)(A)(ii) did not authorize the action. See 695 F. 3d 406, 413–415 (2012). That provision of IGRA, the Sixth Circuit reasoned, permitted a suit against the Tribe to enjoin only gaming activity located on Indian lands, whereas the State’s complaint alleged that the Vanderbilt casino was outside such territory. See id., at 412.[2] Accordingly, the Court of Appeals concluded that Michigan could proceed, if at all, solely against the individual defendants, and it remanded to the District Court to consider those claims. See id., at 416–417.[3] Although no injunc-tion is currently in effect, Bay Mills has not reopened the Vanderbilt casino. We granted certiorari to consider whether tribal sovereign immunity bars Michigan’s suit against Bay Mills, 570 U. S. __ (2013), and we now affirm the Court of Appeals’ judgment. II Indian tribes are “ ‘domestic dependent nations’ ” that exercise “inherent sovereign authority.” Oklahoma Tax Comm’n v. Citizen Band Potawatomi Tribe of Okla., 498 U. S. 505, 509 (1991) (Potawatomi) (quoting Cherokee Nation v. Georgia, 5 Pet. 1, 17 (1831)). As dependents, the tribes are subject to plenary control by Congress. See United States v. Lara, 541 U. S. 193, 200 (2004) (“[T]he Constitution grants Congress” powers “we have consistently described as ‘plenary and exclusive’ ” to “legislate in respect to Indian tribes”). And yet they remain “separate sovereigns pre-existing the Constitution.” Santa Clara Pueblo v. Martinez, 436 U. S. 49, 56 (1978) . Thus, unless and “until Congress acts, the tribes retain” their historic sovereign authority. United States v. Wheeler, 435 U. S. 313, 323 (1978) . Among the core aspects of sovereignty that tribes possess—subject, again, to congressional action—is the “common-law immunity from suit traditionally enjoyed by sovereign powers.” Santa Clara Pueblo, 436 U. S., at 58. That immunity, we have explained, is “a necessary corollary to Indian sovereignty and self-governance.” Three Affiliated Tribes of Fort Berthold Reservation v. Wold Engineering, P. C., 476 U. S. 877, 890 (1986) ; cf. The Federalist No. 81, p. 511 (B. Wright ed. 1961) (A. Hamilton) (It is “inherent in the nature of sovereignty not to be amenable” to suit without consent). And the qualified nature of Indian sovereignty modifies that principle only by placing a tribe’s immunity, like its other governmental powers and attributes, in Congress’s hands. See United States v. United States Fidelity & Guaranty Co., 309 U. S. 506, 512 (1940) (USF&G) (“It is as though the immunity which was theirs as sovereigns passed to the United States for their benefit”). Thus, we have time and again treated the “doctrine of tribal immunity [as] settled law” and dismissed any suit against a tribe absent congressional authorization (or a waiver). Kiowa Tribe of Okla. v. Manufacturing Technologies, Inc., 523 U. S. 751, 756 (1998) . In doing so, we have held that tribal immunity applies no less to suits brought by States (including in their own courts) than to those by individuals. First in Puyallup Tribe, Inc. v. Department of Game of Wash., 433 U. S. 165 –168, 172–173 (1977), and then again in Potawatomi, 498 U. S., at 509–510, we barred a State seeking to enforce its laws from filing suit against a tribe, rejecting arguments grounded in the State’s own sovereignty. In each case, we said a State must resort to other remedies, even if they would be less “efficient.” Id., at 514; see Kiowa, 523 U. S., at 755 (“There is a difference between the right to demand compliance with state laws and the means available to enforce them”). That is because, as we have often stated (and contrary to the dissent’s novel pronouncement, see post, at 3 (opinion of Thomas, J.) (hereinafter the dissent)), tribal immunity “is a matter of federal law and is not subject to diminution by the States.” 523 U. S., at 756 (citing Three Affiliated Tribes, 476 U. S., at 891; Washington v. Confederated Tribes of Colville Reservation, 447 U. S. 134, 154 (1980) ). Or as we elsewhere explained: While each State at the Constitutional Convention surrendered its immunity from suit by sister States, “it would be absurd to suggest that the tribes”—at a conference “to which they were not even parties”—similarly ceded their immunity against state-initiated suits. Blatchford v. Native Village of Noatak, 501 U. S. 775, 782 (1991) . Equally important here, we declined in Kiowa to make any exception for suits arising from a tribe’s commercial activities, even when they take place off Indian lands. In that case, a private party sued a tribe in state court for defaulting on a promissory note. The plaintiff asked this Court to confine tribal immunity to suits involving conduct on “reservations or to noncommercial activities.” 523 U. S., at 758. We said no. We listed Puyallup, Potawa-tomi, and USF&G as precedents applying immunity to a suit predicated on a tribe’s commercial conduct—respectively, fishing, selling cigarettes, and leasing coal mines. 523 U. S., at 754–755. Too, we noted that Puyallup involved enterprise “both on and off [the Tribe’s] reservation.” 523 U. S., at 754 (quoting 433 U. S., at 167). “[O]ur precedents,” we thus concluded, have not previously “drawn the[ ] distinctions” the plaintiff pressed in the case. 523 U. S., at 755. They had established a broad principle, from which we thought it improper suddenly to start carving out exceptions. Rather, we opted to “defer” to Congress about whether to abrogate tribal immunity for off-reservation commercial conduct. Id., at 758, 760; see infra, at 17–18. Our decisions establish as well that such a congressional decision must be clear. The baseline position, we have often held, is tribal immunity; and “[t]o abrogate [such] immunity, Congress must ‘unequivocally’ express that purpose.” C & L Enterprises, Inc. v. Citizen Band Potawatomi Tribe of Okla., 532 U. S. 411, 418 (2001) (quoting Santa Clara Pueblo, 436 U. S., at 58). That rule of construction reflects an enduring principle of Indian law: Although Congress has plenary authority over tribes, courts will not lightly assume that Congress in fact intends to undermine Indian self-government. See, e.g., id., at 58–60; Iowa Mut. Ins. Co. v. LaPlante, 480 U. S. 9, 18 (1987) ; United States v. Dion, 476 U. S. 734 –739 (1986). The upshot is this: Unless Congress has authorized Michigan’s suit, our precedents demand that it be dismissed.[4] And so Michigan, naturally enough, makes two arguments: first, that IGRA indeed abrogates the Tribe’s immunity from the State’s suit; and second, that if it does not, we should revisit—and reverse—our decision in Kiowa, so that tribal immunity no longer applies to claims arising from commercial activity outside Indian lands. We consider—and reject—each contention in turn. III IGRA partially abrogates tribal sovereign immunity in §2710(d)(7)(A)(ii)—but this case, viewed most naturally, falls outside that term’s ambit. The provision, as noted above, authorizes a State to sue a tribe to “enjoin a class III gaming activity located on Indian lands and conducted in violation of any Tribal-State compact.” See supra, at 2; Kiowa, 523 U. S., at 758 (citing the provision as an example of legislation “restrict[ing] tribal immunity from suit in limited circumstances”). A key phrase in that abrogation is “on Indian lands”—three words reflecting IGRA’s overall scope (and repeated some two dozen times in the statute). A State’s suit to enjoin gaming activity on Indian lands (assuming other requirements are met, see n. 6, infra) falls within §2710(d)(7)(A)(ii); a similar suit to stop gaming activity off Indian lands does not. And that creates a fundamental problem for Michigan. After all, the very premise of this suit—the reason Michigan thinks Bay Mills is acting unlawfully—is that the Vanderbilt casino is outside Indian lands. See App. to Pet. for Cert. 59a–60a. By dint of that theory, a suit to enjoin gaming in Vanderbilt is correspondingly outside §2710(d)(7)(A)(ii)’s abrogation of immunity. Michigan first attempts to fit this suit within §2710(d)(7)(A)(ii) by relocating the “class III gaming activity” to which it is objecting. True enough, Michigan states, the Vanderbilt casino lies outside Indian lands. But Bay Mills “authorized, licensed, and operated” that casino from within its own reservation. Brief for Michigan 20. According to the State, that necessary administrative action—no less than, say, dealing craps—is “class III gaming activ-ity,” and because it occurred on Indian land, this suit to enjoin it can go forward. But that argument comes up snake eyes, because numerous provisions of IGRA show that “class III gaming activity” means just what it sounds like—the stuff involved in playing class III games. For example, §2710(d)(3)(C)(i) refers to “the licensing and regulation of [a class III gaming] activity” and §2710(d)(9) concerns the “operation of a class III gaming activity.” Those phrases make perfect sense if “class III gaming activity” is what goes on in a casino—each roll of the dice and spin of the wheel. But they lose all meaning if, as Michigan argues, “class III gaming activity” refers equally to the off-site licensing or operation of the games. (Just plug in those words and see what happens.) See also §§2710(b)(2)(A), (b)(4)(A), (c)(4), (d)(1)(A) (similarly referring to class II or III “gaming activity”). The same holds true throughout the statute. Section 2717(a)(1) specifies fees to be paid by “each gaming operation that conducts a class II or class III gaming activity”—signifying that the gaming activity is the gambling in the poker hall, not the proceedings of the off-site administrative authority. And §§2706(a)(5) and 2713(b)(1) together describe a federal agency’s power to “clos[e] a gaming activity” for “substantial violation[s]” of law—e.g., to shut down crooked blackjack tables, not the tribal regulatory body meant to oversee them. Indeed, consider IGRA’s very first finding: Many tribes, Congress stated, “have licensed gaming activities on Indian lands,” thereby necessitating federal regulation. §2701(1). The “gaming activit[y]” is (once again) the gambling. And that means §2710(d)(7)(A)(ii) does not allow Michigan’s suit even if Bay Mills took action on its reservation to license or oversee the Vanderbilt facility. Stymied under §2710(d)(7)(A)(ii), Michigan next urges us to adopt a “holistic method” of interpreting IGRA that would allow a State to sue a tribe for illegal gaming off, no less than on, Indian lands. Brief for Michigan 30. Michi-gan asks here that we consider “IGRA’s text and structure as a whole.” Id., at 28. But (with one briefly raised exception) Michigan fails to identify any specific textual or structural features of the statute to support its proposed result.[5] Rather, Michigan highlights a (purported) anomaly of the statute as written: that it enables a State to sue a tribe for illegal gaming inside, but not outside, Indian country. “[W]hy,” Michigan queries, “would Congress authorize a state to obtain a federal injunction against illegal tribal gaming on Indian lands, but not on lands subject to the state’s own sovereign jurisdiction?” Reply Brief 1. That question has no answer, Michigan argues: Whatever words Congress may have used in IGRA, it could not have intended that senseless outcome. See Brief for Michigan 28. But this Court does not revise legislation, as Michigan proposes, just because the text as written creates an apparent anomaly as to some subject it does not address. Truth be told, such anomalies often arise from statutes, if for no other reason than that Congress typically legislates by parts—addressing one thing without examining all others that might merit comparable treatment. Rejecting a similar argument that a statutory anomaly (between property and non-property taxes) made “not a whit of sense,” we explained in one recent case that “Congress wrote the statute it wrote”—meaning, a statute going so far and no further. See CSX Transp., Inc. v. Alabama Dept. of Revenue, 562 U. S. ___, ___ (2011) (slip op., at 17–18). The same could be said of IGRA’s abrogation of tribal immunity for gaming “on Indian lands.” This Court has no roving license, in even ordinary cases of statutory interpretation, to disregard clear language simply on the view that (in Michigan’s words) Congress “must have intended” something broader. Brief for Michigan 32. And still less do we have that warrant when the consequence would be to expand an abrogation of immunity, because (as explained earlier) “Congress must ‘unequivocally’ express [its] purpose” to subject a tribe to litigation. C & L Enterprises, 532 U. S., at 418; see supra, at 7. In any event, IGRA’s history and design provide a more than intelligible answer to the question Michigan poses about why Congress would have confined a State’s authority to sue a tribe as §2710(d)(7)(A)(ii) does. Congress adopted IGRA in response to this Court’s decision in California v. Cabazon Band of Mission Indians, 480 U. S. 202 –222 (1987), which held that States lacked any regulatory authority over gaming on Indian lands. Cabazon left fully intact a State’s regulatory power over tribal gaming outside Indian territory—which, as we will soon show, is capacious. See infra, at 12–13. So the problem Congress set out to address in IGRA (Cabazon’s ouster of state authority) arose in Indian lands alone. And the solution Congress devised, naturally enough, reflected that fact. See, e.g., Seminole Tribe of Fla. v. Florida, 517 U. S. 44, 58 (1996) (“[T]he Act grants the States a power that they would not otherwise have, viz., some measure of author-ity over gaming on Indian lands”). Everything—literally everything—in IGRA affords tools (for either state or federal officials) to regulate gaming on Indian lands, and nowhere else. Small surprise that IGRA’s abrogation of tribal immunity does that as well.[6] And the resulting world, when considered functionally, is not nearly so “enigma[tic]” as Michigan suggests. Reply Brief 1. True enough, a State lacks the ability to sue a tribe for illegal gaming when that activity occurs off the reservation. But a State, on its own lands, has many other powers over tribal gaming that it does not possess (absent consent) in Indian territory. Unless federal law provides differently, “Indians going beyond reservation boundaries” are subject to any generally applicable state law. See Wagnon v. Prairie Band Potawatomi Nation, 546 U. S. 95, 113 (2005) (quoting Mescalero Apache Tribe v. Jones, 411 U. S. 145, 148 (1973) ). So, for example, Michigan could, in the first instance, deny a license to Bay Mills for an off-reservation casino. See Mich. Comp. Laws Ann. §§432.206–432.206a (West 2001). And if Bay Mills went ahead anyway, Michigan could bring suit against tribal officials or employees (rather than the Tribe itself ) seeking an injunction for, say, gambling without a license. See §432.220; see also §600.3801(1)(a) (West 2013) (designating illegal gambling facilities as public nuisances). As this Court has stated before, analogizing to Ex parte Young, 209 U. S. 123 (1908) , tribal immunity does not bar such a suit for injunctive relief against individuals, includ-ing tribal officers, responsible for unlawful conduct. See Santa Clara Pueblo, 436 U. S., at 59. And to the extent civil remedies proved inadequate, Michigan could resort to its criminal law, prosecuting anyone who maintains—or even frequents—an unlawful gambling establishment. See Mich. Comp. Laws Ann. §§432.218 (West 2001), 750.303, 750.309 (West 2004). In short (and contrary to the dissent’s unsupported assertion, see post, at 11), the panoply of tools Michigan can use to enforce its law on its own lands—no less than the suit it could bring on Indian lands under §2710(d)(7)(A)(ii)—can shutter, quickly and permanently, an illegal casino.[7] Finally, if a State really wants to sue a tribe for gaming outside Indian lands, the State need only bargain for a waiver of immunity. Under IGRA, a State and tribe negotiating a compact “may include . . . remedies for breach of contract,” 25 U. S. C. §2710(d)(3)(C)(v)—including a provision allowing the State to bring an action against the tribe in the circumstances presented here. States have more than enough leverage to obtain such terms because a tribe cannot conduct class III gaming on its lands without a compact, see §2710(d)(1)(C), and cannot sue to enforce a State’s duty to negotiate a compact in good faith, see Seminole Tribe, 517 U. S., at 47 (holding a State immune from such suits). So as Michigan forthrightly acknowl-edges, “a party dealing with a tribe in contract negotiations has the power to protect itself by refusing to deal absent the tribe’s waiver of sovereign immunity from suit.” Brief for Michigan 40. And many States have taken that path. See Brief for Seminole Tribe of Florida et al. as Amici Curiae 12–22 (listing compacts with waivers of tribal immunity). To be sure, Michigan did not: As noted earlier, the compact at issue here, instead of authorizing judicial remedies, sends disputes to arbitration and expressly retains each party’s sovereign immunity. See supra, at 2. But Michigan—like any State—could have insisted on a different deal (and indeed may do so now for the future, because the current compact has expired and remains in effect only until the parties negotiate a new one, see Tr. of Oral Arg. 21). And in that event, the limitation Congress placed on IGRA’s abrogation of tribal immunity—whether or not anomalous as an abstract matter—would have made no earthly difference. IV Because IGRA’s plain terms do not abrogate Bay Mills’ immunity from this suit, Michigan (and the dissent) must make a more dramatic argument: that this Court should “revisit[ ] Kiowa’s holding” and rule that tribes “have no immunity for illegal commercial activity outside their sovereign territory.” Reply Brief 8, 10; see post, at 1. Michigan argues that tribes increasingly participate in off-reservation gaming and other commercial activity, and operate in that capacity less as governments than as private businesses. See Brief for Michigan 38 (noting, among other things, that “tribal gaming revenues have more than tripled” since Kiowa). Further, Michigan contends, tribes have broader immunity from suits arising from such conduct than other sovereigns—most notably, because Congress enacted legislation limiting foreign nations’ immunity for commercial activity in the United States. See id., at 41; 28 U. S. C. §1605(a)(2). It is time, Michigan concludes, to “level[ ] the playing field.” Brief for Michigan 38. But this Court does not overturn its precedents lightly. Stare decisis, we have stated, “is the preferred course because it promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process.” Payne v. Tennessee, 501 U. S. 808, 827 (1991) . Although “not an inexorable command,” id., at 828, stare decisis is a foundation stone of the rule of law, necessary to ensure that legal rules develop “in a principled and intelligible fashion,” Vasquez v. Hillery, 474 U. S. 254, 265 (1986) . For that reason, this Court has always held that “any departure” from the doctrine “demands special justification.” Arizona v. Rumsey, 467 U. S. 203, 212 (1984) . And that is more than usually so in the circumstances here. First, Kiowa itself was no one-off: Rather, in rejecting the identical argument Michigan makes, our decision reaffirmed a long line of precedents, concluding that “the doctrine of tribal immunity”—without any exceptions for commercial or off-reservation conduct—“is settled law and controls this case.” 523 U. S., at 756; see id., at 754–755; supra, at 5–7. Second, we have relied on Kiowa subsequently: In another case involving a tribe’s off-reservation commercial conduct, we began our analysis with Kiowa’s holding that tribal immunity applies to such activity (and then found that the Tribe had waived its protection). See C & L Enterprises, 532 U. S., at 418. Third, tribes across the country, as well as entities and individuals doing business with them, have for many years relied on Kiowa (along with its forebears and progeny), negotiating their contracts and structuring their transactions against a backdrop of tribal immunity. As in other cases involving contract and property rights, concerns of stare decisis are thus “at their acme.” State Oil Co. v. Khan, 522 U. S. 3, 20 (1997) . And fourth (a point we will later revisit, see infra, at 17–20), Congress exercises primary authority in this area and “remains free to alter what we have done”—another factor that gives “special force” to stare decisis. Patterson v. McLean Credit Union, 491 U. S. 164 –173 (1989). To overcome all these reasons for this Court to stand pat, Michigan would need an ace up its sleeve.[8] But instead, all the State musters are retreads of assertions we have rejected before. Kiowa expressly considered the view, now offered by Michigan, that “when tribes take part in the Nation’s commerce,” immunity “extends beyond what is needed to safeguard tribal self-governance.” 523 U. S., at 758. (Indeed, as Kiowa noted, see id., at 757, Potawatomi had less than a decade earlier rejected Oklahoma’s identical contention that “because tribal business activities . . . are now so detached from traditional tribal interests,” immunity “no longer makes sense in [the commercial] context,” 498 U. S., at 510.) So too, the Kiowa Court comprehended the trajectory of tribes’ commercial activity (which is the dissent’s exclusive rationale for ignoring stare decisis, see post, at 10–13). In the preceding decade, tribal gaming revenues had increased more than thirty fold[9] (dwarfing the still strong rate of growth since that time, see supra, at 14–15); and Kiowa noted the flourishing of other tribal enterprises, ranging from cigarette sales to ski resorts, see 523 U. S., at 758. Moreover, the Kiowa Court understood that other sovereigns did not enjoy similar immunity for commercial activities outside their territory; that seeming “anomal[y]” was a principal point in the dissenting opinion. See id., at 765 (Stevens, J., dissenting). Kiowa did more, in fact, than acknowledge those arguments; it expressed a fair bit of sympathy toward them. See id., at 758 (noting “reasons to doubt the wisdom of perpetuating the doctrine” as to off-reservation commercial conduct). Yet the decision could not have been any clearer: “We decline to draw [any] distinction” that would “confine [immunity] to reservations or to noncommercial activities.” Ibid. We ruled that way for a single, simple reason: because it is fundamentally Congress’s job, not ours, to determine whether or how to limit tribal immunity. The special brand of sovereignty the tribes retain—both its nature and its extent—rests in the hands of Congress. See Lara, 541 U. S., at 200; Wheeler, 435 U. S., at 323. Kiowa chose to respect that congressional responsibility (as Potawatomi had a decade earlier) when it rejected the precursor to Michigan’s argument: Whatever our view of the merits, we explained, “we defer to the role Congress may wish to exercise in this important judgment.” 523 U. S., at 758; see Potawatomi, 498 U. S., at 510 (stating that because “Congress has always been at liberty to dispense with” or limit tribal immunity, “we are not disposed to modify” its scope). Congress, we said—drawing an analogy to its role in shaping foreign sovereign immunity[10]—has the greater capacity “to weigh and accommodate the competing policy concerns and reliance interests” involved in the issue. 523 U. S., at 759. And Congress repeatedly had done just that: It had restricted tribal immunity “in limited circum-stances” (including, we noted, in §2710(d)(7)(A)(ii)), while “in other statutes” declaring an “intention not to alter” the doctrine. Id., at 758; see Potawatomi, 498 U. S., at 510 (citing statutory provisions involving tribal immunity). So too, we thought, Congress should make the call whether to curtail a tribe’s immunity for off-reservation commer-cial conduct—and the Court should accept Congress’s judgment. All that we said in Kiowa applies today, with yet one more thing: Congress has now reflected on Kiowa and made an initial (though of course not irrevocable) decision to retain that form of tribal immunity. Following Kiowa, Congress considered several bills to substantially modify tribal immunity in the commercial context. Two in particular—drafted by the chair of the Senate Appropriations Subcommittee on the Interior—expressly referred to Kiowa and broadly abrogated tribal immunity for most torts and breaches of contract. See S. 2299, 105th Cong., 2d Sess. (1998); S. 2302, 105th Cong., 2d Sess. (1998). But instead of adopting those reversals of Kiowa, Congress chose to enact a far more modest alternative requiring tribes either to disclose or to waive their immunity in contracts needing the Secretary of the Interior’s approval. See Indian Tribal Economic Development and Contract Encouragement Act of 2000, §2, 114Stat. 46 (codified at 25 U. S. C. §81(d)(2)); see also F. Cohen, Handbook of Federal Indian Law §7.05[1][b], p. 643 (2012). Since then, Congress has continued to exercise its plenary authority over tribal immunity, specifically preserving immunity in some contexts and abrogating it in others, but never adopting the change Michigan wants.[11] So rather than confronting, as we did in Kiowa, a legislative vacuum as to the precise issue presented, we act today against the backdrop of a congressional choice: to retain tribal immunity (at least for now) in a case like this one.[12] Reversing Kiowa in these circumstances would scale the heights of presumption: Beyond upending “long-established principle[s] of tribal sovereign immunity,” that action would replace Congress’s considered judgment with our contrary opinion. Potawatomi, 498 U. S., at 510. As Kiowa recognized, a fundamental commitment of Indian law is judicial respect for Congress’s primary role in defining the contours of tribal sovereignty. See 523 U. S., at 758–760; see also Santa Clara Pueblo, 436 U. S., at 60 (“[A] proper respect . . . for the plenary authority of Congress in this area cautions that [the courts] tread lightly”); Cohen, supra, §2.01[1], at 110 (“Judicial deference to the paramount authority of Congress in matters concerning Indian policy remains a central and indispensable principle of the field of Indian law”). That commitment gains only added force when Congress has already reflected on an issue of tribal sovereignty, including immunity from suit, and declined to change settled law. And that force must grow greater still when Congress considered that issue partly at our urging. See Kiowa, 523 U. S., at 758 (hinting, none too subtly, that “Congress may wish to exercise” its authority over the question presented). Having held in Kiowa that this issue is up to Congress, we cannot reverse ourselves because some may think its conclusion wrong. Congress of course may always change its mind—and we would readily defer to that new decision. But it is for Congress, now more than ever, to say whether to create an exception to tribal immunity for off-reservation commercial activity. As in Kiowa—except still more so—“we decline to revisit our case law[,] and choose” instead “to defer to Congress.” Id., at 760. V As “domestic dependent nations,” Indian tribes exercise sovereignty subject to the will of the Federal Government. Cherokee Nation, 5 Pet., at 17. Sovereignty implies immunity from lawsuits. Subjection means (among much else) that Congress can abrogate that immunity as and to the extent it wishes. If Congress had authorized this suit, Bay Mills would have no valid grounds to object. But Congress has not done so: The abrogation of immunity in IGRA applies to gaming on, but not off, Indian lands. We will not rewrite Congress’s handiwork. Nor will we create a freestanding exception to tribal immunity for all off-reservation commercial conduct. This Court has declined that course once before. To choose it now would entail both overthrowing our precedent and usurping Congress’s current policy judgment. Accordingly, Michigan may not sue Bay Mills to enjoin the Vanderbilt casino, but must instead use available alternative means to accomplish that object. We affirm the Sixth Circuit’s judgment and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 The Act defines “Indian lands” as “(A) all lands within the limits of any Indian reservation; and (B) any lands title to which is either held in trust by the United States for the benefit of any Indian tribe or individual[,] or held by any Indian tribe or individual subject to restriction by the United States against alienation and over which an Indian tribe exercises governmental power.” §2703(4). 2 The Sixth Circuit framed part of its analysis in jurisdictional terms, holding that the District Court had no authority to consider Michigan’s IGRA claim because §2710(d)(7)(A)(ii) provides federal jurisdiction only over suits to enjoin gaming on Indian lands (and Michigan’s suit was not that). See 695 F. 3d, at 412–413. That reasoning is wrong, as all parties agree. See Brief for Michigan 22–25; Brief for Bay Mills 23–24; Brief for United States as 16–17. The general federal-question statute, , gives a district court subject matter jurisdiction to decide any claim alleging a violation of IGRA. Nothing in §2710(d)(7)(A)(ii) or any other provision of IGRA limits that grant of jurisdiction (although those provisions may indicate that a party has no statutory right of action). See v. , –644 (2002). 3 The Court of Appeals’ decision applied not only to Michigan’s case, but also to a consolidated case brought by the Little Traverse Bay Bands of Odawa Indians, which operates a casino about 40 miles from the Vanderbilt property. Little Traverse subsequently dismissed its suit, rather than seek review in this Court. 4 Michigan does not argue here that Bay Mills waived its immunity from suit. Recall that the compact expressly preserves both the Tribe’s and the State’s sovereign immunity. See , at 2. 5 Michigan’s single reference to another statutory provision, , does not advance its argument, because that term includes a geographical limitation similar to the one appearing in §2710(d)(7)(A)(ii). Section 1166 makes a State’s gambling laws applicable “in Indian country” as federal law, and then gives the Federal Government “exclusive jurisdiction over criminal prosecutions” for violating those laws. , (d). Michigan briefly argues that, by negative implication, §1166 gives a State the power “to bring a suit to enforce [its] anti-gambling laws in Indian country,” and that this power applies “even when the defendant is an Indian tribe.” Brief for Michigan 26 (emphasis added). Bay Mills and the United States vigorously contest both those propositions, arguing that §1166 gives States no civil enforcement authority at all, much less as against a tribe. See Brief for Bay Mills 30–31; Brief for United States as 20–22. But that dispute is irrelevant here. Even assuming Michigan’s double inference were valid, §1166 would still allow a State to sue a tribe for gaming only “in Indian country.” So Michigan’s suit, alleging that illegal gaming occurred on lands, could no more proceed under §1166 than under §2710(d)(7)(A)(ii). 6 Indeed, the statutory abrogation does not even cover all suits to enjoin gaming on Indian lands, thus refuting the very premise of Michigan’s argument-from-anomaly. Section 2710(d)(7)(A)(ii), recall, allows a State to sue a tribe not for all “class III gaming activity located on Indian lands” (as Michigan suggests), but only for such gaming as is “conducted in violation of any Tribal-State compact. . . that is in effect.” Accordingly, if a tribe opens a casino on Indian lands before negotiating a compact, the surrounding State cannot sue; only the Federal Government can enforce the law. See . To be precise, then, IGRA’s authorization of suit mirrors not the full problem created (a vacuum of state authority over gaming in Indian country) but, more particularly, Congress’s “carefully crafted” compact-based solution to that difficulty. v. , –74 (1996). So Michigan’s binary challenge—if a State can sue to stop gaming in Indian country, why not off?—fails out of the starting gate. In fact, a State sue to enjoin all gaming in Indian country; that gaming must, in addition, violate an agreement that the State and tribe have mutually entered. 7 Michigan contends that these alternative remedies may be more intrusive on, or less respectful of, tribal sovereignty than the suit it wants to bring. See Brief for Michigan 15; Tr. of Oral Arg. 18. Bay Mills, which presumably is better positioned to address that question, emphatically disagrees. See at 32–33. And the law supports Bay Mills’ position: Dispensing with the immunity of a sovereign for fear of pursuing available remedies against its officers or other individuals would upend all known principles of sovereign immunity. 8 Adhering to is particularly appropriate here given that the State, as we have shown, has many alternative remedies: It has no need to sue the Tribe to right the wrong it alleges. See , at 12–13. We need not consider whether the situation would be different if no alternative remedies were available. We have never, for example, specifically addressed (nor, so far as we are aware, has Congress) whether immunity should apply in the ordinary way if a tort victim, or other plaintiff who has not chosen to deal with a tribe, has no alternative way to obtain relief for off-reservation commercial conduct. The argument that such cases would present a “special justification” for abandoning precedent is not before us. v. , . 9 See Nat. Gambling Impact Study Comm’n, Final Report, pp. 6–1 to 6–2 (1999), online at http://govinfo.library.unt.edu/ngisc/reports/6.pdf (as visited Apr. 30, 2014, and available in Clerk of Court’s case file). 10 explained that Congress, in the Foreign Sovereign Immunities Act of 1976, , “den[ied] immunity for the commercial acts of a foreign nation,” codifying an earlier State Department document, known as the Tate Letter, announcing that policy. 523 U. S., at 759. Michigan takes issue with ’s account, maintaining that this Court took the lead in crafting the commercial exception to foreign sovereign immunity, and so should feel free to do the same thing here. See Reply Brief 6–7. But the decision Michigan cites, v. , , does not show what the State would like. First, Michigan points to a part of the opinion commanding only four votes, see , at 695–706 (opinion of White, J.); the majority’s decision was based on the act of state doctrine, not on anything to do with foreign sovereign immunity, see at 690–695. And second, even the plurality opinion relied heavily on the views of the Executive Branch as expressed in the Tate Letter—going so far as to attach that document as an appendix. See at 696–698 (opinion of White, J.); , at 711–715 (appendix 2 to opinion of the Court). The opinion therefore illustrates what highlighted: this Court’s historic practice of “deferr[ing] to the decisions of the political branches,” rather than going it alone, when addressing foreign sovereign immunity. v. , . 11 Compare, , Prevent All Cigarette Trafficking Act of 2009, §§2(e), (3)(a), (preserving immunity), with Arizona Water Settlements Act, §§213(a)(2), 301, (abrogating immunity). The dissent’s claim that “Congress has never granted tribal sovereign immunity in shape or form,” ,at 13, apparently does not take into account the many statutes in which Congress preserved or otherwise ratified tribal immunity. See, , ; see generally , 498 U. S., at 510 (“Congress has consistently reiterated its approval of the immunity doctrine”). 12 The dissent principally counters that this history is not “relevan[t]” because was a “common-law decision.” ,at 14. But that is to ignore what (in line with prior rulings) specifically told Congress: that tribal immunity, far from any old common law doctrine, lies in Congress’s hands to configure. See 523 U. S., at 758; , 498 U. S., at 510; v. , –60 (1978). When we inform Congress that it has primary responsibility over a sphere of law, and invite Congress to consider a specific issue within that sphere, we cannot deem irrelevant how Congress responds.

571.US.161

Congress enacted the Class Action Fairness Act of 2005 (CAFA) to lower diversity jurisdiction requirements in class actions and, as relevant here, in mass actions, i.e., civil actions “in which monetary relief claims of 100 or more persons are proposed to be tried jointly on the ground that the plaintiffs’ claims involve common questions of law or fact,” 28 U. S. C. §1332(d)(11)(B)(i). Petitioner Mississippi sued respondent liquid crystal display (LCD) manufacturers in state court, alleging violations of state law and seeking, inter alia, restitution for LCD purchases made by itself and its citizens. Respondents sought to remove the case to federal court. The District Court held that the suit qualified as a mass action under §1332(d)(11)(B)(i), but remanded the suit to state court on the ground that it fell within CAFA’s “general public” exception, §1332(d)(11)(B)(ii)(III). The Fifth Circuit reversed, agreeing with the District Court that the suit was a mass action but finding the general public exception inapplicable. Held: Because Mississippi is the only named plaintiff, this suit does not constitute a mass action under CAFA. Pp. 5–14. (a) Contrary to respondents’ argument, CAFA’s “100 or more persons” phrase does not encompass unnamed persons who are real parties in interest to claims brought by named plaintiffs. Congress knew how to draft language to that effect when it intended such a meaning, see, e.g., §§1332(d)(5)(B), 1332(d)(1)(D). That it did not do so in the mass action provision indicates that Congress did not want the provision’s numerosity requirement to be satisfied by counting unnamed individuals who possess an interest in the suit. Respondents’ understanding also cannot be reconciled with the fact that the “100 or more persons” are not unspecified individuals with no participation in the suit but are the “plaintiffs” subsequently referred to in the provision, i.e., the very parties proposing to join their claims in a single trial. This is evident in two key ways. First, CAFA uses “persons” and “plaintiffs” the same way they are used in Federal Rule of Civil Procedure 20, which refers to “persons” as individuals who are proposing to join as “plaintiffs” in a single action. Second, it is difficult to imagine how the “claims of 100 or more” unnamed individuals could be “proposed to be tried jointly on the ground that the . . . claims” of some completely different group of named plaintiffs “involve common questions of law or fact.” Construing “plaintiffs” to include both named and unnamed real parties in interest would stretch the meaning of “plaintiff” beyond recognition. A “plaintiff” is commonly understood to be a party who brings a civil suit in a court of law, not anyone, named or unnamed, whom a suit may benefit. Moreover, respondents’ definition would also have to apply to the mass action provision’s subsequent reference to “plaintiffs” in the phrase “jurisdiction shall exist only over those plaintiffs whose claims [exceed $75,000],” §1332(d)(11)(B)(i). See Brown v. Gardner, 513 U.S. 115, 118. This would result in an administrative nightmare that Congress could not possibly have intended, see Griffin v. Oceanic Contractors, Inc., 458 U.S. 564, 575, where district courts would have to identify hundreds (or in this case, hundreds of thousands) of unnamed parties whose claims are for less than $75,000 and then decide how to dispose of their claims. Pp. 5–10. (b) Statutory context reinforces this Court’s reading of the mass action provision. CAFA provides that once removal occurs, a case shall not be transferred to another court “unless a majority of the plaintiffs in the action request transfer.” §1332(d)(11)(C)(i). If “plaintiffs” included unnamed parties, it would be surpassingly difficult for a court to poll the enormous number of real parties in interest to decide whether an action may be transferred. Moreover, respondents’ position that the action here should be removed because it is similar to a class action fails to recognize that the mass action provision functions largely as a backstop to ensure that CAFA’s relaxed class action jurisdictional rules cannot be evaded by a suit that names a host of plaintiffs rather than using the class device. Had Congress wanted CAFA to authorize removal of representative actions brought by States as sole plaintiffs on respondents’ theory, it would have done so through the class action provision, not the mass action provision. Pp. 10–11. (c) This Court has interpreted the diversity jurisdiction statute to require courts in certain contexts to look behind the pleadings to ensure that parties are not improperly creating or destroying diversity jurisdiction, see, e.g., Wecker v. National Enameling & Stamping Co., 204 U.S. 176, 185–186, but Congress did not intend this background inquiry to apply to the mass action provision. First, it could make sense to incorporate the background inquiry into the mass action provision if the inquiry had previously been applied in a similar manner. That is not the case here, however, and so any presumption that Congress wanted to incorporate the inquiry, if it exists at all, would be comparatively weak. Second, even if the background principle had previously been applied in this manner, Congress expressly indicated that it did not want the principle to apply to the mass action provision both through the textual indicators described above and by prohibiting defendants from joining unnamed individuals to a lawsuit in order to turn it into a mass action, §1332(d)(11)(B)(ii)(II). Requiring district courts to identify unnamed persons interested in the suit would run afoul of that intent. Pp. 11–13. 701 F.3d 796, reversed and remanded. Sotomayor, J., delivered the opinion for a unanimous Court.

Under the Class Action Fairness Act of 2005 (CAFA or Act), defendants in civil suits may remove “mass actions” from state to federal court. CAFA defines a “mass action” as “any civil action . . . in which monetary relief claims of 100 or more persons are proposed to be tried jointly on the ground that the plaintiffs’ claims involve common questions of law or fact.” 28 U. S. C. §1332(d)(11)(B)(i). The question presented is whether a suit filed by a State as the sole plaintiff constitutes a “mass action” under CAFA where it includes a claim for restitution based on injuries suffered by the State’s citizens. We hold that it does not. According to CAFA’s plain text, a “mass action” must involve monetary claims brought by 100 or more persons who propose to try those claims jointly as named plaintiffs. Because the State of Mississippi is the only named plaintiff in the instant action, the case must be remanded to state court. I A Congress enacted CAFA in order to “amend the procedures that apply to consideration of interstate class actions.” 119Stat. 4. In doing so, Congress recognized that “[c]lass action lawsuits are an important and valuable part of the legal system.” CAFA §2. It was concerned, however, that certain requirements of federal diversity jurisdic-tion, 28 U. S. C. §1332, had functioned to “kee[p] cases of national importance” in state courts rather than federal courts. CAFA §2. CAFA accordingly loosened the requirements for diversity jurisdiction for two types of cases—“class actions” and “mass actions.” The Act defines “class action” to mean “any civil action filed under rule 23 of the Federal Rules of Civil Procedure or similar State statute or rule of judicial procedure.” 28 U. S. C. §1332(d)(1)(B). And it defines “mass action” to mean “any civil action . . . in which monetary relief claims of 100 or more persons are proposedto be tried jointly on the ground that the plaintiffs’claims involve common questions of law or fact.” §1332(d)(11)(B)(i). For class and mass actions, CAFA expanded diversity jurisdiction in two key ways. First, it replaced the ordinary requirement of complete diversity of citizenship among all plaintiffs and defendants, see State Farm Fire & Casualty Co. v. Tashire, 386 U. S. 523 –531 (1967), with a requirement of minimal diversity. Under that re-quirement, a federal court may exercise jurisdiction over a class action if “any member of a class of plaintiffs isa citizen of a State different from any defendant.” §1332(d)(2)(A). The same rule applies to mass actions. See §1332(d)(11)(A) (“[A] mass action shall be deemed . . . removable under [§§1332(d)(2) through (d)(10)]”). Second, whereas §1332(a) ordinarily requires each plaintiff’s claim to exceed the sum or value of $75,000, see Exxon Mobil Corp. v. Allapattah Services, Inc., 545 U. S. 546 –555 (2005), CAFA grants federal jurisdiction over class and mass actions in which the aggregate amount in contro-versy exceeds $5 million. §§1332(d)(2), (d)(6), (d)(11)(A). Class and mass actions filed in state court that satisfy CAFA’s requirements may be removed to federal court, 28 U. S. C. §1453, but federal jurisdiction in a mass action, unlike a class action, “shall exist only over those plaintiffs” whose claims individually satisfy the $75,000 amount in controversy requirement, §1332(d)(11)(B)(i).[1] B Respondents manufacture liquid crystal displays, or LCDs. In March 2011, the State of Mississippi sued them in state court, alleging that they had formed an international cartel to restrict competition and raise prices inthe LCD market. The State claimed that these actions vio-lated two Mississippi statutes: the Mississippi Antitrust Act, Miss. Code Ann. §75–21–1 et seq. (2009), and the Mississippi Consumer Protection Act, §75–24–1 et seq. (2009 and Cum. Supp. 2013). The State sought injunctive relief and civil penalties under both statutes, along with punitive damages, costs, and attorney’s fees. It also sought restitution for its own purchases “of LCD products and the purchases of its citizens.” App. to Brief in Opposition 65a; §75–24–11. Respondents filed a notice to remove the case from state to federal court, arguing that the case was removable under CAFA as either a “class action” or a “mass action.” The District Court ruled that the suit did not qualify asa “class action” because it was “not brought pursuant to Federal Rule of Civil Procedure 23 or a ‘similar State statute or rule of judicial procedure.’ ” 876 F. Supp. 2d 758, 769 (SD Miss. 2012). But it held that the suit did qualify as a “mass action,” because “[i]t is a civil action ‘in which monetary relief claims of 100 or more persons are proposed to be tried jointly on the ground that the plaintiffs’ claims involve common questions of law or fact.’ ” Id., at 771. The District Court reached that conclusion on the basis of Fifth Circuit precedent in Louisiana ex rel. Caldwell v. Allstate Ins. Co., 536 F. 3d 418 (CA5 2008), which it understood to “stan[d] for the proposition that the words ‘persons’ and ‘plaintiffs’ in [the mass action definition] are to be defined as ‘real parties in interest.’ ” 876 F. Supp. 2d, at 771. Applying that rule, the court found that 100 or more unidentified Mississippi consumers had purchased LCD screens and were therefore real parties in interest to the State’s restitution claim. Ibid. The court noted the “possibility that a ‘mass action’ should be thought of as a ‘mass joinder,’ ”—that is, as a suit involving 100 or more “named plaintiffs.” Ibid., n. 9. But it deemed that interpretation to be foreclosed by Caldwell. The District Court nonetheless remanded the case to state court on the basis of CAFA’s “general public exception,” which excludes from the “mass action” definition “any civil action in which . . . all of the claims in the action are asserted on behalf of the general public (and not on behalf of individual claimants or members of a purported class) pursuant to a State statute specifically authorizing such action.” 28 U. S. C. §1332(d)(11)(B)(ii)(III). The Court of Appeals reversed. 701 F. 3d 796 (CA5 2012). It agreed with the District Court’s determination that Mississippi’s suit is not a “class action” under CAFA.[2] Id., at 799. It also agreed that, under Caldwell, the suit qualifies as a “mass action” because “[t]he real parties in interest in Mississippi’s suit are those more than 100 . . . individual citizens who purchased the [LCD] products within Mississippi.” 701 F. 3d, at 800. It disagreed, however, with the District Court’s ruling that the suit falls within the general public exception. Id., at 802–803.[3] Judge Elrod concurred in the judgment, noting that after the Fifth Circuit’s decision in Caldwell, three Courts of Appeals had deemed similar lawsuits not to be mass actions removable under CAFA.[4] We granted certiorari to resolve this split of authority, 569 U. S. ___ (2013), and now reverse. II A Our analysis begins with the statutory text. Sebelius v. Cloer, 569 U. S. ___, ___ (2013) (slip op., at 6). The statute provides: “[T]he term mass action means any civil action (except a [class action]) in which monetary relief claims of 100 or more persons are proposed to be tried jointly onthe ground that the plaintiffs’ claims involve common questions of law or fact, except that jurisdiction shall exist only over those plaintiffs whose claims in a mass action satisfy the jurisdictional amount requirements under subsection (a).” §1332(d)(11)(B)(i). The parties do not dispute that this provision encompasses suits that are brought jointly by 100 or more named plaintiffs who propose to try their claims together. The question is whether the provision also includes suits brought by fewer than 100 named plaintiffs on the theory that there may be 100 or more unnamed persons who are real parties in interest as beneficiaries to any of the plaintiffs’ claims. Respondents argue that the provision covers such suits because “claims of 100 or more persons” refers to “the persons to whom the claim belongs, i.e., the real parties in interest to the claims,” regardless of whether those persons are named or unnamed. Brief for Respondents 19 (emphasis in original). We disagree. To start, the statute says “100 or more persons,” not “100 or more named or unnamed real parties in interest.” Had Congress intended the latter, it easily could have drafted language to that effect. Indeed, when Congress wanted a numerosity requirement in CAFA to be satisfied by counting unnamed parties in interest in addition to named plaintiffs, it explicitly said so: CAFA provides that in order for a class action to be removable, “the numberof members of all proposed plaintiff classes” must be 100 or greater, §1332(d)(5)(B), and it defines “class members” to mean “the persons (named or unnamed) who fall within the definition of the proposed or certified class,” §1332(d)(1)(D). Congress chose not to use the phrase “named or unnamed” in CAFA’s mass action provision, a decision we understand to be intentional. See Dean v. United States, 556 U. S. 568, 573 (2009) (“ ‘[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion’ ”). More fundamentally, respondents’ interpretation cannot be reconciled with the fact that the “100 or more persons” referred to in the statute are not unspecified individuals who have no actual participation in the suit, but instead the very “plaintiffs” referred to later in the sentence—the parties who are proposing to join their claims in a single trial. Congress made this understanding evident in two key ways. First, we presume that “ ‘Congress is aware of existing law when it passes legislation.’ ” Hall v. United States, 566 U. S. ___, ___ (2012) (slip op., at 9). Here, Congress used the terms “persons” and “plaintiffs” just as they are used in Federal Rule of Civil Procedure 20, governing party joinder. Where §1332(d)(11)(B)(i) requires that the “claims of 100 or more persons [must be] proposed to be tried jointly on the ground that the plaintiffs’ claims involve common questions of law or fact,” Rule 20 provides that “[p]ersons may join in one action as plaintiffs if they assert any right to relief jointly . . . and any question of law or fact common to all plaintiffs will arise in the action.” Thus, just as it is used in Rule 20, the term “persons” in §1332(d)(11)(B)(i) refers to the individuals who are proposing to join as plaintiffs in a single action. Second, respondents’ interpretation of “persons” cannot square with the statute’s requirement that the claims of the “100 or more persons” must be proposed for joint trial “on the ground that the plaintiffs’ claims involve common questions of law or fact.” §1332(d)(11)(B)(i). It is difficult to imagine how the claims of one set of unnamed individ-uals could be proposed for joint trial on the ground that the claims of some completely different group of named plaintiffs share common questions. The better understanding is that Congress meant for the “100 or more persons” and the proposed “plaintiffs” to be one and the same. Recognizing that the statute’s use of the term “persons” could be a reference to proposed plaintiffs, respondents assert that “plaintiffs,” like “persons,” should be construed to “includ[e] both named and unnamed real parties in in-terest.” Brief for Respondents 24. But that stretchesthe meaning of “plaintiff” beyond recognition. The term “plaintiff” is among the most commonly understood of legal terms of art: It means a “party who brings a civil suit in a court of law.” Black’s Law Dictionary 1267 (9th ed. 2009); see also Webster’s Third New International Dictionary 1729 (1961) (defining “plaintiff” to mean “one who commences a personal action or lawsuit,” or “the complaining party in any litigation”). It certainly does not mean “anyone, named or unnamed, whom a suit may benefit,” as respondents suggest.[5] Moreover, Congress used the term “plaintiffs” twice in the mass action provision. The provision encompasses actions in which monetary “claims of 100 or more persons are proposed to be tried jointly on the ground that the plaintiffs’ claims involve common questions,” and it then provides that “jurisdiction shall exist only over those plaintiffs whose claims in a mass action satisfy the jurisdictional amount requiremen[t]” of $75,000. §1332(d)(11)(B)(i). If respondents are correct that “plaintiffs” means unnamed parties in interest where it is used the first time, then so too the second. After all, the “pre-sumption that a given term is used to mean the same thing throughout a statute” is “at its most vigorous when a term is repeated within a given sentence.” Brown v. Gardner, 513 U. S. 115, 118 (1994) . Yet if the term “plaintiffs” is stretched to include all unnamed individuals with an interest in the suit, then §1332(d)(11)(B)(i)’s requirement that “jurisdiction shall exist only over those plaintiffs whose claims [exceed $75,000]” becomes an administrative nightmare that Congress could not possibly have intended, see Griffin v. Oceanic Contractors, Inc., 458 U. S. 564, 575 (1982) . How is a district court to identify the unnamed parties whose claims in a given case are for less than $75,000? Would the court in this case, for instance, have to hold an evidentiary hearing to determine the identity of each of the hundreds of thousands of unnamed Mississippi citizens who purchased one of respondents’ LCD products between 1996 and 2006 (the period alleged in the complaint)? Even if it could identify every such person, how would it ascertain the amount in controversy for each individual claim? Respondents suggest that “[i]n some circumstances, defendants may be able to identify from their payment records any persons who may have claims for overpayments,” but they stop notably short of claiming to possess such decades-old records themselves. Brief for Respondents 25. Furthermore, what would happen with individuals whose claims were valued at less than $75,000? The District Court in this case suggested that if the suit were deemed a mass action, it would sever the claim for “restitution for losses incurred by individuals claiming less than or equal to $75,000 each” and remand that claim back to state court, while allowing the other claims (including the restitution claims exceeding $75,000) to proceed in federal court. 876 F. Supp. 2d, at 775. Even respondents do not defend that outcome, presumably because it would mean that much of the State’s lawsuit could proceed in state court after all, simultaneously with the newly severed parallel federal action.[6] We think it unlikely that Congress intended that federal district courts engage in these unwieldy inquiries. By contrast, interpreting “plaintiffs” in accordance with its usual meaning—to refer to the actual named parties who bring an action—leads to a straightforward, easy to administer rule under which a court would examine whether the plaintiffs have pleaded in good faith the requisite amount. See Horton v. Liberty Mut. Ins. Co., 367 U. S. 348, 353 (1961) . Our decision thus comports with the commonsense observation that “when judges must decide jurisdictional matters, simplicity is a virtue.” See Standard Fire Ins. Co. v. Knowles, 568 U. S. ___, ___ (2013) (slip op., at 6). B Our reading of the mass action provision’s text is reinforced by the statutory context. See Mohamad v. Palestinian Authority, 566 U. S. ___, ___ (2012) (slip op., at 5–6). First, the provision of CAFA governing transfer motions confirms our view that the term “plaintiffs” refers to actual named parties as opposed to unnamed real parties in interest. That provision, §1332(d)(11)(C)(i), provides that once a mass action has been removed to federal court, it “shall not thereafter be transferred to any other court . . . unless a majority of the plaintiffs in the action request transfer.” If respondents are correct that “plaintiffs” means “unnamed parties in interest,” it will be surpassingly difficult for a court to decide in a case like this one whether an action may be transferred. The District Court itself acknowledged this problem, noting that it would have to identify and communicate with “hundreds of thousands if not millions of real parties in interest” to “pol[l] [them] about their preferred forum” if respondents’ interpretation were correct. 876 F. Supp. 2d, at 777. The context in which the mass action provision was enacted lends further support to our conclusion. Congress’ overriding concern in enacting CAFA was with class actions. See Preamble, 119Stat. 4 (describing CAFA as an “[a]ct to amend the procedures that apply to consideration of interstate class actions”); CAFA §2 (Congress’ findings with respect to class actions). The mass action provision thus functions largely as a backstop to ensure that CAFA’s relaxed jurisdictional rules for class actions cannot be evaded by a suit that names a host of plaintiffs rather than using the class device. Respondents’ argument fails to recognize this key distinction. Their position is ultimately that “[t]his action is similar to a class action,” such that it should be removed. Brief for Respondents 27. But if Congress had wanted representative actions brought by States as sole plaintiffs to be removable under CAFA on the theory that they are in substance no different from class actions, it would have done so through the class action provision, not the one governing mass actions.[7] III Rather than relying on the text of CAFA as the source of its real party in interest inquiry, the Court of Appeals appeared to find such an inquiry necessary on the basis of what it understood to be a background principle: that “federal courts look to the substance of the action and not only at the labels that the parties may attach.” Caldwell, 536 F. 3d, at 424. This was error. We have interpreted the diversity jurisdiction statute to require courts in certain contexts to look behind the pleadings to ensure that parties are not improperly creating or destroying diversity jurisdiction. We have held, for example, that a plaintiff may not keep a case out of federal court by fraudulently naming a nondiverse defendant. Wecker v. National Enameling & Stamping Co., 204 U. S. 176 –186 (1907). Nor may a plaintiff create diversity by collusively assigning his interest in an action. Kramer v. Caribbean Mills, Inc., 394 U. S. 823 –830 (1969); see also 28 U. S. C. §1359. And in cases involving a State or state official, we have inquired into the real party in interest because a State’s presence as a party will destroy complete diversity. Missouri, K. & T. R. Co. v. Missouri Railroad and Warehouse Comm’rs, 183 U. S. 53 –59 (1901). But the question in this case is not simply whether there exists some background principle of analyzing the real parties in interest to a suit; the question is whether Congress intended that courts engage in that analysis when deciding whether a suit is a mass action. Recognizing this fact, respondents do not argue that the real party in interest inquiry employed in the above cases somehow supersedes the text of CAFA; they instead argue that we should read CAFA in light of those cases because “ ‘Congress expects its statutes to be read in conformity with this Court’s precedents.’ ” Brief for Respondents 19 (quoting United States v. Wells, 519 U. S. 482, 495 (1997) ). For two reasons, however, we conclude that Congress did not intend the background inquiry to apply to the mass action provision. First, it makes sense to infer Congress’ intent to incorporate a background principle into a new statute where the principle has previously been applied in a similar manner. But that is not the case here. The background real party in interest inquiry identifies what party’s (or parties’) citizenship should be considered in determining diversity. The inquiry that respondents urge is quite dif-ferent: It is an attempt to count up additional unnamed parties in order to satisfy the mass action provision’s numerosity requirement. Respondents offer no reason to believe that Congress intended to extend the real party inquiry to this new circumstance, and so any presumption that Congress wanted to incorporate the inquiry, if it exists in this case at all, would be comparatively weak. Cf. Meyer v. Holley, 537 U. S. 280, 286 (2003) (“Congress’ silence, while permitting an inference that Congress intended to apply ordinary background tort principles, cannot show that it intended to apply an unusual modification of those rules”).[8] Second, even if the background principle had previously been applied in the manner sought by respondents, Congress provided express indications that it did not want the principle to apply to the mass action provision. It specified that “the term ‘mass action’ shall not include any civil action in which . . . the claims are joined upon motion of a defendant.” §1332(d)(11)(B)(ii)(II). By prohibiting defendants from joining unnamed individuals to a lawsuit in order to turn it into a mass action, Congress demonstrated its focus on the persons who are actually proposing to join together as named plaintiffs in the suit. Requiring district courts to pierce the pleadings to identify unnamed persons interested in the suit would run afoul of that intent. Moreover, as already discussed, Congress repeatedly used the word “plaintiffs” to describe the 100 or more persons whose claims must be proposed for a joint trial. That word refers to actual, named parties—a concept inherently at odds with the background inquiry into unnamed real parties in interest, who by definition are never plaintiffs. Congress thus clearly displaced a background real party in interest inquiry, even assuming one might otherwise apply. Cf. Barnhart v. Sigmon Coal Co., 534 U. S. 438, 459, n. 16 (2002) . * * * For the foregoing reasons, the judgment of the United States Court of Appeals for the Fifth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 CAFA provides certain exceptions for class actions that involve matters of principally local or state concern. See 28 U. S. C. §§1332(d)(3)–(5). None of them are at issue in this case. 2 Respondents do not challenge this ruling before this Court. 3 The Court of Appeals did so on the rationale that because individual Mississippi consumers are real parties in interest to the State’s restitution claim, the general public exception’s requirement that “all ofthe claims” must be “asserted on behalf of the general public (and not on behalf of individual claimants)” was not satisfied. . 4 See .v., 699 F. 3d 385 (CA4 2012); v., 672 F. 3d 661 (CA9 2012); v., 665 F. 3d 768 (CA7 2011). 5 Congress could of course require a real party in interest inquiry in a statute that uses the term “plaintiff” simply by saying so. But it has not done that here. 6 Respondents suggest that a district court might be able to exercise supplemental jurisdiction over the claims that fall beneath $75,000, thereby avoiding the problem of identifying and remanding such claims to the state court. We need not decide the issue here, but we note that at least one Court of Appeals has rejected that view. See v., 483 F. 3d 1184, 1206, n. 51 (CA11 2007) (holding that because supplemental jurisdiction does not apply where a federal statute “ ‘expressly provide[s] otherwise,’ ” , the mass action provision’s explicit exclusion of jurisdiction over claims beneath $75,000 negates supplemental jurisdiction over such claims). 7 The parties both point to the “general public exception,” §1332(d)(11)(B)(ii)(III), in support of their respective positions. But because the foregoing arguments resolve this case, we need not construe that provision here. 8 We have also applied a real party in interest inquiry in contexts other than that of determining citizenship for purposes of diversity jurisdiction. See v. , –376 (1923) (state sovereign immunity); v. , –393 (1938) (original jurisdiction). But even if we were to indulge in a presumption that Congress somehow intended to import the inquiry applied in those particular contexts into the mass action provision’s distinct numerosity requirement, we would find any such presumption overridden by CAFA’s text.

573.US.513

Respondent Noel Canning, a Pepsi-Cola distributor, asked the D. C. Circuit to set aside an order of the National Labor Relations Board, claiming that the Board lacked a quorum because three of the five Board members had been invalidly appointed. The nominations of the three members in question were pending in the Senate when it passed a December 17, 2011, resolution providing for a series of “pro forma session[s],” with “no business . . . transacted,” every Tuesday and Friday through January 20, 2012. S. J., 112th Cong., 1st Sess., 923. Invoking the Recess Appointments Clause—which gives the President the power “to fill up all Vacancies that may happen during the Recess of the Senate,” Art. II, §2, cl. 3—the President appointed the three members in question between the January 3 and January 6 pro forma sessions. Noel Canning argued primarily that the appointments were invalid because the 3-day adjournment between those two sessions was not long enough to trigger the Recess Appointments Clause. The D. C. Circuit agreed that the appointments fell outside the scope of the Clause, but on different grounds. It held that the phrase “the recess,” as used in the Clause, does not include intra-session recesses, and that the phrase “vacancies that may happen during the recess” applies only to vacancies that first come into existence during a recess. Held: 1. The Recess Appointments Clause empowers the President to fill any existing vacancy during any recess—intra-session or inter-session—of sufficient length. Pp. 5–33. (a) Two background considerations are relevant to the questions here. First, the Recess Appointments Clause is a subsidiary method for appointing officers of the United States. The Founders intended the norm to be the method of appointment in Article II, §2, cl. 2, which requires Senate approval of Presidential nominations, at least for principal officers. The Recess Appointments Clause reflects the tension between the President’s continuous need for “the assistance of subordinates,” Myers v. United States, 272 U.S. 52, 117, and the Senate’s early practice of meeting for a single brief session each year. The Clause should be interpreted as granting the President the power to make appointments during a recess but not offering the President the authority routinely to avoid the need for Senate confirmation. Second, in interpreting the Clause, the Court puts significant weight upon historical practice. The longstanding “practice of the government,” McCulloch v. Maryland, 4 Wheat. 316, 401, can inform this Court’s determination of “what the law is” in a separation-of-powers case, Marbury v. Madison, 1 Cranch 137, 176. See also, e.g., Mistretta v. United States, 488 U.S. 361, 401; The Pocket Veto Case, 279 U.S. 655, 689–690. There is a great deal of history to consider here, for Presidents have made recess appointments since the beginning of the Republic. Their frequency suggests that the Senate and President have recognized that such appointments can be both necessary and appropriate in certain circumstances. The Court, in interpreting the Clause for the first time, must hesitate to upset the compromises and working arrangements that the elected branches of Government themselves have reached. Pp. 5–9. (b) The phrase “the recess of the Senate” applies to both inter-session recess (i.e., breaks between formal sessions of the Senate) and intra-session recesses (i.e., breaks in the midst of a formal session) of substantial length. The constitutional text is ambiguous. Founding-era dictionaries and usages show that the phrase “the recess” can encompass intra-session breaks. And this broader interpretation is demanded by the purpose of the Clause, which is to allow the President to make appointments so as to ensure the continued functioning of the Government while the Senate is away. The Senate is equally away and unavailable to participate in the appointments process during both an inter-session and an intra-session recess. History offers further support for this interpretation. From the founding until the Great Depression, every time the Senate took a substantial, non-holiday intra-session recess, the President made recess appointments. President Andrew Johnson made the first documented intra-session recess appointments in 1867 and 1868, and Presidents made similar appointments in 1921 and 1929. Since 1929, and particularly since the end of World War II, Congress has shortened its inter-session breaks and taken longer and more frequent intra-session breaks; Presidents accordingly have made more intra-session recess appointments. Meanwhile, the Senate has never taken any formal action to deny the validity of intra-session recess appointments. In 1905, the Senate Judiciary Committee defined “the recess” as “the period of time when the Senate” is absent and cannot “participate as a body in making appointments,” S. Rep. No. 4389, 58th Cong., 3d Sess., p. 2, and that functional definition encompasses both intra-session and inter-session recesses. A 1940 law regulating the payment of recess appointees has also been interpreted functionally by the Comptroller General (an officer of the Legislative Branch). In sum, Presidents have made intra-session recess appointments for a century and a half, and the Senate has never taken formal action to oppose them. That practice is long enough to entitle it to “great weight in a proper interpretation” of the constitutional provision. The Pocket Veto Case, supra, at 689. The Clause does not say how long a recess must be in order to fall within the Clause, but even the Solicitor General concedes that a 3-day recess would be too short. The Adjournments Clause, Art. I, §5, cl. 4, reflects the fact that a 3-day break is not a significant interruption of legislative business. A Senate recess that is so short that it does not require the consent of the House under that Clause is not long enough to trigger the President’s recess-appointment power. Moreover, the Court has not found a single example of a recess appointment made during an intra-session recess that was shorter than 10 days. There are a few examples of inter-session recess appointments made during recesses of less than 10 days, but these are anomalies. In light of historical practice, a recess of more than 3 days but less than 10 days is presumptively too short to fall within the Clause. The word “presumptively” leaves open the possibility that a very unusual circumstance could demand the exercise of the recess-appointment power during a shorter break. Pp. 9–21. (c) The phrase “vacancies that may happen during the recess of the Senate,” Art. II, §2, cl. 3, applies both to vacancies that first come into existence during a recess and to vacancies that initially occur before a recess but continue to exist during the recess. Again, the text is ambiguous. As Thomas Jefferson observed, the Clause is “certainly susceptible of [two] constructions.” Letter to Wilson Cary Nicholas (Jan. 26, 1802), in 36 Papers of Thomas Jefferson 433. It “may mean ‘vacancies that may happen to be’ or ‘may happen to fall’ ” during a recess. Ibid. And, as Attorney General Wirt wrote in 1821, the broader reading is more consonant with the “reason and spirit” of the Clause. 1 Op. Atty. Gen. 632. The purpose of the Clause is to permit the President, who is always acting to execute the law, to obtain the assistance of subordinate officers while the Senate, which acts only in intervals, is unavailable to confirm them. If a vacancy arises too late in the session for the President and Senate to have an opportunity to select a replacement, the narrower reading could paralyze important functions of the Federal Government, particularly at the time of the founding. The broader interpretation ensures that offices needing to be filled can be filled. It does raise a danger that the President may attempt to use the recess-appointment power to circumvent the Senate’s advice and consent role. But the narrower interpretation risks undermining constitutionally conferred powers more seriously and more often. It would prevent a President from making any recess appointment to fill a vacancy that arose before a recess, no matter who the official, how dire the need, how uncontroversial the appointment, and how late in the session the office fell vacant. Historical practice also strongly favors the broader interpretation. The tradition of applying the Clause to pre-recess vacancies dates at least to President Madison. Nearly every Attorney General to consider the question has approved the practice, and every President since James Buchanan has made recess appointments to pre-existing vacancies. It is a fair inference from the historical data that a large proportion of recess appointments over our Nation’s history have filled pre-recess vacancies. The Senate Judiciary Committee in 1863 did issue a report disagreeing with the broader interpretation, and Congress passed a law known as the Pay Act prohibiting payment of recess appointments to pre-recess vacancies soon after. However, the Senate subsequently abandoned its hostility. In 1940, the Senate amended the Pay Act to permit payment of recess appointees in circumstances that would be unconstitutional under the narrower interpretation. In short, Presidents have made recess appointments to preexisting vacancies for two centuries, and the Senate as a body has not countered this practice for nearly three-quarters of a century, perhaps longer. The Court is reluctant to upset this traditional practice where doing so would seriously shrink the authority that Presidents have believed existed and have exercised for so long. Pp. 21–33. 2. For purposes of the Recess Appointments Clause, the Senate is in session when it says that it is, provided that, under its own rules, it retains the capacity to transact Senate business. This standard is consistent with the Constitution’s broad delegation of authority to the Senate to determine how and when to conduct its business, as recognized by this Court’s precedents. See Art. I, §5, cl. 2; Marshall Field & Co. v. Clark, 143 U.S. 649, 672; United States v. Ballin, 144 U.S. 1, 5, 9. Although the Senate’s own determination of when it is and is not in session should be given great weight, the Court’s deference cannot be absolute. When the Senate is without the capacity to act, under its own rules, it is not in session even if it so declares. Under the standard set forth here, the Senate was in session during the pro forma sessions at issue. It said it was in session, and Senate rules make clear that the Senate retained the power to conduct business. The Senate could have conducted business simply by passing a unanimous consent agreement. In fact, it did so; it passed a bill by unanimous consent during its pro forma session on December 23, 2011. See 2011 S. J. 924; Pub. L. 112–78. The Court will not, as the Solicitor General urges, engage in an in-depth factual appraisal of what the Senate actually did during its pro forma sessions in order to determine whether it was in recess or in session for purposes of the Recess Appointments Clause. Because the Senate was in session during its pro forma sessions, the President made the recess appointments at issue during a 3-day recess. Three days is too short a time to bring a recess within the scope of the Clause, so the President lacked the authority to make those appointments. Pp. 33–41. 705 F.3d 490, affirmed. Breyer, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Scalia, J., filed an opinion concurring in the judgment, in which Roberts, C. J., and Thomas and Alito, JJ., joined.

Ordinarily the President must obtain “the Advice and Consent of the Senate” before appointing an “Office[r] of the United States.” U. S. Const., Art. II, §2, cl. 2. But the Recess Appointments Clause creates an exception. It gives the President alone the power “to fill up all Vacancies that may happen during the Recess of the Senate, by granting Commissions which shall expire at the End of their next Session.” Art. II, §2, cl. 3. We here consider three questions about the application of this Clause. The first concerns the scope of the words “recess of the Senate.” Does that phrase refer only to an inter-session recess (i.e., a break between formal sessions of Congress), or does it also include an intra-session recess, such as a summer recess in the midst of a session? We conclude that the Clause applies to both kinds of recess. The second question concerns the scope of the words “vacancies that may happen.” Does that phrase refer only to vacancies that first come into existence during a recess, or does it also include vacancies that arise prior to a recess but continue to exist during the recess? We conclude that the Clause applies to both kinds of vacancy. The third question concerns calculation of the length of a “recess.” The President made the appointments here at issue on January 4, 2012. At that time the Senate was in recess pursuant to a December 17, 2011, resolution providing for a series of brief recesses punctuated by “pro forma session[s],” with “no business . . . transacted,” every Tuesday and Friday through January 20, 2012. S. J., 112th Cong., 1st Sess., 923 (2011) (hereinafter 2011 S. J.). In calculating the length of a recess are we to ignore the pro forma sessions, thereby treating the series of brief recesses as a single, month-long recess? We conclude that we cannot ignore these pro forma sessions. Our answer to the third question means that, when the appointments before us took place, the Senate was in the midst of a 3-day recess. Three days is too short a time to bring a recess within the scope of the Clause. Thus we conclude that the President lacked the power to make the recess appointments here at issue. I The case before us arises out of a labor dispute. The National Labor Relations Board (NLRB) found that a Pepsi-Cola distributor, Noel Canning, had unlawfully refused to reduce to writing and execute a collective-bargaining agreement with a labor union. The Board ordered the distributor to execute the agreement and to make employees whole for any losses. Noel Canning, 358 N. L. R. B. No. 4 (2012). The Pepsi-Cola distributor subsequently asked the Court of Appeals for the District of Columbia Circuit to set the Board’s order aside. It claimed that three of the five Board members had been invalidly appointed, leaving the Board without the three lawfully appointed members necessary for it to act. See 29 U. S. C. §160(f) (providing for judicial review); §153(a) (providing for a 5-member Board); §153(b) (providing for a 3-member quorum); New Process Steel, L. P. v. NLRB, 560 U. S. 674 –688 (2010) (in the absence of a lawfully appointed quorum, the Board cannot exercise its powers). The three members in question were Sharon Block, Richard Griffin, and Terence Flynn. In 2011 the President had nominated each of them to the Board. As of January 2012, Flynn’s nomination had been pending in the Senate awaiting confirmation for approximately a year. The nominations of each of the other two had been pending for a few weeks. On January 4, 2012, the President, invoking the Recess Appointments Clause, appointed all three to the Board. The distributor argued that the Recess Appointments Clause did not authorize those appointments. It pointed out that on December 17, 2011, the Senate, by unanimous consent, had adopted a resolution providing that it would take a series of brief recesses beginning the following day. See 2011 S. J. 923. Pursuant to that resolution, the Senate held pro forma sessions every Tuesday and Friday until it returned for ordinary business on January 23, 2012. Ibid.; 158 Cong. Rec. S1–S11 (Jan. 3–20, 2012). The President’s January 4 appointments were made between the January 3 and January 6 pro forma sessions. In the distributor’s view, each pro forma session terminated the immediately preceding recess. Accordingly, the appointments were made during a 3-day adjournment, which is not long enough to trigger the Recess Appointments Clause. The Court of Appeals agreed that the appointments fell outside the scope of the Clause. But the court set forth different reasons. It held that the Clause’s words “the recess of the Senate” do not include recesses that occur within a formal session of Congress, i.e., intra-session recesses. Rather those words apply only to recesses between those formal sessions, i.e., inter-session recesses. Since the second session of the 112th Congress began on January 3, 2012, the day before the President’s appointments, those appointments occurred during an intra-session recess, and the appointments consequently fell outside the scope of the Clause. 705 F. 3d 490, 499–507 (CADC 2013). The Court of Appeals added that, in any event, the phrase “vacancies that may happen during the recess” applies only to vacancies that come into existence during a recess. Id., at 507–512. The vacancies that Members Block, Griffin, and Flynn were appointed to fill had arisen before the beginning of the recess during which they were appointed. For this reason too the President’s appointments were invalid. And, because the Board lacked a quorum of validly appointed members when it issued its order, the order was invalid. 29 U. S. C. §153(b); New Process Steel, supra. We granted the Solicitor General’s petition for certio-rari. We asked the parties to address not only the Court of Appeals’ interpretation of the Clause but also the distributor’s initial argument, namely, “[w]hether the President’s recess-appointment power may be exercised when the Senate is convening every three days in pro forma sessions.” 570 U. S. ___ (2013). We shall answer all three questions presented. We recognize that the President has nominated others to fill the positions once occupied by Members Block, Griffin, and Flynn, and that the Senate has confirmed these successors. But, as the parties recognize, the fact that the Board now unquestionably has a quorum does not moot the controversy about the validity of the previously entered Board order. And there are pending before us petitions from decisions in other cases involving challenges to the appointment of Board Member Craig Becker. The President appointed Member Becker during an intra-session recess that was not punctuated by pro forma ses-sions, and the vacancy Becker filled had come into existence prior to the recess. See Congressional Research Service, H. Hogue, M. Carey, M. Greene, & M. Bearden, The Noel Canning Decision and Recess Appointments Made from 1981–2013, p. 28 (Feb. 4, 2013) (hereinaf-ter The Noel Canning Decision); NLRB, Members ofthe NLRB since 1935, online at http://www.nlrb.gov/who-we-are/board/members-nlrb-1935 (all Internet materials as visited June 24, 2014, and available in Clerk of Court’s case file). Other cases involving similar challenges are also pending in the Courts of Appeals. E.g., NLRB v. New Vista Nursing & Rehabilitation, No. 11–3440 etc. (CA3). Thus, we believe it is important to answer all three questions that this case presents. II Before turning to the specific questions presented, we shall mention two background considerations that we find relevant to all three. First, the Recess Appointments Clause sets forth a subsidiary, not a primary, method for appointing officers of the United States. The immediately preceding Clause—Article II, Section 2, Clause 2—provides the primary method of appointment. It says that the President “shall nominate, and by and with the Advice and Consent of the Senate, shall appoint Ambassadors, other public Ministers and Consuls, Judges of the supreme Court, and all other Officers of the United States” (emphasis added). The Federalist Papers make clear that the Founders intended this method of appointment, requiring Senate approval, to be the norm (at least for principal officers). Alexander Hamilton wrote that the Constitution vests the power of nomination in the President alone because “one man of discernment is better fitted to analise and estimate the peculiar qualities adapted to particular offices, than a body of men of equal, or perhaps even of superior dis-cernment.” The Federalist No. 76, p. 510 (J. Cooke ed. 1961). At the same time, the need to secure Senate approval provides “an excellent check upon a spirit of favoritism in the President, and would tend greatly to preventing the appointment of unfit characters from State prejudice, from family connection, from personal attachment, or from a view to popularity.” Id., at 513. Hamilton further explained that the “ordinary power of appointment is confided to the President and Senate jointly, and can therefore only be exercised during the session of the Senate; but as it would have been improper to oblige this body to be continually in session for the appointment of officers; and as vacancies might happen in their recess, which it might be necessary for the public service to fill without delay, the succeeding clause is evidently intended to authorise the President singly to make temporary appointments.” Id., No. 67, at 455. Thus the Recess Appointments Clause reflects the tension between, on the one hand, the President’s continuous need for “the assistance of subordinates,” Myers v. United States, 272 U. S. 52, 117 (1926) , and, on the other, the Senate’s practice, particularly during the Republic’s early years, of meeting for a single brief session each year, see Art. I, §4, cl. 2; Amdt. 20, §2 (requiring the Senate to “assemble” only “once in every year”); 3 J. Story, Commentaries on the Constitution of the United States §1551, p. 410 (1833) (it would be “burthensome to the senate, and expensive to the public” to require the Senate to be “perpetually in session”). We seek to interpret the Clause as granting the President the power to make appointments during a recess but not offering the President the author-ity routinely to avoid the need for Senate confirmation. Second, in interpreting the Clause, we put significant weight upon historical practice. For one thing, the inter-pretive questions before us concern the allocation of power between two elected branches of Government. Long ago Chief Justice Marshall wrote that “a doubtful question, one on which human reason may pause, and the human judgment be suspended, in the decision of which the great principles of liberty are not concerned, but the respective powers of those who are equally the representatives of the people, are to be adjusted; if not put at rest by the practice of the government, ought to receive a considerable impression from that practice.” McCulloch v. Maryland, 4 Wheat. 316, 401 (1819). And we later confirmed that “[l]ong settled and established practice is a consideration of great weight in a proper interpretation of constitutional provisions” regulating the relationship between Congress and the President. The Pocket Veto Case, 279 U. S. 655, 689 (1929) ; see also id., at 690 (“[A] practice of at least twenty years duration ‘on the part of the executive department, acquiesced in by the legislative department, . . . is entitled to great regard in determining the true construction of a constitutional provision the phraseology of which is in any respect of doubtful meaning’ ” (quoting State v. South Norwalk, 77 Conn. 257, 264, 58 A. 759, 761 (1904))). We recognize, of course, that the separation of powers can serve to safeguard individual liberty, Clinton v. City of New York, 524 U. S. 417 –450 (1998) (Kennedy, J., concurring), and that it is the “duty of the judicial department”—in a separation-of-powers case as in any other—“to say what the law is,” Marbury v. Madison, 1 Cranch 137, 177 (1803). But it is equally true that the longstanding “practice of the government,” McCulloch, supra, at 401, can inform our determination of “what the law is,” Marbury, supra, at 177. That principle is neither new nor controversial. As James Madison wrote, it “was foreseen at the birth of the Constitution, that difficulties and differences of opinion might occasionally arise in expounding terms & phrases necessarily used in such a charter . . . and that it might require a regular course of practice to liquidate & settle the meaning of some of them.” Letter to Spencer Roane (Sept. 2, 1819), in 8 Writings of James Madison 450 (G. Hunt ed. 1908). And our cases have continually confirmed Madison’s view. E.g., Mistretta v. United States, 488 U. S. 361, 401 (1989) ; Dames & Moore v. Regan, 453 U. S. 654, 686 (1981) ; Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579 –611 (1952) (Frankfurter, J., concurring); The Pocket Veto Case, supra, at 689–690; Ex parte Grossman, 267 U. S. 87 –119 (1925); United States v. Midwest Oil Co., 236 U. S. 459 –474 (1915); McPherson v. Blacker, 146 U. S. 1, 27 (1892) ; McCulloch, supra; Stuart v. Laird, 1 Cranch 299 (1803). These precedents show that this Court has treated practice as an important interpretive factor even when the nature or longevity of that practice is subject to dispute, and even when that practice began after the founding era. See Mistretta, supra, 400–401 (“While these [practices] spawned spirited discussion and frequent criticism, . . . ‘traditional ways of conducting government . . . give meaning’ to the Constitution” (quoting Youngstown, supra, at 610) (Frankfurter, J., concurring)); Regan, supra, at 684 (“[E]ven if the pre-1952 [practice] should be disregarded, congressional acquiescence in [a practice] since that time supports the President’s power to act here”); The Pocket Veto Case, supra, at 689–690 (postfounding practice is entitled to “great weight”); Grossman, supra, at 118–119 (postfounding practice “strongly sustains” a “construction” of the Constitution). There is a great deal of history to consider here. Presidents have made recess appointments since the beginning of the Republic. Their frequency suggests that the Senate and President have recognized that recess appointments can be both necessary and appropriate in certain circumstances. We have not previously interpreted the Clause, and, when doing so for the first time in more than 200 years, we must hesitate to upset the compromises and working arrangements that the elected branches of Government themselves have reached. III The first question concerns the scope of the phrase “the recess of the Senate.” Art. II, §2, cl. 3 (emphasis added). The Constitution provides for congressional elections every two years. And the 2-year life of each elected Congress typically consists of two formal 1-year sessions, each separated from the next by an “inter-session recess.” Congressional Research Service, H. Hogue, Recess Appointments: Frequently Asked Questions 2 (2013). The Senate or the House of Representatives announces an inter-session recess by approving a resolution stating that it will “adjourn sine die,” i.e., without specifying a date to return (in which case Congress will reconvene when the next formal session is scheduled to begin). The Senate and the House also take breaks in the midst of a session. The Senate or the House announces any such “intra-session recess” by adopting a resolution stating that it will “adjourn” to a fixed date, a few days or weeks or even months later. All agree that the phrase “the recess of the Senate” covers inter-session recesses. The question is whether it includes intra-session recesses as well. In our view, the phrase “the recess” includes an intra-session recess of substantial length. Its words taken literally can refer to both types of recess. Founding-era dictionaries define the word “recess,” much as we do today, simply as “a period of cessation from usual work.” 13 The Oxford English Dictionary 322–323 (2d ed. 1989) (hereinafter OED) (citing 18th- and 19th-century sources for that definition of “recess”); 2 N. Webster, An American Dictionary of the English Language (1828) (“[r]emission or suspension of business or procedure”); 2 S. Johnson, A Dictionary of the English Language 1602–1603 (4th ed. 1773) (hereinafter Johnson) (same). The Founders themselves used the word to refer to intra-session, as well as to inter-session, breaks. See, e.g., 3 Records of the Federal Convention of 1787, p. 76 (M. Farrand rev. 1966) (hereinafter Farrand) (letter from George Washington to John Jay using “the recess” to refer to an intra-session break of the Constitutional Convention); id., at 191 (speech of Luther Martin with a similar usage); 1 T. Jefferson, A Manualof Parliamentary Practice §LI, p. 165 (2d ed. 1812) (describing a “recess by adjournment” which did not end a session). We recognize that the word “the” in “the recess” might suggest that the phrase refers to the single break separating formal sessions of Congress. That is because the word “the” frequently (but not always) indicates “a particular thing.” 2 Johnson 2003. But the word can also refer “to a term used generically or universally.” 17 OED 879. The Constitution, for example, directs the Senate to choose a President pro tempore “in the Absence of the Vice-President.” Art. I, §3, cl. 5 (emphasis added). And the Federalist Papers refer to the chief magistrate of an ancient Achaean league who “administered the government in the recess of the Senate.” The Federalist No. 18, at 113 (J. Madison) (emphasis added). Reading “the” generically in this way, there is no linguistic problem applying the Clause’s phrase to both kinds of recess. And, in fact, the phrase “the recess” was used to refer to intra-session recesses at the time of the founding. See, e.g., 3 Farrand 76 (letter from Washington to Jay); New Jersey Legislative-Council Journal, 5th Sess., 1st Sitting 70, 2d Sitting 9 (1781) (twice referring to a 4-month, intra-session break as “the Recess”); see also Brief for Petitioner 14–16 (listing examples). The constitutional text is thus ambiguous. And we believe the Clause’s purpose demands the broader interpretation. The Clause gives the President authority to make appointments during “the recess of the Senate” so that the President can ensure the continued functioning of the Federal Government when the Senate is away. The Senate is equally away during both an inter-session and an intra-session recess, and its capacity to participate in the appointments process has nothing to do with the words it uses to signal its departure. History also offers strong support for the broad interpretation. We concede that pre-Civil War history is not helpful. But it shows only that Congress generally took long breaks between sessions, while taking no significant intra-session breaks at all (five times it took a break of a week or so at Christmas). See Appendix A, infra. Obviously, if there are no significant intra-session recesses, there will be no intra-session recess appointments. In 1867 and 1868, Congress for the first time took substantial, nonholiday intra-session breaks, and President Andrew Johnson made dozens of recess appointments. The Federal Court of Claims upheld one of those specific appointments, writing “[w]e have no doubt that a vacancy occurring while the Senate was thus temporarily adjourned” during the “first session of the Fortieth Congress” was “legally filled by appointment of the President alone.” Gould v. United States, 19 Ct. Cl. 593, 595–596 (1884) (emphasis added). Attorney General Evarts also issued three opinions concerning the constitutionality of President Johnson’s appointments, and it apparently did not occur to him that the distinction between intra-session and inter-session recesses was significant. See 12 Op. Atty. Gen. 449 (1868); 12 Op. Atty. Gen. 455 (1868); 12 Op. Atty. Gen. 469 (1868). Similarly, though the 40th Congress impeached President Johnson on charges relating to his appointment power, he was not accused of violating the Constitution by mak-ing intra-session recess appointments. Hartnett, Recess Appointments of Article III Judges: Three Constitutional Questions, 26 Cardozo L. Rev. 377, 409 (2005). In all, between the founding and the Great Depression, Congress took substantial intra-session breaks (other than holiday breaks) in four years: 1867, 1868, 1921, and 1929. Appendix A, infra. And in each of those years the President made intra-session recess appointments. See App. to Brief for Petitioner 1a–11a. Since 1929, and particularly since the end of World War II, Congress has shortened its inter-session breaks as it has taken longer and more frequent intra-session breaks; Presidents have correspondingly made more intra-session recess appointments. Indeed, if we include military appointments, Presidents have made thousands of intra-session recess appointments. Id., at 11a–64a. President Franklin Roosevelt, for example, commissioned Dwight Eisenhower as a permanent Major General during an intra-session recess; President Truman made Dean Acheson Under Secretary of State; and President George H. W. Bush reappointed Alan Greenspan as Chairman of the Federal Reserve Board. Id., at 11a, 12a, 40a. Justice Scalia does not dispute any of these facts. Not surprisingly, the publicly available opinions of Presidential legal advisers that we have found are nearly unanimous in determining that the Clause authorizes these appointments. In 1921, for example, Attorney General Daugherty advised President Harding that he could make intra-session recess appointments. He reasoned: “If the President’s power of appointment is to be defeated because the Senate takes an adjournment to a specified date, the painful and inevitable result will be measurably to prevent the exercise of governmental functions. I can not bring myself to believe that the framers of the Constitution ever intended such a catastrophe to happen.” 33 Op. Atty. Gen. 20, 23. We have found memoranda offering similar advice to President Eisenhower and to every President from Carter to the present. See 36 Opinion of Office of Legal Counsel (Op. OLC) ___, ___ (2012), online at www.justice.gov/olc/opiniondocslpro-forma-sessions-opinion.pdf; 25 Op. OLC 182 (2001); 20 Op. OLC 124, 161 (1996); 16 Op. OLC 15 (1992); 13 Op. OLC 271 (1989); 6 Op. OLC 585, 586 (1982); 3 Op. OLC 314, 316 (1979); 41 Op. Atty. Gen. 463, 466 (1960). We must note one contrary opinion authored by President Theodore Roosevelt’s Attorney General Philander Knox. Knox advised the President that the Clause did not cover a 19–day intra-session Christmas recess. 23 Op. Atty. Gen. 599 (1901). But in doing so he relied heavily upon the use of the word “the,” a linguistic point that we do not find determinative. See supra, at 10. And Knox all but confessed that his interpretation ran contrary to the basic purpose of the Clause. For it would permit the Senate to adjourn for “several months,” to a fixed date, and thereby “seriously curtail the President’s power of making recess appointments.” 23 Op. Atty. Gen., at 603. Moreover, only three days before Knox gave his opinion, the Solicitor of the Treasury came to the opposite conclusion. Reply Brief 7, n. 5. We therefore do not think Knox’s isolated opinion can disturb the consensus advice within the Executive Branch taking the opposite position. What about the Senate? Since Presidents began making intra-session recess appointments, individual Senators have taken differing views about the proper definition of “the recess.” See, e.g., 130 Cong. Rec. 23234 (1984) (resolution introduced by Senator Byrd urging limits on the length of applicable intra-session recesses); Brief for Sen. Mitch McConnell et al. as Amici Curiae 26 (an intra-session adjournment does not count as “the recess”); Brief for Sen. Edward M. Kennedy as Amicus Curiae in Franklin v. United States, O. T. 2004, No. 04–5858, p. 5 (same). But neither the Senate considered as a body nor its committees, despite opportunities to express opposition to the practice of intra-session recess appointments, has done so. Rather, to the extent that the Senate or a Senate committee has expressed a view, that view has favored a functional definition of “recess,” and a functional definition encompasses intra-session recesses. Most notably, in 1905 the Senate Committee on the Judiciary objected strongly to President Theodore Roosevelt’s use of the Clause to make more than 160 recess appointments during a “fictitious” inter-session recess. S. Rep. No. 4389, 58th Cong., 3d Sess., p. 2 (hereinafter 1905 Senate Report). At noon on December 7, 1903, the Senate President pro tempore had “declare[d]” a formal, “extraordinary session” of the Senate “adjourned without day,” and the next formal Senate session began immediately afterwards. 37 Cong. Rec. 544 (1903). President Roosevelt made over 160 recess appointments during the instantaneous inter-session interval. The Judiciary Committee, when stating its strong objection, defined “recess” in functional terms as “the period of time when the Senate is not sitting in regular or extraordinary session as a branch of the Congress . . . ; when its members owe no duty of attendance; when its Chamber is empty; when, because of its absence, it can not receive communications from the President or participate as a body in making appointments.” 1905 Senate Report, at 2 (emphasisdeleted). That functional definition encompasses intra-session, as well as inter-session, recesses. Justice Scalia is right that the 1905 Report did not specifically address the dis-tinction between inter-session and intra-session recesses. But the animating principle of the Report—that “recess” should be practically construed to mean a time when the Senate is unavailable to participate in the appointments process—is inconsistent with the formalistic approach that Justice Scalia endorses. Similarly, in 1940 the Senate helped to enact a law regulating the payment of recess appointees, and the Comptroller General of the United States has interpreted that law functionally. An earlier 1863 statute had denied pay to individuals appointed to fill up vacancies first arising prior to the beginning of a recess. The Senate Judiciary Committee then believed that those vacancies fell outside the scope of the Clause. See infra, at 30. In 1940, however, the Senate amended the law to permit many of those recess appointees to be paid. Act of July 11, 54Stat. 751. Interpreting the amendments in 1948, the Comptroller General—who, unlike the Attorney General, is an “officer of the Legislative Branch,” Bowsher v. Synar, 478 U. S. 714, 731 (1986) —wrote: “I think it is clear that [the Pay Act amendments’] primary purpose was to relieve ‘recess appointees’ of the burden of serving without compensation during periods when the Senate is not actually sitting and is not available to give its advice and consent in respect to the appointment, irrespective of whether the recess of the Senate is attributable to a final adjournment sine die or to an adjournment to a specified date.” 28 Comp. Gen. 30, 37. We recognize that the Senate cannot easily register opposition as a body to every governmental action that many, perhaps most, Senators oppose. But the Senate has not been silent or passive regarding the meaning of the Clause: A Senate Committee did register opposition to President Theodore Roosevelt’s use of the Clause, and the Senate as a whole has legislated in an effort to discourage certain kinds of recess appointments. And yet we are not aware of any formal action it has taken to call into question the broad and functional definition of “recess” firstset out in the 1905 Senate Report and followed by the Executive Branch since at least 1921. Nor has Justice Scalia identified any. All the while, the President has made countless recess appointments during intra-session recesses. The upshot is that restricting the Clause to inter-session recesses would frustrate its purpose. It would make the President’s recess-appointment power dependent on a formalistic distinction of Senate procedure. Moreover, the President has consistently and frequently interpreted the word “recess” to apply to intra-session recesses, and has acted on that interpretation. The Senate as a body has done nothing to deny the validity of this practice for at least three-quarters of a century. And three-quarters of a century of settled practice is long enough to entitle a practice to “great weight in a proper interpretation” of the constitutional provision. The Pocket Veto Case, 279 U. S., at 689. We are aware of, but we are not persuaded by, three important arguments to the contrary. First, some argue that the Founders would likely have intended the Clause to apply only to inter-session recesses, for they hardly knew any other. See, e.g., Brief for Originalist Scholars as Amici Curiae 27–29. Indeed, from the founding until the Civil War inter-session recesses were the only kind of significant recesses that Congress took. The problem with this argument, however, is that it does not fully describe the relevant founding intent. The question is not: Did the Founders at the time think about intra-session recesses? Perhaps they did not. The question is: Did the Founders intend to restrict the scope of the Clause to the form of congressional recess then prevalent, or did they intend a broader scope permitting the Clause to apply, where appropriate, to somewhat changed circumstances? The Founders knew they were writing a document designed to apply to ever-changing circumstances over centuries. After all, a Constitution is “intended to endure for ages to come,” and must adapt itself to a future that can only be “seen dimly,” if at all. McCulloch, 4 Wheat., at 415. We therefore think the Framers likely did intend the Clause to apply to a new circumstance that so clearly falls within its essential purposes, where doing so is consistent with the Clause’s language. Second, some argue that the intra-session interpretation permits the President to make “illogic[ally]” long recess appointments. Brief for Respondent Noel Canning 13; post, at 10 (Scalia, J., concurring in judgment). A recess appointment made between Congress’ annual sessions would permit the appointee to serve for about a year, i.e., until the “end” of the “next” Senate “session.” Art. II, §2, cl. 3. But an intra-session appointment made at the beginning or in the middle of a formal session could permit the appointee to serve for 1½ or almost 2 years (until the end of the following formal session). We agree that the intra-session interpretation permits somewhat longer recess appointments, but we do not agree that this consequence is “illogical.” A President who makes a recess appointment will often also seek to make a regular appointment, nominating the appointee and securing ordinary Senate confirmation. And the Clause ensures that the President and Senate always have at least a full session to go through the nomination and confirmation process. That process may take several months. See O’Connell, Vacant Offices: Delays in Staffing Top Agency Positions, 82 S. Cal. L. Rev. 913, 967 (2009) (from 1987 to 2005 the nomination and confirmation process took an average of 236 days for noncabinet agency heads). A recess appointment that lasts somewhat longer than a year will ensure the President the continued assistance of subordinates that the Clause permits him to obtain while he and the Senate select a regular appointee. An appointment should last until the Senate has “an opportunity to act on the subject,” Story, §1551, at 410, and the Clause embodies a determination that a full session is needed to select and vet a replacement. Third, the Court of Appeals believed that application of the Clause to intra-session recesses would introduce “vagueness” into a Clause that was otherwise clear. 705 F. 3d, at 504. One can find problems of uncertainty, however, either way. In 1867, for example, President Andrew Johnson called a special session of Congress, which took place during a lengthy intra-session recess. Consider the period of time that fell just after the conclusion of that special session. Did that period remain an intra-session recess, or did it become an inter-session recess? Historians disagree about the answer. Compare Hartnett, 26 Cardozo L. Rev., at 408–409, with Brief for Constitutional Law Scholars as Amici Curiae 23–24. Or suppose that Congress adjourns sine die, but it does so conditionally, so that the leadership can call the members back into session when “the public interest shall warrant it.” E.g., 155 Cong. Rec. 33429 (2009); 152 Cong. Rec. 23731–23732 (2006); 150 Cong. Rec. 25925–25926 (2004). If the Senate Majority Leader were to reconvene the Senate, how would we characterize the preceding recess? Is it still inter-session? On the narrower interpretation the label matters; on the broader it does not. The greater interpretive problem is determining how long a recess must be in order to fall within the Clause. Is a break of a week, or a day, or an hour too short to count as a “recess”? The Clause itself does not say. And Justice Scalia claims that this silence itself shows that the Framers intended the Clause to apply only to an inter-session recess. Post, at 12–13. We disagree. For one thing, the most likely reason the Framers did not place a textual floor underneath the word “recess” is that they did not foresee the need for one. They might have expected that the Senate would meet for a single session lasting at most half a year. The Federalist No. 84, at 596 (A. Hamilton). And they might not have anticipated that intra-session recesses would become lengthier and more significant than inter-session ones. The Framers’ lack of clairvoyance on that point is not dispositive. Unlike Justice Scalia, we think it most consistent with our constitutional structure to presume that the Framers would have allowed intra-session recess appointments where there was a long history of such practice. Moreover, the lack of a textual floor raises a problem that plagues both interpretations—Justice Scalia’s and ours. Today a brief inter-session recess is just as possible as a brief intra-session recess. And though Justice Scalia says that the “notion that the Constitution empowers the President to make unilateral appointments every time the Senate takes a half-hour lunch break is so absurd as to be self-refuting,” he must immediately concede (in a footnote) that the President “can make recess appointments during any break between sessions, no matter how short.” Post, at 11, 15, n. 4 (emphasis added). Even the Solicitor General, arguing for a broader interpretation, acknowledges that there is a lower limit applicable to both kinds of recess. He argues that the lower limit should be three days by analogy to the Adjournments Clause of the Constitution. Tr. of Oral Arg. 11. That Clause says: “Neither House, during the Session of Congress, shall, without the Consent of the other, adjourn for more than three days.” Art. I, §5, cl. 4. We agree with the Solicitor General that a 3-day recess would be too short. (Under Senate practice, “Sunday is generally not considered a day,” and so is not counted for purposes of the Adjournments Clause. S. Doc. No. 101–28, F. Riddick & A. Frumin, Riddick’s Senate Procedure: Precedents and Practices 1265 (hereinafter Riddick’s).) The Adjournments Clause reflects the fact that a 3-day break is not a significant interruption of legislative business. As the Solicitor General says, it is constitutionally de minimis. Brief for Petitioner 18. A Senate recess that is so short that it does not require the consent of the House is not long enough to trigger the President’s recess-appointment power. That is not to say that the President may make recess appointments during any recess that is “more than three days.” Art. I, §5, cl. 4. The Recess Appointments Clause seeks to permit the Executive Branch to function smoothly when Congress is unavailable. And though Congress has taken short breaks for almost 200 years, and there have been many thousands of recess appointments in that time, we have not found a single example of a recess ap-pointment made during an intra-session recess that was shorter than 10 days. Nor has the Solicitor General. Reply Brief 23. Indeed, the Office of Legal Counsel once informally advised against making a recess appointment during a 6-day intra-session recess. 3 Op. OLC, at 315–316. The lack of examples suggests that the recess-appointment power is not needed in that context. (The length of a recess is “ordinarily calculated by counting the calendar days running from the day after the recess begins and including the day the recess ends.” 36 Op. OLC, at ___, n. 1 (citation omitted).) There are a few historical examples of recess appointments made during inter-session recesses shorter than 10 days. We have already discussed President Theodore Roosevelt’s appointments during the instantaneous, “fictitious” recess. President Truman also made a recess appointment to the Civil Aeronautics Board during a 3-day inter-session recess. Hogue, Recess Appointments: Fre-quently Asked Questions, at 5–6. President Taft made a few appointments during a 9-day recess following his inauguration, and President Lyndon Johnson made sev-eral appointments during an 8-day recess several weeks after assuming office. Hogue, The Law: Recess Appointments to Article III Courts, 34 Presidential Studies Q. 656, 671 (2004); 106 S. Exec. J. 2 (1964); 40 S. Exec. J. 12 (1909). There may be others of which we are unaware. But when considered against 200 years of settled practice, we regard these few scattered examples as anomalies. We therefore conclude, in light of historical practice, that a recess of more than 3 days but less than 10 days is presumptively too short to fall within the Clause. We add the word “presumptively” to leave open the possibility that some very unusual circumstance—a national catastrophe, for instance, that renders the Senate unavailable but calls for an urgent response—could demand the exercise of the recess-appointment power during a shorter break. (It should go without saying—except that Justice Scalia compels us to say it—that political opposition in the Senate would not qualify as an unusual circumstance.) In sum, we conclude that the phrase “the recess” applies to both intra-session and inter-session recesses. If a Senate recess is so short that it does not require the consent of the House, it is too short to trigger the Recess Appointments Clause. See Art. I, §5, cl. 4. And a recess lasting less than 10 days is presumptively too short as well. IV The second question concerns the scope of the phrase “vacancies that may happen during the recess of the Senate.” Art. II, §2, cl. 3 (emphasis added). All agree that the phrase applies to vacancies that initially occur during a recess. But does it also apply to vacancies that initially occur before a recess and continue to exist during the recess? In our view the phrase applies to both kinds of vacancy. We believe that the Clause’s language, read literally, permits, though it does not naturally favor, our broader interpretation. We concede that the most natural meaning of “happens” as applied to a “vacancy” (at least to a modern ear) is that the vacancy “happens” when it ini-tially occurs. See 1 Johnson 913 (defining “happen” in relevant part as meaning “[t]o fall out; to chance; to come to pass”). But that is not the only possible way to use the word. Thomas Jefferson wrote that the Clause is “certainly susceptible of [two] constructions.” Letter to Wilson Cary Nicholas (Jan. 26, 1802), in 36 Papers of Thomas Jefferson 433 (B. Oberg ed., 2009). It “may mean ‘vacancies that may happen to be’ or ‘may happen to fall’ ” during a recess. Ibid. Jefferson used the phrase in the first sense when he wrote to a job seeker that a particular position was unavailable, but that he (Jefferson) was “happy that another vacancy happens wherein I can . . . avail the public of your integrity & talents,” for “the office of Treasurer of the US. is vacant by the resignation of mr Meredith.” Letter to Thomas Tudor Tucker (Oct. 31, 1801), in 35 id., at 530 (B. Oberg ed. 2008) (emphasis added). See also Laws Passed by the Legislature of Florida, No. 31, An Act to Organize and Regulate the Militia of the Territory of Florida §13, H. R. Exec. Doc. No. 72, 27th Cong., 3d Sess., 22 (1842) (“[W]hen any vacancy shall take place in the office of any lieutenant colonel, it shall be the duty of the colonel of the regiment in which such vacancy may happen to order an election to be held at the several precincts in the battalion in which such vacancy may happen” (emphasis added)). Similarly, when Attorney General William Wirt advised President Monroe to follow the broader interpretation, he wrote that the “expression seems not perfectly clear. It may mean ‘happen to take place:’ that is, ‘to originate,’ ” or it “may mean, also, without violence to the sense, ‘happen to exist.’ ” 1 Op. Atty. Gen. 631, 631–632 (1823). The broader interpretation, he added, is “most accordant with” the Constitution’s “reason and spirit.” Id., at 632. We can still understand this earlier use of “happen” if we think of it used together with another word that, like “vacancy,” can refer to a continuing state, say, a financial crisis. A statute that gives the President authority to act in respect to “any financial crisis that may happen during his term” can easily be interpreted to include crises that arise before, and continue during, that term. Perhaps that is why the Oxford English Dictionary defines “happen” in part as “chance to be,” rather than “chance to occur.” 6 OED 1096 (emphasis added); see also 19 OED 383 (defining “vacancy” as the “condition of an office or post being . . . vacant”). In any event, the linguistic question here is not whether the phrase can be, but whether it must be, read more narrowly. The question is whether the Clause is ambiguous. The Pocket Veto Case, 279 U. S., at 690. And the broader reading, we believe, is at least a permissible reading of a “ ‘doubtful’ ” phrase. Ibid. We consequently go on to consider the Clause’s purpose and historical practice. The Clause’s purpose strongly supports the broader interpretation. That purpose is to permit the President to obtain the assistance of subordinate officers when the Senate, due to its recess, cannot confirm them. Attorney General Wirt clearly described how the narrower interpretation would undermine this purpose: “Put the case of a vacancy occurring in an office, held in a distant part of the country, on the last day of the Senate’s session. Before the vacancy is made known to the President, the Senate rises. The office may be an important one; the vacancy may paralyze a whole line of action in some essential branch of our internal police; the public interests may imperiously demand that it shall be immediately filled. But the vacancy happened to occur during the session of the Senate; and if the President’s power is to be limited to such vacancies only as happen to occur during the recess of the Senate, the vacancy in the case put must continue, however ruinous the consequences may be to the public.” 1 Op. Atty. Gen., at 632. Examples are not difficult to imagine: An ambassadorial post falls vacant too soon before the recess begins for the President to appoint a replacement; the Senate rejects a President’s nominee just before a recess, too late to select another. Wirt explained that the “substantial purpose of the constitution was to keep these offices filled,” and “if the President shall not have the power to fill a vacancy thus circumstanced, . . . the substance of the constitution will be sacrificed to a dubious construction of its letter.” Ibid. Thus the broader construction, encompassing vacancies that initially occur before the beginning of a recess, is the “only construction of the constitution which is compatible with its spirit, reason, and purposes; while, at the same time, it offers no violence to its language.” Id., at 633. We do not agree with Justice Scalia’s suggestion that the Framers would have accepted the catastrophe envisioned by Wirt because Congress can always provide for acting officers, see 5 U. S. C. §3345, and the President can always convene a special session of Congress, see U. S. Const., Art. II, §3. Acting officers may have less authority than Presidential appointments. 6 Op. OLC 119, 121 (1982). Moreover, to rely on acting officers would lessen the President’s ability to staff the Executive Branch with people of his own choosing, and thereby limit the President’s control and political accountability. Cf. Free Enterprise Fund v. Public Company Accounting Oversight Bd., 561 U. S. 477 –498 (2010). Special sessions areburdensome (and would have been especially so at the time of the founding). The point of the Recess Appointments Clause was to avoid reliance on these inadequate expedients. At the same time, we recognize one important purpose-related consideration that argues in the opposite direction. A broad interpretation might permit a President to avoid Senate confirmations as a matter of course. If the Clause gives the President the power to “fill up all vacancies” that occur before, and continue to exist during, the Senate’s recess, a President might not submit any nominations to the Senate. He might simply wait for a recess and then provide all potential nominees with recess appointments. He might thereby routinely avoid the constitutional need to obtain the Senate’s “advice and consent.” Wirt thought considerations of character and politics would prevent Presidents from abusing the Clause in this way. 1 Op. Atty. Gen., at 634. He might have added that such temptations should not often arise. It is often less desirable for a President to make a recess appointment. A recess appointee only serves a limited term. That, combined with the lack of Senate approval, may diminish the recess appointee’s ability, as a practical matter, to get a controversial job done. And even where the President and Senate are at odds over politically sensitive appointments, compromise is normally possible. Indeed, the 1940 Pay Act amendments represent a general compromise, for they foresee payment of salaries to recess appointees where vacancies occur before the recess began but not too long before (namely, within 30 days before). 5 U. S. C. §5503(a)(1); see infra, at 32. Moreover, the Senate, like the President, has institutional “resources,” including political resources, “available to protect and assert its interests.” Goldwater v. Carter, 444 U. S. 996, 1004 (1979) (Rehnquist, J., concurring in judgment). In an unusual instance, where a matter is important enough to the Sen-ate, that body can remain in session, preventing recess appointments by refusing to take a recess. See Part V, infra. In any event, the Executive Branch has adhered to the broader interpretation for two centuries, and Senate confirmation has always remained the norm for officers that require it. While we concede that both interpretations carry with them some risk of undesirable consequences, we believe the narrower interpretation risks undermining constitutionally conferred powers more seriously and more often. It would prevent the President from making any recess appointment that arose before a recess, no matter who the official, no matter how dire the need, no matter how uncontroversial the appointment, and no matter how late in the session the office fell vacant. Overall, like Attorney General Wirt, we believe the broader interpretation more consistent with the Constitution’s “reason and spirit.” 1 Op. Atty. Gen., at 632. Historical practice over the past 200 years strongly favors the broader interpretation. The tradition of applying the Clause to pre-recess vacancies dates at least to President James Madison. There is no undisputed record of Presidents George Washington, John Adams, or Thomas Jefferson making such an appointment, though the Solicitor General believes he has found records showing that Presidents Washington and Jefferson did so. We know that Edmund Randolph, Washington’s Attorney General, favored a narrow reading of the Clause. Randolph believed that the “Spirit of the Constitution favors the participation of the Senate in all appointments,” though he did not address—let alone answer—the powerful purposive and structural arguments subsequently made by Attorney General Wirt. See Edmund Randolph’s Opinion on Recess Appointments (July 7, 1792), in 24 Papers of Thomas Jefferson 166 (J. Catanzariti ed. 1990). President Adams seemed to endorse the broader view of the Clause in writing, though we are not aware of any appointments he made in keeping with that view. See Letter to J. McHenry (Apr. 16, 1799), in 8 Works of John Adams 632–633 (C. Adams ed. 1853). His Attorney General, Charles Lee, later informed Jefferson that, in the Adams administration, “whenever an office became vacant so short a time before Congress rose, as not to give an opportunity of enquiring for a proper character, they let it lie always till recess.” 36 Papers of Thomas Jefferson 433. We know that President Jefferson thought that the broad interpretation was linguistically supportable, though his actual practice is not clear. But the evidence suggests that James Madison—as familiar as anyone with the workings of the Constitutional Convention—appointed Theodore Gaillard to replace a district judge who had left office before a recess began. Hartnett, 26 Cardozo L. Rev., at 400–401. It also appears that in 1815 Madison signed a bill that created two new offices prior to a recess which he then filled later during the recess. See Act of Mar. 3, ch. 95, 3Stat. 235; S. J. 13th Cong., 3d Sess., 689–690 (1815); 3 S. Exec. J. 19 (1828) (for Monday, Jan. 8, 1816). He also made recess appointments to “territorial” United States attorney and marshal positions, both of which had been created when the Senate was in session more than two years before. Act of Feb. 27, 1813, ch. 35, 2Stat. 806; 3 S. Exec. J. 19. Justice Scalia refers to “written evidence of Madison’s own beliefs,” post, at 36, but in fact we have no direct evidence of what President Madison believed. We only know that he declined to make one appointment to a pre-recess vacancy after his Secretary of War advised him that he lacked the power. On the other hand, he did apparently make at least five other appointments to pre-recess vacancies, as Justice Scalia does not dispute. The next President, James Monroe, received and presumably acted upon Attorney General Wirt’s advice, namely that “all vacancies which, from any casualty, happen to exist at a time when the Senate cannot be consulted as to filling them, may be temporarily filled by the President.” 1 Op. Atty. Gen., at 633. Nearly every subsequent Attorney General to consider the question throughout the Nation’s history has thought the same. E.g., 2 Op. Atty. Gen. 525, 528 (1832); 7 Op. Atty. Gen. 186, 223 (1855); 10 Op. Atty. Gen. 356, 356–357 (1862); 12 Op. Atty. Gen. 32, 33 (1866); 12 Op. Atty. Gen., at 452; 14 Op. Atty. Gen. 562, 564 (1875); 15 Op. Atty. Gen. 207 (1877); 16 Op. Atty. Gen. 522, 524 (1880); 17 Op. Atty. Gen. 521 (1883); 18 Op. Atty. Gen. 29, 29–30 (1884); 19 Op. Atty. Gen. 261, 262 (1889); 26 Op. Atty. Gen. 234, 234–235 (1907); 30 Op. Atty. Gen. 314, 315 (1914); 41 Op. Atty. Gen. 463, 465 (1960); 3 Op. OLC 314 (1979); 6 Op. OLC 585, 586 (1982); 20 Op. OLC 124, 161 (1996); 36 Op. OLC ___ (2012). Indeed, as early as 1862, Attorney General Bates advised President Lincoln that his power to fill pre-recess vacancies was “settled . . . as far . . . as a constitutional question can be settled,” 10 Op. Atty. Gen., at 356, and a century later Acting Attorney General Walsh gave President Eisenhower the same advice “without any doubt,” 41 Op. Atty. Gen., at 466. This power is important. The Congressional Research Service is “unaware of any official source of information tracking the dates of vacancies in federal offices.” The Noel Canning Decision 3, n. 6. Nonetheless, we have enough information to believe that the Presidents since Madison have made many recess appointments filling vacancies that initially occurred prior to a recess. As we have just said, nearly every 19th- and 20th-century Attorney General expressing a view on the matter has agreed with William Wirt, and Presidents tend to follow the legal advice of their chief legal officers. Moreover, the Solicitor General has compiled a list of 102 (mostly uncontested) recess appointments made by Presidents going back to the founding. App. to Brief for Petitioner 65a–89a. Given the difficulty of finding accurate information about vacancy dates, that list is undoubtedly far smaller than the actual number. No one disputes that every President since James Buchanan has made recess appointments to pre-existing vacancies. Common sense also suggests that many recess appointees filled vacancies that arose before the recess began. We have compared the list of intra-session recess appointments in the Solicitor General’s brief with the chart of congressional recesses. Where a specific date of appointment can be ascertained, more than half of those intra-session appointments were made within two weeks of the beginning of a recess. That short window strongly suggests that many of the vacancies initially arose prior to the recess. See App. to Brief for Petitioner 1a–64a; Appendix A, infra. Thus, it is not surprising that the Congressional Research Service, after examining the vacancy dates associated with a random sample of 24 inter-session recess appointments since 1981, concluded that “[i]n most of the 24 cases, the preponderance of evidence indicated that the vacancy arose prior to the recess during which the appointment was made.” The Noel Canning Decision 3. Further, with research assistance from the Supreme Court Library, we have examined a random sample of the recess appointments made by our two most recent Presidents, and have found that almost all of those appointments filled pre-recess vacancies: Of a sample of 21 recess appointments, 18 filled pre-recess vacancies and only 1 filled a vacancy that arose during the recess in which he was appointed. The precise date on which 2 of the vacancies arose could not be determined. See Appendix B, infra. Taken together, we think it is a fair inference that a large proportion of the recess appointments in the history of the Nation have filled pre-existing vacancies. Did the Senate object? Early on, there was some sporadic disagreement with the broad interpretation. In 1814 Senator Gore said that if “the vacancy happen at another time, it is not the case described by the Constitution.” 26 Annals of Cong. 653. In 1822 a Senate committee, while focusing on the President’s power to fill a new vacancy created by statute, used language to the same effect. 38 id., at 489, 500. And early Congresses enacted statutes authorizing certain recess appointments, see post, at 31, a fact that may or may not suggest they accepted the narrower interpretation of the Clause. Most of those statutes—including the one passed by the First Congress—authorized appointments to newly created offices, and may have been addressed to the separate question of whether new offices are vacancies within the meaning of the Clause. See Letter from Alexander Hamilton to James McHenry (May 3, 1799), in 23 Papers of Alexander Hamilton 94 (H. Syrett ed. 1976) (“Vacancy is a relative term, and presupposes that the Office has been once filled”); Reply Brief 17. In any event, by 1862 Attorney General Bates could still refer to “the unbroken acquiescence of the Senate” in support of the broad interpretation. 10 Op. Atty. Gen., at 356. Then in 1863 the Senate Judiciary Committee disagreed with the broad interpretation. It issued a report concluding that a vacancy “must have its inceptive point after one session has closed and before another session has begun.” S. Rep. No. 80, 37th Cong., 3d Sess., p. 3. And the Senate then passed the Pay Act, which provided that “no money shall be paid . . . as a salary, to any person appointed during the recess of the Senate, to fill a vacancy . . . which . . . existed while the Senate was in session.” Act of Feb. 9, 1863, §2, 12Stat. 646. Relying upon the floor statement of a single Senator, Justice Scalia suggests that the passage of the Pay Act indicates that the Senate as a whole endorsed the position in the 1863 Report. But the circumstances are more equivocal. During the floor debate on the bill, not a single Senator referred to the Report. Cong. Globe, 37th Cong., 3d Sess. 564–565 (1863). Indeed, Senator Trumbull, who introduced the Pay Act, acknowledged that there was disagreement about the underlying constitutional question. Id., at 565 (“[S]ome other persons think he has that power”). Further, if a majority of the Senate had believed appointments to pre-recess vacancies were unconstitutional, it could have attempted to do far more than temporarily dock the appointees’ pay. Cf. Tenure of Office Act of 1867, §5, 14Stat. 431 (making it a federal crime for “any person” to “accept any appointment” in certain circumstances). In any event, the Senate subsequently abandoned its hostility. In the debate preceding the 1905 Senate Report regarding President Roosevelt’s “constructive” recess appointments, Senator Tillman—who chaired the Committee that authored the 1905 Report—brought up the 1863 Report, and another Senator responded: “Whatever that report may have said in 1863, I do not think that has been the view the Senate has taken” of the issue. 38 Cong. Rec. 1606 (1904). Senator Tillman then agreed that “the Senate has acquiesced” in the President’s “power to fill” pre-recess vacancies. Ibid. And Senator Tillman’s 1905 Report described the Clause’s purpose in terms closely echoing Attorney General Wirt. 1905 Senate Report, at 2 (“Its sole purpose was to render it certain that at all times there should be, whether the Senate was in session or not, an officer for every office” (emphasis added)). In 1916 the Senate debated whether to pay a recess appointee who had filled a pre-recess vacancy and had not subsequently been confirmed. Both Senators to address the question—one on each side of the payment debate—agreed that the President had the constitutional power to make the appointment, and the Senate voted to pay the appointee for his service. 53 Cong. Rec. 4291–4299; 39Stat. 818–819. In 1927 the Comptroller General, a legislative officer, wrote that “there is no question but that the President has authority to make a recess appointment to fill any vacancy,” including those that “existed while the Senate was in session.” 7 Comp. Gen. 10, 11 (emphasis added). Meanwhile, Presidents continued to make appointments to pre-recess vacancies. The Solicitor General has identified 40 between 1863 and 1940, but that number is clearly not comprehensive. See, e.g., 32 Op. Atty. Gen. 271–272 (1920) (listing 5 appointments that are not in the Solicitor General’s appendix); Recess Appointments, Washington Post, July 7, 1880, p. 1 (noting that President Hayes had made “quite a number of appointments” to pre-recess vacancies). Then in 1940 Congress amended the Pay Act to authorize salary payments (with some exceptions) where (1) the “vacancy arose within thirty days prior to the termination of the session,” (2) “at the termination of the session” a nomination was “pending,” or (3) a nominee was “rejected by the Senate within thirty days prior to the termination of the session.” Act of July 11, 54Stat. 751 (codified, as amended, at 5 U. S. C. §5503). All three circumstances concern a vacancy that did not initially occur during a recess but happened to exist during that recess. By paying salaries to this kind of recess appointee, the 1940 Senate (and later Senates) in effect supported the President’s interpretation of the Clause. The upshot is that the President has consistently and frequently interpreted the Recess Appointments Clause to apply to vacancies that initially occur before, but continue to exist during, a recess of the Senate. The Senate as a body has not countered this practice for nearly three-quarters of a century, perhaps longer. See A. Amar, The Unwritten Constitution 576–577, n. 16 (2012) (for nearly 200 years “the overwhelming mass of actual practice” supports the President’s interpretation); Mistretta v. United States, 488 U. S. 361, 401 (1989) (a “200–year tradition” can “ ‘give meaning’ to the Constitution” (quot-ing Youngstown, 343 U. S., at 610 (Frankfurter, J., concurring))). The tradition is long enough to entitle the practice “to great regard in determining the true construction” of the constitutional provision. The Pocket Veto Case, 279 U. S., at 690. And we are reluctant to upset this traditional practice where doing so would seriously shrink the authority that Presidents have believed existed and have exercised for so long. In light of some linguistic ambiguity, the basic purpose of the Clause, and the historical practice we have described, we conclude that the phrase “all vacancies” includes vacancies that come into existence while the Senate is in session. V The third question concerns the calculation of the length of the Senate’s “recess.” On December 17, 2011, the Senate by unanimous consent adopted a resolution to convene “pro forma session[s]” only, with “no business . . . transacted,” on every Tuesday and Friday from December 20, 2011, through January 20, 2012. 2011 S. J. 923. At the end of each pro forma session, the Senate would “adjourn until” the following pro forma session. Ibid. During that period, the Senate convened and adjourned as agreed. It held pro forma sessions on December 20, 23, 27, and 30, and on January 3, 6, 10, 13, 17, and 20; and at the end of each pro forma session, it adjourned until the time and date of the next. Id., at 923–924; 158 Cong. Rec. S1–S11. The President made the recess appointments before us on January 4, 2012, in between the January 3 and the January 6 pro forma sessions. We must determine the significance of these sessions—that is, whether, for purposes of the Clause, we should treat them as periods when the Senate was in session or as periods when it was in recess. If the former, the period between January 3 and January 6 was a 3-day recess, which is too short to trigger the President’s recess-appointment power, see supra, at 19–21. If the latter, however, then the 3-day period was part of a much longer recess during which the President did have the power to make recess appointments, see ibid. The Solicitor General argues that we must treat the pro forma sessions as periods of recess. He says that these “sessions” were sessions in name only because the Senate was in recess as a functional matter. The Senate, he contends, remained in a single, unbroken recess from January 3, when the second session of the 112th Congress began by operation of the Twentieth Amendment, until January 23, when the Senate reconvened to do regular business. In our view, however, the pro forma sessions count as sessions, not as periods of recess. We hold that, for purposes of the Recess Appointments Clause, the Senate is in session when it says it is, provided that, under its own rules, it retains the capacity to transact Senate business. The Senate met that standard here. The standard we apply is consistent with the Constitution’s broad delegation of authority to the Senate to determine how and when to conduct its business. The Constitution explicitly empowers the Senate to “determine the Rules of its Proceedings.” Art. I, §5, cl. 2. And we have held that “all matters of method are open to the determination” of the Senate, as long as there is “a reasonable relation between the mode or method of proceeding established by the rule and the result which is sought to be attained” and the rule does not “ignore constitutional restraints or violate fundamental rights.” United States v. Ballin, 144 U. S. 1, 5 (1892) . In addition, the Constitution provides the Senate with extensive control over its schedule. There are only limited exceptions. See Amdt. 20, §2 (Congress must meet once a year on January 3, unless it specifies another day by law); Art. II, §3 (Senate must meet if the President calls it into special session); Art. I, §5, cl. 4 (neither House may adjourn for more than three days without consent of the other). See also Art. II, §3 (“[I]n Case of Disagreement between [the Houses], with Respect to the Time of Adjournment, [the President] may adjourn them to such Time as he shall think proper”). The Constitution thus gives the Senate wide latitude to determine whether and when to have a session, as well as how to conduct the session. This suggests that the Senate’s determination about what constitutes a session should merit greatrespect. Furthermore, this Court’s precedents reflect the breadth of the power constitutionally delegated to the Senate. We generally take at face value the Senate’s own report of its actions. When, for example, “the presiding officers” of the House and Senate sign an enrolled bill (and the President “approve[s]” it), “its authentication as a bill that has passed Congress should be deemed complete and unimpeachable.” Marshall Field & Co. v. Clark, 143 U. S. 649, 672 (1892) . By the same principle, when the Journal of the Senate indicates that a quorum was present, under a valid Senate rule, at the time the Senate passed a bill, we will not consider an argument that a quorum was not, in fact, present. Ballin, supra, at 9. The Constitution requires the Senate to keep its Journal, Art. I, §5, cl. 3 (“Each House shall keep a Journal of its proceedings . . .”), and “if reference may be had to” it, “it must be assumed to speak the truth,” Ballin, supra, at 4. For these reasons, we conclude that we must give great weight to the Senate’s own determination of when it is and when it is not in session. But our deference to the Senate cannot be absolute. When the Senate is without the capacity to act, under its own rules, it is not in session even if it so declares. See Tr. of Oral Arg. 69 (acknowledgment by counsel for amici Senators that if the Senate had left the Capitol and “effectively given up . . . the business of legislating” then it might be in recess, even if it said it was not). In that circumstance, the Senate is not simply unlikely or unwilling to act upon nominations of the President. It is unable to do so. The purpose of the Clause is to ensure the continued functioning of the Federal Government while the Senate is unavailable. See supra, at 5–6. This purpose would count for little were we to treat the Senate as though it were in session even when it lacks the ability to provide its “advice and consent.” Art. II, §2, cl. 2. Accordingly, we conclude that when the Senate declares that it is in session and possesses the capacity, under its own rules, to conduct business, it is in session for purposes of the Clause. Applying this standard, we find that the pro forma sessions were sessions for purposes of the Clause. First, the Senate said it was in session. The Journal of the Senate and the Congressional Record indicate that the Senate convened for a series of twice-weekly “sessions” from December 20 through January 20. 2011 S. J. 923–924; 158 Cong. Rec. S1–S11. (The Journal of the Senate for 2012 has not yet been published.) And these reports of the Senate “must be assumed to speak the truth.” Ballin, supra, at 4. Second, the Senate’s rules make clear that during its pro forma sessions, despite its resolution that it would conduct no business, the Senate retained the power to conduct business. During any pro forma session, the Senate could have conducted business simply by passing a unanimous consent agreement. See Riddick’s 1313. The Senate in fact conducts much of its business through unanimous consent. Id., at 1311–1312. Senate rules presume that a quorum is present unless a present Senator questions it. Id., at 1041–1042. And when the Senate has a quorum, an agreement is unanimously passed if, upon its proposal, no present Senator objects. Id., at 1329–1330. It is consequently unsurprising that the Senate has enacted legislation during pro forma sessions even when it has said that no business will be transacted. Indeed, the Senate passed a bill by unanimous consent during the second pro forma session after its December 17 adjournment. 2011 S. J. 924. And that bill quickly became law. Pub. L. 112–78, 125Stat. 1280. By way of contrast, we do not see how the Senate could conduct business during a recess. It could terminate the recess and then, when in session, pass a bill. But in that case, of course, the Senate would no longer be in recess. It would be in session. And that is the crucial point. Senate rules make clear that, once in session, the Senate can act even if it has earlier said that it would not. The Solicitor General argues that more is required. He contends that what counts is not the Senate’s capacity to conduct business but what the Senate actually does (or here, did) during its pro forma sessions. And he looks for support to the functional definition of “recess” set forth in the 1905 Senate Report discussed above. See supra, at 14. That Report describes a “recess” of the Senate as “the period of time . . . when its members owe no duty of attendance; when its Chamber is empty; when, because of its absence, it can not receive communications from the President or participate as a body in making appointments.” 1905 Senate Report, at 2. Even were we, for argument’s sake, to accept all of these criteria as authoritative, they would here be met. Taking the last criterion first, could the Senate, during its pro forma sessions, “participate as a body in making appointments”? It could. It could confirm nominees by unanimous consent, just as it passed the bill mentioned above. See Riddick’s 1313. Could the Senate “receive communications from the President”? It could. The Congressional Record indicates that the Senate “received” a message from the President on January 12, during a 3-day adjournment between two pro forma sessions. See 158 Cong. Rec. S37 (Jan. 23, 2012). If the Senate could receive Presidential messages between two pro forma sessions, it could receive them during a pro forma session. Was the Senate’s Chamber “empty”? It was not. By its official rules, the Senate operates under the presumption that a quorum is present until a present Senator suggests the absence of a quorum, Riddick’s 1041–1042, and nothing in the Journal of the Senate or the Congressional Record reflects any such suggestion. Did Senators “owe [a] duty of attendance”? They did. The Senate’s rules dictate that Senators are under a duty to attend every session. See Riddick’s 214; Standing Rule of the Senate VI(2), S. Doc. No. 112–1, p. 5 (2011) (“No Senator shall absent himself from the service of the Senate without leave”). Nothing excused the Senators from this duty during the Senate’s pro forma sessions. If any present Senator had raised a question as to the presence of a quorum, and by roll call it had become clear that a quorum was missing, the Senators in attendance could have directed the Sergeant at Arms to bring in the missing Senators. Rule VI(4). The Solicitor General asks us to engage in a more realistic appraisal of what the Senate actually did. He argues that, during the relevant pro forma sessions, business was not in fact conducted; messages from the President could not be received in any meaningful way because they could not be placed before the Senate; the Senate Chamber was, according to C-SPAN coverage, almost empty; and in practice attendance was not required. See Brief for Petitioner 48–49, 54–55. We do not believe, however, that engaging in the kind of factual appraisal that the Solicitor General suggests is either legally or practically appropriate. From a legal perspective, this approach would run contrary to prece-dent instructing us to “respect . . . coequal and independent departments” by, for example, taking the Senate’s report of its official action at its word. Field, 143 U. S., at 672; see Ballin, 144 U. S., at 4. From a practical perspective, judges cannot easily determine such matters as who is, and who is not, in fact present on the floor during a particular Senate session. Judicial efforts to engage in these kinds of inquiries would risk undue judicial interference with the functioning of the Legislative Branch. Finally, the Solicitor General warns that our holding may “ ‘disrup[t] the proper balance between the coordinate branches by preventing the Executive Branch from accomplishing its constitutionally assigned functions.’ ” Brief for Petitioner 64 (quoting Morrison v. Olson, 487 U. S. 654, 695 (1988) ; alteration in original). We do not see, however, how our holding could significantly alter the constitutional balance. Most appointments are not controversial and do not produce friction between the branches. Where political controversy is serious, the Senate unquestionably has other methods of preventing recess appointments. As the Solicitor General concedes, the Senate could preclude the President from making recess appointments by holding a series of twice-a-week ordinary (not pro forma) sessions. And the nature of the business conducted at those ordinary sessions—whether, for example, Senators must vote on nominations, or may return totheir home States to meet with their constituents—is a matter for the Senate to decide. The Constitution also gives the President (if he has enough allies in Congress) a way to force a recess. Art. II, §3 (“[I]n Case of Disagreement between [the Houses], with Respect to the Time of Adjournment, [the President] may adjourn them to such Time as he shall think proper”). Moreover, the President and Senators engage with each other in many different ways and have a variety of methods of encouraging each other to accept their points of view. Regardless, the Recess Appointments Clause is not designed to overcome serious institutional friction. It simply provides a subsidiary method for appointing officials when the Senate is away during a recess. Here, as in other contexts, friction between the branches is an inevitable consequence of our constitutional structure. See Myers, 272 U. S., at 293 (Brandeis, J., dissenting). That structure foresees resolution not only through judicial interpretation and compromise among the branches but also by the ballot box. VI The Recess Appointments Clause responds to a structural difference between the Executive and Legislative Branches: The Executive Branch is perpetually in operation, while the Legislature only acts in intervals separated by recesses. The purpose of the Clause is to allow the Executive to continue operating while the Senate is unavailable. We believe that the Clause’s text, standing alone, is ambiguous. It does not resolve whether the President may make appointments during intra-session recesses, or whether he may fill pre-recess vacancies. But the broader reading better serves the Clause’s structural function. Moreover, that broader reading is reinforced by centuries of history, which we are hesitant to disturb. We thus hold that the Constitution empowers the President to fill any existing vacancy during any recess—intra-session or inter-session—of sufficient length. Justice Scalia would render illegitimate thousands of recess appointments reaching all the way back to the founding era. More than that: Calling the Clause an “anachronism,” he would basically read it out of the Constitution. Post, at 12. He performs this act of judicial excision in the name of liberty. We fail to see how excising the Recess Appointments Clause preserves freedom. In fact, Alexander Hamilton observed in the very first Feder-alist Paper that “the vigour of government is essential to the security of liberty.” The Federalist No. 1, at 5. And the Framers included the Recess Appointments Clause to preserve the “vigour of government” at times when an important organ of Government, the United States Senate, is in recess. Justice Scalia’s interpretation of the Clause would defeat the power of the Clause to achieve that objective. The foregoing discussion should refute Justice Scalia’s claim that we have “embrace[d]” an “adverse-possession theory of executive power.” Post, at 48. Instead, as inall cases, we interpret the Constitution in light of itstext, purposes, and “our whole experience” as a Nation. Missouri v. Holland, 252 U. S. 416, 433 (1920) . And we look to the actual practice of Government to inform our interpretation. Given our answer to the last question before us, we conclude that the Recess Appointments Clause does not give the President the constitutional authority to make the appointments here at issue. Because the Court of Appeals reached the same ultimate conclusion (though for reasons we reject), its judgment is affirmed. It is so ordered. APPENDIXES A The following table contains the dates of all the intra-session and inter-session recesses that Congress has taken since the founding. The information (including the footnotes) is taken from 2011–2012 Official Congressional Directory, 112th Cong., 522–539. SESSIONS OF CONGRESS, 1st–112th CONGRESSES, 1789–2011 1 For the purposes of this table, a session’s ‘‘length in days’’ is defined as the total number of calendar days from the convening date to the adjournment date, inclusive. It does not mean the actual number of days that Congress met during that session. 2 For the purposes of this table, a ‘‘recess’’ is defined as a break in House or Senate proceedings of three or more days, excluding Sundays. According to Article I, section 5 of the U. S. Constitution, neither house may adjourn for more than three days without the consent of the other. On occasion, both chambers have held one or more pro forma sessions because of this constitutional obligation or for other purposes. Treated here as recesses, usually no business is conducted during these time periods. On this table, beginning in the 1990s, such pro forma sessions are indicated with a P. B The following table shows the proportion of recent appointments that have filled pre-recess vacancies. It was compiled with research assistance from the Supreme Court Library. It contains a random sample of the recess appointments by President George W. Bush and President Barack Obama. The last column indicates whether the vacancy arose during the recess in which it was filled. “A” indicates a vacancy that arose during the recess, “P” indicates a vacancy that arose before the recess, and “U” indicates that the vacancy date could not be ascertained.

572.US.898

The Patent Act requires that a patent specification “conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as [the] invention.” 35 U. S. C. §112, ¶2. This case concerns the proper reading of the statute’s clarity and precision demand. Assigned to respondent Biosig Instruments, Inc., the patent in dispute (the ’753 patent) involves a heart-rate monitor used with exercise equipment. Prior heart-rate monitors, the patent asserts, were often inaccurate in measuring the electrical signals accompanying each heartbeat (electrocardiograph or ECG signals) because of the presence of other electrical signals (electromyogram or EMG signals), generated by the user’s skeletal muscles, that can impede ECG signal detection. The invention claims to improve on prior art by detecting and processing ECG signals in a way that filters out the EMG interference. Claim 1 of the ’753 patent, which contains the limitations critical to this dispute, refers to a “heart rate monitor for use by a user in association with exercise apparatus and/or exercise procedures.” The claim “comprise[s],” among other elements, a cylindrical bar fitted with a display device; “electronic circuitry including a difference amplifier”; and, on each half of the cylindrical bar, a “live” electrode and a “common” electrode “mounted . . . in spaced relationship with each other.” Biosig filed this patent infringement suit, alleging that Nautilus, Inc., without obtaining a license, sold exercise machines containing Biosig’s patented technology. The District Court, after conducting a hearing to determine the proper construction of the patent’s claims, granted Nautilus’ motion for summary judgment on the ground that the claim term “in spaced relationship with each other” failed §112, ¶2’s definiteness requirement. The Federal Circuit reversed and remanded, concluding that a patent claim passes the §112, ¶2 threshold so long as the claim is “amenable to construction,” and the claim, as construed, is not “insolubly ambiguous.” Under that standard, the court determined, the ’753 patent survived indefiniteness review. Held: 1. A patent is invalid for indefiniteness if its claims, read in light of the patent’s specification and prosecution history, fail to inform, with reasonable certainty, those skilled in the art about the scope of the invention. The parties agree that definiteness is to be evaluated from the perspective of a person skilled in the relevant art, that claims are to be read in light of the patent’s specification and prosecution history, and that definiteness is to be measured as of the time of the patent application. The parties disagree as to how much imprecision §112, ¶2 tolerates. Section 112’s definiteness requirement must take into account the inherent limitations of language. See Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722, 731. On the one hand, some modicum of uncertainty is the “price of ensuring the appropriate incentives for innovation,” id., at 732; and patents are “not addressed to lawyers, or even to the public generally,” but to those skilled in the relevant art, Carnegie Steel Co. v. Cambria Iron Co., 185 U.S. 403, 437. At the same time, a patent must be precise enough to afford clear notice of what is claimed, thereby “ ‘appris[ing] the public of what is still open to them,’ ” Markman v. Westview Instruments, Inc., 517 U.S. 370, 373, in a manner that avoids “[a] zone of uncertainty which enterprise and experimentation may enter only at the risk of infringement claims,” United Carbon Co. v. Binney & Smith Co., 317 U.S. 228, 236. The standard adopted here mandates clarity, while recognizing that absolute precision is unattainable. It also accords with opinions of this Court stating that “the certainty which the law requires in patents is not greater than is reasonable, having regard to their subject-matter.” Minerals Separation, Ltd. v. Hyde, 242 U.S. 261, 270. Pp. 8–11. 2. The Federal Circuit’s standard, which tolerates some ambiguous claims but not others, does not satisfy the statute’s definiteness requirement. The Court of Appeals inquired whether the ’753 patent’s claims were “amenable to construction” or “insolubly ambiguous,” but such formulations lack the precision §112, ¶2 demands. To tolerate imprecision just short of that rendering a claim “insolubly ambiguous” would diminish the definiteness requirement’s public-notice function and foster the innovation-discouraging “zone of uncertainty,” United Carbon, 317 U. S., at 236, against which this Court has warned. While some of the Federal Circuit’s fuller explications of the term “insolubly ambiguous” may come closer to tracking the statutory prescription, this Court must ensure that the Federal Circuit’s test is at least “probative of the essential inquiry.” Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17, 40. The expressions “insolubly ambiguous” and “amenable to construction,” which permeate the Federal Circuit’s recent decisions concerning §112, ¶2, fall short in this regard and can leave courts and the patent bar at sea without a reliable compass. Pp. 11–13. 3. This Court, as “a court of review, not of first view,” Cutter v. Wilkinson, 544 U.S. 709, 718, n. 7, follows its ordinary practice of remanding so that the Federal Circuit can reconsider, under the proper standard, whether the relevant claims in the ’753 patent are sufficiently definite, see, e.g., Johnson v. California, 543 U.S. 499, 515. Pp. 13–14. 715 F.3d 891, vacated and remanded. Ginsburg, J., delivered the opinion for a unanimous Court.

The Patent Act requires that a patent specification “conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as [the] invention.” 35 U. S. C. §112, ¶2 (2006 ed.) (emphasis added). This case, involving a heart-rate monitor used with exercise equipment, concerns the proper reading of the statute’s clarity and precision demand. According to the Federal Circuit, a patent claim passes the §112, ¶2 threshold so long as the claim is “amenable to construction,” and the claim, as construed, is not “insolubly ambiguous.” 715 F. 3d 891, 898–899 (2013). We conclude that the Federal Circuit’s formulation, which tolerates some ambiguous claims but not others, does not satisfy the statute’s definiteness requirement. In place of the “insolubly ambiguous” standard, we hold that a patent is invalid for indefiniteness if its claims, read in light of the specification delineating the patent, and the prosecution history, fail to inform, with reasonable certainty, those skilled in the art about the scope of the invention. Expressing no opinion on the validity of the patent-in-suit, we remand, instructing the Federal Circuit to decide the case employing the standard we have prescribed. I Authorized by the Constitution “[t]o promote the Progress of Science and useful Arts, by securing for limited Times to . . . Inventors the exclusive Right to their . . . Discoveries,” Art. I, §8, cl. 8, Congress has enacted patent laws rewarding inventors with a limited monopoly. “Th[at] monopoly is a property right,” and “like any property right, its boundaries should be clear.” Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U. S. 722, 730 (2002) . See also Markman v. Westview Instruments, Inc., 517 U. S. 370, 373 (1996) (“It has long been understood that a patent must describe the exact scope of an invention and its manufacture . . . .”). Thus, when Congress enacted the first Patent Act in 1790, it directed that patent grantees file a written specification “containing a description . . . of the thing or things . . . invented or discovered,” which “shall be so particular” as to “distinguish the invention or discovery from other things before known and used.” Act of Apr. 10, 1790, §2, 1Stat. 110. The patent laws have retained this requirement of definiteness even as the focus of patent construction has shifted. Under early patent practice in the United States, we have recounted, it was the written specification that “represented the key to the patent.” Markman, 517 U. S., at 379. Eventually, however, patent applicants began to set out the invention’s scope in a separate section known as the “claim.” See generally 1 R. Moy, Walker on Patents §4.2, pp. 4–17 to 4–20 (4th ed. 2012). The Patent Act of 1870 expressly conditioned the receipt of a patent on the inventor’s inclusion of one or more such claims, described with particularity and distinctness. See Act of July 8, 1870, §26, 16Stat. 201 (to obtain a patent, the inventor must “particularly point out and distinctly claim the part, improvement, or combination which [the inventor] claims as his invention or discovery”). The 1870 Act’s definiteness requirement survives today, largely unaltered. Section 112 of the Patent Act of 1952, applicable to this case, requires the patent applicant to conclude the specification with “one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.” 35 U. S. C. §112, ¶2 (2006 ed.). A lack of definiteness renders invalid “the patent or any claim in suit.” §282, ¶2(3).[1] II A The patent in dispute, U. S. Patent No. 5,337,753 (’753 patent), issued to Dr. Gregory Lekhtman in 1994 and assigned to respondent Biosig Instruments, Inc., concerns a heart-rate monitor for use during exercise. Previous heart-rate monitors, the patent asserts, were often inaccurate in measuring the electrical signals accompanying each heartbeat (electrocardiograph or ECG signals). The inaccuracy was caused by electrical signals of a different sort, known as electromyogram or EMG signals, generated by an exerciser’s skeletal muscles when, for example, she moves her arm, or grips an exercise monitor with her hand. These EMG signals can “mask” ECG signals and thereby impede their detection. App. 52, 147. Dr. Lekhtman’s invention claims to improve on prior art by eliminating that impediment. The invention focuses on a key difference between EMG and ECG waveforms: while ECG signals detected from a user’s left hand have a polarity opposite to that of the signals detected from her right hand,[2] EMG signals from each hand have the same polar-ity. The patented device works by measuring equalized EMG signals detected at each hand and then using circuitry to subtract the identical EMG signals from each other, thus filtering out the EMG interference. As relevant here, the ’753 patent describes a heart-rate monitor contained in a hollow cylindrical bar that a user grips with both hands, such that each hand comes into contact with two electrodes, one “live” and one “common.” The device is illustrated in figure 1 of the patent, id., at 41, reproduced in the Appendix to this opinion. Claim 1 of the ’753 patent, which contains the limitations critical to this dispute, refers to a “heart rate monitor for use by a user in association with exercise apparatus and/or exercise procedures.” Id., at 61. The claim “comprise[s],” among other elements, an “elongate member” (cylindrical bar) with a display device; “electronic circuitry including a difference amplifier”; and, on each half of the cylindrical bar, a live electrode and a common electrode “mounted . . . in spaced relationship with each other.” Ibid.[3] The claim sets forth additional elements, including that the cylindrical bar is to be held in such a way that each of the user’s hands “contact[s]” both electrodes on each side of the bar. Id., at 62. Further, the EMG signals detected by the two electrode pairs are to be “of substan-tially equal magnitude and phase” so that the difference amplifier will “produce a substantially zero [EMG] signal” upon subtracting the signals from one another. Ibid. B The dispute between the parties arose in the 1990’s, when Biosig allegedly disclosed the patented technology to StairMaster Sports Medical Products, Inc. According to Biosig, StairMaster, without ever obtaining a license, sold exercise machines that included Biosig’s patented technology, and petitioner Nautilus, Inc., continued to do so after acquiring the StairMaster brand. In 2004, based on these allegations, Biosig brought a patent infringement suit against Nautilus in the U. S. District Court for the Southern District of New York. With Biosig’s lawsuit launched, Nautilus asked the U. S. Patent and Trademark Office (PTO) to reexamine the ’753 patent. The reexamination proceedings centered on whether the patent was anticipated or rendered obvious by prior art—principally, a patent issued in 1984 to an inventor named Fujisaki, which similarly disclosed a heart-rate monitor using two pairs of electrodes and a difference amplifier. Endeavoring to distinguish the ’753 patent from prior art, Biosig submitted a declaration from Dr. Lekhtman. The declaration attested, among other things, that the ’753 patent sufficiently informed a person skilled in the art how to configure the detecting electrodes so as “to produce equal EMG [signals] from the left and right hands.” Id., at 160. Although the electrodes’ design variables—including spacing, shape, size, and material—cannot be standardized across all exercise machines, Dr. Lekhtman explained, a skilled artisan could undertake a “trial and error” process of equalization. This would entail experimentation with different electrode configurations in order to optimize EMG signal cancellation. Id., at 155–156, 158.[4] In 2010, the PTO issued a determination confirming the patentability of the ’753 patent’s claims. Biosig thereafter reinstituted its infringement suit, which the parties had voluntarily dismissed without prejudice while PTO reexamination was underway. In 2011, the District Court conducted a hearing to determine the proper construction of the patent’s claims, see Markman v. Westview Instruments, Inc., 517 U. S. 370 (1996) (claim construction is a matter of law reserved for court decision), including the claim term “in spaced relationship with each other.” According to Biosig, that “spaced relationship” referred to the distance between the live electrode and the common electrode in each electrode pair. Nautilus, seizing on Biosig’s submissions to the PTO during the reexamination, maintained that the “spaced relationship” must be a distance “greater than the width of each electrode.” App. 245. The District Court ultimately construed the term to mean “there is a defined relationship between the live electrode and the common electrode on one side of the cylindrical bar and the same or a different defined relationship between the live electrode and the common electrode on the other side of the cylindrical bar,” without any reference to the electrodes’ width. App. to Pet. for Cert. 43a–44a. Nautilus moved for summary judgment, arguing that the term “spaced relationship,” as construed, was indefinite under §112, ¶2. The District Court granted the motion. Those words, the District Court concluded, “did not tell [the court] or anyone what precisely the space should be,” or even supply “any parameters” for determining the appropriate spacing. Id., at 72a. The Federal Circuit reversed and remanded. A claim is indefinite, the majority opinion stated, “only when it is ‘not amenable to construction’ or ‘insolubly ambiguous.’ ” 715 F. 3d 891, 898 (2013) (quoting Datamize, LLC v. Plumtree Software, Inc., 417 F. 3d 1342, 1347 (CA Fed. 2005)). Under that standard, the majority determined, the ’753 patent survived indefiniteness review. Considering first the “intrinsic evidence”—i.e., the claim language, the specification, and the prosecution history—the majority discerned “certain inherent parameters of the claimed apparatus, which to a skilled artisan may be sufficient to understand the metes and bounds of ‘spaced relationship.’ ” 715 F. 3d, at 899. These sources of meaning, the majority explained, make plain that the distance separating the live and common electrodes on each half of the bar “cannot be greater than the width of a user’s hands”; that is so “because claim 1 requires the live and common electrodes to independently detect electrical signals at two distinct points of a hand.” Ibid. Furthermore, the major-ity noted, the intrinsic evidence teaches that this distance cannot be “infinitesimally small, effectively merging the live and common electrodes into a single electrode with one detection point.” Ibid. The claim’s functional provisions, the majority went on to observe, shed additional light on the meaning of “spaced relationship.” Surveying the record before the PTO on reexamination, the majority concluded that a skilled artisan would know that she could attain the indicated functions of equalizing and removing EMG signals by adjusting design variables, including spacing. In a concurring opinion, Judge Schall reached the majority’s result employing “a more limited analysis.” Id., at 905. Judge Schall accepted the majority’s recitation of the definiteness standard, under which claims amenable to construction are nonetheless indefinite when “the construction remains insolubly ambiguous.” Ibid. (internal quotation marks omitted). The District Court’s construction of “spaced relationship,” Judge Schall maintained, was sufficiently clear: the term means “there is a fixed spatial relationship between the live electrode and the common electrode” on each side of the cylindrical bar. Ibid. Judge Schall agreed with the majority that the intrinsic evidence discloses inherent limits of that spacing. But, unlike the majority, Judge Schall did not “presum[e] a functional linkage between the ‘spaced relationship’ limitation and the removal of EMG signals.” Id., at 906. Other limitations of the claim, in his view, and not the “ ‘spaced relationship’ limitation itself,” “included a functional requirement to remove EMG signals.” Ibid. We granted certiorari, 571 U. S. ___ (2014), and now vacate and remand. III A Although the parties here disagree on the dispositive question—does the ’753 patent withstand definiteness scrutiny—they are in accord on several aspects of the §112, ¶2 inquiry. First, definiteness is to be evaluated from the perspective of someone skilled in the relevant art. See, e.g., General Elec. Co. v. Wabash Appliance Corp., 304 U. S. 364, 371 (1938) . See also §112, ¶1 (patent’s specification “shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same” (emphasis added)). Second, in assessing definiteness, claims are to be read in light of the patent’s specification and prosecution history. See, e.g., United States v. Adams, 383 U. S. 39 –49 (1966) (specification); Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U. S. 722, 741 (2002) (prosecution history). Third, “[d]efiniteness is measured from the viewpoint of a person skilled in [the] art at the time the patent was filed.” Brief for Respondent 55 (emphasis added). See generally E. Manzo, Patent Claim Construction in the Federal Circuit §0.2, p. 9 (2014) (“Patent claims . . . should be construed from an objective perspective of a [skilled artisan], based on what the applicant actually claimed, disclosed, and stated during the application process.”). The parties differ, however, in their articulations of just how much imprecision §112, ¶2 tolerates. In Nautilus’ view, a patent is invalid when a claim is “ambiguous, such that readers could reasonably interpret the claim’s scope differently.” Brief for Petitioner 37. Biosig and the Solicitor General would require only that the patent provide reasonable notice of the scope of the claimed invention. See Brief for Respondent 18; Brief for United States as Amicus Curiae 9–10. Section 112, we have said, entails a “delicate balance.” Festo, 535 U. S., at 731. On the one hand, the definiteness requirement must take into account the inherent limitations of language. See ibid. Some modicum of uncertainty, the Court has recognized, is the “price of ensuring the appropriate incentives for innovation.” Id., at 732. One must bear in mind, moreover, that patents are “not addressed to lawyers, or even to the public generally,” but rather to those skilled in the relevant art. Carnegie Steel Co. v. Cambria Iron Co., 185 U. S. 403, 437 (1902) (also stating that “any description which is sufficient to apprise [steel manufacturers] in the language of the art of the definite feature of the invention, and to serve as a warning to others of what the patent claims as a monopoly, is sufficiently definite to sustain the patent”).[5] At the same time, a patent must be precise enough to afford clear notice of what is claimed, thereby “ ‘appris[ing] the public of what is still open to them.’ ” Markman, 517 U. S., at 373 (quoting McClain v. Ortmayer, 141 U. S. 419, 424 (1891) ).[6] Otherwise there would be “[a] zone of uncertainty which enterprise and experimentation may enter only at the risk of infringement claims.” United Carbon Co. v. Binney & Smith Co., 317 U. S. 228, 236 (1942) . And absent a meaningful definiteness check, we are told, patent applicants face powerful incentives to inject ambigu-ity into their claims. See Brief for Petitioner 30–32 (citing patent treatises and drafting guides). See also Federal Trade Commission, The Evolving IP Marketplace: Aligning Patent Notice and Remedies With Competition 85 (2011) (quoting testimony that patent system fosters “an incentive to be as vague and ambiguous as you can with your claims” and “defer clarity at all costs”).[7] Eliminating that temptation is in order, and “the patent drafter is in the best position to resolve the ambiguity in . . . patent claims.” Halliburton Energy Servs., Inc. v. M–I LLC, 514 F. 3d 1244, 1255 (CA Fed. 2008). See also Hormone Research Foundation, Inc. v. Genentech, Inc., 904 F. 2d 1558, 1563 (CA Fed. 1990) (“It is a well-established axiom in patent law that a patentee is free to be his or her own lexicographer . . . .”). To determine the proper office of the definiteness command, therefore, we must reconcile concerns that tug in opposite directions. Cognizant of the competing concerns, we read §112, ¶2 to require that a patent’s claims, viewed in light of the specification and prosecution history, inform those skilled in the art about the scope of the invention with reasonable certainty. The definiteness requirement, so understood, mandates clarity, while recognizing that absolute precision is unattainable. The standard we adopt accords with opinions of this Court stating that “the certainty which the law requires in patents is not greater than is reasonable, having regard to their subject-matter.” Minerals Separation, Ltd. v. Hyde, 242 U. S. 261, 270 (1916) . See also United Carbon, 317 U. S., at 236 (“claims must be reasonably clear-cut”); Markman, 517 U. S., at 389 (claim construction calls for “the necessarily sophisticated analysis of the whole document,” and may turn on evaluations of expert testimony). B In resolving Nautilus’ definiteness challenge, the Fed-eral Circuit asked whether the ’753 patent’s claims were “amenable to construction” or “insolubly ambiguous.” Those formulations can breed lower court confusion,[8] for they lack the precision §112, ¶2 demands. It cannot be sufficient that a court can ascribe some meaning to a patent’s claims; the definiteness inquiry trains on the understanding of a skilled artisan at the time of the patent application, not that of a court viewing matters post hoc. To tolerate imprecision just short of that rendering a claim “insolubly ambiguous” would diminish the definiteness requirement’s public-notice function and foster the innovation-discouraging “zone of uncertainty,” United Carbon, 317 U. S., at 236, against which this Court has warned. Appreciating that “terms like ‘insolubly ambiguous’ may not be felicitous,” Brief for Respondent 34, Biosig argues the phrase is a shorthand label for a more probing inquiry that the Federal Circuit applies in practice. The Federal Circuit’s fuller explications of the term “insolubly ambiguous,” we recognize, may come closer to tracking the statutory prescription. See, e.g., 715 F. 3d, at 898 (case below) (“[I]f reasonable efforts at claim construction result in a definition that does not provide sufficient particularity and clarity to inform skilled artisans of the bounds of the claim, the claim is insolubly ambiguous and invalid for indefiniteness.” (internal quotation marks omitted)). But although this Court does not “micromanag[e] the Federal Circuit’s particular word choice” in applying patent-law doctrines, we must ensure that the Federal Circuit’s test is at least “probative of the essential inquiry.” Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U. S. 17, 40 (1997) . Falling short in that regard, the expressions “insolubly ambiguous” and “amenable to construction” permeate the Federal Circuit’s recent decisions concerning §112, ¶2’s requirement.[9] We agree with Nautilus and its amici that such terminology can leave courts and the patent bar at sea without a reliable compass.[10] IV Both here and in the courts below, the parties have advanced conflicting arguments as to the definiteness of the claims in the ’753 patent. Nautilus maintains that the claim term “spaced relationship” is open to multiple interpretations reflecting markedly different understandings of the patent’s scope, as exemplified by the disagreement among the members of the Federal Circuit panel.[11] Biosig responds that “spaced relationship,” read in light of the specification and as illustrated in the accompanying drawings, delineates the permissible spacing with sufficient precision. “[M]indful that we are a court of review, not of first view,” Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005), we decline to apply the standard we have announced to the controversy between Nautilus and Biosig. As we have explained, the Federal Circuit invoked a standard more amorphous than the statutory definiteness requirement allows. We therefore follow our ordinary practice of remanding so that the Court of Appeals can reconsider, under the proper standard, whether the relevant claims in the ’753 patent are sufficiently definite. See, e.g., Johnson v. California, 543 U. S. 499, 515 (2005) ; Gasperini v. Center for Humanities, Inc., 518 U. S. 415, 438 (1996) . * * * For the reasons stated, we vacate the judgment of the United States Court of Appeals for the Federal Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered. APPENDIX Patent No. 5,337,753, Figure 1Notes 1 In the Leahy-Smith America Invents Act, Pub. L. 112–29, , enacted in 2011, Congress amended several parts of the Patent Act. Those amendments modified §§112 and 282 in minor respects not pertinent here. In any event, the amended versions of those provisions are inapplicable to patent applications filed before September 16, 2012, and proceedings commenced before September 16, 2011. See §§4(e), 15(c), 20(), , notes following 35 U. S. C. §§2, 111, 119. Here, the application for the patent-in-suit was filed in 1992, and the relevant court proceedings were initiated in 2010. Accordingly, this opinion’s citations to the Patent Act refer to the 2006 edition of the United States Code. 2 This difference in polarity occurs because the heart is not aligned vertically in relation to the center of the body; the organ tilts leftward from apex to bottom. App. 213. 3 As depicted in figure 1 of the patent, at 41, reproduced in the Appendix to this opinion, the live electrodes are identified by numbers 9 and 13, and the common electrodes, by 11 and 15. 4 Dr. Lekhtman’s declaration also referred to an expert report prepared by Dr. Henrietta Galiana, Chair of the Department of Biomedical Engineering at McGill University, for use in the infringement litigation. That report described how Dr. Galiana’s laboratory technician, equipped with a wooden dowel, wire, metal foil, glue, electrical tape, and the drawings from the ’753 patent, was able in two hours to build a monitor that “worked just as described in the . . . patent.” , at 226. 5 See also v. , –66 (1923) (upholding as definite a patent for an improvement to a paper-making machine, which provided that a wire be placed at a “high” or “substantial elevation,” where “readers . . . skilled in the art of paper making and versed in the use of the . . . machine” would have “no difficulty . . . in determining . . . the substantial [elevation] needed” for the machine to operate as specified). 6 See also v. , (“The statutory requirement of particularity and distinctness in claims is met only when they clearly distinguish what is claimed from what went before in the art and clearly circumscribe what is foreclosed from future enterprise.”); v. , (“The limits of a patent must be known for the protection of the patentee, the encouragement of the inventive genius of others and the assurance that the subject of the patent will be dedicated ultimately to the public.”). 7 Online at http: / / www. ftc.gov / sites / default / files / documents /reports / evolving - ip - marketplace - aligning - patent - notice - and - remedies -competition - report - federal - trade / 110307patentreport.pdf (as visited May 30, 2014, and available in Clerk of Court’s case file). 8 See, v. , ___ F. Supp. 2d ___, ___, 2014 WL 869092, *4 (MD Fla., Mar. 5, 2014) (finding that “the account,” as used in claim, “lacks definiteness,” because it might mean several different things and “no informed and confident choice is available among the contending definitions,” but that “the extent of the indefiniteness . . . falls far short of the ‘insoluble ambiguity’ required to invalidate the claim”). 9 , v. , 600 F. 3d 1357, 1366 (CA Fed. 2010) (“the definiteness of claim terms depends on whether those terms can be given any reasonable meaning”); v. , 417 F. 3d 1342, 1347 (CA Fed. 2005) (“Only claims ‘not amenable to construction’ or ‘insolubly ambiguous’ are indefinite.”); v. , 265 F. 3d 1371, 1375 (CA Fed. 2001) (“If a claim is insolubly ambiguous, and no narrowing construction can properly be adopted, we have held the claim indefinite.”). See also Dept. of Commerce, Manual of Patent Examining Procedure §2173.02(I), p. 294 (9th ed. 2014) (PTO manual describing Federal Circuit’s test as upholding a claim’s validity “if some meaning can be gleaned from the language”). 10 The Federal Circuit suggests that a permissive definiteness standard “ ‘accord[s] respect to the statutory presumption of patent validity.’ ” 715 F. 3d 891, 902 (2013) (quoting , 265 F. 3d, at 1375). See also §282, ¶1 (“[a] patent shall be presumed valid,” and “[t]he burden of establishing invalidity of a patent or any claim thereof shall rest on the party asserting such invalidity”); v. , 564 U. S. ___, ___ (2011) (slip op., at 1) (invalidity defenses must be proved by “clear and convincing evidence”).As the parties appear to agree, however, this presumption of validity does not alter the degree of clarity that §112, ¶2 demands from patent applicants; to the contrary, it incorporates that definiteness requirement by reference. See §282, ¶2(3) (defenses to infringement actions include “[i]nvalidity of the patent or any claim in suit for failure to comply with . . . any requirement of [§112]”). 11 Notably, however, all three panel members found Nautilus’ arguments unavailing.

572.US.273

Petitioner Northwest, Inc., terminated respondent’s membership in its frequent flyer program, apparently based on a provision in the frequent flyer agreement that gave Northwest sole discretion to determine whether a participant had abused the program. Respondent filed suit, asserting, as relevant here, that Northwest had breached its contract by revoking his membership status without valid cause and had violated the duty of good faith and fair dealing because it terminated his membership in a way that contravened his reasonable expectations. The District Court found that the Airline Deregulation Act of 1978 (ADA) pre-empted the breach of the duty of good faith and fair dealing claim and dismissed the breach of contract claim without prejudice. Respondent appealed only the dismissal of his breach of the duty of good faith and fair dealing claim. The Ninth Circuit reversed, finding that claim “ ‘too tenuously connected to airline regulation to trigger’ ” ADA pre-emption. Held: 1. The ADA pre-empts a state-law claim for breach of the implied covenant of good faith and fair dealing if it seeks to enlarge the contractual obligations that the parties voluntarily adopt. Pp. 4–10. (a) Before the ADA was enacted, air carriers’ routes, rates, and services were regulated under the Federal Aviation Act of 1958. And because that Act contained a saving provision preserving pre-existing statutory and common-law remedies, air carriers were also regulated by the States. The ADA did not repeal that saving provision, but it did include a pre-emption provision to prohibit States from “enact[ing] or enforc[ing] a law, regulation, or other provision having the force and effect of law related to [an air carrier’s] price, route, or service,” 49 U. S. C. §41713(b)(1), thus ensuring that “States would not undo federal deregulation with regulation of their own,” Morales v. Trans World Airlines, Inc., 504 U.S. 374, 378. In Morales, the Court recognized that the key phrase “related to” expresses a “broad pre-emptive purpose,” id., at 383, and held that the ADA pre-empted the use of state consumer protection laws to regulate airline advertising, concluding that “relat[es] to” means “ha[s] a connection with, or reference to, airline ‘rates, routes, or services,’ ” id., at 384. And in American Airlines, Inc. v. Wolens, 513 U.S. 219, the Court found that the ADA pre-empted the use of an Illinois consumer law to challenge an airline’s devaluation of frequent flyer earned miles. But it did not pre-empt breach of contract claims because “terms and conditions airlines offer and passengers accept are privately ordered obligations” not “ ‘a State’s “enact[ment] or enforce[ment] [of] any law, rule, regulation, standard, or other provision having the force and effect of law” within the [pre-emption provision’s] meaning.’ ” Id., at 228–229. Pp. 4–6. (b) The phrase “other provision having the force and effect of law” includes state common-law rules like the implied covenant at issue. Common-law rules are routinely called “provisions,” see, e.g., Madsen v. Women’s Health Center, Inc., 512 U.S. 753, 765, n. 3, and they clearly have “the force and effect of law.” The pre-emption provision’s original language confirms this understanding. As first enacted, the provision also applied to “rule[s]” and “standard[s],” a formulation encompassing common-law rules. See CSX Transp., Inc. v. Easterwood, 507 U.S. 658, 664. And Congress made clear that the deletion of those terms as part of Title 49’s wholesale recodification effected no “substantive change.” §1(a), 108Stat. 745. Respondent’s reliance on Sprietsma v. Mercury Marine, 537 U.S. 51, is misplaced. There, the Court held that the Federal Boat Safety Act of 1971 did not pre-empt a common-law tort claim, but that Act’s pre-emption provision is more narrowly worded than the ADA provision. The Boat Safety Act’s saving and pre-emption provisions were also enacted at the same time, while the Federal Aviation Act’s general remedies saving clause is “a relic of the pre-ADA/no pre-emption regime,” Morales, 504 U. S., at 385, that “cannot be allowed to supersede the specific substantive pre-emption provision,” ibid. Exempting common-law claims would also disserve the ADA’s central purpose, which was to eliminate federal regulation of rates, routes, and services so they could be set by market forces. Finally, if all state common-law rules fell outside the pre-emption provision’s ambit, Wolens would not have singled out a subcategory, for common-law claims based on the parties’ voluntary undertaking, as falling outside that provision’s coverage. Pp. 6–9. (c) Respondent’s claim “relates to” “rates, routes, or services.” It clearly has “a connection with or reference to airline” prices, routes, or services, Morales, 504 U. S., at 384. As in Wolens, Northwest’s program connects to the airline’s “rates” by awarding mileage credits redeemable for tickets and upgrades, thus eliminating or reducing ticket prices. It also connects to “services,” i.e., access to flights and higher service categories. Respondent’s counterarguments are unpersuasive. His claim that he is contesting his termination, not access to flights or upgrades, ignores his reason for seeking reinstatement: to obtain reduced rates and enhanced services. Although respondent and amici claim there have been fundamental changes in the way that frequent flyer miles are earned since Wolens was decided, that does not matter here where respondent did not assert that he earned miles from any activity but taking flights or that he attempted to redeem miles for anything but tickets and upgrades. Pp. 9–10. 2. Because respondent’s implied covenant claim seeks to enlarge his contractual agreement with petitioners, it is pre-empted by §41713(b)(1). Under Minnesota law, which controls here, the implied covenant must be regarded as a state-imposed obligation. Minnesota law does not permit parties to contract out of the covenant. And when a State’s law does not authorize parties to free themselves from the covenant, a breach of covenant claim is pre-empted under Wolens. As an independent basis for this conclusion, if, as Minnesota law provides, the implied covenant applies to “every contract” except employment contracts for “policy reasons,” then the decision not to exempt other types of contracts must likewise be based on a policy determination, namely, that the policy reason for the employment contract rule does not apply in other contexts. Petitioners claim that the refusal to pre-empt all implied covenant claims, regardless of state law, will lead to a patchwork of rules that will frustrate the ADA’s deregulatory aim. But airlines can avoid such a result if they contract out of covenants where permitted by state law. Nor are participants in frequent flyer programs left without protection. They can avoid an airline with a poor reputation and possibly enroll in a more favorable rival program. Moreover, the Department of Transportation has the authority to investigate complaints about frequent flyer programs. Finally, respondent might have been able to vindicate his claim of ill treatment by Northwest had he appealed his breach of contract claim. Pp. 10–14. 695 F.3d 873, reversed and remanded. Alito, J., delivered the opinion for a unanimous Court.

We must decide in this case whether the Airline Deregulation Act pre-empts a state-law claim for breach of the implied covenant of good faith and fair dealing. Following our interpretation of the Act in American Airlines, Inc. v. Wolens, 513 U. S. 219 (1995) , we hold that such a claim is pre-empted if it seeks to enlarge the contractual obligations that the parties voluntarily adopt. And because the doctrine is invoked in the present case in an attempt to expand those obligations, we reverse the judgment of the Court of Appeals. I A Like many airlines, petitioner Northwest, Inc. (Northwest), established a frequent flyer program, its World-Perks Airline Partners Program, to attract loyal cus-tomers. Under this program, members are able to earn “miles” by taking flights operated by Northwest and other “partner” airlines. Members can then redeem these miles for tickets and service upgrades with Northwest or its airline partners. Respondent became a member of Northwest’s World-Perks program in 1999, and as a result of extensivetravel on Northwest flights, he achieved “Platinum Elite” status (the highest level available) in 2005. In 2008, however, Northwest terminated respondent’s membership, apparently in reliance on a provision of the WorldPerks agreement that provided that “[a]buse of the . . . program (including . . . improper conduct as determined by [Northwest] in its sole judgment[ ) ] . . . may result in cancellation of the member’s account.” App. 64–65. According to respondent, a Northwest representative telephoned him in June 2008 and informed him that his“Platinum Elite” status was being revoked because he had “ ‘abused’ ” the program. Id., at 35. In a letter sent about two weeks later, Northwest wrote: “[Y]ou have contacted our office 24 times since December 3, 2007 regarding travel problems, including9 incidents of your bag arriving late at the luggage carousel. . . . . . . . . “ Since December 3, 2007, you have continually asked for compensation over and above our guidelines. We have awarded you $1,925.00 in travel credit vouchers, 78,500 WorldPerks bonus miles, a voucher extension for your son, and $491.00 in cash reimbursements. . . . “ Due to our past generosity, we must respectfully advise that we will no longer be awarding you compensation each time you contact us.” Id., at 58–59. Respondent requested clarification of his status, but a Northwest representative sent him an e-mail stating that “[a]fter numerous conversations with not only the Legal Department, but with members of the WorldPerks department, I believe your status with the program should be very clear.” Id., at 60. B Alleging that Northwest had ended his membership asa cost-cutting measure tied to Northwest’s merger with Delta Air Lines, respondent filed a class action in the United States District Court for the Southern District of California on behalf of himself and all other similarly situated WorldPerks members. Respondent’s complaint asserted four separate claims.[1] First, his complaint alleged that Northwest had breached its contract by re-voking his “Platinum Elite” status without valid cause. Second, the complaint claimed that Northwest violated the duty of good faith and fair dealing because it terminated his membership in a way that contravened his reasonable expectations with respect to the manner in which Northwest would exercise its discretion. Third, the complaint asserted a claim for negligent misrepresentation, and fourth, the complaint alleged intentional misrepresentation. Respondent sought damages in excess of $5 million, as well as injunctive relief requiring Northwest to restore the class members’ WorldPerks status and prohibiting Northwest from future revocations of membership. The District Court held that respondent’s claims for breach of the covenant of good faith and fair dealing, negligent misrepresentation, and intentional misrepresentation were pre-empted by the Airline Deregulation Act of 1978 (ADA or Act) as amended, 49 U. S. C. §41713. These claims, the court concluded, were “relate[d] to” Northwest’s rates and services and thus fell within the ADA’s express pre-emption clause. App. to Pet. for Cert. 69. Respondent’s remaining claim—for breach of contract—was dismissed without prejudice under Federal Rule of Civil Procedure 12(b)(6). The court held that respondent had failed to identify any material breach because the frequent flyer agreement gave Northwest sole discretion to determine whether a participant had abused the program. Respondent appealed the dismissal of his breach of the duty of good faith and fair dealing claim but not the other claims that the court had dismissed. The Ninth Circuit reversed. 695 F. 3d 873 (2012). Relying on pre-Wolens Circuit precedent, the Ninth Circuit first held that a breach of implied covenant claim is “ ‘too tenuously connected to airline regulation to trigger preemption under the ADA.’ ” 695 F. 3d, at 879. Sucha claim, the Ninth Circuit wrote, “does not interfere with the [Act’s] deregulatory mandate” and does not “ ‘forcethe Airlines to adopt or change their prices, routes or services—the prerequisite for . . . preemption.’ ” Id., at 880. In addition, the Court held that the covenant of good faith and fair dealing does not fall within the terms of the Act’s pre-emption provision because it does not have a “direct effect” on either “prices” or “services.” Id., at 877, 881. We granted certiorari. 569 U. S. ___ (2013). II A Before the enactment of the ADA, the Federal Aviation Act of 1958 empowered the Civil Aeronautics Board to regulate the interstate airline industry. Pursuant to this authority, the Board closely regulated air carriers, controlling, among other things, routes, rates, and services. See, e.g., Western Air Lines, Inc. v. CAB, 347 U. S. 67 (1954) ; Federal Aviation Act of 1958, 72Stat. 731. And since the Federal Aviation Act contained a saving provision preserving pre-existing statu-tory and common-law remedies, §1106, id., at 798, air carriers were also regulated by the States. See Morales v. Trans World Airlines, Inc., 504 U. S. 374, 378 (1992) . In 1978, however, Congress enacted the ADA, which sought to promote “efficiency, innovation, and low prices” in the airline industry through “maximum reliance on competitive market forces and on actual and potential competition.” 49 U. S. C. §§40101(a)(6), (12)(A). While the ADA did not repeal the predecessor law’s saving provision, it included a pre-emption provision in order to “ensure that the States would not undo federal deregulation with regulation of their own.” Morales supra, at 378. In its current form, this provision states that “a State, political subdivision of a State, or political authority of at least 2 States may not enact or enforce a law, regulation, or other provision having the force and effect of law related to a price, route, or service of an air carrier that may provide air transportation under this subpart.” §41713(b)(1). We have had two occasions to consider the ADA’s pre-emptive reach. In Morales, we held that the ADApre-empted the use of state consumer protection laws to regulate airline advertising. We recognized that the key phrase “related to” expresses a “broad pre-emptive purpose.” 504 U. S., at 383. Noting our interpretation of similar language in the pre-emption provision of the Employee Retirement and Income Security Act of 1974, 29 U. S. C. §1144(a), we held that a claim “relat[es] to rates, routes, or services,” within the meaning of the ADA, if the claim “ha[s] a connection with, or reference to, airline ‘rates, routes, or services.’ ” 504 U. S., at 384. The older saving provision, we concluded, did not undermine this conclusion. Id., at 384–385. Subsequently, in American Airlines, Inc. v. Wolens, 513 U. S. 219 (1995) , we considered the application of the ADA pre-emption provision to two types of claims concerning an airline’s frequent flyer program: first, claims under the Illinois Consumer Fraud and Deceptive Business Practices Act challenging an airline’s devaluation of earned miles (chiefly as the result of the imposition of “blackout dates” and limits on the number of seats available for customers wishing to obtain tickets by using those miles) and, second, breach of contract claims. We reaffirmed Morales’ broad interpretation of the ADA pre-emption provision and held that this provision barred the claims based on the Illinois statute but not the breach-of-contract claims. “[T]erms and conditions airlines offer and passengers accept,” we wrote, “are privately ordered obligations and thus do not amount to a State’s ‘enact[ment] or enforce[ment] [of] any law, rule, regulation, standard, or other provision having the force and effect of law’ within the meaning of [the ADA pre-emption provision].” 513 U. S., at 228–229. With this background in mind, we turn to the question whether the ADA pre-empts respondent’s claim for breach of the implied covenant of good faith and fair dealing. B The first question we address is whether, as respondent now maintains, the ADA’s pre-emption provision applies only to legislation enacted by a state legislature and regulations issued by a state administrative agency but not to a common-law rule like the implied covenant of good faith and fair dealing. We have little difficulty rejecting this argument. To begin, state common-law rules fall comfortably within the language of the ADA pre-emption provision. As noted above, the current version of this provision applies to state “law[s], regulation[s], or other provision[s] having the force and effect of law,” 49 U. S. C. §41713(b)(1). It is routine to call common-law rules “provisions.” See, e.g., Madsen v. Women’s Health Center, Inc., 512 U. S. 753, 765, n. 3 (1994) ; United States v. Barnett, 376 U. S. 681 –700 (1964); Brown v. United Airlines, Inc., 720 F. 3d 60, 68 (CA1 2013) (“[W]hen read in context, the word ‘provision’ in the ADA preemption provision can most appropriately be construed to include common law”). And a common-law rule clearly has “the force and effect of law.” In Wolens, we noted that this phrase is most naturally read to “ ‘refe[r] to binding standards of conduct that operate irrespective of any private agreement,’ ” 513 U. S., at 229, n. 5, and we see no basis for holding that such standards must be based on a statute or regulation as opposed to the common law. This understanding becomes even clearer when the original wording of the pre-emption provision is taken into account. When first enacted in 1978, this provision also applied to “rule[s]” and “standard[s],” and there surely can be no doubt that this formulation encompassed common-law rules. Indeed, we held in CSX Transp., Inc. v. Easterwood, 507 U. S. 658, 664 (1993) , that virtually identical language in the Federal Railroad Safety Act of 1970 includes “[l]egal duties imposed . . . by the common law.” See also Riegel v. Medtronic, Inc., 552 U. S. 312, 324 (2008) (holding that a State’s “ ‘requirements’ ” “includ[e] [the state’s] common-law duties”). While “rule[s]” and “standard[s]” are not mentioned in the current version of the statute, this omission is the result of a recodification that was not meant to affect the provision’s meaning. Those additional terms were deleted as part of a wholesale recodification of Title 49 in 1994, but Congress made it clear that this recodification did not effect any “substantive change.” § 1(a), 108Stat. 745. In arguing that common-law rules fall outside the scope of the ADA pre-emption provision, respondent relies on our decision in Sprietsma v. Mercury Marine, 537 U. S. 51 (2002) , which held that the Federal Boat Safety Act of 1971 did not pre-empt a common-law tort claim, but there are critical differences between the pre-emption provisions in the Boat Safety Act and the ADA. The Boat Safety Act provision applies only to “a law or regulation,” 46 U. S. C. §4306, whereas the ADA provision, as just explained, is much more broadly worded. In addition, the relationship between the ADA’s pre-emption provision and the saving provision carried over from the prior law is also quite different. The Sprietsma decision placed substantial weight on the Boat Safety Act’s saving provision, which was enacted at the same time as the pre-emption provision, but we have described the Federal Aviation Act saving clause as “a relic of the pre-ADA/no pre-emption regime.” Morales, 504 U. S., at 385. That provision applies to the entire, sprawling Federal Aviation Act, and not just to the ADA, and as we held in Morales, this “general ‘remedies’ saving clause cannot be allowed to supersede the specific substantive pre-emption provision.” Ibid. See also Wolens, supra, at 245 (O’Connor, J., concurring in judgment in part and dissenting in part). For these reasons, respondent’s interpretation of the ADA pre-emption provision cannot be squared with the provision’s terms. Exempting common-law claims would also disserve the central purpose of the ADA. The Act eliminated federal regulation of rates, routes, and services in order to allow those aspects of air transportation to be set by market forces, and the pre-emption provision was included to prevent the States from undoing what the Act was meant to accomplish. Morales, supra, at 378. What is important, therefore, is the effect of a state law, regulation, or provision, not its form, and the ADA’s deregulatory aim can be undermined just as surely by a state common-law rule as it can by a state statute or regulation. See Medtronic, Inc., supra, at 325 (recognizing that state tort law that imposes certain requirements would “disrup[t] the federal scheme no less than state regulatory law to the same effect”). As the First Circuit has recognized, “[i]t defies logic to think that Congress would disregard real-world consequences and give dispositive effect to the form of a clear intrusion into a federally regulated industry.” Brown, 720 F. 3d, at 66–67. Finally, if all state common-law rules fell outside the ambit of the ADA’s pre-emption provision, we would have had no need in Wolens to single out a subcategory of common-law claims, i.e., those based on the parties’ voluntary undertaking, as falling outside that provision’s coverage. Accordingly, we conclude that the phrase “other provision having the force and effect of law” includes common-law claims. C We must next determine whether respondent’s breach of implied covenant claim “relates to” “rates, routes, or services.” A claim satisfies this requirement if it has “a connection with, or reference to, airline” prices, routes, or services, Morales, supra, at 384, and the claim at issue here clearly has such a connection. That claim seeks respondent’s reinstatement in Northwest’s frequent flyer program so that he can access the program’s “valuable . . . benefits,” including “flight upgrades, accumulated mileage, loyalty program status or benefits on other airlines, and other advantages.” App. 49–50. Like the frequent flyer program in Wolens, the Northwest program is connected to the airline’s “rates” because the program awards mileage credits that can be redeemed for tickets and upgrades. See 513 U. S., at 226. When miles are used in this way, the rate that a customer pays, i.e., the price of a particular ticket, is either eliminated or reduced. The program is also connected to “services,” i.e., access to flights and to higher service categories. Ibid. Respondent argues that his claim differs from the claims in Wolens because he “does not challenge access to flights and upgrades or the number of miles needed to obtain air tickets” but instead contests “the termination of his WorldPerks elite membership,” Brief for Respondent 12, but this argument ignores respondent’s reason for seeking reinstatement of his membership, i.e., to obtain reduced rates and enhanced services. Respondent’s proffered distinction has no substance. Respondent and amici suggest that Wolens is not controlling because frequent flyer programs have fundamentally changed since the time of that decision. We are told that “most miles [are now] earned without consuming airline services” and are “spent without consuming airline services.” Brief for State of California et al. 18 (emphasis deleted). But whether or not this alleged change might have some impact in a future case, it is not implicated here. In this case, respondent did not assert that he earned his miles from any activity other than taking flights or that he attempted to redeem miles for anything other than tickets and upgrades. See Tr. of Oral Arg. 47–48. III With these preliminary issues behind us, we turn to the central issue in this case, i.e., whether respondent’s implied covenant claim is based on a state-imposed obligation or simply one that the parties voluntarily undertook. Petitioners urge us to hold that implied covenant claims are always pre-empted, and respondent suggests that such claims are generally not pre-empted, but the reasoning of Wolens neither dooms nor spares all such claims. While most States recognize some form of the good faith and fair dealing doctrine, it does not appear that there is any uniform understanding of the doctrine’s precise meaning. “[T]he concept of good faith in the performance of contracts ‘is a phrase without general meaning (or meanings) of its own.’ ” Tymshare, Inc. v. Covell, 727 F. 2d 1145, 1152 (CADC 1984) (Scalia, J.) (quoting Summers, “Good Faith” in General Contract Law and the Sales Provisions of the Uniform Commercial Code, 54 Va. L. Rev. 195, 201 (1968)); see also Burton, Breach of Contract and the Common Law Duty to Perform in Good Faith, 94 Harv. L. Rev. 369, 371 (1980). Of particular importance here, while some States are said to use the doctrine “to effectuate the intentions of parties or to protect their rea-sonable expectations,” ibid., other States clearly em-ploy the doctrine to ensure that a party does not “ ‘violate community standards of decency, fairness, or reasonableness.’ ” Universal Drilling Co., LLC v. R & R Rig Service, LLC, 2012 WY 31, 37, 271 P. 3d 987, 999; DDP Roofing Services, Inc. v. Indian River School Dist., 2010 WL 4657161, *3 (Del. Super. Ct., Nov. 16, 2010); Allworth v. Howard Univ., 890 A. 2d 194, 201–202 (D. C. 2006); Brunswick Hills Racquet Club, Inc. v. Route 18 Shopping Center Assocs., 182 N. J. 210, 224, 864 A. 2d 387, 395396 (2005); Harper v. Healthsource New Hampshire, 140 N. H. 770, 776, 674 A. 2d 962, 965–966 (1996); Borys v. Josada Builders, Inc., 110 Ill. App. 3d 29, 32–33, 441 N. E. 2d 1263, 1265–1266 (1982); Restatement (Second) of Contracts §205, Comment a (1979). See also Summers, The General Duty of Good Faith—Its Recognition and Conceptualization, 67 Cornell L. Rev. 810, 812 (1982). Whatever may be the case under the law of other jurisdictions, it seems clear that under Minnesota law, which is controlling here, see n. 1, supra, the implied covenant must be regarded as a state-imposed obligation.[2] Re-spondent concedes that under Minnesota law parties cannot contract out of the covenant. See Tr. of Oral Arg. 33–34; see also In re Hennepin Cty. 1986 Recycling Bond Litigation, 540 N. W. 2d 292, 502 (Minn. 1995); Sterling Capital Advisors, Inc. v. Herzog, 575 N. W. 2d 121, 125 (Minn. App. 1998); Minnwest Bank Central v. Flagship Properties LLC, 689 N. W. 2d 295, 303 (Minn. App. 2004). And as a leading commentator has explained, a State’s “unwillingness to allow people to disclaim the obligation of good faith . . . shows that the obligation cannot be implied, but is law imposed.” 3A A. Corbin, Corbin on Contracts §654A, p. 88 (L. Cunningham & A. Jacobsen eds. Supp. 1994). When the law of a State does not authorize parties to free themselves from the covenant, a breach of covenant claim is pre-empted under the reasoning of Wolens. Another feature of Minnesota law provides an additional, independent basis for our conclusion. Minnesota law holds that the implied covenant applies to “every contract,” In re Hennepin Cty., supra, at 502, with the notable exception of employment contracts. Hunt v. IBM Mid America Employees Fed. Credit Union, 384 N. W. 2d 853, 857–858 (Minn. 1986). The exception for employment contracts is based, in significant part, on “policy reasons,” id., at 858, and therefore the decision not to exempt other types of contracts must be based on a policy determination, namely, that the “policy reasons” that support the rule for employment contracts do not apply (at least with the same force) in other contexts. When the application of the implied covenant depends on state policy, a breachof implied covenant claim cannot be viewed as simply an attempt to vindicate the parties’ implicit understanding of the contract. For these reasons, the breach of implied covenant claim in this case cannot stand, but petitioners exhort us to go further and hold that all such claims, no matter the content of the law of the relevant jurisdiction, are pre-empted. If pre-emption depends on state law, petitioners warn, airlines will be faced with a baffling patchwork of rules, and the deregulatory aim of the ADA will be frustrated. But the airlines have means to avoid such a result. A State’s implied covenant rules will escape pre-emption only if the law of the relevant State permits an airline to contract around those rules in its frequent flyer program agreement, and if an airline’s agreement is governed by the law of such a State, the airline can specify that the agreement does not incorporate the covenant. While the inclusion of such a provision may impose transaction costs and presumably would not enhance the attractiveness of the program, an airline can decide whether the benefits of such a provision are worth the potential costs. Our holding also does not leave participants in frequent flyer programs without protection. The ADA is based on the view that the best interests of airline passengers are most effectively promoted, in the main, by allowing the free market to operate. If an airline acquires a reputation for mistreating the participants in its frequent flyer program (who are generally the airline’s most loyal and valuable customers), customers can avoid that program and may be able to enroll in a more favorable rival program. Federal law also provides protection for frequent flyer program participants. Congress has given the Department of Transportation (DOT) the general authority to prohibit and punish unfair and deceptive practices in air transportation and in the sale of air transportation, 49 U. S. C. §41712(a), and Congress has specifically authorized the DOT to investigate complaints relating to frequent flyer programs. See FAA Modernization andReform Act of 2012, §408(6), 126Stat. 87. Pursuant to these provisions, the DOT regularly entertains and acts on such complaints.[3] We note, finally, that respondent’s claim of ill treatment by Northwest might have been vindicated if he had pursued his breach-of-contract claim after its dismissal by the District Court. Respondent argues that, contrary to the holding of the District Court, the frequent flyer agreement did not actually give Northwest unfettered discretion to terminate his membership in the program, see Brief for Respondent 20–21, and the United States makes a related argument, namely, that even if the agreement gave Northwest complete discretion with respect to a determination regarding abuse of the program, the agreement did not necessarily bar a claim asserting that membership was ended for an ulterior reason, such as an effort to cut costs. If respondent had appealed the dismissal of his breach-of-contract claim, he could have presented these arguments to the Court of Appeals, but he chose not to press that claim. He voluntarily dismissed the breach-of-contract claim and instead appealed only the breach of implied covenant claim, which we hold to be pre-empted. * * * Because respondent’s implied covenant of good faithand fair dealing claim seeks to enlarge his contractual agreement with petitioners, we hold that 49 U. S. C. §41713(b)(1) pre-empts the claim. The judgment of the Court of Appeals for the Ninth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Applying California choice-of-law rules, the District Court held that Minnesota law applies because respondent “was a resident of Minneapolis, appears to fly in and out of Minnesota, and Northwest’s principal place of business is Minnesota.” App. to Pet. for Cert. 70. That determination was not challenged on appeal. 2 Like Minnesota, some other States preclude a party from waiving the obligations of good faith and fair dealing. v. , 927 So. 2d 810, 813, n. 5 (Ala. 2005); v. , 240 P. 3d 834, 844 (Alaska 2010); v. , 201 Ariz. 474, 491, 38 P. 3d 12, 29 (2002); v. , 224 Conn. 231, 238, 618 A. 2d 501, 505 (1992); v. , 878 A. 2d 434, 442 (Del. 2005); v. , 933 A. 2d 314, 333 (D. C. 2007); v. , 873 F. Supp. 808, 815 (SDNY 1994); v. , 83 S. W. 3d 647, 652 (Mo. App. 2002) (“When terms are present that directly nullify the implied covenants of good faith and reasonable efforts, . . . the contract is void for lack of mutuality”); v. , 2008 WY 134, ¶31, 196 P. 3d 184, 196. 3 See DOT, Air Travel Consumer Report 44 (Feb. 2014), online at http://www.dot.gov/sites/dot.dev/files/docs/2014_February_ATCR.pdf (as visited Mar. 31, 2014, and available in Clerk of Court’s case file).

572.US.545

The Patent Act’s fee-shifting provision authorizes district courts to award attorney’s fees to prevailing parties in “exceptional cases.” 35 U. S. C. §285. In Brooks Furniture Mfg., Inc. v. Dutailier Int’l, Inc., 393 F.3d 1378, 1381, the Federal Circuit defined an “exceptional case” as one which either involves “material inappropriate conduct” or is both “objectively baseless” and “brought in subjective bad faith.” Brooks Furniture also requires that parties establish the “exceptional” nature of a case by “clear and convincing evidence.” Id., at 1382. Respondent ICON Health & Fitness, Inc., sued petitioner Octane Fitness, LLC, for patent infringement. The District Court granted summary judgment to Octane. Octane then moved for attorney’s fees under §285. The District Court denied the motion under the Brooks Furniture framework, finding ICON’s claim to be neither objectively baseless nor brought in subjective bad faith. The Federal Circuit affirmed. Held: The Brooks Furniture framework is unduly rigid and impermissibly encumbers the statutory grant of discretion to district courts. Pp. 7–12. (a) Section 285 imposes one and only one constraint on district courts’ discretion to award attorney’s fees: The power is reserved for “exceptional” cases. Because the Patent Act does not define “exceptional,” the term is construed “in accordance with [its] ordinary meaning.” Sebelius v. Cloer, 569 U. S. ___, ___. In 1952, when Congress used the word in §285 (and today, for that matter), “[e]xceptional” meant “uncommon,” “rare,” or “not ordinary.” Webster’s New International Dictionary 889 (2d ed. 1934). An “exceptional” case, then, is simply one that stands out from others with re-spect to the substantive strength of a party’s litigating position (considering both the governing law and the facts of the case) or the unreasonable manner in which the case was litigated. District courts may determine whether a case is “exceptional” in the case-by-case exercise of their discretion, considering the totality of the circumstances. Cf. Fogerty v. Fantasy, Inc., 510 U.S. 517. Pp. 7–8. (b) The Brooks Furniture framework superimposes an inflexible framework onto statutory text that is inherently flexible. Pp. 8–11. (1) Brooks Furniture is too restrictive in defining the two categories of cases in which fee awards are allowed. The first category—cases involving litigation or certain other misconduct—appears to extend largely to independently sanctionable conduct. But that is not the appropriate benchmark. A district court may award fees in the rare case in which a party’s unreasonable, though not independently sanctionable, conduct is so “exceptional” as to justify an award. For litigation to fall within the second category, a district court must determine that the litigation is both objectively baseless and brought in subjective bad faith. But a case presenting either subjective bad faith or exceptionally meritless claims may sufficiently set itself apart from mine-run cases to be “exceptional.” The Federal Circuit imported this second category from Professional Real Estate Investors, Inc. v. Columbia Pictures Industries, Inc., 508 U.S. 49, but that case’s standard finds no roots in §285’s text and makes little sense in the context of the exceptional-case determination. Pp. 8–10. (2) Brooks Furniture is so demanding that it would appear to render §285 largely superfluous. Because courts already possess the inherent power to award fees in cases involving misconduct or bad faith, see Alyeska Pipeline Service Co. v. Wilderness Society, 421 U.S. 240, 258–259, this Court has declined to construe fee-shifting provisions narrowly so as to avoid rendering them superfluous. See, e.g., Christiansburg Garment Co. v. EEOC, 434 U.S. 412, 419. Pp. 10–11. (3) Brooks Furniture’s requirement that proof of entitlement to fees be made by clear and convincing evidence is not justified by §285, which imposes no specific evidentiary burden. Nor has this Court interpreted comparable fee-shifting statutes to require such a burden of proof. See, e.g., Fogerty, 510 U. S, at 519. P. 11. 496 Fed. Appx. 57, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, Breyer, Alito, and Kagan, JJ., joined, and in which Scalia, J., joined except as to footnotes 1–3.

bjective baselessness alone can create a sufficient inference of bad faith to establish exceptionality under §285, unless the circumstances as a whole show a lack of recklessness on the patentee’s part.” , at 1314. Chief Judge Rader wrote a concurring opinion that sharply criticized , 738 F. 3d, at 1318–1320; the court, he said, “should have remained true to its original reading of” §285, at 1320. 6 One e-mail, sent from ICON’s Vice President of Global Sales to two employees, read: “ ‘We are suing Octane. Not only are we coming out with a greater product to go after them, but throwing a lawsuit on top of that.’ ” 2011 WL 3900975, *4. One of the recipients then forwarded that e-mail to a third party, along with the accompanying message: “ ‘Just clearing the way and making sure you guys have all your guns loaded!’ ” . More than a year later, that same employee sent ane-mail to the Vice President of Global Sales with the subject, “ ‘I heard we are suing Octane!’ ” The executive responded as follows: “ ‘Yes—old patent we had for a long time that was sitting on the shelf. They are just looking for royalties.’ ” The District Court wrote that “in the light most favorable to Octane, these remarks are stray comments by employees with no demonstrated connection to the lawsuit.” 7 In v. , , for example, we explained that in determining whether to award fees under a similar provision in the Copyright Act, district courts could consider a “nonexclusive” list of “factors,” including “frivolousness, motivation, objective unreasonableness (both in the factual and legal components of the case) and the need in particular circumstances to advance considerations of compensation and deterrence.” , at 534, n. 19 (internal quotation marks omitted).

572.US.434

The respondent victim in this case was sexually abused as a young girl in order to produce child pornography. When she was 17, she learned that images of her abuse were being trafficked on the Internet, in effect repeating the original wrongs, for she knew that her humiliation and hurt would be renewed well into the future as thousands of additional wrongdoers witnessed those crimes. Petitioner Paroline pleaded guilty in federal court to possessing images of child pornography, which included two of the victim, in violation of 18 U. S. C. §2252. The victim then sought restitution under §2259, requesting nearly $3 million in lost income and about $500,000 in future treatment and counseling costs. The District Court declined to award restitution, concluding that the Government had not met its burden of proving what losses, if any, were proximately caused by Paroline’s offense. The victim sought a writ of mandamus, asking the Fifth Circuit to direct the District Court to order Paroline to pay restitution. Granting the writ on rehearing en banc, the Fifth Circuit held, inter alia, that §2259 did not limit restitution to losses proximately caused by the defendant, and that each defendant who possessed the victim’s images should be made liable for the victim’s entire losses from the trade in her images. Held: 1. Restitution is proper under §2259 only to the extent the defendant’s offense proximately caused a victim’s losses. This provision has a broad restitutionary purpose, stating that a district court “shall order restitution for any offense” under Chapter 110 of Title 18, such as Paroline’s possession offense; requiring district courts to order defendants “to pay the victim . . . the full amount of the victim’s losses as determined by the court,” §2259(b)(1); and expressly making “issuance of a restitution order . . . mandatory,” §2259(b)(4)(A). The Government has the “burden of demonstrating the amount of the [victim’s] loss.” §3664(e). To say one event proximately caused another means, first, that the former event caused the latter, i.e., actual cause or cause in fact; and second, that it is a proximate cause, i.e., it has a sufficient connection to the result. The concept of proximate causation is applicable in both criminal and tort law, and the analysis is parallel in many instances. Section 2259(c) defines a victim as “the individual harmed as a result of a commission of a crime under this chapter.” The words “as a result of” plainly suggest causation, and the referent of “a crime” is the offense of conviction. The “full amount of the victim’s losses,” §2259(b)(1), includes “any costs incurred by the victim” for six enumerated categories of expense, §2259(b)(3). The reference to “costs incurred by the victim” is most naturally understood as costs arising “as a result of” the offense of conviction, i.e., the defendant’s conduct. And the last of the six enumerated categories—for “other losses suffered . . . as a proximate result of the offense,” §2259(b)(3)(F)—clearly states that the causal requirement is one of proximate cause. This reading is supported by the canon of construction that, “[w]hen several words are followed by a clause which is applicable as much to the first and other words as to the last, the natural construction of the language demands that the clause be read as applicable to all.” Porto Rico Railway, Light & Power Co. v. Mor, 253 U.S. 345, 348. The reading also presents a commonsense way to impose sensible limitations on claims for attenuated costs. Pp. 5–11. 2. Applying the statute’s causation requirements in this case, victims should be compensated and defendants should be held to account for the impact of their conduct on those victims, but defendants should only be made liable for the consequences and gravity of their own conduct, not the conduct of others. Pp. 11–25. (a) A somewhat atypical causal process underlies the losses here. It may be simple to prove aggregate losses, i.e., “general losses,” stemming from the ongoing traffic in the victim’s images, but the question for §2259 purposes is how much of these general losses were the “proximate result” of an individual defendant’s offense. Here, the victim’s costs of treatment and lost income resulting from the trauma of knowing that images of her abuse are being viewed over and over are direct and foreseeable results of child-pornography crimes, provided the prerequisite of factual causation is satisfied. The primary problem, then, is the proper standard of causation in fact. Pp. 11–12. (b) A showing of but-for causation is not the proper standard here, for it is not possible to prove that the victim’s losses would be less but for one possessor’s individual role in the large, loosely connected network through which her images circulate. The victim and the Government urge the Court to read §2259 to require a less restrictive causation standard in child-pornography cases like this. They endorse the theory of “aggregate causation,” one formulation of which finds factual causation satisfied where a wrongdoer’s conduct, though alone “insufficient . . . to cause the plaintiff’s harm,” is, “when combined with conduct by other persons,” “more than sufficient to cause the harm.” 1 Restatement (Third) of Torts: Liability for Physical and Emotional Harm §27, Comment f. Tort law teaches that such alternative causal tests, though a kind of legal fiction, may be necessary to vindicate the law’s purposes, for it would be anomalous to turn away a person harmed by the combined acts of many wrongdoers simply because none of those wrongdoers alone caused the harm, and nonsensical to adopt a rule whereby individuals hurt by the combined wrongful acts of many would have no redress, while those hurt by the acts of one person alone would. These are sound principles. Taken too far, however, such alternative causal standards would treat each possessor as the cause in fact of all the trauma and attendant losses incurred as a result of all the ongoing traffic in the victim’s images. Aggregate causation logic should not be adopted in an incautious manner in the context of criminal restitution, which differs from tort law in numerous respects. Paroline’s contribution to the causal process underlying the victim’s losses was very minor, both compared to the combined acts of all other relevant offenders and compared to the contributions of other individual offenders, particularly distributors and the initial producer of the child pornography. Congress gave no indication that it intended the statute to be applied in an expansive manner so starkly contrary to the principle that restitution should reflect the consequences of the defendant’s own conduct. The victim claims that holding each possessor liable for her entire losses would be fair and practical in part because offenders can seek contribution from one another, but there is no general federal right to contribution and no specific statutory authorization for contribution here. Her severe approach could also raise questions under the Excessive Fines Clause of the Eighth Amendment. Pp. 12–19. (c) While the victim’s expansive reading must be rejected, that does not mean the broader principles underlying aggregate causation theories are irrelevant to determining the proper outcome in cases like this. The cause of the victim’s general losses is the trade in her images, and Paroline is a part of that cause. Just as it undermines the purposes of tort law to turn away plaintiffs harmed by several wrongdoers, it would undermine §2259’s purposes to turn away victims in cases like this. With respect to the statute’s remedial purpose, there is no question that it would produce anomalous results to say that no restitution is appropriate in these circumstances, for harms of the kind the victim endured here are a major reason why child pornography is outlawed. The unlawful conduct of everyone who reproduces, distributes, or possesses images of the victim’s abuse—including Paroline—plays a part in sustaining and aggravating this tragedy. And there is no doubt Congress wanted restitution for such victims. Denying restitution would also be at odds with §2259’s penological purposes, which include the need to impress upon offenders that their conduct produces concrete and devastating harms for real, identifiable victims. Thus, where it can be shown both that a defendant possessed a victim’s images and that a victim has outstanding losses caused by the continuing traffic in her images but where it is impossible to trace a particular amount of those losses to the individual defendant utilizing a more traditional causal inquiry, a court should order restitution in an amount that comports with the defendant’s relative role in the causal process underlying the victim’s general losses. District courts should use discretion and sound judgment in determining the proper amount of restitution. A variety of factors may serve as guideposts. Courts might, as a start, determine the amount of the victim’s losses caused by the continuing traffic in the victim’s images, and then base an award on factors bearing on the relative causal significance of the defendant’s conduct in producing those losses. The victim finds this approach untenable because her losses are “indivisible,” but the Court is required to define a causal standard that effects the statute’s purposes, not to apply tort-law concepts in a mechanical way in the criminal restitution context. She also argues she will be consigned to “piecemeal” restitution that may never lead to full recovery, but Congress has not promised victims full and swift restitution at the cost of holding a defendant liable for an amount drastically out of proportion to his individual causal relation to those losses. Furthermore, this approach better effects the need to impress upon defendants that their acts are not irrelevant or victimless. Pp. 19–25. (d) Though this approach is not without difficulties, courts can only do their best to apply the statute as written in a workable manner, faithful to the competing principles at stake: that victims should be compensated and that defendants should be held to account for the impact of their own conduct, not the conduct of others. District courts, which routinely exercise wide discretion both in sentencing generally and in fashioning restitution orders, should be able to apply the causal standard defined here without further detailed guidance. P. 25. 701 F.3d 749, vacated and remanded. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Alito, and Kagan, JJ., joined. Roberts, C. J., filed a dissenting opinion, in which Scalia and Thomas, JJ., joined. Sotomayor, J., filed a dissenting opinion.

This case presents the question of how to determine the amount of restitution a possessor of child pornography must pay to the victim whose childhood abuse appears in the pornographic materials possessed. The relevant statutory provisions are set forth at 18 U. S. C. §2259. Enacted as a component of the Violence Against Women Act of 1994, §2259 requires district courts to award restitution for certain federal criminal offenses, including child-pornography possession. Petitioner Doyle Randall Paroline pleaded guilty to such an offense. He admitted to possessing between 150 and 300 images of child pornography, which included two that depicted the sexual exploitation of a young girl, now a young woman, who goes by the pseudonym “Amy” for this litigation. The question is what causal relationship must be established between the defendant’s conduct and a victim’s losses for purposes of determining the right to, and the amount of, restitution under §2259. I Three decades ago, this Court observed that “the exploitive use of children in the production of pornography has become a serious national problem.” New York v. Ferber, 458 U. S. 747, 749 (1982) . The demand for child pornography harms children in part because it drives production, which involves child abuse. The harms caused by child pornography, however, are still more extensive because child pornography is “a permanent record” of the depicted child’s abuse, and “the harm to the child is exacerbated by [its] circulation.” Id., at 759. Because child pornography is now traded with ease on the Internet, “the number of still images and videos memorializing the sexual assault and other sexual exploitation of children, many very young in age, has grown exponentially.” United States Sentencing Comm’n, P. Saris et al., FederalChild Pornography Offenses 3 (2012) (hereinafter Sentencing Comm’n Report). One person whose story illustrates the devastating harm caused by child pornography is the respondent victim in this case. When she was eight and nine years old, she was sexually abused by her uncle in order to produce child pornography. Her uncle was prosecuted, required to pay about $6,000 in restitution, and sentenced to a lengthy prison term. The victim underwent an initial course of therapy beginning in 1998 and continuing into 1999. By the end of this period, her therapist’s notes reported that she was “ ‘back to normal’ ”; her involvement in dance and other age-appropriate activities, and the support of her family, justified an optimistic assessment. App. 70–71. Her functioning appeared to decline in her teenage years, however; and a major blow to her recovery came when, at the age of 17, she learned that images of her abuse were being trafficked on the Internet. Id., at 71. The digital images were available nationwide and no doubt worldwide. Though the exact scale of the trade in her images is unknown, the possessors to date easily number in the thousands. The knowledge that her images were circulated far and wide renewed the victim’s trauma and made it difficult for her to recover from her abuse. As she explained in a victim impact statement submitted to the District Court in this case: “Every day of my life I live in constant fear that someone will see my pictures and recognize me and that I will be humiliated all over again. It hurts me to know someone is looking at them—at me—when I was just a little girl being abused for the camera. I did not choose to be there, but now I am there forever in pictures that people are using to do sick things. I want it all erased. I want it all stopped. But I am powerless to stop it just like I was powerless to stop my uncle. . . . My life and my feelings are worse now because the crime has never really stopped and will never really stop. . . . It’s like I am being abused over and over and over again.” Id., at 60–61. The victim says in her statement that her fear and trauma make it difficult for her to trust others or to feel that she has control over what happens to her. Id., at 63. The full extent of this victim’s suffering is hard to grasp. Her abuser took away her childhood, her self-conception of her innocence, and her freedom from the kind of nightmares and memories that most others will never know. These crimes were compounded by the distribution of images of her abuser’s horrific acts, which meant the wrongs inflicted upon her were in effect repeated; for she knew her humiliation and hurt were and would be renewed into the future as an ever-increasing number of wrongdoers witnessed the crimes committed against her. Petitioner Paroline is one of the individuals who possessed this victim’s images. In 2009, he pleaded guilty in federal court to one count of possession of material involving the sexual exploitation of children in violation of 18 U. S. C. §2252. 672 F. Supp. 2d 781, 783 (ED Tex. 2009). Paroline admitted to knowing possession of between 150 and 300 images of child pornography, two of which depicted the respondent victim. Ibid. The victim sought restitution under §2259, asking for close to $3.4 million, consistingof nearly $3 million in lost income and about $500,000in future treatment and counseling costs. App. 52, 104. She also sought attorney’s fees and costs. 672 F. Supp. 2d, at 783. The parties submitted competing expert reports. They stipulated that the victim did not know who Paroline was and that none of her claimed losses flowed from any specific knowledge about him or his offense conduct. Id., at 792, and n. 11; App. 230. After briefing and hearings, the District Court declined to award restitution. 672 F. Supp. 2d, at 793. The District Court observed that “everyone involved with child pornography—from the abusers and producers to the end-users and possessors—contribute[s] to [the victim’s] ongoing harm.” Id., at 792. But it concluded that the Government had the burden of proving the amount of the victim’s losses “directly produced by Paroline that would not have occurred without his possession of her images.” Id., at 791. The District Court found that, under this standard, the Government had failed to meet its burden of proving what losses, if any, were proximately caused by Paroline’s offense. It thus held that “an award of restitution is not appropriate in this case.” Id., at 793. The victim sought a writ of mandamus, asking the United States Court of Appeals for the Fifth Circuit to direct the District Court to order Paroline to pay restitution in the amount requested. In re Amy, 591 F. 3d 792, 793 (2009). The Court of Appeals denied relief. Id., at 795. The victim sought rehearing. Her rehearing request was granted, as was her petition for a writ of mandamus. In re Amy Unknown, 636 F. 3d 190, 201 (2011). The Fifth Circuit reheard the case en banc along with another case, in which the defendant, Michael Wright, had raised similar issues in appealing an order of restitution under §2259, see United States v. Wright, 639 F. 3d 679, 681 (2011) (per curiam). As relevant, the Court of Appeals set out to determine the level of proof required to award restitution to victims in cases like this. It held that §2259 did not limit restitution to losses proximately caused by the defendant, and each defendant who possessed the victim’s images should be made liable for the victim’s entire losses from the trade in her images, even though other offenders played a role in causing those losses. In re Amy Unknown, 701 F. 3d 749, 772–774 (2012) (en banc). Paroline sought review here. Certiorari was granted to resolve a conflict in the Courts of Appeals over the proper causation inquiry for purposes of determining the entitlement to and amount of restitution under §2259. 570 U. S. ___ (2013). For the reasons set forth, the decision of the Court of Appeals is vacated. II Title 18 U. S. C. §2259(a) provides that a district court “shall order restitution for any offense” under Chapter 110 of Title 18, which covers a number of offenses involving the sexual exploitation of children and child pornography in particular. Paroline was convicted of knowingly possessing child pornography under §2252, a Chapter 110 offense. Section 2259 states a broad restitutionary purpose: It requires district courts to order defendants “to pay the victim . . . the full amount of the victim’s losses as determined by the court,” §2259(b)(1), and expressly states that “[t]he issuance of a restitution order under this section is mandatory,” §2259(b)(4)(A). Section 2259(b)(2) provides that “[a]n order of restitution under this section shall be issued and enforced in accordance with section 3664,” which in turn provides in relevant part that “[t]he burden of demonstrating the amount of the loss sustained by a victim as a result of the offense shall be on the attorney for the Government,” §3664(e). The threshold question the Court faces is whether §2259 limits restitution to those losses proximately caused by the defendant’s offense conduct. The Fifth Circuit held that it does not, contrary to the holdings of other Courts of Appeals to have addressed the question. Compare, e.g., 701 F. 3d, at 752 (no general proximate-cause requirement applies under §2259), with United States v. Rogers, 714 F. 3d 82, 89 (CA1 2013) (general proximate-cause requirement applies under §2259); United States v. Benoit, 713 F. 3d 1, 20 (CA10 2013) (same); United States v. Fast, 709 F. 3d 712, 721–722 (CA8 2013) (same); United States v. Laraneta, 700 F. 3d 983, 989–990 (CA7 2012) (same); United States v. Burgess, 684 F. 3d 445, 456–457 (CA4 2012) (same); United States v. Evers, 669 F. 3d 645, 659 (CA6 2012) (same); United States v. Aumais, 656 F. 3d 147, 153 (CA2 2011) (same); United States v. Kennedy, 643 F. 3d 1251, 1261 (CA9 2011) (same); United States v. Monzel, 641 F. 3d 528, 535 (CADC 2011) (same); United States v. McDaniel, 631 F. 3d 1204, 1208–1209 (CA11 2011) (same). As a general matter, to say one event proximately caused another is a way of making two separate but related assertions. First, it means the former event caused the latter. This is known as actual cause or cause in fact. The concept of actual cause “is not a metaphysical one but an ordinary, matter-of-fact inquiry into the existence . . . of a causal relation as laypeople would view it.” 4 F. Harper, F. James, & O. Gray, Torts §20.2, p. 100 (3d ed. 2007). Every event has many causes, however, see ibid., and only some of them are proximate, as the law uses that term. So to say that one event was a proximate cause of another means that it was not just any cause, but one with a sufficient connection to the result. The idea of proximate cause, as distinct from actual cause or cause in fact, defies easy summary. It is “a flexible concept,” Bridge v. Phoenix Bond & Indemnity Co., 553 U. S. 639, 654 (2008) , that generally “refers to the basic requirement that . . . there must be ‘some direct relation between the injury asserted and the injurious conduct alleged,’ ” CSX Transp., Inc. v. McBride, 564 U. S. ___, ___ (2011) (Roberts, C. J., dissenting) (slip op., at 3) (quoting Holmes v. Securities Investor Protection Corporation, 503 U. S. 258, 268 (1992) ). The concept of proximate causation is applicable in both criminal and tort law, and the analysis is parallel in many instances. 1 W. LaFave, Substantive Criminal Law §6.4(c), p. 471 (2d ed. 2003) (hereinafter LaFave). Proximate cause is often explicated in terms of foreseeability or the scope of the risk created by the predicate conduct. See, e.g., ibid.; 1 Restatement (Third) of Torts: Liability for Physical and Emotional Harm §29, p. 493 (2005) (hereinafter Restatement). A requirement of proximate cause thus serves, inter alia, to preclude liability in situations where the causal link between conduct and result is so attenuated that the consequence is more aptly described as mere fortuity. Exxon Co., U. S. A. v. Sofec, Inc., 517 U. S. 830 –839 (1996). All parties agree §2259 imposes some causation requirement. The statute defines a victim as “the individual harmed as a result of a commission of a crime under this chapter.” §2259(c). The words “as a result of” plainly suggest causation. See Pacific Operators Offshore, LLP v. Valladolid, 565 U. S. ___, ___ (2012) (slip op., at 13); see also Burrage v. United States, 571 U. S. ___, ___ (2014) (slip op., at 5). And a straightforward reading of §2259(c) indicates that the term “a crime” refers to the offense of conviction. Cf. Hughey v. United States, 495 U. S. 411, 416 (1990) . So if the defendant’s offense conduct did not cause harm to an individual, that individual is by definition not a “victim” entitled to restitution under §2259. As noted above, §2259 requires a court to order restitution for “the full amount of the victim’s losses,” §2259(b)(1), which the statute defines to include “any costs incurred by the victim” for six enumerated categories of expense, §2259(b)(3). The reference to “costs incurred by the victim” is most naturally understood as costs stemming from the source that qualifies an individual as a “victim” in the first place—namely, ones arising “as a result of” the offense. Thus, as is typically the case with criminal restitution, §2259 is intended to compensate victims for losses caused by the offense of conviction. See id., at 416. This is an important point, for it means the central concern of the causal inquiry must be the conduct of the particular defendant from whom restitution is sought. But there is a further question whether restitution under §2259 is limited to losses proximately caused by the offense. As noted, a requirement of proximate cause is more restrictive than a requirement of factual cause alone. Even if §2259 made no express reference to proximate causation, the Court might well hold that a showing of proximate cause was required. Proximate cause is a standard aspect of causation in criminal law and the law of torts. See 1 LaFave §6.4(a), at 464–466; W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts §41, p. 263 (5th ed. 1984) (hereinafter Prosser and Keeton). Given proximate cause’s traditional role in causation analysis, this Court has more than once found a proximate-cause requirement built into a statute that did not expressly impose one. See Holmes, supra, at 265–268; Associated Gen. Contractors of Cal., Inc. v. Carpenters, 459 U. S. 519 –536 (1983); see also CSX Transp., Inc., supra, at ___ (Roberts, C. J., dissenting) (slip op., at 4) (“We have applied the standard requirement of proximate cause to actions under federal statutes where the text did not expressly provide for it”); Lexmark Int’l, Inc. v. Static Control Components, Inc., ante, at 13–14. Here, however, the interpretive task is easier, for the requirement of proximate cause is in the statute’s text. The statute enumerates six categories of covered losses. §2259(b)(3). These include certain medical services, §2259(b)(3)(A); physical and occupational therapy, §2259(b)(3)(B); transportation, temporary housing, and child care, §2259(b)(3)(C); lost income, §2259(b)(3)(D); attorney’s fees and costs, §2259(b)(3)(E); and a final catchall category for “any other losses suffered by the victim as a proximate result of the offense,” §2259(b)(3)(F). The victim argues that because the “proximate result” language appears only in the final, catchall category of losses set forth at §2259(b)(3)(F), the statute has noproximate-cause requirement for losses falling within the prior enumerated categories. She justifies this reading of §2259(b) in part on the grammatical rule of the last antecedent, “according to which a limiting clause or phrase . . . should ordinarily be read as modifying only the noun or phrase that it immediately follows.” Barnhart v. Thomas, 540 U. S. 20, 26 (2003) . But that rule is “not an absolute and can assuredly be overcome by other indicia of meaning.” Ibid. The Court has not applied it in a mechanical way where it would require accepting “unlikely premises.” United States v. Hayes, 555 U. S. 415, 425 (2009) . Other canons of statutory construction, moreover, work against the reading the victim suggests. “When several words are followed by a clause which is applicable as much to the first and other words as to the last, the natural construction of the language demands that the clause be read as applicable to all.” Porto Rico Railway, Light & Power Co. v. Mor, 253 U. S. 345, 348 (1920) . Furthermore, “[i]t is . . . a familiar canon of statutory construction that [catchall] clauses are to be read as bringing within a statute categories similar in type to those specifically enumerated.” Federal Maritime Comm’n v. Seatrain Lines, Inc., 411 U. S. 726, 734 (1973) . Here, §2259(b)(3)(F) defines a broad, final category of “other losses suffered . . . as a proximate result of the offense.” That category is most naturally understood as a summary of the type of losses covered—i.e., losses suffered as a proximate result of the offense. The victim says that if Congress had wanted to limit the losses recoverable under §2259 to those proximately caused by the offense, it could have written the statute the same way it wrote §2327, which provides for restitution to victims of telemarketing fraud. Section 2327, which is written and structured much like §2259, simply defines the term “full amount of the victim’s losses” as “all losses suffered by the victim as a proximate result of the offense.” §2327(b)(3). In essence the victim argues that the first five categories of losses enumerated in §2259(b)(3) would be superfluous if all were governed by a proximate-cause requirement. That, however, is unpersuasive. The first five categories provide guidance to district courts as to the specific types of losses Congress thought would often be the proximate result of a Chapter 110 offense and could as a general matter be included in an award of restitution. Reading the statute to impose a general proximate-cause limitation accords with common sense. As noted above, proximate cause forecloses liability in situations where the causal link between conduct and result is so attenuated that the so-called consequence is more akin to mere fortuity. For example, suppose the traumatized victim of a Chapter 110 offender needed therapy and had a car accident on the way to her therapist’s office. The resulting medical costs, in a literal sense, would be a factual result of the offense. But it would be strange indeed to make a defendant pay restitution for these costs. The victim herself concedes Congress did not intend costs like these to be recoverable under §2259. Brief for Respondent Amy 45. But she claims that it is unnecessary to “read . . . into” §2259 a proximate-cause limitation in order to exclude costs of that sort. Ibid. She says the statute “contextually and inferentially require[s] a nexus for why” the losses were sustained—i.e., a sufficient connection to child pornography. Id., at 46. The victim may be right that the concept of proximate cause is not necessary to impose sensible limitations on restitution for remote consequences. But one very effective way, and perhaps the most obvious way, of excluding costs like those arising from the hypothetical car accident described above would be to incorporate a proximate-cause limitation into the statute. Congress did so, and for reasons given above the proximate-cause requirement applies to all the losses described in §2259. Restitution is therefore proper under §2259 only to the extent the defendant’s offense proximately caused a victim’s losses. III There remains the difficult question of how to apply the statute’s causation requirements in this case. The problem stems from the somewhat atypical causal process underlying the losses the victim claims here. It is perhaps simple enough for the victim to prove the aggregate losses, including the costs of psychiatric treatment and lost income, that stem from the ongoing traffic in her images as a whole. (Complications may arise in disaggregating losses sustained as a result of the initial physical abuse, but those questions may be set aside for present purposes.) These losses may be called, for convenience’s sake, a victim’s “general losses.” The difficulty is in determining the “full amount” of those general losses, if any, that are the proximate result of the offense conduct of a particular defendant who is one of thousands who have possessed and will in the future possess the victim’s images but who has no other connection to the victim. In determining the amount of general losses a defendant must pay under §2259 the ultimate question is how much of these losses were the “proximate result,” §2259(b)(3)(F), of that individual’s offense. But the most difficult aspect of this inquiry concerns the threshold requirement of causation in fact. To be sure, the requirement of proximate causation, as distinct from mere causation in fact, would prevent holding any possessor liable for losses caused in only a remote sense. But the victim’s costs of treatment and lost income resulting from the trauma of knowing that images of her abuse are being viewed over and over are direct and foreseeable results of child-pornography crimes, including possession, assuming the prerequisite of factual causation is satisfied. The primary problem, then, is the proper standard of causation in fact. A The traditional way to prove that one event was a fac-tual cause of another is to show that the latter would not have occurred “but for” the former. This approach is a familiar part of our legal tradition, see 1 LaFave §6.4(b), at 467–468; Prosser and Keeton §41, at 266, and no party disputes that a showing of but-for causation would satisfy §2259’s factual-causation requirement. Sometimes that showing could be made with little difficulty. For example, but-for causation could be shown with ease in many cases involving producers of child pornography, see §2251(a); parents who permit their children to be used for child-pornography production, see §2251(b); individuals who sell children for such purposes, see §2251A; or the initial distributor of the pornographic images of a child, see §2252. In this case, however, a showing of but-for causation cannot be made. The District Court found that the Government failed to prove specific losses caused by Paroline in a but-for sense and recognized that it would be “incredibly difficult” to do so in a case like this. 672 F. Supp. 2d, at 791–793. That finding has a solid foundation in the record, and it is all but unchallenged in this Court. See Brief for Respondent Amy 63; Brief for United States 19, 25. But see Supp. Brief for United States 8–10. From the victim’s perspective, Paroline was just one of thousands of anonymous possessors. To be sure, the victim’s precise degree of trauma likely bears a relation to the total number of offenders; it would probably be less if only 10 rather than thousands had seen her images. But it is not possible to prove that her losses would be less (and by how much) but for one possessor’s individual role in the large, loosely connected network through which her images circulate. See Sentencing Comm’n Report, at ii, xx. Even without Paroline’s offense, thousands would have viewed and would in the future view the victim’s images, so it cannot be shown that her trauma and attendant losses would have been any different but for Paroline’s offense. That is especially so given the parties’ stipulation that the victim had no knowledge of Paroline. See supra, at 4. Recognizing that losses cannot be substantiated under a but-for approach where the defendant is an anonymous possessor of images in wide circulation on the Internet, the victim and the Government urge the Court to read §2259 to require a less restrictive causation standard, at least in this and similar child-pornography cases. They are correct to note that courts have departed from the but-for standard where circumstances warrant, especially where the combined conduct of multiple wrongdoers produces a bad outcome. See Burrage, 571 U. S., at ___ (slip op., at 10) (acknowledging “the undoubted reality that courts have not always required strict but-for causality, even where criminal liability is at issue”). The victim and the Government look to the literature on criminal and tort law for alternatives to the but-for test. The Court has noted that the “most common” exception to the but-for causation requirement is applied where “multiple sufficient causes independently . . . produce a result,” ibid.; see also 1 LaFave §6.4(b), at 467–469; 1 Restatement §27, at 376. This exception is an ill fit here, as all parties seem to recognize. Paroline’s possession of two images of the victim was surely not sufficient to cause her entire losses from the ongoing trade in her images. Nor is there a practical way to isolate some subset of the victim’s general losses that Paroline’s conduct alone would have been sufficient to cause. See Brief for United States 26, n. 11. Understandably, the victim and the Government thus concentrate on a handful of less demanding causation tests endorsed by authorities on tort law. One prominent treatise suggests that “[w]hen the conduct of two or more actors is so related to an event that their combined conduct, viewed as a whole, is a but-for cause of the event, and application of the but-for rule to them individually would absolve all of them, the conduct of each is a cause in fact of the event.” Prosser and Keeton §41, at 268. The Restatement adopts a similar exception for “[m]ultiple sufficient causal sets.” 1 Restatement §27, Comment f, at 380–381. This is where a wrongdoer’s conduct, though alone “insufficient . . . to cause the plaintiff’s harm,” is, “when combined with conduct by other persons,” “more than sufficient to cause the harm.” Ibid. The Restatement offers as an example a case in which three people independently but simultaneously lean on a car, creating enough combined force to roll it off a cliff. Ibid. Even if each exerted too little force to move the car, and the force exerted by any two was sufficient to the move the car, each individual is a factual cause of the car’s destruction. Ibid. The Government argues that these authorities “provide ample support for an ‘aggregate’ causation theory,” Brief for United States 18, and that such a theory would best effectuate congressional intent in cases like this, id., at 18–19. The victim says much the same. Brief for Respondent Amy 42–43. These alternative causal tests are a kind of legal fiction or construct. If the conduct of a wrongdoer is neither necessary nor sufficient to produce an outcome, that conduct cannot in a strict sense be said to have caused the outcome. Nonetheless, tort law teaches that alternative and less demanding causal standards are necessary in certain circumstances to vindicate the law’s purposes. It would be anomalous to turn away a person harmed by the combined acts of many wrongdoers simply because none of those wrongdoers alone caused the harm. And it would be nonsensical to adopt a rule whereby individuals hurt by the combined wrongful acts of many (and thus in many instances hurt more badly than otherwise) would haveno redress, whereas individuals hurt by the acts of one person alone would have a remedy. Those are the prin-ciples that underlie the various aggregate causation tests the victim and the Government cite, and they are sound principles. These alternative causal standards, though salutary when applied in a judicious manner, also can be taken too far. That is illustrated by the victim’s suggested approach to applying §2259 in cases like this. The victim says that under the strict logic of these alternative causal tests, each possessor of her images is a part of a causal set sufficient to produce her ongoing trauma, so each possessor should be treated as a cause in fact of all the trauma and all the attendant losses incurred as a result of the entire ongoing traffic in her images. Id., at 43. And she argues that if this premise is accepted the further requirement of proximate causation poses no barrier, for she seeks restitution only for those losses that are the direct and foreseeable result of child-pornography offenses. Because the statute requires restitution for the “full amount of the victim’s losses,” including “any . . . losses suffered by the victim as a proximate result of the offense,” §2259(b), she argues that restitution is required for the entire aggregately caused amount. The striking outcome of this reasoning—that each possessor of the victim’s images would bear the consequences of the acts of the many thousands who possessed those images—illustrates why the Court has been reluctant to adopt aggregate causation logic in an incautious manner, especially in interpreting criminal statutes where there is no language expressly suggesting Congress intended that approach. See Burrage, 571 U. S., at ___ (slip op., at 11–12). Even if one were to refer just to the law of torts, it would be a major step to say there is a sufficient causal link between the injury and the wrong so that all the victim’s general losses were “suffered . . . as a proximate result of [Paroline’s] offense,” §2259(b)(3)(F). And there is special reason not to do so in the context of criminal restitution. Aside from the manifest procedural differences between criminal sentencing and civil tort lawsuits, restitution serves purposes that differ from (though they overlap with) the purposes of tort law. See, e.g., Kelly v. Robinson, 479 U. S. 36 , n. 10 (1986) (noting that restitution is, inter alia, “an effective rehabilitative penalty”). Legal fictions developed in the law of torts cannot be imported into criminal restitution and applied to their utmost limits without due consideration of these differences. Contrary to the victim’s suggestion, this is not akin to a case in which a “gang of ruffians” collectively beats a person, or in which a woman is “gang raped by five men on one night or by five men on five sequential nights.” Brief for Respondent Amy 55. First, this case does not involve a set of wrongdoers acting in concert, see Prosser and Keeton §52, at 346 (discussing full liability for a joint enter-prise); for Paroline had no contact with the overwhelming majority of the offenders for whose actions the victim would hold him accountable. Second, adopting the victim’s approach would make an individual possessor liable for the combined consequences of the acts of not just 2, 5, or even 100 independently acting offenders; but instead, a number that may reach into the tens of thousands. See Brief for Respondent Amy 65. It is unclear whether it could ever be sensible to embrace the fiction that this victim’s entire losses were the “proximate result,” §2259(b)(3)(F), of a single possessor’s offense. Paroline’s contribution to the causal process underlying the victim’s losses was very minor, both compared to the combined acts of all other relevant offenders, and in comparison to the contributions of other individual offenders, particularly distributors (who may have caused hundreds or thousands of further viewings) and the initial producer of the child pornography. See 1 Restatement §36, and Comment a, at 597–598 (recognizing a rule excluding from liability individuals whose contribution to a causal set that factually caused the outcome “pales by comparison to the other contributions to that causal set”). But see id., §27, Reporters’ Note, Comment i, at 395 (“The conclusion that none of” two dozen small contributions to a sufficient causal set was a cause of the outcome “is obviously untenable”). Congress gave no indication that it intended its statute to be applied in the expansive manner the victim suggests, a manner contrary to the bedrock principle that restitution should reflect the consequences of the defendant’s own conduct, see Hughey, 495 U. S., at 416, not the conduct of thousands of geographically and temporally distant offenders acting independently, and with whom the defendant had no contact. The victim argues that holding each possessor liable for her entire losses would be fair and practical, in part because offenders may seek contribution from one another. Brief for Respondent Amy 58. If that were so, it might mitigate to some degree the concerns her approach presents. But there is scant authority for her contention that offenders convicted in different proceedings in different jurisdictions and ordered to pay restitution to the same victim may seek contribution from one another. There is no general federal right to contribution. Northwest Airlines, Inc. v. Transport Workers, 451 U. S. 77 –97 (1981). Nor does the victim point to any clear statutory basis for a right to contribution in these circumstances. She thus suggests that this Court should imply a cause of action. Brief for Respondent Amy 58. But that is a rare step in any circumstance. See, e.g., Stoneridge Investment Partners, LLC v. Scientific-Atlanta, Inc., 552 U. S. 148 –165 (2008); Musick, Peeler & Garrett v. Employers Ins. of Wausau, 508 U. S. 286, 291 (1993) (noting that this Court’s precedents “teach that the creation of new rights ought to be left to legislatures, not courts”). And it would do little to address the practical problems offenders would face in seeking contribution in any event, see Brief for United States 45–46, problems with which the victim fails to grapple. The reality is that the victim’s suggested approach would amount to holding each possessor of her images liable for the conduct of thousands of other independently acting possessors and distributors, with no legal or practical avenue for seeking contribution. That approach is so severe it might raise questions under the Excessive Fines Clause of the Eighth Amendment. To be sure, this Court has said that “the Excessive Fines Clause was intended to limit only those fines directly imposed by, and payable to, the government.” Browning-Ferris Industries of Vt., Inc. v. Kelco Disposal, Inc., 492 U. S. 257, 268 (1989) . But while restitution under §2259 is paid to a victim, it is imposed by the Government “at the culmination of a criminal proceeding and requires conviction of an underlying” crime, United States v. Bajakajian, 524 U. S. 321, 328 (1998) . Thus, despite the differences between restitution and a traditional fine, restitution still implicates “the prosecutorial powers of government,” Browning-Ferris, supra, at 275. The primary goal of restitution is remedial or compensatory, cf. Bajakajian, supra, at 329, but it also serves punitive purposes, see Pasquantino v. United States, 544 U. S. 349, 365 (2005) (“The purpose of awarding restitution” under 18 U. S. C. §3663A “is . . . to mete out appropriate criminal punishment”); Kelly, 479 U. S., at 49, n. 10. That may be “sufficient to bring [it] within the purview of the Excessive Fines Clause,” Bajakajian, supra, at 329, n. 4. And there is a real question whether holding a single possessor liable for millions of dollars in losses collectively caused by thousands of independent actors might be excessive and disproportionate in these circumstances. These concerns offer further reason not to interpret the statute the way the victim suggests. B The contention that the victim’s entire losses from the ongoing trade in her images were “suffered . . . as a proximate result” of Paroline’s offense for purposes of §2259 must be rejected. But that does not mean the broader principles underlying the aggregate causation theories the Government and the victim cite are irrelevant to determining the proper outcome in cases like this. The cause of the victim’s general losses is the trade in her images. And Paroline is a part of that cause, for he is one of those who viewed her images. While it is not possible to identify a discrete, readily definable incremental loss he caused, it is indisputable that he was a part of the overall phenomenon that caused her general losses. Just as it undermines the purposes of tort law to turn away plaintiffs harmed by several wrongdoers, it would undermine the remedial and penological purposes of §2259 to turn away victims in cases like this. With respect to the statute’s remedial purpose, there can be no question that it would produce anomalous results to say that no restitution is appropriate in these circumstances. It is common ground that the victim suffers continuing and grievous harm as a result of her knowledge that a large, indeterminate number of individuals have viewed and will in the future view images of the sexual abuse she endured. Brief for Petitioner 50; Brief for Respondent Wright 4; Brief for United States 23; Brief for Respondent Amy 60. Harms of this sort are a major reason why child pornography is outlawed. See Ferber, 458 U. S., at 759. The unlawful conduct of everyone who reproduces, distributes, or possesses the images of the victim’s abuse—including Paroline—plays a part in sustaining and aggravating this tragedy. And there can be no doubt Congress wanted victims to receive restitution for harms like this. The law makes restitution “mandatory,” §2259(b)(4), for child-pornography offenses under Chapter 110, language that indicates Congress’ clear intent that victims of child pornography be compensated by the perpetrators who contributed to their anguish. It would undermine this intent to apply the statute in a way that would render it a dead letter in child-pornography prosecutions of this type. Denying restitution in cases like this would also be at odds with the penological purposes of §2259’s mandatory restitution scheme. In a sense, every viewing of child pornography is a repetition of the victim’s abuse. One reason to make restitution mandatory for crimes like this is to impress upon offenders that their conduct produces concrete and devastating harms for real, identifiable victims. See Kelly, supra, at 49, n. 10 (“Restitution is an effective rehabilitative penalty because it forces the defendant to confront, in concrete terms, the harm his actions have caused”). It would be inconsistent with this purpose to apply the statute in a way that leaves offenders with the mistaken impression that child-pornography possession (at least where the images are in wide circulation) is a victimless crime. If the statute by its terms required a showing of strict but-for causation, these purposes would be beside the point. But the text of the statute is not so limited. Although Congress limited restitution to losses that are the “proximate result” of the defendant’s offense, such unelaborated causal language by no means requires but-for causation by its terms. See Burrage, 571 U. S., at ___ (slip op., at 8) (courts need not read phrases like “results from” to require but-for causality where there is “textual or contextual” reason to conclude otherwise). As the authorities the Government and the victim cite show, the availability of alternative causal standards where circumstances warrant is, no less than the but-for test itself as a default, part of the background legal tradition against which Congress has legislated, cf. id., at ___ (slip op., at 10). It would be unacceptable to adopt a causal standard so strict that it would undermine congressional intent where neither the plain text of the statute nor legal tradition demands such an approach. In this special context, where it can be shown both that a defendant possessed a victim’s images and that a victim has outstanding losses caused by the continuing traffic in those images but where it is impossible to trace a particular amount of those losses to the individual defendant by recourse to a more traditional causal inquiry, a court applying §2259 should order restitution in an amount that comports with the defendant’s relative role in the causal process that underlies the victim’s general losses. The amount would not be severe in a case like this, given the nature of the causal connection between the conduct of a possessor like Paroline and the entirety of the victim’s general losses from the trade in her images, which are the product of the acts of thousands of offenders. It would not, however, be a token or nominal amount. The required restitution would be a reasonable and circumscribed award imposed in recognition of the indisputable role of the offender in the causal process underlying the victim’s losses and suited to the relative size of that causal role. This would serve the twin goals of helping the victim achieve eventual restitution for all her child-pornography losses and impressing upon offenders the fact that child-pornography crimes, even simple possession, affect real victims. There remains the question of how district courts should go about determining the proper amount of restitution. At a general level of abstraction, a court must assess as best it can from available evidence the significance of the individual defendant’s conduct in light of the broader causal process that produced the victim’s losses. This cannot be a precise mathematical inquiry and involves the use of discretion and sound judgment. But that is neither unusual nor novel, either in the wider context of criminal sentencing or in the more specific domain of restitution. It is well recognized that district courts by necessity “exercise . . . discretion in fashioning a restitution order.” §3664(a). Indeed, a district court is expressly authorized to conduct a similar inquiry where multiple defendants who have “contributed to the loss of a victim” appear before it. §3664(h). In that case it may “apportion liability among the defendants to reflect the level of contribution to the victim’s loss . . . of each defendant.” Ibid. Assessing an individual defendant’s role in the causal process behind a child-pornography victim’s losses does not involve a substantially different or greater exercise of discretion. There are a variety of factors district courts might consider in determining a proper amount of restitution, and it is neither necessary nor appropriate to prescribe a precise algorithm for determining the proper restitution amount at this point in the law’s development. Doing so would unduly constrain the decisionmakers closest to the facts of any given case. But district courts might, as a starting point, determine the amount of the victim’s losses caused by the continuing traffic in the victim’s images (excluding, of course, any remote losses like the hypothetical car accident described above, see supra, at 10), then set an award of restitution in consideration of factors that bear on the relative causal significance of the defendant’s conduct in producing those losses. These could include the number of past criminal defendants found to have contributed to the victim’s general losses; reasonable predictions of the number of future offenders likely to be caught and convicted for crimes contributing to the victim’s general losses; any available and reasonably reliable estimate of the broader number of offenders involved (most of whom will, of course, never be caught or convicted); whether the defendant reproduced or distributed images of the victim; whether the defendant had any connection to the initial production of the images; how many images of the victim the defendant possessed; and other facts relevant to the defendant’s relative causal role. See Brief for United States 49. These factors need not be converted into a rigid formula, especially if doing so would result in trivial restitution orders. They should rather serve as rough guideposts for determining an amount that fits the offense. The resulting amount fixed by the court would be deemed the amount of the victim’s general losses that were the “proximate result of the offense” for purposes of §2259, and thus the “full amount” of such losses that should be awarded. The court could then set an appropriate payment schedule in consideration of the defendant’s financial means. See §3664(f)(2). The victim says this approach is untenable because her losses are “indivisible” in the sense that term is used by tort law, i.e., that there is no “reasonable basis for the factfinder to determine . . . the amount of damages separately caused by” any one offender’s conduct. Restatement (Third) of Torts: Apportionment of Liability §26, p. 320 (1999). The premise of her argument is that because it is in a sense a fiction to say Paroline caused $1,000 in losses, $10,000 in losses, or any other lesser amount, it is necessary to embrace the much greater fiction that Paroline caused all the victim’s losses from the ongoing trade in her images. But that is a non sequitur. The Court is required to define a causal standard that effects the statute’s purposes, not to apply tort-law concepts in a mechanical way in the criminal restitution context. Even if the victim’s losses are fully “indivisible” in this sense (which is debatable), treating Paroline as a proximate cause of all the victim’s losses—especially in the absence of a workable system of contribution—stretches the fiction of aggregate causation to its breaking point. Treating him as a cause of a smaller amount of the victim’s general losses, taking account of his role in the overall causal process behind those losses, effects the statute’s purposes; avoids the nonsensical result of turning away victims emptyhanded; and does so without sacrificing the need for proportional-ity in sentencing. The victim also argues that this approach would consign her to “piecemeal” restitution and leave her to face “decades of litigation that might never lead to full recovery,” Brief for Respondent Amy 57, which “would convert Congress’s promise to child pornography victims into an empty gesture,” id., at 66. But Congress has not promisedvictims full and swift restitution at all costs. To be sure, the statute states a strong restitutionary purpose; but that purpose cannot be twisted into a license to hold a defendant liable for an amount drastically out of proportion to his own individual causal relation to the victim’s losses. Furthermore, an approach of this sort better effects the need to impress upon defendants that their acts are not irrelevant or victimless. As the Government observes, Reply Brief for United States 18, it would undermine this important purpose of criminal restitution if the victim simply collected her full losses from a handful of wealthy possessors and left the remainder to pay nothing because she had already fully collected. Of course the victim should someday collect restitution for all her child-pornography losses, but it makes sense to spread payment among a larger number of offenders in amounts more closely in proportion to their respective causal roles and their own circumstances so that more are made aware, through the concrete mechanism of restitution, of the impact of child-pornography possession on victims. C This approach is not without its difficulties. Restitution orders should represent “an application of law,” not “a decisionmaker’s caprice,” Philip Morris USA v. Williams, 549 U. S. 346, 352 (2007) (internal quotation marks omitted), and the approach articulated above involves discretion and estimation. But courts can only do their best to apply the statute as written in a workable manner, faithful to the competing principles at stake: that victims should be compensated and that defendants should be held to account for the impact of their conduct on those victims, but also that defendants should be made liable for the consequences and gravity of their own conduct, not the conduct of others. District courts routinely exercise wide discretion both in sentencing as a general matter and more specifically in fashioning restitution orders. There is no reason to believe they cannot apply the causal standard defined above in a reasonable manner without further detailed guidance at this stage in the law’s elaboration. Based on its experience in prior cases of this kind, the Government—which, as noted above, see supra, at 5–6, bears the burden of proving the amount of the victim’s losses, §3664(e)—could also inform district courts of restitution sought and ordered in other cases. * * * The Fifth Circuit’s interpretation of the requirements of §2259 was incorrect. The District Court likewise erred in requiring a strict showing of but-for causation. The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

572.US.663

The Copyright Act (Act) protects copyrighted works published before 1978 for an initial period of 28 years, renewable for a period of up to 67 years. 17 U. S. C. §304(a). The author’s heirs inherit the renewal rights. See §304(a)(1)(C)(ii)–(iv). When an author who has assigned her rights away “dies before the renewal period, . . . the assignee may continue to use the original work only if the author’s successor transfers the renewal rights to the assignee,” Stewart v. Abend, 495 U.S. 207, 221. The Act provides both equitable and legal remedies for infringement: an injunction “on such terms as [a court] may deem reasonable to prevent or restrain infringement of a copyright,” §502(a); and, at the copyright owner’s election, either (1) the “owner’s actual damages and any additional profits of the infringer,” §504(a)(1), which petitioner seeks in this case, or (2) specified statutory damages, §504(c). The Act’s statute of limitations provides: “No civil action shall be maintained under the [Act] unless it is commenced within three years after the claim accrued.” §507(b). A claim ordinarily accrues when an infringing act occurs. Under the separate-accrual rule that attends the copyright statute of limitations, when a defendant has committed successive violations, each infringing act starts a new limitations period. However, under §507(b), each infringement is actionable only within three years of its occurrence. Here, the allegedly infringing work is the motion picture Raging Bull, based on the life of boxing champion Jake LaMotta, who, with Frank Petrella, told his story in, inter alia, a screenplay copyrighted in 1963. In 1976, the pair assigned their rights and renewal rights, which were later acquired by respondent United Artists Corporation, a subsidiary of respondent Metro-Goldwyn-Mayer, Inc. (collectively, MGM). In 1980, MGM released, and registered a copyright in, the film Raging Bull, and it continues to market the film today. Frank Petrella died during the initial copyright term, so renewal rights reverted to his heirs. Plaintiff below, petitioner here, Paula Petrella (Petrella), his daughter, renewed the 1963 copyright in 1991, becoming its sole owner. Seven years later, she advised MGM that its exploitation of Raging Bull violated her copyright and threatened suit. Some nine years later, on January 6, 2009, she filed an infringement suit, seeking monetary and injunctive relief limited to acts of infringement occurring on or after January 6, 2006. Invoking the equitable doctrine of laches, MGM moved for summary judgment. Petrella’s 18-year delay in filing suit, MGM argued, was unreasonable and prejudicial to MGM. The District Court granted MGM’s motion, holding that laches barred Petrella’s complaint. The Ninth Circuit affirmed. Held: 1. Laches cannot be invoked as a bar to Petrella’s pursuit of a claim for damages brought within §507(b)’s three-year window. Pp. 11–19. (a) By permitting a successful plaintiff to gain retrospective relief only three years back from the time of suit, the copyright statute of limitations itself takes account of delay. Brought to bear here, §507(b) directs that Petrella cannot reach MGM’s returns on its investment in Raging Bull in years before 2006. Moreover, if infringement within the three-year window is shown, a defendant may offset against profits made in that period expenses incurred in generating those profits. See §504(b). In addition, a defendant may retain the return on investment shown to be attributable to its own enterprise, as distinct from the value created by the infringed work. See ibid. Both before and after the merger of law and equity in 1938, this Court has cautioned against invoking laches to bar legal relief. See, e.g., Holmberg v. Armbrecht, 327 U.S. 392, 395, 396. Pp. 11–14. (b) MGM’s principal arguments regarding the contemporary scope of the laches defense are unavailing. Pp. 14–19. (1) MGM urges that, because laches is listed in Federal Rule of Civil Procedure 8(c) as an affirmative defense discrete from a statute of limitations defense, the plea should be “available . . . in every civil action” to bar all forms of relief. Such an expansive role careens away from understandings, past and present, of the essentially gap-filling, not legislation-overriding, office of laches. This Court has never applied laches to bar in their entirety claims for discrete wrongs occurring within a federally prescribed limitations period. Inviting individual judges to set a time limit other than the one Congress prescribed would tug against the uniformity Congress sought to achieve in enacting §507(b). Pp. 14–15. (2) MGM contends that laches, like equitable tolling, should be “read into every federal statute of limitation,” Holmberg, 327 U. S., at 397. However, tolling lengthens the time for commencing a civil action where there is a statute of limitations and is, in effect, a rule of interpretation tied to that statutory limit. See, e.g., Young v. United States, 535 U.S. 43, 49–50. In contrast, laches, which originally served as a guide when no statute of limitations controlled, can scarcely be described as a rule for interpreting a statutory prescription. Pp. 15–16. (3) MGM insists that the laches defense must be available to prevent a copyright owner from sitting still, doing nothing, waiting to see what the outcome of an alleged infringer’s investment will be. It is hardly incumbent on copyright owners, however, to challenge each and every actionable infringement. And there is nothing untoward about waiting to see whether an infringer’s exploitation undercuts the value of the copyrighted work, has no effect on that work, or even complements it. Section 507(b)’s limitations period, coupled to the separate-accrual rule, allows a copyright owner to defer suit until she can estimate whether litigation is worth the candle. Pp. 16–17. (4) MGM is concerned that evidence needed or useful to defend against liability will be lost during a copyright owner’s inaction. But Congress must have been aware that the passage of time and the author’s death could cause evidentiary issues when it provided for reversionary renewal rights that an author’s heirs can exercise long after a work was written and copyrighted. Moreover, because a copyright plaintiff bears the burden of proving infringement, any hindrance caused by evidence unavailability is as likely to affect plaintiffs as defendants. The need for extrinsic evidence is also reduced by the registration mechanism, under which both the certificate and the original work must be on file with the Copyright Office before a copyright owner can sue for infringement. Pp. 17–18. (5) Finally, when a copyright owner engages in intentionally misleading representations concerning his abstention from suit, and the alleged infringer detrimentally relies on such deception, the doctrine of estoppel may bar the copyright owner’s claims completely, eliminating all potential remedies. The gravamen of estoppel, a defense long recognized as available in actions at law, is wrongdoing, overt misleading, and consequent loss. Estoppel does not undermine the statute of limitations, for it rests on misleading, whether engaged in early on, or later in time. P. 19. 2. While laches cannot be invoked to preclude adjudication of a claim for damages brought within the Act’s three-year window, in extraordinary circumstances, laches may, at the very outset of the litigation, curtail the relief equitably awarded. For example, where owners of a copyrighted architectural design, although aware of an allegedly infringing housing project, delayed suit until the project was substantially constructed and partially occupied, an order mandating destruction of the project would not be tolerable. See Chirco v. Crosswinds Communities, Inc., 474 F.3d 227, 236. Nor, in the face of an unexplained delay in commencing suit, would it be equitable to order “total destruction” of a book already printed, packed, and shipped. See New Era Publications Int’l v. Henry Holt & Co., 873 F.2d 576, 584–585. No such extraordinary circumstance is present here. Petrella notified MGM of her copyright claims before MGM invested millions of dollars in creating a new edition of Raging Bull, and the equitable relief she seeks—e.g., disgorgement of unjust gains and an injunction against future infringement—would not result in anything like “total destruction” of the film. Allowing Petrella’s suit to go forward will put at risk only a fraction of the income MGM has earned during the more than three decades Raging Bull has been marketed and will work no unjust hardship on innocent third parties. Should Petrella ultimately prevail on the merits, the District Court, in determining appropriate injunctive relief and assessing profits, may take account of Petrella’s delay in commencing suit. In doing so, however, the court must closely examine MGM’s alleged reliance on Petrella’s delay, taking account of MGM’s early knowledge of her claims, the protection MGM might have achieved through a declaratory judgment action, the extent to which MGM’s investment was protected by the separate-accrual rule, the court’s authority to order injunctive relief “on such terms as it may deem reasonable,” §502(a), and any other relevant considerations. Pp. 19–22. 695 F.3d 946, reversed and remanded. Ginsburg, J., delivered the opinion of the Court, in which Scalia, Thomas, Alito, Sotomayor, and Kagan, JJ., joined. Breyer, J., filed a dissenting opinion, in which Roberts, C. J., and Kennedy, J., joined.

The Copyright Act provides that “[n]o civil action shall be maintained under the [Act] unless it is commenced within three years after the claim accrued.” 17 U. S. C. §507(b). This case presents the question whether the equitable defense of laches (unreasonable, prejudicial delay in commencing suit) may bar relief on a copyright infringement claim brought within §507(b)’s three-year limitations period. Section 507(b), it is undisputed, bars relief of any kind for conduct occurring prior to the three-year limitations period. To the extent that an infringement suit seeks relief solely for conduct occurring within the limitations period, however, courts are not at liberty to jettison Congress’ judgment on the timeliness of suit. Laches, we hold, cannot be invoked to preclude adjudication of a claim for damages brought within the three-year window. As to equitable relief, in extraordinary circumstances, laches may bar at the very threshold the particular relief requested by the plaintiff. And a plaintiff’s delay can always be brought to bear at the remedial stage, in determining appropriate injunctive relief, and in assessing the “profits of the infringer . . . attributable to the infringement.” §504(b).[1] Petitioner Paula Petrella, in her suit for copyright infringement, sought no relief for conduct occurring outside §507(b)’s three-year limitations period. Nevertheless, the courts below held that laches barred her suit in its en-tirety, without regard to the currency of the conduct of which Petrella complains. That position, we hold, is contrary to §507(b) and this Court’s precedent on the province of laches. I The Copyright Act (Act), 17 U. S. C. §101 et seq., grants copyright protection to original works of authorship. §102(a). Four aspects of copyright law bear explanation at the outset. First, the length of a copyright term. Under the Act, a copyright “vests initially in the author or authors of the work,” who may transfer ownership to a third party. §201. The Act confers on a copyright owner certain exclusive rights, including the rights to reproduce and distribute the work and to develop and market derivative works. §106. Copyrighted works published before 1978—as was the work at issue—are protected for an initial period of 28 years, which may be—and in this case was—extended for a renewal period of up to 67 years. §304(a). From and after January 1, 1978, works are generally protected from the date of creation until 70 years after the author’s death. §302(a). Second, copyright inheritance. For works copyrighted under the pre-1978 regime in which an initial period of protection may be followed by a renewal period, Congress provided that the author’s heirs inherit the renewal rights. See §304(a)(1)(C)(ii)–(iv). We held in Stewart v. Abend, 495 U. S. 207 (1990) , that if an author who has assigned her rights away “dies before the renewal period, then the assignee may continue to use the original work [to produce a derivative work] only if the author’s successor transfers the renewal rights to the assignee.” Id., at 221.[2] Third, remedies. The Act provides a variety of civil remedies for infringement, both equitable and legal. See §§502–505, described supra, at 2, n. 1. A court may issue an injunction “on such terms as it may deem reasonable to prevent or restrain infringement of a copyright.” §502(a). At the election of the copyright owner, a court may also award either (1) “the copyright owner’s actual damages and any additional profits of the infringer,” §504(a)(1), which petitioner seeks in the instant case, or (2) statutory damages within a defined range, §504(c). Fourth, and most significant here, the statute of limitations. Until 1957, federal copyright law did not include a statute of limitations for civil suits. Federal courts therefore used analogous state statutes of limitations to determine the timeliness of infringement claims. See S. Rep. No. 1014, 85th Cong., 1st Sess., 2 (1957) (hereinafter Senate Report). And they sometimes invoked laches to abridge the state-law prescription. As explained in Teamsters & Employers Welfare Trust of Ill. v. Gorman Bros. Ready Mix, 283 F. 3d 877, 881 (CA7 2002): “When Congress fails to enact a statute of limitations, a [federal] court that borrows a state statute of limitations but permits it to be abridged by the doctrine of laches is not invading congressional prerogatives. It is merely filling a legislative hole.” (internal citation omitted). In 1957, Congress addressed the matter and filled the hole; it prescribed a three-year look-back limitations period for all civil claims arising under the Copyright Act. See Act of Sept. 7, 1957, Pub. L. 85–313, 71Stat. 633, 17 U. S. C. §115(b) (1958 ed.). The provision, as already noted, reads: “No civil action shall be maintained under the provisions of this title unless it is commenced within three years after the claim accrued.” §507(b).[3] The federal limitations prescription governing copyright suits serves two purposes: (1) to render uniform and certain the time within which copyright claims could be pursued; and (2) to prevent the forum shopping invited by disparate state limitations periods, which ranged from one to eight years. Senate Report 2; see H. R. Rep. No. 2419, 84th Cong., 2d Sess., 2 (1956). To comprehend how the Copyright Act’s limitations period works, one must understand when a copyright infringement claim accrues. A claim ordinarily accrues “when [a] plaintiff has a complete and present cause of action.” Bay Area Laundry and Dry Cleaning Pension Trust Fund v. Ferbar Corp. of Cal., 522 U. S. 192, 201 (1997) (internal quotation marks omitted). In other words, the limitations period generally begins to run at the point when “the plaintiff can file suit and obtain relief.” Ibid. A copyright claim thus arises or “accrue[s]” when an infringing act occurs.[4] It is widely recognized that the separate-accrual rule attends the copyright statute of limitations.[5] Under that rule, when a defendant commits successive violations, the statute of limitations runs separately from each violation. Each time an infringing work is reproduced or distributed, the infringer commits a new wrong. Each wrong gives rise to a discrete “claim” that “accrue[s]” at the time the wrong occurs.[6] In short, each infringing act starts a new limitations period. See Stone v. Williams, 970 F. 2d 1043, 1049 (CA2 1992) (“Each act of infringement is a distinct harm giving rise to an independent claim for relief.”). Under the Act’s three-year provision, an infringement is actionable within three years, and only three years, of its occurrence. And the infringer is insulated from liability for earlier infringements of the same work. See 3 M. Nimmer & D. Nimmer, Copyright §12.05[B][1][b], p. 12–150.4 (2013) (“If infringement occurred within three years prior to filing, the action will not be barred even if prior infringements by the same party as to the same work are barred because they occurred more than three years previously.”). Thus, when a defendant has engaged (or is alleged to have engaged) in a series of discrete infringing acts, the copyright holder’s suit ordinarily will be timely under §507(b) with respect to more recent acts of infringement (i.e., acts within the three-year window), but untimely with respect to prior acts of the same or similar kind.[7] In sum, Congress provided two controlling time prescriptions: the copyright term, which endures for decades, and may pass from one generation to another; and §507(b)’s limitations period, which allows plaintiffs during that lengthy term to gain retrospective relief running only three years back from the date the complaint was filed. II A The allegedly infringing work in this case is the criti-cally acclaimed motion picture Raging Bull, based on the life of boxing champion Jake LaMotta. After retiring from the ring, LaMotta worked with his longtime friend, Frank Petrella, to tell the story of the boxer’s career. Their venture resulted in three copyrighted works: two screenplays, one registered in 1963, the other in 1973, and a book, registered in 1970. This case centers on the screenplay registered in 1963. The registration identified Frank Petrella as sole author, but also stated that the screenplay was written “in collaboration with” LaMotta. App. 164. In 1976, Frank Petrella and LaMotta assigned their rights in the three works, including renewal rights, to Chartoff-Winkler Productions, Inc. Two years later, respondent United Artists Corporation, a subsidiary of respond-ent Metro-Goldwyn-Mayer, Inc. (collectively, MGM), acquired the motion picture rights to the book and both screenplays, rights stated by the parties to be “exclusiv[e] and forever, including all periods of copyright and renewals and extensions thereof.” Id., at 49. In 1980, MGM released, and registered a copyright in, the film Raging Bull, directed by Martin Scorcese and starring Robert De Niro, who won a Best Actor Academy Award for his portrayal of LaMotta. MGM continues to market the film, and has converted it into formats unimagined in 1980, including DVD and Blu-ray. Frank Petrella died in 1981, during the initial terms of the copyrights in the screenplays and book. As this Court’s decision in Stewart confirmed, Frank Petrella’s renewal rights reverted to his heirs, who could renew the copyrights unburdened by any assignment previously made by the author. See 495 U. S., at 220–221 (relying on Court’s earlier decision in Miller Music Corp. v. Charles N. Daniels, Inc., 362 U. S. 373 (1960) ). Plaintiff below, petitioner here, Paula Petrella (Petrella) is Frank Petrella’s daughter. Learning of this Court’s decision in Stewart, Petrella engaged an attorney who, in 1991, renewed the copyright in the 1963 screenplay. Because the copyrights in the 1973 screenplay and the 1970 book were not timely renewed, the infringement claims in this case rest exclusively on the screenplay registered in 1963. Petrella is now sole owner of the copyright in that work.[8] In 1998, seven years after filing for renewal of the copyright in the 1963 screenplay, Petrella’s attorney informed MGM that Petrella had obtained the copyright to that screenplay. Exploitation of any derivative work, including Raging Bull, the attorney asserted, infringed on the copyright now vested in Petrella. During the next two years, counsel for Petrella and MGM exchanged letters in which MGM denied the validity of the infringement claims, and Petrella repeatedly threatened to take legal action. B Some nine years later, on January 6, 2009, Petrella filed a copyright infringement suit in the United States District Court for the Central District of California. She alleged that MGM violated and continued to violate her copyright in the 1963 screenplay by using, producing, and distributing Raging Bull, a work she described as derivative of the 1963 screenplay. Petrella’s complaint sought monetary and injunctive relief. Because the statute of limitations for copyright claims requires commencement of suit “within three years after the claim accrued,” §507(b), Petrella sought relief only for acts of infringement occurring on or after January 6, 2006. No relief, she recognizes, can be awarded for infringing acts prior to that date. MGM moved for summary judgment on several grounds, among them, the equitable doctrine of laches. Petrella’s 18-year delay, from the 1991 renewal of the copyright on which she relied, until 2009, when she commenced suit, MGM maintained, was unreasonable and prejudicial to MGM. See Memorandum of Points and Authorities in Support of Defendants’ Motion for Summary Judgment in No. CV 09–0072 (CD Cal.). The District Court granted MGM’s motion. See App. to Pet. for Cert. 28a–48a. As to the merits of the infringement claims, the court found, disputed issues of material fact precluded summary adjudication. See id., at 34a–42a. Even so, the court held, laches barred Petrella’s complaint. Id., at 42a–48a. Petrella had unreasonably delayed suit by not filing until 2009, the court concluded, and further determined that MGM was prejudiced by the delay. Id., at 42a–46a. In particular, the court stated, MGM had shown “expectations-based prejudice,” because the company had “made significant investments in exploiting the film”; in addition, the court accepted that MGM would encounter “evidentiary prejudice,” because Frank Petrella had died and LaMotta, then aged 88, appeared to have sustained a loss of memory. Id., at 44a–46a.[9] The U. S. Court of Appeals for the Ninth Circuit affirmed the laches-based dismissal. 695 F. 3d 946 (2012). Under Ninth Circuit precedent, the Court of Appeals first observed, “[i]f any part of the alleged wrongful conduct occurred outside of the limitations period, courts presume that the plaintiff’s claims are barred by laches.” Id., at 951 (internal quotation marks omitted). The presumption was applicable here, the court indicated, because “[t]he statute of limitations for copyright claims in civil cases is three years,” ibid. (citing §507(b)), and Petrella was aware of her potential claims many years earlier (as was MGM), id., at 952. “[T]he true cause of Petrella’s delay,” the court suggested, “was, as [Petrella] admits, that ‘the film hadn’t made money’ [in years she deferred suit].” Id., at 953.[10] Agreeing with the District Court, the Ninth Circuit determined that MGM had established expectations-based prejudice: the company had made a large investment in Raging Bull, believing it had complete ownership and control of the film. Id., at 953–954.[11] Judge Fletcher concurred only because Circuit precedent obliged him to do so. Id., at 958. Laches in copyright cases, he observed, is “entirely a judicial creation,” one notably “in tension with Congress’ [provision of a three-year limitations period].” Ibid. We granted certiorari to resolve a conflict among the Circuits on the application of the equitable defense of laches to copyright infringement claims brought within the three-year look-back period prescribed by Congress.[12] 570 U. S. ___ (2013). III We consider first whether, as the Ninth Circuit held, laches may be invoked as a bar to Petrella’s pursuit of legal remedies under 17 U. S. C. §504(b). The Ninth Circuit erred, we hold, in failing to recognize that the copyright statute of limitations, §507(b), itself takes account of delay. As earlier observed, see supra, at 5–6, a successful plaintiff can gain retrospective relief only three years back from the time of suit. No recovery may be had for infringement in earlier years. Profits made in those years remain the defendant’s to keep. Brought to bear here, §507(b) directs that MGM’s returns on its investment in Raging Bull in years outside the three-year window (years before 2006) cannot be reached by Petrella. Only by disregarding that feature of the statute, and the separate-accrual rule attending §507(b), see supra, at 4–5, could the Court of Appeals presume that infringing acts occurring before January 6, 2006 bar all relief, monetary and injunctive, for infringement occurring on and after that date. See 695 F. 3d, at 951; supra, at 9–10.[13] Moreover, if infringement within the three-year look-back period is shown, the Act allows the defendant to prove and offset against profits made in that period “deductible expenses” incurred in generating those profits. §504(b). In addition, the defendant may prove and offset “elements of profit attributable to factors other than the copyrighted work.” §504(b). The defendant thus may retain the return on investment shown to be attributable to its own enterprise, as distinct from the value created by the infringed work. See Sheldon v. Metro-Goldwyn Pictures Corp., 309 U. S. 390, 402, 407 (1940) (equitably apportioning profits to account for independent contributions of infringing defendant). See also infra, at 19–22 (delay in commencing suit as a factor in determining contours of relief appropriately awarded). Last, but hardly least, laches is a defense developed by courts of equity; its principal application was, and remains, to claims of an equitable cast for which the Legislature has provided no fixed time limitation. See 1 D. Dobbs, Law of Remedies §2.4(4), p. 104 (2d ed. 1993) (hereinafter Dobbs) (“laches . . . may have originated in equity because no statute of limitations applied, . . . suggest[ing] that laches should be limited to cases in which no statute of limitations applies”). Both before and after the merger of law and equity in 1938,[14] this Court has cautioned against invoking laches to bar legal relief. See Holmberg v. Armbrecht, 327 U. S. 392, 395, 396 (1946) (in actions at law, “[i]f Congress explicitly puts a limit upon the time for enforcing a right which it created, there is an end of the matter,” but “[t]raditionally . . . , statutes of limitation are not controlling measures of equitable relief”); Merck & Co. v. Reynolds, 559 U. S. 633, 652 (2010) (quoting, for its current relevance, statement in United States v. Mack, 295 U. S. 480, 489 (1935) , that “[l]aches within the term of the statute of limitations is no defense [to an action] at law”); County of Oneida v. Oneida Indian Nation of N. Y., 470 U. S. 226 , n. 16 (1985) (“[A]pplication of the equitable defense of laches in an action at law would be novel indeed.”).[15] Because we adhere to the position that, in face of a statute of limitations enacted by Congress, laches cannot be invoked to bar legal relief, the dissent thinks we “plac[e] insufficient weight upon the rules and practice of modern litigation.” Post, at 12. True, there has been, since 1938, only “one form of action—the civil action.” Fed. Rule Civ. Proc. 2. But “the substantive and remedial principles [applicable] prior to the advent of the federal rules [have] not changed.” 4 C. Wright & A. Miller, Federal Practice and Procedure §1043, p. 177 (3d ed. 2002). Holmberg, Merck, and Oneida so illustrate. The dissent presents multiple citations, see post, at 1, 3–4, 7–8, 10–11, many of them far afield from the issue at hand, others obscuring what the cited decisions in fact ruled. Compare, e.g., post, at 1, 11, with infra, at 20–21 (describing Chirco v. Crosswinds Communities, Inc., 474 F. 3d 227 (CA6 2007)); post, at 1, 10–11, with infra, at 15, n. 16 (describing National Railroad Passenger Corporation v. Morgan, 536 U. S. 101 (2002) ); post, at 8, with infra, at 15, n. 16 (describing Patterson v. Hewitt, 195 U. S. 309 (1904) ). Yet tellingly, the dissent has come up with no case in which this Court has approved the application of laches to bar a claim for damages brought within the time allowed by a federal statute of limitations. There is nothing at all “differen[t],” see post, at 12, about copyright cases in this regard. IV We turn now to MGM’s principal arguments regarding the contemporary scope of the laches defense, all of them embraced by the dissent. A Laches is listed among affirmative defenses, along with, but discrete from, the statute of limitations, in Federal Rule of Civil Procedure 8(c). Accordingly, MGM maintains, the plea is “available . . . in every civil action” to bar all forms of relief. Tr. of Oral Arg. 43; see Brief for Respondents 40. To the Court’s question, could laches apply where there is an ordinary six-year statute of limitations, MGM’s counsel responded yes, case-specific circumstances might warrant a ruling that a suit brought in year five came too late. Tr. of Oral Arg. 52; see id., at 41. The expansive role for laches MGM envisions careens away from understandings, past and present, of the essentially gap-filling, not legislation-overriding, office of laches. Nothing in this Court’s precedent suggests a doctrineof such sweep. Quite the contrary, we have never applied laches to bar in their entirety claims for discrete wrongs occurring within a federally prescribed limitations pe-riod.[16] Inviting individual judges to set a time limit other than the one Congress prescribed, we note, would tug against the uniformity Congress sought to achieve when it enacted §507(b). See supra, at 3–4. B MGM observes that equitable tolling “is read into every federal statute of limitation,” Holmberg, 327 U. S., at 397, and asks why laches should not be treated similarly. See Brief for Respondents 23–26; post, at 7–8. Tolling, which lengthens the time for commencing a civil action in appropriate circumstances,[17] applies when there is a statute of limitations; it is, in effect, a rule of interpretation tied to that limit. See Young v. United States, 535 U. S. 43 –50 (2002); Johnson v. Railway Express Agency, Inc., 421 U. S. 454, 464 (1975) .[18] Laches, in contrast, originally served as a guide when no statute of limitations controlled the claim; it can scarcely be described as a rule for interpreting a statutory prescription. That is so here, because the statute, §507(b), makes the starting trigger an infringing act committed three years back from the commencement of suit, while laches, as conceived by the Ninth Circuit and advanced by MGM, makes the presumptive trigger the defendant’s initial infringing act. See 695 F. 3d, at 951; Brief for United States 16. C MGM insists that the defense of laches must be avail-able to prevent a copyright owner from sitting still, doing nothing, waiting to see what the outcome of an alleged infringer’s investment will be. See Brief for Respondents 48. In this case, MGM stresses, “[Petrella] conceded that she waited to file because ‘the film was deeply in debt and in the red and would probably never recoup.’ ” Id., at 47 (quoting from App. 110). The Ninth Circuit similarly faulted Petrella for waiting to sue until the film Raging Bull “made money.” 695 F. 3d, at 953 (internal quotation marks omitted). See also post, at 3–6 (deploring plaintiffs who wait to see whether the allegedly infringing work makes money). It is hardly incumbent on copyright owners, however, to challenge each and every actionable infringement. And there is nothing untoward about waiting to see whether an infringer’s exploitation undercuts the value of the copyrighted work, has no effect on the original work, or even complements it. Fan sites prompted by a book or film, for example, may benefit the copyright owner. See Wu, Tolerated Use, 31 Colum. J. L. & Arts 617, 619–620 (2008). Even if an infringement is harmful, the harm may be too small to justify the cost of litigation. If the rule were, as MGM urges, “sue soon, or forever hold your peace,” copyright owners would have to mount a federal case fast to stop seemingly innocuous infringements, lest those infringements eventually grow in magnitude. Section 507(b)’s three-year limitations period, however, coupled to the separate-accrual rule, see supra, at3–6, avoids such litigation profusion. It allows a copyright owner to defer suit until she can estimate whether litigation is worth the candle. She will miss out on damages for periods prior to the three-year look-back, but her right to prospective injunctive relief should, in most cases, remain unaltered.[19] D MGM points to the danger that evidence needed or useful to defend against liability will be lost during a copyright owner’s inaction. Brief for Respondents 37–38; see post, at 2–4.[20] Recall, however, that Congress provided for reversionary renewal rights exercisable by an author’s heirs, rights that can be exercised, at the earliest for pre-1978 copyrights, 28 years after a work was written and copyrighted. See, supra, at 2–3. At that time, the author, and perhaps other witnesses to the creation of the work, will be dead. See supra, at 7. Congress must have been aware that the passage of time and the author’s death could cause a loss or dilution of evidence. Congress chose, nonetheless, to give the author’s family “a second chance to obtain fair remuneration.” Stewart, 495 U. S., at 220. Moreover, a copyright plaintiff bears the burden of proving infringement. See 3 W. Patry, Copyright §9.4, p. 9–18 (2013) (hereinafter Patry) (“As in other civil litigation, a copyright owner bears the burden of establishing a prima facie case.”). But cf. post, at 4 (overlooking plain-tiff’s burden to show infringement and the absence of any burden upon the defendant “to prove that it did not infringe”). Any hindrance caused by the unavailability of evidence, therefore, is at least as likely to affect plaintiffs as it is to disadvantage defendants. That is so in cases of the kind Petrella is pursuing, for a deceased author most probably would have supported his heir’s claim. The registration mechanism, we further note, reduces the need for extrinsic evidence. Although registration is “permissive,” both the certificate and the original work must be on file with the Copyright Office before a copyright owner can sue for infringement. §§408(b), 411(a). Key evidence in the litigation, then, will be the certificate, the original work, and the allegedly infringing work. And the adjudication will often turn on the factfinder’s direct comparison of the original and the infringing works, i.e., on the factfinder’s “good eyes and common sense” in comparing the two works’ “total concept and overall feel.” Peter F. Gaito Architecture, LLC v. Simone Development Corp., 602 F. 3d 57, 66 (CA2 2010) (internal quotation marks omitted). E Finally, when a copyright owner engages in intention-ally misleading representations concerning his abstention from suit, and the alleged infringer detrimentally relies on the copyright owner’s deception, the doctrine of estoppel may bar the copyright owner’s claims completely, eliminating all potential remedies. See 6 Patry §20:58, at 20–110 to 20–112.[21] The test for estoppel is more exacting than the test for laches, and the two defenses are differently oriented. The gravamen of estoppel, a defense long recognized as available in actions at law, see Wehrman v. Conklin, 155 U. S. 314, 327 (1894) , is misleading and consequent loss, see 6 Patry §20:58, at 20–110 to 20–112. Delay may be involved, but is not an element of the defense. For laches, timeliness is the essential element. In contrast to laches, urged by MGM entirely to override the statute of limitations Congress prescribed, estoppel does not undermine Congress’ prescription, for it rests on misleading, whether engaged in early on, or later in time. Stating that the Ninth Circuit “ha[d] taken a wrong turn in its formulation and application of laches in copyright cases,” Judge Fletcher called for fresh consideration of the issue. 695 F. 3d, at 959. “A recognition of the distinction between . . . estoppel and laches,” he suggested, “would be a good place to start.” Ibid. We agree. V The courts below summarily disposed of Petrella’s case based on laches, preventing adjudication of any of her claims on the merits and foreclosing the possibility of any form of relief. That disposition, we have explained, was erroneous. Congress’ time provisions secured to authors a copyright term of long duration, and a right to sue for infringement occurring no more than three years back from the time of suit. That regime leaves “little place” for a doctrine that would further limit the timeliness of a copyright owner’s suit. See 1 Dobbs §2.6(1), at 152. In extraordinary circumstances, however, the consequences of a delay in commencing suit may be of sufficient magnitude to warrant, at the very outset of the litigation, curtailment of the relief equitably awardable. Chirco v. Crosswinds Communities, Inc., 474 F. 3d 227 (CA6 2007), is illustrative. In that case, the defendants were alleged to have used without permission, in planning and building a housing development, the plaintiffs’ copyrighted architectural design. Long aware of the defendants’ project, the plaintiffs took no steps to halt the housing development until more than 168 units were built, 109 of which were occupied. Id., at 230. Although the action was filed within §507(b)’s three-year statute of limitations, the District Court granted summary judgment to the defendants, dismissing the entire case on grounds of laches. The trial court’s rejection of the entire suit could not stand, the Court of Appeals explained, for it was not within the Judiciary’s ken to debate the wisdom of §507(b)’s three-year look-back prescription. Id., at 235. Nevertheless, the Court of Appeals affirmed the District Court’s judgment to this extent: The plaintiffs, even if they might succeed in proving infringement of their copyrighted design, would not be entitled to an order mandating destruction of the housing project. That relief would be inequit-able, the Sixth Circuit held, for two reasons: the plaintiffs knew of the defendants’ construction plans before the de-fendants broke ground, yet failed to take readily available measures to stop the project; and the requested relief would “work an unjust hardship” upon the defendants and innocent third parties. Id., at 236. See also New Era Publications Int’l v. Henry Holt & Co., 873 F. 2d 576, 584–585 (CA2 1989) (despite awareness since 1986 that book containing allegedly infringing material would be published in the United States, copyright owner did not seek a restraining order until 1988, after the book had been printed, packed, and shipped; as injunctive relief “would [have] result[ed] in the total destruction of the work,” the court “relegat[ed plaintiff] to its damages remedy”). In sum, the courts below erred in treating laches as a complete bar to Petrella’s copyright infringement suit. The action was commenced within the bounds of §507(b), the Act’s time-to-sue prescription, and does not present extraordinary circumstances of the kind involved in Chirco and New Era. Petrella notified MGM of her copyright claims before MGM invested millions of dollars in creating a new edition of Raging Bull. And the equitable relief Petrella seeks—e.g., disgorgement of unjust gains and an injunction against future infringement—would not result in “total destruction” of the film, or anything close to it. See New Era, 873 F. 2d, at 584. MGM released Raging Bull more than three decades ago and has marketed it continuously since then. Allowing Petrella’s suit to go forward will put at risk only a fraction of the income MGM has earned during that period and will work no unjust hardship on innocent third parties, such as consumers who have purchased copies of Raging Bull. Cf. Chirco, 474 F. 3d, at 235–236 (destruction remedy would have ousted families from recently purchased homes). The circumstances here may or may not (we need not decide) warrant limiting relief at the remedial stage, but they are not sufficiently extraordinary to justify threshold dismissal. Should Petrella ultimately prevail on the merits, the District Court, in determining appropriate injunctive relief and assessing profits, may take account of her delay in commencing suit. See supra, at 1–2, 11–12. In doing so, however, that court should closely examine MGM’s alleged reliance on Petrella’s delay.[22] This examination should take account of MGM’s early knowledge of Petrella’s claims, the protection MGM might have achieved through pursuit of a declaratory judgment action, the extent to which MGM’s investment was protected by the separate-accrual rule, the court’s authority to order injunctive relief “on such terms as it may deem reasonable,” §502(a), and any other considerations that would justify adjusting injunctive relief or profits. See Haas v. Leo Feist, Inc., 234 F. 105, 107–108 (SDNY 1916) (adjudicating copyright infringement suit on the merits and decreeing injunctive relief, but observing that, in awarding profits, account may be taken of copyright owner’s inaction until infringer had spent large sums exploiting the work at issue). See also Tr. of Oral Arg. 23 (Government observation that, in fashioning equitable remedies, court has considerable leeway; it could, for example, allow MGM to continue using Raging Bull as a derivative work upon payment of a reasonable royalty to Petrella). Whatever adjustments may be in order in awarding injunctive relief, and in accounting for MGM’s gains and profits, on the facts thus far presented, there is no evident basis for immunizing MGM’s present and future uses of the copyrighted work, free from any obligation to pay royalties. * * * For the reasons stated, the judgment of the United States Court of Appeals for the Ninth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 As infringement remedies, the Copyright Act provides for injunctions, §502, impoundment and disposition of infringing articles, §503, damages and profits, §504, costs and attorney’s fees, §505. Like other restitutional remedies, recovery of profits “is not easily characterized as legal or equitable,” for it is an “amalgamation of rights and remedies drawn from both systems.” Restatement (Third) of Restitution and Unjust Enrichment §4, Comment , p. 28 (2010). Given the “protean character” of the profits-recovery remedy, see Comment , at 30, we regard as appropriate its treatment as “equitable” in this case. 2 For post-1978 works, heirs still have an opportunity to recapture rights of the author. See 3 M. Nimmer & D. Nimmer, Copyright §11.01[A], p. 11–4 (2013) (hereinafter Nimmer). 3 The Copyright Act was pervasively revised in 1976, but the three-year look-back statute of limitations has remained materially unchanged. See Act of Oct. 19, 1976, §101, . 4 Although we have not passed on the question, nine Courts of Appeals have adopted, as an alternative to the incident of injury rule, a “discovery rule,” which starts the limitations period when “the plaintiff discovers, or with due diligence should have discovered, the injury that forms the basis for the claim.” v. , 568 F. 3d 425, 433 (CA3 2009) (internal quotation marks omitted). See also 6 W. Patry, Copyright §20:19, p. 20–28 (2013) (hereinafter Patry) (“The overwhelming majority of courts use discovery accrual in copyright cases.”). 5 See generally ., §20:23, at 20–44; 3 Nimmer §12.05[B][1][b], at 12–150.2 to 12–150.4. See also, , 568 F. 3d, at 433; v. , 533 F. 3d 1287, 1320, n. 39 (CA11 2008); v. , 376 F. 3d 615, 621 (CA6 2004); v. , 37 F. 3d 180, 182 (CA5 1994); v. , 19 F. 3d 479, 481 (CA9 1994). 6 Separately accruing harm should not be confused with harm from past violations that are continuing. Compare v. , (for separately accruing harm, each new act must cause “harm [to the plaintiff] over and above the harm that the earlier acts caused”), with v. , –381 (1982) (“[W]here a plaintiff . . . challenges . . . an unlawful practice that continues into the limitations period, the complaint is timely when it is filed within [the limitations period, measured from] the last asserted occurrence of that practice.” (footnote omitted)). 7 A case arising outside of the copyright context is illustrative. In v. , , an employer was delinquent in making a series of scheduled payments to an underfunded pension plan. See at 198–199. The trustees filed suit just over six years after the first missed payment, barely outside of the applicable six-year statute of limitations. See at 198. Because the first missed payment in the series fell outside the statute of limitations, the employer argued that the subsequent missed payments were also time barred. See at 206. We rejected that argument. The remaining claims were timely, we held, because “each missed payment create[d] a separate cause of action with its own six-year limitations period.” Cf. , 521 U. S., at 190 (for civil Racketeer Influenced and Corrupt Organizations Act claims, plaintiff may recover for acts occurring within the limitations period, but may not use an “independent, new predicate act as a bootstrap to recover for injuries caused by other earlier predicate acts that took place outside the limitations period”); v. , –121 (2002) (distinguishing discrete acts, each independently actionable, from conduct “cumulative [in] effect,” hostile environment claims pursued under Title VII of the Civil Rights Act of 1964, ; “in direct contrast to discrete acts, a single [instance of hostility] may not be actionable on its own”). But cf. at 10–11 (ignoring the distinction took care to draw between discrete acts independently actionable and conduct cumulative in effect). 8 Petrella’s attorney filed the renewal application on behalf of Frank Petrella’s heirs. When Petrella’s mother died and her brother assigned his rights to her, Petrella became the sole owner of all rights in the 1963 screenplay. 9 LaMotta, the court noted, “ha[d] suffered myriad blows to his head as a fighter years ago,” and “no longer recognize[d Petrella], even though he ha[d] known her for forty years.” App. to Pet. for Cert.45a–46a. 10 In her declaration, Petrella stated that MGM told her in 2001 that the film was in “a huge deficit financially,” “would never show a profit,” and, for that reason, “MGM would not continue to send [financial] statements [to her].” App. 234. 11 The Court of Appeals did not consider whether MGM had also shown evidentiary prejudice. 695 F. 3d 946, 953 (CA9 2012). 12 See v. , 243 F. 3d 789, 798 (CA4 2001) (laches defense unavailable in copyright infringement cases, regardless of remedy sought); , 533 F. 3d, at 1320 (“[T]here is a strong presumption [in copyright cases] that a plaintiff’s suit is timely if it is filed before the statute of limitations has run. Only in the most extraordinary circumstances will laches be recognized as a defense.”); v. , 474 F. 3d 227, 233 (CA6 2007) (in copyright litigation, laches applies only to “the most compelling of cases”); v. , 287 F. 3d 936, 950 (CA10 2002) (“Rather than deciding copyright cases on the issue of laches, courts should generally defer to the three-year statute of limitations.”); v. , 873 F. 2d 576, 584–585 (CA2 1989) (“severe prejudice, coupled with . . . unconscionable delay . . . mandates denial of . . . injunction for laches and relegation of [plaintiff] to its damages remedy”). Cf. at 1, 13 (acknowledging that application of laches should be “extraordinary,” confined to “few and unusual cases”). 13 Assuming Petrella had a winning case on the merits, the Court of Appeals’ ruling on laches would effectively give MGM a cost-free license to exploit Raging Bull throughout the long term of the copyright. The value to MGM of such a free, compulsory license could exceed by far MGM’s expenditures on the film. 14 See Fed. Rule Civ. Proc. 2 (“There is one form of action—the civil action.”); Rule 8(c) (listing among affirmative defenses both “laches” and “statute of limitations”). 15 In contrast to the Copyright Act, the Lanham Act, which governs trademarks, contains no statute of limitations, and expressly provides for defensive use of “equitable principles, including laches.” . But cf. at 8, 11 (citing v. 191 F. 3d 813 (CA7 1999), but failing to observe that Lanham Act contains no statute of limitations). 16 MGM pretends otherwise, but the cases on which it relies do not carry the load MGM would put on them. , described , at 6, n. 7, is apparently MGM’s best case, for it is cited 13 times in MGM’s brief. See Brief for Respondents 8, 9, 14, 16, 18, 19, 25, 31, 34, 35, 36, 40, 47; at 1, 7, 10–11. , however, does not so much as hint that laches may bar claims for discrete wrongs, all of them occurring within a federal limitations period. Part II–A of that opinion, dealing with the separate-accrual rule, held that “[e]ach discrete discriminatory act starts a new clock for filing charges alleging that act,” regardless of whether “past acts” are time barred. 536 U. S., at 113. Parts II–B and II–C of the opinion then separately accruing wrongs from hostile-work-environment claims, cumulative in effect and extending over long periods of time. , at 115–117, 121. Laches could be invoked, the Court reasoned, to limit the continuing violation doctrine’s potential to rescue claims, not claims accruing separately within the limitations period. 17 a party’s infancy or mental disability, absence of the defendant from the jurisdiction, fraudulent concealment. See S. Rep. No. 1014, 85th Cong., 1st Sess., 2–3 (1957) (hereinafter Senate Report). 18 The legislative history to which the dissent refers, at 7, speaks of “equitable situations on which the statute of limitations is generally suspended,” Senate Report 3, and says nothing about laches shrinking the time Congress allowed. 19 The dissent worries that a plaintiff might sue for profits “every three years . . . until the copyright expires.” at 5; see at 2. That suggestion neglects to note that a plaintiff who proves infringement will likely gain forward-looking injunctive relief stopping the defendant’s repetition of infringing acts. 20 As earlier noted, see at 10, n. 11, the Court of Appeals did not reach the question whether evidentiary prejudice existed. 695 F. 3d, at 953. 21 Although MGM, in its answer to Petrella’s complaint, separately raised both laches and estoppel as affirmative defenses, see Defendants’ Answer to Plaintiff’s Complaint in No. CV 09–0072 (CD Cal.), the courts below did not address the estoppel plea. 22 While reliance or its absence may figure importantly in this case, we do not suggest that reliance is in all cases a for adjustment of injunctive relief or profits.

572.US.765

Donald Rickard led police officers on a high-speed car chase that came to a temporary halt when Rickard spun out into a parking lot. Rickard resumed maneuvering his car, and as he continued to use the accelerator even though his bumper was flush against a patrol car, an officer fired three shots into Rickard’s car. Rickard managed to drive away, almost hitting an officer in the process. Officers fired 12 more shots as Rickard sped away, striking him and his passenger, both of whom died from some combination of gunshot wounds and injuries suffered when the car eventually crashed. Respondent, Rickard’s minor daughter, filed a 42 U. S. C. §1983 action, alleging that the officers used excessive force in violation of the Fourth and Fourteenth Amendments. The District Court denied the officers’ motion for summary judgment based on qualified immunity, holding that their conduct violated the Fourth Amendment and was contrary to clearly established law at the time in question. After finding that it had appellate jurisdiction, the Sixth Circuit held that the officers’ conduct violated the Fourth Amendment. It affirmed the District Court’s order, suggesting that it agreed that the officers violated clearly established law. Held: 1. The Sixth Circuit properly exercised jurisdiction under 28 U. S. C. §1291, which gives courts of appeals jurisdiction to hear appeals from “final decisions” of the district courts. The general rule that an order denying a summary judgment motion is not a “final decision[n],” and thus not immediately appealable, does not apply when it is based on a qualified immunity claim. Johnson v. Jones, 515 U.S. 304, 311. Respondent argues that Johnson forecloses appellate jurisdiction here, but the order in Johnson was not immediately appealable because it merely decided “a question of ‘evidence sufficiency,’ ” id., at 313, while here, petitioners’ qualified immunity claims raise legal issues quite different from any purely factual issues that might be confronted at trial. Deciding such legal issues is a core responsibility of appellate courts and does not create an undue burden for them. See, e.g., Scott v. Harris, 550 U.S. 372. Pp. 5–7. 2. The officers’ conduct did not violate the Fourth Amendment. Pp. 7–15. (a) Addressing this question first will be “beneficial” in “develop[ing] constitutional precedent” in an area that courts typically consider in cases in which the defendant asserts a qualified immunity defense, Pearson v. Callahan, 555 U.S. 223, 236. Pp. 7–8. (b) Respondent’s excessive-force argument requires analyzing the totality of the circumstances from the perspective “of a reasonable officer on the scene.” Graham v. Connor, 490 U.S. 386, 396. Respondent contends that the Fourth Amendment did not allow the officers to use deadly force to terminate the chase, and that, even if they were permitted to fire their weapons, they went too far when they fired as many rounds as they did. Pp. 8–12. (1) The officers acted reasonably in using deadly force. A “police officer’s attempt to terminate a dangerous high-speed car chase that threatens the lives of innocent bystanders does not violate the Fourth Amendment, even when it places the fleeing motorist at risk of serious injury or death.” Scott, supra, at 385. Rickard’s outrageously reckless driving—which lasted more than five minutes, exceeded 100 miles per hour, and included the passing of more than two dozen other motorists—posed a grave public safety risk, and the record conclusively disproves that the chase was over when Rickard’s car came to a temporary standstill and officers began shooting. Under the circumstances when the shots were fired, all that a reasonable officer could have concluded from Rickard’s conduct was that he was intent on resuming his flight, which would again pose a threat to others on the road. Pp. 9–11. (2) Petitioners did not fire more shots than necessary to end the public safety risk. It makes sense that, if officers are justified in firing at a suspect in order to end a severe threat to public safety, they need not stop shooting until the threat has ended. Here, during the 10-second span when all the shots were fired, Rickard never abandoned his attempt to flee and eventually managed to drive away. A passenger’s presence does not bear on whether officers violated Rickard’s Fourth Amendment rights, which “are personal rights [that] may not be vicariously asserted.” Alderman v. United States, 394 U.S. 165, 174. Pp. 11–12. 3. Even if the officers’ conduct had violated the Fourth Amendment, petitioners would still be entitled to summary judgment based on qualified immunity. An official sued under §1983 is entitled to qualified immunity unless it is shown that the official violated a statutory or constitutional right that was “ ‘clearly established’ ” at the time of the challenged conduct. Ashcroft v. al-Kidd, 563 U. S. ___, ___. Brosseau v. Haugen, 543 U.S. 194, 201, where an officer shot at a fleeing vehicle to prevent possible harm, makes plain that no clearly established law precluded the officer’s conduct there. Thus, to prevail, respondent must meaningfully distinguish Brosseau or point to any “controlling authority” or “robust ‘consensus of cases of persuasive authority,’ ” al-Kidd, supra, at ___, that emerged between the events there and those here that would alter the qualified-immunity analysis. Respondent has made neither showing. If anything, the facts here are more favorable to the officers than the facts in Brosseau; and respondent points to no cases that could be said to have clearly established the unconstitutionality of using lethal force to end a high-speed car chase. Pp. 12–15. 509 Fed. Appx. 388, reversed and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Thomas, Sotomayor, and Kagan, JJ., joined, in which Ginsburg, J., joined as to the judgment and Parts I, II, and III–C, and in which Breyer, J., joined except as to Part III–B–2.

judgment.” 509 Fed. Appx., at 393. 4 In holding that petitioners’ conduct violated the , the District Court relied on reasoning that is irreconcilable with our decision in . The District Court held that the danger presented by a high-speed chase cannot justify the use of deadly force because that danger was caused by the officers’ decision to continue the chase. v. , 2011 WL 197426, *8 (WD Tenn., Jan. 20, 2011). In , however, we declined to “lay down a rule requiring the police to allow fleeing suspects to get away whenever they drive so recklessly that they put other people’s lives in danger,” concluding that the Constitution “assuredly does not impose this invitation to impunity-earned-by-recklessness.” 550 U. S., at 385–386. 5 There seems to be some disagreement among lower courts as to whether a passenger in Allen’s situation can recover under a theory. Compare v. , 343 F. 3d 1323 (CA11 2003) (suggesting yes), and v. , 234 F. 3d 312 (CA6 2000) (same), with v. , 243 F. 3d 157 (CA4 2001) (suggesting no), and v. , 906 F. 2d 791 (CA1 1990) (same). We express no view on this question. We also note that in v. , , the Court held that a passenger killed as a result of a police chase could recover under a substantive due process theory only if the officer had “a purpose to cause harm unrelated to the legitimate object of arrest.”

573.US.102

This case involves the intersection of two federal statutes. The Lanham Act permits one competitor to sue another for unfair competition arising from false or misleading product descriptions. 15 U. S. C. §1125. The Federal Food, Drug, and Cosmetic Act (FDCA) prohibits the misbranding of food and drink. 21 U. S. C. §§321(f), 331. To implement the FDCA’s provisions, the Food and Drug Administration (FDA) has promulgated regulations regarding food and beverage labeling, including one concerning juice blends. Unlike the Lanham Act, which, relies in large part for its enforcement on private suits brought by injured competitors, the FDCA and its regulations give the United States nearly exclusive enforcement authority and do not permit private enforcement suits. The FDCA also pre-empts certain state misbranding laws. Petitioner POM Wonderful LLC, which produces, markets, and sells, inter alia, a pomegranate-blueberry juice blend, filed a Lanham Act suit against respondent Coca-Cola Company, alleging that the name, label, marketing, and advertising of one of Coca-Cola’s juice blends mislead consumers into believing the product consists predominantly of pomegranate and blueberry juice when it in fact consists predominantly of less expensive apple and grape juices, and that the ensuing confusion causes POM to lose sales. The District Court granted partial summary judgment to Coca-Cola, ruling that the FDCA and its regulations preclude Lanham Act challenges to the name and label of Coca-Cola’s juice blend. The Ninth Circuit affirmed in relevant part. Held: Competitors may bring Lanham Act claims like POM’s challenging food and beverage labels regulated by the FDCA. Pp. 7–17. (a) This result is based on the following premises. First, this is not a pre-emption case, for it does not raise the question whether state law is pre-empted by a federal law, see Wyeth v. Levine, 555 U.S. 555, 563, but instead concerns the alleged preclusion of a cause of action under one federal statute by the provisions of another federal statute. Pre-emption principles may nonetheless be instructive insofar as they are designed to assess the interaction of laws bearing on the same subject. Second, this is a statutory interpretation case; and analysis of the statutory text, aided by established interpretation rules, controls. See Chickasaw Nation v. United States, 534 U.S. 84, 94. While a principle of interpretation may be countered “by some maxim pointing in a different direction,” Circuit City Stores, Inc. v. Adams, 532 U.S. 105, 115, this Court need not decide what maxim establishes the proper framework here: Even assuming that Coca-Cola is correct that the Court’s task is to reconcile or harmonize the statutes instead of to determine whether one statute is an implied repeal in part of another statute, Coca-Cola is incorrect that the best way to do that is to bar POM’s Lanham Act claim. Pp. 7–9. (b) Neither the Lanham Act nor the FDCA, in express terms, forbids or limits Lanham Act claims challenging labels that are regulated by the FDCA. The absence of such a textual provision when the Lanham Act and the FDCA have coexisted for over 70 years is “powerful evidence that Congress did not intend FDA oversight to be the exclusive means” of ensuring proper food and beverage labeling. See Wyeth, supra, at 575. In addition, and contrary to Coca-Cola’s argument, Congress, by taking care to pre-empt only some state laws, if anything indicated it did not intend the FDCA to preclude requirements arising from other sources. See Setser v. United States, 566 U. S. ___, ___. The structures of the FDCA and the Lanham Act reinforce this conclusion. Where two statutes are complementary, it would show disregard for the congressional design to hold that Congress intended one federal statute nonetheless to preclude the operation of the other. See J. E. M. Ag Supply, Inc. v. Pioneer Hi-Bred Int’l, Inc., 534 U.S. 124, 144. The Lanham Act and the FDCA complement each other in major respects, for each has its own scope and purpose. Both touch on food and beverage labeling, but the Lanham Act protects commercial interests against unfair competition, while the FDCA protects public health and safety. They also complement each other with respect to remedies. The FDCA’s enforcement is largely committed to the FDA, while the Lanham Act empowers private parties to sue competitors to protect their interests on a case-by-case basis. Allowing Lanham Act suits takes advantage of synergies among multiple methods of regulation. A holding that the FDCA precludes Lanham Act claims challenging food and beverage labels also could lead to a result that Congress likely did not intend. Be-cause the FDA does not necessarily pursue enforcement measures regarding all objectionable labels, preclusion of Lanham Act claims could leave commercial interests—and indirectly the public at large—with less effective protection in the food and beverage labeling realm than in other less regulated industries. Pp. 9–12. (c) Coca-Cola’s arguments do not support its claim that preclusion is proper because Congress intended national uniformity in food and beverage labeling. First, the FDCA’s delegation of enforcement authority to the Federal Government does not indicate that Congress intended to foreclose private enforcement of other federal statutes. Second, the FDCA’s express pre-emption provision applies by its terms to state, not federal, law. Even if it were proper to stray from that text, it not clear that Coca-Cola’s national uniformity assertions reflect the congressional design. Finally, the FDCA and its implementing regulations may address food and beverage labeling with more specificity than the Lanham Act, but this specificity would matter only if the two Acts cannot be implemented in full at the same time. Here, neither the statutory structure nor the empirical evidence of which the Court is aware indicates there will be any difficulty in fully enforcing each statute according to its terms. Pp. 13–15. (d) The Government’s intermediate position—that a Lanham Act claim is precluded “to the extent the FDCA or FDA regulations specifically require or authorize the challenged aspects of [the] label,” and that this rule precludes POM’s challenge to the name of Coca-Cola’s product—is flawed, for the Government assumes that the FDCA and its regulations are a ceiling on the regulation of food and beverage labeling when Congress intended the Lanham Act and the FDCA to complement each other with respect to labeling. Though the FDA’s rulemaking alludes at one point to a balance of interests, it neither discusses nor cites the Lanham Act; and the Government points to no other statement suggesting that the FDA considered the full scope of interests protected by the Lanham Act. Even if agency regulations with the force of law that purport to bar other legal remedies may do so, it is a bridge too far to accept an agency’s after-the-fact statement to justify that result here. An agency may not reorder federal statutory rights without congressional authorization. Pp. 15–17. 679 F.3d 1170, reversed and remanded. Kennedy, J., delivered the opinion of the Court, in which all other Members joined, except Breyer, J., who took no part in the consideration or decision of the case.

POM Wonderful LLC makes and sells pomegranate juice products, including a pomegranate-blueberry juice blend. App. 23a. One of POM’s competitors is the Coca-Cola Company. Coca-Cola’s Minute Maid Division makes a juice blend sold with a label that, in describing the contents, displays the words “pomegranate blueberry” with far more prominence than other words on the label that show the juice to be a blend of five juices. In truth, the Coca-Cola product contains but 0.3% pomegranate juice and 0.2% blueberry juice. Alleging that the use of that label is deceptive and misleading, POM sued Coca-Cola under §43 of the Lanham Act. 60Stat. 441, as amended, 15 U. S. C. §1125. That provision allows one competitor to sue another if it alleges unfair competition arising from false or misleading product descriptions. The Court of Appeals for the Ninth Circuit held that, in the realm of labeling for food and beverages, a Lanham Act claim like POM’s is precluded by a second federal statute. The second statute is the Federal Food, Drug, and Cosmetic Act (FDCA), which forbids the misbranding of food, including by means of false or misleading labeling. §§301, 403, 52Stat. 1042, 1047, as amended, 21 U. S. C. §§331, 343. The ruling that POM’s Lanham Act cause of action is precluded by the FDCA was incorrect. There is no statutory text or established interpretive principle to support the contention that the FDCA precludes Lanham Act suits like the one brought by POM in this case. Nothing in the text, history, or structure of the FDCA or the Lanham Act shows the congressional purpose or design to forbid these suits. Quite to the contrary, the FDCA and the Lanham Act complement each other in the federal regulation of misleading food and beverage labels. Competitors, in their own interest, may bring Lanham Act claims like POM’s that challenge food and beverage labels that are regulated by the FDCA. I A This case concerns the intersection and complementar-ity of these two federal laws. A proper beginning point is a description of the statutes. Congress enacted the Lanham Act nearly seven decades ago. See 60Stat. 427 (1946). As the Court explained earlier this Term, it “requires no guesswork” to ascertain Congress’ intent regarding this federal law, for Congress included a “detailed statement of the statute’s purposes.” Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U. S. ___, ___ (2014) (slip op., at 12). Section 45 of the Lanham Act provides: “The intent of this chapter is to regulate commerce within the control of Congress by making actionable the deceptive and misleading use of marks in such commerce; to protect registered marks used in such commerce from interference by State, or territorial legislation; to protect persons engaged in such commerce against unfair competition; to prevent fraud and deception in such commerce by the use of reproductions, copies, counterfeits, or colorable imitations of registered marks; and to provide rights and remedies stipulated by treaties and conventions respecting trademarks, trade names, and unfair competition entered into between the United States and foreign nations.” 15 U. S. C. §1127. The Lanham Act’s trademark provisions are the primary means of achieving these ends. But the Act also creates a federal remedy “that goes beyond trademark protection.” Dastar Corp. v. Twentieth Century Fox Film Corp., 539 U. S. 23, 29 (2003) . The broader remedy is at issue here. The Lanham Act creates a cause of action for unfair competition through misleading advertising or labeling. Though in the end consumers also benefit from the Act’s proper enforcement, the cause of action is for competitors, not consumers. The term “competitor” is used in this opinion to indicate all those within the class of persons and entities protected by the Lanham Act. Competitors are within the class that may invoke the Lanham Act because they may suffer “an injury to a commercial interest in sales or business reputation proximately caused by [a] defendant’s misrepresentations.” Lexmark, supra, at ___ (slip op., at 22). The petitioner here asserts injury as a competitor. The cause of action the Act creates imposes civil liability on any person who “uses in commerce any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact, which . . . misrepresents the nature, characteristics, qualities, or geographic origin of his or her or another person’s goods, services, or commercial activities.” 15 U. S. C. §1125(a)(1). As the Court held this Term, the private remedy may be invoked only by those who “allege an injury to a commercial interest in reputation or sales. A consumer who is hoodwinked into purchasing a disappointing product may well have an injury-in-fact cognizable under Article III, but he cannot invoke the protection of the Lanham Act.” Lexmark, 572 U. S., at ___ (slip op., at 13). This principle reflects the Lanham Act’s purpose of “ ‘protect[ing] persons engaged in [commerce within the control of Congress] against unfair competition.’ ” Id., at ___ (slip op., at 12). POM’s cause of action would be straightforward enough but for Coca-Cola’s contention that a separate federal statutory regime, the FDCA, allows it to use the label in question and in fact precludes the Lanham Act claim. So the FDCA is the second statute to be discussed. The FDCA statutory regime is designed primarily to protect the health and safety of the public at large. See 62 Cases of Jam v. United States, 340 U. S. 593, 596 (1951) ; FDCA, §401, 52Stat. 1046, 21 U. S. C. §341 (agency may issue certain regulations to “promote honesty and fair dealing in the interest of consumers”). The FDCA prohibits the misbranding of food and drink. 21 U. S. C. §§321(f), 331. A food or drink is deemed misbranded if, inter alia, “its labeling is false or misleading,” §343(a), information required to appear on its label “is not prominently placed thereon,” §343(f), or a label does not bear “the common or usual name of the food, if any there be,” §343(i). To implement these provisions, the Food and Drug Administration (FDA) promulgated regulations regarding food and beverage labeling, including the labeling of mixes of different types of juice into one juice blend. See 21 CFR §102.33 (2013). One provision of those regulations is particularly relevant to this case: If a juice blend does not name all the juices it contains and mentions only juices that are not predominant in the blend, then it must either declare the percentage content of the named juice or “[i]ndicate that the named juice is present as a flavor or flavoring,” e.g., “raspberry and cranberry flavored juice drink.” §102.33(d). The Government represents that the FDA does not preapprove juice labels under these regulations. See Brief for United States as Amicus Curiae in Opposition 16. That contrasts with the FDA’s regulation of other types of labels, such as drug labels, see 21 U. S. C. §355(d), and is consistent with the less extensive role the FDA plays in the regulation of food than in the regulation of drugs. Unlike the Lanham Act, which relies in substantial part for its enforcement on private suits brought by injured competitors, the FDCA and its regulations provide the United States with nearly exclusive enforcement author-ity, including the authority to seek criminal sanctions in some circumstances. 21 U. S. C. §§333(a), 337. Private parties may not bring enforcement suits. §337. Also unlike the Lanham Act, the FDCA contains a provision pre-empting certain state laws on misbranding. That provision, which Congress added to the FDCA in the Nutrition Labeling and Education Act of 1990, §6, 104Stat. 2362–2364, forecloses a “State or political subdivision of a State” from establishing requirements that are of the type but “not identical to” the requirements in some of the misbranding provisions of the FDCA. 21 U. S. C. §343–1(a). It does not address, or refer to, other federal statutes or the preclusion thereof. B POM Wonderful LLC is a grower of pomegranates anda distributor of pomegranate juices. Through its POM Wonderful brand, POM produces, markets, and sells a variety of pomegranate products, including a pomegranate-blueberry juice blend. App. 23a. POM competes in the pomegranate-blueberry juice market with the Coca-Cola Company. Coca-Cola, under its Minute Maid brand, created a juice blend containing 99.4% apple and grape juices, 0.3% pomegranate juice, 0.2% blueberry juice, and 0.1% raspberry juice. Id., at 38a; Brief for Respondent 8. Despite the minuscule amount of pomegranate and blueberry juices in the blend, the front label of the Coca-Cola product displays the words “pomegranate blueberry” in all capital letters, on two separate lines. App. 38a. Below those words, Coca-Cola placed the phrase “flavored blend of 5 juices” in much smaller type. Ibid. And below that phrase, in still smaller type, were the words “from concentrate with added ingredients”—and, with a line break before the final phrase— “and other natural flavors.” Ibid. The product’s front label also displays a vignette of blueberries, grapes, and raspberries in front of a halved pomegranate and a halved apple. Ibid. Claiming that Coca-Cola’s label tricks and deceives consumers, all to POM’s injury as a competitor, POM brought suit under the Lanham Act. POM alleged that the name, label, marketing, and advertising of Coca-Cola’s juice blend mislead consumers into believing the product consists predominantly of pomegranate and blueberry juice when it in fact consists predominantly of less expensive apple and grape juices. Id., at 27a. That confusion, POM complained, causes it to lose sales. Id., at 28a. POM sought damages and injunctive relief. Id., at 32a–33a. The District Court granted partial summary judgment to Coca-Cola on POM’s Lanham Act claim, ruling that the FDCA and its regulations preclude challenges to the name and label of Coca-Cola’s juice blend. The District Court reasoned that in the juice blend regulations the “FDA has directly spoken on the issues that form the basis of Pom’s Lanham Act claim against the naming and labeling of” Coca-Cola’s product, but has not prohibited any, and indeed expressly has permitted some, aspects of Coca-Cola’s label. 727 F. Supp. 2d 849, 871–873 (CD Cal. 2010). The Court of Appeals for the Ninth Circuit affirmed in relevant part. Like the District Court, the Court of Appeals reasoned that Congress decided “to entrust matters of juice beverage labeling to the FDA”; the FDA has promulgated “comprehensive regulation of that labeling”; and the FDA “apparently” has not imposed the requirements on Coca-Cola’s label that are sought by POM. 679 F. 3d 1170, 1178 (2012). “[U]nder [Circuit] precedent,” the Court of Appeals explained, “for a court to act when the FDA has not—despite regulating extensively in this area—would risk undercutting the FDA’s expert judgments and authority.” Id., at 1177. For these reasons, and “[o]utof respect for the statutory and regulatory scheme,” the Court of Appeals barred POM’s Lanham Act claim. Id., at 1178. II A This Court granted certiorari to consider whether a private party may bring a Lanham Act claim challenging a food label that is regulated by the FDCA. 571 U. S. ___ (2014). The answer to that question is based on the following premises. First, this is not a pre-emption case. In pre-emption cases, the question is whether state law is pre-empted by a federal statute, or in some instances, a federal agency action. See Wyeth v. Levine, 555 U. S. 555, 563 (2009) . This case, however, concerns the alleged preclusion of a cause of action under one federal statute by the provisions of another federal statute. So the state-federal balance does not frame the inquiry. Because this is a preclusion case, any “presumption against pre-emption,” id., at 565, n. 3, has no force. In addition, the preclusion analysis is not governed by the Court’s complex categorization of the types of pre-emption. See Crosby v. National Foreign Trade Council, 530 U. S. 363 –373 (2000). Although the Court’s pre-emption precedent does not govern preclusion analysis in this case, its principles are instructive insofar as they are designed to assess the interaction of laws that bear on the same subject. Second, this is a statutory interpretation case and the Court relies on traditional rules of statutory interpretation. That does not change because the case involves multiple federal statutes. See FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120 –139 (2000). Nor does it change because an agency is involved. See ibid. Analysis of the statutory text, aided by established principles of interpretation, controls. See Chickasaw Nation v. United States, 534 U. S. 84, 94 (2001) . A principle of interpretation is “often countered, of course, by some maxim pointing in a different direction.” Circuit City Stores, Inc. v. Adams, 532 U. S. 105, 115 (2001) . It is thus unsurprising that in this case a threshold dispute has arisen as to which of two competing maxims establishes the proper framework for decision. POM argues that this case concerns whether one statute, the FDCA as amended, is an “implied repeal” in part of an-other statute, i.e., the Lanham Act. See, e.g., Carcieri v. Salazar, 555 U. S. 379, 395 (2009) . POM contends that in such cases courts must give full effect to both statutes unless they are in “irreconcilable conflict,” see ibid., and that this high standard is not satisfied here. Coca-Cola resists this canon and its high standard. Coca-Cola argues that the case concerns whether a more specific law, the FDCA, clarifies or narrows the scope of a more general law, the Lanham Act. See, e.g., United States v. Fausto, 484 U. S. 439, 453 (1988) ; Brief for Respondent 18. The Court’s task, it claims, is to “reconcil[e]” the laws, ibid., and it says the best reconciliation is that the more specific provisions of the FDCA bar certain causes of action authorized in a general manner by the Lanham Act. The Court does not need to resolve this dispute. Even assuming that Coca-Cola is correct that the Court’s task is to reconcile or harmonize the statutes and not, as POM urges, to enforce both statutes in full unless there is a genuinely irreconcilable conflict, Coca-Cola is incorrect that the best way to harmonize the statutes is to bar POM’s Lanham Act claim. B Beginning with the text of the two statutes, it must be observed that neither the Lanham Act nor the FDCA, in express terms, forbids or limits Lanham Act claims challenging labels that are regulated by the FDCA. By its terms, the Lanham Act subjects to suit any person who “misrepresents the nature, characteristics, qualities, or geographic origin” of goods or services. 15 U. S. C. §1125(a). This comprehensive imposition of liability extends, by its own terms, to misrepresentations on labels, including food and beverage labels. No other provision in the Lanham Act limits that understanding or purports to govern the relevant interaction between the Lanham Act and the FDCA. And the FDCA, by its terms, does not preclude Lanham Act suits. In consequence, food and beverage labels regulated by the FDCA are not, under the terms of either statute, off limits to Lanham Act claims. No textual provision in either statute discloses a purpose to bar unfair competition claims like POM’s. This absence is of special significance because the Lanham Act and the FDCA have coexisted since the passage of the Lanham Act in 1946. 60Stat. 427 (1946); ch. 675, 52Stat. 1040 (1938). If Congress had concluded, in light of experience, that Lanham Act suits could interfere with the FDCA, it might well have enacted a provision addressing the issue during these 70 years. See Wyeth, supra, at 574 (“If Congress thought state-law suits posed an obstacle to its objectives, it surely would have enacted an express pre-emption provision at some point during the FDCA’s 70-year history”). Congress enacted amendments to the FDCA and the Lanham Act, see, e.g., Nutrition Labeling and Education Act of 1990, 104Stat. 2353; Trademark Law Revision Act of 1988, §132, 102Stat. 3946, including an amendment that added to the FDCA an express pre-emption provision with respect to state laws addressing food and beverage misbranding, §6, 104Stat. 2362. Yet Congress did not enact a provision addressing the preclusion of other federal laws that might bear on food and beverage labeling. This is “powerful evidence that Congress did not intend FDA oversight to be the exclusive means” of ensuring proper food and beverage labeling. See Wyeth, 555 U. S., at 575. Perhaps the closest the statutes come to addressing the preclusion of the Lanham Act claim at issue here is the pre-emption provision added to the FDCA in 1990 as part of the Nutrition Labeling and Education Act. See 21 U. S. C. §343–1. But, far from expressly precluding suits arising under other federal laws, the provision if anything suggests that Lanham Act suits are not precluded. This pre-emption provision forbids a “State or political subdivision of a State” from imposing requirements that are of the type but “not identical to” corresponding FDCA requirements for food and beverage labeling. Ibid. It is significant that the complex pre-emption provision distinguishes among different FDCA requirements. It forbids state-law requirements that are of the type but not identical to only certain FDCA provisions with respect to food and beverage labeling. See §§343–1(a)(1)–(5) (citing some but not all of the subsections of §343); §6, 104Stat. 2362–2364 (codified at 21 U. S. C. §343–1, and note following). Just as significant, the provision does not refer to requirements imposed by other sources of law, such as federal statutes. For purposes of deciding whether the FDCA displaces a regulatory or liability scheme in another statute, it makes a substantial difference whether that other statute is state or federal. By taking care to mandate express pre-emption of some state laws, Congress if anything indicated it did not intend the FDCA to preclude requirements arising from other sources. See Setser v. United States, 566 U. S. ___, ___ (2012) (slip op., at 6–7) (applying principle of expressio unius est exclusio alterius). Pre-emption of some state requirements does not suggest an intent to preclude federal claims. The structures of the FDCA and the Lanham Act reinforce the conclusion drawn from the text. When two statutes complement each other, it would show disregard for the congressional design to hold that Congress nonetheless intended one federal statute to preclude the operation of the other. See J. E. M. Ag Supply, Inc. v. Pioneer Hi-Bred Int’l, Inc., 534 U. S. 124, 144 (2001) (“[W]e can plainly regard each statute as effective because of its different requirements and protections”); see also Wyeth, supra, at 578–579. The Lanham Act and the FDCA complement each other in major respects, for each has its own scope and purpose. Although both statutes touch on food and beverage labeling, the Lanham Act protects commercial interests against unfair competition, while the FDCA protects public health and safety. Compare Lexmark, 572 U. S., at ___ (slip op., at 12–13), with 62 Cases of Jam, 340 U. S., at 596. The two statutes impose “different requirements and protections.” J. E. M. Ag Supply, supra, at 144. The two statutes complement each other with respect to remedies in a more fundamental respect. Enforcement of the FDCA and the detailed prescriptions of its implementing regulations is largely committed to the FDA. The FDA, however, does not have the same perspective or expertise in assessing market dynamics that day-to-day competitors possess. Competitors who manufacture or distribute products have detailed knowledge regarding how consumers rely upon certain sales and marketing strategies. Their awareness of unfair competition prac-tices may be far more immediate and accurate than that of agency rulemakers and regulators. Lanham Act suits draw upon this market expertise by empowering private parties to sue competitors to protect their interests on a case-by-case basis. By “serv[ing] a distinct compensatory function that may motivate injured persons to come forward,” Lanham Act suits, to the extent they touch on the same subject matter as the FDCA, “provide incentives” for manufacturers to behave well. See id., at 579. Allowing Lanham Act suits takes advantage of synergies among multiple methods of regulation. This is quite consistent with the congressional design to enact two different statutes, each with its own mechanisms to enhance the protection of competitors and consumers. A holding that the FDCA precludes Lanham Act claims challenging food and beverage labels would not only ignore the distinct functional aspects of the FDCA and the Lanham Act but also would lead to a result that Congress likely did not intend. Unlike other types of labels regu-lated by the FDA, such as drug labels, see 21 U. S. C. §355(d), it would appear the FDA does not preapprove food and beverage labels under its regulations and instead relies on enforcement actions, warning letters, and other measures. See Brief for United States as Amicus Curiae in Opposition 16. Because the FDA acknowledges that it does not necessarily pursue enforcement measures regarding all objectionable labels, ibid., if Lanham Act claims were to be precluded then commercial interests—and indirectly the public at large—could be left with less effective protection in the food and beverage labeling realm than in many other, less regulated industries. It is un-likely that Congress intended the FDCA’s protection of health and safety to result in less policing of misleading food and beverage labels than in competitive markets for other products. C Coca-Cola argues the FDCA precludes POM’s Lanham Act claim because Congress intended national uniformity in food and beverage labeling. Coca-Cola notes three aspects of the FDCA to support that position: delegation of enforcement authority to the Federal Government rather than private parties; express pre-emption with respect to state laws; and the specificity of the FDCA and its implementing regulations. But these details of the FDCA do not establish an intent or design to preclude Lanham Act claims. Coca-Cola says that the FDCA’s delegation of enforcement authority to the Federal Government shows Congress’ intent to achieve national uniformity in labeling. But POM seeks to enforce the Lanham Act, not the FDCA or its regulations. The centralization of FDCA enforcement authority in the Federal Government does not indicate that Congress intended to foreclose private enforcement of other federal statutes. Coca-Cola next appeals to the pre-emption provision added to the FDCA in 1990. See §343–1. It argues that allowing Lanham Act claims to proceed would undermine the pre-emption provision’s goal of ensuring that food and beverage manufacturers can market nationally without the burden of complying with a patchwork of requirements. A significant flaw in this argument is that the pre-emption provision by its plain terms applies only to certain state-law requirements, not to federal law. See Part II–B, supra. Coca-Cola in effect asks the Court to ignore the words “State or political subdivision of a State” in the statute. Even if it were proper to stray from the text in this way, it is far from clear that Coca-Cola’s assertions about national uniformity in fact reflect the congressional design. Although the application of a federal statute such as the Lanham Act by judges and juries in courts throughout the country may give rise to some variation in outcome, this is the means Congress chose to enforce a national policy to ensure fair competition. It is quite different from the disuniformity that would arise from the multitude of state laws, state regulations, state administrative agency rulings, and state-court decisions that are partially forbidden by the FDCA’s pre-emption provision. Congress not infrequently permits a certain amount of variability by authorizing a federal cause of action even in areas of law where national uniformity is important. Compare Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U. S. 141, 162 (1989) (“One of the fundamental purposes behind the Patent and Copyright Clauses of the Constitution was to promote national uniformity in the realm of intellectual property”), with 35 U. S. C. §281 (private right of action for patent infringement); see Wyeth, 555 U. S., at 570 (“[T]he [FDCA] contemplates that federal juries will resolve most misbranding claims”). The Lanham Act itself is an example of this design: Despite Coca-Cola’s protestations, the Act is uniform in extending its protection against unfair competition to the whole class it describes. It is variable only to the extent that those rights are enforced on a case-by-case basis. The variability about which Coca-Cola complains is no different than the variability that any industry covered by the Lanham Act faces. And, as noted, Lanham Act actions are a means to implement a uniform policy to prohibit unfair competition in all covered markets. Finally, Coca-Cola urges that the FDCA, and particu-larly its implementing regulations, addresses food and bev-erage labeling with much more specificity than is foundin the provisions of the Lanham Act. That is true. The pages of FDA rulemakings devoted only to juice-blend labeling attest to the level of detail with which the FDA has examined the subject. E.g., Food Labeling; Declaration of Ingredients; Common or Usual Name for Nonstandardized Foods; Diluted Juice Beverages, 58 Fed. Reg. 2897–2926 (1993). Because, as we have explained, the FDCA and the Lanham Act are complementary and have separate scopes and purposes, this greater specificity would matter only if the Lanham Act and the FDCA cannot be implemented in full at the same time. See RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 566 U. S. ___, ___ (2012) (slip op., at 5–7). But neither the statutory structure nor the empirical evidence of which the Court is aware indicates there will be any difficulty in fully enforcing each statute according to its terms. See Part II–B, supra. D The Government disagrees with both Coca-Cola and POM. It submits that a Lanham Act claim is precluded “to the extent the FDCA or FDA regulations specifically require or authorize the challenged aspects of [the] label.” Brief for United States as Amicus Curiae 11. Applying that standard, the Government argues that POM may not bring a Lanham Act challenge to the name of Coca-Cola’s product, but that other aspects of the label may be challenged. That is because, the Government argues, the FDA regulations specifically authorize the names of juice blends but not the other aspects of the label that are at issue. In addition to raising practical concerns about drawing a distinction between regulations that “specifically . . . authorize” a course of conduct and those that merely tolerate that course, id., at 10–11, the flaw in the Government’s intermediate position is the same as that in Coca-Cola’s theory of the case. The Government assumes that the FDCA and its regulations are at least in some circumstances a ceiling on the regulation of food and beverage labeling. But, as discussed above, Congress intended the Lanham Act and the FDCA to complement each other with respect to food and beverage labeling. The Government claims that the “FDA’s juice-naming regulation reflects the agency’s ‘weigh[ing of] the competing interests relevant to the particular requirement in question.’ ” Id., at 19 (quoting Medtronic, Inc. v. Lohr, 518 U. S. 470, 501 (1996) ). The rulemaking indeed does allude, at one point, to a balancing of interests: It styles a particular requirement as “provid[ing] manufacturers with flexibility for labeling products while providing consumers with information that they need.” 58 Fed. Reg. 2919–2920. But that rulemaking does not discuss or even cite the Lanham Act, and the Government cites no other statement in the rulemaking suggesting that the FDA considered the full scope of the interests the Lanham Act protects. In addition, and contrary to the language quoted above, the FDA explicitly encouraged manufacturers to include material on their labels that is not required by the regulations. Id., at 2919. A single isolated reference to a desire for flexibility is not sufficient to transform a rulemaking that is otherwise at best inconclusive as to its interaction with other federal laws into one with preclusive force, even on the assumption that a federal regulation in some instances might preclude application of a federal statute. Cf. Williamson v. Mazda Motor of Amer-ica, Inc., 562 U. S. ___, ___ (2011) (slip op., at 10–11). In addition, Geier v. American Honda Motor Co., 529 U. S. 861 (2000) , does not support the Government’s argument. In Geier, the agency enacted a regulation deliberately allowing manufacturers to choose between different options because the agency wanted to encourage diversity in the industry. A subsequent lawsuit challenged one of those choices. The Court concluded that the action was barred because it directly conflicted with the agency’s policy choice to encourage flexibility to foster innovation. Id., at 875. Here, by contrast, the FDA has not made a policy judgment that is inconsistent with POM’s Lanham Act suit. This is not a case where a lawsuit is undermining an agency judgment, and in any event the FDA does not have authority to enforce the Lanham Act. It is necessary to recognize the implications of the United States’ argument for preclusion. The Government asks the Court to preclude private parties from availing themselves of a well-established federal remedy because an agency enacted regulations that touch on similar subject matter but do not purport to displace that remedy or even implement the statute that is its source. Even if agency regulations with the force of law that purport to bar other legal remedies may do so, see id., at 874; see also Wyeth, 555 U. S., at 576, it is a bridge too far to accept an agency’s after-the-fact statement to justify that result here. An agency may not reorder federal statutory rights without congressional authorization. * * * Coca-Cola and the United States ask the Court to elevate the FDCA and the FDA’s regulations over the private cause of action authorized by the Lanham Act. But the FDCA and the Lanham Act complement each other in the federal regulation of misleading labels. Congress did not intend the FDCA to preclude Lanham Act suits like POM’s. The position Coca-Cola takes in this Court that because food and beverage labeling is involved it has no Lanham Act liability here for practices that allegedly mislead and trick consumers, all to the injury of competitors, finds no support in precedent or the statutes. The judgment of the Court of Appeals for the Ninth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Breyer took no part in the consideration or decision of this case.

572.US.393

A California Highway Patrol officer stopped the pickup truck occupied by petitioners because it matched the description of a vehicle that a 911 caller had recently reported as having run her off the road. As he and a second officer approached the truck, they smelled marijuana. They searched the truck’s bed, found 30 pounds of marijuana, and arrested petitioners. Petitioners moved to suppress the evidence, arguing that the traffic stop violated the Fourth Amendment. Their motion was denied, and they pleaded guilty to transporting marijuana. The California Court of Appeal affirmed, concluding that the officer had reasonable suspicion to conduct an investigative stop. Held: The traffic stop complied with the Fourth Amendment because, under the totality of the circumstances, the officer had reasonable suspicion that the truck’s driver was intoxicated. Pp. 3–11. (a) The Fourth Amendment permits brief investigative stops when an officer has “a particularized and objective basis for suspecting the particular person stopped of . . . criminal activity.” United States v. Cortez, 449 U.S. 411, 417–418. Reasonable suspicion takes into account “the totality of the circumstances,” id., at 417, and depends “upon both the content of information possessed by police and its degree of reliability,” Alabama v. White, 496 U.S. 325, 330. An anonymous tip alone seldom demonstrates sufficient reliability, White, 496 U. S., at 329, but may do so under appropriate circumstances, id., at 327. Pp. 3–5. (b) The 911 call in this case bore adequate indicia of reliability for the officer to credit the caller’s account. By reporting that she had been run off the road by a specific vehicle, the caller necessarily claimed an eyewitness basis of knowledge. The apparently short time between the reported incident and the 911 call suggests that the caller had little time to fabricate the report. And a reasonable officer could conclude that a false tipster would think twice before using the 911 system, which has several technological and regulatory features that safeguard against making false reports with immunity. Pp. 5–8. (c) Not only was the tip here reliable, but it also created reasonable suspicion of drunk driving. Running another car off the road suggests the sort of impairment that characterizes drunk driving. While that conduct might be explained by another cause such as driver distraction, reasonable suspicion “need not rule out the possibility of innocent conduct.” United States v. Arvizu, 534 U.S. 266, 277. Finally, the officer’s failure to observe additional suspicious conduct during the short period that he followed the truck did not dispel the reasonable suspicion of drunk driving, and the officer was not required to surveil the truck for a longer period. Pp. 8–10. Affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, and Alito, JJ., joined. Scalia, J., filed a dissenting opinion, in which Ginsburg, Sotomayor, and Kagan, JJ., joined.

After a 911 caller reported that a vehicle had run her off the road, a police officer located the vehicle she identified during the call and executed a traffic stop. We hold that the stop complied with the Fourth Amendment because, under the totality of the circumstances, the officer had reasonable suspicion that the driver was intoxicated. I On August 23, 2008, a Mendocino County 911 dispatch team for the California Highway Patrol (CHP) received a call from another CHP dispatcher in neighboring Humboldt County. The Humboldt County dispatcher relayed a tip from a 911 caller, which the Mendocino County team recorded as follows: “ ‘Showing southbound Highway 1 at mile marker 88, Silver Ford 150 pickup. Plate of 8-David-94925. Ran the reporting party off the roadway and was last seen approximately five [minutes] ago.’ ” App. 36a. The Mendocino County team then broadcast that information to CHP officers at 3:47 p.m. A CHP officer heading northbound toward the reported vehicle responded to the broadcast. At 4:00 p.m., the officer passed the truck near mile marker 69. At about 4:05 p.m., after making a U-turn, he pulled the truck over. A second officer, who had separately responded to the broadcast, also arrived on the scene. As the two officers approached the truck, they smelled marijuana. A search of the truck bed revealed 30 pounds of marijuana. The officers arrested the driver, petitioner Lorenzo Prado Navarette, and the passenger, petitioner José Prado Navarette. Petitioners moved to suppress the evidence, arguing that the traffic stop violated the Fourth Amendment because the officer lacked reasonable suspicion of criminal activity. Both the magistrate who presided over the suppression hearing and the Superior Court disagreed.[1] Petitioners pleaded guilty to transporting marijuana and were sentenced to 90 days in jail plus three years ofprobation. The California Court of Appeal affirmed, concluding that the officer had reasonable suspicion to conduct an investigative stop. 2012 WL 4842651 (Oct. 12, 2012). The court reasoned that the content of the tip indicated that it came from an eyewitness victim of reckless driving, and that the officer’s corroboration of the truck’s description, location, and direction established that the tip was reliable enough to justify a traffic stop. Id., at *7. Finally, the court concluded that the caller reported driving that was sufficiently dangerous to merit an investigative stop without waiting for the officer to observe additional reckless driving himself. Id., at *9. The California Supreme Court denied review. We granted certiorari, 570 U. S. ___ (2013), and now affirm. II The Fourth Amendment permits brief investigative stops—such as the traffic stop in this case—when a law enforcement officer has “a particularized and objective basis for suspecting the particular person stopped of criminal activity.” United States v. Cortez, 449 U. S. 411 –418 (1981); see also Terry v. Ohio, 392 U. S. 1 –22 (1968). The “reasonable suspicion” necessary to justify such a stop “is dependent upon both the content of information possessed by police and its degree of reliability.” Alabama v. White, 496 U. S. 325, 330 (1990) . The standard takes into account “the totality of the circumstances—the whole picture.” Cortez, supra, at 417. Although a mere “ ‘hunch’ ” does not create reasonable suspicion, Terry, supra, at 27, the level of suspicion the standard requires is “considerably less than proof of wrongdoing by a preponderance of the evidence,” and “obviously less” than is necessary for probable cause, United States v. Sokolow, 490 U. S. 1, 7 (1989) . A These principles apply with full force to investigative stops based on information from anonymous tips. We have firmly rejected the argument “that reasonable cause for a[n investigative stop] can only be based on the officer’s personal observation, rather than on information supplied by another person.” Adams v. Williams, 407 U. S. 143, 147 (1972) . Of course, “an anonymous tip alone seldom demonstrates the informant’s basis of knowledge or veracity.” White, 496 U. S., at 329 (emphasis added). That is because “ordinary citizens generally do not provide extensive recitations of the basis of their everyday observations,” and an anonymous tipster’s veracity is “ ‘by hypoth-esis largely unknown, and unknowable.’ ” Ibid. But under appropriate circumstances, an anonymous tip can demonstrate “sufficient indicia of reliability to provide reasonable suspicion to make [an] investigatory stop.” Id., at 327. Our decisions in Alabama v. White, 496 U. S. 325 (1990) , and Florida v. J. L., 529 U. S. 266 (2000) , are useful guides. In White, an anonymous tipster told the police that a woman would drive from a particular apartment building to a particular motel in a brown Plymouth station wagon with a broken right tail light. The tipster further asserted that the woman would be transporting cocaine. 496 U. S., at 327. After confirming the innocent details, officers stopped the station wagon as it neared the motel and found cocaine in the vehicle. Id., at 331. We held that the officers’ corroboration of certain details made the anonymous tip sufficiently reliable to create reasonable suspicion of criminal activity. By accurately predicting future behavior, the tipster demonstrated “a special familiarity with respondent’s affairs,” which in turn implied that the tipster had “access to reliable information about that individual’s illegal activities.” Id., at 332. We also recognized that an informant who is proved to tell the truth about some things is more likely to tell the truth about other things, “including the claim that the object of the tip is engaged in criminal activity.” Id., at 331 (citing Illinois v. Gates, 462 U. S. 213, 244 (1983) ). In J. L., by contrast, we determined that no reasonable suspicion arose from a bare-bones tip that a young black male in a plaid shirt standing at a bus stop was carrying a gun. 529 U. S., at 268. The tipster did not explain how he knew about the gun, nor did he suggest that he had any special familiarity with the young man’s affairs. Id., at 271. As a result, police had no basis for believing “that the tipster ha[d] knowledge of concealed criminal activity.” Id., at 272. Furthermore, the tip included no predictions of future behavior that could be corroborated to assess the tipster’s credibility. Id., at 271. We accordingly concluded that the tip was insufficiently reliable to justify a stop and frisk. B The initial question in this case is whether the 911 call was sufficiently reliable to credit the allegation that petitioners’ truck “ran the [caller] off the roadway.” Even assuming for present purposes that the 911 call was anonymous, see n. 1, supra, we conclude that the call bore adequate indicia of reliability for the officer to credit the caller’s account. The officer was therefore justified in proceeding from the premise that the truck had, in fact, caused the caller’s car to be dangerously diverted from the highway. By reporting that she had been run off the road by a specific vehicle—a silver Ford F-150 pickup, license plate 8D94925—the caller necessarily claimed eyewitness knowledge of the alleged dangerous driving. That basis of knowledge lends significant support to the tip’s reliability. See Gates, supra, at 234 (“[An informant’s] explicit and detailed description of alleged wrongdoing, along with a statement that the event was observed firsthand, entitles his tip to greater weight than might otherwise be the case”); Spinelli v. United States, 393 U. S. 410, 416 (1969) (a tip of illegal gambling is less reliable when “it is not alleged that the informant personally observed [the defendant] at work or that he had ever placed a bet with him”). This is in contrast to J. L., where the tip provided no basis for concluding that the tipster had actually seen the gun. 529 U. S., at 271. Even in White, where we upheld the stop, there was scant evidence that the tipster had actually observed cocaine in the station wagon. We called White a “ ‘close case’ ” because “[k]nowledge about a person’s future movements indicates some familiarity with that person’s affairs, but having such knowledge does not necessarily imply that the informant knows, in particular, whether that person is carrying hidden contraband.” 529 U. S., at 271. A driver’s claim that another vehicle ran her off the road, however, necessarily implies that the informant knows the other car was driven dangerously. There is also reason to think that the 911 caller in this case was telling the truth. Police confirmed the truck’s location near mile marker 69 (roughly 19 highway miles south of the location reported in the 911 call) at 4:00 p.m. (roughly 18 minutes after the 911 call). That timeline of events suggests that the caller reported the incident soon after she was run off the road. That sort of contemporaneous report has long been treated as especially reliable. In evidence law, we generally credit the proposition that statements about an event and made soon after perceiving that event are especially trustworthy because “substantial contemporaneity of event and statement negate the likelihood of deliberate or conscious misrepresentation.” Advisory Committee’s Notes on Fed. Rule Evid. 803(1), 28 U. S. C. App., p. 371 (describing the rationale for the hearsay exception for “present sense impression[s]”). A similar rationale applies to a “statement relating to a startling event”—such as getting run off the road—“made while the declarant was under the stress of excitement that it caused.” Fed. Rule Evid. 803(2) (hearsay exception for “excited utterances”). Unsurprisingly, 911 calls that would otherwise be inadmissible hearsay have often been admitted on those grounds. See D. Binder, Hearsay Handbook §8.1, pp. 257–259 (4th ed. 2013–2014) (citing cases admitting 911 calls as present sense impressions); id., §9.1, at 274–275 (911 calls admitted as excited utterances). There was no indication that the tip in J. L. (or even in White) was contemporaneous with the observation of criminal activity or made under the stress of excitement caused by a startling event, but those considerations weigh in favor of the caller’s veracity here. Another indicator of veracity is the caller’s use of the 911 emergency system. See Brief for Respondent 40–41, 44; Brief for United States as Amicus Curiae 16–18. A 911 call has some features that allow for identifying and tracing callers, and thus provide some safeguards against making false reports with immunity. See J. L., supra, at 276 (Kennedy, J., concurring). As this case illustrates, see n. 1, supra, 911 calls can be recorded, which provides victims with an opportunity to identify the false tipster’s voice and subject him to prosecution, see, e.g., Cal. Penal Code Ann. §653x (West 2010) (makes “telephon[ing] the 911 emergency line with the intent to annoy or harass” punishable by imprisonment and fine); see also §148.3 (2014 West Cum. Supp.) (prohibits falsely reporting “that an ‘emergency’ exists”); §148.5 (prohibits falsely reporting “that a felony or misdemeanor has been committed”). The 911 system also permits law enforcement to verify important information about the caller. In 1998, the Federal Communications Commission (FCC) began to require cellular carriers to relay the caller’s phone number to 911 dispatchers. 47 CFR §20.18(d)(1) (2013) (FCC’s “Phase I enhanced 911 services” requirements). Beginning in 2001, carriers have been required to identify the caller’s geographic location with increasing specificity. §§20.18(e)–(h) (“Phase II enhanced 911 service” requirements). And although callers may ordinarily block call recipients from obtaining their identifying information, FCC regulations exempt 911 calls from that privilege. §§64.1601(b), (d)(4)(ii) (“911 emergency services” exemption from rule that, when a caller so requests, “a carrier may not reveal that caller’s number or name”). None of this is to suggest that tips in 911 calls are per se reliable. Given the foregoing technological and regulatory developments, however, a reasonable officer could conclude that a false tipster would think twice before using such a system. The caller’s use of the 911 system is therefore one of the relevant circum-stances that, taken together, justified the officer’s reliance on the information reported in the 911 call. C Even a reliable tip will justify an investigative stop only if it creates reasonable suspicion that “criminal activity may be afoot.” Terry, 392 U. S., at 30. We must therefore determine whether the 911 caller’s report of being run off the roadway created reasonable suspicion of an ongoing crime such as drunk driving as opposed to an isolated episode of past recklessness. See Cortez, 449 U. S., at 417 (“An investigatory stop must be justified by some objective manifestation that the person stopped is, or is about to be, engaged in criminal activity”). We conclude that the behavior alleged by the 911 caller, “viewed from the standpoint of an objectively reasonable police officer, amount[s] to reasonable suspicion” of drunk driving. Ornelas v. United States, 517 U. S. 690, 696 (1996) . The stop was therefore proper.[2] Reasonable suspicion depends on “ ‘ “the factual and practical considerations of everyday life on which reason-able and prudent men, not legal technicians, act.” ’ ” Id., at 695. Under that commonsense approach, we can appropriately recognize certain driving behaviors as sound indicia of drunk driving. See, e.g., People v. Wells, 38 Cal. 4th 1078, 1081, 136 P. 3d 810, 811 (2006) (“ ‘weaving all over the roadway’ ”); State v. Prendergast, 103 Haw. 451, 452–453, 83 P. 3d 714, 715–716 (2004) (“cross[ing] over the center line” on a highway and “almost caus[ing] several head-on collisions”); State v. Golotta, 178 N. J. 205, 209, 837 A. 2d 359, 361 (2003) (driving “ ‘all overthe road’ ” and “ ‘weaving back and forth’ ”); State v. Walshire, 634 N. W. 2d 625, 626 (Iowa 2001) (“driving in the median”). Indeed, the accumulated experience of thousands of officers suggests that these sorts of erratic behaviors are strongly correlated with drunk driving. See Nat. Highway Traffic Safety Admin., The Visual Detection of DWI Motorists 4–5 (Mar. 2010), online at http://nhtsa.gov/staticfiles/nti/pdf/808677.pdf (as visited Apr. 18, 2014, and available in Clerk of Court’s case file). Of course, not all traffic infractions imply intoxication. Unconfirmed reports of driving without a seatbelt or slightly over the speed limit, for example, are so tenuously connected to drunk driving that a stop on those grounds alone would be constitutionally suspect. But a reliable tip alleging the dangerous behaviors discussed above gener-ally would justify a traffic stop on suspicion of drunk driving. The 911 caller in this case reported more than a minor traffic infraction and more than a conclusory allegation of drunk or reckless driving. Instead, she alleged a specific and dangerous result of the driver’s conduct: running another car off the highway. That conduct bears too great a resemblance to paradigmatic manifestations of drunk driving to be dismissed as an isolated example of recklessness. Running another vehicle off the road suggests lane-positioning problems, decreased vigilance, impaired judgment, or some combination of those recognized drunk driving cues. See Visual Detection of DWI Motorists 4–5. And the experience of many officers suggests that a driver who almost strikes a vehicle or another object—the exact scenario that ordinarily causes “running [another vehicle] off the roadway”—is likely intoxicated. See id., at 5, 8.As a result, we cannot say that the officer acted unreasonably under these circumstances in stopping a driverwhose alleged conduct was a significant indicator of drunk driving. Petitioners’ attempts to second-guess the officer’s rea-sonable suspicion of drunk driving are unavailing. It is true that the reported behavior might also be explained by, for example, a driver responding to “an unruly child or other distraction.” Brief for Petitioners 21. But we have consistently recognized that reasonable suspicion “need not rule out the possibility of innocent conduct.” United States v. Arvizu, 534 U. S. 266, 277 (2002) . Nor did the absence of additional suspicious conduct, after the vehicle was first spotted by an officer, dispel the reasonable suspicion of drunk driving. Brief for Petitioners 23–24. It is hardly surprising that the appearance of a marked police car would inspire more careful driving for a time. Cf. Arvizu, supra, at 275 (“ ‘[s]lowing down after spotting a law enforcement vehicle’ ” does not dispel reasonable suspicion of criminal activity). Extended observation of an allegedly drunk driver might eventually dispel a reasonable suspicion of intoxication, but the 5-minute period in this case hardly sufficed in that regard. Of course, an officer who already has such a reasonable suspicion need not surveil a vehicle at length in order to personally observe suspicious driving. See Adams v. Williams, 407 U. S., at 147 (repudiating the argument that “reasonable cause for a[n investigative stop] can only be based on the officer’s personal observation”). Once reasonable suspicion of drunk driving arises, “[t]he reasonableness of the officer’s decision to stop a suspect does not turn on the availability of less intrusive investigatory techniques.” Sokolow, 490 U. S., at 11. This would be a particularly inappropriate context to depart from that settled rule, because allowing a drunk driver a second chance for dangerous conduct could have disastrousconsequences. III Like White, this is a “close case.” 496 U. S., at 332. As in that case, the indicia of the 911 caller’s reliability here are stronger than those in J. L., where we held that a bare-bones tip was unreliable. 529 U. S., at 271. Although the indicia present here are different from those we found sufficient in White, there is more than one way to demonstrate “a particularized and objective basis for suspecting the particular person stopped of criminal activity.” Cortez, 449 U. S., at 417–418. Under the totality of the circumstances, we find the indicia of reliability in this case sufficient to provide the officer with reasonable suspicion that the driver of the reported vehicle had run another vehicle off the road. That made it reasonable under the circumstances for the officer to execute a traffic stop. We accordingly affirm. It is so ordered.Notes 1 At the suppression hearing, counsel for petitioners did not dispute that the reporting party identified herself by name in the 911 call recording. Because neither the caller nor the Humboldt County dispatcher who received the call was present at the hearing, however, the prosecution did not introduce the recording into evidence. The prosecution proceeded to treat the tip as anonymous, and the lower courts followed suit. See 2012 WL 4842651, *6 (Cal. Ct. App., Oct. 12, 2012). 2 Because we conclude that the 911 call created reasonable suspicion of an ongoing crime, we need not address under what circumstances a stop is justified by the need to investigate completed criminal activity. Cf. v. , .

573.US.373

In No. 13–132, petitioner Riley was stopped for a traffic violation, which eventually led to his arrest on weapons charges. An officer searching Riley incident to the arrest seized a cell phone from Riley’s pants pocket. The officer accessed information on the phone and noticed the repeated use of a term associated with a street gang. At the police station two hours later, a detective specializing in gangs further examined the phone’s digital contents. Based in part on photographs and videos that the detective found, the State charged Riley in connection with a shooting that had occurred a few weeks earlier and sought an enhanced sentence based on Riley’s gang membership. Riley moved to suppress all evidence that the police had obtained from his cell phone. The trial court denied the motion, and Riley was convicted. The California Court of Appeal affirmed. In No. 13–212, respondent Wurie was arrested after police observed him participate in an apparent drug sale. At the police station, the officers seized a cell phone from Wurie’s person and noticed that the phone was receiving multiple calls from a source identified as “my house” on its external screen. The officers opened the phone, accessed its call log, determined the number associated with the “my house” label, and traced that number to what they suspected was Wurie’s apartment. They secured a search warrant and found drugs, a firearm and ammunition, and cash in the ensuing search. Wurie was then charged with drug and firearm offenses. He moved to suppress the evidence obtained from the search of the apartment. The District Court denied the motion, and Wurie was convicted. The First Circuit reversed the denial of the motion to suppress and vacated the relevant convictions. Held: The police generally may not, without a warrant, search digital information on a cell phone seized from an individual who has been arrested. Pp. 5–28. (a) A warrantless search is reasonable only if it falls within a specific exception to the Fourth Amendment’s warrant requirement. See Kentucky v. King, 563 U. S. ___, ___. The well-established exception at issue here applies when a warrantless search is conducted incident to a lawful arrest. Three related precedents govern the extent to which officers may search property found on or near an arrestee. Chimel v. California, 395 U.S. 752, requires that a search incident to arrest be limited to the area within the arrestee’s immediate control, where it is justified by the interests in officer safety and in preventing evidence destruction. In United States v. Robinson, 414 U.S. 218, the Court applied the Chimel analysis to a search of a cigarette pack found on the arrestee’s person. It held that the risks identified in Chimel are present in all custodial arrests, 414 U. S., at 235, even when there is no specific concern about the loss of evidence or the threat to officers in a particular case, id., at 236. The trilogy concludes with Arizona v. Gant, 556 U.S. 332, which permits searches of a car where the arrestee is unsecured and within reaching distance of the passenger compartment, or where it is reasonable to believe that evidence of the crime of arrest might be found in the vehicle, id., at 343. Pp. 5–8. (b) The Court declines to extend Robinson’s categorical rule to searches of data stored on cell phones. Absent more precise guidance from the founding era, the Court generally determines whether to exempt a given type of search from the warrant requirement “by assessing, on the one hand, the degree to which it intrudes upon an individual’s privacy and, on the other, the degree to which it is needed for the promotion of legitimate governmental interests.” Wyoming v. Houghton, 526 U.S. 295, 300. That balance of interests supported the search incident to arrest exception in Robinson. But a search of digital information on a cell phone does not further the government interests identified in Chimel, and implicates substantially greater individual privacy interests than a brief physical search. Pp. 8–22. (1) The digital data stored on cell phones does not present either Chimel risk. Pp. 10–15. (i) Digital data stored on a cell phone cannot itself be used as a weapon to harm an arresting officer or to effectuate the arrestee’s escape. Officers may examine the phone’s physical aspects to ensure that it will not be used as a weapon, but the data on the phone can endanger no one. To the extent that a search of cell phone data might warn officers of an impending danger, e.g., that the arrestee’s confederates are headed to the scene, such a concern is better addressed through consideration of case-specific exceptions to the warrant requirement, such as exigent circumstances. See, e.g., Warden, Md. Penitentiary v. Hayden, 387 U.S. 294, 298–299. Pp. 10–12. (ii) The United States and California raise concerns about the destruction of evidence, arguing that, even if the cell phone is physically secure, information on the cell phone remains vulnerable to remote wiping and data encryption. As an initial matter, those broad concerns are distinct from Chimel’s focus on a defendant who responds to arrest by trying to conceal or destroy evidence within his reach. The briefing also gives little indication that either problem is prevalent or that the opportunity to perform a search incident to arrest would be an effective solution. And, at least as to remote wiping, law enforcement currently has some technologies of its own for combatting the loss of evidence. Finally, law enforcement’s remaining concerns in a particular case might be addressed by responding in a targeted manner to urgent threats of remote wiping, see Missouri v. McNeely, 569 U. S. ___, ___, or by taking action to disable a phone’s locking mechanism in order to secure the scene, see Illinois v. McArthur, 531 U.S. 326, 331–333. Pp. 12–15. (2) A conclusion that inspecting the contents of an arrestee’s pockets works no substantial additional intrusion on privacy beyond the arrest itself may make sense as applied to physical items, but more substantial privacy interests are at stake when digital data is involved. Pp. 15–22. (i) Cell phones differ in both a quantitative and a qualitative sense from other objects that might be carried on an arrestee’s person. Notably, modern cell phones have an immense storage capacity. Before cell phones, a search of a person was limited by physical realities and generally constituted only a narrow intrusion on privacy. But cell phones can store millions of pages of text, thousands of pictures, or hundreds of videos. This has several interrelated privacy consequences. First, a cell phone collects in one place many distinct types of information that reveal much more in combination than any isolated record. Second, the phone’s capacity allows even just one type of information to convey far more than previously possible. Third, data on the phone can date back for years. In addition, an element of pervasiveness characterizes cell phones but not physical records. A decade ago officers might have occasionally stumbled across a highly personal item such as a diary, but today many of the more than 90% of American adults who own cell phones keep on their person a digital record of nearly every aspect of their lives. Pp. 17–21. (ii) The scope of the privacy interests at stake is further complicated by the fact that the data viewed on many modern cell phones may in fact be stored on a remote server. Thus, a search may extend well beyond papers and effects in the physical proximity of an arrestee, a concern that the United States recognizes but cannot definitively foreclose. Pp. 21–22. (c) Fallback options offered by the United States and California are flawed and contravene this Court’s general preference to provide clear guidance to law enforcement through categorical rules. See Michigan v. Summers, 452 U.S. 692, 705, n. 19. One possible rule is to import the Gant standard from the vehicle context and allow a warrantless search of an arrestee’s cell phone whenever it is reasonable to believe that the phone contains evidence of the crime of arrest. That proposal is not appropriate in this context, and would prove no practical limit at all when it comes to cell phone searches. Another possible rule is to restrict the scope of a cell phone search to information relevant to the crime, the arrestee’s identity, or officer safety. That proposal would again impose few meaningful constraints on officers. Finally, California suggests an analogue rule, under which officers could search cell phone data if they could have obtained the same information from a pre-digital counterpart. That proposal would allow law enforcement to search a broad range of items contained on a phone even though people would be unlikely to carry such a variety of information in physical form, and would launch courts on a difficult line-drawing expedition to determine which digital files are comparable to physical records. Pp. 22–25. (d) It is true that this decision will have some impact on the ability of law enforcement to combat crime. But the Court’s holding is not that the information on a cell phone is immune from search; it is that a warrant is generally required before a search. The warrant requirement is an important component of the Court’s Fourth Amendment jurisprudence, and warrants may be obtained with increasing efficiency. In addition, although the search incident to arrest exception does not apply to cell phones, the continued availability of the exigent circumstances exception may give law enforcement a justification for a warrantless search in particular cases. Pp. 25–27. No. 13–132, reversed and remanded; No. 13–212, 728 F.3d 1, affirmed. Roberts, C. J., delivered the opinion of the Court, in which Scalia, Kennedy, Thomas, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in part and concurring in the judgment. Notes 1 Together with No. 13–212, United States v. Wurie, on certiorari to the United States Court of Appeals for the First Circuit.

These two cases raise a common question: whether the police may, without a warrant, search digital information on a cell phone seized from an individual who has been arrested. I A In the first case, petitioner David Riley was stopped by a police officer for driving with expired registration tags. In the course of the stop, the officer also learned that Riley’s license had been suspended. The officer impounded Riley’s car, pursuant to department policy, and another officer conducted an inventory search of the car. Riley was arrested for possession of concealed and loaded firearms when that search turned up two handguns under the car’s hood. See Cal. Penal Code Ann. §§12025(a)(1), 12031(a)(1) (West 2009). An officer searched Riley incident to the arrest and found items associated with the “Bloods” street gang. He also seized a cell phone from Riley’s pants pocket. According to Riley’s uncontradicted assertion, the phone was a “smart phone,” a cell phone with a broad range of other functions based on advanced computing capability, large storage capacity, and Internet connectivity. The officer accessed information on the phone and noticed that some words (presumably in text messages or a contacts list) were preceded by the letters “CK”—a label that, he believed, stood for “Crip Killers,” a slang term for members of the Bloods gang. At the police station about two hours after the arrest, a detective specializing in gangs further examined the contents of the phone. The detective testified that he “went through” Riley’s phone “looking for evidence, because . . . gang members will often video themselves with guns or take pictures of themselves with the guns.” App. in No. 13–132, p. 20. Although there was “a lot of stuff” on the phone, particular files that “caught [the detective’s] eye” included videos of young men sparring while someone yelled encouragement using the moniker “Blood.” Id., at 11–13. The police also found photographs of Riley standing in front of a car they suspected had been involved in a shooting a few weeks earlier. Riley was ultimately charged, in connection with that earlier shooting, with firing at an occupied vehicle, assault with a semiautomatic firearm, and attempted murder. The State alleged that Riley had committed those crimes for the benefit of a criminal street gang, an aggravating factor that carries an enhanced sentence. Compare Cal. Penal Code Ann. §246 (2008) with §186.22(b)(4)(B) (2014). Prior to trial, Riley moved to suppress all evidence that the police had obtained from his cell phone. He contended that the searches of his phone violated the Fourth Amendment, because they had been performed without a warrant and were not otherwise justified by exigent circumstances. The trial court rejected that argument. App. in No. 13–132, at 24, 26. At Riley’s trial, police officers testified about the photographs and videos found on the phone, and some of the photographs were admitted into evidence. Riley was convicted on all three counts and received an enhanced sentence of 15 years to life in prison. The California Court of Appeal affirmed. No. D059840 (Cal. App., Feb. 8, 2013), App. to Pet. for Cert. in No. 13–132, pp. 1a–23a. The court relied on the California Supreme Court’s decision in People v. Diaz, 51 Cal. 4th 84, 244 P. 3d 501 (2011), which held that the Fourth Amendment permits a warrantless search of cell phone data incident to an arrest, so long as the cell phone was immediately associated with the arrestee’s person. See id., at 93, 244 P. 3d, at 505–506. The California Supreme Court denied Riley’s petition for review, App. to Pet. for Cert. in No. 13–132, at 24a, and we granted certiorari, 571 U. S. ___ (2014). B In the second case, a police officer performing routine surveillance observed respondent Brima Wurie make an apparent drug sale from a car. Officers subsequently arrested Wurie and took him to the police station. At the station, the officers seized two cell phones from Wurie’s person. The one at issue here was a “flip phone,” a kind of phone that is flipped open for use and that generally has a smaller range of features than a smart phone. Five to ten minutes after arriving at the station, the officers noticed that the phone was repeatedly receiving calls from a source identified as “my house” on the phone’s external screen. A few minutes later, they opened the phone and saw a photograph of a woman and a baby set as the phone’s wallpaper. They pressed one button on the phone to access its call log, then another button to determine the phone number associated with the “my house” label. They next used an online phone directory to trace that phone number to an apartment building. When the officers went to the building, they saw Wurie’s name on a mailbox and observed through a window a woman who resembled the woman in the photograph on Wurie’s phone. They secured the apartment while obtaining a search warrant and, upon later executing the warrant, found and seized 215 grams of crack cocaine, mari-juana, drug paraphernalia, a firearm and ammunition, and cash. Wurie was charged with distributing crack cocaine, possessing crack cocaine with intent to distribute, and being a felon in possession of a firearm and ammunition. See 18 U. S. C. §922(g); 21 U. S. C. §841(a). He moved to suppress the evidence obtained from the search of the apartment, arguing that it was the fruit of an unconstitutional search of his cell phone. The District Court denied the motion. 612 F. Supp. 2d 104 (Mass. 2009). Wurie was convicted on all three counts and sentenced to 262 months in prison. A divided panel of the First Circuit reversed the denial of Wurie’s motion to suppress and vacated Wurie’s convictions for possession with intent to distribute and possession of a firearm as a felon. 728 F. 3d 1 (2013). The court held that cell phones are distinct from other physical possessions that may be searched incident to arrest without a warrant, because of the amount of personal data cell phones contain and the negligible threat they pose to law enforcement interests. See id., at 8–11. We granted certiorari. 571 U. S. ___ (2014). II The Fourth Amendment provides: “The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.” As the text makes clear, “the ultimate touchstone of the Fourth Amendment is ‘reasonableness.’ ” Brigham City v. Stuart, 547 U. S. 398, 403 (2006) . Our cases have determined that “[w]here a search is undertaken by law enforcement officials to discover evidence of criminal wrongdoing, . . . reasonableness generally requires the obtaining of a judicial warrant.” Vernonia School Dist. 47J v. Acton, 515 U. S. 646, 653 (1995) . Such a warrant ensures that the inferences to support a search are “drawn by a neutral and detached magistrate instead of being judged by the officer engaged in the often competitive enterprise of ferreting out crime.” Johnson v. United States, 333 U. S. 10, 14 (1948) . In the absence of a warrant, a search is reasonable only if it falls within a specific exception to the warrant requirement. See Kentucky v. King, 563 U. S. ___, ___ (2011) (slip op., at 5–6). The two cases before us concern the reasonableness of a warrantless search incident to a lawful arrest. In 1914, this Court first acknowledged in dictum “the right on the part of the Government, always recognized under English and American law, to search the person of the accused when legally arrested to discover and seize the fruits or evidences of crime.” Weeks v. United States, 232 U. S. 383 . Since that time, it has been well accepted that such a search constitutes an exception to the warrant requirement. Indeed, the label “exception” is something of a misnomer in this context, as warrantless searches incident to arrest occur with far greater frequency than searches conducted pursuant to a warrant. See 3 W. LaFave, Search and Seizure §5.2(b), p. 132, and n. 15 (5th ed. 2012). Although the existence of the exception for such searches has been recognized for a century, its scope has been de-bated for nearly as long. See Arizona v. Gant, 556 U. S. 332, 350 (2009) (noting the exception’s “checkered his-tory”). That debate has focused on the extent to which officers may search property found on or near the arrestee. Three related precedents set forth the rules governing such searches: The first, Chimel v. California, 395 U. S. 752 (1969) , laid the groundwork for most of the existing search incident to arrest doctrine. Police officers in that case arrested Chimel inside his home and proceeded to search his entire three-bedroom house, including the attic and garage. In particular rooms, they also looked through the contents of drawers. Id., at 753–754. The Court crafted the following rule for assessing the reasonableness of a search incident to arrest: “When an arrest is made, it is reasonable for the arresting officer to search the person arrested in order to remove any weapons that the latter might seek to use in order to resist arrest or effect his escape. Otherwise, the officer’s safety might well be endangered, and the arrest itself frustrated. In addition, it is entirely reasonable for the arresting officer to search for and seize any evidence on the arrestee’s person in order to prevent its concealment or destruction. . . . There is ample justification, therefore, for a search of the arrestee’s person and the area ‘within his immediate control’—construing that phrase to mean the area from within which he might gain possession of a weapon or destructible evidence.” Id., at 762–763. The extensive warrantless search of Chimel’s home did not fit within this exception, because it was not needed to protect officer safety or to preserve evidence. Id., at 763, 768. Four years later, in United States v. Robinson, 414 U. S. 218 (1973) , the Court applied the Chimel analysis in the context of a search of the arrestee’s person. A police officer had arrested Robinson for driving with a revoked license. The officer conducted a patdown search and felt an object that he could not identify in Robinson’s coat pocket. He removed the object, which turned out to be a crumpled cigarette package, and opened it. Inside were 14 capsules of heroin. Id., at 220, 223. The Court of Appeals concluded that the search was unreasonable because Robinson was unlikely to have evidence of the crime of arrest on his person, and because it believed that extracting the cigarette package and opening it could not be justified as part of a protective search for weapons. This Court reversed, rejecting the notion that “case-by-case adjudication” was required to determine “whether or not there was present one of the reasons supporting the authority for a search of the person incident to a lawful arrest.” Id., at 235. As the Court explained, “[t]he authority to search the person incident to a lawful custodial arrest, while based upon the need to disarm and to discover evidence, does not depend on what a court may later decide was the probability in a particular arrest situation that weapons or evidence would in fact be found upon the person of the suspect.” Ibid. Instead, a “custodial arrest of a suspect based on probable cause is a reasonable intrusion under the Fourth Amendment; that intrusion being lawful, a search incident to the arrest requires no additional justification.” Ibid. The Court thus concluded that the search of Robinson was reasonable even though there was no concern about the loss of evidence, and the arresting officer had no specific concern that Robinson might be armed. Id., at 236. In doing so, the Court did not draw a line between a search of Robinson’s person and a further examination of the cigarette pack found during that search. It merely noted that, “[h]aving in the course of a lawful search come upon the crumpled package of cigarettes, [the officer] was entitled to inspect it.” Ibid. A few years later, the Court clarified that this exception was limited to “personal property . . . immediately associated with the person of the arrestee.” United States v. Chadwick, 433 U. S. 1, 15 (1977) (200-pound, locked footlocker could not be searched incident to arrest), abrogated on other grounds by California v. Acevedo, 500 U. S. 565 (1991) . The search incident to arrest trilogy concludes with Gant, which analyzed searches of an arrestee’s vehicle. Gant, like Robinson, recognized that the Chimel concerns for officer safety and evidence preservation underlie the search incident to arrest exception. See 556 U. S., at 338. As a result, the Court concluded that Chimel could authorize police to search a vehicle “only when the arrestee is unsecured and within reaching distance of the passenger compartment at the time of the search.” 556 U. S., at 343. Gant added, however, an independent exception for a warrantless search of a vehicle’s passenger compartment “when it is ‘reasonable to believe evidence relevant to the crime of arrest might be found in the vehicle.’ ” Ibid. (quoting Thornton v. United States, 541 U. S. 615, 632 (2004) (Scalia, J., concurring in judgment)). That exception stems not from Chimel, the Court explained, but from “circumstances unique to the vehicle context.” 556 U. S., at 343. III These cases require us to decide how the search incident to arrest doctrine applies to modern cell phones, which are now such a pervasive and insistent part of daily life that the proverbial visitor from Mars might conclude they were an important feature of human anatomy. A smart phone of the sort taken from Riley was unheard of ten years ago; a significant majority of American adults now own such phones. See A. Smith, Pew Research Center, Smartphone Ownership—2013 Update (June 5, 2013). Even less sophisticated phones like Wurie’s, which have already faded in popularity since Wurie was arrested in 2007, have been around for less than 15 years. Both phones are based on technology nearly inconceivable just a few decades ago, when Chimel and Robinson were decided. Absent more precise guidance from the founding era, we generally determine whether to exempt a given type of search from the warrant requirement “by assessing, on the one hand, the degree to which it intrudes upon an individual’s privacy and, on the other, the degree to which it is needed for the promotion of legitimate governmental interests.” Wyoming v. Houghton, 526 U. S. 295, 300 (1999) . Such a balancing of interests supported the search incident to arrest exception in Robinson, and a mechanical application of Robinson might well support the warrantless searches at issue here. But while Robinson’s categorical rule strikes the appropriate balance in the context of physical objects, neither of its rationales has much force with respect to digital content on cell phones. On the government interest side, Robinson concluded that the two risks identified in Chimel—harm to officers and destruction of evidence—are present in all custodial arrests. There are no comparable risks when the search is of digital data. In addition, Robinson regarded any privacy interests retained by an individual after arrest as significantly diminished by the fact of the arrest itself. Cell phones, however, place vast quantities of personal information literally in the hands of individuals. A search of the information on a cell phone bears little resemblance to the type of brief physical search considered in Robinson. We therefore decline to extend Robinson to searches of data on cell phones, and hold instead that officers must generally secure a warrant before conducting such a search. A We first consider each Chimel concern in turn. In doing so, we do not overlook Robinson’s admonition that searches of a person incident to arrest, “while based upon theneed to disarm and to discover evidence,” are reasonable regardless of “the probability in a particular arrest situation that weapons or evidence would in fact be found.” 414 U. S., at 235. Rather than requiring the “case-by-case adjudication” that Robinson rejected, ibid., we ask instead whether application of the search incident to arrest doctrine to this particular category of effects would “untether the rule from the justifications underlying the Chimel exception,” Gant, supra, at 343. See also Knowles v. Iowa, 525 U. S. 113, 119 (1998) (declining to extend Robinson to the issuance of citations, “a situation where the concern for officer safety is not present to the same extent and the concern for destruction or loss of evidence is not present at all”). 1 Digital data stored on a cell phone cannot itself be used as a weapon to harm an arresting officer or to effectuate the arrestee’s escape. Law enforcement officers remain free to examine the physical aspects of a phone to ensure that it will not be used as a weapon—say, to determine whether there is a razor blade hidden between the phone and its case. Once an officer has secured a phone and eliminated any potential physical threats, however, data on the phone can endanger no one. Perhaps the same might have been said of the cigarette pack seized from Robinson’s pocket. Once an officer gained control of the pack, it was unlikely that Robinson could have accessed the pack’s contents. But unknown physical objects may always pose risks, no matter how slight, during the tense atmosphere of a custodial arrest. The officer in Robinson testified that he could not identify the objects in the cigarette pack but knew they were not cigarettes. See 414 U. S., at 223, 236, n. 7. Given that, a further search was a reasonable protective measure. No such unknowns exist with respect to digital data. As the First Circuit explained, the officers who searched Wurie’s cell phone “knew exactly what they would find therein: data. They also knew that the data could not harm them.” 728 F. 3d, at 10. The United States and California both suggest that a search of cell phone data might help ensure officer safety in more indirect ways, for example by alerting officers that confederates of the arrestee are headed to the scene. There is undoubtedly a strong government interest in warning officers about such possibilities, but neither the United States nor California offers evidence to suggest that their concerns are based on actual experience. The proposed consideration would also represent a broadening of Chimel’s concern that an arrestee himself might grab a weapon and use it against an officer “to resist arrest or effect his escape.” 395 U. S., at 763. And any such threats from outside the arrest scene do not “lurk[ ] in all custodial arrests.” Chadwick, 433 U. S., at 14–15. Accordingly, the interest in protecting officer safety does not justify dispensing with the warrant requirement across the board. To the extent dangers to arresting officers may be implicated in a particular way in a particular case, they are better addressed through consideration of case-specific exceptions to the warrant requirement, such as the one for exigent circumstances. See, e.g., Warden, Md. Penitentiary v. Hayden, 387 U. S. 294 –299 (1967) (“The Fourth Amendment does not require police officers to delay in the course of an investigation if to do so would gravely endanger their lives or the lives of others.”). 2 The United States and California focus primarily on the second Chimel rationale: preventing the destruction of evidence. Both Riley and Wurie concede that officers could have seized and secured their cell phones to prevent destruction of evidence while seeking a warrant. See Brief for Petitioner in No. 13–132, p. 20; Brief for Respondent in No. 13–212, p. 41. That is a sensible concession. See Illinois v. McArthur, 531 U. S. 326 –333 (2001); Chadwick, supra, at 13, and n. 8. And once law enforcement officers have secured a cell phone, there is no longer any risk that the arrestee himself will be able to delete incriminating data from the phone. The United States and California argue that information on a cell phone may nevertheless be vulnerable to two types of evidence destruction unique to digital data—remote wiping and data encryption. Remote wiping occurs when a phone, connected to a wireless network, receives a signal that erases stored data. This can happen when a third party sends a remote signal or when a phone is preprogrammed to delete data upon entering or leaving certain geographic areas (so-called “geofencing”). See Dept. of Commerce, National Institute of Standards and Technology, R. Ayers, S. Brothers, & W. Jansen, Guidelines on Mobile Device Forensics (Draft) 29, 31 (SP 800–101 Rev. 1, Sept. 2013) (hereinafter Ayers). Encryption is a security feature that some modern cell phones use in addition to password protection. When such phones lock, data becomes protected by sophisticated encryption that renders a phone all but “unbreakable” unless police know the password. Brief for United States as Amicus Curiae in No. 13–132, p. 11. As an initial matter, these broader concerns about the loss of evidence are distinct from Chimel’s focus on a defendant who responds to arrest by trying to conceal or destroy evidence within his reach. See 395 U. S., at 763–764. With respect to remote wiping, the Government’s primary concern turns on the actions of third parties who are not present at the scene of arrest. And data encryption is even further afield. There, the Government focuses on the ordinary operation of a phone’s security features, apart from any active attempt by a defendant or his associates to conceal or destroy evidence upon arrest. We have also been given little reason to believe that either problem is prevalent. The briefing reveals only a couple of anecdotal examples of remote wiping triggered by an arrest. See Brief for Association of State Criminal Investigative Agencies et al. as Amici Curiae in No. 13–132, pp. 9–10; see also Tr. of Oral Arg. in No. 13–132,p. 48. Similarly, the opportunities for officers to search a password-protected phone before data becomes encrypted are quite limited. Law enforcement officers are very unlikely to come upon such a phone in an unlocked state because most phones lock at the touch of a button or, as a default, after some very short period of inactivity. See, e.g., iPhone User Guide for iOS 7.1 Software 10 (2014) (default lock after about one minute). This may explain why the encryption argument was not made until the merits stage in this Court, and has never been considered by the Courts of Appeals. Moreover, in situations in which an arrest might trigger a remote-wipe attempt or an officer discovers an unlocked phone, it is not clear that the ability to conduct a warrantless search would make much of a difference. The need to effect the arrest, secure the scene, and tend to other press-ing matters means that law enforcement officers may well not be able to turn their attention to a cell phone right away. See Tr. of Oral Arg. in No. 13–132, at 50; see also Brief for United States as Amicus Curiae in No. 13–132, at 19. Cell phone data would be vulnerable to remote wiping from the time an individual anticipates arrest to the time any eventual search of the phone is completed, which might be at the station house hours later. Likewise, an officer who seizes a phone in an unlocked state might not be able to begin his search in the short time remaining before the phone locks and data becomes encrypted. In any event, as to remote wiping, law enforcement is not without specific means to address the threat. Remote wiping can be fully prevented by disconnecting a phone from the network. There are at least two simple ways to do this: First, law enforcement officers can turn the phone off or remove its battery. Second, if they are concerned about encryption or other potential problems, they can leave a phone powered on and place it in an enclosure that isolates the phone from radio waves. See Ayers 30–31. Such devices are commonly called “Faraday bags,” after the English scientist Michael Faraday. They are essentially sandwich bags made of aluminum foil: cheap, lightweight, and easy to use. See Brief for Criminal Law Professors as Amici Curiae 9. They may not be a complete answer to the problem, see Ayers 32, but at least for now they provide a reasonable response. In fact, a number of law enforcement agencies around the country already encourage the use of Faraday bags. See, e.g., Dept. of Justice, National Institute of Justice, Electronic Crime Scene Investigation: A Guide for First Responders 14, 32 (2d ed. Apr. 2008); Brief for Criminal Law Professors as Amici Curiae 4–6. To the extent that law enforcement still has specific concerns about the potential loss of evidence in a particular case, there remain more targeted ways to address those concerns. If “the police are truly confronted with a ‘now or never’ situation,”—for example, circumstances suggesting that a defendant’s phone will be the target of an imminent remote-wipe attempt—they may be able to rely on exigent circumstances to search the phone immediately. Missouri v. McNeely, 569 U. S. ___, ___ (2013) (slip op., at 10) (quoting Roaden v. Kentucky, 413 U. S. 496, 505 (1973) ; some internal quotation marks omitted). Or, if officers happen to seize a phone in an unlocked state, they may be able to disable a phone’s automatic-lock feature in order to prevent the phone from locking and encrypting data. See App. to Reply Brief in No. 13–132, p. 3a (diagramming the few necessary steps). Such a preventive measure could be analyzed under the principles set forth in our decision in McArthur, 531 U. S. 326 , which approved officers’ reasonable steps to secure a scene to preserve evidence while they awaited a warrant. See id., at 331–333. B The search incident to arrest exception rests not only on the heightened government interests at stake in a volatile arrest situation, but also on an arrestee’s reduced privacy interests upon being taken into police custody. Robinson focused primarily on the first of those rationales. But it also quoted with approval then-Judge Cardozo’s account of the historical basis for the search incident to arrest exception: “Search of the person becomes lawful when grounds for arrest and accusation have been discovered, and the law is in the act of subjecting the body of the accused to its physical dominion.” 414 U. S., at 232 (quoting People v. Chiagles, 237 N. Y. 193, 197, 142 N. E. 583, 584 (1923)); see also 414 U. S., at 237 (Powell, J., concurring) (“an individual lawfully subjected to a custodial arrest retains no significant Fourth Amendment interest in the privacy of his person”). Put simply, a patdown of Robinson’s cloth-ing and an inspection of the cigarette pack found in his pocket constituted only minor additional intrusions compared to the substantial government authority exercised in taking Robinson into custody. See Chadwick, 433 U. S., at 16, n. 10 (searches of a person are justified in part by “reduced expectations of privacy caused by the arrest”). The fact that an arrestee has diminished privacy interests does not mean that the Fourth Amendment falls out of the picture entirely. Not every search “is acceptable solely because a person is in custody.” Maryland v. King, 569 U. S. ___, ___ (2013) (slip op., at 26). To the contrary, when “privacy-related concerns are weighty enough” a “search may require a warrant, notwithstanding the diminished expectations of privacy of the arrestee.” Ibid. One such example, of course, is Chimel. Chimel refused to “characteriz[e] the invasion of privacy that results from a top-to-bottom search of a man’s house as ‘minor.’ ” 395 U. S., at 766–767, n. 12. Because a search of the arrestee’s entire house was a substantial invasion beyond the arrest itself, the Court concluded that a warrant was required. Robinson is the only decision from this Court applying Chimel to a search of the contents of an item found on an arrestee’s person. In an earlier case, this Court had approved a search of a zipper bag carried by an arrestee, but the Court analyzed only the validity of the arrest itself. See Draper v. United States, 358 U. S. 307 –311 (1959). Lower courts applying Robinson and Chimel, however, have approved searches of a variety of personal items carried by an arrestee. See, e.g., United States v. Carrion, 809 F. 2d 1120, 1123, 1128 (CA5 1987) (billfold and address book); United States v. Watson, 669 F. 2d 1374, 1383–1384 (CA11 1982) (wallet); United States v. Lee, 501 F. 2d 890, 892 (CADC 1974) (purse). The United States asserts that a search of all data stored on a cell phone is “materially indistinguishable” from searches of these sorts of physical items. Brief for United States in No. 13–212, p. 26. That is like saying a ride on horseback is materially indistinguishable from a flight to the moon. Both are ways of getting from point A to point B, but little else justifies lumping them together. Modern cell phones, as a category, implicate privacy concerns far beyond those implicated by the search of a cigarette pack, a wallet, or a purse. A conclusion that inspecting the contents of an arrestee’s pockets works no substantial additional intrusion on privacy beyond the arrest itself may make sense as applied to physical items, but any extension of that reasoning to digital data has to rest on its own bottom. 1 Cell phones differ in both a quantitative and a qualitative sense from other objects that might be kept on an arrestee’s person. The term “cell phone” is itself misleading shorthand; many of these devices are in fact minicomputers that also happen to have the capacity to be used as a telephone. They could just as easily be called cameras, video players, rolodexes, calendars, tape recorders, libraries, diaries, albums, televisions, maps, or newspapers. One of the most notable distinguishing features of modern cell phones is their immense storage capacity. Before cell phones, a search of a person was limited by physical realities and tended as a general matter to constitute only a narrow intrusion on privacy. See Kerr, Foreword: Accounting for Technological Change, 36 Harv. J. L. & Pub. Pol’y 403, 404–405 (2013). Most people cannot lug around every piece of mail they have received for the past several months, every picture they have taken, or every book or article they have read—nor would they have any reason to attempt to do so. And if they did, they would have to drag behind them a trunk of the sort held to require a search warrant in Chadwick, supra, rather than a container the size of the cigarette package in Robinson. But the possible intrusion on privacy is not physically limited in the same way when it comes to cell phones. The current top-selling smart phone has a standard capacity of 16 gigabytes (and is available with up to 64 gigabytes). Sixteen gigabytes translates to millions of pages of text, thousands of pictures, or hundreds of videos. See Kerr, supra, at 404; Brief for Center for Democracy & Technol-ogy et al. as Amici Curiae 7–8. Cell phones couple that capacity with the ability to store many different types of information: Even the most basic phones that sell for less than $20 might hold photographs, picture messages, text messages, Internet browsing history, a calendar, a thousand-entry phone book, and so on. See id., at 30; United States v. Flores-Lopez, 670 F. 3d 803, 806 (CA7 2012). We expect that the gulf between physical practicability and digital capacity will only continue to widen in the future. The storage capacity of cell phones has several interrelated consequences for privacy. First, a cell phone collects in one place many distinct types of information—an address, a note, a prescription, a bank statement, a video—that reveal much more in combination than any isolated record. Second, a cell phone’s capacity allows even just one type of information to convey far more than previously possible. The sum of an individual’s private life can be reconstructed through a thousand photographs labeled with dates, locations, and descriptions; the same cannot be said of a photograph or two of loved ones tucked into a wallet. Third, the data on a phone can date back to the purchase of the phone, or even earlier. A person might carry in his pocket a slip of paper reminding him to call Mr. Jones; he would not carry a record of all his communications with Mr. Jones for the past several months, as would routinely be kept on a phone.[1] Finally, there is an element of pervasiveness that characterizes cell phones but not physical records. Prior to the digital age, people did not typically carry a cache of sensitive personal information with them as they went about their day. Now it is the person who is not carrying a cell phone, with all that it contains, who is the exception. According to one poll, nearly three-quarters of smart phone users report being within five feet of their phones most of the time, with 12% admitting that they even use their phones in the shower. See Harris Interactive, 2013 Mobile Consumer Habits Study (June 2013). A decade ago police officers searching an arrestee might have occasionally stumbled across a highly personal item such as a diary. See, e.g., United States v. Frankenberry, 387 F. 2d 337 (CA2 1967) (per curiam). But those discoveries were likely to be few and far between. Today, by contrast, it is no exaggeration to say that many of the more than 90% of American adults who own a cell phone keep on their person a digital record of nearly every aspect of their lives—from the mundane to the intimate. See Ontario v. Quon, 560 U. S. 746, 760 (2010) . Allowing the police to scrutinize such records on a routine basis is quite different from allowing them to search a personal item or two in the occasional case. Although the data stored on a cell phone is distinguished from physical records by quantity alone, certain types of data are also qualitatively different. An Internet search and browsing history, for example, can be found on an Internet-enabled phone and could reveal an individual’s private interests or concerns—perhaps a search for certain symptoms of disease, coupled with frequent visits to WebMD. Data on a cell phone can also reveal where a person has been. Historic location information is a stand-ard feature on many smart phones and can reconstruct someone’s specific movements down to the minute, not only around town but also within a particular building. See United States v. Jones, 565 U. S. ___, ___ (2012) (Sotomayor, J., concurring) (slip op., at 3) (“GPS monitoring generates a precise, comprehensive record of a person’s public movements that reflects a wealth of detail about her familial, political, professional, religious, and sexual associations.”). Mobile application software on a cell phone, or “apps,” offer a range of tools for managing detailed information about all aspects of a person’s life. There are apps for Democratic Party news and Republican Party news; apps for alcohol, drug, and gambling addictions; apps for sharing prayer requests; apps for tracking pregnancy symptoms; apps for planning your budget; apps for every conceivable hobby or pastime; apps for improving your romantic life. There are popular apps for buying or selling just about anything, and the records of such transactions may be accessible on the phone indefinitely. There are over a million apps available in each of the two major app stores; the phrase “there’s an app for that” is now part of the popular lexicon. The average smart phone user has installed 33 apps, which together can form a revealing montage of the user’s life. See Brief for Electronic Privacy Information Center as Amicus Curiae in No. 13–132, p. 9. In 1926, Learned Hand observed (in an opinion later quoted in Chimel) that it is “a totally different thing to search a man’s pockets and use against him what they contain, from ransacking his house for everything which may incriminate him.” United States v. Kirschenblatt, 16 F. 2d 202, 203 (CA2). If his pockets contain a cell phone, however, that is no longer true. Indeed, a cell phone search would typically expose to the government far more than the most exhaustive search of a house: A phone not only contains in digital form many sensitive records previ-ously found in the home; it also contains a broad array of private information never found in a home in any form—unless the phone is. 2 To further complicate the scope of the privacy interests at stake, the data a user views on many modern cell phones may not in fact be stored on the device itself. Treating a cell phone as a container whose contents may be searched incident to an arrest is a bit strained as an initial matter. See New York v. Belton, 453 U. S. 454, 460, n. 4 (1981) (describing a “container” as “any object capable of holding another object”). But the analogy crumbles entirely when a cell phone is used to access data located elsewhere, at the tap of a screen. That is what cell phones, with increasing frequency, are designed to do by taking advantage of “cloud computing.” Cloud computing is the capacity of Internet-connected devices to display data stored on remote servers rather than on the device itself. Cell phone users often may not know whether particular information is stored on the device or in the cloud, and it generally makes little difference. See Brief for Electronic Privacy Information Center in No. 13–132, at 12–14, 20. Moreover, the same type of data may be stored locally on the device for one user and in the cloud for another. The United States concedes that the search incident to arrest exception may not be stretched to cover a search of files accessed remotely—that is, a search of files stored in the cloud. See Brief for United States in No. 13–212, at 43–44. Such a search would be like finding a key in a suspect’s pocket and arguing that it allowed law enforcement to unlock and search a house. But officers searching a phone’s data would not typically know whether the information they are viewing was stored locally at the time of the arrest or has been pulled from the cloud. Although the Government recognizes the problem, its proposed solutions are unclear. It suggests that officers could disconnect a phone from the network before searching the device—the very solution whose feasibility it contested with respect to the threat of remote wiping. Compare Tr. of Oral Arg. in No. 13–132, at 50–51, with Tr. of Oral Arg. in No. 13–212, pp. 13–14. Alternatively, the Government proposes that law enforcement agencies “develop protocols to address” concerns raised by cloud computing. Reply Brief in No. 13–212, pp. 14–15. Probably a good idea, but the Founders did not fight a revolution to gain the right to government agency protocols. The possibility that a search might extend well beyond papers and effects in the physical proximity of an arrestee is yet another reason that the privacy interests here dwarf those in Robinson. C Apart from their arguments for a direct extension of Robinson, the United States and California offer various fallback options for permitting warrantless cell phone searches under certain circumstances. Each of the proposals is flawed and contravenes our general preference to provide clear guidance to law enforcement through categorical rules. “[I]f police are to have workable rules, the balancing of the competing interests . . . ‘must in large part be done on a categorical basis—not in an ad hoc, case-by-case fashion by individual police officers.’ ” Michigan v. Summers, 452 U. S. 692, 705, n. 19 (1981) (quoting Dunaway v. New York, 442 U. S. 200 –220 (1979) (White, J., concurring)). The United States first proposes that the Gant standard be imported from the vehicle context, allowing a warrantless search of an arrestee’s cell phone whenever it is reasonable to believe that the phone contains evidence of the crime of arrest. But Gant relied on “circumstances unique to the vehicle context” to endorse a search solely for the purpose of gathering evidence. 556 U. S., at 343. Justice Scalia’s Thornton opinion, on which Gant was based, explained that those unique circumstances are “a reduced expectation of privacy” and “heightened law enforcement needs” when it comes to motor vehicles. 541 U. S., at 631; see also Wyoming v. Houghton, 526 U. S., at 303–304. For reasons that we have explained, cell phone searches bear neither of those characteristics. At any rate, a Gant standard would prove no practical limit at all when it comes to cell phone searches. In the vehicle context, Gant generally protects against searches for evidence of past crimes. See 3 W. LaFave, Search and Seizure §7.1(d), at 709, and n. 191. In the cell phone context, however, it is reasonable to expect that incriminating information will be found on a phone regardless of when the crime occurred. Similarly, in the vehicle context Gant restricts broad searches resulting from minor crimes such as traffic violations. See id., §7.1(d), at 713, and n. 204. That would not necessarily be true for cell phones. It would be a particularly inexperienced or unimaginative law enforcement officer who could not come up with sev-eral reasons to suppose evidence of just about any crime could be found on a cell phone. Even an individual pulled over for something as basic as speeding might well have locational data dispositive of guilt on his phone. An individual pulled over for reckless driving might have evidence on the phone that shows whether he was texting while driving. The sources of potential pertinent information are virtually unlimited, so applying the Gant standard to cell phones would in effect give “police officers unbridled discretion to rummage at will among a person’s private effects.” 556 U. S., at 345. The United States also proposes a rule that would restrict the scope of a cell phone search to those areas of the phone where an officer reasonably believes that infor-mation relevant to the crime, the arrestee’s identity, or officer safety will be discovered. See Brief for United States in No. 13–212, at 51–53. This approach would again impose few meaningful constraints on officers. The proposed categories would sweep in a great deal of information, and officers would not always be able to discern in advance what information would be found where. We also reject the United States’ final suggestion that officers should always be able to search a phone’s call log, as they did in Wurie’s case. The Government relies on Smith v. Maryland, 442 U. S. 735 (1979) , which held that no warrant was required to use a pen register at telephone company premises to identify numbers dialed by a particular caller. The Court in that case, however, concluded that the use of a pen register was not a “search” at all under the Fourth Amendment. See id., at 745–746. There is no dispute here that the officers engaged in a search of Wurie’s cell phone. Moreover, call logs typically contain more than just phone numbers; they include any identifying information that an individual might add, such as the label “my house” in Wurie’s case. Finally, at oral argument California suggested a different limiting principle, under which officers could search cell phone data if they could have obtained the same information from a pre-digital counterpart. See Tr. of Oral Arg. in No. 13–132, at 38–43; see also Flores-Lopez, 670 F. 3d, at 807 (“If police are entitled to open a pocket diary to copy the owner’s address, they should be entitled to turn on a cell phone to learn its number.”). But the fact that a search in the pre-digital era could have turned up a photograph or two in a wallet does not justify a search of thousands of photos in a digital gallery. The fact that someone could have tucked a paper bank statement in a pocket does not justify a search of every bank statement from the last five years. And to make matters worse, such an analogue test would allow law enforcement to search a range of items contained on a phone, even though people would be unlikely to carry such a variety of information in physical form. In Riley’s case, for example, it is implausible that he would have strolled around with video tapes, photo albums, and an address book all crammed into his pockets. But because each of those items has a pre-digital analogue, police under California’s proposal would be able to search a phone for all of those items—a significant diminution of privacy. In addition, an analogue test would launch courts on a difficult line-drawing expedition to determine which digital files are comparable to physical records. Is an e-mail equivalent to a letter? Is a voicemail equivalent to a phone message slip? It is not clear how officers could make these kinds of decisions before conducting a search, or how courts would apply the proposed rule after the fact. An analogue test would “keep defendants and judges guessing for years to come.” Sykes v. United States, 564 U. S. 1 , ___ (2011) (Scalia, J., dissenting) (slip op., at 7) (discussing the Court’s analogue test under the Armed Career Criminal Act). IV We cannot deny that our decision today will have an impact on the ability of law enforcement to combat crime. Cell phones have become important tools in facilitating coordination and communication among members of criminal enterprises, and can provide valuable incriminating information about dangerous criminals. Privacy comes at a cost. Our holding, of course, is not that the information on a cell phone is immune from search; it is instead that a warrant is generally required before such a search, even when a cell phone is seized incident to arrest. Our cases have historically recognized that the warrant requirement is “an important working part of our machinery of gov-ernment,” not merely “an inconvenience to be somehow ‘weighed’ against the claims of police efficiency.” Coolidge v. New Hampshire, 403 U. S. 443, 481 (1971) . Recent technological advances similar to those discussed here have, in addition, made the process of obtaining a warrant itself more efficient. See McNeely, 569 U. S., at ___ (slip op., at 11–12); id., at ___ (Roberts, C. J., concurring in part and dissenting in part) (slip op., at 8) (describing jurisdiction where “police officers can e-mail warrant requests to judges’ iPads [and] judges have signed such warrants and e-mailed them back to officers in less than 15 minutes”). Moreover, even though the search incident to arrest exception does not apply to cell phones, other case-specific exceptions may still justify a warrantless search of a particular phone. “One well-recognized exception applies when ‘ “the exigencies of the situation” make the needs of law enforcement so compelling that [a] warrantless search is objectively reasonable under the Fourth Amendment.’ ” Kentucky v. King, 563 U. S., at ___ (slip op., at 6) (quoting Mincey v. Arizona, 437 U. S. 385, 394 (1978) ). Such exigencies could include the need to prevent the imminent destruction of evidence in individual cases, to pursue a fleeing suspect, and to assist persons who are seriously injured or are threatened with imminent injury. 563 U. S., at ___. In Chadwick, for example, the Court held that the exception for searches incident to arrest did not justify a search of the trunk at issue, but noted that “if officers have reason to believe that luggage contains some immediately dangerous instrumentality, such as explosives, it would be foolhardy to transport it to the station house without opening the luggage.” 433 U. S., at 15, n. 9. In light of the availability of the exigent circumstances exception, there is no reason to believe that law enforcement officers will not be able to address some of the more extreme hypotheticals that have been suggested: a suspect texting an accomplice who, it is feared, is preparing to detonate a bomb, or a child abductor who may have information about the child’s location on his cell phone. The defendants here recognize—indeed, they stress—that such fact-specific threats may justify a warrantless search of cell phone data. See Reply Brief in No. 13–132, at 8–9; Brief for Respondent in No. 13–212, at 30, 41. The critical point is that, unlike the search incident to arrest exception, the exigent circumstances exception requires a court to examine whether an emergency justified a warrantless search in each particular case. See McNeely, supra, at ___ (slip op., at 6).[2] * * * Our cases have recognized that the Fourth Amendment was the founding generation’s response to the reviled “general warrants” and “writs of assistance” of the colonial era, which allowed British officers to rummage through homes in an unrestrained search for evidence of criminal activity. Opposition to such searches was in fact one of the driving forces behind the Revolution itself. In 1761, the patriot James Otis delivered a speech in Boston denouncing the use of writs of assistance. A young John Adams was there, and he would later write that “[e]very man of a crowded audience appeared to me to go away, as I did, ready to take arms against writs of assistance.” 10 Works of John Adams 247–248 (C. Adams ed. 1856). According to Adams, Otis’s speech was “the first scene of the first act of opposition to the arbitrary claims of Great Britain. Then and there the child Independence was born.” Id., at 248 (quoted in Boyd v. United States, 116 U. S. 616, 625 (1886) ). Modern cell phones are not just another technological convenience. With all they contain and all they may reveal, they hold for many Americans “the privacies of life,” Boyd, supra, at 630. The fact that technology now allows an individual to carry such information in his hand does not make the information any less worthy of the protection for which the Founders fought. Our answer to the question of what police must do before searching a cell phone seized incident to an arrest is accordingly simple—get a warrant. We reverse the judgment of the California Court of Appeal in No. 13–132 and remand the case for further proceedings not inconsistent with this opinion. We affirm the judgment of the First Circuit in No. 13–212. It is so ordered.Notes 1 Because the United States and California agree that these cases involve incident to arrest, these cases do not implicate the question whether the collection or inspection of aggregated digital information amounts to a search under other circumstances. 2 In Wurie’s case, for example, the dissenting First Circuit judge argued that exigent circumstances could have justified a search of Wurie’s phone. See 728 F. 3d 1, 17 (2013) (opinion of Howard, J.) (discussing the repeated unanswered calls from “my house,” the suspected location of a drug stash). But the majority concluded that the Government had not made an exigent circumstances argument. See at 1. The Government acknowledges the same in this Court. See Brief for United States in No. 13–212, p. 28, n. 8.

572.US.639

Petitioner Robers was convicted of a federal crime for submitting fraudulent mortgage loan applications to two banks. On appeal, he argued that the District Court had miscalculated his restitution obligation under the Mandatory Victims Restitution Act of 1996, 18 U. S. C. §§3663A–3664, a provision of which requires property crime offenders to pay “an amount equal to . . . the value of the property” less “the value (as of the date the property is returned) of any part of the property that is returned,” §3663A(b)(1)(B). The District Court had ordered Robers to pay the difference between the amount lent to him and the amount the banks received in selling the houses that had served as collateral for the loans. Robers claimed that the District Court should have instead reduced the restitution amount by the value of the houses on the date the banks took title to them since that was when “part of the property” was “returned.” The Seventh Circuit rejected Robers’ argument. Held: The phrase “any part of the property . . . returned” refers to the property the banks lost, namely, the money they lent to Robers, and not to the collateral the banks received, namely, the houses. Read naturally, the words “the property,” which appear seven times in §3663A(b)(1), refer to the property that was lost as a result of the crime, here, the money. Because “[g]enerally, ‘identical words used in different parts of the same statute are . . . presumed to have the same meaning,’ ” Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 547 U.S. 71, 86 (quoting IBP, Inc. v. Alvarez, 546 U.S. 21, 34), “the property . . . returned” must also be the property lost as a result of the crime. Any awkwardness or redundancy that comes from substituting an amount of money for the words “the property” is the linguistic price paid for having a single statutory provision that covers different kinds of property. Since valuing money is easier than valuing other types of property, the natural reading also facilitates the statute’s administration. Robers’ contrary arguments are unconvincing. First, other provisions of the statute, see, e.g., §§3664(f)(2), (3)(A), (4), seem to give courts adequate authority to avoid Robers’ false dichotomy of having to choose between refusing to award restitution and requiring the offender to pay the full amount lent where a victim has not sold the collateral by the time of sentencing. Second, for purposes of the statute’s proximate-cause requirement, see §§3663A(a)(2), 3664(e), normal market fluctuations do not break the causal chain between the offender’s fraud and the losses incurred by the victim. Third, even assuming that the return of collateral compensates lenders for their losses under state mortgage law, the issue here is whether the statutory provision, which does not purport to track state mortgage law, requires that collateral received be valued at the time the victim received it. Finally, the rule of lenity does not apply here. See Muscarello v. United States, 524 U.S. 125, 139. Pp. 3–7. 698 F.3d 937, affirmed. Breyer, J., delivered the opinion for a unanimous Court. Sotomayor, J., filed a concurring opinion, in which Ginsburg, J., joined.

The Mandatory Victims Restitution Act of 1996 requires certain offenders to restore property lost by their victims as a result of the crime. 18 U. S. C. §3663A. A provision in the statue says that, when return of the property lost by the victim is “impossible, impracticable, or inadequate,” the offender must pay the victim “an amount equal to . . . the value of the property” less “the value (as of the date the property is returned) of any part of the property that is returned.” §3663A(b)(1)(B). The question before us is whether “any part of the property” is “returned” when a victim takes title to collateral securing a loan that an offender fraudulently obtained from the victim. We hold that it is not. In our view, the statutory phrase “any part of the property” refers only to the specific property lost by a victim, which, in the case of a fraudulently obtained loan, is the money lent. Therefore, no “part of the property” is “returned” to the victim until the collateral is sold and the victim receives money from the sale. The import of our holding is that a sentencing court must reduce the restitution amount by the amount of money the victim received in selling the collateral, not the value of the collateral when the victim received it. I The relevant facts, as simplified, are the following: In 2005 petitioner Benjamin Robers, acting as a straw buyer, submitted fraudulent loan applications to two banks. The banks lent Robers about $470,000 for the purchase of two houses, upon which the banks took mortgages. When Robers failed to make loan payments, the banks foreclosed on the mortgages. In 2006 they took title to the two houses. In 2007 they sold one house for about $120,000. Andin 2008 they sold the other house for about $160,000. The sales took place in a falling real estate market. In 2010 Robers was convicted in federal court of conspiracy to commit wire fraud. See §§371, 1343. He was sentenced to three years of probation. And the court ordered him to pay restitution of about $220,000, roughly the $470,000 the banks lent to Robers less the $280,000 the banks received from the sale of the two houses (minus certain expenses incurred in selling them). On appeal Robers argued that the sentencing court had miscalculated his restitution obligation. In his view, “part of the property” was “returned” to the banks when they took title to the houses. And, since the statute says that “returned” property shall be valued “as of the date the property is returned,” the sentencing court should have reduced the restitution amount by more than $280,000: $280,000 was what the banks received from the sale of the houses, but since the banks sold the houses in a falling real estate market, the houses had been worth more when the banks took title to them. The Court of Appeals rejected Robers’ argument. 698 F. 3d 937 (CA7 2012). And, because different Circuits have come to different conclusions about this kind of matter, we granted Robers’ petition for certiorari. Compare id., at 942 (case below) (restitution obligation reduced by money received from sale of collateral), with United States v. Yeung, 672 F. 3d 594, 604 (CA9 2012) (restitution obligation reduced by value of collateral at time lender took title). II In our view, the phrase “any part of the property . . . returned” refers to the property the banks lost, namely, the money they lent to Robers, and not to the collateral the banks received, namely, the two houses. For one thing, that is what the statute says. The phrase is part of a long sentence that reads as follows: “(b) The order of restitution shall require that [the] defendant— “(1) in the case of an offense resulting in damage to or loss or destruction of property of a victim of theoffense— “(A) return the property to the owner of the property . . . ; or “(B) if return of the property under subparagraph (A) is impossible, impracticable, or inadequate, pay an amount equal to— “(i) the greater of— “(I) the value of the property on the date of the damage, loss, or destruction; or “(II) the value of the property on the date of sentencing, less “(ii) the value (as of the date the property is returned) of any part of the property that is returned . . . .” §3663A (emphasis added). The words “the property” appear seven times in this sentence. If read naturally, they refer to the “property” that was “damage[d],” “los[t],” or “destr[oyed]” as a result of the crime. §3663A(b)(1). “Generally, ‘identical words used in different parts of the same statute are . . . presumed to have the same meaning.’ ” Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 547 U. S. 71, 86 (2006) (quoting IBP, Inc. v. Alvarez, 546 U. S. 21, 34 (2005) ). And, if the “property” that was “damage[d],” “los[t],” or “destr[oyed]” was the money, then “the property . . . returned” must also be the money. Money being fungible, however, see, e.g., Ransom v. FIA Card Services, N. A., 562 U. S. ___, ___ (2011) (slip op., at 17); Sabri v. United States, 541 U. S. 600, 606 (2004) , “the property . . . returned” need not be the very same bills or checks. We concede that substituting an amount of money, say, $1,000, for the words “the property” will sometimes seem awkward or unnecessary as, for example: “[I]f return of [$1,000] . . . is impossible, . . . pay an amount equal to . . . the greater of . . . the value of [$1,000] on the date of the . . . loss . . . or . . . the value of [$1,000] on the date of sentencing . . . .” §3663A(b)(1)(B). But any such awkwardness or redundancy is the linguistic price paid for having a single statutory provision that covers property of many different kinds. The provision is not awkward as applied to, say, a swindler who obtains jewelry, is unable to return all of the jewelry, and must then instead pay an amount equal to the value of all of the jewelry obtained less the value (as of the date of the return) of any of the jewelry that he did return. It directs the court to value the returned jewelry as of the date it was returned and subtract that amount from the value of all of the jewelry the swindler obtained. As applied to money, the provision is in part unnecessary but reading the statute similarly does no harm. And the law does not require legislators to write extra language specifically exempting, phrase by phrase, applications in respect to which a portion of a phrase is not needed. The natural reading also facilitates the statute’s administration. Many victims who lose money but subsequently receive other property (e.g., collateral securing a loan) will sell that other property and receive money from the sale. And often that sale will take place fairly soon after the victim receives the property. Valuing the money from the sale is easy. But valuing other property as of the time it was received may provoke argument, requiring time, expense, and expert testimony to resolve. We are not convinced by Robers’ arguments to the contrary. First, Robers says that, when a victim has not sold the collateral by the time of sentencing, our interpretation will lead to unfair results. A sentencing court will have only two choices, both undesirable. The court will either have to refuse to award restitution, thereby undercompensating the victim, or have to require the offender to pay the full amount lent to him, thereby giving the victim a windfall. In our view, however, the dilemma is a false one. Other provisions of the statute allow the court to avoid an undercompensation or a windfall. Where, for example, a sale of the collateral is foreseen but has not yet taken place, the court may postpone determination of the restitution amount for two to three months after sentencing, thereby providing the victim with additional time to sell. See §3664(d)(5). Where a victim receives, say, collateral, but does not intend to sell it, other provisions of the statute may come into play. Section 3664(f)(2) provides that upon “determination of the amount of restitution owed to each victim, the court shall . . . specify in the restitution order the manner in which, and the schedule according to which, the restitution is to be paid.” Section 3664(f)(3)(A) says that a “restitution order may direct the defendant to make a single, lump-sum payment, partial payments at spec-ified intervals, in-kind payments, or a combinationof payments at specified intervals and in-kindpayments.” And §3664(f)(4) defines “in-kind payment” as including “replacement of property.” These provisions would seem to give a court adequate authority to count, as part of the restitution paid, the value of collateral previously received but not sold. Regardless, Robers has not pointed us to any case suggesting an unfairness problem. And the Government has conceded that the statute (whether through these or other provisions) provides room for “credit[s]” against an offender’s restitution obligation “to prevent double recovery to the victim.” Brief for United States 30 (emphasis deleted). Robers also points out, correctly, that the statute has a proximate cause requirement. See §3663A(a)(2) (defining “victim” as “a person directly and proximately harmed as a result of the commission of” the offense (emphasis added)); §3664(e) (Government bears the “burden of demon-strating the amount of the loss sustained by a victim as a result of the offense” (emphasis added)). Cf. Paroline v. United States, ante, at 6–11. And Robers argues that where, as here, a victim receives less money from a later sale than the collateral was worth when received, the market and not the offender is the proximate cause of the deficiency. We are not convinced. The basic question that a proximate cause requirement presents is “whether the harm alleged has a sufficiently close connection to the conduct” at issue. Lexmark Int’l, Inc. v. Static Control Components, Inc., ante, at 14. Here, it does. Fluctuations in property values are common. Their existence (though not direction or amount) is foreseeable. And losses in part incurred through a decline in the value of collateral sold are di-rectly related to an offender’s having obtained collateralized property through fraud. That is not to say that an offender is responsible for everything that reduces the amount of money a victim receives for collateral. Market fluctuations are normally unlike, say, an unexpected natural disaster that destroys collateral or a victim’s donation of collateral or its sale to a friend for a nominal sum—any of which, as the Government concedes, could break the causal chain. See Tr. of Oral Arg. 25–27, 38–39, 46, 50–51. Further, Robers argues that “principles” of state mortgage law “confirm that the return of mortgage collateral compensates a lender for its losses.” Brief for Petitioner 30. But whether the collateral compensates a victim for its losses is not the question before us. That question is whether the particular statutory provision at issue here requires that collateral received be valued at the time the victim received it. That statutory provision does not purport to track the details of state mortgage law. Thus, even were we to assume that Robers is right about the details of state mortgage law, we would not find them sufficient to change our interpretation. Finally, Robers invokes the rule of lenity. To apply this rule, we would have to assume that we could interpret the statutory provision to help an offender like Robers, who is hurt when the market for collateral declines, without harming other offenders, who would be helped when the market for collateral rises. We cannot find such an interpretation. Regardless, the rule of lenity applies only if, after using the usual tools of statutory construction, we are left with a “grievous ambiguity or uncertainty in the statute.” Muscarello v. United States, 524 U. S. 125, 139 (1998) (internal quotation marks omitted). Having come to the end of our analysis, we are left with no such ambiguity or uncertainty here. The statutory provision refers to the money lost, not to the collateral received. * * * For these reasons, the judgment of the Court of Appeals is affirmed. It is so ordered.

572.US.65

Petitioner Justus Rosemond took part in a drug deal in which either he or one of his confederates fired a gun. Because the shooter’s identity was disputed, the Government charged Rosemond with violating 18 U. S. C. §924(c) by using or carrying a gun in connection with a drug trafficking crime, or, in the alternative, aiding and abetting that offense under 18 U. S. C. §2. The trial judge instructed the jury that Rosemond was guilty of aiding and abetting the §924(c) offense if he (1) “knew his cohort used a firearm in the drug trafficking crime” and (2) “knowingly and actively participated in the drug trafficking crime.” This deviated from Rosemond’s proposed instruction that the jury must find that he acted intentionally “to facilitate or encourage” the firearm’s use, as opposed to merely the predicate drug offense. Rosemond was convicted, and the Tenth Circuit affirmed, rejecting his argument that the District Court’s aiding and abetting instructions were erroneous. Held: 1. The Government establishes that a defendant aided and abetted a §924(c) violation by proving that the defendant actively participated in the underlying drug trafficking or violent crime with advance knowledge that a confederate would use or carry a gun during the crime’s commission. Pp. 5–16. (a) The federal aiding and abetting statute, which derives from common-law standards for accomplice liability, has two components. A person is liable under §2 only if he (1) takes an affirmative act in furtherance of the underlying offense (2) with the intent to facilitate that offense’s commission. Pp. 5–6. (b) The first question is whether Rosemond’s conduct was sufficient to satisfy the affirmative act requirement of aiding and abetting. Section 924(c) has two elements: a drug deal or violent crime, and using or carrying a firearm in connection with that crime. The instructions permitted the jury to convict Rosemond of aiding and abetting even if he facilitated only the drug element, and not the gun element, of the §924(c) offense. Those instructions were correct. The common law imposed aiding and abetting liability on a person who facilitated any element of a criminal offense, even if he did not facilitate all elements. That principle continues to govern §2. See, e.g., United States v. Johnson, 319 U.S. 503, 515. Pp. 6–11. (c) In addition to conduct extending to some part of the crime, aiding and abetting requires intent extending to the whole crime. The defendant must not just associate himself with the venture, but also participate in it as something that he wishes to bring about and seek by his actions to make it succeed. Nye & Nissen v. United States, 336 U.S. 613, 619. That requirement is satisfied when a person actively participates in a criminal venture with full knowledge of the circumstances constituting the charged offense. See Pereira v. United States, 347 U.S. 1, 12. An active participant in a drug transaction has the intent needed to aid and abet a §924(c) violation when he knows that one of his confederates will carry a gun. This must be advance knowledge—meaning, knowledge at a time when the accomplice has a reasonable opportunity to walk away. Pp. 11–16. 2. The trial court’s jury instructions were erroneous because they failed to require that Rosemond knew in advance that one of his cohorts would be armed. In telling the jury to consider merely whether Rosemond “knew his cohort used a firearm,” the court did not direct the jury to determine when Rosemond obtained the requisite knowledge—i.e., to decide whether Rosemond knew about the gun in sufficient time to withdraw from the crime. The case is remanded to permit the Tenth Circuit to address whether this objection was properly preserved and whether any error was harmless. Pp. 16–19. 695 F.3d 1151, vacated and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined, and in which Scalia, J., joined in all but footnotes 7 and 8. Alito, J., filed an opinion concurring in part and dissenting in part, in which Thomas, J., joined.

he government need not prove assistance related to every element of the underlying offense”). And so forth and so on. 7 Consider a hypothetical similar to and (and a modern variant of the Wharton treatise’s, see n. 4, ). Suppose that as part of a kidnapping scheme, one accomplice lures the victim into a car under false pretenses; another drives the vehicle; a third allows the use of her house to hold the victim captive; and still a fourth keeps watch outside to divert potential witnesses. None would have personally completed, or even assisted with, all elements of the offense. See, ,v., 504 F. 3d 576, 580 (CA5 2007) (listing elements). But (if they had the requisite intent) all would be liable under §2. 8 Some authorities suggest an exception to the general rule when another crime is the “natural and probable consequence” of the crime the defendant intended to abet. See, ,2 LaFave §13.3(b), at 356 (citing cases); but see , §13.3 (“Under the better view, one is not an accomplice to a crime merely because . . . that crime was a natural and probable consequence of another offense as to which he is an accomplice”). That question is not implicated here, because no one contends that a §924(c) violation is a natural and probable consequence of simple drug trafficking. We therefore express no view on the issue. 9 We did not deal in these cases, nor do we here, with defendants who incidentally facilitate a criminal venture rather than actively participate in it. A hypothetical case is the owner of a gun store who sells a firearm to a criminal, knowing but not caring how the gun will be used. We express no view about what sort of facts, if any, would suffice to show that such a third party has the intent necessary to be convicted of aiding and abetting. 10 Of course, if a defendant continues to participate in a crime after a gun was displayed or used by a confederate, the jury can permissibly infer from his failure to object or withdraw that he had such knowledge. In any criminal case, after all, the factfinder can draw inferences about a defendant’s intent based on all the facts and circumstances of a crime’s commission. 11 Contrary to the dissent’s view, see , at 3–4,nothing in this holding changes the way the defenses of duress and necessity operate. Neither does our decision remotely deny that the “intent to undertake some act is . . . perfectly consistent with the motive of avoiding adverse consequences which would otherwise occur.” at 5. Our holding is grounded in the distinctive intent standard for aiding and abetting someone else’s act—in the words of Judge Hand, that a defendant must not just “in some sort associate himself with the venture” (as seems to be good enough for the dissent), but also “participate in it as in something that he wishes to bring about” and “seek by his action to make it succeed.” v. , (quoting , 100 F. 2d, at 402). For the reasons just given, see at 13, 15–16, we think that intent standard cannot be satisfied if a defendant charged with aiding and abetting a §924(c) offense learns of a gun only after he can realistically walk away—when he has no opportunity to decide whether “he wishes to bring about” (or make succeed) an drug transaction, rather than a simple drug crime. And because a defendant’s prior knowledge is part of the intent required to aid and abet a §924(c) offense, the burden to prove it resides with the Government.

571.US.220

Petitioner Sandifer and others filed a putative collective action under the Fair Labor Standards Act of 1938, seeking backpay for time spent donning and doffing pieces of protective gear that they assert respondent United States Steel Corporation requires workers to wear because of hazards at its steel plants. U. S. Steel contends that this donning-and-doffing time, which would otherwise be compensable under the Act, is noncompensable under a provision of its collective-bargaining agreement with petitioners’ union. That provision’s validity depends on 29 U. S. C. §203(o), which allows parties to collectively bargain over whether “time spent in changing clothes . . . at the beginning or end of each workday” must be compensated. The District Court granted U. S. Steel summary judgment in pertinent part, holding that petitioners’ donning and doffing constituted “changing clothes” under §203(o). It also assumed that any time spent donning and doffing items that were not “clothes” was “de minimis” and hence noncompensable. The Seventh Circuit affirmed. Held: The time petitioners spend donning and doffing their protective gear is not compensable by operation of §203(o). Pp. 3–15. (a) This Court initially construed compensability under the Fair Labor Standards Act expansively. See, e.g., Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680. The Act was amended in 1949, however, to provide that the compensability of time spent “changing clothes or washing at the beginning or end of each workday” is a subject appropriately committed to collective bargaining, §203(o). Whether petitioners’ donning and doffing qualifies as “changing clothes” depends on the meaning of that statutory phrase. Pp. 3–6. (b) The term “clothes,” which is otherwise undefined, is “interpreted as taking [its] ordinary, contemporary, common meaning.” Perrin v. United States, 444 U.S. 37, 42. In dictionaries from the era of §203(o)’s enactment, “clothes” denotes items that are both designed and used to cover the body and are commonly regarded as articles of dress. Nothing in §203(o)’s text or context suggests anything other than this ordinary meaning. There is no basis for petitioners’ proposition that the unmodified term “clothes” somehow omits protective clothing. Section 203(o)’s exception applies only when the changing of clothes is “an integral and indispensable part of the principal activities for which covered workmen are employed,” Steiner v. Mitchell, 350 U.S. 247, 256, and thus otherwise compensable under the Act. See 29 U. S. C. §254(a). And protective gear is the only clothing that is integral and indispensable to the work of many occupations, such as butchers and longshoremen. Petitioners’ position is also incompatible with the historical context of §203(o)’s passage, contradicting contemporaneous Labor Department regulations and dictum in Steiner, see 350 U. S., at 248, 254–255. The interpretation adopted here leaves room for distinguishing between clothes and wearable items that are not clothes, such as some equipment and devices. The view of respondent and its amici that “clothes” encompasses the entire outfit that one puts on to be ready for work is also devoid of any textual foundation. Pp. 6–10. (c) While the normal meaning of “changing clothes” connotes substitution, “changing” also carried the meaning to “alter” at the time of §203(o)’s enactment. The broader statutory context makes plain that “time spent in changing clothes” includes time spent in altering dress. Whether one exchanges street clothes for work clothes or simply chooses to layer one over the other may be a matter of purely personal choice, and §203(o) should not be read to allow workers to opt into or out of its coverage at random or at will when another reading is textually permissible. Pp. 10–11. (d) Applying these principles here, it is evident that the donning and doffing in this case qualifies as “changing clothes” under §203(o). Of the 12 items at issue, only 3—safety glasses, earplugs, and a respirator—do not fit within the elaborated interpretation of “clothes.” Apparently concerned that federal judges would have to separate the minutes spent clothes-changing and washing from the minutes devoted to other activities during the relevant period, some Courts of Appeals have invoked the doctrine de minimis non curat lex (the law does not take account of trifles). But that doctrine does not fit comfortably within this statute, which is all about trifles. A more appropriate way to proceed is for courts to ask whether the period at issue can, on the whole, be fairly characterized as “time spent in changing clothes or washing.” If an employee devotes the vast majority of that time to putting on and off equipment or other non-clothes items, the entire period would not qualify as “time spent in changing clothes” under §203(o), even if some clothes items were also donned and doffed. But if the vast majority of the time is spent in donning and doffing “clothes” as defined here, the entire period qualifies, and the time spent putting on and off other items need not be subtracted. Here, the Seventh Circuit agreed with the District Court’s conclusion that the time spent donning and doffing safety glasses and earplugs was minimal. And this Court is disinclined to disturb the District Court’s additional factual finding, not addressed by the Seventh Circuit, that the respirators were donned and doffed as needed during the normal workday and thus fell beyond §203(o)’s scope. Pp. 12–15. 678 F.3d 590, affirmed. Scalia, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, Breyer, Alito, and Kagan, JJ., joined, and in which Sotomayor, J., joined except as to footnote 7.

n employer may not arbitrarily fail to count as hours worked any part, however small, of the employee’s fixed or regular working time or practically ascertainable period of time he is regularly required to spend on duties assigned to him.” 29 CFR §785.47.

572.US.291

After this Court decided that the University of Michigan’s undergraduate admissions plan’s use of race-based preferences violated the Equal Protection Clause, Gratz v. Bollinger, 539 U.S. 244, 270, but that the law school admission plan’s more limited use did not, Grutter v. Bollinger, 539 U.S. 306, 343, Michigan voters adopted Proposal 2, now Art. I, §26, of the State Constitution, which, as relevant here, prohibits the use of race-based preferences as part of the admissions process for state universities. In consolidated challenges, the District Court granted summary judgment to Michigan, thus upholding Proposal 2, but the Sixth Circuit reversed, concluding that the proposal violated the principles of Washington v. Seattle School Dist. No. 1, 458 U.S. 457. Held: The judgment is reversed. 701 F.3d 466, reversed. Justice Kennedy, joined by The Chief Justice and Justice Alito, concluded that there is no authority in the Federal Constitution or in this Court’s precedents for the Judiciary to set aside Michigan laws that commit to the voters the determination whether racial preferences may be considered in governmental decisions, in particular with respect to school admissions. Pp. 4–18. (a) This case is not about the constitutionality, or the merits, of race-conscious admissions policies in higher education. Here, the principle that the consideration of race in admissions is permissible when certain conditions are met is not being challenged. Rather, the question concerns whether, and in what manner, voters in the States may choose to prohibit the consideration of such racial preferences. Where States have prohibited race-conscious admissions policies, universities have responded by experimenting “with a wide variety of alternative approaches.” Grutter, supra, at 342. The decision by Michigan voters reflects the ongoing national dialogue about such practices. Pp. 4–5. (b) The Sixth Circuit’s determination that Seattle controlled here extends Seattle’s holding in a case presenting quite different issues to reach a mistaken conclusion. Pp. 5–18. (1) It is necessary to consider first the relevant cases preceding Seattle and the background against which Seattle arose. Both Reitman v. Mulkey, 387 U.S. 369, and Hunter v. Erickson, 393 U.S. 385, involved demonstrated injuries on the basis of race that, by reasons of state encouragement or participation, became more aggravated. In Mulkey, a voter-enacted amendment to the California Constitution prohibiting state legislative interference with an owner’s prerogative to decline to sell or rent residential property on any basis barred the challenging parties, on account of race, from invoking the protection of California’s statutes, thus preventing them from leasing residential property. In Hunter, voters overturned an Akron ordinance that was enacted to address widespread racial discrimination in housing sales and rentals had forced many to live in “ ‘unhealthful, unsafe, unsanitary and overcrowded’ ” segregated housing, 393 U. S., at 391. In Seattle, after the school board adopted a mandatory busing program to alleviate racial isolation of minority students in local schools, voters passed a state initiative that barred busing to desegregate. This Court found that the state initiative had the “practical effect” of removing “the authority to address a racial problem . . . from the existing decisionmaking body, in such a way as to burden minority interests” of busing advocates who must now “seek relief from the state legislature, or from the statewide electorate.” 458 U. S., at 474. Pp. 5–8. (2) Seattle is best understood as a case in which the state action had the serious risk, if not purpose, of causing specific injuries on account of race as had been the case in Mulkey and Hunter. While there had been no judicial finding of de jure segregation with respect to Seattle’s school district, a finding that would be required today, see Parents Involved in Community Schools v. Seattle School Dist. No. 1, 551 U.S. 701, 720–721, Seattle must be understood as Seattle understood itself, as a case in which neither the State nor the United States “challenge[d] the propriety of race-conscious student assignments for the purpose of achieving integration, even absent a finding of prior de jure segregation.” 458 U.S. at 472, n. 15. Seattle’s broad language, however, went well beyond the analysis needed to resolve the case. Seizing upon the statement in Justice Harlan’s concurrence in Hunter that the procedural change in that case had “the clear purpose of making it more difficult for certain racial and religious minorities to achieve legislation that is in their interest,” 385 U. S., at 395, the Seattle Court established a new and far-reaching rationale: Where a government policy “inures primarily to the benefit of the minority” and “minorities . . . consider” the policy to be “ ‘in their interest,’ ” then any state action that “place[s] effective decisionmaking authority over” that policy “at a different level of government” is subject to strict scrutiny. 458 U. S., at 472, 474. Pp. 8–11. (3) To the extent Seattle is read to require the Court to determine and declare which political policies serve the “interest” of a group defined in racial terms, that rationale was unnecessary to the decision in Seattle; it has no support in precedent; and it raises serious equal protection concerns. In cautioning against “impermissible racial stereotypes,” this Court has rejected the assumption that all individuals of the same race think alike, see Shaw v. Reno, 509 U.S. 630, 647, but that proposition would be a necessary beginning point were the Seattle formulation to control. And if it were deemed necessary to probe how some races define their own interest in political matters, still another beginning point would be to define individuals according to race. Such a venture would be undertaken with no clear legal standards or accepted sources to guide judicial decision. It would also result in, or impose a high risk of, inquiries and categories dependent upon demeaning stereotypes, classifications of questionable constitutionality on their own terms. Assuming these steps could be taken, the court would next be required to determine the policy realms in which groups defined by race had a political interest. That undertaking, again without guidance from accepted legal standards, would risk the creation of incentives for those who support or oppose certain policies to cast the debate in terms of racial advantage or disadvantage. Adoption of the Seattle formulation could affect any number of laws or decisions, involving, e.g., tax policy or housing subsidies. And racial division would be validated, not discouraged. It can be argued that objections to the larger consequences of the Seattle formulation need not be confronted here, for race was an undoubted subject of the ballot issue. But other problems raised by Seattle, such as racial definitions, still apply. And the principal flaw in the Sixth Circuit’s decision remains: Here there was no infliction of a specific injury of the kind at issue in Mulkey and Hunter and in the history of the Seattle schools, and there is no precedent for extending these cases to restrict the right of Michigan voters to determine that race-based preferences granted by state entities should be ended. The Sixth Circuit’s judgment also calls into question other States’ long-settled rulings on policies similar to Michigan’s. Unlike the injuries in Mulkey, Hunter, and Seattle, the question here is not how to address or prevent injury caused on account of race but whether voters may determine whether a policy of race-based preferences should be continued. By approving Proposal 2 and thereby adding §26 to their State Constitution, Michigan voters exercised their privilege to enact laws as a basic exercise of their democratic power, bypassing public officials they deemed not responsive to their concerns about a policy of granting race-based preferences. The mandate for segregated schools, Brown v. Board of Education, 347 U.S. 483, and scores of other examples teach that individual liberty has constitutional protection. But this Nation’s constitutional system also embraces the right of citizens to speak and debate and learn and then, as a matter of political will, to act through a lawful electoral process, as Michigan voters have done here. These precepts are not inconsistent with the well-established principle that when hurt or injury is inflicted on racial minorities by the encouragement or command of laws or other state action, the Constitution requires redress by the courts. Such circumstances were present in Mulkey, Hunter, and Seattle, but they are not present here. Pp. 11–18. Justice Scalia, joined by Justice Thomas, agreed that §26 rightly stands, though not because it passes muster under the political-process doctrine. It likely does not, but the cases establishing that doctrine should be overruled. They are patently atextual, unadministrable, and contrary to this Court’s traditional equal protection jurisprudence. The question here, as in every case in which neutral state action is said to deny equal protection on account of race, is whether the challenged action reflects a racially discriminatory purpose. It plainly does not. Pp. 1–18. (a) The Court of Appeals for the Sixth Circuit held §26 unconstitutional under the so-called political-process doctrine, derived from Washington v. Seattle School Dist. No. 1, 458 U.S. 457, and Hunter v. Erickson, 393 U.S. 385. In those cases, one level of government exercised borrowed authority over an apparently “racial issue” until a higher level of government called the loan. This Court deemed each revocation an equal-protection violation, without regard to whether there was evidence of an invidious purpose to discriminate. The relentless, radical logic of Hunter and Seattle would point to a similar conclusion here, as in so many other cases. Pp. 3–7. (b) The problems with the political-process doctrine begin with its triggering prong, which assigns to a court the task of determining whether a law that reallocates policymaking authority concerns a “racial issue,” Seattle, 458 U. S., at 473, i.e., whether adopting one position on the question would “at bottom inur[e] primarily to the benefit of the minority, and is designed for that purpose,” id., at 472. Such freeform judicial musing into ethnic and racial “interests” involves judges in the dirty business of dividing the Nation “into racial blocs,” Metro Broadcasting, Inc. v. FCC, 497 U.S. 547, 603, 610 (O’Connor, J., dissenting), and promotes racial stereotyping, see Shaw v. Reno, 509 U.S. 630, 647. More fundamentally, the analysis misreads the Equal Protection Clause to protect particular groups, a construction that has been repudiated in a “long line of cases understanding equal protection as a personal right.” Adarand Constructors, Inc. v. Peña, 515 U.S. 200, 224, 230. Pp. 7–12. (c) The second part of the Hunter-Seattle analysis directs a court to determine whether the challenged act “place[s] effective decisionmaking authority over [the] racial issue at a different level of government,” Seattle, supra, at 474; but, in another line of cases, the Court has emphasized the near-limitless sovereignty of each State to design its governing structure as it sees fit, see, e.g., Holt Civic Club v. Tuscaloosa, 439 U.S. 60, 71. Taken to the limits of its logic, Hunter-Seattle is the gaping exception that nearly swallows the rule of structural state sovereignty, which would seem to permit a State to give certain powers to cities, later assign the same powers to counties, and even reclaim them for itself. Pp. 12–15. (d) Hunter and Seattle also endorse a version of the proposition that a facially neutral law may deny equal protection solely because it has a disparate racial impact. That equal-protection theory has been squarely and soundly rejected by an “unwavering line of cases” holding “that a violation of the Equal Protection Clause requires state action motivated by discriminatory intent,” Hernandez v. New York, 500 U.S. 352, 372–373 (O’Connor, J., concurring in judgment), and that “official action will not be held unconstitutional solely because it results in a racially disproportionate impact,” Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 264–265. Respondents cannot prove that the action here reflects a racially discriminatory purpose, for any law expressly requiring state actors to afford all persons equal protection of the laws does not—cannot—deny “to any person . . . equal protection of the laws,” U. S. Const., Amdt. 14, §1. Pp. 15–17. Justice Breyer agreed that the amendment is consistent with the Equal Protection Clause, but for different reasons. First, this case addresses the amendment only as it applies to, and forbids, race-conscious admissions programs that consider race solely in order to obtain the educational benefits of a diverse student body. Second, the Constitution permits, but does not require, the use of the kind of race-conscious programs now barred by the Michigan Constitution. It foresees the ballot box, not the courts, as the normal instrument for resolving debates about the merits of these programs. Third, Hunter v. Erickson, 393 U.S. 385, and Washington v. Seattle School Dist. No. 1, 458 U.S. 457, which reflect the important principle that an individual’s ability to participate meaningfully in the political process should be independent of his race, do not apply here. Those cases involved a restructuring of the political process that changed the political level at which policies were enacted, while this case involves an amendment that took decisionmaking authority away from unelected actors and placed it in the hands of the voters. Hence, this case does not involve a diminution of the minority’s ability to participate in the political process. Extending the holding of Hunter and Seattle to situations where decisionmaking authority is moved from an administrative body to a political one would also create significant difficulties, given the nature of the administrative process. Furthermore, the principle underlying Hunter and Seattle runs up against a competing principle favoring decisionmaking through the democratic process. Pp. 1–6. Kennedy, J., announced the judgment of the Court and delivered an opinion, in which Roberts, C. J., and Alito, J., joined. Roberts, C. J., filed a concurring opinion. Scalia, J., filed an opinion concurring in the judgment, in which Thomas, J., joined. Breyer, J., filed an opinion concurring in the judgment. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, J., joined. Kagan, J., took no part in the consideration or decision of the case.

in which The Chief Justice and Justice Alito join. The Court in this case must determine whether an amendment to the Constitution of the State of Michigan, approved and enacted by its voters, is invalid under the Equal Protection Clause of the Fourteenth Amendment to the Constitution of the United States. In 2003 the Court reviewed the constitutionality of two admissions systems at the University of Michigan, one for its undergraduate class and one for its law school. The undergraduate admissions plan was addressed in Gratz v. Bollinger, 539 U. S. 244 . The law school admission plan was addressed in Grutter v. Bollinger, 539 U. S. 306 . Each admissions process permitted the explicit consideration of an applicant’s race. In Gratz, the Court invalidated the undergraduate plan as a violation of the Equal Protection Clause. 539 U. S., at 270. In Grutter, the Court found no constitutional flaw in the law school admission plan’s more limited use of race-based preferences. 539 U. S., at 343. In response to the Court’s decision in Gratz, the university revised its undergraduate admissions process, but the revision still allowed limited use of race-based preferences. After a statewide debate on the question of racial preferences in the context of governmental decisionmaking, the voters, in 2006, adopted an amendment to the State Constitution prohibiting state and other governmental entities in Michigan from granting certain preferences, including race-based preferences, in a wide range of actions and decisions. Under the terms of the amendment, race-based preferences cannot be part of the admissions process for state universities. That particular prohibition is central to the instant case. The ballot proposal was called Proposal 2 and, after it passed by a margin of 58 percent to 42 percent, the resulting enactment became Article I, §26, of the Michigan Constitution. As noted, the amendment is in broad terms. Section 26 states, in relevant part, as follows: “(1) The University of Michigan, Michigan State University, Wayne State University, and any other public college or university, community college, or school district shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting. “(2) The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting. “(3) For the purposes of this section ‘state’ includes, but is not necessarily limited to, the state itself, any city, county, any public college, university, or community college, school district, or other political subdivision or governmental instrumentality of or within the State of Michigan not included in sub-section 1.” Section 26 was challenged in two cases. Among the plaintiffs in the suits were the Coalition to Defend Affirmative Action, Integration and Immigrant Rights and Fight for Equality By Any Means Necessary (BAMN); students; faculty; and prospective applicants to Michigan public universities. The named defendants included then-Governor Jennifer Granholm, the Board of Regents of the University of Michigan, the Board of Trustees of Michigan State University, and the Board of Governors of Wayne State University. The Michigan Attorney General was granted leave to intervene as a defendant. The United States District Court for the Eastern District of Michigan consolidated the cases. In 2008, the District Court granted summary judgment to Michigan, thus upholding Proposal 2. BAMN v. Regents of Univ. of Mich., 539 F. Supp. 2d 924. The District Court denied a motion to reconsider the grant of summary judgment. 592 F. Supp. 2d 948. A panel of the United States Court of Appeals for the Sixth Circuit reversed the grant of summary judgment. 652 F. 3d 607 (2011). Judge Gibbons dissented from that holding. Id., at 633–646. The panel majority held that Proposal 2 had violated the principles elaborated by this Court in Washington v. Seattle School Dist. No. 1, 458 U. S. 457 (1982) , and in the cases that Seattle relied upon. The Court of Appeals, sitting en banc, agreed with the panel decision. 701 F. 3d 466 (CA6 2012). The majority opinion determined that Seattle “mirrors the [case] before us.” Id., at 475. Seven judges dissented in a number of opinions. The Court granted certiorari. 568 U. S. ___ (2013). Before the Court addresses the question presented, it is important to note what this case is not about. It is not about the constitutionality, or the merits, of race-conscious admissions policies in higher education. The consideration of race in admissions presents complex questions, in part addressed last Term in Fisher v. University of Texas at Austin, 570 U. S. ––– (2013). In Fisher, the Court did not disturb the principle that the consideration of race in admissions is permissible, provided that certain conditions are met. In this case, as in Fisher, that principle is not challenged. The question here concerns not the permissibility of race-conscious admissions policies under the Constitution but whether, and in what manner, voters in the States may choose to prohibit the consideration of racial preferences in governmental decisions, in particular with respect to school admissions. This Court has noted that some States have decided to prohibit race-conscious admissions policies. In Grutter, the Court noted: “Universities in California, Florida, and Washington State, where racial preferences in admissions are prohibited by state law, are currently engaged in experimenting with a wide variety of alternative approaches. Universities in other States can and should draw on the most promising aspects of these race-neutral alternatives as they develop.” 539 U. S., at 342 (citing United States v. Lopez, 514 U. S. 549, 581 (1995) (Kennedy, J., concurring) (“[T]he States may perform their role as laboratories for experimentation to devise various solutions where the best solution is far from clear”)). In this way, Grutter acknowledged the significance of a dialogue regarding this contested and complex policy question among and within States. There was recognition that our federal structure “permits ‘ innovation and experimentation’ ” and “enables greater citizen ‘ involvement in democratic processes.’ ” Bond v. United States, 564 U. S. –––, ––– (2011) (slip op., at 9) (quoting Gregory v. Ashcroft, 501 U. S. 452, 458 (1991) ). While this case arises in Michigan, the decision by the State’s voters reflects in part the national dialogue regarding the wisdom and practicality of race-conscious admissions policies in higher education. See, e.g., Coalition for Economic Equity v. Wilson, 122 F. 3d 692 (CA9 1997). In Michigan, the State Constitution invests independent boards of trustees with plenary authority over public universities, including admissions policies. Mich. Const., Art. VIII, §5; see also Federated Publications, Inc. v. Board of Trustees of Mich. State Univ., 460 Mich. 75, 86–87, 594 N. W. 2d 491, 497 (1999). Although the members of the boards are elected, some evidence in the record suggests they delegated authority over admissions policy to the faculty. But whether the boards or the faculty set the specific policy, Michigan’s public universities did consider race as a factor in admissions decisions before 2006. In holding §26 invalid in the context of student admissions at state universities, the Court of Appeals relied in primary part on Seattle, supra, which it deemed to control the case. But that determination extends Seattle’s holding in a case presenting quite different issues to reach a conclusion that is mistaken here. Before explaining this further, it is necessary to consider the relevant cases that preceded Seattle and the background against which Seattle itself arose. Though it has not been prominent in the arguments of the parties, this Court’s decision in Reitman v. Mulkey, 387 U. S. 369 (1967) , is a proper beginning point for discussing the controlling decisions. In Mulkey, voters amended the California Constitution to prohibit any state legislative interference with an owner’s prerogative to decline to sell or rent residential property on any basis. Two different cases gave rise to Mulkey. In one a couple could not rent an apartment, and in the other a couple were evicted from their apartment. Those adverse actions were on account of race. In both cases the complaining parties were barred, on account of race, from invoking the protection of California’s statutes; and, as a result, they were unable to lease residential property. This Court concluded that the state constitutional provision was a denial of equal protection. The Court agreed with the California Supreme Court that the amendment operated to insinuate the State into the decision to discriminate by encouraging that practice. The Court noted the “immediate design and intent” of the amendment was to “establis[h] a purported constitutional right to privately discriminate.” Id., at 374 (internal quotation marks omitted and emphasis deleted). The Court agreed that the amendment “expressly authorized and constitutionalized the private right to discriminate.” Id., at 376. The effect of the state constitutional amendment was to “significantly encourage and involve the State in private racial discriminations.” Id., at 381. In a dissent joined by three other Justices, Justice Harlan disagreed with the majority’s holding. Id., at 387. The dissent reasoned that California, by the action of its voters, simply wanted the State to remain neutral in this area, so that the State was not a party to discrimination. Id., at 389. That dissenting voice did not prevail against the majority’s conclusion that the state action in question encouraged discrimination, causing real and specific injury. The next precedent of relevance, Hunter v. Erickson, 393 U. S. 385 (1969) , is central to the arguments the respondents make in the instant case. In Hunter, the Court for the first time elaborated what the Court of Appeals here styled the “political process” doctrine. There, the Akron City Council found that the citizens of Akron consisted of “ ‘people of different race[s], . . . many of whom live in circumscribed and segregated areas, under sub-standard unhealthful, unsafe, unsanitary and overcrowded conditions, because of discrimination in the sale, lease, rental and financing of housing.’ ” Id., at 391. To address the problem, Akron enacted a fair housing ordinance to prohibit that sort of discrimination. In response, voters amended the city charter to overturn the ordinance and to require that any additional antidiscrimination housing ordinance be approved by referendum. But most other ordinances “regulating the real property market” were not subject to those threshold requirements. Id., at 390. The plaintiff, a black woman in Akron, Ohio, alleged that her real estate agent could not show her certain residences because the owners had specified they would not sell to black persons. Central to the Court’s reasoning in Hunter was that the charter amendment was enacted in circumstances where widespread racial discrimination in the sale and rental of housing led to segregated housing, forcing many to live in “ ‘unhealthful, unsafe, unsanitary and overcrowded conditions.’ ” Id., at 391. The Court stated: “It is against this background that the referendum required by [the charter amendment] must be assessed.” Ibid. Akron attempted to characterize the charter amendment “simply as a public decision to move slowly in the delicate area of race relations” and as a means “to allow the people of Akron to participate” in the decision. Id., at 392. The Court rejected Akron’s flawed “justifications for its discrimination,”justifications that by their own terms had the effect of acknowledging the targeted nature of the charter amendment. Ibid. The Court noted, furthermore, that the charter amendment was unnecessary as a general means of public control over the city council; for the people of Akron already were empowered to overturn ordinances by referendum. Id., at 390, n. 6. The Court found that the city charter amendment, by singling out antidiscrimination ordinances, “places special burden on racial minorities within the governmental process,” thus becoming as impermissible as any other government action taken with the invidious intent to injure a racial minority. Id., at 391. Justice Harlan filed a concurrence. He argued the city charter amendment “has the clear purpose of making it more difficult for certain racial and religious minorities to achieve legislation that is in their interest.” Id., at 395. But without regard to the sentence just quoted, Hunter rests on the unremarkable principle that the State may not alter the procedures of government to target racial minorities. The facts in Hunter established that invidious discrimination would be the necessary result of the procedural restructuring. Thus, in Mulkey and Hunter, there was a demonstrated injury on the basis of race that, by reasons of state encouragement or participation, became more aggravated. Seattle is the third case of principal relevance here. There, the school board adopted a mandatory busing program to alleviate racial isolation of minority students in local schools. Voters who opposed the school board’s busing plan passed a state initiative that barred busing to desegregate. The Court first determined that, although “white as well as Negro children benefit from” diversity, the school board’s plan “inures primarily to the benefit of the minority.” 458 U. S., at 472. The Court next found that “the practical effect” of the state initiative was to “remov[e] the authority to address a racial problem—and only a racial problem—from the existing decisionmaking body, in such a way as to burden minority interests” because advocates of busing “now must seek relief from the state legislature, or from the statewide electorate.” Id., at 474. The Court therefore found that the initiative had “explicitly us[ed] the racial nature of a decision to determine the decisionmaking process.” Id., at 470 (emphasis deleted). Seattle is best understood as a case in which the state action in question (the bar on busing enacted by the State’s voters) had the serious risk, if not purpose, of causing specific injuries on account of race, just as had been the case in Mulkey and Hunter. Although there had been no judicial finding of de jure segregation with respect to Seattle’s school district, it appears as though school segregation in the district in the 1940’s and 1950’s may have been the partial result of school board policies that “permitted white students to transfer out of black schools while restricting the transfer of black students into white schools.” Parents Involved in Community Schools v. Seattle School Dist. No. 1, 551 U. S. 701 –808 (2007) (Breyer, J., dissenting). In 1977, the National Association for the Advancement of Colored People (NAACP) filed a complaint with the Office for Civil Rights, a federal agency. The NAACP alleged that the school board had maintained a system of de jure segregation. Specifically, the complaint alleged “that the Seattle School Board had created or perpetuated unlawful racial segregation through, e.g., certain school-transfer criteria, a construction program that needlessly built new schools in white areas, district line-drawing criteria, the maintenance of inferior facilities at black schools, the use of explicit racial criteria in the assignment of teachers and other staff, and a general pattern of delay in respect to the implementation of promised desegregation efforts.” Id., at 810. As part of a settlement with the Office for Civil Rights, the school board implemented the “Seattle Plan,” which used busing and mandatory reassignments between elementary schools to reduce racial imbalance and which was the subject of the state initiative at issue in Seattle. See 551 U. S., at 807–812. As this Court held in Parents Involved, the school board’s purported remedial action would not be permissible today absent a showing of de jure segregation. Id., at 720–721. That holding prompted Justice Breyer to observe in dissent, as noted above, that one permissible reading of the record was that the school board had maintained policies to perpetuate racial segregation in the schools. In all events we must understand Seattle as Seattle understood itself, as a case in which neither the State nor the United States “challenge[d] the propriety of race-conscious student assignments for the purpose of achieving integration, even absent a finding of prior de jure segregation.” 458 U. S. at 472, n. 15. In other words the legitimacy and constitutionality of the remedy in question (busing for desegregation) was assumed, and Seattle must be understood on that basis. Ibid. Seattle involved a state initiative that “was carefully tailored to interfere only with desegregative busing.” Id., at 471. The Seattle Court, accepting the validity of the school board’s busing remedy as a predicate to its analysis of the constitutional question, found that the State’s disapproval of the school board’s busing remedy was an aggravation of the very racial injury in which the State itself was complicit. The broad language used in Seattle, however, went well beyond the analysis needed to resolve the case. The Court there seized upon the statement in Justice Harlan’s concurrence in Hunter that the procedural change in that case had “the clear purpose of making it more difficult for certain racial and religious minorities to achieve legislation that is in their interest.” 385 U. S., at 395. That language, taken in the context of the facts in Hunter, is best read simply to describe the necessity for finding an equal protection violation where specific injuries from hostile discrimination were at issue. The Seattle Court, however, used the language from the Hunter concurrence to establish a new and far-reaching rationale. Seattle stated that where a government policy “inures primarily to the benefit of the minority” and “minorities . . . con-sider” the policy to be “ ‘in their interest,’ ” then any state action that “place[s] effective decisionmaking authority over” that policy “at a different level of government” must be reviewed under strict scrutiny. 458 U. S., at 472, 474. In essence, according to the broad reading of Seattle, any state action with a “racial focus” that makes it “more difficult for certain racial minorities than for other groups” to “achieve legislation that is in their interest” is subject to strict scrutiny. It is this reading of Seattle that the Court of Appeals found to be controlling here. And that reading must be rejected. The broad rationale that the Court of Appeals adopted goes beyond the necessary holding and the meaning of the precedents said to support it; and in the instant case neither the formulation of the general rule just set forth nor the precedents cited to authenticate it suffice to invalidate Proposal 2. The expansive reading of Seattle has no principled limitation and raises serious questions of compatibility with the Court’s settled equal protection jurisprudence. To the extent Seattle is read to require the Court to determine and declare which political policies serve the “interest” of a group defined in racial terms, that rationale was unnecessary to the decision in Seattle; it has no support in precedent; and it raises serious constitu-tional concerns. That expansive language does not provide a proper guide for decisions and should not be deemed authoritative or controlling. The rule that the Court of Appeals elaborated and respondents seek to establish here would contradict central equal protection principles. In cautioning against “impermissible racial stereotypes,” this Court has rejected the assumption that “members of the same racial group—regardless of their age, education, economic status, or the community in which they live—think alike, share the same political interests, and will prefer the same candidates at the polls.” Shaw v. Reno, 509 U. S. 630, 647 (1993) ; see also Metro Broadcasting, Inc. v. FCC, 497 U. S. 547, 636 (1990) (Kennedy, J., dissenting) (rejecting the “demeaning notion that members of . . . defined racial groups ascribe to certain ‘minority views’ that must be different from those of other citizens”). It cannot be entertained as a serious proposition that all individuals of the same race think alike. Yet that proposition would be a necessary beginning point were the Seattle formulation to control, as the Court of Appeals held it did in this case. And if it were deemed necessary to probe how some races define their own interest in political matters, still another beginning point would be to define individuals according to race. But in a society in which those lines are becoming more blurred, the attempt to define race-based categories also raises serious questions of its own. Government action that classifies individuals on the basis of race is inherently suspect and carries the danger of perpetuating the very racial divisions the polity seeks to transcend. Cf. Ho v. San Francisco Unified School Dist., 147 F. 3d 854, 858 (CA9 1998) (school district delineating 13 racial categories for purposes of racial balancing). Were courts to embark upon this venture not only would it be undertaken with no clear legal standards or accepted sources to guide judicial decision but also it would result in, or at least impose a high risk of, inquiries and categories dependent upon demeaning stereotypes, classifications of questionable constitutionality on their own terms. Even assuming these initial steps could be taken in a manner consistent with a sound analytic and judicial framework, the court would next be required to determine the policy realms in which certain groups—groups defined by race—have a political interest. That undertaking, again without guidance from any accepted legal standards, would risk, in turn, the creation of incentives for those who support or oppose certain policies to cast the debate in terms of racial advantage or disadvantage. Thus could racial antagonisms and conflict tend to arise in the context of judicial decisions as courts undertook to announce what particular issues of public policy should be classified as advantageous to some group defined by race. This risk is inherent in adopting the Seattle formulation. There would be no apparent limiting standards defining what public policies should be included in what Seattle called policies that “inur[e] primarily to the benefit of the minority” and that “minorities . . . consider” to be “ ‘in their interest.’ ” 458 U. S., at 472, 474. Those who seek to represent the interests of particular racial groups could attempt to advance those aims by demanding an equal protection ruling that any number of matters be foreclosed from voter review or participation. In a nation in which governmental policies are wide ranging, those who seek to limit voter participation might be tempted, were this Court to adopt the Seattle formulation, to urge that a group they choose to define by race or racial stereotypes are advantaged or disadvantaged by any number of laws or decisions. Tax policy, housing subsidies, wage regulations, and even the naming of public schools, highways, and monuments are just a few examples of what could become a list of subjects that some organizations could insist should be beyond the power of voters to decide, or beyond the power of a legislature to decide when enacting limits on the power of local authorities or other governmental entities to address certain subjects. Racial division would be validated, not discouraged, were the Seattle formulation, and the reasoning of the Court of Appeals in this case, to remain in force. Perhaps, when enacting policies as an exercise of democratic self-government, voters will determine that race-based preferences should be adopted. The constitutional validity of some of those choices regarding racial preferences is not at issue here. The holding in the instant case is simply that the courts may not disempower the voters from choosing which path to follow. In the realm of policy discussions the regular give-and-take of debate ought to be a context in which rancor or discord based on race are avoided, not invited. And if these factors are to be interjected, surely it ought not to be at the invitation or insistence of the courts. One response to these concerns may be that objections to the larger consequences of the Seattle formulation need not be confronted in this case, for here race was an undoubted subject of the ballot issue. But a number of problems raised by Seattle, such as racial definitions, still apply. And this principal flaw in the ruling of the Court of Appeals does remain: Here there was no infliction of a specific injury of the kind at issue in Mulkey and Hunter and in the history of the Seattle schools. Here there is no precedent for extending these cases to restrict the right of Michigan voters to determine that race-based preferences granted by Michigan governmental entities should be ended. It should also be noted that the judgment of the Court of Appeals in this case of necessity calls into question other long-settled rulings on similar state policies. The California Supreme Court has held that a California constitutional amendment prohibiting racial preferences in public contracting does not violate the rule set down by Seattle. Coral Constr., Inc. v. City and County of San Francisco, 50 Cal. 4th 315, 235 P. 3d 947 (2010). The Court of Appeals for the Ninth Circuit has held that the same amendment, which also barred racial preferences in public education, does not violate the Equal Protection Clause. Wilson, 122 F. 3d 692 (1997). If the Court were to affirm the essential rationale of the Court of Appeals in the instant case, those holdings would be invalidated, or at least would be put in serious question. The Court, by affirming the judgment now before it, in essence would announce a finding that the past 15 years of state public debate on this issue have been improper. And were the argument made that Coral might still stand because it involved racial preferences in public contracting while this case concerns racial preferences in university admissions, the implication would be that the constitutionality of laws forbidding racial preferences depends on the policy interest at stake, the concern that, as already explained, the voters deem it wise to avoid because of its divisive potential. The instant case presents the question involved in Coral and Wilson but not involved in Mulkey, Hunter, and Seattle. That question is not how to address or prevent injury caused on account of race but whether voters may determine whether a policy of race-based preferences should be continued. By approving Proposal 2 and thereby adding §26 to their State Constitution, the Michigan voters exercised their privilege to enact laws as a basic exercise of their democratic power. In the federal system States “respond, through the enactment of positive law, to the initiative of those who seek a voice in shaping the destiny of their own times.” Bond, 564 U. S., at ––– (slip op., at 9). Michigan voters used the initiative system to bypass public officials who were deemed not responsive to the concerns of a majority of the voters with respect to a policy of granting race-based preferences that raises difficult and delicate issues. The freedom secured by the Constitution consists, in one of its essential dimensions, of the right of the individual not to be injured by the unlawful exercise of governmental power. The mandate for segregated schools, Brown v. Board of Education, 347 U. S. 483 (1954) ; a wrongful invasion of the home, Silverman v. United States, 365 U. S. 505 (1961) ; or punishing a protester whose views offend others, Texas v. Johnson, 491 U. S. 397 (1989) ; and scores of other examples teach that individual liberty has constitutional protection, and that liberty’s full extent and meaning may remain yet to be discovered and affirmed. Yet freedom does not stop with individual rights. Our constitutional system embraces, too, the right of citizens to debate so they can learn and decide and then, through the political process, act in concert to try to shape the course of their own times and the course of a nation that must strive always to make freedom ever greater and more secure. Here Michigan voters acted in concert and statewide to seek consensus and adopt a policy on a difficult subject against a historical background of race in America that has been a source of tragedy and persisting injustice. That history demands that we continue to learn, to listen, and to remain open to new approaches if we are to aspire always to a constitutional order in which all persons are treated with fairness and equal dignity. Were the Court to rule that the question addressed by Michigan voters is too sensitive or complex to be within the grasp of the electorate; or that the policies at issue remain too delicate to be resolved save by university officials or faculties, acting at some remove from immediate public scru-tiny and control; or that these matters are so arcane that the electorate’s power must be limited because the people cannot prudently exercise that power even after a full debate, that holding would be an unprecedented restriction on the exercise of a fundamental right held not just by one person but by all in common. It is the right to speak and debate and learn and then, as a matter of political will, to act through a lawful electoral process. The respondents in this case insist that a difficult question of public policy must be taken from the reach of the voters, and thus removed from the realm of public discussion, dialogue, and debate in an election campaign. Quite in addition to the serious First Amendment implications of that position with respect to any particular election, it is inconsistent with the underlying premises of a responsible, functioning democracy. One of those premises is that a democracy has the capacity—and the duty—to learn from its past mistakes; to discover and confront persisting biases; and by respectful, rationale deliberation to rise above those flaws and injustices. That process is impeded, not advanced, by court decrees based on the proposition that the public cannot have the requisite repose to discuss certain issues. It is demeaning to the democratic process to presume that the voters are not capable of deciding an issue of this sensitivity on decent and rational grounds. The process of public discourse and political debate should not be foreclosed even if there is a risk that during a public campaign there will be those, on both sides, who seek to use racial division and discord to their own political advantage. An informed public can, and must, rise above this. The idea of democracy is that it can, and must, mature. Freedom embraces the right, indeed the duty, to engage in a rational, civic discourse in order to determine how best to form a consensus to shape the destiny of the Nation and its people. These First Amendment dynamics would be disserved if this Court were to say that the question here at issue is beyond the capacity of the voters to debate and then to determine. These precepts are not inconsistent with the well-established principle that when hurt or injury is inflicted on racial minorities by the encouragement or command of laws or other state action, the Constitution requires redress by the courts. Cf. Johnson v. California, 543 U. S. 499 –512 (2005) (“[S]earching judicial review . . . is necessary to guard against invidious discrimination”); Edmonson v. Leesville Concrete Co., 500 U. S. 614, 619 (1991) (“Racial discrimination” is “invidious in all contexts”). As already noted, those were the circumstances that the Court found present in Mulkey, Hunter, and Seattle. But those circumstances are not present here. For reasons already discussed, Mulkey, Hunter, and Seattle are not precedents that stand for the conclusion that Michigan’s voters must be disempowered from acting. Those cases were ones in which the political restriction in question was designed to be used, or was likely to be used, to encourage infliction of injury by reason of race. What is at stake here is not whether injury will be inflicted but whether government can be instructed not to follow a course that entails, first, the definition of racial categories and, second, the grant of favored status to persons in some racial categories and not others. The electorate’s instruction to governmental entities not to embark upon the course of race-defined and race-based preferences was adopted, we must assume, because the voters deemed a preference system to be unwise, on account of what voters may deem its latent potential to become itself a source of the very resentments and hostilities based on race that this Nation seeks to put behind it. Whether those adverse results would follow is, and should be, the subject of debate. Voters might likewise consider, after debate and reflection, that programs designed to increase diversity—consistent with the Constitution—are a necessary part of progress to transcend the stigma of past racism. This case is not about how the debate about racial preferences should be resolved. It is about who may resolve it. There is no authority in the Constitution of the United States or in this Court’s precedents for the Judiciary to set aside Michigan laws that commit this policy determination to the voters. See Sailors v. Board of Ed. of County of Kent, 387 U. S. 105, 109 (1967) (“Save and unless the state, county, or municipal government runs afoul of a federally protected right, it has vast leeway in the management of its internal affairs”). Deliberative debate on sensitive issues such as racial preferences all too often may shade into rancor. But that does not justify removing certain court-determined issues from the voters’ reach. Democracy does not presume that some subjects are either too divisive or too profound for public debate. The judgment of the Court of Appeals for the Sixth Circuit is reversed. It is so ordered. Justice Kagan took no part in the consideration or decision of this case.

573.US.41

The Immigration and Nationality Act permits qualifying U. S. citizens and lawful permanent residents (LPRs) to petition for certain family members to obtain immigrant visas. A sponsored individual, known as the principal beneficiary, is placed into a “family preference” category based on his relationship with the petitioner. 8 U. S. C. §§1153(a)(1)–(4). The principal beneficiary’s spouse and minor children in turn qualify as derivative beneficiaries, “entitled to the same status” and “order of consideration” as the principal. §1153(d). The beneficiaries then become eligible to apply for visas in order of “priority date”—that is, the date a petition was filed. §1153(e)(1). Because the immigration process often takes years or decades to complete, a child seeking to immigrate may “age out”—i.e., reach adulthood and lose her immigration status—before she reaches the front of the visa queue. The Child Status Protection Act (CSPA) sets forth a remedy in that circumstance, providing that “[i]f the age of an alien is determined . . . to be 21 years of age or older,” notwithstanding certain allowances for bureaucratic delay, §§1153(h)(1)–(2), “the alien’s petition shall automatically be converted to the appropriate category and the alien shall retain the original priority date issued upon receipt of the original petition.” §1153(h)(3). Respondents, principal beneficiaries who became LPRs, filed petitions for their aged-out children, asserting that the newly filed petitions should receive the same priority date as their original petitions. Instead, U. S. Citizenship and Immigration Services (USCIS) gave the new petitions current priority dates. The District Court granted the Government summary judgment, deferring to the Board of Immigration Appeals’ (BIA’s) determination that only those petitions that can be seamlessly converted from one family preference category to another without the need for a new sponsor are entitled to conversion under §1153(h)(3). The en banc Ninth Circuit reversed, holding that the provision unambiguously entitled all aged-out derivative beneficiaries to automatic conversion and priority date retention. Held: The judgment is reversed, and the case is remanded. 695 F.3d 1003, reversed and remanded. Justice Kagan, joined by Justice Kennedy and Justice Ginsburg, concluded that the BIA’s textually reasonable construction of §1153(h)(3)’s ambiguous language was entitled to deference. Pp. 13–33. (a) Because §1153(h)(3) does not speak unambiguously to the issue here, a court must defer to the BIA’s reasonable interpretation. See Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 844. The first clause of §1153(h)(3) states a condition that encompasses every aged-out beneficiary of a family preference petition. The second clause, however, does not easily cohere with the first. It prescribes a remedy that can apply to only a subset of the beneficiaries described in the first clause. This remedial prescription directs immigration officials to take the alien’s petition and convert it from a category benefitting a child to an appropriate category for adults, without any change in the petition, including its sponsor, or any new filing. Moreover, this conversion is to be “automati[c]”—that is, one involving no additional decisions, contingencies, or delays. Thus, the only aliens who may benefit from §1153(h)(3)’s back half are those for whom automatic conversion is possible. The understanding that “automatic conversion” entails nothing more than picking up the petition from one category and dropping it into another for which the alien now qualifies matches the exclusive way immigration law used the term when §1153(h)(3) was enacted. See 8 CFR §204.2(i)(1)–(3) (2002). And Congress used the word “conversion” in the identical way elsewhere in the CSPA. See, e.g., §§1151(f)(2), (3). If the term meant more than that in §1153(h)(3), it would undermine the family preference system’s core premise: that each immigrant must have a qualified and willing sponsor. See §§1154(a), (b). If an original sponsor does not have a legally recognized relationship with the aged-out derivative beneficiary, another sponsor, e.g., the old principal beneficiary, must be swapped in for the alien to qualify for a new family preference category. But immigration officials cannot assume that a new sponsor is eligible and willing to petition on the alien’s behalf, given the numerous requirements the law imposes on family preference petitioners. See, e.g., §1154(a)(1)(B)(i)(II). Nei-ther can they figure out whether a valid sponsor exists unless he files and USCIS approves a new petition—the very thing §1153(h)(3) says is not required. In any case, a new qualified sponsor will rarely exist at the requisite time. An alien is deemed to age out on “the date on which an immigrant visa number became available for the alien’s parent.” §1153(h)(1)(A). Since aging out triggers automatic conversion, the date of automatic conversion is best viewed as the same. But at that time, the aged-out beneficiary’s parent cannot yet be a citizen or LPR, and so no new, qualified sponsor will be ready to step into the old one’s shoes. On the above account, §1153(h)(3)’s second clause provides a remedy to those principal and derivative beneficiaries who had a qualifying relationship with an LPR both before and after they aged out. In contrast, aliens like respondents’ children—the nieces, nephews, and grandchildren of the initial sponsors—cannot qualify for “automatic conversion”: they lacked a qualifying preference relationship with the initial petitioner, and so cannot fit into a new preference category without obtaining a new sponsor. The ambiguity created by §1153(h)(3)’s ill-fitting clauses left the BIA to choose how to reconcile the statute’s different commands. It reasonably opted to abide by the inherent limits of §1153(h)(3)’s remedial clause, rather than go beyond those limits so as to match the sweep of the first clause’s condition. When an agency thus resolves statutory tension, ordinary principles of administrative deference require this Court to defer. See National Assn. of Home Builders v. Defenders of Wildlife, 551 U.S. 644, 666. Pp. 13–22. (b) Respondents take issue with the BIA’s interpretation, but none of their contentions is persuasive. Pp. 22–33. (1) Respondents aver that every aged-out beneficiary could be automatically converted if immigration officials substituted new sponsors and managed the timing of conversion so that a new sponsor existed on the relevant date. These administrative maneuvers are not in keeping with the natural and long-established meaning of “automatic conversion,” they require conversion to occur on a date that has no connection to the alien’s aging out, and they demand administrative juggling to make automatic conversion work. And that painstakingly managed process still cannot succeed because a derivative’s parent may never become able to sponsor a visa—and immigration officials cannot practicably tell whether a given parent has done so. Pp. 22–27. (2) Respondents argue that the word “and” in the second clause of §1153(h)(3) indicates that priority date retention is a benefit wholly independent of automatic conversion. But “and” does not necessarily disjoin two phrases, and context suggests that the instructions work in tandem. In other statutory and regulatory provisions respecting “conversions,” retention of a priority date is conditional on a conversion occurring. See, e.g., §§1154(k)(1)–(3). Respondent’s reading would make priority date retention conditional on something the statute nowhere mentions. And it would engender unusual results that, without some clearer statement, the Court cannot conclude that Congress intended. Pp. 27–30. (3) Finally, respondents contend that, assuming §1153(h)(3) is ambiguous, the BIA acted unreasonably in choosing the more restrictive reading. But the BIA’s interpretation benefits from administrative simplicity and fits with immigration law’s basic first-come, first-served rule. By contrast, respondents would scramble the priority order Congress established by allowing aged-out derivative beneficiaries, like respondents’ sons and daughters, to enter the visa queue ahead of beneficiaries who had a qualifying relationship with an LPR for a far longer time. Pp. 31–33. The Chief Justice, joined by Justice Scalia, agreed that the BIA’s interpretation was reasonable, but not because an agency has authority to resolve direct conflicts within a statute. There is no conflict or internal tension in §1153(h)(3). The first clause of the provision defines the persons potentially affected, but does not grant anything to anyone. The particular benefit provided by the statute—automatic conversion and retention of priority date—is found exclusively in the second clause, and that relief requires, at minimum, that an aged-out beneficiary have his own eligible sponsor who is committed to providing financial support for the beneficiary. Beyond that, Congress did not speak clearly to which petitions can be automatically converted. The BIA’s reasonable interpretation of §1153(h)(3) is consistent with the ordinary meaning of the statutory terms, with the established meaning of automatic conversion in immigration law, and with the structure of the family-based immigration system. Pp. 1–4. Kagan, J., announced the judgment of the Court and delivered an opinion, in which Kennedy and Ginsburg, JJ., joined. Roberts, C. J., filed an opinion concurring in the judgment, in which Scalia, J., joined. Alito, J., filed a dissenting opinion. Sotomayor, J., filed a dissenting opinion, in which Breyer, J., joined, and in which Thomas, J., joined except as to footnote 3.

in which Justice Kennedy and Justice Ginsburg join. Under the Immigration and Nationality Act, 8 U. S. C. §1101 et seq., citizens and lawful permanent residents (LPRs) of the United States may petition for certain fam-ily members—spouses, siblings, and children of various ages—to obtain immigrant visas. Such a sponsored individual is known as the petition’s principal beneficiary. In turn, any principal beneficiary’s minor child—meaning an unmarried child under the age of 21—qualifies as a derivative beneficiary, “entitled to the same [immigration] status” and “order of consideration” as his parent. §1153(d). Accordingly, when a visa becomes available to the petition’s principal beneficiary, one also becomes available to her minor child. But what happens if, sometime after the relevant petition was filed, a minor child (whether a principal or a derivative beneficiary) has turned 21—or, in immigration lingo, has “aged out”? The immigration process may take years or even decades to complete, due in part to bureaucratic delays associated with reviewing immigration documents and in (still greater) part to long queues for the limited number of visas available each year. So someone who was a youngster at the start of the process may be an adult at the end, and no longer qualify for an immigration status given to minors. The Child Status Protection Act (CSPA), 116Stat. 927, ensures that the time Government officials have spent processing immigration papers will not count against the beneficiary in assessing his status. See 8 U. S. C. §1153(h)(1). But even with that provision, the beneficiary may age out solely because of the time he spent waiting in line for a visa to become available. The question presented in this case is whether the CSPA grants a remedy to all aliens who have thus outpaced the immigration process—that is, all aliens who counted as child beneficiaries when a sponsoring petition was filed, but no longer do so (even after excluding administrative delays) by the time they reach the front of the visa queue. The Board of Immigration Appeals (BIA or Board) said no. It interpreted the CSPA as providing relief to only a subset of that group—specifically, those aged-out aliens who qualified or could have qualified as principal beneficiaries of a visa petition, rather than only as derivative beneficiaries piggy-backing on a parent. We now uphold the Board’s determination as a permissible construction of the statute. I A An alien needs an immigrant visa to enter and permanently reside in the United States. See §1181(a).[1] To obtain that highly sought-after document, the alien must fall within one of a limited number of immigration cate-gories. See §§1151(a)–(b). The most favored is for the “immediate relatives” of U. S. citizens—their parents, spouses, and unmarried children under the age of 21.See §§1151(b)(2)(A)(i), 1101(b)(1). Five other categories—crucial to this case, and often denominated “preference” categories—are for “family-sponsored immigrants,” who include more distant or independent relatives of U. S. citizens, and certain close relatives of LPRs.[2] Specifically, those family preference categories are: F1: the unmarried, adult (21 or over) sons and daughters of U. S. citizens; F2A: the spouses and unmarried, minor (under 21) children of LPRs; F2B: the unmarried, adult (21 or over) sons and daughters of LPRs; F3: the married sons and daughters of U. S. citizens; F4: the brothers and sisters of U. S. citizens. §§1151(a)(1), 1153(a)(1)–(4).[3] (A word to the wise: Dog-ear this page for easy reference, because these categories crop up regularly throughout this opinion.) The road to obtaining any family-based immigrant visa begins when a sponsoring U. S. citizen or LPR files a petition on behalf of a foreign relative, termed the principal beneficiary. See §§1154(a)(1)(A)(i), (a)(1)(B)(i)(I), (b); 8 CFR §204.1(a)(1) (2014). The sponsor (otherwise knownas the petitioner—we use the words interchangeably) must provide U. S. Citizenship and Immigration Services (USCIS) with evidence showing, among other things, that she has the necessary familial relationship with thebeneficiary, see §§204.2(a)(2), (d)(2), (g)(2), and that she has not committed any conduct disqualifying her from sponsoring an alien for a visa, see, e.g., 8 U. S. C. §1154(a)(1)(B)(i)(II) (barring an LPR from submitting a petition if she has committed certain offenses against minors). USCIS thereafter reviews the petition, and approves it if found to meet all requirements. See §1154(b). For a family preference beneficiary, that approval results not in getting a visa then and there, but only in getting a place in line. (The case is different for “immediate relatives” of U. S. citizens, who can apply for and receive a visa as soon as a sponsoring petition is approved.) The law caps the number of visas issued each year in the five family preference categories, see §§1151(c)(1), 1152, 1153(a)(1)–(4), and demand regularly exceeds the supply. As a consequence, the principal beneficiary of an approved petition is placed in a queue with others in her category (F1, F2A, or what have you) in order of “priority date”—that is, the date a petition was filed with USCIS. See §1153(e)(1); 8 CFR §204.1(b); 22 CFR 42.53(a) (2013). Every month, the Department of State sets a cut-off date for each family preference category, indicating that visas (sometimes referred to by “visanumbers”) are available for beneficiaries with priority dates earlier than the cut-off. See 8 CFR §245.1(g)(1); 22 CFR §42.51(b). The system is thus first-come, first-served within each preference category, with visas becoming available in order of priority date. Such a date may benefit not only the principal beneficiary of a family preference petition, but also her spouse and minor children. Those persons, labeled the petition’s “derivative beneficiar[ies],” are “entitled to the same status, and the same order of consideration” as the principal. 8 U. S. C. §§1153(d), (h). Accordingly, when a visa becomes available for the principal, one becomes available for her spouse and minor children too. And that is so even when (as is usually but not always the case) the spouse and children would not qualify for any family preference category on their own. For example, the child of an F4 petition’s principal beneficiary is the niece or nephew of a U. S. citizen, and federal immigration law does not recognize that relationship. Nonetheless, the child can piggy-back on his qualifying parent in seeking an immigrant visa—although, as will be further discussed, he may not immigrate without her. See 22 CFR §40.1(a)(2); infra, at 6, 20–21, 31–32. Once visas become available, the principal and any derivative beneficiaries must separately file visa applications. See 8 U. S. C. §1202(a). Such an application requires an alien to demonstrate in various ways her ad-missibility to the United States. See, e.g., §1182(a)(1)(A) (alien may not have serious health problems); §1182(a)(2)(A) (alien may not have been convicted of certain crimes); §1182(a)(3)(B) (alien may not have engaged in terrorist activity). Notably, one necessary showing involves the U. S. citizen or LPR who filed the initial petition: To mitigate any possibility of becoming a “public charge,” the visa applicant (whether a principal or de-rivative beneficiary) must append an “affidavit of sup-port” executed by that sponsoring individual. §§1182(a)(4)(C)(ii), 1183a(a)(1). Such an affidavit legally commits the sponsor to support the alien, usually for at least 10 years, with an annual income “not less than125% of the federal poverty line.” §1183a(a)(1)(A); see §§1183a(a)(2)–(3). After the beneficiaries have filed their applications, a consular official reviews the documents and, if everything is in order, schedules in-person interviews. See §1202(h). The interviews for a principal and her children (or spouse) usually occur back-to-back, although those for the children may also come later.[4] The consular official will determine first whether the principal should receive a visa; if (but only if ) the answer is yes, the official will then consider the derivatives’ applications. See 22 CFR §§40.1(a)(2), 42.62, 42.81(a). Provided all goes well, everyone exits the consulate with visas in hand—but that still does not make them LPRs. See 8 U. S. C. §1154(e). Each approved alien must then travel to the United States within a set time, undergo inspection, and confirm her admissibility. See §§1201(c), 1222, 1225(a)–(b). Once again, a derivative’s fate is tied to the principal’s: If the principal cannot enter the country, neither can her children (or spouse). See §1153(d); 22 CFR §40.1(a)(2). When, but only when, an alien with an immigrant visa is approved at the border does she finally become an LPR.[5] B All of this takes time—and often a lot of it. At the front end, many months may go by before USCIS approves the initial sponsoring petition.[6] On the back end, several additional months may elapse while a consular official considers the alien’s visa application and schedules an interview.[7] And the middle is the worst. After a sponsoring petition is approved but before a visa application can be filed, a family-sponsored immigrant may stand in line for years—or even decades—just waiting for an immigrant visa to become available. See, e.g., Dept. of State, Bureau of Consular Affairs, 9 Visa Bulletin, Immigrant Numbers for December 2013 (Nov. 8, 2013). And as the years tick by, young people grow up, and thereby endanger their immigration status. Remember that not all offspring, but only those under the age of 21 can qualify as an “immediate relative” of a U. S. citizen, or as the principal beneficiary of an LPR’s F2A petition, or (most crucially here) as the derivative beneficiary of any family preference petition. See supra, at 3, 5. So an alien eligible to immigrate at the start of the process (when a sponsor files a petition) might not be so at the end (when an immigration official reviews his documents for admission). He may have “aged out” of his original immigration status by the simple passage of time. In 2002, Congress enacted the Child Status Protection Act (CSPA), 116Stat. 927, to address the treatment of those once-but-no-longer-minor aliens. One section of the Act neatly eliminates the “aging out” problem for the offspring of U. S. citizens seeking to immigrate as “immediate relatives.” Under that provision, the “determination of whether [such] an alien satisfies the [immigration law’s] age requirement . . . shall be made using [his] age” on the date the initial petition was filed. 8 U. S. C. §1151(f )(1). The section thus halts the flow of time for that group of would-be immigrants: If an alien was young when a U. S. citizen sponsored his entry, then Peter Pan-like, he remains young throughout the immigration process. A different scheme—and one not nearly so limpid—applies to the offspring of LPRs and aliens who initially qualified as either principal beneficiaries of F2A petitions or derivative beneficiaries of any kind of family preference petition. Section 3 of the CSPA, now codified at 8 U. S. C. §1153(h), contains three interlinked paragraphs that mitigate the “aging out” problem for those prospective immigrants. The first two are complex but, with some perseverance, comprehensible. The third—the key provision here—is through and through perplexing.[8] The first paragraph, §1153(h)(1), contains a formula for calculating the age of an alien “[f ]or purposes of subsections (a)(2)(A) and (d)”—that is, for any alien seeking an immigrant visa directly under F2A or as a derivative beneficiary of any preference category. The “determination of whether [such] an alien satisfies the [immigration law’s] age requirement”—that is, counts as under 21—“shall be made using— “(A) the age of the alien on the date on which an immigrant visa number becomes available for such alien (or, in the case of [derivative beneficiaries], the date on which an immigrant visa number became available for the alien’s parent) . . . ; reduced by “(B) the number of days in the period during which the applicable petition described in paragraph (2) was pending.” §1153(h)(1). The cross-referenced second paragraph, §1153(h)(2), then explains that the “applicable petition” mentioned is the petition covering the given alien—so again, either an F2A petition filed on his own behalf or any petition extending to him as a derivative. Taken together, those two paragraphs prevent an alien from “aging out” because of—but only because of—bureaucratic delays: the time Government officials spend reviewing (or getting around to reviewing) paperwork at what we have called the front and back ends of the immigration process. See supra, at 6–7. The months that elapse before USCIS personnel approve a family preference petition (“the period during which the applicable petition described in paragraph (2) was pending”) do not count against an alien in determining his statutory “age.” Neither do the months a consular officer lets pass before adjudicating the alien’s own visa application (the period after “an immigrant visa number becomes available for such alien (or . . . [his] parent)”). But the time in between—the months or, more likely, years the alien spends simply waiting for a visa to become available—is not similarly excluded in calculating his age: Every day the alien stands in that line is a day he grows older, under the immigration laws no less than in life. And so derivative beneficiaries, as well as principal beneficiaries of F2A petitions, can still “age out”—in other words, turn 21, notwithstanding §1153(h)(1)’s dual age adjustments—prior to receiving an opportunity to immigrate. What happens then (if anything) is the subject of §1153(h)’s third paragraph—the provision at issue in this case. That paragraph states: “If the age of an alien is determined under paragraph (1) to be 21 years of age or older for the purposes of subsections (a)(2)(A) and (d) of this section, thealien’s petition shall automatically be converted to the appropriate category and the alien shall retain the original priority date issued upon receipt of the original petition.” The provision thus first references the aged-out beneficiaries of family preference petitions, and then directs immigration officials to do something whose meaning this opinion will further consider—i.e., “automatically convert” an alien’s petition to an “appropriate category.” The Board of Immigration Appeals (BIA) addressed the meaning of §1153(h)(3) in Matter of Wang, 25 I. & N. Dec. 28 (2009); its interpretation there is what we review in this case. Wang was the principal beneficiary of an F4 petition that his sister, a U. S. citizen, filed in 1992. At that time, Wang’s daughter was 10 years old, and thus qualified as a derivative beneficiary. But Wang waited in line for a visa for more than a decade, and by the time his priority date finally came up, his daughter had turned 22 (even after applying §1153(h)(1)’s age-reduction formula). Wang thus obtained a visa for himself, boarded a plane alone, and entered the United States as an LPR. He then filed a new preference petition on his daughter’s behalf—this one under F2B, the category for LPRs’ adult sons and daughters. USCIS approved that petition, with a priority date corresponding to the date of Wang’s filing. Wang contended that under §1153(h)(3), his daughter was instead entitled to “retain the original priority date” given to his sister’s old F4 petition, because that petition could “automatically be converted” to the F2B category. The Board rejected that argument. It explained that “the language of [§1153(h)(3)] does not expressly state which petitions qualify for automatic conversion and retention of priority dates.” Id., at 33. Given that “ambiguity,” the BIA looked to the “recognized meaning” of “the phrase ‘automatic conversion’ ” in immigration statutes and regulations—which it “presume[d]” Congress understood when enacting the CSPA. Id., at 33–35. “Historically,” the BIA showed, that language applied only when apetition could move seamlessly from one family preference category to another—not when a new sponsor was needed to fit a beneficiary into a different category. Id., at 35. Some aged-out aliens’ petitions could accomplish that maneuver, because the alien had a qualifying relationship with the original sponsor, and continued to do so upon aging out; in that event, the Board held, §1153(h)(3) ensured that the alien would retain his original priority date. See id., at 34–35. But the F4 petition filed by Wang’s sister could not “automatically be converted” in that way because Wang’s daughter never had a qualifying relationship with the sponsor: “[N]o category exists for the niece of a United States citizen.” Id., at 35–36. That is why Wang himself had to file a new petition on his daughter’s behalf once she aged out and could no longer ride on his sibling status. The Board saw no evidence that Congress meant “to expand the use of the concept[ ] of automatic conversion” to reach such a case. Id., at 36. And the Board thought such an expansion unwarranted because it would allow aliens like Wang’s daughter, who lacked any independent entitlement to a visa during the years her father spent standing on the F4 queue, to “cut[ ] in line ahead of others awaiting visas in other preference categories.” Id., at 38. C The respondents in this case are similarly situated to Wang, and they seek the same relief. Each was once the principal beneficiary of either an F3 petition filed by a U. S. citizen parent or an F4 petition filed by a U. S. citizen sibling. Each also has a son or daughter who, on the date of filing, was under 21 and thus qualified as a derivative beneficiary of the petition. But as was true of Wang’s daughter, the respondents’ offspring had all turned 21 (even accounting for §1153(h)(1)’s age adjustments) by the time visas became available. Accordingly, the respondents immigrated to the United States alone and, as new LPRs, filed F2B petitions for their sons and daughters. Each argued that under §1153(h)(3), those petitions should get the same priority date as the original F3 and F4 petitions once had. USCIS instead gave the new F2B petitions current priority dates, meaning that the sons and daughters could not leapfrog over others in the F2B line. This case began as two separate suits, one joining many individual plaintiffs and the other certified as a class action. In each suit, the District Court deferred to the BIA’s interpretation of §1153(h)(3) in Wang, and accordingly granted summary judgment to the Government. See Zhang v. Napolitano, 663 F. Supp. 2d 913, 919 (CD Cal. 2009); Costelo v. Chertoff, No. SA08–00688, 2009 WL 4030516 (CD Cal., Nov. 10, 2009). After consolidating the two cases on appeal, a panel of the Ninth Circuit affirmed: Like the lower courts, it found §1153(h)(3) ambiguous and acceded to the BIA’s construction. 656 F. 3d 954, 965–966 (2011). The Ninth Circuit then granted rehearing en banc and reversed in a 6-to-5 decision. 695 F. 3d 1003 (2012). The majority concluded that “the plain language of the CSPA unambiguously grants automatic conversion and priority date retention to [all] aged-out derivative beneficiaries,” and that the Board’s contrary conclusion “is not entitled to deference.” Id., at 1006. We granted certiorari, 570 U. S. ___ (2013), to resolve a Circuit split on the meaning of §1153(h)(3),[9] and we now reverse the Ninth Circuit’s decision. II Principles of Chevron deference apply when the BIA interprets the immigration laws. See Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 –844 (1984); INS v. Aguirre-Aguirre, 526 U. S. 415 –425 (1999). Indeed, “judicial deference to the Executive Branch is especially appropriate in the immigration context,” where decisions about a complex statu-tory scheme often implicate foreign relations. Id., at 425. (Those hardy readers who have made it this far will surely agree with the “complexity” point.) Under Chevron, the statute’s plain meaning controls, whatever the Board might have to say. See 467 U. S., at 842–843. But if the law does not speak clearly to the question at issue, a court must defer to the Board’s reasonable interpretation, rather than substitute its own reading. Id., at 844. And §1153(h)(3) does not speak unambiguously to the issue here—or more precisely put, it addresses that issue in divergent ways. We might call the provision Janus-faced. Its first half looks in one direction, toward the sweeping relief the respondents propose, which would reach every aged-out beneficiary of a family preference petition. But as the BIA recognized, and we will further explain, the section’s second half looks another way, toward a remedy that can apply to only a subset of those beneficiaries—and one not including the respondents’ offspring. The two faces of the statute do not easily cohere with each other: Read either most naturally, and the other appears to mean not what it says. That internal tension makes possible alternative reasonable constructions, bringing into correspondence in one way or another the section’s different parts. And when that is so, Chevron dictates that a court defer to the agency’s choice—here, to the Board’s expert judgment about which interpretation fits best with, and makes most sense of, the statutory scheme. Begin by reading the statute from the top—the part favoring the respondents. Section 1153(h)(3)’s first clause—“If the age of an alien is determined under paragraph (1) to be 21 years of age or older for the purposes of subsections (a)(2)(A) and (d)”—states a condition that every aged-out beneficiary of a preference petition satisfies. That is because all those beneficiaries have had their ages “determined under paragraph (1)” (and have come up wanting): Recall that the age formula of §1153(h)(1) applies to each alien child who originally qualified (under “subsections (a)(2)(A) and (d)”) as the principal beneficiary of an F2A petition or the derivative beneficiary of any family preference petition. On its own, then, §1153(h)(3)’s opening clause encompasses the respondents’ sons and daughters, along with every other once-young beneficiary of a family preference petition now on the wrong side of 21. If the next phrase said something like “the alien shall be treated as though still a minor” (much as the CSPA did to ensure U. S. citizens’ children, qualifying as “immediate relatives,” would stay forever young, see supra, at 7–8), all those aged-out beneficiaries would prevail in this case. But read on, because §1153(h)(3)’s second clause instead prescribes a remedy containing its own limitation on the eligible class of recipients. “[T]he alien’s petition,” that part provides, “shall automatically be converted to the appropriate category and the alien shall retain the original priority date.” That statement directs immigration officials to take the initial petition benefitting an alien child, and now that he has turned 21, “convert[ ]” that same petition from a category for children to an “appropriate category” for adults (while letting him keep the old priority date). The “conversion,” in other words, is merely from one category to another; it does not entail any change in the petition, including its sponsor, let alone any new filing. And more, that category shift is to be “automatic”—that is, one involving no additional decisions, contingencies, or delays. See, e.g., Random House Webster’s Unabridged Dictionary 140 (2d ed. 2001) (defining “automatic” as “having the capability of starting, operating, moving, etc., independently”); The American Heritage Dictionary 122 (4th ed. 2000) (“[a]cting or operating in a manner essentially independent of external influence”). The operation described is, then, a mechanical cut-and-paste job—moving a petition, without any substantive alteration, from one (no-longer-appropriate, child-based) category to another (now-appropriate, adult) compartment. And so the aliens who may benefit from §1153(h)(3)’s back half are only those for whom that procedure is possible. The clause offers relief not to every aged-out beneficiary, but just to those covered by petitions that can roll over, seamlessly and promptly, into a category for adult relatives. That understanding of §1153(h)(3)’s “automatic conversion” language matches the exclusive way immigration law used the term when Congress enacted the CSPA. For many years before then (as today), a regulation entitled “Automatic conversion of preference classification” instructed immigration officials to change the preference category of a petition’s principal beneficiary when either his or his sponsor’s status changed in specified ways. See 8 CFR §§204.2(i)(1)–(3) (2002). For example, the regulation provided that when a U. S. citizen’s child aged out, his “immediate relative” petition converted to an F1 petition, with his original priority date left intact. See §204.2(i)(2). Similarly, when a U. S. citizen’s adult son married, his original petition migrated from F1 to F3, see §204.2(i) (1)(i); when, conversely, such a person divorced, his petition converted from F3 to F1, see §204.2(i)(1)(iii); and when a minor child’s LPR parent became a citizen, his F2A petition became an “immediate relative” petition, see §204.2(i)(3)—all again with their original priority dates. Most notable here, what all of those authorized changes had in common was that they could occur without any change in the petitioner’s identity, or otherwise in the petition’s content. In each circumstance, the “automatic conversion” entailed nothing more than picking up the petition from one category and dropping it into another for which the alien now qualified.[10] Congress used the word “conversion” (even without the modifier “automatic”) in the identical way in two other sections of the CSPA. See Law v. Siegel, 571 U. S. ___, ___ (slip op., at 7) (2014) (“[W]ords repeated in different parts of the same statute generally have the same meaning”). Section 2 refers to occasions on which, by virtue of the above-described regulation, a petition “converted” from F2A to the “immediate relative” category because of the sponsor parent’s naturalization, or from the F3 to theF1 box because of the beneficiary’s divorce. 8 U. S. C. §§1151(f )(2), (3). Then, in §6, Congress authorized an additional conversion of the same nature: It directed that when an LPR parent-sponsor naturalizes, the petition he has filed for his adult son or daughter “shall be converted,” unless the beneficiary objects, from the F2B to the F1 compartment—again with the original priority date unchanged. 8 U. S. C. §§1154(k)(1)–(3). (That opt-out mechanism itself underscores the otherwise mechanical nature of the conversion.) Once again, in those cases, all that is involved is a recategorization—moving the same petition, filed by the same petitioner, from one preference classification to another, so as to reflect a change in either the alien’s or his sponsor’s status. In the rest of the CSPA, as in the prior immigration regulation, that is what “conversion” means. And if the term meant more than that in §1153(h)(3), it would undermine the family preference system’s core premise: that each immigrant must have a qualified sponsor. Consider the alternative addressed in Wang—if “automatic conversion” were also to encompass the substitution of a new petitioner for the old one, to make sure the aged-out alien’s petition fits into a new preference category. In a case like Wang, recall, the original sponsor doesnot have a legally recognized relationship with the aged-out derivative beneficiary (they are aunt and niece); accordingly, the derivative’s father—the old principal beneficiary—must be swapped in as the petitioner to enable his daughter to immigrate. But what if, at that point, the father is in no position to sponsor his daughter? Suppose he decided in the end not to immigrate, or failed to pass border inspection, or died in the meanwhile. Or suppose he entered the country, but cannot sponsor a relative’s visa because he lacks adequate proof of parentage or committed a disqualifying crime. See §1154(a)(1)(B)(i)(II); 8 CFR §204.2(d)(2); supra, at 4. Or suppose he does not want to—or simply cannot—undertake the significant financial obligations that the law imposes on someone petitioning for an alien’s admission. See 8 U. S. C. §§1183a(a)(1)(A), (f )(1)(D); supra, at 5. Immigration officials cannot assume away all those potential barriers to entry: That would run counter to the family preference system’s insistence that a qualified and willing sponsor back every immigrant visa. See §§1154(a)–(b). But neither can they easily, or perhaps at all, figure out whether such a sponsor exists unless he files and USCIS approves a new petition—the very thing §1153(h)(3) says is not required. Indeed, in cases like Wang, the problem is broader: Under the statute’s most natural reading, a new qualified sponsor will hardly ever exist at the moment the petition is to be “converted.” Section 1153(h)(3), to be sure, does not explicitly identify that point in time. But §1153(h)(1) specifies the date on which a derivative beneficiary is deemed to have either aged out or not: It is “the date on which an immigrant visa number became available for the alien’s parent.” See §§1153(h)(1)(A)–(B). Because that statutory aging out is the one and only thing that triggers automatic conversion for eligible aliens, the date of conversion is best viewed as the same. That reading, more-over, comports with the “automatic conversion” regulation on which Congress drew in enacting the CSPA, see supra, at 16–17: The rule authorizes conversions “upon” or “as of the date” of the relevant change in the alien’s status (including turning 21))—regardless when USCIS may receive notice of the change. 8 CFR §204.2(i); but cf. post, at 14 (Sotomayor, J., dissenting) (wrongly stating that under that rule conversion occurs upon the agency’s receipt of proof of the change). But on that date, no new petitioner will be ready to step into the old one’s shoes if such a substitution is needed to fit an aged-out beneficiary into a different category. The beneficiary’s parent, on the day a “visa number became available,” cannot yet be an LPR or citizen; by definition, she has just become eligible to apply for a visa, and faces a wait of at least several months before she can sponsor an alien herself. Nor, except in a trivial number of cases, is any hitherto unidentified person likely to have a legally recognized relationship to the alien. So if an aged-out beneficiary has lost his qualifying connection to the original petitioner, no conversion to an “appropriate category” can take place at the requisite time. As long as immigration law demands some valid sponsor, §1153(h)(3) cannot give such an alien the designated relief. On the above account—in which conversion entails a simple reslotting of an original petition into a now-appropriate category—§1153(h)(3)’s back half provides a remedy to two groups of aged-out beneficiaries. First, any child who was the principal beneficiary of an F2A petition (filed by an LPR parent on his behalf) can take advantage of that clause after turning 21. He is, upon aging out, the adult son of the same LPR who sponsored him as a child; his petition can therefore be moved seamlessly—without the slightest alteration or delay—into the F2B category. Second, any child who was the derivative beneficiary of an F2A petition (filed by an LPR on his spouse’s behalf) can similarly claim relief, provided that under the statute, he is not just the spouse’s but also the petitioner’s child.[11] Such an alien is identically situated to the aged-out principal beneficiary of an F2A petition; indeed, for the price of another filing fee, he could just as easily have been named a principal himself. He too is now the adult son of the original LPR petitioner, and his petition can also be instantly relabeled an F2B petition, without any need to substitute a new sponsor or make other revisions. In each case, the alien had a qualifying relationship before he was 21 and retains it afterward; all that must be changed is the label affixed to his petition.[12] In contrast, as the Board held in Wang, the aged-out derivative beneficiaries of the other family preference categories—like the sons and daughters of the respondents here—cannot qualify for “automatic conversion.” Recall that the respondents themselves were principal beneficiaries of F3 and F4 petitions; their children, when under 21, counted as derivatives, but lacked any qualifying preference relationship of their own. The F3 derivatives were the petitioners’ grandsons and granddaughters; the F4 derivatives their nephews and nieces; and none of those are relationships Congress has recognized as warranting a family preference. See 8 U. S. C. §§1153(a)(3)–(4). Now that the respondents’ children have turned 21, and they can no longer ride on their parents’ coattails, that lack of independent eligibility makes a difference. For them, unlike for the F2A beneficiaries, it is impossible simply to slide the original petitions from a (no-longer-appropriate) child category to a (now-appropriate) adult one. To fit into a new category, those aged-out derivatives, like Wang’s daughter, must have new sponsors—and for all the reasons already stated, that need means they cannot benefit from “automatic conversion.” All that said, we hold only that §1153(h)(3) permits—not that it requires—the Board’s decision to so distinguish among aged-out beneficiaries. That is because, as we explained earlier, the two halves of §1153(h)(3) face in different directions. See supra, at 14. Section 1153(h)(3)’s first part—its conditional phrase—encompasses every aged-out beneficiary of a family preference petition, and thus points toward broad-based relief. But as just shown, §1153(h)(3)’s second part—its remedial prescription—applies only to a narrower class of beneficiaries: those aliens who naturally qualify for (and so can be “automatically converted” to) a new preference classification when they age out. Were there an interpretation that gave each clause full effect, the Board would have been required to adopt it. But the ambiguity those ill-fitting clauses create instead left the Board with a choice—essentially of how to reconcile the statute’s different commands. The Board, recognizing the need to make that call, opted to abide by the inherent limits of §1153(h)(3)’s remedial clause, rather than go beyond those limits so as to match the sweep of the section’s initial condition. On the Board’s reasoned view, the only beneficiaries entitled to statutory relief are those capable of obtaining the remedy designated. When an agency thus resolves statutory tension, ordinary principles of administrative deference require us to defer. See National Assn. of Home Builders v. Defenders of Wildlife, 551 U. S. 644, 666 (2007) (When a statutory scheme contains “a fundamental ambiguity” arising from “the differing mandates” of two provisions, “it is appropriate to look to the implementing agency’s expert interpretation” to determine which “must give way”). III The respondents urge us to overturn the Board’s judgment for three independent reasons. First, and principally, they take issue with the Board’s—and now our—viewof the limits associated with “automatic conversion”: They argue that every aged-out beneficiary’s petition can “automatically be converted” to an “appropriate category,” and that the two halves of §1153(h)(3) are thus reconcilable. Second, the respondents contend that even if “automatic conversion” does not extend so far, §1153(h)(3) separately entitles each such beneficiary to the benefit of his original petition’s priority date. And third, they claim that the Board’s way of resolving whatever ambiguity inheres in §1153(h)(3) is arbitrary and capricious. The dissenting opinion reiterates the first two arguments, though with slight variation and in opposite order, while forgoing the third. See post, at 9–19 (opinion of Sotomayor, J.) (hereinafter the dissent). We find none of the contentions persuasive. A The respondents (and the dissent) initially aver that every aged-out beneficiary (including their own sons and daughters) can “automatically be converted” to an “appropriate” immigration category, if only immigration officials try hard enough. The Government, in the respondents’ view, can accomplish that feat by substituting new sponsors for old ones, and by “managing the timing” of every conversion to ensure such a new petitioner exists on the relevant date. Brief for Respondents 33. And because, the respondents say, it is thus possible to align the two halves of §1153(h)(3)—even if through multiple administrative maneuvers—immigration officials are under an obligation to do so. We disagree, for reasons that should sound familiar: Several are the same as those we have just given for upholding the Board’s interpretation. But still, we walk through the respondents’ argument step-by-step, to show how far it departs from any ordinary understanding of “automatic conversion.” The first (and necessary) premise of that argument does not augur well for the remainder: It is the view that the “automatic conversion” procedure permits a change in the petitioner’s identity. According to the respondents, the aged-out beneficiaries’ parents, upon becoming LPRs, can be subbed in for the original sponsors (i.e., the beneficiaries’ grandparents, aunts, and uncles), and the petitions then converted to the F2B category. But as we have shown, the “automatic conversion” language—as most naturally read and as long used throughout immigration law—contemplates merely moving a petition into a new and valid category, not changing its most essential feature. See supra, at 15–17. That alone defeats the respondents’ position. And a further problem follows—this one concerning the date of automatic conversion. The respondents need that date to come at a time when the derivative beneficiaries’ parents (the substitute petitioners) are already living in the United States as LPRs; otherwise, the petitions could not qualify for the F2B box. In an attempt to make that possible, the respondents propose that conversion be viewed as taking place when “the derivative beneficiary’s visa . . . application is adjudicated.” Brief for Respondents 29. But as we have (again) demonstrated, the statute is best read as establishing a different date: that “on which an immigrant visa number became available for the alien’s parent”—when, by definition, the parent is not yet an LPR. §1153(h)(1); see supra, at 18–19. That is the moment when a derivative ages out, which is the single change conversion reflects. By contrast, the respondents’ suggested date has no connection to that metamorphosis; the date of adjudication is merely when an immigration official later discovers that a child has turned 21. And that date is itself fortuitous, reflecting no more than when an immigration officer got around to reviewing a visa application: The possibility of conversion would thus depend on bureaucratic vagaries attending the visa process. So the respondents’ mistaken view of the timing of conversion is another off-ramp from their argument.[13] Yet there is more—because even after substituting a new petitioner and delaying the conversion date in a way the statute does not contemplate, the respondents must propose yet further fixes to make “automatic” conversion work for their sons and daughters. The respondents’ next problem is that even on the conversion date they propose, most of them (and other derivatives’ parents) were not yet LPRs, and so could not possibly be sponsors. In the ordinary course, principal and derivative beneficiaries living abroad apply for their visas at the same time and go to the consulate together for back-to-back interviews. See supra, at 6. And even if the parent is approved first, that alone does not make her an LPR; she still must come to this country, demonstrate her continued eligibility, and pass an inspection. See ibid. Thus, the respondents must recommend changes to the visa process to get the timing to work—essentially, administrative juggling to hold off the derivative beneficiary’s visa adjudication until his parent has become an LPR. In particular, they suggest that the consular official defer the derivative’s interview, or that the official nominally “reject the application” and then instruct the derivative to “reapply after the principal beneficiary immigrates.” Brief for Respondents 30. But the need for that choreography (which, in any event, few if any of the respondents conformed to) renders the conversion process only less “automatic,” because now it requires special intervention, purposeful delay, and deviation from standard administrative practice. Conversion has become not a machine that would go of itself, but a process painstakingly managed. And after all this fancy footwork, the respondents’ scheme still cannot succeed, because however long a visa adjudication is postponed, a derivative’s parent may never become able to sponsor a relative’s visa—and immigration officials cannot practicably tell whether a given parent has done so. We have noted before the potential impediments to serving as a petitioner—including that a parent may not immigrate, may not qualify as a sponsor, or may not be able to provide the requisite financial support. See supra, at 17–18. The respondents offer no way to deal with those many contingencies. Require the parent to submit a new petition? But the entire point of automatic conversion (as the respondents themselves agree) is to obviate the need for such a document. See Brief for Respondents 30, 42. Investigate the parent’s eligibility in some other way? But even were that possible (which we doubt) such an inquiry would not square with the essential idea of an automatic process. Disregard the possibility that no legal sponsor exists? But then visas would go, inevitably and not infrequently, to ineligible aliens. And so the workarounds have well and truly run out on the respondents’ argument.[14] That leaves us with the same statutory inconsistency with which we began. Having followed each step of the respondents’ resourceful (if Rube Goldbergish) argument, we still see no way to apply the concept of automatic conversion to the respondents’ children and others like them. And that means we continue to face a statute whose halves do not correspond to each other—giving rise to an ambiguity that calls for Chevron deference. B The respondents, however, have another idea for reconciling §1153(h)(3)’s front and back parts (and this back-up claim becomes the dissent’s principal argument). Recall that the section’s remedial clause instructs that “the alien’s petition shall automatically be converted to the appropriate category and the alien shall retain the original priority date issued upon receipt of the original petition.” The respondents (and the dissent) ask us to read the italicized language as conferring a benefit wholly independent of automatic conversion. On that view, aged-out derivatives, even though ineligible for conversion, could “retain the[ir] original priority date[s]” if their parents file a new petition (as the respondents in fact did here “as a protective matter,” Tr. of Oral Arg. 55). And then, everyone encompassed in §1153(h)(3)’s first clause would get at least some form of relief (even if not both forms) from the section’s second. For this argument, the respondents principally rely on the word “and”: “Where the word ‘and’ connects two” phrases as in §1153(h)(3)’s back half, the respondents contend, those terms “operate independently.” Brief for Respondents 39; see post, at 9. But the conjunction “and” does not necessarily disjoin two phrases in the way the respondents say. In some sentences, no doubt, the respondents have a point. They use as their primary example: “[I]f the boat takes on water, then you shall operate the bilge pump and you shall distribute life jackets.” Brief for Respondents 39; see also post, at 10 (offering further examples). We agree that “you shall distribute life jackets” functions in that sentence as an independent command. But we can come up with many paired dictates in which the second is conditional on the first. “If the price is reasonable, buy two tickets and save a receipt.” “If you have time this summer, read this book and give me a report.” Or, shades of this case: “If your cell-phone contract expires, buy a new phone and keep the old number.”[15] In each case, the second command functions only once the first is accomplished. Whether “and” works in that way or in the respondents’ depends, like many questions of usage, on the context. See, e.g., Caraco Pharmaceutical Laboratories, Ltd. v. Novo Nordisk A/S, 566 U. S. ___, ___ (2012). Here, we think, context compels the Board’s view that the instructions work in tandem. The first phrase instructs immigration officials to convert a petition (when an “appropriate category” exists); the next clarifies that such a converted petition will retain the original priority date, rather than receive a new one corresponding to the date of conversion. That reading comports with the way retention figures in other statutory and regulatory provisions respecting “conversions”; there too, retention of a priority date is conditional on a conversion occurring. See 8 U. S. C. §§1154(k)(1)–(3); 8 CFR §204.2(i); supra, at 16. The respondents wish to unhook the “retention” phrase from that mooring, and use it to explain what will attend a different event—that is, the filing of a new petition. But that is to make “retention” conditional on something the statute nowhere mentions—a highly improbable thing for Congress to have done. (If, once again, a teacher says to “read this book and give me a report,” no one would think he wants a report on some unidentified subject.) And indeed, the respondents’ and dissent’s own examples prove this point: In not a single one of their proffered sentences is the second command contingent on the occurrence of some additional, unstated event, as it would have to be under the respondents’ construction of §1153(h)(3); rather, each such command (e.g., “distribute life jackets”) flows directly from the stated condition (e.g., “if the boat takes on water”). So by far the more natural understanding of §1153(h)(3)’s text is that retention follows conversion, and nothing else. The respondents’ contrary view would also engender unusual results, introducing uncertainty into the immigration system’s operation and thus interfering with statutory goals. Were their theory correct, an aged-out alien could hold on to a priority date for years or even decades while waiting for a relative to file a new petition. Even if that filing happened, say, 20 years after the alien aged out, the alien could take out his priority-date token, and assert a right to spring to the front of any visa line. At that point, USCIS could well have a hard time confirming the old priority date, in part because the names of derivative beneficiaries need not be listed on a visa petition. And the possibility of such leap-frogging from many years past would impede USCIS’s publication of accurate waiting times. As far as we know, immigration law nowhere else allows an alien to keep in his pocket a priority date untethered to any existing valid petition. Without some clearer statement, we cannot conclude Congress intended here to create such a free-floating, open-ended entitlement to a defunct petition’s priority date. See Wang, 25 I. & N. Dec., at 36.[16] C Finally, the respondents contend that even if §1153(h)(3) points at once in two directions—toward a broader scope in its first half and a narrower one in its second—the BIA acted unreasonably in choosing the more restrictive reading. In their view, the Board has offered no valid reason, consistent with “the purposes and concerns of the immigration laws,” to treat their own sons and daughters less favorably than aliens who were principal and derivative beneficiaries of F2A petitions. Brief for Respondents 47. Indeed, the respondents suggest that the BIA, “for its own unfathomable reasons, disapproves of Congress’s decision to allow any aged-out” aliens to get relief, and has thus “limited [§1153(h)(3)] to as few derivative beneficiaries as possible.” Id., at 55. We cannot agree. At the least, the Board’s interpretation has administrative simplicity to recommend it. Under that view, immigration authorities need only perform the kind of straightforward (i.e., “automatic”) conversion they have done for decades—moving a petition from one box to another to reflect a given status change like aging out. See Wang, 25 I. & N. Dec., at 36. The respondents, as we have shown, would transform conversion into a managed, multi-stage process, requiring immigration and consular officials around the world to sequence and delay every aged-out alien’s visa adjudication until they are able to confirm that one of his parents had become a qualifying and willing F2B petitioner. And according to the Government’s (incomplete) statistics, that would have to happen in, at a minimum, tens of thousands of cases every year. See Reply Brief 18, n. 13. Still more important, the Board offered a cogent argument, reflecting statutory purposes, for distinguishing between aged-out beneficiaries of F2A petitions and the respondents’ sons and daughters. See Wang, 35 I. & N. Dec., at 38. As earlier explained, the F2A beneficiaries have all had a qualifying relationship with an LPR for the entire period they have waited in line—i.e., since their original priority dates. See supra, at 19–20. That means that when immigration authorities convert their petitions, they will enter the F2B line at the same place as others who have had a comparable relationship for an equal time. The conversion thus fits with the immigration law’s basic first-come-first-served rule. See 8 U. S. C. §1153(e); supra, at 4. By contrast, the derivative beneficiaries of F3 and F4 petitions, like the respondents’ sons and daughters, lacked any qualifying relationship with a citizen or LPR during the period they waited in line. See supra, at 20–21. They were, instead, the grandchildren, nieces, or nephews of citizens, and those relationships did not independently entitle them to visas. If such aliens received relief under §1153(h)(3), they would jump over thousands of others in the F2B line who had a qualifying relationship with an LPR for a far longer time. That displacement would, the Board reasonably found, scramble the priority order Congress prescribed. The argument to the contrary assumes that the respondents’ sons and daughters should “receive credit” for all the time the respondents themselves stood in line. Brief for Respondents 50. But first, the time the respondents spent waiting for a visa may diverge substantially from the time their children did. Suppose, for example, that one of the respondents had stood in the F4 queue for 15 years, and with just 4 years to go, married someone with a 17-year-old son. Under the respondents’ reading, that derivative beneficiary, after aging out, would get the full benefit of his new parent’s wait, and so displace many thousands of aliens who (unlike him) had stood in an immigration queue for nearly two decades. And second, even when the derivative qualified as such for all the time his parent stood in line, his status throughout that period hinged on his being that parent’s minor child. If his parent had obtained a visa before he aged out, he would have been eligible for a visa too, because the law does not demand that a prospective immigrant abandon a minor child. But if the parent had died while waiting for a visa, or had been found ineligible, or had decided not to immigrate after all, the derivative would have gotten nothing for the time spent in line. See supra, at 5–6. Similarly, the Board could reasonably conclude, he should not receive credit for his parent’s wait when he has become old enough to live independently. In the unavoidably zero-sum world of allocating a limited number of visas, the Board could decide that he belongs behind any alien who has had a lengthier stand-alone entitlement to immigrate. IV This is the kind of case Chevron was built for. What-ever Congress might have meant in enacting §1153(h)(3), it failed to speak clearly. Confronted with a self-contradictory, ambiguous provision in a complex statutory scheme, the Board chose a textually reasonable construction consonant with its view of the purposes and policies underlying immigration law. Were we to overturn the Board in that circumstance, we would assume as our own the responsible and expert agency’s role. We decline that path, and defer to the Board. We therefore reverse the judgment of the Ninth Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 An alien already in the United States—for example, on a student or temporary worker visa—must obtain “adjustment of status” rather than an immigrant visa to become a lawful permanent resident. See . Because the criteria for securing adjustment of status and obtaining an immigrant visa are materially identical, we use the single term “immigrant visa” to refer to both. 2 The “family preference” label, as used by immigration officials, applies only to these five classifications, and not to the category for “im-mediate relatives” of U. S. citizens. See Brief for Petitioners 3, n. 1. 3 Immigrant visas can also go to aliens with special, marketable skills, see §§1151(a)(2), 1153(b), or to aliens from countries with historically low immigration to the United States, see §§1151(a)(3), 1153(c). None of the respondents here sought visas under those “employment-based” or “diversity” categories. 4 See Dept. of State, The Immigrant Visa Process: Visa Applicant Interview, online at http://travel.state.gov/content/visas/english/immigrate/immigrant-process/interview/applicant_interview.html (all Internet materials as visited June 5, 2014, and available in Clerk of Court’s case file). 5 The last part of the immigration process is streamlined for aliens already residing in the United States who have applied for adjustment of status. See n. 1, . The immigration officer interviewing such an alien, upon finding her visa-eligible, may declare her an LPR on the spot. See . But here too, the officer will not make a derivative beneficiary an LPR unless and until he approves that status for the principal. See 22 CFR §40.1(a)(2). 6 See USCIS, Processing Time Information, online at https://egov.uscis.gov/cris/processingTimesDisplayInit.do. 7 See The Immigrant Visa Process: Interview, online at http://travel.state.gov/content / visas / english / immigrate / immigrant-process /interview.html. 8 The full text of these three paragraphs, for the masochists among this opinion’s readers, is as follows: 9 Compare 695 F. 3d 1003, 1006 (CA9 2012) (case below) (holding that §1153(h)(3) extends relief to all aged-out derivative beneficiaries); v. , 655 F. 3d 363, 365 (CA5 2011) (same), with v. , 654 F. 3d 376, 385 (CA2 2011) (holding that §1153(h)(3) not merely permits, but requires the Board’s contrary interpretation). 10 The dissent responds to this fact only with a pair of non-sequiturs. at 18–19 (opinion of ). First, the dissent cites a statutory provision that does not use the word “conversion” at all, so can hardly attest to its meaning. See . And next, the dissent cites a regulation that post-dated the CSPA by years, and thus is equally irrelevant to what Congress intended. See 71 Fed. Reg. 35732, 35749 (2006) (adding 8 CFR §204.2(i)(1)(iv)). More-over, both provisions relate to a circumstance in which a person can -petition for a visa because her U. S. citizen or LPR relative either died or engaged in domestic abuse. In that situation, the alien’s eligibility rests throughout on her connection to the deceased or abusive relative; no new party must ever come in, as one has to in a case like , to salvage a no-longer-effective petition. See , at 18 (addressing the problems that the substitution of a new petitioner raises). 11 Given the statute’s broad definition of “child,” the only F2A derivative beneficiaries who fall outside that proviso are stepchildren who were over the age of 18 when the petitioner married the spousal beneficiary. See §1101(b)(1)(B). The Government represents that thousands of children are designated as F2A derivatives every year. See Reply Brief 18, n. 13. 12 It is, therefore, impossible to understand the dissent’s statement that conversion of such a petition to an appropriate category requires “ ‘substantive alteration’ to [the] petition.” See , at 19, n. 8 (opinion of . 13 Still, the respondents’ view of the timing of conversion is better than the dissent’s. As an initial matter, the dissent’s objection to assessing conversion as of the date a visa becomes available hinges on an imaginary difficulty. That approach, the dissent complains, cannot be right because that date always “occurs before the point at which the child is determined to have aged-out.” at 15Well, yes. The date a visa becomes available is, under the statute, the date an alien ages out (or not); and that status change of course occurs before an immigration official, reviewing a visa application, finds that it has done so. But what of it? When an official determines that an alien was no longer a child on the date a visa became available, he also assesses whether automatic conversion was available to the alien that prior date. In other words, here as elsewhere in immigration law, conversion occurs (or not) upon the date of the relevant status change—and no other. See at 19. And once that is understood, the supposed difficulties the dissent throws up all melt away. At the time of the status change, F2A petitions can be converted without further contingencies, decisions, or delays, whereas no other petitions can. But cf. at 16, 17, n. 7 (countering, irrelevantly, that an F2A petition is automatically converted, additional steps remain in the immigration process). And immigration officials later reviewing visa applications know that fact, and can treat the different classes of aged-out beneficiaries accordingly. 14 Nor does the dissent offer any serious aid to the respondents. The dissent initially acknowledges that automatic conversion cannot involve “additional decisions, contingencies, or delays.” , at 13. But no worries, the dissent continues: “[O]nce [an alien’s parent] provides confirmation of her eligibility to sponsor” the aged-out alien, the original petition “can automatically be converted to an F2B petition, with no additional decision or contingency” or (presumably) delay. ,at 14. Think about that: Once every decision, contingency, and delay we have just described is over (and a parent has at long last turned out to be a viable sponsor), the dissent assures us that no further decisions, contingencies, and delays remain. Or, put differently, there are no contingencies after all the contingencies have been resolved; no decisions after all the decisions have been made; and no delay after all the delay has transpired. And as if that argument were not awkward enough, consider that it would make automatic conversion turn on the filing of a new document that shows the parent’s eligibility to sponsor her aged-out son or daughter—the very thing, as all parties agree, that conversion is supposed to render unnecessary. See at 18, 26. 15 The dissent appears to think that something helpful to its view follows from repeating the word “shall” and changing the subject of the commands. See at 9–10. But that is not so, as some further examples show. “If you advance to the next round, my assistant shall schedule an interview and you shall come in to answer questions.” “If the plane is low on fuel, the tanks shall be refilled and the pilot shall fly the route as scheduled.” In these sentences, as in our prior ones, the second command is conditional on the first; all that differs is that these sentences are (much like statutes) more formal and stilted. And the dissent’s citation of v. , , adds nothing to its argument. There, we construed the following provision: “[T]here shall be allowed to the holder of [a secured] claim, interest on such claim, and any reasonable fees, costs, or charges provided for under the agreement under which such claim arose.” , at 241. We held that the phrase “provided for under the agreement” qualifies the words “any reasonable fees, costs, or charges,” but not the words “interest on such claim.” at 241–242. What relevance that interpretation bears to this case eludes us. 16 The dissent claims that USCIS “administered priority date retention in exactly this manner” before the CSPA’s enactment, , at 10, but that confident assertion is just not so—or at least not in any way that assists the respondents. The dissent principally relies on 8 CFR §204.2(a)(4), which prior to the CSPA’s enactment permitted an aged-out F2A derivative beneficiary to retain his old priority date “if [a] subsequent petition is filed ” as filed the original. Far from authorizing an open-ended, free-floating entitlement, that now-superseded regulation allowed an alien to keep his priority date only if he (unlike the respondents’ offspring) had a qualifying relationship with the initial petitioner—that is, only if he fell within the group that the BIA in thought entitled to reliefSee 25 I. & N. Dec., at 34–35. And the other provisions the dissent cites (which, unlike §204.2(a)(4), continue to operate) similarly fail to support the dissent’s position, because they enable an alien to retain a priority date only if attached to an existing valid petition. See (permitting an alien to retain a priority date associated with an ex-isting F2B petition); 8 CFR §204.5(e) (permitting an alien to retain apriority date associated with an existing employment-based peti-tion); §204.12(f)(1) (permitting an alien to retain a priority dateassociated with an existing employment-based petition for immigrating physicians).

571.US.69

Sprint Communications, Inc. (Sprint), a national telecommunications service provider, withheld payment of intercarrier access fees imposed by Windstream Iowa Communications, Inc. (Windstream), a local telecommunications carrier, for long distance Voice over Internet Protocol (VoIP) calls, after concluding that the Telecommunications Act of 1996 preempted intrastate regulation of VoIP traffic. Windstream responded by threatening to block all Sprint customer calls, which led Sprint to ask the Iowa Utilities Board (IUB) to enjoin Windstream from discontinuing service to Sprint. Windstream retracted its threat, and Sprint moved to withdraw its complaint. Concerned that the dispute would recur, the IUB continued the proceedings in order to resolve the question whether VoIP calls are subject to intrastate regulation. Rejecting Sprint’s argument that this question was governed by federal law, the IUB ruled that intrastate fees applied to VoIP calls. Sprint sued respondents, IUB members (collectively IUB), in Federal District Court, seeking a declaration that the Telecommunications Act of 1996 preempted the IUB’s decision. As relief, Sprint sought an injunction against enforcement of the IUB’s order. Sprint also sought review of the IUB’s order in Iowa state court, reiterating the preemption argument made in Sprint’s federal-court complaint and asserting several other claims. Invoking Younger v. Harris, 401 U.S. 37, the Federal District Court abstained from adjudicating Sprint’s complaint in deference to the parallel state-court proceeding. The Eighth Circuit affirmed the District Court’s abstention decision, concluding that Younger abstention was required because the ongoing state-court review concerned Iowa’s important interest in regulating and enforcing state utility rates. Held: This case does not fall within any of the three classes of exceptional cases for which Younger abstention is appropriate. Pp. 6–12. (a) The District Court had jurisdiction to decide whether federal law preempted the IUB’s decision, see Verizon Md. Inc. v. Public Serv. Comm’n of Md., 535 U.S. 635, 642, and thus had a “virtually unflagging obligation” to hear and decide the case, Colorado River Water Conservation Dist. v. United States, 424 U.S. 800, 817. In Younger, this Court recognized an exception to that obligation for cases in which there is a parallel, pending state criminal proceeding. This Court has extended Younger abstention to particular state civil proceedings that are akin to criminal prosecutions, see Huffman v. Pursue, Ltd., 420 U.S. 592, or that implicate a State’s interest in enforcing the orders and judgments of its courts, see Pennzoil Co. v. Texaco Inc., 481 U.S. 1, but has reaffirmed that “only exceptional circumstances justify a federal court’s refusal to decide a case in deference to the States,” New Orleans Public Service, Inc. v. Council of City of New Orleans, 491 U.S. 350, 368 (NOPSI). NOPSI identified three such “exceptional circumstances.” First, Younger precludes federal intrusion into ongoing state criminal prosecutions. See 491 U. S., at 368. Second, certain “civil enforcement proceedings” warrant Younger abstention. Ibid. Finally, federal courts should refrain from interfering with pending “civil proceedings involving certain orders . . . uniquely in furtherance of the state courts’ ability to perform their judicial functions.” Ibid. This Court has not applied Younger outside these three “exceptional” categories, and rules, in accord with NOPSI, that they define Younger’s scope. Pp. 6–8. (b) The initial IUB proceeding does not fall within any of NOPSI’s three exceptional categories and therefore does not trigger Younger abstention. The first and third categories plainly do not accommodate the IUB’s proceeding, which was civil, not criminal in character, and which did not touch on a state court’s ability to perform its judicial function. Nor is the IUB’s order an act of civil enforcement of the kind to which Younger has been extended. The IUB proceeding is not “akin to a criminal prosecution.” Huffman, 420 U. S., at 604. Nor was it initiated by “the State in its sovereign capacity,” Trainor v. Hernandez, 431 U.S. 434, 444, to sanction Sprint for some wrongful act, see, e.g., Middlesex County Ethics Comm. v. Garden State Bar Assn., 457 U.S. 423, 433–434. Rather, the action was initiated by Sprint, a private corporation. No state authority conducted an investigation into Sprint’s activities or lodged a formal complaint against Sprint. Once Sprint withdrew the complaint that commenced administrative proceedings, the IUB argues, those proceedings became, essentially, a civil enforcement action. However, the IUB’s adjudicative authority was invoked to settle a civil dispute between two private parties, not to sanction Sprint for a wrongful act. In holding that abstention was the proper course, the Eighth Circuit misinterpreted this Court’s decision in Middlesex to mean that Younger abstention is warranted whenever there is (1) “an ongoing state judicial proceeding, which (2) implicates important state interests, and (3) . . . provide[s] an adequate opportunity to raise [federal] challenges.” In Middlesex, the Court invoked Younger to bar a federal court from entertaining a lawyer’s challenge to a state ethics committee’s pending investigation of the lawyer. Unlike the IUB’s proceeding, however, the state ethics committee’s hearing in Middlesex was plainly “akin to a criminal proceeding”: An investigation and formal complaint preceded the hearing, an agency of the State’s Supreme Court initiated the hearing, and the hearing’s purpose was to determine whether the lawyer should be disciplined for failing to meet the State’s professional conduct standards. 457 U. S., at 433–435. The three Middlesex conditions invoked by the Court of Appeals were therefore not dispositive; they were, instead, additional factors appropriately considered by the federal court before invoking Younger. Younger extends to the three “exceptional circumstances” identified in NOPSI, but no further. Pp. 8–11. 690 F.3d 864, reversed. Ginsburg, J., delivered the opinion for a unanimous Court.

This case involves two proceedings, one pending in state court, the other in federal court. Each seeks review of an Iowa Utilities Board (IUB or Board) order. And each presents the question whether Windstream Iowa Communications, Inc. (Windstream), a local telecommunications carrier, may impose on Sprint Communications, Inc. (Sprint), intrastate access charges for telephone calls transported via the Internet. Federal-court jurisdiction over controversies of this kind was confirmed in Verizon Md. Inc. v. Public Serv. Comm’n of Md., 535 U. S. 635 (2002) . Invoking Younger v. Harris, 401 U. S. 37 (1971) , the U. S. District Court for the Southern District of Iowa abstained from adjudicating Sprint’s complaint in deference to the parallel state-court proceeding, and the Court of Appeals for the Eighth Circuit affirmed the District Court’s abstention decision. We reverse the judgment of the Court of Appeals. In the main, federal courts are obliged to decide cases within the scope of federal jurisdiction. Abstention is not in order simply because a pending state-court proceeding involves the same subject matter. New Orleans Public Service, Inc. v. Council of City of New Orleans, 491 U. S. 350, 373 (1989) (NOPSI) (“[T]here is no doctrine that . . . pendency of state judicial proceedings excludes the federal courts.”). This Court has recognized, however, certain instances in which the prospect of undue interference with state proceedings counsels against federal relief. See id., at 368. Younger exemplifies one class of cases in which federal-court abstention is required: When there is a parallel, pending state criminal proceeding, federal courts must refrain from enjoining the state prosecution. This Court has extended Younger abstention to particular state civil proceedings that are akin to criminal prosecutions, see Huffman v. Pursue, Ltd., 420 U. S. 592 (1975) , or that implicate a State’s interest in enforcing the orders and judgments of its courts, see Pennzoil Co. v. Texaco Inc., 481 U. S. 1 (1987) . We have cautioned, however, that federal courts ordinarily should entertain and resolve on the merits an action within the scope of a jurisdictional grant, and should not “refus[e] to decide a case in deference to the States.” NOPSI, 491 U. S., at 368. Circumstances fitting within the Younger doctrine, we have stressed, are “exceptional”; they include, as catalogued in NOPSI, “state criminal prosecutions,” “civil enforcement proceedings,” and “civil proceedings involving certain orders that are uniquely in furtherance of the state courts’ ability to perform their judicial functions.” Id., at 367–368. Because this case presents none of the circumstances the Court has ranked as “exceptional,” the general rule governs: “[T]he pendency of an action in [a] state court is no bar to proceedings concerning the same matter in the Federal court having jurisdiction.” Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, 817 (1976) (quoting McClellan v. Carland, 217 U. S. 268, 282 (1910) ). I Sprint, a national telecommunications service provider, has long paid intercarrier access fees to the Iowa communications company Windstream (formerly Iowa Telecom) for certain long distance calls placed by Sprint customers to Windstream’s in-state customers. In 2009, however, Sprint decided to withhold payment for a subset of those calls, classified as Voice over Internet Protocol (VoIP), after concluding that the Telecommunications Act of 1996 preempted intrastate regulation of VoIP traffic. [ 1 ] In response, Windstream threatened to block all calls to and from Sprint customers. Sprint filed a complaint against Windstream with the IUB asking the Board to enjoin Windstream from discontinuing service to Sprint. In Sprint’s view, Iowa law entitled it to withhold payment while it contested the access charges and prohibited Windstream from carrying out its disconnection threat. In answer to Sprint’s complaint, Windstream retracted its threat to discontinue serving Sprint, and Sprint moved, successfully, to withdraw its complaint. Because the conflict between Sprint and Windstream over VoIP calls was “likely to recur,” however, the IUB decided to continue the proceedings to resolve the underlying legal question, i.e., whether VoIP calls are subject to intrastate regulation. Order in Sprint Communications Co. v. Iowa Telecommunications Servs., Inc., No. FCU–2010–0001 (IUB, Feb. 1, 2010), p. 6 (IUB Order). The question retained by the IUB, Sprint argued, was governed by federal law, and was not within the IUB’s adjudicative jurisdiction. The IUB disagreed, ruling that the intrastate fees applied to VoIP calls. [ 2 ] Seeking to overturn the Board’s ruling, Sprint commenced two lawsuits. First, Sprint sued the members of the IUB (respondents here) [ 3 ] in their official capacities in the United States District Court for the Southern District of Iowa. In its federal-court complaint, Sprint sought a declaration that the Telecommunications Act of 1996 preempted the IUB’s decision; as relief, Sprint requested an injunction against enforcement of the IUB’s order. Second, Sprint petitioned for review of the IUB’s order in Iowa state court. The state petition reiterated the preemption argument Sprint made in its federal-court complaint; in addition, Sprint asserted state law and procedural due process claims. Because Eighth Circuit precedent effectively required a plaintiff to exhaust state remedies before proceeding to federal court, see Alleghany Corp. v. McCartney, 896 F. 2d 1138 (1990), Sprint urges that it filed the state suit as a protective measure. Failing to do so, Sprint explains, risked losing the opportunity to obtain any review, federal or state, should the federal court decide to abstain after the expiration of the Iowa statute of limitations. See Brief for Petitioner 7–8. [ 4 ] As Sprint anticipated, the IUB filed a motion asking the Federal District Court to abstain in light of the state suit, citing Younger v. Harris, 401 U. S. 37 (1971) . The District Court granted the IUB’s motion and dismissed the suit. The IUB’s decision, and the pending state-court review of it, the District Court said, composed one “uninterruptible process” implicating important state interests. On that ground, the court ruled, Younger abstention was in order. Sprint Communications Co. v. Berntsen, No. 4:11–cv–00183–JAJ (SD Iowa, Aug. 1, 2011), App. to Pet. for Cert. 24a. For the most part, the Eighth Circuit agreed with the District Court’s judgment. The Court of Appeals rejected the argument, accepted by several of its sister courts, that Younger abstention is appropriate only when the parallel state proceedings are “coercive,” rather than “remedial,” in nature. 690 F. 3d 864, 868 (2012); cf. Guillemard-Ginorio v. Contreras-Gómez, 585 F. 3d 508, 522 (CA1 2009) (“[P]roceedings must be coercive, and in most cases, state-initiated, in order to warrant abstention.”). Instead, the Eighth Circuit read this Court’s precedent to require Younger abstention whenever “an ongoing state judicial proceeding . . . implicates important state interests, and . . . the state proceedings provide adequate opportunity to raise [federal] challenges.” 690 F. 3d, at 867 (citing Middlesex County Ethics Comm. v. Garden State Bar Assn., 457 U. S. 423, 432 (1982) ). Those criteria were satisfied here, the appeals court held, because the ongoing state-court review of the IUB’s decision concerned Iowa’s “important state interest in regulating and enforcing its intrastate utility rates.” 690 F. 3d, at 868. Recognizing the “possibility that the parties [might] return to federal court,” however, the Court of Appeals vacated the judgment dismissing Sprint’s complaint. In lieu of dismissal, the Eighth Circuit remanded the case, instructing the District Court to enter a stay during the pendency of the state-court action. Id., at 869. We granted certiorari to decide whether, consistent with our delineation of cases encompassed by the Younger doctrine, abstention was appropriate here. 569 U. S. ___ (2013). [ 5 ] II A Neither party has questioned the District Court’s jurisdiction to decide whether federal law preempted the IUB’s decision, and rightly so. In Verizon Md. Inc. v. Public Serv. Comm’n of Md., 535 U. S. 635 (2002) , we reviewed a similar federal-court challenge to a state administrative adjudication. In that case, as here, the party seeking federal-court review of a state agency’s decision urged that the Telecommunications Act of 1996 preempted the state action. We had “no doubt that federal courts ha[d federal question] jurisdiction under [28 U. S. C.] §1331 to entertain such a suit,” id., at 642, and nothing in the Telecommunications Act detracted from that conclusion, see id., at 643. Federal courts, it was early and famously said, have “no more right to decline the exercise of jurisdiction which is given, than to usurp that which is not given.” Cohens v. Virginia, 6 Wheat. 264, 404 (1821). Jurisdiction existing, this Court has cautioned, a federal court’s “obligation” to hear and decide a case is “virtually unflagging.” Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, 817 (1976) . Parallel state-court proceedings do not detract from that obligation. See ibid. In Younger, we recognized a “far-from-novel” exception to this general rule. New Orleans Public Service, Inc. v. Council of City of New Orleans, 491 U. S. 350, 364 (1989) (NOPSI). The plaintiff in Younger sought federal-court adjudication of the constitutionality of the California Criminal Syndicalism Act. Requesting an injunction against the Act’s enforcement, the federal-court plaintiff was at the time the defendant in a pending state criminal prosecution under the Act. In those circumstances, we said, the federal court should decline to enjoin the prosecution, absent bad faith, harassment, or a patently invalid state statute. See 401 U. S., at 53–54. Abstention was in order, we explained, under “the basic doctrine of equity jurisprudence that courts of equity should not act . . . to restrain a criminal prosecution, when the moving party has an adequate remedy at law and will not suffer irreparably injury if denied equitable relief.” Id., at 43–44. “[R]estraining equity jurisdiction within narrow limits,” the Court observed, would “prevent erosion of the role of the jury and avoid a duplication of legal proceedings and legal sanctions.” Id., at 44. We explained as well that this doctrine was “reinforced” by the notion of “ ‘comity,’ that is, a proper respect for state functions.” Ibid. We have since applied Younger to bar federal relief in certain civil actions. Huffman v. Pursue, Ltd., 420 U. S. 592 (1975) , is the pathmarking decision. There, Ohio officials brought a civil action in state court to abate the showing of obscene movies in Pursue’s theater. Because the State was a party and the proceeding was “in aid of and closely related to [the State’s] criminal statutes,” the Court held Younger abstention appropriate. Id., at 604. More recently, in NOPSI, 491 U. S., at 368, the Court had occasion to review and restate our Younger jurisprudence. NOPSI addressed and rejected an argument that a federal court should refuse to exercise jurisdiction to review a state council’s ratemaking decision. “[O]nly ex- ceptional circumstances,” we reaffirmed, “justify a fed- eral court’s refusal to decide a case in deference to the States.” Ibid. Those “exceptional circumstances” exist, the Court determined after surveying prior decisions, in three types of proceedings. First, Younger precluded federal intrusion into ongoing state criminal prosecutions. See ibid. Second, certain “civil enforcement proceedings” warranted abstention. Ibid. (citing, e.g., Huffman, 420 U. S., at 604). Finally, federal courts refrained from interfering with pending “civil proceedings involving certain orders . . . uniquely in furtherance of the state courts’ ability to perform their judicial functions.” 491 U. S., at 368 (citing Juidice v. Vail, 430 U. S. 327 , n. 12 (1977), and Pennzoil Co. v. Texaco Inc., 481 U. S. 1, 13 (1987) ). We have not applied Younger outside these three “exceptional” categories, and today hold, in accord with NOPSI, that they define Younger’s scope. B The IUB does not assert that the Iowa state court’s review of the Board decision, considered alone, implicates Younger. Rather, the initial administrative proceeding justifies staying any action in federal court, the IUB contends, until the state review process has concluded. The same argument was advanced in NOPSI. 491 U. S., at 368. We will assume without deciding, as the Court did in NOPSI, that an administrative adjudication and the subsequent state court’s review of it count as a “unitary process” for Younger purposes. Id., at 369. The question remains, however, whether the initial IUB proceeding is of the “sort . . . entitled to Younger treatment.” Ibid. The IUB proceeding, we conclude, does not fall within any of the three exceptional categories described in NOPSI and therefore does not trigger Younger abstention. The first and third categories plainly do not accommodate the IUB’s proceeding. That proceeding was civil, not criminal in character, and it did not touch on a state court’s ability to perform its judicial function. Cf. Juidice, 430 U. S., at 336, n. 12 (civil contempt order); Pennzoil, 481 U. S., at 13 (requirement for posting bond pending appeal). Nor does the IUB’s order rank as an act of civil enforcement of the kind to which Younger has been extended. Our decisions applying Younger to instances of civil enforcement have generally concerned state proceedings “akin to a criminal prosecution” in “important respects.” Huffman, 420 U. S., at 604. See also Middlesex, 457 U. S., at 432 (Younger abstention appropriate where “noncriminal proceedings bear a close relationship to proceedings criminal in nature”). Such enforcement actions are characteristically initiated to sanction the federal plaintiff, i.e., the party challenging the state action, for some wrongful act. See, e.g., Middlesex, 457 U. S., at 433–434 (state-initiated disciplinary proceedings against lawyer for violation of state ethics rules). In cases of this genre, a state actor is routinely a party to the state proceeding and often initiates the action. See, e.g., Ohio Civil Rights Comm’n v. Dayton Christian Schools, Inc., 477 U. S. 619 (1986) (state-initiated administrative proceedings to enforce state civil rights laws); Moore v. Sims, 442 U. S. 415 –420 (1979) (state-initiated proceeding to gain custody of children allegedly abused by their parents); Trainor v. Hernandez, 431 U. S. 434, 444 (1977) (civil proceeding “brought by the State in its sovereign capacity” to recover welfare payments defendants had allegedly obtained by fraud); Huffman, 420 U. S., at 598 (state-initiated proceeding to enforce obscenity laws). Investigations are commonly involved, often culminating in the filing of a formal complaint or charges. See, e.g., Dayton, 477 U. S., at 624 (noting preliminary investigation and complaint); Middlesex, 457 U. S., at 433 (same). The IUB proceeding does not resemble the state enforcement actions this Court has found appropriate for Younger abstention. It is not “akin to a criminal prosecution.” Huffman, 420 U. S., at 604. Nor was it initiated by “the State in its sovereign capacity.” Trainor, 431 U. S., at 444. A private corporation, Sprint, initiated the action. No state authority conducted an investigation into Sprint’s activities, and no state actor lodged a formal complaint against Sprint. In its brief, the IUB emphasizes Sprint’s decision to withdraw the complaint that commenced proceedings before the Board. At that point, the IUB argues, Sprint was no longer a willing participant, and the proceedings became, essentially, a civil enforcement action. See Brief for Respondents 31. [ 6 ] The IUB’s adjudicative authority, however, was invoked to settle a civil dispute between two private parties, not to sanction Sprint for commission of a wrongful act. Although Sprint withdrew its complaint, administrative efficiency, not misconduct by Sprint, prompted the IUB to answer the underlying federal question. By determining the intercarrier compensation regime applicable to VoIP calls, the IUB sought to avoid renewed litigation of the parties’ dispute. Because the underlying legal question remained unsettled, the Board observed, the controversy was “likely to recur.” IUB Order 6. Nothing here suggests that the IUB proceeding was “more akin to a criminal prosecution than are most civil cases.” Huffman, 420 U. S., at 604. In holding that abstention was the proper course, the Eighth Circuit relied heavily on this Court’s decision in Middlesex. Younger abstention was warranted, the Court of Appeals read Middlesex to say, whenever three conditions are met: There is (1) “an ongoing state judicial proceeding, which (2) implicates important state interests, and (3) . . . provide[s] an adequate opportunity to raise [federal] challenges.” 690 F. 3d, at 867 (citing Middlesex, 457 U. S., at 432). Before this Court, the IUB has endorsed the Eighth Circuit’s approach. Brief for Respondents 13. The Court of Appeals and the IUB attribute to this Court’s decision in Middlesex extraordinary breadth. We invoked Younger in Middlesex to bar a federal court from entertaining a lawyer’s challenge to a New Jersey state ethics committee’s pending investigation of the lawyer. Unlike the IUB proceeding here, the state ethics committee’s hearing in Middlesex was indeed “akin to a criminal proceeding.” As we noted, an investigation and formal complaint preceded the hearing, an agency of the State’s Supreme Court initiated the hearing, and the purpose of the hearing was to determine whether the lawyer should be disciplined for his failure to meet the State’s standards of professional conduct. 457 U. S., at 433–435. See also id., at 438 (Brennan, J., concurring in judgment) (noting the “quasi-criminal nature of bar disciplinary proceedings”). The three Middlesex conditions recited above were not dispositive; they were, instead, additional factors appropriately considered by the federal court before invoking Younger. Divorced from their quasi-criminal context, the three Middlesex conditions would extend Younger to virtually all parallel state and federal proceedings, at least where a party could identify a plausibly important state interest. See Tr. of Oral Arg. 35–36. That result is irreconcilable with our dominant instruction that, even in the presence of parallel state proceedings, abstention from the exercise of federal jurisdiction is the “exception, not the rule.” Hawaii Housing Authority v. Midkiff, 467 U. S. 229, 236 (1984) (quoting Colorado River, 424 U. S., at 813). In short, to guide other federal courts, we today clarify and affirm that Younger extends to the three “exceptional circumstances” identified in NOPSI, but no further. * * * For the reasons stated, the judgment of the United States Court of Appeals for the Eighth Circuit is Reversed. Notes 1 The Federal Communications Commission has yet to provide its view on whether the Telecommunications Act categorically preempts intrastate access charges for VoIP calls. See In re Connect America Fund, 26 FCC Rcd. 17663, 18002, ¶934 (2011) (reserving the ques-tion whether all VoIP calls “must be subject exclusively to federal regulation”). 2 At the conclusion of the IUB proceedings, Sprint paid Windstream all contested fees. 3 For convenience, we refer to respondents collectively as the IUB. 4 Since we granted certiorari, the Iowa state court issued an opinion rejecting Sprint’s preemption claim on the merits. Sprint Communications Co. v. Iowa Utils. Bd., No. CV–8638, App. to Joint Supp. Brief 20a–36a (Iowa Dist. Ct., Sept. 16, 2013). The Iowa court decision does not, in the parties’ view, moot this case, see Joint Supp. Brief 1, and we agree. Because Sprint intends to appeal the state-court decision, the “controversy . . . remains live.” Exxon Mobil Corp. v. Saudi Basic Industries Corp., . 5 The IUB agrees with Sprint that our decision in Burford v. Sun Oil Co., , cannot independently sustain the Eighth Circuit’s abstention analysis. See Brief for Respondents 9; cf. New Orleans Public Service, Inc. v. Council of City of New Orleans, . 6 To determine whether a state proceeding is an enforcement action under Younger, several Courts of Appeals, as noted, see supra, at 5, inquire whether the underlying state proceeding is “coercive” rather than “remedial.” See, e.g., Devlin v. Kalm, 594 F. 3d 893, 895 (CA6 2010). Though we referenced this dichotomy once in a footnote, see Ohio Civil Rights Comm’n v. Dayton Christian Schools, Inc., , n. 2 (1986), we do not find the inquiry necessary or inevitably helpful, given the susceptibility of the designations to manipulation.

573.US.149

Respondent Driehaus, a former Congressman, filed a complaint with the Ohio Elections Commission alleging that petitioner Susan B. Anthony List (SBA) violated an Ohio law that criminalizes certain false statements made during the course of a political campaign. Specifically, Driehaus alleged that SBA violated the law when it stated that his vote for the Patient Protection and Affordable Care Act (ACA) was a vote in favor of “taxpayer funded abortion.” After Driehaus lost his re-election bid, the complaint was dismissed, but SBA continued to pursue a separate suit in Federal District Court challenging the law on First Amendment grounds. Petitioner Coalition Opposed to Additional Spending and Taxes (COAST) also filed a First Amendment challenge to the Ohio law, alleging that it had planned to disseminate materials presenting a similar message but refrained due to the proceedings against SBA. The District Court consolidated the two lawsuits and dismissed them as nonjusticiable, concluding that neither suit presented a sufficiently concrete injury for purposes of standing or ripeness. The Sixth Circuit affirmed on ripeness grounds. Held: Petitioners have alleged a sufficiently imminent injury for Article III purposes. Pp. 7–18. (a) To establish Article III standing, a plaintiff must show, inter alia, an “injury in fact,” which must be “concrete and particularized” and “actual or imminent, not ‘conjectural’ or ‘hypothetical.’ ” Lujan v. Defenders of Wildlife, 504 U.S. 555, 560. When challenging a law prior to its enforcement, a plaintiff satisfies the injury-in-fact requirement where he alleges “an intention to engage in a course of conduct arguably affected with a constitutional interest, but proscribed by a statute, and there exists a credible threat of prosecution thereunder.” Babbitt v. Farm Workers, 442 U.S. 289, 298. Pp. 7–11. (b) Petitioners have alleged a credible threat of enforcement of the Ohio law. Pp. 11–17. (1) Petitioners have alleged “an intention to engage in a course of conduct arguably affected with a constitutional interest” by pleading specific statements they intend to make in future election cycles. Pp. 11–12. (2) Petitioners’ intended future conduct is also “arguably . . . proscribed by [the] statute.” The Ohio false statement statute sweeps broadly, and a panel of the Ohio Elections Commission already found probable cause to believe that SBA violated the law when it made statements similar to those petitioners plan to make in the future. Golden v. Zwickler, 394 U.S. 103, is distinguishable; the threat of prosecution under an electoral leafletting ban in that case was wholly conjectural because the plaintiff’s “sole concern” related to a former Congressman who was unlikely to run for office again. Here, by contrast, petitioners’ speech focuses on the broader issue of support for the ACA, not on the voting record of a single candidate. Nor does SBA’s insistence that its previous statements were true render its fears of enforcement misplaced. After all, that insistence did not prevent the Commission from finding probable cause for a violation the first time. Pp. 12–13. (3) Finally, the threat of future enforcement is substantial. There is a history of past enforcement against petitioners. Past enforcement against the same conduct is good evidence that the threat of enforcement is not “ ‘chimerical.’ ” Steffel v. Thompson, 415 U.S. 452, 459. The credibility of that threat is bolstered by the fact that a complaint may be filed with the State Commission by “any person,” Ohio Rev. Code Ann. §3517.153(A), not just a prosecutor or agency. The threatened Commission proceedings are of particular concern because of the burden they impose on electoral speech. Moreover, the target of a complaint may be forced to divert significant time and resources to hire legal counsel and respond to discovery requests in the crucial days before an election. But this Court need not decide whether the threat of Commission proceedings standing alone is sufficient; here, those proceedings are backed by the additional threat of criminal prosecution. Pp. 14–17. (c) The Sixth Circuit separately considered two other “prudential factors”: “fitness” and “hardship.” This Court need not resolve the continuing vitality of the prudential ripeness doctrine in this case because those factors are easily satisfied here. See Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U. S. ___. Pp. 17–18. 525 Fed. Appx. 415, reversed and remanded. Thomas, J., delivered the opinion for a unanimous Court.

Petitioners in this case seek to challenge an Ohio statute that prohibits certain “false statements” during the course of a political campaign. The question in this caseis whether their preenforcement challenge to that law is justiciable—and in particular, whether they have alleged a sufficiently imminent injury for the purposes of Article III. We conclude that they have. I The Ohio statute at issue prohibits certain “false statement[s]” “during the course of any campaign for nomination or election to public office or office of a political party.” Ohio Rev. Code Ann. §3517.21(B) (Lexis 2013). As relevant here, the statute makes it a crime for any person to “[m]ake a false statement concerning the voting record of a candidate or public official,” §3517.21(B)(9), or to “[p]ost, publish, circulate, distribute, or otherwise disseminate a false statement concerning a candidate, either knowing the same to be false or with reckless disregard of whether it was false or not,” §3517.21(B)(10).[1] “[A]ny person” acting on personal knowledge may file a complaint with the Ohio Elections Commission (or Commission) alleging a violation of the false statement statute. §3517.153(A) (Lexis Supp. 2014). If filed within 60 days of a primary election or 90 days of a general election, the complaint is referred to a panel of at least three Commission members. §§3517.156(A), (B)(1) (Lexis 2013). The panel must then hold an expedited hearing, generally within two business days, §3517.156(B)(1), to determine whether there is probable cause to believe the alleged violation occurred, §3517.156(C). Upon a finding of probable cause, the full Commission must, within 10 days, hold a hearing on the complaint. §3517.156(C)(2); see also Ohio Admin. Code §3517–1–10(E) (2008). The statute authorizes the full Commission to subpoena witnesses and compel production of documents. Ohio Rev. Code Ann. §3517.153(B) (Lexis Supp. 2014). At the full hearing, the parties may make opening and closing statements and present evidence. Ohio Admin. Code §§3517–1–11(B)(2)(c), (d), (g). If the Commission determines by “clear and convincing evidence” that a party has violated the false statement law, the Commission “shall” refer the matter to the relevant county prosecutor. Ohio Rev. Code Ann. §§3517.155(D)(1)–(2) (Lexis Supp. 2014). Alternatively, the Commission’s regulations state that it may simply issue a reprimand. See Ohio Admin. Code §3517–1–14(D). Violation of the false statement statute is a first-degree misdemeanor punishable by up to six months of imprisonment, a fine up to $5,000, or both. Ohio Rev. Code Ann. §§3599.40 (Lexis 2013), 3517.992(V) (Lexis Supp. 2014). A second conviction under the false statement statute is a fourth-degree felony that carries a mandatory penalty of disfranchisement. §3599.39. II Petitioner Susan B. Anthony List (SBA) is a “pro-life advocacy organization.” 525 Fed. Appx. 415, 416 (CA6 2013). During the 2010 election cycle, SBA publicly criticized various Members of Congress who voted for the Patient Protection and Affordable Care Act (ACA). In particular, it issued a press release announcing its plan to “educat[e] voters that their representative voted for a health care bill that includes taxpayer-funded abortion.” App. 49–50. The press release listed then-Congressman Steve Driehaus, a respondent here, who voted for the ACA. SBA also sought to display a billboard in Driehaus’ district condemning that vote. The planned billboard would have read: “Shame on Steve Driehaus! Driehaus voted FOR taxpayer-funded abortion.” Id., at 37. The advertising company that owned the billboard space refused to display that message, however, after Driehaus’ counsel threatened legal action. On October 4, 2010, Driehaus filed a complaint with the Ohio Elections Commission alleging, as relevant here, that SBA had violated §§3517.21(B)(9) and (10) by falsely stating that he had voted for “taxpayer-funded abortion.”[2] Because Driehaus filed his complaint 29 days before the general election, a Commission panel held an expedited hearing. On October 14, 2010, the panel voted 2 to 1 to find probable cause that a violation had been committed. The full Commission set a hearing date for 10 business days later, and the parties commenced discovery. Driehaus noticed depositions of three SBA employees as well as individuals affiliated with similar advocacy groups. He also issued discovery requests for all evidence that SBA would rely on at the Commission hearing, as well as SBA’s communications with allied organizations, political party committees, and Members of Congress and their staffs. On October 18, 2010—after the panel’s probable-cause determination, but before the scheduled Commission hearing—SBA filed suit in Federal District Court, seek-ing declaratory and injunctive relief on the ground that §§3517.21(B)(9) and (10) violate the First and Fourteenth Amendments of the United States Constitution. The District Court stayed the action under Younger v. Harris, 401 U. S. 37 (1971) , pending completion of the Commission proceedings. The Sixth Circuit denied SBA’s motion for an injunction pending appeal. Driehaus and SBA eventually agreed to postpone the full Commission hearing until after the election. When Driehaus lost the election in November 2010, he moved to withdraw his complaint against SBA. The Commission granted the motion with SBA’s consent. Once the Commission proceedings were terminated, the District Court lifted the stay and SBA amended its complaint. As relevant here, the amended complaint alleged that Ohio Rev. Code Ann. §§3517.21(B)(9) and (10) are unconstitutional both facially and as applied. Specifically, the complaint alleged that SBA’s speech about Driehaus had been chilled; that SBA “intends to engage in substantially similar activity in the future”; and that it “face[d] the prospect of its speech and associational rights again being chilled and burdened,” because “[a]ny complainant can hale [it] before the [Commission], forcing it to expend time and resources defending itself.” App. 121–122. The District Court consolidated SBA’s suit with a separate suit brought by petitioner Coalition Opposed to Ad-ditional Spending and Taxes (COAST), an advocacy orga-nization that also alleged that the same Ohio falsestatement provisions are unconstitutional both facially and as applied.[3] According to its amended complaint, COAST intended to disseminate a mass e-mail and other materials criticizing Driehaus’ vote for the ACA as a vote “to fund abortions with tax dollars,” but refrained from doing so because of the Commission proceedings against SBA. Id., at 146, 148, 162. COAST further alleged that it “desires to make the same or similar statements about other federal candidates who voted for” the ACA, but that fear “of finding itself subject to the same fate” as SBA has deterred it from doing so. Id., at 149, 157.[4] The District Court dismissed both suits as non-justiciable, concluding that neither suit presented a sufficiently concrete injury for purposes of standing or ripeness. The Sixth Circuit affirmed on ripeness grounds. 525 Fed. Appx. 415. The Court of Appeals analyzed three factors to assess whether the case was ripe for review: (1) the likelihood that the alleged harm would come to pass; (2) whether the factual record was sufficiently developed; and (3) the hardship to the parties if judicial relief were denied. Regarding the first factor, the Sixth Circuit concluded that SBA’s prior injuries—the probable-cause determination and the billboard rejection—“do not help it show an imminent threat of future prosecution,” particularly where “the Commission never found that SBA . . . violated Ohio’s false-statement law.” Id., at 420. The court further reasoned that it was speculative whether any person would file a complaint with the Commission in the future, in part because Driehaus took a 2-year assignment with the Peace Corps in Africa after losing the election. Finally, the court noted that SBA has not alleged that “it plans to lie or recklessly disregard the veracity of its speech” in the future, but rather maintains that the statements it intends to make are factually true. Id., at 422. As for the remaining factors, the court concluded that the factual record was insufficiently developed with respect to the content of SBA’s future speech, and that withholding judicial relief would not result in undue hardship because, in the time period leading up to the 2010 election, SBA continued to communicate its message even after Commission proceedings were initiated. The Sixth Circuit therefore determined that SBA’s suit was not ripe for review, and that its analysis as to SBA compelled the same conclusion with respect to COAST. We granted certiorari, 571 U. S. ___ (2014), and now reverse. III A Article III of the Constitution limits the jurisdiction of federal courts to “Cases” and “Controversies.” U. S. Const., Art. III, §2. The doctrine of standing gives meaning to these constitutional limits by “identify[ing] those disputes which are appropriately resolved through the judicial process.”[5] Lujan v. Defenders of Wildlife, 504 U. S. 555, 560 (1992) . “The law of Article III standing, which is built on separation-of-powers principles, serves to prevent the judicial process from being used to usurp the powers of the political branches.” Clapper v. Amnesty Int’l USA, 568 U. S. ___, ___, (2013) (slip op., at 9). To establish Article III standing, a plaintiff must show (1) an “injury in fact,” (2) a sufficient “causal connection between the injuryand the conduct complained of,” and (3) a “likel[ihood]” that the injury “will be redressed by a favorable decision.” Lujan, supra, at 560–561 (internal quotation marksomitted). This case concerns the injury-in-fact requirement, which helps to ensure that the plaintiff has a “personal stake in the outcome of the controversy.” Warth v. Seldin, 422 U. S. 490, 498 (1975) (internal quotation marks omitted). An injury sufficient to satisfy Article III must be “concrete and particularized” and “actual or imminent, not ‘conjectural’ or ‘hypothetical.’ ” Lujan, supra, at 560 (some internal question marks omitted). An allegation of future injury may suffice if the threatened injury is “certainly impending,” or there is a “ ‘substantial risk’ that the harm will occur.” Clapper, 568 U. S., at ___, ___, n. 5 (slip op., at 10, 15, n. 5) (emphasis deleted and internal quotation marks omitted). “ ‘ The party invoking federal jurisdiction bears the burden of establishing’ standing.” Id., at ___ (slip op., at 12). “[E]ach element must be supported in the same way as any other matter on which the plaintiff bears the burden of proof, i.e., with the manner and degree of evidence required at the successive stages of the litigation.” Lujan, supra, at 561. B One recurring issue in our cases is determining when the threatened enforcement of a law creates an Article III injury. When an individual is subject to such a threat, an actual arrest, prosecution, or other enforcement action is not a prerequisite to challenging the law. See Steffel v. Thompson, 415 U. S. 452, 459 (1974) (“[I]t is not necessary that petitioner first expose himself to actual arrest or prosecution to be entitled to challenge a statute that he claims deters the exercise of his constitutional rights”); see also MedImmune, Inc. v. Genentech, Inc., 549 U. S. 118 –129 (2007) (“[W]here threatened action by government is concerned, we do not require a plaintiff to expose himself to liability before bringing suit to challenge the basis for the threat”). Instead, we have permitted pre-enforcement review under circumstances that render the threatened enforcement sufficiently imminent. Specifically, we have held that a plaintiff satisfies the injury-in-fact requirement where he alleges “an intention to engage in a course of conduct arguably affected with a constitutional interest, but proscribed by a statute, and there exists a credible threat of prosecution thereunder.” Babbitt v. Farm Workers, 442 U. S. 289, 298 (1979) . Several of our cases illustrate the circumstances under which plaintiffs may bring a preenforcement challenge consistent with Article III. In Steffel, for example, police officers threatened to arrest petitioner and his companion for distributing handbills protesting the Vietnam War. Petitioner left to avoid arrest; his companion remained and was arrested and charged with criminal trespass. Petitioner sought a declaratory judgment that the trespass statute was unconstitutional as applied to him. We determined that petitioner had alleged a credible threat of enforcement: He had been warned to stop handbilling and threatened with prosecution if he disobeyed; he stated his desire to continue handbilling (an activity he claimed was constitutionally protected); and his companion’s prosecution showed that his “concern with arrest” was not “ ‘ chimerical.’ ” 415 U. S., at 459. Under those circumstances, we said, “it is not necessary that petitioner first expose himself to actual arrest or prosecution to be entitled to challenge a statute that he claims deters the exercise of his constitutional rights.” Ibid. In Babbitt, we considered a preenforcement challenge to a statute that made it an unfair labor practice to encourage consumers to boycott an “agricultural product . . . by the use of dishonest, untruthful and deceptive publicity.’ ” 442 U. S., at 301. The plaintiffs contended that the law “unconstitutionally penalize[d] inaccuracies inadvertently uttered in the course of consumer appeals.” Ibid. Building on Steffel, we explained that a plaintiff could bring a preenforcement suit when he “has alleged an intention to engage in a course of conduct arguably af-fected with a constitutional interest, but proscribed by a statute, and there exists a credible threat of prosecution thereunder.” Babbit, supra, at 298. We found those circumstances present in Babbitt. In that case, the law “on its face proscribe[d] dishonest, untruthful, and deceptive publicity.” 442 U. S., at 302. The plaintiffs had “actively engaged in consumer publicity campaigns in the past” and alleged “an intention to continue” those campaigns in the future. Id., at 301. And although they did not “plan to propagate untruths,” they argued that “ ‘ erroneous statement is inevitable in free debate.’ ” Ibid. We concluded that the plaintiffs’ fear of prosecution was not “imaginary or wholly speculative,” and that their challenge to the consumer publicity provision presented an Article III case or controversy. Id., at 302. Two other cases bear mention. In Virginia v. American Booksellers Assn. Inc., 484 U. S. 383 (1988) , we held that booksellers could seek preenforcement review of a law making it a crime to “ ‘knowingly display for commercial purpose’ ” material that is “ ‘harmful to juveniles’ ” as defined by the statute. Id., at 386. At trial, the booksellers introduced 16 books they believed were covered by the statute and testified that costly compliance measures would be necessary to avoid prosecution for displaying such books. Just as in Babbitt and Steffel, we determined that the “pre-enforcement nature” of the suit was not “troubl[ing]” because the plaintiffs had “alleged an actual and well-founded fear that the law will be enforced against them.” 484 U. S., at 393. Finally, in Holder v. Humanitarian Law Project, 561 U. S. 1 (2010) , we considered a preenforcement challenge to a law that criminalized “ ‘ knowingly provid[ing] mate-rial support or resources to a foreign terrorist organization.’ ” Id., at 8. The plaintiffs claimed that they had provided support to groups designated as terrorist organizations prior to the law’s enactment and would provide similar support in the future. The Government had charged 150 persons with violating the law and declined to disavow prosecution if the plaintiffs resumed their support of the designated organizations. We held that the claims were justiciable: The plaintiffs faced a “ ‘credible threat’ ” of enforcement and “ ‘should not be required to await and undergo a criminal prosecution as the sole means of seeking relief.’ ” Id., at 15. IV Here, SBA and COAST contend that the threat of enforcement of the false statement statute amounts to an Article III injury in fact. We agree: Petitioners have alleged a credible threat of enforcement. See Babbitt, 442 U. S., at 298. A First, petitioners have alleged “an intention to engage in a course of conduct arguably affected with a constitutional interest.” Ibid. Both petitioners have pleaded specific statements they intend to make in future election cycles. SBA has already stated that representatives who voted for the ACA supported “taxpayer-funded abortion,” and it has alleged an “inten[t] to engage in substantially similar activity in the future.” App. 49–50, 122. See also Humanitarian Law Project, supra, at 15–16 (observing that plaintiffs had previously provided support to groups designated as terrorist organizations and alleged they “would provide similar support [to the same terrorist organizations] again if the statute’s allegedly unconstitutional bar were lifted”). COAST has alleged that it previously intended to disseminate materials criticizing a vote for the ACA as a vote “to fund abortions with tax dollars,” and that it “desires to make the same or similar statements about other federal candidates who voted for [the ACA].” App. 146, 149, 162. Because petitioners’ intended future conduct concerns political speech, it is certainly “affected with a constitutional interest.” Babbitt, supra, at 298; see also Monitor Patriot Co. v. Roy, 401 U. S. 265, 272 (1971) (“[T]he constitutional guarantee has its fullest and most urgent application precisely to the conduct of campaigns for political office”). B Next, petitioners’ intended future conduct is “arguably. . . proscribed by [the] statute” they wish to challenge. Babbitt, supra, at 298. The Ohio false statement law sweeps broadly, see supra, at 1–2, and n. 1., and covers the subject matter of petitioners’ intended speech. Both SBA and COAST have alleged an intent to “[m]ake” statements “concerning the voting record of a candidate or public official,” §3517.21(B)(9), and to “disseminate” statements “concerning a candidate . . . to promote the election, nomination, or defeat of the candidate,” §3517.21(B)(10). And, a Commission panel here already found probable cause to believe that SBA violated the statute when it stated that Driehaus had supported“taxpayer-funded abortion”—the same sort of statement petitioners plan to disseminate in the future. Under these circumstances, we have no difficulty concluding that petitioners’ intended speech is “arguably proscribed” by the law. Respondents incorrectly rely on Golden v. Zwickler, 394 U. S. 103 (1969) . In that case, the plaintiff had previously distributed anonymous leaflets criticizing a particular Congressman who had since left office. Id., at 104–106, and n. 2. The Court dismissed the plaintiff’s challenge to the electoral leafletting ban as nonjusticiable because his “sole concern was literature relating to the Congressman and his record,” and “it was most unlikely that the Congressman would again be a candidate.” Id., at 109 (emphasis added). Under those circumstances, any threat of future prosecution was “wholly conjectural.” Ibid. Here, by contrast, petitioners’ speech focuses on the broader issue of support for the ACA, not on the voting record of a single candidate. See Reply Brief 4–5 (identifying other elected officials who plan to seek reelection as potential objects of SBA’s criticisms). Because petitioners’ alleged future speech is not directed exclusively at Driehaus, it does not matter whether he “may run for office again.” Brief for Respondents 33 (internal quotation marks omitted). As long as petitioners continue to engage in comparable electoral speech regarding support for the ACA, that speech will remain arguably proscribed by Ohio’s false statement statute. Respondents, echoing the Sixth Circuit, contend that SBA’s fears of enforcement are misplaced because SBA has not said it “ ‘plans to lie or recklessly disregard the veracity of its speech.’ ” Id., at 15 (quoting 525 Fed. Appx., at 422). The Sixth Circuit reasoned that because SBA “can only be liable for making a statement ‘knowing’ it is false,” SBA’s insistence that its speech is factually true “makes the possibility of prosecution for uttering such statements exceedingly slim.” Id., at 422. The Sixth Circuit misses the point. SBA’s insistence that the allegations in its press release were true did not prevent the Commission panel from finding probable cause to believe that SBA had violated the law the first time around. And, there is every reason to think that similar speech in the future will result in similar proceedings, notwithstanding SBA’s belief in the truth of its allegations. Nothing in this Court’s decisions requires a plaintiff who wishes to challenge the constitutionality of a law to confess that he will in fact violate that law. See, e.g., Babbitt, 442 U. S., at 301 (case was justiciable even though plaintiffs disavowed any intent to “propagate untruths”). C Finally, the threat of future enforcement of the false statement statute is substantial. Most obviously, there is a history of past enforcement here: SBA was the subject of a complaint in a recent election cycle. We have observed that past enforcement against the same conduct is good evidence that the threat of enforcement is not “ ‘ chimerical.’ ” Steffel, 415 U. S., at 459; cf. Clapper, 568 U. S., at ___ (slip op., at 12) (plaintiffs’ theory of standing was “substantially undermine[d]” by their “fail[ure] to offer any evidence that their communications ha[d] been monitored” under the challenged statute). Here, the threat is even more substantial given that the Commission panel actually found probable cause to believe that SBA’s speech violated the false statement statute. Indeed future complainants may well “invoke the prior probable-cause finding to prove that SBA knowingly lied.” Brief for Petitioners 32. The credibility of that threat is bolstered by the fact that authority to file a complaint with the Commission is not limited to a prosecutor or an agency. Instead, the false statement statute allows “any person” with knowledge of the purported violation to file a complaint. §3517.153(A). Because the universe of potential complainants is not restricted to state officials who are constrained by explicit guidelines or ethical obligations, there is a real risk of complaints from, for example, political opponents. See Brief for Michael DeWine, Attorney General of Ohio, as Amicus Curiae 8 (hereinafter DeWine Brief); see also id., at 6 (noting that “the Commission has no system for weeding out frivolous complaints”). And petitioners, who intend to criticize candidates for political office, are easy targets. Finally, Commission proceedings are not a rare occurrence. Petitioners inform us that the Commission “ ‘handles about 20 to 80 false statement complaints per year,’ ” Brief for Petitioners 46, and respondents do not deny that the Commission frequently fields complaints alleging violations of the false statement statute. Cf. Humani-tarian Law Project, 561 U. S., at 16 (noting that there had been numerous prior prosecutions under the challenged statute). Moreover, respondents have not disavowed enforcement if petitioners make similar statements in the future. See Tr. of Oral Arg. 29–30; see also Humanitarian Law Project, supra, at 16 (“The Government has not argued to this Court that plaintiffs will not be prosecuted if they do what they say they wish to do”). In fact, the specter of enforcement is so substantial that the owner of the billboard refused to display SBA’s message after receiving a letter threatening Commission proceedings. On these facts, the prospect of future enforcement is far from “imaginary or speculative.” Babbitt, supra, at 298. We take the threatened Commission proceedings into account because administrative action, like arrest or prosecution, may give rise to harm sufficient to justify pre-enforcement review. See Ohio Civil Rights Comm’n v. Dayton Christian Schools, Inc., 477 U. S. 619 –626, n. 1 (1986) (“If a reasonable threat of prosecution creates a ripe controversy, we fail to see how the actual filing of the administrative action threatening sanctions in this case does not”). The burdens that Commission proceedings can impose on electoral speech are of particular concern here. As the Ohio Attorney General himself notes, the “practical effect” of the Ohio false statement scheme is “to permit a private complainant . . . to gain a campaign advantage without ever having to prove the falsity of a statement.” DeWine Brief 7. “[C]omplainants may time their submissions to achieve maximum disruption of their political opponents while calculating that an ultimate decision on the merits will be deferred until after the relevant election.” Id., at 14–15. Moreover, the target of a false statement complaint may be forced to divert significant time and resources to hire legal counsel and respond to discovery requests in the crucial days leading up to an election. And where, as here, a Commission panel issues a preelection probable-cause finding, “such a determination itself may be viewed [by the electorate] as a sanction by the State.” Id., at 13. Although the threat of Commission proceedings is a substantial one, we need not decide whether that threat standing alone gives rise to an Article III injury. The burdensome Commission proceedings here are backed by the additional threat of criminal prosecution. We conclude that the combination of those two threats suffices to create an Article III injury under the circumstances of this case. See Babbitt, supra, at 302, n. 13 (In addition to the threat of criminal sanctions, “the prospect of issuance of an administrative cease-and-desist order or a court-ordered injunction against such prohibited conduct provides substantial additional support for the conclusion that appellees’ challenge . . . is justiciable” (citations omitted)). That conclusion holds true as to both SBA and COAST. Respondents, relying on Younger v. Harris, 401 U. S. 37 (1971) , appear to suggest that COAST lacks standing because it refrained from actually disseminating its planned speech in order to avoid Commission proceedings of its own. See Brief for Respondents 26–27, 34. In Younger, the plaintiff had been indicted for distributing leaflets in violation of the California Criminal Syndicalism Act. When he challenged the constitutionality of the law in federal court, several other plaintiffs intervened, arguing that their own speech was inhibited by Harris’ prosecution. The Court concluded that only the plaintiff had standing because the intervenors “d[id] not claim that they ha[d] ever been threatened with prosecution, that a prosecution [wa]s likely, or even that a prosecution [wa]s remotely possible.” 401 U. S., at 42. That is not this case. Unlike the intervenors in Younger, COAST has alleged an intent to engage in the same speech that was the subject of a prior enforcement proceeding. Also unlike the intervenors in Younger, who had never been threatened with prosecution, COAST has been the subject of Commission proceedings in the past. See, e.g., COAST Candidates PAC v. Ohio Elections Comm’n, 543 Fed. Appx. 490 (CA6 2013). COAST is far more akin to the plaintiff in Steffel, who was not arrested alongside his handbilling companion but was nevertheless threatened with prosecution for similar speech. 415 U. S., at 459. In sum, we find that both SBA and COAST have alleged a credible threat of enforcement. V In concluding that petitioners’ claims were not justiciable, the Sixth Circuit separately considered two other factors: whether the factual record was sufficiently developed, and whether hardship to the parties would result if judicial relief is denied at this stage in the proceedings. 525 Fed. Appx., at 419. Respondents contend that these “prudential ripeness” factors confirm that the claims at issue are nonjusticiable. Brief for Respondents 17. But we have already concluded that petitioners have alleged a sufficient Article III injury. To the extent respondents would have us deem petitioners’ claims nonjusticiable “on grounds that are ‘prudential,’ rather than constitutional,” “[t]hat request is in some tension with our recent reaffirmation of the principle that ‘a federal court’s obligation to hear and decide’ cases within its jurisdiction ‘is virtually unflagging.’ ” Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U. S. ___, ___ (2014) (slip op., at 6) (quoting Sprint Communications, Inc. v. Jacobs, 571 U. S. ___, ___ (2013) (slip op., at 6); some internal quotation marks omitted). In any event, we need not resolve the continuing vitality of the prudential ripeness doctrine in this case because the “fitness” and “hardship” factors are easily satisfied here. First, petitioners’ challenge to the Ohio false statement statute presents an issue that is “purely legal, and will not be clarified by further factual development.” Thomas v. Union Carbide Agricultural Products Co., 473 U. S. 568, 581 (1985) . And denying prompt judicial review would impose a substantial hardship on petitioners, forcing them to choose between refraining from core political speech on the one hand, or engaging in that speech and risking costly Commission proceedings and criminal prosecution on the other. * * * Petitioners in this case have demonstrated an injury in fact sufficient for Article III standing. We accordingly reverse the judgment of the United States Court of Appeals for the Sixth Circuit and remand the case for further proceedings consistent with this opinion, including a determination whether the remaining Article III standing requirements are met. It is so ordered.Notes 1 Section 3517.21(B) provides in relevant part: 2 The dispute about the falsity of SBA’s speech concerns two different provisions of the ACA: (1) the subsidy to assist lower income individ-uals in paying insurance premiums, and (2) the direct appropriation of federal money for certain health programs such as community health centers. See Brief for Petitioners 4–5. 3 Petitioners also challenged a related “disclaimer provision,” App. 126–127, 156–157, under Ohio Rev. Code Ann. §3517.20, and COAST raised pre-emption and due process claims. Reply Brief 21, n. 7. Petitioners do not pursue their “disclaimer,” pre-emption, or due process claims before us. We also need not address SBA’s separate challenge to the Commission’s investigatory procedures; petitioners have conceded that the procedures claim stands or falls with the substantive prohibition on false statements. see Tr. of Oral Arg. 19. Finally, the parties agree that petitioners’ as-applied claims “are better read as facial objections to Ohio’s law.” Reply Brief 19. Accordingly, we do not separately address the as-applied claims. 4 SBA named Driehaus, the Commission’s members and its staff attorney (in their official capacities), and the Ohio Secretary of State (in her official capacity) as defendants. COAST named the Commission, the Commission’s members and its staff attorney (in their official capacities), and the Ohio Secretary of State (in her official capacity) as defendants. All named defendants are respondents here. 5 The doctrines of standing and ripeness “originate” from the same Article III limitation. v. , . As the parties acknowledge, the Article III standing and ripeness issues in this case “boil down to the same question.” v. , , n. 8 (2007); see Brief for Petitioners 28; Brief for Respondents 22. Consistent with our practice in cases like v. , , and v. , n. 11 (1979), we use the term “standing” in this opinion.

572.US.565

Since 1999, the monthly town board meetings in Greece, New York, have opened with a roll call, a recitation of the Pledge of Allegiance, and a prayer given by clergy selected from the congregations listed in a local directory. While the prayer program is open to all creeds, nearly all of the local congregations are Christian; thus, nearly all of the participating prayer givers have been too. Respondents, citizens who attend meetings to speak on local issues, filed suit, alleging that the town violated the First Amendment’s Establishment Clause by preferring Christians over other prayer givers and by sponsoring sectarian prayers. They sought to limit the town to “inclusive and ecumenical” prayers that referred only to a “generic God.” The District Court upheld the prayer practice on summary judgment, finding no impermissible preference for Christianity; concluding that the Christian identity of most of the prayer givers reflected the predominantly Christian character of the town’s congregations, not an official policy or practice of discriminating against minority faiths; finding that the First Amendment did not require Greece to invite clergy from congregations beyond its borders to achieve religious diversity; and rejecting the theory that legislative prayer must be nonsectarian. The Second Circuit reversed, holding that some aspects of the prayer program, viewed in their totality by a reasonable observer, conveyed the message that Greece was endorsing Christianity. Held: The judgment is reversed. 681 F.3d 20, reversed. Justice Kennedy delivered the opinion of the Court, except as to Part II–B, concluding that the town’s prayer practice does not violate the Establishment Clause. Pp. 6–18. (a) Legislative prayer, while religious in nature, has long been understood as compatible with the Establishment Clause. Marsh v. Chambers, 463 U.S. 783, 792. In Marsh, the Court concluded that it was not necessary to define the Establishment Clause’s precise boundary in order to uphold Nebraska’s practice of employing a legislative chaplain because history supported the conclusion that the specific practice was permitted. The First Congress voted to appoint and pay official chaplains shortly after approving language for the First Amendment, and both Houses have maintained the office virtually uninterrupted since then. See id., at 787–789, and n. 10. A majority of the States have also had a consistent practice of legislative prayer. Id., at 788–790, and n. 11. There is historical precedent for the practice of opening local legislative meetings with prayer as well. Marsh teaches that the Establishment Clause must be interpreted “by reference to historical practices and understandings.” County of Allegheny v. American Civil Liberties Union, Greater Pittsburgh Chapter, 492 U.S. 573, 670 (opinion of Kennedy, J.). Thus, any test must acknowledge a practice that was accepted by the Framers and has withstood the critical scrutiny of time and political change. The Court’s inquiry, then, must be to determine whether the prayer practice in the town of Greece fits within the tradition long followed in Congress and the state legislatures. Pp. 6–9. (b) Respondents’ insistence on nonsectarian prayer is not consistent with this tradition. The prayers in Marsh were consistent with the First Amendment not because they espoused only a generic theism but because the Nation’s history and tradition have shown that prayer in this limited context could “coexis[t] with the principles of disestablishment and religious freedom.” 463 U. S., at 786. Dictum in County of Allegheny suggesting that Marsh permitted only prayer with no overtly Christian references is irreconcilable with the facts, holding, and reasoning of Marsh, which instructed that the “content of the prayer is not of concern to judges,” provided “there is no indication that the prayer opportunity has been exploited to proselytize or advance any one, or to disparage any other, faith or belief.” 463 U. S., at 794–795. To hold that invocations must be nonsectarian would force the legislatures sponsoring prayers and the courts deciding these cases to act as supervisors and censors of religious speech, thus involving government in religious matters to a far greater degree than is the case under the town’s current practice of neither editing nor approving prayers in advance nor criticizing their content after the fact. Respondents’ contrary arguments are unpersuasive. It is doubtful that consensus could be reached as to what qualifies as a generic or nonsectarian prayer. It would also be unwise to conclude that only those religious words acceptable to the majority are permis-sible, for the First Amendment is not a majority rule and government may not seek to define permissible categories of religious speech. In rejecting the suggestion that legislative prayer must be nonsectarian, the Court does not imply that no constraints remain on its content. The relevant constraint derives from the prayer’s place at the opening of legislative sessions, where it is meant to lend gravity to the occasion and reflect values long part of the Nation’s heritage. From the Nation’s earliest days, invocations have been addressed to assemblies comprising many different creeds, striving for the idea that people of many faiths may be united in a community of tolerance and devotion, even if they disagree as to religious doctrine. The prayers delivered in Greece do not fall outside this tradition. They may have invoked, e.g., the name of Jesus, but they also invoked universal themes, e.g., by calling for a “spirit of cooperation.” Absent a pattern of prayers that over time denigrate, proselytize, or betray an impermissible government purpose, a challenge based solely on the content of a particular prayer will not likely establish a constitutional violation. See 463 U. S., at 794–795. Finally, so long as the town maintains a policy of nondiscrimination, the Constitution does not require it to search beyond its borders for non-Christian prayer givers in an effort to achieve religious balancing. Pp. 9–18. Justice Kennedy, joined by The Chief Justice and Justice Alito, concluded in Part II–B that a fact-sensitive inquiry that considers both the setting in which the prayer arises and the audience to whom it is directed shows that the town is not coercing its citizens to engage in a religious observance. The prayer opportunity is evaluated against the backdrop of a historical practice showing that prayer has become part of the Nation’s heritage and tradition. It is presumed that the reasonable observer is acquainted with this tradition and understands that its purposes are to lend gravity to public proceedings and to acknowledge the place religion holds in the lives of many private citizens. Furthermore, the principal audience for these invocations is not the public, but the lawmakers themselves. And those lawmakers did not direct the public to participate, single out dissidents for opprobrium, or indicate that their decisions might be influenced by a person’s acquiescence in the prayer opportunity. Respondents claim that the prayers gave them offense and made them feel excluded and disrespected, but offense does not equate to coercion. In contrast to Lee v. Weisman, 505 U.S. 577, where the Court found coercive a religious invocation at a high school graduation, id., at 592–594, the record here does not suggest that citizens are dissuaded from leaving the meeting room during the prayer, arriving late, or making a later protest. That the prayer in Greece is delivered during the opening ceremonial portion of the town’s meeting, not the policymaking portion, also suggests that its purpose and effect are to acknowledge religious leaders and their institutions, not to exclude or coerce nonbelievers. Pp. 18–23. Justice Thomas, joined by Justice Scalia as to Part II, agreed that the town’s prayer practice does not violate the Establishment Clause, but concluded that, even if the Establishment Clause were properly incorporated against the States through the Fourteenth Amendment, the Clause is not violated by the kind of subtle pressures respondents allegedly suffered, which do not amount to actual legal coercion. The municipal prayers in this case bear no resemblance to the coercive state establishments that existed at the founding, which exercised government power in order to exact financial support of the church, compel religious observance, or control religious doctrine. Pp. 1–8. Kennedy, J., delivered the opinion of the Court, except as to Part II–B. Roberts, C. J., and Alito, J., joined the opinion in full, and Scalia and Thomas, JJ., joined except as to Part II–B. Alito, J., filed a concurring opinion, in which Scalia, J., joined. Thomas, J., filed an opinion concurring in part and concurring in the judgment, in which Scalia, J., joined as to Part II. Breyer, J., filed a dissenting opinion. Kagan, J., filed a dissenting opinion, in which Ginsburg, Breyer, and Sotomayor, JJ., joined.

except as to Part II–B.[1]* The Court must decide whether the town of Greece, New York, imposes an impermissible establishment of religion by opening its monthly board meetings with a prayer. It must be concluded, consistent with the Court’s opinion in Marsh v. Chambers, 463 U. S. 783 (1983) , that no violation of the Constitution has been shown. I Greece, a town with a population of 94,000, is in upstate New York. For some years, it began its monthly town board meetings with a moment of silence. In 1999, the newly elected town supervisor, John Auberger, decided to replicate the prayer practice he had found meaningful while serving in the county legislature. Following the roll call and recitation of the Pledge of Allegiance, Auberger would invite a local clergyman to the front of the room to deliver an invocation. After the prayer, Auberger would thank the minister for serving as the board’s “chaplain for the month” and present him with a commemorative plaque. The prayer was intended to place town board members in a solemn and deliberative frame of mind, invoke divine guidance in town affairs, and follow a tradition practiced by Congress and dozens of state legislatures. App. 22a–25a. The town followed an informal method for selecting prayer givers, all of whom were unpaid volunteers. A town employee would call the congregations listed in a local directory until she found a minister available for that month’s meeting. The town eventually compiled a list of willing “board chaplains” who had accepted invitations and agreed to return in the future. The town at no point excluded or denied an opportunity to a would-be prayer giver. Its leaders maintained that a minister or layperson of any persuasion, including an atheist, could give the invocation. But nearly all of the congregations in town were Christian; and from 1999 to 2007, all of the participating ministers were too. Greece neither reviewed the prayers in advance of the meetings nor provided guidance as to their tone or content, in the belief that exercising any degree of control over the prayers would infringe both the free exercise and speech rights of the ministers. Id., at 22a. The town instead left the guest clergy free to compose their own devotions. The resulting prayers often sounded both civic and religious themes. Typical were invocations that asked the divinity to abide at the meeting and bestow blessings on the community: “Lord we ask you to send your spirit of servanthood upon all of us gathered here this evening to do your work for the benefit of all in our community. We ask you to bless our elected and appointed officials so they may deliberate with wisdom and act with courage. Bless the members of our community who come here to speak before the board so they may state their cause with honesty and humility. . . . Lord we ask you to bless us all, that everything we do here tonight will move you to welcome us one day into your kingdom as good and faithful servants. We ask this in the name of our brother Jesus. Amen.” Id., at 45a. Some of the ministers spoke in a distinctly Christian idiom; and a minority invoked religious holidays, scripture, or doctrine, as in the following prayer: “Lord, God of all creation, we give you thanks and praise for your presence and action in the world. We look with anticipation to the celebration of Holy Week and Easter. It is in the solemn events of next week that we find the very heart and center of our Christian faith. We acknowledge the saving sacrifice ofJesus Christ on the cross. We draw strength, vitality, and confidence from his resurrection at Easter. . . . We pray for peace in the world, an end to terrorism, violence, conflict, and war. We pray for stability, democracy, and good government in those countries in which our armed forces are now serving, especially in Iraq and Afghanistan. . . . Praise and glory be yours, O Lord, now and forever more. Amen.” Id., at 88a–89a. Respondents Susan Galloway and Linda Stephens attended town board meetings to speak about issues of local concern, and they objected that the prayers violated their religious or philosophical views. At one meeting, Galloway admonished board members that she foundthe prayers “offensive,” “intolerable,” and an affront to a “diverse community.” Complaint in No. 08–cv–6088 (WDNY), ¶66. After respondents complained that Christian themes pervaded the prayers, to the exclusion of citizens who did not share those beliefs, the town invited a Jewish layman and the chairman of the local Baha’i temple to deliver prayers. A Wiccan priestess who had read press reports about the prayer controversy requested, and was granted, an opportunity to give the invocation. Galloway and Stephens brought suit in the United States District Court for the Western District of New York. They alleged that the town violated the First Amendment’s Establishment Clause by preferring Christians over other prayer givers and by sponsoring sectarian prayers, such as those given “in Jesus’ name.” 732 F. Supp. 2d 195, 203 (2010). They did not seek an end to the prayer practice, but rather requested an injunction that would limit the town to “inclusive and ecumenical” prayers that referred only to a “generic God” and would not associate the government with any one faith or belief. Id., at 210, 241. The District Court on summary judgment upheld the prayer practice as consistent with the First Amendment. It found no impermissible preference for Christianity, noting that the town had opened the prayer program to all creeds and excluded none. Although most of the prayer givers were Christian, this fact reflected only the predominantly Christian identity of the town’s congregations, rather than an official policy or practice of discriminating against minority faiths. The District Court found no authority for the proposition that the First Amendment required Greece to invite clergy from congregations beyond its borders in order to achieve a minimum level of religious diversity. The District Court also rejected the theory that legislative prayer must be nonsectarian. The court began its inquiry with the opinion in Marsh v. Chambers, 463 U. S. 783 , which permitted prayer in state legislatures by a chaplain paid from the public purse, so long as the prayer opportunity was not “exploited to proselytize or advance any one, or to disparage any other, faith or belief,” id., at 794–795. With respect to the prayer in Greece, the District Court concluded that references to Jesus, and the occasional request that the audience stand for the prayer, did not amount to impermissible proselytizing. It located in Marsh no additional requirement that the prayers be purged of sectarian content. In this regard the court quoted recent invocations offered in the U. S. House of Representatives “in the name of our Lord Jesus Christ,” e.g., 156 Cong Rec. H5205 (June 30, 2010), and situated prayer in this context as part a long tradition. Finally, the trial court noted this Court’s statement in County of Allegheny v. American Civil Liberties Union, Greater Pittsburgh Chapter, 492 U. S. 573, 603 (1989) , that the prayers in Marsh did not offend the Establishment Clause “because the particular chaplain had ‘removed all references to Christ.’ ” But the District Court did not read that statement to mandate that legislative prayer be nonsectarian, at least in circumstances where the town permitted clergy from a variety of faiths to give invocations. By welcoming many viewpoints, the District Court concluded, the town would be unlikely to give the impression that it was affiliating itself with any one religion. The Court of Appeals for the Second Circuit reversed. 681 F. 3d 20, 34 (2012). It held that some aspects of the prayer program, viewed in their totality by a reasonable observer, conveyed the message that Greece was endorsing Christianity. The town’s failure to promote the prayer opportunity to the public, or to invite ministers from congregations outside the town limits, all but “ensured a Christian viewpoint.” Id., at 30–31. Although the court found no inherent problem in the sectarian content of the prayers, it concluded that the “steady drumbeat” of Christian prayer, unbroken by invocations from other faith traditions, tended to affiliate the town with Christianity. Id., at 32. Finally, the court found it relevant that guest clergy sometimes spoke on behalf of all present at the meeting, as by saying “let us pray,” or by asking audience members to stand and bow their heads: “The invitation . . . to participate in the prayer . . . placed audience members who are nonreligious or adherents of non-Christian religion in the awkward position of either participating in prayers invoking beliefs they did not share or appearing to show disrespect for the invocation.” Ibid. That board members bowed their heads or made the sign of the cross further conveyed the message that the town endorsed Christianity. The Court of Appeals emphasized that it was the “interaction of the facts present in this case,” rather than any single element, that rendered the prayer unconstitutional. Id., at 33. Having granted certiorari to decide whether the town’s prayer practice violates the Establishment Clause, 569 U. S. ___ (2013), the Court now reverses the judgment of the Court of Appeals. II In Marsh v. Chambers, 463 U. S. 783 , the Court found no First Amendment violation in the Nebraska Legislature’s practice of opening its sessions with a prayer delivered by a chaplain paid from state funds. The decision concluded that legislative prayer, while religious in nature, has long been understood as compatible with the Establishment Clause. As practiced by Congress since the framing of the Constitution, legislative prayer lends grav-ity to public business, reminds lawmakers to transcend petty differences in pursuit of a higher purpose, and expresses a common aspiration to a just and peaceful soci-ety. See Lynch v. Donnelly, 465 U. S. 668, 693 (1984) (O’Connor, J., concurring); cf. A. Adams & C. Emmerich, A Nation Dedicated to Religious Liberty 83 (1990). The Court has considered this symbolic expression to be a “tolerable acknowledgement of beliefs widely held,” Marsh, 463 U. S., at 792, rather than a first, treacherous step towards establishment of a state church. Marsh is sometimes described as “carving out an exception” to the Court’s Establishment Clause jurisprudence, because it sustained legislative prayer without subjecting the practice to “any of the formal ‘tests’ that have traditionally structured” this inquiry. Id., at 796, 813 (Brennan, J., dissenting). The Court in Marsh found those tests unnecessary because history supported the conclusion that legislative invocations are compatible with the Establishment Clause. The First Congress made it an early item of business to appoint and pay official chaplains, and both the House and Senate have maintained the office virtually uninterrupted since that time. See id., at 787–789, and n. 10; N. Feldman, Divided by God 109 (2005). But see Marsh, supra, at 791–792, and n. 12 (noting dissenting views among the Framers); Madison, “Detached Memoranda”, 3 Wm. & Mary Quarterly 534, 558–559 (1946) (hereinafter Madison’s Detached Memoranda). When Marsh was decided, in 1983, legislative prayer had persisted in the Nebraska Legislature for more than a cen-tury, and the majority of the other States also had the same, consistent practice. 463 U. S., at 788–790, and n. 11. Although no information has been cited by the parties to indicate how many local legislative bodies open their meetings with prayer, this practice too has historical precedent. See Reports of Proceedings of the City Council of Boston for the Year Commencing Jan. 1, 1909, and Ending Feb. 5, 1910, pp. 1–2 (1910) (Rev. Arthur Little) (“And now we desire to invoke Thy presence, Thy blessing, and Thy guidance upon those who are gathered here this morning . . .”). “In light of the unambiguous and unbroken history of more than 200 years, there can be no doubt that the practice of opening legislative sessions with a prayer has become part of the fabric of our society.” Marsh, supra, at 792. Yet Marsh must not be understood as permitting a practice that would amount to a constitutional violation if not for its historical foundation. The case teaches instead that the Establishment Clause must be interpreted “by reference to historical practices and understandings.” County of Allegheny, 492 U. S., at 670 (Kennedy, J., concurring in judgment in part and dissenting in part). That the First Congress provided for the appointment of chaplains only days after approving language for the First Amendment demonstrates that the Framers considered legislative prayer a benign acknowledgment of religion’s role in society. D. Currie, The Constitution in Congress: The Federalist Period 1789–1801, pp. 12–13 (1997). In the 1850’s, the judiciary committees in both the House and Senate reevaluated the practice of official chaplaincies after receiving petitions to abolish the office. The committees concluded that the office posed no threat of an establishment because lawmakers were not compelled to attend the daily prayer, S. Rep. No. 376, 32d Cong., 2d Sess., 2 (1853); no faith was excluded by law, nor any favored, id., at 3; and the cost of the chaplain’s salary imposed a vanishingly small burden on taxpayers, H. Rep. No. 124, 33d Cong., 1st Sess., 6 (1854). Marsh stands for the proposition that it is not necessary to define the precise boundary of the Establishment Clause where history shows that the specific practice is permitted. Any test the Court adopts must acknowledge a practice that was accepted by the Framers and has withstood the critical scrutiny of time and political change. County of Allegheny, supra, at 670 (opinion of Kennedy, J.); see also School Dist. of Abington Township v. Schempp, 374 U. S. 203, 294 (1963) (Brennan, J., concurring) (“[T]he line we must draw between the permissible and the impermissible is one which accords with history and faithfully reflects the understanding of the Founding Fathers”). A test that would sweep away what has so long been settled would create new controversy and begin anew the very divisions along religious lines that the Establishment Clause seeks to prevent. See Van Orden v. Perry, 545 U. S. 677 –704 (2005) (Breyer, J., concurring in judgment). The Court’s inquiry, then, must be to determine whether the prayer practice in the town of Greece fits within the tradition long followed in Congress and the state legislatures. Respondents assert that the town’s prayer exercise falls outside that tradition and transgresses the Establishment Clause for two independent but mutually reinforcing reasons. First, they argue that Marsh did not approve prayers containing sectarian language or themes, such as the prayers offered in Greece that referred to the “death, resurrection, and ascension of the Savior Jesus Christ,” App. 129a, and the “saving sacrifice of Jesus Christ on the cross,” id., at 88a. Second, they argue that the setting and conduct of the town board meetings create social pressures that force nonadherents to remain in the room or even feign participation in order to avoid offending the representatives who sponsor the prayer and will vote on matters citizens bring before the board. The sectarian content of the prayers compounds the subtle coercive pressures, they argue, because the nonbeliever who might tolerate ecumenical prayer is forced to do the same for prayer that might be inimical to his or her beliefs. A Respondents maintain that prayer must be nonsectarian, or not identifiable with any one religion; and they faultthe town for permitting guest chaplains to deliver prayers that “use overtly Christian terms” or “invoke specifics of Christian theology.” Brief for Respondents 20. A prayer is fitting for the public sphere, in their view, only if it contains the ‘ “most general, nonsectarian reference to God,’ ” id., at 33 (quoting M. Meyerson, Endowed by Our Creator: The Birth of Religious Freedom in America 11–12 (2012)), and eschews mention of doctrines associated with any one faith, Brief for Respondents 32–33. They argue that prayer which contemplates “the workings of the Holy Spirit, the events of Pentecost, and the belief that God ‘has raisedup the Lord Jesus’ and ‘will raise us, in our turn, and put us by His side’ ” would be impermissible, as would any prayer that reflects dogma particular to a single faith tradition. Id., at 34 (quoting App. 89a and citing id., at 56a, 123a, 134a). An insistence on nonsectarian or ecumenical prayer as a single, fixed standard is not consistent with the tradition of legislative prayer outlined in the Court’s cases. The Court found the prayers in Marsh consistent with the First Amendment not because they espoused only a ge-neric theism but because our history and tradition have shown that prayer in this limited context could “coexis[t] with the principles of disestablishment and religious freedom.” 463 U. S., at 786. The Congress that drafted the First Amendment would have been accustomed to invocations containing explicitly religious themes of the sort respondents find objectionable. One of the Senate’s first chaplains, the Rev. William White, gave prayers in a series that included the Lord’s Prayer, the Collect forAsh Wednesday, prayers for peace and grace, a general thanksgiving, St. Chrysostom’s Prayer, and a prayer seeking “the grace of our Lord Jesus Christ, &c.” Letter from W. White to H. Jones (Dec. 29, 1830), in B. Wilson, Memoir of the Life of the Right Reverend William White, D. D., Bishop of the Protestant Episcopal Church in the State of Pennsylvania 322 (1839); see also New Hampshire Patriot & State Gazette, Dec. 15, 1823, p. 1 (describing a Senate prayer addressing the “Throne of Grace”); Cong. Globe, 37th Cong., 1st Sess., 2 (1861) (reciting the Lord’s Prayer). The decidedly Christian nature of these prayers must not be dismissed as the relic of a time when our Nation was less pluralistic than it is today. Congress continues to permit its appointed and visiting chaplains to express themselves in a religious idiom. It acknowledges our growing diversity not by proscribing sectarian content but by welcoming ministers of many creeds. See, e.g., 160 Cong. Rec. S1329 (Mar. 6, 2014) (Dalai Lama) (“I am a Buddhist monk—a simple Buddhist monk—so we pray to Buddha and all other Gods”); 159 Cong. Rec. H7006 (Nov. 13, 2013) (Rabbi Joshua Gruenberg) (“Our God and God of our ancestors, Everlasting Spirit of the Universe . . .”); 159 Cong. Rec. H3024 (June 4, 2013) (Satguru Bodhinatha Veylanswami) (“Hindu scripture declares, without equivocation, that the highest of high ideals is to never know-ingly harm anyone”); 158 Cong. Rec. H5633 (Aug. 2, 2012) (Imam Nayyar Imam) (“The final prophet of God, Muhammad, peace be upon him, stated: ‘The leaders of a people are a representation of their deeds’ ”). The contention that legislative prayer must be generic or nonsectarian derives from dictum in County of Allegheny, 492 U. S. 573 , that was disputed when written and has been repudiated by later cases. There the Court held that a crèche placed on the steps of a county courthouse to celebrate the Christmas season violated the Establishment Clause because it had “the effect of endorsing a patently Christian message.” Id., at 601. Four dissenting Justices disputed that endorsement could be the proper test, as it likely would condemn a host of traditional practices that recognize the role religion plays in our society, among them legislative prayer and the “forthrightly religious” Thanksgiving proclamations issued by nearly every President since Washington. Id., at 670–671. The Court sought to counter this criticism by recasting Marsh to permit only prayer that contained no overtly Christian references: “However history may affect the constitutionality of nonsectarian references to religion by the government, history cannot legitimate practices that demonstrate the government’s allegiance to a particular sect or creed . . . . The legislative prayers involved in Marsh did not violate this principle because the particular chaplain had ‘removed all references to Christ.’ ” Id., at 603 (quoting Marsh, supra, at 793, n. 14; footnote omitted). This proposition is irreconcilable with the facts of Marsh and with its holding and reasoning. Marsh nowhere suggested that the constitutionality of legislative prayer turns on the neutrality of its content. The opinion noted that Nebraska’s chaplain, the Rev. Robert E. Palmer, modu-lated the “explicitly Christian” nature of his prayer and “removed all references to Christ” after a Jewish law-maker complained. 463 U. S., at 793, n. 14. With this foot-note, the Court did no more than observe the practical demands placed on a minister who holds a permanent, appointed position in a legislature and chooses to write his or her prayers to appeal to more members, or at least to give less offense to those who object. See Mallory, “An Officer of the House Which Chooses Him, and Nothing More”: How Should Marsh v. Chambers Apply to Rotating Chaplains?, 73 U. Chi. L. Rev. 1421, 1445 (2006). Marsh did not suggest that Nebraska’s prayer practice would have failed had the chaplain not acceded to the legislator’s request. Nor did the Court imply the rule that prayer violates the Establishment Clause any time it is given in the name of a figure deified by only one faith or creed. See Van Orden, 545 U. S., at 688, n. 8 (recognizing that the prayers in Marsh were “often explicitly Christian” and rejecting the view that this gave rise to an establishment violation). To the contrary, the Court instructed that the “content of the prayer is not of concern to judges,” provided “there is no indication that the prayer opportunity has been exploited to proselytize or advance any one, or to disparage any other, faith or belief.” 463 U. S., at 794–795. To hold that invocations must be nonsectarian would force the legislatures that sponsor prayers and the courts that are asked to decide these cases to act as supervisors and censors of religious speech, a rule that would involve government in religious matters to a far greater degree than is the case under the town’s current practice of neither editing or approving prayers in advance nor criticizing their content after the fact. Cf. Hosanna-Tabor Evangelical Lutheran Church and School v. EEOC, 565 U. S. ___, ___ (2012) (slip op., at 13–14). Our Government is prohibited from prescribing prayers to be recited in our public institutions in order to promote a preferred system of belief or code of moral behavior. Engel v. Vitale, 370 U. S. 421, 430 (1962) . It would be but a few steps removed from that prohibition for legislatures to require chaplains to redact the religious content from their message in order to make it acceptable for the public sphere. Government may not mandate a civic religion that stifles any but the most generic reference to the sacred any more than it may prescribe a religious orthodoxy. See Lee v. Weisman, 505 U. S. 577, 590 (1992) (“The suggestion that government may establish an official or civic religion as a means of avoiding the establishment of a religion with more specific creeds strikes us as a contradiction that cannot be ac-cepted”); Schempp, 374 U. S., at 306 (Goldberg, J., concurring) (arguing that “untutored devotion to the concept of neutrality” must not lead to “a brooding and pervasive devotion to the secular”). Respondents argue, in effect, that legislative prayer may be addressed only to a generic God. The law and the Court could not draw this line for each specific prayer or seek to require ministers to set aside their nuanced and deeply personal beliefs for vague and artificial ones. There is doubt, in any event, that consensus might be reached as to what qualifies as generic or nonsectarian. Honorifics like “Lord of Lords” or “King of Kings” might strike a Christian audience as ecumenical, yet these titles may have no place in the vocabulary of other faith tradi-tions. The difficulty, indeed the futility, of sifting sectarian from nonsectarian speech is illustrated by a letter thata lawyer for the respondents sent the town in the early stages of this litigation. The letter opined that references to “Father, God, Lord God, and the Almighty” would be acceptable in public prayer, but that references to “Jesus Christ, the Holy Spirit, and the Holy Trinity” would not. App. 21a. Perhaps the writer believed the former grouping would be acceptable to monotheists. Yet even seemingly general references to God or the Father might alienate nonbelievers or polytheists. McCreary County v. American Civil Liberties Union of Ky., 545 U. S. 844, 893 (2005) (Scalia, J., dissenting). Because it is unlikely that prayer will be inclusive beyond dispute, it would be unwise to adopt what respondents think is the next-best option: permitting those religious words, and only those words, that are acceptable to the majority, even if they will exclude some. Torcaso v. Watkins, 367 U. S. 488, 495 (1961) . The First Amendment is not a majority rule, and government may not seek to define permissible categories of religious speech. Once it invites prayer into the public sphere, government must permit a prayer giver to address his or her own God or gods as conscience dictates, unfettered by what an administrator or judge considers to be nonsectarian. In rejecting the suggestion that legislative prayer must be nonsectarian, the Court does not imply that no constraints remain on its content. The relevant constraint derives from its place at the opening of legislative sessions, where it is meant to lend gravity to the occasion and reflect values long part of the Nation’s heritage. Prayer that is solemn and respectful in tone, that invites lawmakers to reflect upon shared ideals and common ends before they embark on the fractious business of governing, serves that legitimate function. If the course and practice over time shows that the invocations denigrate nonbeliev-ers or religious minorities, threaten damnation, or preach conversion, many present may consider the prayer to fall short of the desire to elevate the purpose of the occasion and to unite lawmakers in their common effort. That circumstance would present a different case than the one presently before the Court. The tradition reflected in Marsh permits chaplains to ask their own God for blessings of peace, justice, and freedom that find appreciation among people of all faiths. That a prayer is given in the name of Jesus, Allah, or Jehovah, or that it makes passing reference to religious doctrines, does not remove it from that tradition. These religious themes provide particular means to universal ends. Prayer that reflects beliefs specific to only some creeds can still serve to solemnize the occasion, so long as the practice over time is not “exploited to proselytize or advance any one, or to disparage any other, faith or belief.” Marsh, 463 U. S., at 794–795. It is thus possible to discern in the prayers offered to Congress a commonality of theme and tone. While these prayers vary in their degree of religiosity, they often seek peace for the Nation, wisdom for its lawmakers, and justice for its people, values that count as universal and that are embodied not only in religious traditions, but in our founding documents and laws. The first prayer delivered to the Continental Congress by the Rev. Jacob Duché on Sept. 7, 1774, provides an example: “Be Thou present O God of Wisdom and direct the counsel of this Honorable Assembly; enable them to settle all things on the best and surest foundations; that the scene of blood may be speedily closed; that Order, Harmony, and Peace be effectually restored, and the Truth and Justice, Religion and Piety, prevail and flourish among the people. “Preserve the health of their bodies, and the vigor of their minds, shower down on them, and the millions they here represent, such temporal Blessings as Thou seest expedient for them in this world, and crown them with everlasting Glory in the world to come. All this we ask in the name and through the merits ofJesus Christ, Thy Son and our Saviour, Amen.” W. Federer, America’s God and Country 137 (2000). From the earliest days of the Nation, these invocations have been addressed to assemblies comprising many different creeds. These ceremonial prayers strive for the idea that people of many faiths may be united in a community of tolerance and devotion. Even those who dis-agree as to religious doctrine may find common ground in the desire to show respect for the divine in all aspects of their lives and being. Our tradition assumes that adult citizens, firm in their own beliefs, can tolerate and perhaps appreciate a ceremonial prayer delivered by a person of a different faith. See Letter from John Adams to Abigail Adams (Sept. 16, 1774), in C. Adams, Familiar Letters of John Adams and His Wife Abigail Adams, During the Revolution 37–38 (1876). The prayers delivered in the town of Greece do not fall outside the tradition this Court has recognized. A number of the prayers did invoke the name of Jesus, the Heavenly Father, or the Holy Spirit, but they also invoked universal themes, as by celebrating the changing of the seasons or calling for a “spirit of cooperation” among town leaders. App. 31a, 38a. Among numerous examples of such prayer in the record is the invocation given by the Rev. Richard Barbour at the September 2006 board meeting: “Gracious God, you have richly blessed our nationand this community. Help us to remember your generosity and give thanks for your goodness. Bless the elected leaders of the Greece Town Board as they conduct the business of our town this evening. Give them wisdom, courage, discernment and a single-minded desire to serve the common good. We ask your blessing on all public servants, and especially on our police force, firefighters, and emergency medical personnel. . . . Respectful of every religious tradition, I offer this prayer in the name of God’s only son Jesus Christ, the Lord, Amen.” Id., at 98a–99a. Respondents point to other invocations that disparaged those who did not accept the town’s prayer practice. One guest minister characterized objectors as a “minority” who are “ignorant of the history of our country,” id., at 108a, while another lamented that other towns did not have “God-fearing” leaders, id., at 79a. Although these two remarks strayed from the rationale set out in Marsh, they do not despoil a practice that on the whole reflects and embraces our tradition. Absent a pattern of prayers that over time denigrate, proselytize, or betray an impermissible government purpose, a challenge based solely on the content of a prayer will not likely establish a constitutional violation. Marsh, indeed, requires an inquiry into the prayer opportunity as a whole, rather than into the contents of a single prayer. 463 U. S., at 794–795. Finally, the Court disagrees with the view taken by the Court of Appeals that the town of Greece contravened the Establishment Clause by inviting a predominantly Christian set of ministers to lead the prayer. The town made reasonable efforts to identify all of the congregations located within its borders and represented that it would welcome a prayer by any minister or layman who wished to give one. That nearly all of the congregations in town turned out to be Christian does not reflect an aversion or bias on the part of town leaders against minority faiths. So long as the town maintains a policy of nondiscrimination, the Constitution does not require it to search beyond its borders for non-Christian prayer givers in an effort to achieve religious balancing. The quest to promote “a ‘diversity’ of religious views” would require the town “to make wholly inappropriate judgments about the number of religions [it] should sponsor and the relative frequency with which it should sponsor each,” Lee, 505 U. S., at 617 (Souter, J., concurring), a form of government entanglement with religion that is far more troublesome than the current approach. B Respondents further seek to distinguish the town’s prayer practice from the tradition upheld in Marsh on the ground that it coerces participation by nonadherents. They and some amici contend that prayer conducted in the intimate setting of a town board meeting differs in fundamental ways from the invocations delivered in Congress and state legislatures, where the public remains segregated from legislative activity and may not address the body except by occasional invitation. Citizens attend town meetings, on the other hand, to accept awards; speak on matters of local importance; and petition the board for action that may affect their economic interests, such as the granting of permits, business licenses, and zoning variances. Respondents argue that the public may feel subtle pressure to participate in prayers that violate their beliefs in order to please the board members from whom they are about to seek a favorable ruling. In their view the fact that board members in small towns know many of their constituents by name only increases the pressure to conform. It is an elemental First Amendment principle that government may not coerce its citizens “to support or participate in any religion or its exercise.” County of Allegheny, 492 U. S., at 659 (Kennedy, J., concurring in judgment in part and dissenting in part); see also Van Orden, 545 U. S., at 683 (plurality opinion) (recognizing that our “institutions must not press religious observances upon their citizens”). On the record in this case the Court is not persuaded that the town of Greece, through the act of offering a brief, solemn, and respectful prayer to open its monthly meetings, compelled its citizens to engage in a religious observance. The inquiry remains a fact-sensitive one that considers both the setting in which the prayer arises and the audience to whom it is directed. The prayer opportunity in this case must be evaluated against the backdrop of historical practice. As a practice that has long endured, legislative prayer has become part of our heritage and tradition, part of our expressive idiom, similar to the Pledge of Allegiance, inaugural prayer, or the recitation of “God save the United States and this honorable Court” at the opening of this Court’s sessions. See Lynch, 465 U. S., at 693 (O’Connor, J., concurring). It is presumed that the reasonable observer is acquainted with this tradition and understands that its purposes are to lend gravity to public proceedings and to acknowledge the place religion holds in the lives of many private citizens, not to afford government an opportunity to proselytize or force truant constituents into the pews. See Salazar v. Buono, 559 U. S. 700 –721 (2010) (plurality opinion); Santa Fe Independent School Dist. v. Doe, 530 U. S. 290, 308 (2000) . That many appreciate these acknowledgments of the divine in our public institutions does not suggest that those who disagree are compelled to join the expression or approve its content. West Virginia Bd. of Ed. v. Barnette, 319 U. S. 624, 642 (1943) . The principal audience for these invocations is not, indeed, the public but lawmakers themselves, who may find that a moment of prayer or quiet reflection sets the mind to a higher purpose and thereby eases the task of governing. The District Court in Marsh described the prayer exercise as “an internal act” directed at the Nebraska Legislature’s “own members,” Chambers v. Marsh, 504 F. Supp. 585, 588 (Neb. 1980), rather than an effort to promote religious observance among the public. See also Lee, 505 U. S., at 630, n. 8 (Souter, J., concurring) (describing Marsh as a case “in which government officials invoke[d] spiritual inspiration entirely for their own benefit”); Atheists of Fla., Inc. v. Lakeland, 713 F. 3d 577, 583 (CA11 2013) (quoting a city resolution providing for prayer “for the benefit and blessing of” elected leaders); Madison’s Detached Memoranda 558 (characterizing prayer in Congress as “religious worship for national representatives”); Brief for U. S. Senator Marco Rubio et al. as Amici Curiae 30–33; Brief for 12 Members of Congress as Amici Curiae 6. To be sure, many members of the public find these prayers meaningful and wish to join them. But their purpose is largely to accommodate the spiritual needs of lawmakers and connect them to a tradition dating to the time of the Framers. For members of town boards and commissions, who often serve part-time and as volunteers, ceremonial prayer may also reflect the values they hold as private citizens. The prayer is an opportunity for them to show who and what they are without denying the right to dissent by those who disagree. The analysis would be different if town board members directed the public to participate in the prayers, singled out dissidents for opprobrium, or indicated that their decisions might be influenced by a person’s acquiescence in the prayer opportunity. No such thing occurred in the town of Greece. Although board members themselves stood, bowed their heads, or made the sign of the cross during the prayer, they at no point solicited similar gestures by the public. Respondents point to several occasions where audience members were asked to rise for the prayer. These requests, however, came not from town leaders but from the guest ministers, who presumably are accustomed to directing their congregations in this way and might have done so thinking the action was inclusive, not coercive. See App. 69a (“Would you bow your heads with me as we invite the Lord’s presence here tonight?”); id., at 93a (“Let us join our hearts and minds together in prayer”); id., at 102a (“Would you join me in a moment of prayer?”); id., at 110a (“Those who are willing may join me now in prayer”). Respondents suggest that constituents might feel pressure to join the prayers to avoid irritating the officials who would be ruling on their petitions, but this argument has no evidentiary support. Nothing in the record indicates that town leaders allocated benefits and burdens based on participation in the prayer, or that citizens were received differently depending on whether they joined the invocation or quietly declined. In no instance did town leaders signal disfavor toward nonparticipants or suggest that their stature in the community was in any way diminished. A practice that classified citizens based on their religious views would violate the Constitution, but that is not the case before this Court. In their declarations in the trial court, respondents stated that the prayers gave them offense and made them feel excluded and disrespected. Offense, however, does not equate to coercion. Adults often encounter speech they find disagreeable; and an Establishment Clause violation is not made out any time a person experiences a sense of affront from the expression of contrary religious views in a legislative forum, especially where, as here, any member of the public is welcome in turn to offer an invocation reflecting his or her own convictions. See Elk Grove Unified School Dist. v. Newdow, 542 U. S. 1, 44 (2004) (O’Connor, J., concurring) (“The compulsion of which Justice Jackson was concerned . . . was of the direct sort—the Constitution does not guarantee citizens a rightentirely to avoid ideas with which they disagree”). If circum-stances arise in which the pattern and practice of ceremonial, legislative prayer is alleged to be a means to coerce or intimidate others, the objection can be addressed in the regular course. But the showing has not been made here, where the prayers neither chastised dissenters nor attempted lengthy disquisition on religious dogma. Courts remain free to review the pattern of prayers over time to determine whether they comport with the tradition of solemn, respectful prayer approved in Marsh, or whether coercion is a real and substantial likelihood. But in the general course legislative bodies do not engage in impermissible coercion merely by exposing constituents to prayer they would rather not hear and in which they need not participate. See County of Allegheny, 492 U. S., at 670 (Kennedy, J., concurring in judgment in part and dissenting in part). This case can be distinguished from the conclusions and holding of Lee v. Weisman, 505 U. S. 577 . There the Court found that, in the context of a graduation where school authorities maintained close supervision over the conduct of the students and the substance of the ceremony, a religious invocation was coercive as to an objecting student. Id., at 592–594; see also Santa Fe Independent School Dist., 530 U. S., at 312. Four Justices dissented in Lee, but the circumstances the Court confronted there are not present in this case and do not control its outcome. Nothing in the record suggests that members of the public are dissuaded from leaving the meeting room during the prayer, arriving late, or even, as happened here, making a later protest. In this case, as in Marsh, board members and constituents are “free to enter and leave with little comment and for any number of reasons.” Lee, supra, at 597. Should nonbelievers choose to exit the room during a prayer they find distasteful, their absence will not stand out as disrespectful or even noteworthy. And should they remain, their quiet acquiescence will not, in light of our traditions, be interpreted as an agreement with the words or ideas expressed. Neither choice represents an unconstitutional imposition as to mature adults, who “presumably” are “not readily susceptible to religious indoctrination or peer pressure.” Marsh, 463 U. S., at 792 (internal quotation marks and citations omitted). In the town of Greece, the prayer is delivered during the ceremonial portion of the town’s meeting. Board members are not engaged in policymaking at this time, but in more general functions, such as swearing in new police officers, inducting high school athletes into the town hall of fame, and presenting proclamations to volunteers, civic groups, and senior citizens. It is a moment for town leaders to recognize the achievements of their constituents and the aspects of community life that are worth celebrating. By inviting ministers to serve as chaplain for the month, and welcoming them to the front of the room alongside civic leaders, the town is acknowledging the central place that religion, and religious institutions, hold in the lives of those present. Indeed, some congregations are not simply spiritual homes for town residents but also the provider of social services for citizens regardless of their beliefs. See App. 31a (thanking a pastor for his “community involvement”); id., at 44a (thanking a deacon “for the job that you have done on behalf of our community”). The inclusion of a brief, ceremonial prayer as part of a larger exercise in civic recognition suggests that its purpose and effect are to acknowledge religious leaders and the institutions they represent rather than to exclude or coerce nonbelievers. Ceremonial prayer is but a recognition that, since this Nation was founded and until the present day, many Americans deem that their own existence must be understood by precepts far beyond the authority of government to alter or define and that willing participation in civic affairs can be consistent with a brief acknowledgment of their belief in a higher power, always with due respect for those who adhere to other beliefs. The prayer in this case has a permissible ceremonial purpose. It is not an unconstitutional establishment of religion. * * * The town of Greece does not violate the First Amendment by opening its meetings with prayer that comports with our tradition and does not coerce participation by nonadherents. The judgment of the U. S. Court of Appeals for the Second Circuit is reversed. It is so ordered.Notes 1 * and join this opinion in full. and join this opinion except as to Part II–B.

571.US.359

Vandenberg Air Force Base has been designated a “closed base,” meaning that civilians may not enter without express permission. The Air Force has granted an easement over two areas of the Base, with the result that two public highways traverse the Base. Adjacent to one of those highways is an area that the Government has designated for peaceful protests. The Base commander has enacted several restrictions to control the protest area and has issued an advisory stating that anyone who fails to adhere to the protest area policies may be barred from entering the Base. Petitioner Apel was barred from the Base for trespassing and vandalism, but continued to enter the protest area. A Magistrate Judge convicted him of violating 18 U. S. C. §1382, which makes it a crime to reenter a “military. . . installation” after having been ordered not to do so “by any officer or person in command.” On appeal, the Federal District Court rejected Apel’s defense that §1382 does not apply to the designated protest area. The Ninth Circuit reversed. It held that because the easement through Vandenberg deprived the Government of exclusive possession, §1382 did not cover the portion of the Base where Apel’s protest occurred. Held: A “military. . . installation” for purposes of §1382 encompasses the commanding officer’s area of responsibility, and it includes Vandenberg’s highways and protest area. Pp. 6–14. (a) Contrary to Apel’s argument, §1382 does not require exclusive possession and control. The statute is written broadly to apply to many different kinds of military places, and nothing in its text defines those places in terms of the access granted to the public or the nature of the Government’s possessory interest. See United States v. Albertini, 472 U.S. 675, 682. Nor have military places been defined historically as land withdrawn from public use. The common feature of the places described in §1382 is that they have defined boundaries and are subject to the command authority of a military officer. This conclusion is confirmed by United States v. Phisterer, 94 U.S. 219, 222, which defined the term “military station” as a place “where military duty is performed or military protection afforded.” And while some Executive Branch documents have said that §1382 requires exclusive possession, those opinions are nonbinding, and this Court has never held that the Government’s reading of a criminal statute is entitled to any deference. Pp. 7–10. (b) Section 1382 applies to any place with a defined boundary that is under the command of a military officer. Apel contends that the highways and protest area are outside the Base because they lie outside fenced areas on the Base, but this argument assumes the conclusion. The United States has placed the entire Vandenberg property under the administration of the Air Force. The Air Force’s choice to secure a portion of the Base more closely does not alter its boundaries or diminish its commander’s jurisdiction. Apel’s further contention that the highways and protest area are uncontrolled spaces where military operations are not performed is contrary to the record: The Base commander has enacted rules to restrict the manner of protests in the designated area and has publicly stated that persons barred from Vandenberg may not enter the Base to protest; the District Court found that the Government exercises substantial control over the protest area; the easement itself reserves to the Base commander the authority to restrict access to the entire Base when necessary and reserves to the United States rights of way for all purposes; and the Base commander has occasionally closed the highways to the public for security purposes or when conducting a military launch. In any event, §1382 does not require base commanders to make continuous, uninterrupted use of a place within their jurisdiction, lest they lose authority to exclude certain individuals. Such a use-it-or-lose-it rule would frustrate the administration of military facilities, raise difficult questions for judges, and discourage commanders from opening portions of their bases for public convenience. Pp. 10–13. (c) Apel’s argument that the statute was unconstitutional as applied was not reached by the Ninth Circuit and, thus, is not addressed here. P. 13. 676 F.3d 1202, vacated and remanded. Roberts, C. J., delivered the opinion for a unanimous Court. Ginsburg, J., filed a concurring opinion, in which Sotomayor, J., joined. Alito, J., filed a concurring opinion.

Federal law makes it a crime to reenter a “military . . . installation” after having been ordered not to do so “by any officer or person in command.” 18 U. S. C. §1382. The question presented is whether a portion of an Air Force base that contains a designated protest area and an easement for a public road qualifies as part of a “military installation.” I A Vandenberg Air Force Base is located in central California, near the coast, approximately 170 miles northwest of Los Angeles. The Base sits on land owned by the United States and administered by the Department of the Air Force. It is the site of sensitive missile and space launch facilities. The commander of Vandenberg has designated it a “closed base,” meaning that civilians may not enter without express permission. Memorandum for the General Public Re: Closed Base, from David J. Buck, Commander (Oct. 23, 2008), App. 51; see also 32 CFR §809a.2(b) (2013) (“Each [Air Force] commander is au-thorized to grant or deny access to their installations,and to exclude or remove persons whose presence isunauthorized”). Although the Base is closed, the Air Force has granted to the County of Santa Barbara “an easement for a right-of-way for a road or street” over two areas within Vandenberg. Department of the Air Force, Easement for Road or Street No. DA–04–353–ENG–8284 (Aug. 20, 1962), App. 35. Pursuant to that easement, two state roads traverse the Base. Highway 1 (the Pacific Coast Highway) runs through the eastern part of the Base and provides a route between the towns of Santa Maria and Lompoc. Highway 246 runs through the southern part of the Base and allows access to a beach and a train station on Vandenberg’s western edge. The State of California maintains and po-lices these highways as it does other state roads, except that its jurisdiction is merely “concurrent” with that ofthe Federal Government. Letter from Governor Edmund G. Brown, Jr., to Joseph C. Zengerle, Assistant Secretary of the Air Force (July 21, 1981), App. 40. The easement in-strument states that use of the roads “shall be subject to such rules and regulations as [the Base commander] may prescribe from time to time in order to properly protect the interests of the United States.” Easement, App. 36. The United States also “reserves to itself rights-of-way for all purposes” that would not create “unnecessary interference with . . . highway purposes.” Id., at 37. As relevant to this case, Highway 1 runs northwest several miles inside Vandenberg until it turns northeast at a 90 degree angle. There Highway 1 intersects with Lompoc Casmalia Road, which continues running northwest, and with California Boulevard, which runs southwest. In the east corner of this intersection there is a middle school. In the west corner there is a visitors’ center and a public bus stop. A short way down California Boulevard is the main entrance to the operational areas of the Base where military personnel live and work. Those areas are surrounded by a fence and entered by a security checkpoint. See Appendix, infra (maps from record). In the south corner of the intersection is an area that has been designated by the Federal Government for peaceful protests. A painted green line on the pavement, a temporary fence, Highway 1, and Lompoc Casmalia Road mark the boundaries of the protest area. Memorandum for the General Public Re: Limited Permission for Peaceful Protest Activity Policy, from David J. Buck, Commander (Oct. 23, 2008), App. 57–58. The Base commander has enacted several restrictions to control the protest area, including reserving the authority “for any reason” to withdraw permission to protest and “retain[ing] authority and control over who may access the installation, including access to roadway easements for purposes other than traversing by vehicle through the installation.” Ibid. A public advisory explains other rules for the protest area: demonstrations “must be coordinated and scheduled with [B]ase Public Affairs and [Base] Security Forces at least two (2) weeks in advance”; “[a]nyone failing to vacate installation property upon advisement from Security Forces will be cited for trespass pursuant to [ 18 U. S. C. §1382]”; and “[a]ctivities other than peaceful protests in this area are not permitted and are specifically prohib-ited.” U. S. Air Force Fact Sheet, Protest Advisory, App. 52–53. The advisory states, consistent with federal regulations, that anyone who fails to adhere to these policies may “receive an official letter barring you from entering Vandenberg.” Id., at 55; see also 32 CFR §809a.5 (“Under the authority of 50 U. S. C. [§]797, installation commanders may deny access to the installation through the use of a barment order”). And for any person who is “currently barred from Vandenberg AFB, there is no exception to the barment permitting you to attend peaceful protest activity on Vandenberg AFB property. If you are barred and attend a protest or are otherwise found on base, you will be cited and detained for a trespass violation due to the non-adherence of the barment order.” Protest Advisory, App. 54. B John Dennis Apel is an antiwar activist who demonstrates at Vandenberg. In March 2003, Apel trespassed beyond the designated protest area and threw blood on a sign for the Base. He was convicted for these actions, was sentenced to two months’ imprisonment, and was barred from the Base for three years. In May 2007, Apel returned to Vandenberg to protest. When he trespassed again and was convicted, he received another order barring him from Vandenberg, this time permanently, unless he followed specified procedures “to modify or revoke” the order. Memorandum for John D. Apel Re: Barment Order (Oct. 22, 2007), App. 63–65. The only exception to the barment was limited permission from the Base commander for Apel to “ ‘traverse’, meaning to travel . . . on [Highway] 1 and . . . on [Highway] 246 . . . . You are not authorized to deviate from these paved roadways onto [Vandenberg] property.” Id., at 64. The order informed Apel that ifhe reentered Vandenberg in violation of the order, he would “be subject to detention by Security Forces personnel and prosecution by civilian authorities for a violation of [ 18 U. S. C. §1382].” Ibid. Apel ignored the commander’s order and reentered Vandenberg several times during 2008 and 2009. That led the Base commander to serve Apel with an updated order, which informed him: “You continue to refuse to adhere to the rules and guidelines that have been put in place by me to protect and preserve order and to safeguard the persons and property under my jurisdiction by failing to remain in the area approved by me for peaceful demonstrations pursuant to [50] U. S. C. § 797 and 32 C. F. R. § 809a.0–[809]a.11. You cannot be expectedor trusted to abide by the protest guidance rules based upon this behavior. I consider your presence on this installation to be a risk and detrimental to my responsibility to protect and preserve order and to safeguard the persons and property under my jurisdiction. You are again ordered not to enter onto [Vandenberg] property, as provided in the October 22, 2007 order. The content and basis of that order is hereby incorporated by reference herein, EXCEPT that your barment will be for a period of three (3) years from the dateof this supplemental letter.” Memorandum for John D. Apel Re: Barment Order Dated Oct. 22, 2007 (served Jan. 31, 2010), App. 59–62. Apel ignored this barment order too, and on three occasions in 2010 he reentered Vandenberg to protest in the designated area. Each time Vandenberg security personnel reminded him of the barment order and instructed him to leave. Each time Apel refused. He was cited for violating §1382 and escorted off Base property. A Magistrate Judge convicted Apel and ordered him to pay a total of $355 in fines and fees. Apel appealed to the Federal District Court for the Central District of California. The District Court rejected Apel’s defense that §1382 does not apply to the designated protest area, holding that the military “has a sufficient possessory interest and exercises sufficient control over” the area. App. to Pet. for Cert. 14a. The court also concluded that Apel’s conviction would not violate the First Amendment. Id., at 13a. The United States Court of Appeals for the Ninth Circuit reversed, holding that the statute does not apply. Based on Circuit precedent, the Ninth Circuit interpreted §1382 to require the Government to prove that it has “the exclusive right of possession of the area on which the trespass allegedly occurred.” 676 F. 3d 1202, 1203 (2012) (citing United States v. Parker, 651 F. 3d 1180 (CA9 2011)). The court found that the easement through Vandenberg deprived the Government of exclusive possession of the roadway, so it concluded that §1382 does not cover the portion of the Base where Apel’s protest occurred. We granted certiorari, 569 U. S. ___ (2013), and now vacate the judgment. II Section 1382 provides in full: “Whoever, within the jurisdiction of the United States, goes upon any military, naval, or Coast Guard reservation, post, fort, arsenal, yard, station, or installation, for any purpose prohibited by law or lawful regulation; or “Whoever reenters or is found within any such reservation, post, fort, arsenal, yard, station, or installation, after having been removed therefrom or ordered not to reenter by any officer or person in command or charge thereof— “Shall be fined under this title or imprisoned not more than six months, or both.” Apel does not dispute that he was “found within” the lawful boundaries of Vandenberg, “within the jurisdiction of the United States,” after having been “ordered not to reenter” by the Base commander. §1382. And certainly Vandenberg would naturally be described as a “military installation”: it is an Air Force base, which a military commander has closed to the public (with limited exceptions), located on land owned by the United States and under the jurisdiction of the Air Force, where military personnel conduct sensitive missile operations. Against this straightforward interpretation, Apel insists that §1382 applies only where the military exercises exclusive possession and control, which, he contends, does not include land subject to a roadway easement. Apel further argues that the fence enclosing Vandenberg’s operational facilities marks the real boundary of the Base and that Vandenberg’s commander lacks authority to control the rest, or at least the designated protest area. We take his arguments in turn. A Apel asserts that the Ninth Circuit’s exclusive possession and control requirement “derives directly from the text of §1382.” Brief for Respondent 23. It does not. Section 1382 is written broadly to apply to many different kinds of military places: a “reservation, post, fort, arsenal, yard, station, or installation.” Nothing in the text defines those places in terms of the access granted to the public or the nature of the Government’s possessory interest. See United States v. Albertini, 472 U. S. 675, 682 (1985) (“The language of the statute does not limit §1382 to military bases where access is restricted”). Apel contends that the listed military places have histor-ically been defined as land withdrawn from public use. Not so. Historical sources are replete with references to military “forts” and “posts” that provided services to civilians, and were open for access by them. See, e.g., R. Wooster, Soldiers, Sutlers, and Settlers 64 (1987) (“The frontier forts of Texas were not simply army bases occupied solely by military personnel. They were often bustling communities that attracted merchants, laborers, settlers, and dependents”); Davis, The Sutler at Fort Bridger, 2 Western Hist. Q. 37, 37, 40–41 (Jan. 1971) (describing a 19th-century post in southwestern present Wyoming which included a “sutler,” a civilian merchant who set up shop inside the fort and sold wares both to soldiers and to civilians from outside the base). The common feature of the places described in §1382 is not that they are used exclusively by the military, but that they have defined boundaries and are subject to the command authority of a military officer. That makes sense, because the Solicitor General has informed us that a military commander’s authority is frequently defined by the boundaries of a particular place: When the Department of Defense establishes a base, military commanders assign a military unit to the base, and the commanding officer of the unit becomes the commander of the base. Tr. of Oral Arg. 6–7. Apel responds by invoking our decision in United States v. Phisterer, 94 U. S. 219 (1877) , which held that the term “military station” (in a different statute) did not includea soldier’s off-base home. But Phisterer only confirmsour conclusion that §1382 does not require exclusive use, possession, or control. For there we interpreted “military station” to mean “a place where troops are assembled, where military stores, animate or inanimate, are kept or distributed, where military duty is performed or military protection afforded,—where something, in short, more or less closely connected with arms or war is kept or is to be done.” Id., at 222. To describe a place as “more or less closely connected” with military activities hardly requires that the military hold an exclusive right to the property. Rather, “military duty” and “military protection” are synonymous with the exercise of military jurisdiction. And that, not coincidentally, is precisely how the term “military installation” is used elsewhere in federal law. See, e.g., 10 U. S. C. §2687(g)(1) (defining “military installation” as a “base . . . or other activity under the jurisdiction of the Department of Defense”); §2801(c)(4) (defining “military installation” as a “base . . . or other activity under the jurisdiction of the Secretary of a military department”); 32 CFR §809a.0 (“This part prescribes the commanders’ authority for enforcing order within or near Air Force installations under their jurisdiction and controlling entry to those installations”). Apel also relies on the fact that some Executive Branch documents, including the United States Attorneys’ Man-ual and opinions of the Air Force Judge Advocate General, have said that §1382 requires exclusive possession. Brief for Respondent 44–47. So they have, and that is a point in his favor. But those opinions are not intended to be binding. See Dept. of Justice, United States Attorneys’ Man-ual §1–1.100 (2009) (“The Manual provides only internal Department of Justice guidance. It is not intended to, does not, and may not be relied upon to create any rights, substantive or procedural, enforceable at law by any party in any matter civil or criminal”); 2 Civil Law Opinionsof The Judge Advocate General, United States Air Force 1978–1983 (Preface) (opinions of the Judge Advocate Gen-eral “are good starting points but should not be citedas precedence [sic] without first verifying the validity of the conclusions by independent research”). Their views may reflect overly cautious legal advice based on division in the lower courts. Or they may reflect legal error. Either way, we have never held that the Government’s reading of a criminal statute is entitled to any deference. See Crandon v. United States, 494 U. S. 152, 177 (1990) (Scalia, J., concurring in judgment). Today, as throughout our Nation’s history, there is sig-nificant variation in the ownership status of U. S. mil-itary sites around the world. Some are owned in fee, others are leased. Some are routinely open to the public, others are open for specific occasions or purposes, and no public access whatsoever is permitted on others. Many, including such well-known places as the Washington Navy Yard and the United States Air Force Academy, have roads running through them that are used freely by the public. Nothing in §1382 or our history suggests that the statute does not apply to a military base under the command of the Air Force, merely because the Government has conveyed a limited right to travel through a portion of the base or to assemble in a particular area. B Section 1382 is most naturally read to apply to places with a defined boundary under the command of a military officer. Apel argues, however, that Vandenberg’s commander has no authority on the highways running through the Base or, apparently, in the designated protest area. His arguments more or less reduce to two contentions: that the highways and protest area lie “outside the entrance to [a] closed military installation[ ],” Brief for Respondent 22, and that they are “uncontrolled” spaces where “no military operations are performed,” id., at 23. Neither contention is sound. First, to say that the highway and protest area are “outside” the Vandenberg installation is not a legal ar-gument; it simply assumes the conclusion. Perhaps recognizing as much, Apel tacks: He suggests that because Vandenberg’s operational facilities are surrounded by a fence and guarded by a security checkpoint, the Government has determined that it does not control the rest of the Base. The problem with this argument is that the United States has placed the entire Vandenberg property under the administration of the Air Force, which has defined that property as an Air Force base and designated the Base commander to exercise jurisdiction. Federal law makes the commander responsible “for the protection or security of” “property subject to the jurisdiction, administration, or in the custody of the Department of Defense.” 50 U. S. C. §§797(a)(2), (4); see also 32 CFR §809a.2(a) (“Air Force installation commanders are responsible for protecting personnel and property under their jurisdiction”). And pursuant to that authority, the Base commander has issued an order closing the entire base to the public. Buck Memorandum Re: Closed Base, App. 51; see also 32 CFR §809a.3 (“any directive issued by the commander of a military installation or facility, which includes the parameters for authorized entry to or exit from a military installation, is legally enforceable against all persons”). The fact that the Air Force chooses to securea portion of the Base more closely—be it with a fence, a checkpoint, or a painted green line—does not alter the boundaries of the Base or diminish the jurisdiction of the military commander. As for Apel’s claim that the protest area specifically is uncontrolled, the record is conclusively to the contrary. The Base commander “at all times has retained authority and control over who may access the installation,” including the protest area. Buck Memorandum Re: Protest Activity, App. 58. He has enacted rules to restrict the manner of protests in the designated area. Protest Advisory, App. 53. In particular, he requires two weeks’ notice to schedule a protest and prohibits the distribution of pamphlets or leaflets. Id., at 52–53. The Base com-mander has also publicly stated that persons who are barred from Vandenberg—for whatever reason—may not come onto the Base to protest. Id., at 54. And the District Court found, after hearing testimony, that “the Government exercises substantial control over the designated protest area, including, for example, patrolling the area.” App. to Pet. for Cert. 14a–15a. Apel has never disputed these facts. Instead Apel tells us that, by granting an easement, the military has “relinquished its right to exclude civilians from Highway 1,” Brief for Respondent 36, and that the easement does not “permit[ ]” use by the military, id., at 43. But the easement itself specifically reserves to Vandenberg’s commander the authority to restrict access to the entire Base, including Highway 1, when necessary “to properly protect the interests of the United States,” and likewise “reserves to [the United States] rights-of-way for all purposes.” Easement, App. 36. We simply do not understand how Apel can claim that “[n]othing in the easement contemplates, or even permits, military use or occupation; it provides for exclusive civil use and occupation.” Brief for Respondent 43. Moreover, the Base commander, in an exercise of his command authority, has notified the public that use of the roads is “limited to . . . vehicular travel activity through the base,” which does not include Apel’s protest activity. See Buck Memorandum Re: Closed Base, App. 51. Apel likewise offers no support for his contention that military functions do not occur on the easement highways. The Government has referred us to instances when the commander of Vandenberg has closed the highways to the public for security purposes or when conducting a military launch. Reply Brief 12, and n. 5; Tr. of Oral Arg. 8–9. In any event, there is no indication that Congress intended §1382 to require base commanders to make continuous, uninterrupted use of a place within their jurisdiction, lest they lose authority to exclude individuals who have vandalized military property and been determined to pose a threat to the order and security of the base. In sum, we decline Apel’s invitation to require civilian judges to examine U. S. military sites around the world, parcel by parcel, to determine which have roads, which have fences, and which have a sufficiently important, persistent military purpose. The use-it-or-lose-it rule that Apel proposes would frustrate the administration of military facilities and raise difficult questions for judges, who are not expert in military operations. And it would discourage commanders from opening portions of their bases for the convenience of the public. We think a much better reading of §1382 is that it reaches all property within the defined boundaries of a military place that is under the command of a military officer. III Much of the rest of Apel’s brief is devoted to arguing that §1382 would be unconstitutional as applied to him on this Base. But the Court of Appeals never reached Apel’s constitutional arguments, and we decline to do so in the first instance. Apel also attempts to repackage his First Amendment objections as a statutory interpretation argument based on constitutional avoidance. See Brief for Respondent 54 (“the statute should be interpreted . . . not to apply to peaceful protests on a public road outside of a closed military base over which an easement has been granted and that has been declared a protest zone”). But we do not “interpret” statutes by gerrymandering them with a list of exceptions that happen to describe a party’s case. “The canon [of constitutional avoidance] is not a method of adjudicating constitutional questions by other means.” Clark v. Martinez, 543 U. S. 371, 381 (2005) . Whether §1382 is unconstitutional as applied is a question we need not address. * * * Where a place with a defined boundary is under the administration of a military department, the limits of the “military installation” for purposes of §1382 are coterminous with the commanding officer’s area of responsibility. Those limits do not change when the commander invites the public to use a portion of the base for a road, a school, a bus stop, or a protest area, especially when the commander reserves authority to protect military property by, among other things, excluding vandals and trespassers. The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

572.US.157

Respondent Castleman moved to dismiss his indictment under 18 U. S. C. §922(g)(9), which forbids the possession of firearms by anyone convicted of a “misdemeanor crime of domestic violence.” He argued that his previous conviction for “intentionally or knowingly caus[ing] bodily injury to” the mother of his child, App. 27, did not qualify as a “misdemeanor crime of domestic violence” because it did not involve “the use or attempted use of physical force,” 18 U. S. C. §921(a)(33)(A)(ii). The District Court agreed, reasoning that “physical force” must entail violent contact and that one can cause bodily injury without violent contact, e.g., by poisoning. The Sixth Circuit affirmed on a different rationale. It held that the degree of physical force required for a conviction to constitute a “misdemeanor crime of domestic violence” is the same as that required for a “violent felony” under the Armed Career Criminal Act (ACCA), §924(e)(2)(B)(i)—namely, violent force—and that Castleman could have been convicted for causing slight injury by nonviolent conduct. Held: Castleman’s conviction qualifies as a “misdemeanor crime of domestic violence.” Pp. 4–16. (a) Section 922(g)(9)’s “physical force” requirement is satisfied by the degree of force that supports a common-law battery conviction—namely, offensive touching. Congress presumably intends to incorporate the common-law meaning of terms that it uses, and nothing suggests Congress intended otherwise here. The Sixth Circuit relied upon Johnson v. United States, 559 U.S. 133, in which the common-law meaning of “force” was found to be a “comical misfit,” id., at 145, when read into ACCA’s “violent felony” definition. But Johnson resolves this case in the Government’s favor: The very reasons for rejecting the common-law meaning in Johnson are reasons to embrace it here. First, whereas it was “unlikely” that Congress meant to incorporate in ACCA’s “violent felony” definition “a phrase that the common law gave peculiar meaning only in its definition of a misdemeanor,” id., at 141, it is likely that Congress meant to incorporate the misdemeanor-specific meaning of “force” in defining a “misdemeanor crime of domestic violence.” Second, whereas the word “violent” or “violence” standing alone “connotes a substantial degree of force,” id., at 140, that is not true of “domestic violence,” which is a term of art encompassing acts that one might not characterize as “violent” in a nondomestic context. Third, whereas this Court has hesitated to apply ACCA to “crimes which, though dangerous, are not typically committed by those whom one normally labels ‘armed career criminals,’ ” Begay v. United States, 553 U.S. 137, 146, there is no anomaly in grouping domestic abusers convicted of generic assault or battery offenses together with others whom §922(g) disqualifies from gun ownership. In addition, a contrary reading would have made §922(g)(9) inoperative in at least ten States when it was enacted. Pp. 4–10. (b) Under this definition of “physical force,” Castleman’s conviction qualifies as a “misdemeanor crime of domestic violence.” The application of the modified categorical approach—consulting Castleman’s state indictment to determine whether his conviction entailed the elements necessary to constitute the generic federal offense—is straightforward. Castleman pleaded guilty to “intentionally or knowingly caus[ing] bodily injury to” the mother of his child, and the knowing or intentional causation of bodily injury necessarily involves the use of physical force. First, a “bodily injury” must result from “physical force.” The common-law concept of “force” encompasses even its indirect application, making it impossible to cause bodily injury without applying force in the common-law sense. Second, the knowing or intentional application of force is a “use” of force. Leocal v. Ashcroft, 543 U.S. 1, distinguished. Pp. 10–13. (c) Castleman claims that legislative history, the rule of lenity, and the canon of constitutional avoidance weigh against this Court’s interpretation of §922(g)(9), but his arguments are unpersuasive. Pp. 14–15. 695 F.3d 582, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined. Scalia, J., filed an opinion concurring in part and concurring in the judgment. Alito, J., filed an opinion concurring in the judgment, in which Thomas, J., joined.

Recognizing that “[f]irearms and domestic strife are a potentially deadly combination,” United States v. Hayes, 555 U. S. 415, 427 (2009) , Congress forbade the possession of firearms by anyone convicted of “a misdemeanor crime of domestic violence.” 18 U. S. C. §922(g)(9). The respondent, James Alvin Castleman, pleaded guilty to the misdemeanor offense of having “intentionally or knowingly cause[d] bodily injury to” the mother of his child. App.27. The question before us is whether this conviction qualifies as “a misdemeanor crime of domestic violence.” We hold that it does. I A This country witnesses more than a million acts of domestic violence, and hundreds of deaths from domestic violence, each year.[1] See Georgia v. Randolph, 547 U. S. 103, 117–118 (2006). Domestic violence often escalates in severity over time, see Brief for Major Cities Chiefs Association et al. as Amici Curiae 13–15; Brief for National Network to End Domestic Violence et al. as Amici Curiae 9–12, and the presence of a firearm increases the likelihood that it will escalate to homicide, see id., at 14–15; Campbell et al., Assessing Risk Factors for Intimate Partner Homicide, DOJ, Nat. Institute of Justice J., No. 250, p. 16 (Nov. 2003) (“When a gun was in the house, an abused woman was 6 times more likely than other abused women to be killed”). “[A]ll too often,” as one Senator noted during the debate over §922(g)(9), “the only difference between a battered woman and a dead woman is the presence of a gun.” 142 Cong. Rec. 22986 (1996) (statement of Sen. Wellstone). Congress enacted §922(g)(9), in light of these sobering facts, to “ ‘close [a] dangerous loophole’ ” in the gun control laws: While felons had long been barred from possessing guns, many perpetrators of domestic violence are convicted only of misdemeanors. Hayes, 555 U. S., at 418, 426. Section 922(g)(9) provides, as relevant, that any person “who has been convicted . . . of a misdemeanor crime of domestic violence” may not “possess in or affecting commerc[e] any firearm or ammunition.” With exceptions that do not apply here, the statute defines a “misdemeanor crime of domestic violence” as “an offense that . . . (i) is a misdemeanor under Fed-eral, State, or Tribal law; and (ii) has, as an element, the use or attempted use of physical force, or the threatened use of a deadly weapon, committed by a current or former spouse, parent, or guardian of the victim, by a person with whom the victim shares a child in common, by a person who is cohabiting with or has cohabited with the victim as a spouse, parent, or guardian, or by a person similarly situated toa spouse, parent, or guardian of the victim.” §921(a)(33)(A). This case concerns the meaning of one phrase in this definition: “the use . . . of physical force.” B In 2001, Castleman was charged in a Tennessee court with having “intentionally or knowingly cause[d] bodily injury to” the mother of his child, in violation of Tenn. Code Ann. §39–13–111(b) (Supp. 2002). App. 27. He pleaded guilty. Id., at 29. In 2008, federal authorities learned that Castleman was selling firearms on the black market. A grand jury in the Western District of Tennessee indicted him on two counts of violating §922(g)(9) and on other charges not relevant here. Id., at 13–16. Castleman moved to dismiss the §922(g)(9) charges, arguing that his Tennessee conviction did not qualify as a “misdemeanor crime of domestic violence” because it did not “ha[ve], as an element, the use . . . of physical force,” §921(a)(33)(A)(ii). The District Court agreed, on the the-ory that “the ‘use of physical force’ for §922(g)(9) purposes” must entail “violent contact with the victim.” App. to Pet. for Cert. 40a. The court held that a conviction under the relevant Tennessee statute cannot qualify as a “misdemeanor crime of domestic violence” because one can cause bodily injury without “violent contact”—for example, by “deceiving [the victim] into drinking a poisoned beverage.” Id., at 41a. A divided panel of the U. S. Court of Appeals for the Sixth Circuit affirmed, by different reasoning. 695 F. 3d 582 (2012). The majority held that the degree of physical force required by §921(a)(33)(A)(ii) is the same as required by §924(e)(2)(B)(i), which defines “violent felony.” Id., at 587. Applying our decision in Johnson v. United States, 559 U. S. 133 (2010) , which held that §924(e)(2)(B)(i) re-quires “violent force,” id., at 140, the majority held that Castleman’s conviction did not qualify as a “misdemeanor crime of domestic violence” because Castleman could have been convicted for “caus[ing] a slight, nonserious physical injury with conduct that cannot be described as violent.” 695 F. 3d, at 590. Judge McKeague dissented, arguing both that the majority erred in extending Johnson’s definition of a “violent felony” to the context of a “misdemeanor crime of domestic violence” and that, in any event, Castle-man’s conviction satisfied the Johnson standard. Id., at 593–597. The Sixth Circuit’s decision deepened a split of author-ity among the Courts of Appeals. Compare, e.g., United States v. Nason, 269 F. 3d 10, 18 (CA1 2001) (§922(g)(9) “encompass[es] crimes characterized by the application of any physical force”), with United States v. Belless, 338 F. 3d 1063, 1068 (CA9 2003) (§922(g)(9) covers only “the violent use of force”). We granted certiorari to resolve this split, 570 U. S. ___ (2013), and now reverse the Sixth Cir-cuit’s judgment. II A “It is a settled principle of interpretation that, absent other indication, ‘Congress intends to incorporate the well-settled meaning of the common-law terms it uses.’ ” Sekhar v. United States, 570 U. S. ___, ___ (2013) (slip op., at 3). Seeing no “other indication” here, we hold that Congress incorporated the common-law meaning of “force”—namely, offensive touching—in §921(a)(33)(A)’s definition of a “mis-demeanor crime of domestic violence.” Johnson resolves this case in the Government’s favor—not, as the Sixth Circuit held, in Castleman’s. In Johnson, we considered whether a battery conviction was a “violent felony” under the Armed Career Criminal Act (ACCA), §924(e)(1). As here, ACCA defines such a crime as one that “has as an element the use . . . of physical force,” §924(e)(2)(B)(i). We began by observing that at common law, the element of force in the crime of battery was “satisfied by even the slightest offensive touching.” 559 U. S., at 139 (citing 3 W. Blackstone, Commentaries on the Laws of England 120 (1768)).[2] And we recognized the general rule that “a common-law term of art should be given its established common-law meaning,” except “where that meaning does not fit.” 559 U. S., at 139. We declined to read the common-law meaning of “force” into ACCA’s definition of a “violent felony,” because we found it a “comical misfit with the defined term.” Id., at 145; see United States v. Stevens, 559 U. S. 460, 474 (2010) (“[A]n unclear definitional phrase may take meaning from the term to be defined”). In defining a “ ‘violent felony,’ ” we held, “the phrase ‘physical force’ ” must “mea[n] violent force.” Johnson, 559 U. S., at 140. But here, the common-law meaning of “force” fits perfectly: The very reasons we gave for rejecting that meaning in defining a “violent felony” are reasons to embrace it in defining a “misdemeanor crime of domestic violence.”[3] First, because perpetrators of domestic violence are “routinely prosecuted under generally applicable assault or battery laws,” Hayes, 555 U. S., at 427, it makes sense for Congress to have classified as a “misdemeanor crime of domestic violence” the type of conduct that supports a common-law battery conviction. Whereas it was “unlikely” that Congress meant to incorporate in the definition of a “ ‘violent felony’ a phrase that the common law gave peculiar meaning only in its definition of a misdemeanor,” Johnson, 559 U. S., at 141, it is likely that Congress meant to incorporate that misdemeanor-specific meaning of “force” in defining a “misdemeanor crime of domestic violence.” Second, whereas the word “violent” or “violence” standing alone “connotes a substantial degree of force,” id., at 140,[4] that is not true of “domestic violence.” “Domestic violence” is not merely a type of “violence”; it is a term of art encompassing acts that one might not characterize as “violent” in a nondomestic context. See Brief for National Network to End Domestic Violence et al. as Amici Curiae 4–9; DOJ, Office on Violence Against Women, Domestic Violence (defining physical forms of domestic violence to include “[h]itting, slapping, shoving, grabbing, pinching, biting, [and] hair pulling”), online at http://www.ovw.usdoj.gov/domviolence.htm.[5] Indeed, “most physical assaults committed against women and men by intimates are relatively minor and consist of pushing, grabbing, shoving, slapping, and hitting.” DOJ, P. Tjaden & N. Thoennes, Extent, Nature and Consequences of Intimate Partner Violence 11 (2000). Minor uses of force may not constitute “violence” in the generic sense. For example, in an opinion that we cited with approval in Johnson, the Seventh Circuit noted that it was “hard to describe . . . as ‘violence’ ” “a squeeze of the arm [that] causes a bruise.” Flores v. Ashcroft, 350 F. 3d 666, 670 (2003). But an act of this nature is easy to describe as “domestic violence,” when the accumulation of such acts over time can subject one intimate partner to the other’s control. If a seemingly minor act like this draws the attention of authorities and leads to a successful prosecution for a misdemeanor offense, it does not offend common sense or the English language to characterize the resulting conviction as a “misdemeanor crime of domestic violence.” Justice Scalia’s concurrence discounts our reference to social-science definitions of “domestic violence,” including those used by the organizations most directly engaged with the problem and thus most aware of its dimensions. See post, at 8–11. It is important to keep in mind, how-ever, that the operative phrase we are construing is not “domestic violence”; it is “physical force.” §921(a)(33)(A). “Physical force” has a presumptive common-law meaning, and the question is simply whether that presumptive meaning makes sense in defining a “misdemeanor crime of domestic violence.”[6] A third reason for distinguishing Johnson’s definition of “physical force” is that unlike in Johnson—where a determination that the defendant’s crime was a “violent felony” would have classified him as an “armed career criminal”—the statute here groups those convicted of “misdemeanor crimes of domestic violence” with others whose conduct does not warrant such a designation. Section 922(g) bars gun possession by anyone “addicted to any controlled substance,” §922(g)(3); by most people who have “been admitted to the United States under a nonimmigrant visa,” §922(g)(5)(B); by anyone who has renounced United States citizenship, §922(g)(7); and by anyone subject to a domestic restraining order, §922(g)(8). Whereas we have hesitated (as in Johnson) to apply the Armed Career Criminal Act to “crimes which, though dangerous, are not typically committed by those whom one normally labels ‘armed career criminals,’ ” Begay v. United States, 553 U. S. 137, 146 (2008) , we see no anomaly in grouping domestic abusers convicted of generic assault or battery offenses together with the others whom §922(g) disqualifies from gun ownership. An additional reason to read the statute as we do is that a contrary reading would have rendered §922(g)(9) inoperative in many States at the time of its enactment. The “assault or battery laws” under which “domestic abusers were . . . routinely prosecuted” when Congress enacted §922(g)(9), and under which many are still prosecuted today, Hayes, 555 U. S., at 427, fall generally into two categories: those that prohibit both offensive touching and the causation of bodily injury, and those that prohibit only the latter. See Brief for United States 36–38. Whether or not the causation of bodily injury necessarily entails violent force—a question we do not reach—mere offensive touching does not. See Johnson, 559 U. S., at 139–140. So if offensive touching did not constitute “force” under §921(a)(33)(A), then §922(g)(9) would have been ineffec-tual in at least 10 States—home to nearly thirty percent of the Nation’s population[7]—at the time of its enactment. See post, at 6, and n. 5 (Scalia, J., concurring in part and concurring in judgment) (acknowledging that §922(g)(9) would have been inapplicable in California and nine other States if it did not encompass offensive touching); App. to Brief for United States 10a–16a (listing statutes prohibiting both offensive touching and the causation of bodily injury, only some of which are divisible); cf. Hayes, 555 U. S., at 427 (rejecting an interpretation under which “§922(g)(9) would have been ‘a dead letter’ in some two-thirds of the States from the very moment of its enactment”). In sum, Johnson requires that we attribute the common-law meaning of “force” to §921(a)(33)(A)’s definition of a “misdemeanor crime of domestic violence” as an offense that “has, as an element, the use or attempted useof physical force.” We therefore hold that the requirement of “physical force” is satisfied, for purposes of §922(g)(9), by the degree of force that supports a common-law battery conviction. B Applying this definition of “physical force,” we conclude that Castleman’s conviction qualifies as a “misdemeanor crime of domestic violence.” In doing so, we follow the analytic approach of Taylor v. United States, 495 U. S. 575 (1990) , and Shepard v. United States, 544 U. S. 13 (2005) . We begin with Taylor’s categorical approach, under which we look to the statute of Castleman’s conviction to determine whether that conviction necessarily “ha[d], as an element, the use or attempted use of physical force, or the threatened use of a deadly weapon,” §921(a)(33)(A). The Tennessee statute under which Castleman was convicted made it a crime to “commi[t] an assault . . . against” a “family or household member”—in Castleman’s case, the mother of his child. Tenn. Code Ann. §39–13–111(b). A provision incorporated by reference, §39–13–101, defined three types of assault: “(1) [i]ntentionally, knowingly or recklessly caus[ing] bodily injury to another; (2) [i]ntentionally or knowingly caus[ing] another to reasonably fear imminent bodily injury; or (3) [i]ntentionally or knowingly caus[ing] physical contact with another” in a manner that a “reasonable person would regard . . . as extremely offensive or provocative.” §39–13–101(a). It does not appear that every type of assault defined by §39–13–101 necessarily involves “the use or attempted use of physical force, or the threatened use of a deadly weapon,” §921(a)(33)(A). A threat under §39–13–101(2) maynot necessarily involve a deadly weapon, and the merely reckless causation of bodily injury under §39–13–101(1) may not be a “use” of force.[8] But we need not decide whether a domestic assault conviction in Tennessee categorically constitutes a “misdemeanor crime of domestic violence,” because the parties do not contest that §39–13–101 is a “ ‘divisible statute,’ ” Descamps v. United States, 570 U. S. ___, ___ (2013) (slip op., at 1). We may accordingly apply the modified categorical approach, consulting the indictment to which Castleman pleaded guilty in order to determine whether his conviction did entail the elements necessary to constitute the generic federal offense. Id., at ___ (slip op., at 1–2); see Shepard, 544 U. S., at 26. Here, that analysis is straightforward: Castleman pleaded guilty to having “intentionally or knowingly cause[d] bodily injury” to the mother of his child, App. 27, and the knowing or intentional causation of bodily injury necessarily involves the use of physical force. First, a “bodily injury” must result from “physical force.” Under Tennessee law, “bodily injury” is a broad term: It “includes a cut, abrasion, bruise, burn or disfigurement; physical pain or temporary illness or impairment of the function of a bodily member, organ, or mental faculty.” Tenn. Code Ann. §39–11–106(a)(2) (1997). Justice Sca-lia’s concurrence suggests that these forms of injury ne-cessitate violent force, under Johnson’s definition of that phrase. Post, at 3. But whether or not that is so—aquestion we do not decide—these forms of injury do necessitate force in the common-law sense. The District Court thought otherwise, reasoning that one can cause bodily injury “without the ‘use of physical force’ ”—for example, by “deceiving [the victim] into drinking a poisoned beverage, without making contact of any kind.” App. to Pet. for Cert. 41a. But as we explained in Johnson, “physical force” is simply “force exerted by and through concrete bodies,” as opposed to “intellectual force or emotional force.” 559 U. S., at 138. And the common-law concept of “force” encompasses even its indirect ap-plication. “Force” in this sense “describ[es] one of the elements of the common-law crime of battery,” id., at 139, and “[t]he force used” in battery “need not be applied directly to the body of the victim.” 2 W. LaFave, Substan-tive Criminal Law §16.2(b) (2d ed. 2003). “[A] battery may be committed by administering a poison or by infecting with a disease, or even by resort to some intangible substance,” such as a laser beam. Ibid. (footnote omitted) (citing State v. Monroe, 121 N. C. 677, 28 S. E. 547 (1897) (poison); State v. Lankford, 29 Del. 594, 102 A. 63 (1917) (disease); Adams v. Commonwealth, 33 Va. App. 463, 534 S. E. 2d 347 (2000) (laser beam)). It is impossible to cause bodily injury without applying force in the common-law sense. Second, the knowing or intentional application of force is a “use” of force. Castleman is correct that under Leocal v. Ashcroft, 543 U. S. 1 (2004) , the word “use” “conveys the idea that the thing used (here, ‘physical force’) has been made the user’s instrument.” Brief for Respondent 37. But he errs in arguing that although “[p]oison may have ‘forceful physical properties’ as a matter of organic chemistry, . . . no one would say that a poisoner ‘employs’ force or ‘carries out a purpose by means of force’ when he orshe sprinkles poison in a victim’s drink,” ibid. The “useof force” in Castleman’s example is not the act of “sprinkl[ing]” the poison; it is the act of employing poison knowingly as a device to cause physical harm. That the harm occurs indirectly, rather than directly (as with a kick or punch), does not matter. Under Castleman’s logic, after all, one could say that pulling the trigger on a gun is not a “use of force” because it is the bullet, not the trigger, that actually strikes the victim. Leocal held that the “use” of force must entail “a higher degree of intent than negligent or merely accidental conduct,” 543 U. S., at 9; it did not hold that the word “use” somehow alters the meaning of “force.” Because Castleman’s indictment makes clear that the use of physical force was an element of his conviction, that conviction qualifies as a “misdemeanor crime of domestic violence.” III We are not persuaded by Castleman’s nontextual arguments against our interpretation of §922(g)(9). A First, Castleman invokes §922(g)(9)’s legislative history to suggest that Congress could not have intended for the provision to apply to acts involving minimal force. But to the extent that legislative history can aid in the inter-pretation of this statute, Castleman’s reliance on it is unpersuasive. Castleman begins by observing that during the debate over §922(g)(9), several Senators argued that the provision would help to prevent gun violence by perpetrators of severe domestic abuse. Senator Lautenberg referred to “serious spousal or child abuse” and to “violent individuals”; Senator Hutchison to “ ‘people who batter their wives’ ”; Senator Wellstone to people who “brutalize” their wives or children; and Senator Feinstein to “severe and recurring domestic violence.” 142 Cong. Rec. 22985–22986, 22988. But as we noted above, see supra, at 2, the impetus of §922(g)(9) was that even perpetrators of severe domestic violence are often convicted “under generally applicable assault or battery laws.” Hayes, 555 U. S., at 427. So nothing about these Senators’ isolated references to severe domestic violence suggests that they would not have wanted §922(g)(9) to apply to a misdemeanor assault conviction like Castleman’s. Castleman next observes that §922(g)(9) is the product of a legislative compromise. The provision originally barred gun possession for any “crime of domestic violence,” defined as any “felony or misdemeanor crime of violence, regardless of length, term, or manner of punishment.” 142 Cong. Rec. 5840. Congress rewrote the provision to require the use of physical force in response to the concern “that the term crime of violence was too broad, and could be interpreted to include an act such as cutting up a credit card with a pair of scissors,” id., at 26675. See Hayes, 555 U. S., at 428. Castleman would have us conclude that Congress thus meant “to narrow the scope of the statute to convictions based on especially severe conduct.” Brief for Respondent 24. But all Congress meant to do was address the fear that §922(g)(9) might be triggered by offenses in which no force at all was directed at a person. As Senator Lautenberg noted, the revised text was not only “more precise” than the original but also “probably broader.” 142 Cong. Rec. 26675. B We are similarly unmoved by Castleman’s invocation of the rule of lenity. Castleman is correct that our “construction of a criminal statute must be guided by the need for fair warning.” Crandon v. United States, 494 U. S. 152, 160 (1990) . But “the rule of lenity only applies if, after considering text, structure, history, and purpose, there remains a grievous ambiguity or uncertainty in the statute, such that the Court must simply guess as to what Congress intended.” Barber v. Thomas, 560 U. S. 474, 488 (2010) (citation and internal quotation marks omitted). That is not the case here. C Finally, Castleman suggests—in a single paragraph—that we should read §922(g)(9) narrowly because it implicates his constitutional right to keep and bear arms. But Castleman has not challenged the constitutionality of §922(g)(9), either on its face or as applied to him, and the meaning of the statute is sufficiently clear that we need not indulge Castleman’s cursory nod to constitutional avoidance concerns. * * * Castleman’s conviction for having “intentionally or knowingly cause[d] bodily injury to” the mother of his child qualifies as a “misdemeanor crime of domestic violence.” The judgment of the United States Court of Appeals for the Sixth Circuit is therefore reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 See Dept. of Justice (DOJ), Bureau of Justice Statistics (BJS), J. Truman, L. Langton, & M. Planty, Criminal Victimization 2012 (Oct. 2013) (Table 1) (1,259,390 incidents of domestic violence in 2012), online at http://www.bjs.gov/content/pub/pdf/cv12.pdf (all Internet ma-terials as visited Mar. 19, 2014, and available in Clerk of Court’scase file); DOJ, BJS, C. Rennison, Crime Data Brief, Intimate Partner Violence, 1993–2001, p. 1 (Feb. 2003) (violence among intimate partners caused deaths of 1,247 women and 440 men in 2000), online at http://‌www.bjs.gov/content/pub/pdf/ipv01.pdf. 2 We explained that the word “physical” did not add much to the word “force,” except to distinguish “force exerted by and through concrete bodies . . . from, for example, intellectual force or emotional force.” , 559 U. S., at 138. 3 specifically reserved the question whether our definition of “physical force” would extend to . 559 U. S., at 143–144. And these reasons for declining to extend ’s definition to §922(g)(9) serve equally to rebut the “presumption of consistent usage” on which ’s concurrence heavily relies, at1–2, 4. 4 This portion of ’s analysis relied heavily on v. , , in which we interpreted the meaning of a “crime of violence” under . As in and here, the statute defines a “crime of violence” in part as one “that has as an element the use . . . of physical force,” §16(a). In support of our holding in , we quoted ’s observation that “ ‘[t]he ordinary meaning of [a “crime of violence”] . . . suggests a category of violent, active crimes.’ ” 559 U. S., at 140 (quoting 543 U. S., at 11). 5 See also A. Ganley, Understanding Domestic Violence, in Im-proving the Health Care Response to Domestic Violence: A Re-source Manual for Health Care Providers 18 (2d ed. 1996),online at http://www.‌futureswithoutviolence.org / userfiles / file/ HealthCare / improving_healthcare_manual_1.pdf (physical forms of domestic violence “may include spitting, scratching, biting, grabbing, shaking, shoving, pushing, restraining, throwing, twisting, [or] slapping”); M. McCue, Domestic Violence: A Reference Handbook 6 (1995) (noting that physical forms of domestic violence “may begin with relatively minor assaults such as painful pinching or squeezing”). 6 The concurrence’s reliance on definitions of “domestic violence” in other statutory provisions, see at 8, and n. 7, is similarly unpersuasive. These other provisions show that when Congress wished to define “domestic violence” as a type of “violence” , it knew how to do so. That it did not do so here suggests, if anything, that it did not mean to. See, v., . This also answers the concurrence’s suggestion, at 10, that our holding will somehow make it difficult for Congress to define “domestic violence”—where it wants to—as requiring violent force. 7 See U. S. Census Bureau, Time Series of Intercensal State Population Estimates: April 1, 1990 to April 1, 2000, online at http://www.‌census.gov/popest/data/intercensal/st-co/files/CO-EST2001-12-00.pdf (esti-mating state and national populations as of July 1, 1996). 8 We held in that “ ‘use’ requires active employment,” rather “than negligent or merely accidental conduct.” 543 U. S. at 9. Al-though reserved the question whether a reckless application of force could constitute a “use” of force, at 13, the Courts of Appeals have almost uniformly held that recklessness is not sufficient. See v. , 606 F. 3d 1317, 1335–1336 (CA11 2010);v. , 548 F. 3d 557, 560 (CA7 2008); v. , 527 F. 3d 1110, 1124 (CA10 2008); v. , 487 F. 3d 607, 615–616 (CA8 2007); v. , 469 F. 3d 496, 499 (CA6 2006);v. , 466 F. 3d 1121, 1127–1132 (CA9 2006) (en banc); v. , 455 F. 3d 465, 468–469 (CA4 2006); v. , 418 F. 3d 260, 263–265 (CA3 2005) (Alito, J.); v. , 326 F. 3d 367, 373 (CA2 2003); v. , 243 F. 3d 921, 926 (CA5 2001). But see v. , 644 F. 3d 12, 19–20 (CA1 2011) (noting that the First Circuit had not resolved the recklessness issue under , but declining to extend ’s analysis to §922(g)(9)).

572.US.141

Respondent Quality Stores, Inc., and its affiliates (collectively Quality Stores) made severance payments to employees who were involuntarily terminated as part of Quality Stores’ Chapter 11 bankruptcy. Payments—which were made pursuant to plans that did not tie payments to the receipt of state unemployment insurance—varied based on job seniority and time served. Quality Stores paid and withheld, inter alia, taxes required under the Federal Insurance Contributions Act (FICA), 26 U. S. C. §3101 et seq. Later believing that the payments should not have been taxed as wages under FICA, Quality Stores sought a refund on behalf of itself and about 1,850 former employees. When the Internal Revenue Service (IRS) did not allow or deny the refund, Quality Stores initiated proceedings in the Bankruptcy Court, which granted summary judgment in its favor. The District Court and Sixth Circuit affirmed, concluding that severance payments are not wages under FICA. Held: The severance payments at issue are taxable wages for FICA purposes. Pp. 4–15. (a) FICA defines “wages” broadly as “all remuneration for employment.” §3121(a). As a matter of plain meaning, severance payments fit this definition: They are a form of remuneration made only to employees in consideration for employment. “Employment” is “any service . . . performed . . . by an employee” for an employer. §3121(b). By varying according to a terminated employee’s function and seniority, the severance payments at issue confirm the principle that “service” “mea[ns] not only work actually done but the entire employer-employee relationship for which compensation is paid.” Social Security Bd. v. Nierotko, 327 U.S. 358, 365–366. This broad definition is reinforced by the specificity of FICA’s lengthy list of exemptions. The exemption for severance payments made “because of . . . retirement for disability,” §3121(a)(13)(A), would be unnecessary were severance payments generally not considered wages. FICA’s statutory history sheds further light on the definition. FICA originally contained definitions of “wages” and “employment” identical in substance to the current ones, but in 1939, Congress excepted from “wages” “[d]ismissal payments” not legally required by the employer, 53Stat. 1384. Since that exception was repealed in 1950, FICA has contained no general exception for severance payments. Pp. 4–7. (b) The Internal Revenue Code chapter governing income-tax withholding does not limit the meaning of “wages” for FICA purposes. Like FICA’s definitional section, §3401(a) has a broad definition of “wages” and contains a series of specific exemptions. Section 3402(o) instructs that “supplemental unemployment compensation benefits” or SUBs, which include severance payments, be treated “as if” they were wages. Contrary to Quality Stores’ reading, this “as if” instruction does not mean that severance payments fall outside the definition of “wages” for income-tax withholding purposes and, in turn, are not covered by FICA’s definition. Nor can Quality Stores rely on §3402(o)’s heading, which refers to “certain payments other than wages.” To the extent statutory headings are useful in resolving ambiguity, see FTC v. Mandel Brothers, Inc., 359 U.S. 385, 388–389, §3402(o)’s heading falls short of declaring that all the payments listed in §3402(o) are “other than wages.” Instead, §3402(o) must be understood in terms of the regulatory background against which it was enacted. In the 1950’s and 1960’s, because some States provided unemployment benefits only to terminated employees not earning wages, IRS Rulings took the position that severance payments tied to the receipt of state benefits were not wages. To address the problem that severance payments were still considered taxable income, which could lead to large year-end tax liability for terminated workers, Congress enacted §3402(o), which treats both SUBs and severance payments the IRS considered wages “as if” they were wages subject to withholding. By extending this treatment to all SUBs, Congress avoided the practical problems that might arise if the IRS later determined that SUBs besides severance payments linked to state benefits should be exempt from withholding. Considering this regulatory background, the assumption that Congress meant to exclude all SUBs from the definition of “wages” is unsustainable. That §3402(o) does not narrow FICA’s “wages” definition is also consistent with the major principle of Rowan Cos. v. United States, 452 U. S. 247: that simplicity of administration and consistency of statutory interpretation instruct that the meaning of “wages” should be in general the same for income-tax withholding and for FICA calculations. Pp. 7–14. 693 F.3d 605, reversed and remanded. Kennedy, J., delivered the opinion of the Court, in which all other Members joined, except Kagan, J., who took no part in the consideration or decision of the case.

This case presents the question whether severance pay-ments made to employees terminated against their will are taxable wages under the Federal Insurance Contri-butions Act (FICA), 26 U. S. C. §3101 et seq. The Court of Appeals for the Sixth Circuit held that the payments are not wages taxed by FICA. To reach its holding, the Court of Appeals relied not on FICA’s definition of wages but on §3402(o) of the Internal Revenue Code, a provision governing income-tax withholding. That conclusion, for the reasons to be discussed, was incorrect. FICA’s broad definition of wages includes the severance payments made here. And §3402(o) does not alter that definition. Section 3402(o) instructs that any severance payment “shall be treated as if it were a payment of wages.” According to the Court of Appeals, §3402(o) suggeststhat the definition of wages for income-tax withholding does not extend to severance payments; and so, the argument continues, severance payments also must be beyond the terms of FICA’s similar definition. But §3402(o) is entirely compatible with the proposition that some or all payments do fall within the broad definition of the term wages. Section 3402(o) was enacted in response to a narrow, specific problem regarding income-tax withholding. In addition, were the Court to rule that the severance payments made here are exempt from FICA taxation but not from withholding under §3402 for income-tax pur-poses, it would contravene the holding in Rowan Cos. v. United States, 452 U. S. 247 (1981) , which held there should be congruence in the rules for FICA and income-tax withholding. I Quality Stores, Inc., an agricultural-specialty retailer, entered bankruptcy proceedings in 2001. Before and following the filing of an involuntary Chapter 11 bankruptcy petition, respondents Quality Stores and affiliated companies, all referred to here as Quality Stores, terminated thousands of employees. The employees received severance payments, which all parties to this case stipulate were the result of a reduction in work force or discontinuance of a plant or operation. The payments were made pursuant to one of two different termination plans. (For reasons later to be explained, it should be noted that neither termination plan tied severance payments to the receipt of state unemployment compensation.) Under the first plan, terminated employees received severance pay based on job grade and management level. The president and chief executive officer received 18 months of severance pay, senior managers received 12 months of severance pay, and other employees received one week of severance pay for each year of service. The second plan was designed to facilitate Quality Stores’ postbankruptcy operations and encourage employees to put off their job searches. To receive severance pay, employees had to complete their last day of service as determined by the employer. Officers received between 6 and 12 months of severance pay, and full-time employees and employees paid by the hour received one week of severance pay for every year of service if the employees had been employed for at least two years, up to a stated maximum of severance pay. Workers who had been employed for less than two years received a week of severance pay. Quality Stores reported the severance payments as wages on W–2 tax forms, paid the employer’s required share of FICA taxes, and withheld employees’ share of FICA taxes. Then Quality Stores asked 3,100 former employees to allow it to file FICA tax refund claims for them. About 1,850 former employees agreed to allow Qual-ity Stores to pursue FICA refunds. On its own behalf and on behalf of the former employees, Quality Stores filed for a refund of $1,000,125 in FICA taxes. The Internal Revenue Service neither allowed nor denied the claim. Quality Stores initiated a proceeding in the Bankruptcy Court seeking a refund of the disputed amount. The Bankruptcy Court granted summary judgment in its favor. The District Court and Court of Appeals for the Sixth Circuit affirmed, concluding that severance payments are not “wages” under FICA. See In re Quality Stores, Inc., 693 F. 3d 605 (2012). Other Courts of Appeals, however, have concluded that at least some severance payments do constitute wages subject to FICA tax. See, e.g., CSX Corp. v. United States, 518 F. 3d 1328 (CA Fed. 2008); University of Pittsburgh v. United States, 507 F. 3d 165 (CA3 2007); North Dakota State Univ. v. United States, 255 F. 3d 599 (CA8 2001). The United States, claiming that the FICA taxes must be withheld, sought review here; and certiorari was granted, 570 U. S. ___ (2013). II A The first question is whether FICA’s definition of “wages” encompasses severance payments. The beginning pointis the relevant statutory text. Mississippi ex rel. Hoodv. AU Optronics Corp., 571 U. S. ___, ___ (2014) (slip op., at 5). To fund benefits provided by the Social Security Act and Medicare, FICA taxes “wages” paid by an employer or re-ceived by an employee “with respect to employment.” 26 U. S. C. §§3101(a), (b), 3111(a), (b). Congress chose to define wages under FICA “broadly.” Mayo Foundation for Medical Ed. and Research v. United States, 562 U. S. ___, ___ (2011) (slip op., at 2). FICA defines “wages” as “all remuneration for employment, including the cash value of all remuneration (including benefits) paid in any medium other than cash.” §3121(a). The term “employment” encompasses “any service, of whatever nature, performed . . . by an employee for the person employing him.” §3121(b). Under this definition, and as a matter of plain meaning, severance payments made to terminated employees are “remuneration for employment.” Severance payments are, of course, “remuneration,” and common sense dictates that employees receive the payments “for employment.” Severance payments are made to employees only. It wouldbe contrary to common usage to describe as a severance payment remuneration provided to someone who has not worked for the employer. Severance payments are made in consideration for employment—for a “service . . . performed” by “an employee for the person employing him,” per FICA’s definition of the term “employment.” Ibid. In Social Security Bd. v. Nierotko, 327 U. S. 358 (1946) , the Court interpreted the term “wages” in the Social Security statutory context to have substantial breadth. In that case a worker, who had been wrongfully terminated, sought to have his backpay counted as taxable wages for the purpose of obtaining credits under the Social Security system. The Court stated that the term “service,” used with respect to Social Security, “means not only work actually done but the entire employer-employee relationship for which compensation is paid to the employee by the employer.” Id., at 365–366. As confirmation of that principle, severance payments often vary, as they did here, according to the function and seniority of the particular employee who is terminated. For example, under both termination plans, Quality Stores employees were given severance payments based on job grade and management level. And under the second termination plan, nonofficer employees who had served at least two years with their company received more in severance pay than nonofficer employees who had not—a standard example of a company policy to reward employees for a greater length of good service and loyalty. In this respect severance payments are like many other benefits employers offer to employees above and beyond salary payments. Like health and retirement benefits, stock options, or merit-based bonuses, a competitive severance payment package can help attract talented employees. Here, the terminations leading to the severance payments were triggered by the employer’s involuntary bankruptcy proceeding, a prospect against which employees may wish to protect themselves in an economy that is always subject to changing conditions. Severance payments, moreover, can be desirable from the perspective of the employer as an alternative or supplemental form of remuneration. In situations in which Chapter 11 bankruptcy reorganization is necessary, an employer may seek to retain goodwill by paying its terminated employees well, thus reinforcing its reputation as a worthy employer. Employers who downsize in a period of slow business may wish to retain the ability to rehire employees who have been terminated. A specific exemption under FICA for certain termination-related payments reinforces the conclusion that the payments in question are well within the definition of wages. Section 3121(a)(13)(A) exempts from taxablewages any severance payments made “because of . . . retire-ment for disability.” That exemption would be unnecessary were severance payments in general not within FICA’s definition of “wages.” Cf. American Bank & Trust Co. v. Dallas County, 463 U. S. 855, 864 (1983) (declining to read a statute in a manner that would cause “spe-cific exemptions” to be “superfluous”). FICA’s definitional section, moreover, provides a lengthy list of specific exemptions from the definition of wages. For example, FICA exempts from wages payments on account of disability caused by sickness or accident, cash payments made for domestic service in a private home under a certain amount, and cash tips less than a certain amount. See §§3121(a)(2)(A), (7)(B), (12)(B). The specificity of these exemptions reinforces the broad nature of FICA’s definition of wages. FICA’s statutory history sheds further light on the text of §3121, which defines the term “wages.” FICA was originally enacted in Title VIII of the Social Security Act, 49Stat. 636. (In 1939, Title VIII was transferred to the Internal Revenue Code and became FICA. 53Stat. 1387.) Title VIII contained, in substance, definitions of “wages” and “employment” identical to those FICA now provides. See §811(a), 49Stat. 639; §811(b), ibid. With respect to the Social Security Act, in 1936 the Treasury Department promulgated a regulation stating that the statutory definition of “wages” included “dismissal pay.” Bureau of Internal Revenue, Employees’ Tax and the Employers’ Tax Under Title VIII of the Social Security Act, 1 Fed. Reg. 1764, 1769 (1936). Congress responded a few years later, in 1939, by creating an exception from “wages” for “[d]ismissal payments which the employer is not legally required to make.” Social Security Act Amendments of 1939, §606, 53Stat. 1384 (codified at 26 U. S. C. §1426(a)(4) (1940 ed.)). In 1950, however, Congress repealed that exception. Social Security Act Amendments, §203(a), 64Stat. 525–527. “When Congress acts to amend a statute, we presume it intends its amendment to have real and sub-stantial effect.” Stone v. INS, 514 U. S. 386, 397 (1995) . Congress has not revisited its 1950 amendment; and since that time, FICA has contained no exception for severance payments. B The next question is whether §3402(o) of the Internal Revenue Code relating to income-tax withholding is a limitation on the meaning of “wages” for FICA purposes. Section 3402 provides: “(o) Extension of withholding to certain payments other than wages. “(1) General rule “For purposes of this chapter (and so much of subtitle F as relates to this chapter)— “(A) any supplemental unemployment compensation benefit paid to an individual, . . . . . “shall be treated as if it were a payment of wages by an employer to an employee for a payroll period.” (Pursuant to stipulations by the parties, the Court of Appeals determined that the severance payments constitute “supplemental unemployment compensation benefits,” or SUBs. See §3402(o)(2)(A). The Court assumes, for purposes of this case, that this premise is correct.) Quality Stores argues that §3402(o)’s instruction that SUBs be treated “as if” they were wages for purposes of income-tax withholding is an indirect means of stating that the definition of wages for income-tax withholding does not cover severance payments. It contends, further, that if the definition of wages for purposes of income-tax withholding does not encompass severance payments, then severance payments are not covered by FICA’s similar definition of wages. The Court disagrees that §3402(o) should be read as Quality Stores suggests. The chapter governing income-tax withholding has a broad definition of the term “wages”: “all remuneration . . . for services performed by an employee for his employer, including the cash value of all remuneration (including benefits) paid in any medium other than cash.” §3401(a). The definitional section for income-tax withholding, like the definitional section for FICA, contains a series of specific exemptions that reinforce the broad scope of its definition of wages. The provision exempts from wages, for example, any remuneration paid for domestic service in a private home, for services rendered to a foreign government, and for services performed by a minister of a church in the course of his duties. §§3401(a)(3), (5), (9). Severance payments are not exempted, and they squarely fall within the broad textual definition of wages for purposes of income-tax withholding under §3401(a), for the same reasons outlined above with respect to FICA’s similar definition of wages. Quality Stores contends that, the broad wording of the definition in §3401(a) aside, severance payments must fall outside the definition of wages for income-tax withholding. Otherwise, it argues, §3402(o) would be superfluous. But, as the Government points out, §3402(o)’s command that all severance payments be treated “as if” they were wages for income-tax withholding is in all respects consistent with the proposition that at least some severance payments are wages. As the Federal Circuit explained when construing §3402(o), the statement that “all men shall be treated as if they were six feet tall does not imply that no men are six feet tall.” CSX Corp., 518 F. 3d, at 1342. In the last of its textual arguments, Quality Stores draws attention to the boldface heading of §3402(o), which states, “Extension of withholding to certain payments other than wages.” It contends the heading declares that the payments enumerated within §3402(o) are “other than wages.” Captions, of course, can be “a useful aid in resolving” a statutory text’s “ambiguity.” FTC v. Mandel Brothers, Inc., 359 U. S. 385 –389 (1959). But Quality Stores cannot rely on the statutory heading to support its argument that §3402(o), without ambiguity, excludes all severance payments from the definition of wages. The heading states that withholding is extended to “certain payments.” This falls short of a declaration that all the payments listed in §3402(o) are not wages. Next, the regulatory background against which §3402(o) was enacted illustrates the limited nature of the problem the provision was enacted to address. For this purpose, it is instructive to concentrate on the statutory term “supplemental unemployment benefits,” which defines the scope of §3402(o)’s income-tax withholding mandate. The concept of SUBs originated in labor demands for a guaranteed annual wage. When it became clear this was “impractical in their industries, unions such as the Steelworkers and the United Auto Workers transformed their guaranteed annual wage demands into proposals to supplement existing unemployment compensation programs.” Coffy v. Republic Steel Corp., 447 U. S. 191, 200 (1980) . A SUB plan, as originally conceived, offered “second-level protection against layoff” by supplementing unemployment benefits offered by the States. Ibid. In the 1950’s, major American employers such as Ford Motor Company adopted SUB plans of this type, agreeing to fund trusts that would provide SUBs to terminated employees. For example, Ford’s contract with employees defined the concept of SUBs as the receipt of “both a state system unemployment benefit and a Weekly Supplemental Benefit . . . without reduction of the state system unemployment benefit because of the payment of the Weekly Supplemental Benefit.” Note, Effect of Receiving Supplemental Unemployment Benefits on Eligibility for State Benefits, 69 Harv. L. Rev. 362, 364, n. 11 (1955); seeJ. Becker, Guaranteed Income for the Unemployed: The Story of SUB (1968). Employer plans that provided SUBs sought “to provide economic security for regular employees” and “to assure a stable work force through periods of short-term layoffs.” Coffy, supra, at 200. But an obstacle arose. For these plans to work, it was necessary to avoid having the SUBs defined under federal law as “wages.” That was because some States only provided unemployment benefits if terminated employees were not earning “wages” from their employers. See Brief for United States 29; CSX Corp., supra, at 1334–1335; Note, 69 Harv. L. Rev., at 366 (“The typical state unemployment compensation statute provides that ‘an individual shall be deemed unemployed in any week with respect to which no wages are payable to him and during which he performs no services . . .’ ” (ellipsis and emphasis in original)); id., at 367 (“[S]tates tend to treat as ‘wages’ those items which the federal government treats as ‘wages’ ”). The inability of terminated employees to receive state unemployment benefits, of course, would render SUBs far less useful to them and their employers. Employers, as a consequence, undertook to ensure that the Federal Government did not construe benefits paid out by SUB plans as “wages.” CSX Corp., supra, at 1334–1335. In at least partial response to the prospect of differential treatment of SUBs based on the vagaries of state law, the IRS promulgated a series of Revenue Rulings in the 1950’s and 1960’s that took the position that SUB payments were not “wages” under FICA as well as for purposes of income-tax withholding. Rev. Rul. 56–249, 1956–1 Cum. Bull. 488; see Rev. Rul. 58–128, 1958–1 Cum. Bull. 89; Rev. Rul. 60–330, 1960–2 Cum. Bull. 46; see also IRS Technical Advice Memorandum 9416003, 1993 WL 642695 (Apr. 22, 1994) (hereinafter TAM 9416003). Although the IRS exempted SUBs paid to terminated employees from withholding for income-tax purposes, the payments still were considered taxable income. Rev. Rul. 56–249, 1956–1 Cum. Bull. 488. As a result, terminated employees faced significant tax liability at the end ofthe year. The Treasury Department suggested Congress authorize the agency to promulgate regulations allowing voluntary withholding. Statements and Recommendations of the Department of the Treasury: Hearings on H. R. 13270 before the Senate Committee on Finance, 91st Cong., 1st Sess., 905–906 (1969). In 1969, Congress chose instead to address the withholding problem by enacting §3402(o). It provides that all severance payments—that is, both SUBs as well as severance payments that the IRS considered wages—shall be “treated as if” they were wages for purposes of income-tax withholding. It is apparent that the definition Congress adopted in §3402(o) is not limited to the SUBs that the IRS had deemed exempt from wages under FICA. See §3402(o)(2)(A). It must be presumed that Congress was aware that §3402(o) covered more than the severance payments that were excluded from income-tax withholding. Not all severance payment plans were tied to state unemployment benefits; and, before §3402(o)’s 1969 enactment, the IRS ruled that severance payments not linked to state unemployment benefits were wages for purposes of income-tax withholding. See Rev. Rul. 65–251, 1965–2 Cum. Bull. 395; see also TAM 9416003 (the IRS’ original 1956 exception for SUBs provided “a limited exception from the definition of wages for . . . federal income tax withholding . . . only if the payments are de-signed to supplement the receipt of state unemployment compensation and are actually tied to state unemployment benefits”); ibid. (“SUB-pay plans must be designed to supplement unemployment benefits . . .”). Once this background is understood, the Court of Appeals’ interpretation of §3402(o) as standing for some broad definitional principle is shown to be incorrect. Although Congress need not have agreed with the Revenue Rulings to enact §3402(o), its purpose to eliminatethe withholding problem caused by the differential treatment of severance payments is the necessary background to understand the meaning and purpose of the provision. The problem Congress sought to resolve was the prospect that terminated employees would owe large payments in taxes at the end of the year as a result of the IRS’ exemption of certain SUBs from withholding. It remained possible that the IRS would determine that other forms of SUB plans, perhaps linked differently to state unemployment benefits, should be exempt from withholding. If Congress had only incorporated the Revenue Rulings already in effect, that response may have risked the withholding problem arising once again. On the other hand, by drawing a withholding requirement that was broader than then-current IRS exemptions, Congress avoided these practical problems. A requirement that a form of remuneration already included as wages be treated “as if” it were wages created no administrative difficulties. The Court of Appeals understood Congress’ decision to include within §3402(o) a larger set of SUBs than was already exempt from withholding under IRS Revenue Rulings to mean that all SUBs were excluded from the definition of wages. But that assumption, although in the abstract not necessarily an illogical inference, is unsustainable, considering the regulatory background against which §3402(o) was enacted. Congress interpreted the Revenue Rulings not at all as a definitive gloss on the meaning of the term “wages” in §3401. The better reading is that Congress determined that, whatever position the IRS took with respect to certain categories of severance payments, the problem with withholding should be solved by treating all severance payments as wages requiring withholding. The necessary conclusion is that §3402(o) does not narrow the term “wages” under FICA to exempt all severance payments. This reasoning is consistent with Rowan, a previous decision interpreting FICA. In Rowan, the Court held that Treasury Regulations interpreting “wages”un-der FICA to include the value of meals and lodging were invalid. The Government conceded, for income-tax purposes, that the taxpayer in Rowan was correct to exempt the value of the meals and lodging in computing the wages properly withheld under §3402. 452 U. S., at 250–251. But it argued, nevertheless, that the value of the meals and lodging was taxable as wages under FICA, pursuant to Treasury Regulations. The Rowan Court observed that the definition of wages under FICA was in substance the same as for purposes of withholding. Id., at 255. The Court read that similarity to be “strong evidence that Congress intended ‘wages’ to mean the same thing under FICA . . . and income-tax withholding.” Ibid. To support that conclusion, the Court noted a “congressional concern for ‘the interest of simplicity and ease of administration.’ ” Ibid. (quoting S. Rep. No. 1631, 77th Cong., 2d Sess., 165 (1942)). Because “Congress intended . . . to coordinate the income-tax withholding system with FICA” in order “to promote simplicity and ease of administration,” the Court held that it would be “extraordinary” for Congress to intend the definitions of “wages” to vary between FICA and income-tax withholding. 452 U. S., at 257. The specific holding of Rowan—that regulations governing meals and lodging were invalid—has little or nobearing on the issue confronting us here. What is of im-portance is the major principle recognized in Rowan: that simplicity of administration and consistency of statutory interpretation instruct that the meaning of “wages” should be in general the same for income-tax withholding and for FICA calculations. Quality Stores contends that, under the mandate of §3402(o), severance payments are not subject to FICA taxation but are to be deemed wages for purposes ofincome-tax withholding. It justifies this differential treat-ment in the name of uniformity. But that so-called uniformity as to the definitions of wages (i.e., that severance payments are not wages) is not consistent with the broad textual definitions of wages under FICA and income-tax withholding. Nor is it consistent with this Court’s holding that administrative reasons justify treating severance payments as taxable for both FICA and income-tax purposes. To read Congress’ command to withhold severance payments as an implicit overruling of the broad definition of wages in FICA would disserve the statutory text and the congressional interest in administrative simplicity deemed controlling in Rowan. In concluding, the Court notes that the IRS still provides that severance payments tied to the receipt of state unemployment benefits are exempt not only from income-tax withholding but also from FICA taxation. See, e.g., Rev. Rul. 90–72, 1990–2 Cum. Bull. 211. Those Revenue Rulings are not at issue here. Because the severance payments here were not linked to state unemployment benefits, the Court does not reach the question whether the IRS’ current exemption is consistent with the broad definition of wages under FICA. * * * The severance payments here were made to employees terminated against their will, were varied based on job seniority and time served, and were not linked to the receipt of state unemployment benefits. Under FICA’s broad definition, these severance payments constitute taxable wages. The judgment of the Court of Appeals for the Sixth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Kagan took no part in the consideration or decision of this case.

571.US.31

Respondent Gary Woods and his employer, Billy Joe McCombs, participated in an offsetting-option tax shelter designed to generate large paper losses that they could use to reduce their taxable income. To that end, they purchased from Deutsche Bank a series of currency-option spreads. Each spread was a package consisting of a long option, which Woods and McCombs purchased from Deutsche Bank and for which they paid a premium, and a short option, which Woods and McCombs sold to Deutsche Bank and for which they received a premium. Because the premium paid for the long option was largely offset by the premium received for the short option, the net cost of the package to Woods and McCombs was substantially less than the cost of the long option alone. Woods and McCombs contributed the spreads, along with cash, to two partnerships, which used the cash to purchase stock and currency. When calculating their basis in the partnership interests, Woods and McCombs considered only the long component of the spreads and disregarded the nearly offsetting short component. As a result, when the partnerships’ assets were disposed of for modest gains, Woods and McCombs claimed huge losses. Al-though they had contributed roughly $3.2 million in cash and spreads to the partnerships, they claimed losses of more than $45 million. The Internal Revenue Service sent each partnership a Notice of Final Partnership Administrative Adjustment, disregarding the partnerships for tax purposes and disallowing the related losses. It concluded that the partnerships were formed for the purpose of tax avoidance and thus lacked “economic substance,” i.e., they were shams. As there were no valid partnerships for tax purposes, the IRS determined that the partners could not claim a basis for their partnership interests greater than zero and that any resulting tax underpayments would be subject to a 40-percent penalty for gross valuation misstatements. Woods sought judicial review. The District Court held that the partnerships were properly disregarded as shams but that the valuation-misstatement penalty did not apply. The Fifth Circuit affirmed. Held: 1. The District Court had jurisdiction to determine whether the partnerships’ lack of economic substance could justify imposing a valuation-misstatement penalty on the partners. Pp. 6–11. (a) Because a partnership does not pay federal income taxes, its taxable income and losses pass through to the partners. Under the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA), the IRS initiates partnership-related tax proceedings at the partnership level to adjust “partnership items,” i.e., items relevant to the partnership as a whole. 26 U. S. C. §§6221, 6231(a)(3). Once the adjustments become final, the IRS may undertake further proceedings at the partner level to make any resulting “computational adjustments” in the tax liability of the individual partners. §§6230(a)(1)–(2), (c), 6231(a)(6). Pp. 6–7. (b) Under TEFRA’s framework, a court in a partnership-level proceeding has jurisdiction to determine “the applicability of any penalty . . . which relates to an adjustment to a partnership item.” §6226(f). A determination that a partnership lacks economic substance is such an adjustment. TEFRA authorizes courts in partnership-level proceedings to provisionally determine the applicability of any penalty that could result from an adjustment to a partnership item, even though imposing the penalty requires a subsequent, partner-level proceeding. In that later proceeding, each partner may raise any reasons why the penalty may not be imposed on him specifically. Applying those principles here, the District Court had jurisdiction to determine the applicability of the valuation-misstatement penalty. Pp. 7–11. 2. The valuation-misstatement penalty applies in this case. Pp. 11–16. (a) A penalty applies to the portion of any underpayment that is “attributable to” a “substantial” or “gross” “valuation misstatement,” which exists where “the value of any property (or the adjusted basis of any property) claimed on any return of tax” exceeds by a specified percentage “the amount determined to be the correct amount of such valuation or adjusted basis (as the case may be).” §§6662(a), (b)(3), (e)(1)(A), (h). The penalty’s plain language makes it applicable here. Once the partnerships were deemed not to exist for tax purposes, no partner could legitimately claim a basis in his partnership interest greater than zero. Any underpayment resulting from use of a non-zero basis would therefore be “attributable to” the partner’s having claimed an “adjusted basis” in the partnerships that exceeded “the correct amount of such . . . adjusted basis.” §6662(e)(1)(A). And under the relevant Treasury Regulation, when an asset’s adjusted basis is zero, a valuation misstatement is automatically deemed gross. Pp. 11–12. (b) Woods’ contrary arguments are unpersuasive. The valuation-misstatement penalty encompasses misstatements that rest on legal as well as factual errors, so it is applicable to misstatements that rest on the use of a sham partnership. And the partnerships’ lack of economic substance is not an independent ground separate from the misstatement of basis in this case. Pp. 12–16. 471 Fed. Appx. 320, reversed. Scalia, J., delivered the opinion for a unanimous Court.

We decide whether the penalty for tax underpayments attributable to valuation misstatements, 26 U. S. C. §6662(b)(3), is applicable to an underpayment resulting from a basis-inflating transaction subsequently disregarded for lack of economic substance. I. The Facts A This case involves an offsetting-option tax shelter, variants of which were marketed to high-income taxpayers in the late 1990’s. Tax shelters of this type sought to generate large paper losses that a taxpayer could use to reduce taxable income. They did so by attempting to give the tax-payer an artificially high basis in a partnership interest, which enabled the taxpayer to claim a significant tax loss upon disposition of the interest. See IRS Notice 2000–44, 2000–2 Cum. Bull. 255 (describing offsetting-option tax shelters). The particular tax shelter at issue in this case was developed by the now-defunct law firm Jenkens & Gilchrist and marketed by the accounting firm Ernst & Young under the name “Current Options Bring Reward Alternatives,” or COBRA. Respondent Gary Woods and his employer, Billy Joe McCombs, agreed to participate in COBRA to reduce their tax liability for 1999. To that end, in November 1999 they created two general partnerships: one, Tesoro Drive Partners, to produce ordinary losses, and the other, SA Tesoro Investment Partners, to produce capital losses. Over the next two months, acting through their respective wholly owned, limited liability companies, Woods and McCombs executed a series of transactions. First, they purchased from Deutsche Bank five 30-day currency-option spreads. Each of these option spreads was a package consisting of a so-called long option, which entitled Woods and McCombs to receive a sum of money from Deutsche Bank if a certain currency exchange rate exceeded a certain figure on a certain date, and a so-called short option, which entitled Deutsche Bank to receive a sum of money from Woods and McCombs if the exchange rate for the same currency on the same date exceeded a certain figure so close to the figure triggering the long option that both were likely to be triggered (or not to be triggered) on the fated date. Because the premium paid to Deutsche Bank for purchase of the long option was largely offset by the premium received from Deutsche Bank for sale of the short option, the net cost of the package to Woods and McCombs was substantially less than the cost of the long option alone. Specifically, the premiums paid for all five of the spreads’ long options totaled $46 million, and the premiums received for the five spreads’ short options totaled $43.7 million, so the net cost of the spreads was just $2.3 million. Woods and McCombs contributed the spreads to the partnerships along with about $900,000 in cash. The partnerships used the cash to purchase assets—Canadian dollars for the partnership that sought to produce ordinary losses, and Sun Microsystems stock for the partnership that sought to produce capital losses. The partnerships then terminated the five option spreads in exchange for a lump-sum payment from Deutsche Bank. As the tax year drew to a close, Woods and McCombs transferred their interests in the partnerships to two S corporations. One corporation, Tesoro Drive Investors, Inc., received both partners’ interests in Tesoro Drive Partners; the other corporation, SA Tesoro Drive Investors, Inc., received both partners’ interests in SA Tesoro Investment Partners. Since this left each partnership with only a single partner (the relevant S corporation), the partnerships were liquidated by operation of law, and their assets—the Canadian dollars and Sun Microsystems stock, plus the remaining cash—were deemed distributed to the corporations. The corporations then sold those assets for modest gains of about $2,000 on the Canadian dollars and about $57,000 on the stock. But instead of gains, the corporations reported huge losses: an ordinary loss of more than $13 million on the sale of the Canadian dollars and a capital loss of more than $32 million on the sale of the stock. The losses were allocated between Woods and McCombs as the corporations’ co-owners. The reason the corporations were able to claim such vast losses—the alchemy at the heart of an offsetting-options tax shelter—lay in how Woods and McCombs calculated the tax basis of their interests in the partnerships. Tax basis is the amount used as the cost of an asset when computing how much its owner gained or lost for tax purposes when disposing of it. See J. Downes & J. Goodman, Dictionary of Finance and Investment Terms 736 (2010). A partner’s tax basis in a partnership interest—called “outside basis” to distinguish it from “inside basis,” the partnership’s basis in its own assets—is tied to the value of any assets the partner contributed to acquire the interest. See 26 U. S. C. §722. Collectively, Woods and McCombs contributed roughly $3.2 million in option spreads and cash to acquire their interests in the two partnerships. But for purposes of computing outside basis, Woods and McCombs considered only the long component of the spreads and disregarded the nearly offset-ting short component on the theory that it was “too contingent” to count. Brief for Respondent 14. As a result, they claimed a total adjusted outside basis of more than $48 million. Since the basis of property distributed to a partner by a liquidating partnership is equal to the adjusted basis of the partner’s interest in the partnership (reduced by any cash distributed with the property), see §732(b), the inflated outside basis figure was carried over to the S corporations’ basis in the Canadian dollars and the stock, enabling the corporations to report enormous losses when those assets were sold. At the end of the day, Woods’ and McCombs’ $3.2 million investment generated tax losses that, if treated as valid, could have shielded more than $45 million of income from taxation. B The Internal Revenue Service, however, did not treat the COBRA-generated losses as valid. Instead, after auditing the partnerships’ tax returns, it issued to each partnership a Notice of Final Partnership Administrative Adjustment, or “FPAA.” In the FPAAs, the IRS determined that the partnerships had been “formed and availed of solely for purposes of tax avoidance by artificially overstating basis in the partnership interests of [the] purported partners.” App. 92, 146. Because the partnerships had “no business purpose other than tax avoidance,” the IRS said, they “lacked economic substance”—or, put more starkly, they were “sham[s]”—so the IRS would disregard them for tax purposes and disallow the related losses. Ibid. And because there were no valid partnerships for tax purposes, the IRS determined that the partners had “not established adjusted bases in their respective partnership interests in an amount greater than zero,” id., at 95, ¶7, 149, ¶7 so that any resulting tax underpayments would be subject to a 40-percent penalty for gross valuation misstatements, see 26 U. S. C. §6662(b)(3). Woods, as the tax-matters partner for both partnerships, sought judicial review of the FPAAs pursuant to §6226(a). The District Court held that the partner- ships were properly disregarded as shams but that the valuation-misstatement penalty did not apply. The Govern-ment appealed the decision on the penalty to the Court of Appeals for the Fifth Circuit. While the appeal was pending, the Fifth Circuit held in a similar case that, under Circuit precedent, the valuation-misstatement penalty does not apply when the relevant transaction is disregarded for lacking economic substance. Bemont Invs., LLC v. United States, 679 F. 3d 339, 347–348 (2012). In a concurrence joined by the other members of the panel, Judge Prado acknowledged that this rule was binding Circuit law but suggested that it was mistaken. See id., at 351–355. A different panel subsequently affirmed the District Court’s decision in this case in a one-paragraph opinion, declaring the issue “well settled.” 471 Fed. Appx. 320 (per curiam), reh’g denied (2012). [ 1 ] We granted certiorari to resolve a Circuit split over whether the valuation-misstatement penalty is applicable in these circumstances. 569 U. S. ___ (2013). See Bemont, supra, at 354–355 (Prado, J., concurring) (recognizing “near-unanimous opposition” to the Fifth Circuit’s rule). Because two Courts of Appeals have held that District Courts lacked jurisdiction to consider the valuation-misstatement penalty in similar circumstances, see Jade Trading, LLC v. United States, 598 F. 3d 1372, 1380 (CA Fed. 2010); Petaluma FX Partners, LLC v. Commissioner, 591 F. 3d 649, 655–656 (CADC 2010), we ordered briefing on that question as well. II. District-Court Jurisdiction A We begin with a brief explanation of the statutory scheme for dealing with partnership-related tax matters. A partnership does not pay federal income taxes; instead, its taxable income and losses pass through to the partners. 26 U. S. C. §701. A partnership must report its tax items on an information return, §6031(a), and the partners must report their distributive shares of the partnership’s tax items on their own individual returns, §§702, 704. Before 1982, the IRS had no way of correcting errors on a partnership’s return in a single, unified proceeding. Instead, tax matters pertaining to all the members of a partnership were dealt with just like tax matters pertaining only to a single taxpayer: through deficiency proceedings at the individual-taxpayer level. See generally §§6211–6216 (2006 ed. and Supp. V). Deficiency proceedings require the IRS to issue a separate notice of deficiency to each taxpayer, §6212(a) (2006 ed.), who can file a petition in the Tax Court disputing the alleged deficiency before paying it, §6213(a). Having to use deficiency proceedings for partnership-related tax matters led to du-plicative proceedings and the potential for inconsistent treatment of partners in the same partnership. Congress addressed those difficulties by enacting the Tax Treatment of Partnership Items Act of 1982, as Title IV of the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA). 96Stat. 648 (codified as amended at 26 U. S. C. §§6221–6232 (2006 ed. and Supp. V)). Under TEFRA, partnership-related tax matters are addressed in two stages. First, the IRS must initiate proceedings at the partnership level to adjust “partnership items,” those relevant to the partnership as a whole. §§6221, 6231(a)(3). It must issue an FPAA notifying the partners of any adjustments to partnership items, §6223(a)(2), and the partners may seek judicial review of those adjustments, §6226(a)–(b). Once the adjustments to partnership items have become final, the IRS may undertake further proceedings at the partner level to make any resulting “computational adjustments” in the tax liability of the individual partners. §6231(a)(6). Most computational adjustments may be directly assessed against the partners, bypassing deficiency proceedings and permitting the partners to challenge the assessments only in post-payment refund actions. §6230(a)(1), (c). Deficiency proceedings are still required, however, for certain com-putational adjustments that are attributable to “affected items,” that is, items that are affected by (but are not themselves) partnership items. §§6230(a)(2)(A)(i), 6231(a)(5). B Under the TEFRA framework, a court in a partnership-level proceeding like this one has jurisdiction to determine not just partnership items, but also “the applicability of any penalty . . . which relates to an adjustment to a partnership item.” §6226(f). As both sides agree, a determination that a partnership lacks economic substance is an adjustment to a partnership item. Thus, the jurisdictional question here boils down to whether the valuation-misstatement penalty “relates to” the determination that the partnerships Woods and McCombs created were shams. The Government’s theory of why the penalty was triggered is based on a straightforward relationship between the economic-substance determination and the penalty. In the Government’s view, there can be no outside basis in a sham partnership (which, for tax purposes, does not exist), so any partner who underpaid his individual taxes by declaring an outside basis greater than zero committed a valuation misstatement. In other words, the penalty flows logically and inevitably from the economic-substance determination. Woods, however, argues that because outside basis is not a partnership item, but an affected item, a penalty that would rest on a misstatement of outside basis cannot be considered at the partnership level. He maintains, in short, that a penalty does not relate to a partnership-item adjustment if it “requires a partner-level determination,” regardless of “whether or not the penalty has a connection to a partnership item.” Brief for Respondent 27. Because §6226(f)’s “relates to” language is “essentially indeterminate,” we must resolve this dispute by looking to “the structure of [TEFRA] and its other provisions.” Maracich v. Spears, 570 U. S. ___, ___ (2013) (slip op., at 9) (internal quotation marks and brackets omitted). That inquiry makes clear that the District Court’s jurisdiction is not as narrow as Woods contends. Prohibiting courts in partnership-level proceedings from considering the applicability of penalties that require partner-level inquiries would be inconsistent with the nature of the “applicability” determination that TEFRA requires. Under TEFRA’s two-stage structure, penalties for tax underpayment must be imposed at the partner level, because partnerships themselves pay no taxes. And imposing a penalty always requires some determinations that can be made only at the partner level. Even where a partnership’s return contains significant errors, a partner may not have carried over those errors to his own return; or if he did, the errors may not have caused him to underpay his taxes by a large enough amount to trigger the penalty; or if they did, the partner may nonetheless have acted in good faith with reasonable cause, which is a bar to the imposition of many penalties, see §6664(c)(1). None of those issues can be conclusively determined at the partnership level. Yet notwithstanding that every pen-alty must be imposed in partner-level proceedings after partner-level determinations, TEFRA provides that the applicability of some penalties must be determined at the partnership level. The applicability determination is therefore inherently provisional; it is always contingent upon determinations that the court in a partnership-level proceeding does not have jurisdiction to make. Barring partnership-level courts from considering the applicability of penalties that cannot be imposed without partner-level inquiries would render TEFRA’s authorization to consider some penalties at the partnership level meaningless. Other provisions of TEFRA confirm that conclusion. One requires the IRS to use deficiency proceedings for computational adjustments that rest on “affected items which require partner level determinations (other than penalties . . . that relate to adjustments to partnership items).” §6230(a)(2)(A)(i). Another states that while a partnership-level determination “concerning the applicability of any penalty . . . which relates to an adjustment to a partnership item” is “conclusive” in a subsequent re-fund action, that does not prevent the partner from “assert[ing] any partner level defenses that may apply.” §6230(c)(4). Both these provisions assume that a penalty can relate to a partnership-item adjustment even if the penalty cannot be imposed without additional, partner-level determinations. These considerations lead us to reject Woods’ interpretation of §6226(f). We hold that TEFRA gives courts in partnership-level proceedings jurisdiction to determine the applicability of any penalty that could result from an adjustment to a partnership item, even if imposing the penalty would also require determining affected or non-partnership items such as outside basis. The partnership-level applicability determination, we stress, is provisional: the court may decide only whether adjustments properly made at the partnership level have the potential to trigger the penalty. Each partner remains free to raise, in subsequent, partner-level proceedings, any reasons why the penalty may not be imposed on him specifically. Applying the foregoing principles to this case, we conclude that the District Court had jurisdiction to determine the applicability of the valuation-misstatement penalty—to determine, that is, whether the partnerships’ lack of economic substance (which all agree was properly decided at the partnership level) could justify imposing a valuation-misstatement penalty on the partners. When making that determination, the District Court was obliged to consider Woods’ arguments that the economic-substance determination was categorically incapable of triggering the penalty. Deferring consideration of those arguments until partner-level proceedings would replicate the precise evil that TEFRA sets out to remedy: duplicative proceedings, potentially leading to inconsistent results, on a question that applies equally to all of the partners. To be sure, the District Court could not make a formal ad-justment of any partner’s outside basis in this partnership-level proceeding. See Petaluma, 591 F. 3d, at 655. But it nonetheless could determine whether the adjustments it did make, including the economic-substance deter-mination, had the potential to trigger a penalty; and in doing so, it was not required to shut its eyes to the legal impossibility of any partner’s possessing an outside basis greater than zero in a partnership that, for tax purposes, did not exist. Each partner’s outside basis still must be adjusted at the partner level before the penalty can be imposed, but that poses no obstacle to a partnership-level court’s provisional consideration of whether the economic-substance determination is legally capable of triggering the penalty. [ 2 ] III. Applicability of Valuation-Misstatement Penalty A Taxpayers who underpay their taxes due to a “valuation misstatement” may incur an accuracy-related penalty. A 20-percent penalty applies to “the portion of any underpayment which is attributable to . . . [a]ny substantial valuation misstatement under chapter 1.” 26 U. S. C. §6662(a), (b)(3). Under the version of the penalty statute in effect when the transactions at issue here occurred, “there is a substantial valuation misstatement under chapter 1 if . . . the value of any property (or the adjusted basis of any property) claimed on any return of tax imposed by chapter 1 is 200 percent or more of the amount determined to be the correct amount of such valuation or adjusted basis (as the case may be).” §6662(e)(1)(A) (2000 ed.). If the reported value or adjusted basis exceeds the correct amount by at least 400 percent, the valuation misstatement is considered not merely substantial, but “gross,” and the penalty increases to 40 percent. §6662(h). [ 3 ] The penalty’s plain language makes it applicable here. As we have explained, the COBRA transactions were designed to generate losses by enabling the partners to claim a high outside basis in the partnerships. But once the partnerships were deemed not to exist for tax purposes, no partner could legitimately claim an outside basis greater than zero. Accordingly, if a partner used an outside basis figure greater than zero to claim losses on his tax return, and if deducting those losses caused the partner to underpay his taxes, then the resulting underpayment would be “attributable to” the partner’s having claimed an “adjusted basis” in the partnerships that exceeded “the correct amount of such . . . adjusted basis.” §6662(e)(1)(A). An IRS regulation provides that when an asset’s true value or adjusted basis is zero, “[t]he value or adjusted basis claimed . . . is considered to be 400 percent or more of the correct amount,” so that the resulting valuation misstatement is automatically deemed gross and subject to the 40-percent penalty. Treas. Reg. §1.6662–5(g), 26 CFR §1.6662–5(g) (2013). [ 4 ] B Against this straightforward application of the statute, Woods’ primary argument is that the economic-substance determination did not result in a “valuation misstatement.” He asserts that the statutory terms “value” and “valuation” connote “a factual—rather than legal—concept,” and that the penalty therefore applies only to factual misrepresentations about an asset’s worth or cost, not to misrepresentations that rest on legal errors (like the use of a sham partnership). Brief for Respondent 35. We are not convinced. To begin, we doubt that “value” is limited to factual issues and excludes threshold legal determinations. Cf. Powers v. Commissioner, 312 U. S. 259, 260 (1941) (“[W]hat criterion should be employed for determining the ‘value’ of the gifts is a question of law”); Chapman Glen Ltd. v. Commissioner, 140 T. C. No. 15, 2013 WL2319282, *17 (2013) (“[T]hree approaches are used to determine the fair market value of property,” and “which approach to apply in a case is a question of law”). But even if “value” were limited to factual matters, the statute refers to “value” or “adjusted basis,” and there is no justification for extending that limitation to the latter term, which plainly incorporates legal inquiries. An asset’s “basis” is simply its cost, 26 U. S. C. §1012(a) (2006 ed., Supp. V), but calculating its “adjusted basis” requires the application of a host of legal rules, see §§1011(a) (2006 ed.), 1016 (2006 ed. and Supp. V), including specialized rules for calculating the adjusted basis of a partner’s interest in a partnership, see §705 (2006 ed.). The statute contains no indication that the misapplication of one of those legal rules cannot trigger the penalty. Were we to hold otherwise, we would read the word “adjusted” out of the statute. To overcome the plain meaning of “adjusted basis,” Woods asks us to interpret the parentheses in the statutory phrase “the value of any property (or the adjusted basis of any property)” as a signal that “adjusted basis” is merely explanatory or illustrative and has no meaning inde-pendent of “value.” The parentheses cannot bear that much weight, given the compelling textual evidence to the contrary. For one thing, the terms reappear later in the same sentence sans parentheses—in the phrase “such valuation or adjusted basis.” Moreover, the operative terms are connected by the conjunction “or.” While that can sometimes introduce an appositive—a word or phrase that is synonymous with what precedes it (“Vienna or Wien,” “Batman or the Caped Crusader”)—its ordinary use is almost always disjunctive, that is, the words it connects are to “be given separate meanings.” Reiter v. Sonotone Corp., 442 U. S. 330, 339 (1979) . And, of course, there is no way that “adjusted basis” could be regarded as synonymous with “value.” Finally, the terms’ second disjunctive appearance is followed by “as the case may be,” which eliminates any lingering doubt that the preceding items are alternatives. See New Oxford American Dictionary 269 (3d ed. 2010). The parentheses thus do not justify “rob[bing] the term [‘adjusted basis’] of its independent and ordinary significance.” Reiter, supra, at 338–339. Our holding that the valuation-misstatement penalty encompasses legal as well as factual misstatements of adjusted basis does not make superfluous the new penalty that Congress enacted in 2010 for transactions lacking in economic substance, see §1409(b)(2), 124Stat. 1068–1069 (codified at 26 U. S. C. §6662(b)(6) (2006 ed., Supp. V)). The new penalty covers all sham transactions, including those that do not cause the taxpayer to misrepresent value or basis; thus, it can apply in situations where the valuation-misstatement penalty cannot. And the fact that both penalties are potentially applicable to sham transactions resulting in valuation misstatements is not problematic. Congress recognized that penalties might overlap in a given case, and it addressed that possibility by providing that a taxpayer generally cannot receive more than one accuracy-related penalty for the same underpayment. See §6662(b) (2006 ed. and Supp. V). [ 5 ] C In the alternative, Woods argues that any underpayment of tax in this case would be “attributable,” not to the misstatements of outside basis, but rather to the deter-mination that the partnerships were shams—which he describes as an “independent legal ground.” Brief for Respondent 46. That is the rationale that the Fifth and Ninth Circuits have adopted for refusing to apply the valuation-misstatement penalty in cases like this, although both courts have voiced doubts about it. See Bemont, 679 F. 3d, at 347–348; id., at 351–355 (Prado, J., concurring); Keller v. Commissioner, 556 F. 3d 1056, 1060–1061 (CA9 2009). We reject the argument’s premise: The economic-substance determination and the basis misstatement are not “independent” of one another. This is not a case where a valuation misstatement is a mere side effect of a sham transaction. Rather, the overstatement of outside basis was the linchpin of the COBRA tax shelter and the mechanism by which Woods and McCombs sought to reduce their taxable income. As Judge Prado observed, in this type of tax shelter, “the basis misstatement and the transaction’s lack of economic substance are inextricably intertwined,” so “attributing the tax underpayment only to the artificiality of the transaction and not to the basis overvaluation is making a false distinction.” Bemont, supra, at 354 (concurring opinion). In short, the partners underpaid their taxes because they overstated their outside basis, and they overstated their outside basis because the partnerships were shams. We therefore have no difficulty concluding that any underpayment resulting from the COBRA tax shelter is attributable to the partners’ misrepresentation of outside basis (a valuation misstatement). Woods contends, however, that a document known as the “Blue Book” compels a different result. See General Explanation of the Economic Recovery Tax Act of 1981 (Pub. L. 97–34), 97 Cong., 1st Sess., 333, and n. 2 (Jt. Comm. Print 1980). Blue Books are prepared by the staff of the Joint Committee on Taxation as commentaries on recently passed tax laws. They are “written after passage of the legislation and therefore d[o] not inform the decisions of the members of Congress who vot[e] in favor of the [law].” Flood v. United States, 33 F. 3d 1174, 1178 (CA9 1994). We have held that such “[p]ost-enactment legislative history (a contradiction in terms) is not a legitimate tool of statutory interpretation.” Bruesewitz v. Wyeth LLC, 562 U. S. ___, ___ (2011) (slip op., at 17–18); accord, Federal Nat. Mortgage Assn. v. United States, 379 F. 3d 1303, 1309 (CA Fed. 2004) (dismissing Blue Book as “a post-enactment explanation”). While we have relied on similar documents in the past, see FPC v. Memphis Light, Gas & Water Div., 411 U. S. 458 –472 (1973), our more recent precedents disapprove of that practice. Of course the Blue Book, like a law review article, may be relevant to the extent it is persuasive. But the passage at issue here does not persuade. It concerns a situation quite different from the one we confront: two separate, non-overlapping underpayments, only one of which is attributable to a valuation misstatement. * * * The District Court had jurisdiction in this partnership-level proceeding to determine the applicability of the valuation-misstatement penalty, and the penalty is applicable to tax underpayments resulting from the partners’ participation in the COBRA tax shelter. The judgment of the Court of Appeals is reversed. It is so ordered. Notes 1 The District Court held that the partnerships did not have to be “honored as legitimate for tax purposes” because they did not possess “ ‘economic substance.’ ” App. to Pet. for Cert. 19a. Woods did not appeal the District Court’s application of the economic-substance doctrine, so we express no view on it. 2 Some amici warn that our holding bodes an odd procedural result: The IRS will be able to assess the 40-percent penalty directly, but it will have to use deficiency proceedings to assess the tax underpayment upon which the penalty is imposed. See Brief for New Millennium Trading, LLC, et al. as Amici Curiae 12–13. That criticism assumes that the underpayment would not be exempt from deficiency proceedings because it would rest on outside basis, an “affected ite[m] . . . other than [a] penalt[y],” . We need not resolve that question today, but we do not think amici’s answer necessarily follows. Even an underpayment attributable to an affected item is exempt so long as the affected item does not “require partner level determinations,” ibid.; see Bush v. United States, 655 F. 3d 1323, 1330, 1333–1334 (CA Fed. 2011) (en banc); and it is not readily apparentwhy additional partner-level determinations would be required before adjusting outside basis in a sham partnership. Cf. Petaluma FX Partners, LLC v. Commissioner, 591 F. 3d 649, 655 (CADC 2010)(“If disregarding a partnership leads ineluctably to the conclusion that its partners have no outside basis, that should be just as obvious in partner-level proceedings as it is in partnership-level proceedings”). 3 Congress has since lowered the thresholds for substantial and gross misstatements to 150 percent and 200 percent, respectively. See Pension Protection Act of 2006, §1219(a)(1)–(2), . 4 An amicus suggests that this regulation is in tension with the mathematical rule forbidding division by zero. See Brief for Prof. Amandeep S. Grewal as Amicus Curiae 20, n. 7; cf. Lee’s Summit v. Surface Transp. Bd., 231 F. 3d 39, 41–42 (CADC 2000) (discussing “problems posed by applying [a] 100% increase standard to a baseline of zero”). Woods has not challenged the regulation before this Court, so we assume its validity for purposes of deciding this case. 5 We do not consider Woods’ arguments based on legislative history. Whether or not legislative history is ever relevant, it need not be consulted when, as here, the statutory text is unambiguous. Mohamad v. Palestinian Authority, 566 U. S. ___, ___ (2012) (slip op., at 8). Nor do we evaluate the claim that application of the penalty to legal rather than factual misrepresentations is a recent innovation. An agency’s failure to assert a power, even if prolonged, cannot alter the plain meaning of a statute.

573.US.302

The Clean Air Act imposes permitting requirements on stationary sources, such as factories and powerplants. The Act’s “Prevention of Significant Deterioration” (PSD) provisions make it unlawful to construct or modify a “major emitting facility” in “any area to which [the PSD program] applies” without a permit. §§7475(a)(1), 7479(2)(C). A “major emitting facility” is a stationary source with the potential to emit 250 tons per year of “any air pollutant” (or 100 tons per year for certain types of sources). §7479(1). Facilities seeking to qualify for a PSD permit must, inter alia, comply with emissions limitations that reflect the “best available control technology” (BACT) for “each pollutant subject to regulation under” the Act. §7475(a)(4). In addition, Title V of the Act makes it unlawful to operate any “major source,” wherever located, without a permit. §7661a(a). A “major source” is a stationary source with the potential to emit 100 tons per year of “any air pollutant.” §§7661(2)(B), 7602(j). In response to Massachusetts v. EPA, 549 U.S. 497, EPA promulgated greenhouse-gas emission standards for new motor vehicles, and made stationary sources subject to the PSD program and Title V on the basis of their potential to emit greenhouse gases. It recognized, however, that requiring permits for all sources with greenhouse-gas emissions above the statutory thresholds would radically expand those programs and render them unadministrable. So EPA purported to “tailor” the programs to accommodate greenhouse gases by providing, among other things, that sources would not become newly subject to PSD or Title V permitting on the basis of their potential to emit greenhouse gases in amounts less than 100,000 tons per year. Numerous parties, including several States, challenged EPA’s actions in the D. C. Circuit, which dismissed some of the petitions for lack of jurisdiction and denied the remainder. Held: The judgment is affirmed in part and reversed in part. 684 F.3d 102, affirmed in part and reversed in part. Justice Scalia delivered the opinion of the Court with respect to Parts I and II, concluding: 1. The Act neither compels nor permits EPA to adopt an interpretation of the Act requiring a source to obtain a PSD or Title V permit on the sole basis of its potential greenhouse-gas emissions. Pp. 10–24. (a) The Act does not compel EPA’s interpretation. Massachusetts held that the Act-wide definition of “air pollutant” includes greenhouse gases, 549 U. S., at 529, but where the term “air pollutant” appears in the Act’s operative provisions, including the PSD and Title V permitting provisions, EPA has routinely given it a narrower, context-appropriate meaning. Massachusetts did not invalidate those longstanding constructions. The Act-wide definition is not a command to regulate, but a description of the universe of substances EPA may consider regulating under the Act’s operative provisions. Though Congress’s profligate use of “air pollutant” is not conducive to clarity, the presumption of consistent usage “ ‘readily yields’ ” to context, and a statutory term “may take on distinct characters from association with distinct statutory objects calling for different implementation strategies.” Environmental Defense v. Duke Energy Corp., 549 U.S. 561, 574. Pp. 10–16. (b) Nor does the Act permit EPA’s interpretation. Agencies empowered to resolve statutory ambiguities must operate “within the bounds of reasonable interpretation,” Arlington v. FCC, 569 U. S. ___, ___. EPA has repeatedly acknowledged that applying the PSD and Title V permitting requirements to greenhouse gases would be inconsistent with the Act’s structure and design. A review of the relevant statutory provisions leaves no doubt that the PSD program and Title V are designed to apply to, and cannot rationally be extended beyond, a relative handful of large sources capable of shouldering heavy substantive and procedural burdens. EPA’s interpretation would also bring about an enormous and transformative expansion in EPA’s regulatory authority without clear congressional authorization. FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 160. Pp. 16–20. (c) EPA lacked authority to “tailor” the Act’s unambiguous numerical thresholds of 100 or 250 tons per year to accommodate its greenhouse-gas-inclusive interpretation of the permitting triggers. Agencies must always “ ‘give effect to the unambiguously expressed intent of Congress.’ ” National Assn. of Home Builders v. Defenders of Wildlife, 551 U.S. 644, 665. The power to execute the laws does not include a power to revise clear statutory terms that turn out not to work in practice. Pp. 20–24. 2. EPA reasonably interpreted the Act to require sources that would need permits based on their emission of conventional pollutants to comply with BACT for greenhouse gases. Pp. 24–29. (a) Concerns that BACT, which has traditionally been about end-of-stack controls, is fundamentally unsuited to greenhouse-gas regulation, which is more about energy use, are not unfounded. But an EPA guidance document states that BACT analysis should consider options other than energy efficiency, including “carbon capture and storage,” which EPA contends is reasonably comparable to more traditional, end-of-stack BACT technologies. Moreover, assuming that BACT may be used to force improvements in energy efficiency, important limitations on BACT may work to mitigate concerns about “unbounded” regulatory authority. Pp. 24–27. (b) EPA’s decision to require BACT for greenhouse gases emitted by sources otherwise subject to PSD review is, as a general matter, a permissible interpretation of the statute under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837. The specific phrasing of the BACT provision—which requires BACT “for each pollutant subject to regulation under” the Act, §7475(a)(4)—does not suggest that the provision can bear a narrowing construction. And even if the text were not clear, applying BACT to greenhouse gases is not so disastrously unworkable, and need not result in such a dramatic expansion of agency authority, as to make EPA’s interpretation unreasonable. Pp. 27–29. Scalia, J., announced the judgment of the Court and delivered an opinion, Parts I and II of which were for the Court. Roberts, C. J., and Kennedy, J., joined that opinion in full; Thomas and Alito, JJ., joined as to Parts I, II–A, and II–B–1; and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined as to Part II–B–2. Breyer J., filed an opinion concurring in part and dissenting in part, in which Ginsburg, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in part and dissenting in part, in which Thomas, J., joined. Notes 1 Together with No. 12–1248, American Chemistry Council et al. v. Environmental Protection Agency et al., No. 12–1254, Energy-Intensive Manufacturers Working Group on Greenhouse Gas Regulation et al. v. Environmental Protection Agency et al., No. 12–1268, Southeastern Legal Foundation, Inc., et al. v. Environmental Protection Agency et al., No. 12–1269, Texas et al. v. Environmental Protection Agency et al., and No. 12–1272, Chamber of Commerce of United States States et al. v. Environmental Protection Agency et al., also on certiorari to the same court.

. First, it upheld the Endangerment Finding and Tailpipe Rule. Id., at 119, 126. Next, it held that EPA’s interpretation of the PSD permitting requirement as applying to “any regulated air pollutant,” including greenhouse gases, was “compelled by the statute.” Id., at 133–134. The court also found it “crystal clear that PSD permittees must install BACT for greenhouse gases.” Id., at 137. Because it deemed petitioners’ arguments about the PSD program insufficiently applicable to Title V, it held they had “forfeited any challenges to EPA’s greenhouse gas-inclusive interpretation of Title V.” Id., at 136. Finally, it held that petitioners were without Article III standing to challenge EPA’s efforts to limit the reach of the PSD program and Title V through the Triggering and Tailoring Rules. Id., at 146. The court denied rehearing en banc, with Judges Brown and Kavanaugh each dissenting. No. 09–1322 etc. (Dec. 20, 2012), App. 139, 2012 WL 6621785. We granted six petitions for certiorari but agreed to decide only one question: “ ‘Whether EPA permissibly determined that its regulation of greenhouse gas emissions from new motor vehicles triggered permitting requirements under the Clean Air Act for stationary sources that emit greenhouse gases.’ ” 571 U. S. ____ (2013). II. Analysis This litigation presents two distinct challenges to EPA’s stance on greenhouse-gas permitting for stationary sources. First, we must decide whether EPA permissibly determined that a source may be subject to the PSD and Title V permitting requirements on the sole basis of the source’s potential to emit greenhouse gases. Second, we must decide whether EPA permissibly determined that a source already subject to the PSD program because of its emission of conventional pollutants (an “anyway” source) may be required to limit its greenhouse-gas emissions by employing the “best available control technology” for greenhouse gases. The Solicitor General joins issue on both points but evidently regards the second as more important; he informs us that “anyway” sources account for roughly 83% of American stationary-source greenhouse-gas emissions, compared to just 3% for the additional, non-“anyway” sources EPA sought to regulate at Steps 2 and 3 of the Tailoring Rule. Tr. of Oral Arg. 52. We review EPA’s interpretations of the Clean Air Act using the standard set forth in Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 –843 (1984). Under Chevron, we presume that when an agency-administered statute is ambiguous with respect to what it prescribes, Congress has empowered the agency to resolve the ambiguity. The question for a reviewing court is whether in doing so the agency has acted reasonably and thus has “stayed within the bounds of its statutory authority.” Arlington v. FCC, 569 U. S. ___, ___ (2013) (slip op., at 5) (emphasis deleted). A. The PSD and Title V Triggers We first decide whether EPA permissibly interpreted the statute to provide that a source may be required to obtain a PSD or Title V permit on the sole basis of its potential greenhouse-gas emissions. 1 EPA thought its conclusion that a source’s greenhouse-gas emissions may necessitate a PSD or Title V permit followed from the Act’s unambiguous language. The Court of Appeals agreed and held that the statute “compelled” EPA’s interpretation. 684 F. 3d, at 134. We disagree. The statute compelled EPA’s greenhouse-gas-inclusive interpretation with respect to neither the PSD program nor Title V.[4] The Court of Appeals reasoned by way of a flawed syllogism: Under Massachusetts, the general, Act-wide definition of “air pollutant” includes greenhouse gases; the Act requires permits for major emitters of “any air pollutant”; therefore, the Act requires permits for major emitters of greenhouse gases. The conclusion follows from the premises only if the air pollutants referred to in the permit-requiring provisions (the minor premise) are the same air pollutants encompassed by the Act-wide definition as interpreted in Massachusetts (the major premise). Yet no one—least of all EPA—endorses that proposition, and it is obviously untenable. The Act-wide definition says that an air pollutant is “any air pollution agent or combination of such agents, including any physical, chemical, biological, [or] radioactive . . . substance or matter which is emitted into or otherwise enters the ambient air.” §7602(g). In Massachusetts, the Court held that the Act-wide definition includes greenhouse gases because it is all-encompassing; it “embraces all airborne compounds of whatever stripe.” 549 U. S., at 529. But where the term “air pollutant” appears in the Act’s operative provisions, EPA has routinely given it a narrower, context-appropriate meaning. That is certainly true of the provisions that require PSD and Title V permitting for major emitters of “any air pollutant.” Since 1978, EPA’s regulations have interpreted “air pollutant” in the PSD permitting trigger as limited to regulated air pollutants, 43 Fed. Reg. 26403, codified, as amended, 40 CFR §52.21(b)(1)–(2), (50)—a class much narrower than Massachusetts’ “all airborne compounds of whatever stripe,” 549 U. S., at 529. And since 1993 EPA has informally taken the same position with regard to the Title V permitting trigger, a position the Agency ultimately incorporated into some of the regulations at issue here. See Memorandum from Lydia N. Wegman, Deputy Director, Office of Air Quality Planning and Standards, to Air Division Director, Regions I–X, pp. 4–5 (Apr. 26, 1993); Tailoring Rule 31607–31608 (amending 40 CFR §§70.2, 71.2). Those interpretations were appropriate: It is plain as day that the Act does not envision an elaborate, burdensome permitting process for major emitters of steam, oxygen, or other harmless airborne substances. It takes some cheek for EPA to insist that it cannot possibly give “air pollutant” a reasonable, context-appropriate meaning in the PSD and Title V contexts when it has been doing precisely that for decades. Nor are those the only places in the Act where EPA has inferred from statutory context that a generic reference to air pollutants does not encompass every substance falling within the Act-wide definition. Other examples abound: The Act authorizes EPA to enforce new source performance standards (NSPS) against a pre-existing source if, after promulgation of the standards, the source undergoes a physical or operational change that increases its emission of “any air pollutant.” §7411(a)(2), (4), (b)(1)(B). EPA interprets that provision as limited to air pollutants for which EPA has promulgated new source performance standards. 36 Fed. Reg. 24877 (1971), codified, as amended, 40 CFR §60.2; 40 Fed. Reg. 58419 (1975), codified, as amended, 40 CFR §60.14(a). The Act requires a permit for the construction or operation in a nonattainment area of a source with the potential to emit 100 tons per year of “any air pollutant.” §§7502(c)(5), 7602(j). EPA interprets that provision as limited to pollutants for which the area is designated nonattainment. 45 Fed. Reg. 52745 (1980), promulgating 40 CFR §51.18(j)(2), as amended, §51.165(a)(2). The Act directs EPA to require “enhanced monitoring and submission of compliance certifications” for any source with the potential to emit 100 tons per year of “any air pollutant.” §§7414(a)(3), 7602(j). EPA interprets that provision as limited to regulated pollutants. 62 Fed. Reg. 54941 (1997), codifiedat 40 CFR §§64.1, 64.2. The Act requires certain sources of air pollutants that interfere with visibility to undergo retrofitting if they have the potential to emit 250 tons per year of “any pollutant.” §7491(b)(2)(A), (g)(7). EPA interprets that provision as limited to visibility-impairing air pollutants. 70 Fed. Reg. 39160 (2005), codified at 40 CFR pt. 51, App. Y, §II.A.3. Although these limitations are nowhere to be found in the Act-wide definition, in each instance EPA has concluded—as it has in the PSD and Title V context—that the statute is not using “air pollutant” in Massachusetts’ broad sense to mean any airborne substance whatsoever. Massachusetts did not invalidate all these longstanding constructions. That case did not hold that EPA must always regulate greenhouse gases as an “air pollutant” everywhere that term appears in the statute, but only that EPA must “ground its reasons for action or inaction in the statute,” 549 U. S., at 535 (emphasis added), rather than on “reasoning divorced from the statutory text,” id., at 532. EPA’s inaction with regard to Title II was not sufficiently grounded in the statute, the Court said, in part because nothing in the Act suggested that regulating greenhouse gases under that Title would conflict with the statutory design. Title II would not compel EPA to regulate in any way that would be “extreme,” “counterintuitive,” or contrary to “ ‘common sense.’ ” Id., at 531. At most, it would require EPA to take the modest step of adding greenhouse-gas standards to the roster of new-motor-vehicle emission regulations. Ibid. Massachusetts does not strip EPA of authority to exclude greenhouse gases from the class of regulable air pollutants under other parts of the Act where their inclusion would be inconsistent with the statutory scheme. The Act-wide definition to which the Court gave a “sweeping” and “capacious” interpretation, id., at 528, 532, is not a command to regulate, but a description of the universe of substances EPA may consider regulating under the Act’s operative provisions. Massachusetts does not foreclose the Agency’s use of statutory context to infer that certain of the Act’s provisions use “air pollutant” to denote not every conceivable airborne substance, but only those that may sensibly be encompassed within the particular regulatory program. As certain amici felicitously put it, while Massachusetts “rejected EPA’s categorical contention that greenhouse gases could not be ‘air pollutants’ for any purposes of the Act,” it did not “embrace EPA’s current, equally categorical position that greenhouse gases must be air pollutants for all purposes” regardless of the statutory context. Brief for Administrative Law Professors et al. as Amici Curiae 17.[5] To be sure, Congress’s profligate use of “air pollutant” where what is meant is obviously narrower than the Act-wide definition is not conducive to clarity. One ordinarily assumes “ ‘that identical words used in different parts of the same act are intended to have the same meaning.’ ” Environmental Defense v. Duke Energy Corp., 549 U. S. 561, 574 (2007) . In this respect (as in countless others), the Act is far from a chef d’oeuvre of legislative draftsmanship. But we, and EPA, must do our best, bearing in mind the “ ‘fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.’ ” FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120, 133 (2000) . As we reiterated the same day we decided Massachusetts, the presumption of consistent usage “ ‘readily yields’ ” to context, and a statutory term—even one defined in the statute—“may take on distinct characters from association with distinct statutory objects calling for different implementation strategies.” Duke Energy, supra, at 574. We need not, and do not, pass on the validity of all the limiting constructions EPA has given the term “air pollutant” throughout the Act. We merely observe that taken together, they belie EPA’s rigid insistence that when interpreting the PSD and Title V permitting requirements it is bound by the Act-wide definition’s inclusion of greenhouse gases, no matter how incompatible that inclusion is with those programs’ regulatory structure. In sum, there is no insuperable textual barrier to EPA’s interpreting “any air pollutant” in the permitting triggers of PSD and Title V to encompass only pollutants emitted in quantities that enable them to be sensibly regulated at the statutory thresholds, and to exclude those atypical pollutants that, like greenhouse gases, are emitted in such vast quantities that their inclusion would radically transform those programs and render them unworkable as written.[6] 2 Having determined that EPA was mistaken in thinking the Act compelled a greenhouse-gas-inclusive interpretation of the PSD and Title V triggers, we next consider the Agency’s alternative position that its interpretation was justified as an exercise of its “discretion” to adopt “a reasonable construction of the statute.” Tailoring Rule 31517. We conclude that EPA’s interpretation is not permissible. Even under Chevron’s deferential framework, agencies must operate “within the bounds of reasonable interpretation.” Arlington, 569 U. S., at ___ (slip op., at 5). And reasonable statutory interpretation must account for both “the specific context in which . . . language is used” and “the broader context of the statute as a whole.” Robinson v. Shell Oil Co., 519 U. S. 337, 341 (1997) . A statutory “provision that may seem ambiguous in isolation is often clarified by the remainder of the statutory scheme . . . because only one of the permissible meanings produces a substantive effect that is compatible with the rest of the law.” United Sav. Assn. of Tex. v. Timbers of Inwood Forest Associates, Ltd., 484 U. S. 365, 371 (1988) . Thus, an agency interpretation that is “inconsisten[t] with the design and structure of the statute as a whole,” University of Tex. Southwestern Medical Center v. Nassar, 570 U. S. ___, ___ (2013) (slip op., at 13), does not merit deference. EPA itself has repeatedly acknowledged that applying the PSD and Title V permitting requirements to greenhouse gases would be inconsistent with—in fact, would overthrow—the Act’s structure and design. In the Tailoring Rule, EPA described the calamitous consequences of interpreting the Act in that way. Under the PSD program, annual permit applications would jump from about 800 to nearly 82,000; annual administrative costs would swell from $12 million to over $1.5 billion; and decade-long delays in issuing permits would become common, causing construction projects to grind to a halt nationwide. Tailoring Rule 31557. The picture under Title V was equally bleak: The number of sources required to have permits would jump from fewer than 15,000 to about 6.1 million; annual administrative costs would balloon from $62 million to $21 billion; and collectively the newly covered sources would face permitting costs of $147 billion. Id., at 31562–31563. Moreover, “the great majority of additional sources brought into the PSD and title V programs would be small sources that Congress did not expect would need to undergo permitting.” Id., at 31533. EPA stated that these results would be so “contrary to congressional intent,” and would so “severely undermine what Congress sought to accomplish,” that they necessitated as much as a 1,000-fold increase in the permitting thresholds set forth in the statute. Id., at 31554, 31562. Like EPA, we think it beyond reasonable debate that requiring permits for sources based solely on their emission of greenhouse gases at the 100- and 250-tons-per-year levels set forth in the statute would be “incompatible” with “the substance of Congress’ regulatory scheme.” Brown & Williamson, 529 U. S., at 156. A brief review of the relevant statutory provisions leaves no doubt that the PSD program and Title V are designed to apply to, and cannot rationally be extended beyond, a relative handful of large sources capable of shouldering heavy substantive and procedural burdens. Start with the PSD program, which imposes numerous and costly requirements on those sources that are required to apply for permits. Among other things, the applicant must make available a detailed scientific analysis of the source’s potential pollution-related impacts, demonstrate that the source will not contribute to the violation of any applicable pollution standard, and identify and use the “best available control technology” for each regulated pollutant it emits. §7475(a)(3), (4), (6), (e). The permitting authority (the State, usually) also bears its share of the burden: It must grant or deny a permit within a year, during which time it must hold a public hearing on the application. §7475(a)(2), (c). Not surprisingly, EPA acknowledges that PSD review is a “complicated, resource-intensive, time-consuming, and sometimes contentious process” suitable for “hundreds of larger sources,” not “tens of thousands of smaller sources.” 74 Fed. Reg. 55304, 55321–55322. Title V contains no comparable substantive requirements but imposes elaborate procedural mandates. It requires the applicant to submit, within a year of becoming subject to Title V, a permit application and a “compliance plan” describing how it will comply with “all applicable requirements” under the Act; to certify its compliance annually; and to submit to “inspection, entry, monitoring, . . . and reporting requirements.” §§7661b(b)–(c), 7661c(a)–(c). The procedural burdens on the permitting authority and EPA are also significant. The permitting authority must hold a public hearing on the application, §7661a(b)(6), and it must forward the application and any proposed permit to EPA and neighboring States and respond in writing to their comments, §7661d(a), (b)(1). If it fails to issue or deny the permit within 18 months, any interested party can sue to require a decision “without additional delay.” §§7661a(b)(7), 7661b(c). An interested party also can petition EPA to block issuance of the permit; EPA must grant or deny the petition within 60 days, and its decision may be challenged in federal court. §7661d(b)(2)–(3). As EPA wrote, Title V is “finely crafted for thousands,” not millions, of sources. Tailoring Rule 31563. The fact that EPA’s greenhouse-gas-inclusive interpretation of the PSD and Title V triggers would place plainly excessive demands on limited governmental resources is alone a good reason for rejecting it; but that is not the only reason. EPA’s interpretation is also unreasonable because it would bring about an enormous and transformative expansion in EPA’s regulatory authority without clear congressional authorization. When an agency claims to discover in a long-extant statute an unheralded power to regulate “a significant portion of the American economy,” Brown & Williamson, 529 U. S., at 159, we typically greet its announcement with a measure of skepticism. We expect Congress to speak clearly if it wishes to assign to an agency decisions of vast “economic and political significance.” Id., at 160; see also MCI Telecommunications Corp. v. American Telephone & Telegraph Co., 512 U. S. 218, 231 (1994) ; Industrial Union Dept., AFL–CIO v. American Petroleum Institute, 448 U. S. 607 –646 (1980) (plurality opinion). The power to require permits for the construction and modification of tens of thousands, and the operation of millions, of small sources nationwide falls comfortably within the class of authorizations that we have been reluctant to read into ambiguous statutory text. Moreover, in EPA’s assertion of that authority, we confront a singular situation: an agency laying claim to extravagant statutory power over the national economy while at the same time strenuously asserting that the authority claimed would render the statute “unrecognizable to the Congress that designed” it. Tailoring Rule 31555. Since, as we hold above, the statute does not compel EPA’s interpretation, it would be patently unreason-able—not to say outrageous—for EPA to insist on seizing expansive power that it admits the statute is not designed to grant.[7] 3 EPA thought that despite the foregoing problems, it could make its interpretation reasonable by adjusting the levels at which a source’s greenhouse-gas emissions would oblige it to undergo PSD and Title V permitting. Although the Act, in no uncertain terms, requires permits for sources with the potential to emit more than 100 or 250 tons per year of a relevant pollutant, EPA in its Tailoring Rule wrote a new threshold of 100,000 tons per year for greenhouse gases. Since the Court of Appeals thought the statute unambiguously made greenhouse gases capable of triggering PSD and Title V, it held that petitioners lacked Article III standing to challenge the Tailoring Rule because that rule did not injure petitioners but merely relaxed the pre-existing statutory requirements. Because we, however, hold that EPA’s greenhouse-gas-inclusive interpretation of the triggers was not compelled, and because EPA has essentially admitted that its interpretation would be unreasonable without “tailoring,” we consider the validity of the Tailoring Rule. We conclude that EPA’s rewriting of the statutory thresholds was impermissible and therefore could not validate the Agency’s interpretation of the triggering provisions. An agency has no power to “tailor” legislation to bureaucratic policy goals by rewriting unambiguous statutory terms. Agencies exercise discretion only in the interstices created by statutory silence or ambiguity; they must always “ ‘give effect to the unambiguously expressed intent of Congress.’ ” National Assn. of Home Builders v. Defenders of Wildlife, 551 U. S. 644, 665 (2007) (quoting Chevron, 467 U. S., at 843). It is hard to imagine a statutory term less ambiguous than the precise numerical thresholds at which the Act requires PSD and Title V permitting. When EPA replaced those numbers with others of its own choosing, it went well beyond the “bounds of its statutory authority.” Arlington, 569 U. S., at ___ (slip op., at 5) (emphasis deleted). The Solicitor General does not, and cannot, defend the Tailoring Rule as an exercise of EPA’s enforcement discretion. The Tailoring Rule is not just an announcement of EPA’s refusal to enforce the statutory permitting requirements; it purports to alter those requirements and to establish with the force of law that otherwise-prohibited conduct will not violate the Act. This alteration of the statutory requirements was crucial to EPA’s “tailoring” efforts. Without it, small entities with the potential to emit greenhouse gases in amounts exceeding the statutory thresholds would have remained subject to citizen suits—authorized by the Act—to enjoin their construction, modification, or operation and to impose civil penalties of up to $37,500 per day of violation. §§7413(b), 7604(a), (f)(4); 40 CFR §19.4. EPA itself has recently affirmed that the “independent enforcement authority” furnished by the citizen-suit provision cannot be displaced by a permitting authority’s decision not to pursue enforcement. 78 Fed. Reg. 12477, 12486–12487 (2013). The Solicitor General is therefore quite right to acknowledge that the availability of citizen suits made it necessary for EPA, in seekingto mitigate the unreasonableness of its greenhouse-gas-inclusive interpretation, to go beyond merely exercising its enforcement discretion. See Tr. of Oral Arg. 87–88. For similar reasons, Morton v. Ruiz, 415 U. S. 199 (1974) —to which the Solicitor General points as the best case supporting the Tailoring Rule, see Tr. of Oral Arg. 71, 80–81—is irrelevant. In Ruiz, Congress had appropriated funds for the Bureau of Indian Affairs to spend on providing assistance to “ ‘Indians throughout the United States’ ” and had not “impose[d] any geographical limitation on the availability of general assistance benefits.” Id., at 206–207, and n. 7. Although we held the Bureau could not deny benefits to off-reservation Indians because it had not published its eligibility criteria, we stated in dictum that the Bureau could, if it followed proper administrative procedures, “create reasonable classifications and eligibility requirements in order to allocate the limited funds available.” Id., at 230–231. That dictum stands only for the unremarkable proposition that an agency may adopt policies to prioritize its expenditures within the bounds established by Congress. See also Lincoln v. Vigil, 508 U. S. 182 –193 (1993). Nothing in Ruiz remotely authorizes an agency to modify unambiguous requirements imposed by a federal statute. An agency confronting resource constraints may change its own conduct, but it cannot change the law. Were we to recognize the authority claimed by EPA in the Tailoring Rule, we would deal a severe blow to the Constitution’s separation of powers. Under our system of government, Congress makes laws and the President, acting at times through agencies like EPA, “faithfully execute[s]” them. U. S. Const., Art. II, §3; see Medellín v. Texas, 552 U. S. 491 –527 (2008). The power of executing the laws necessarily includes both authority and responsibility to resolve some questions left open by Congress that arise during the law’s administration. But it does not include a power to revise clear statutory terms that turn out not to work in practice. See, e.g., Barnhart v. Sigmon Coal Co., 534 U. S. 438, 462 (2002) (agency lacked authority “to develop new guidelines or to assign liability in a manner inconsistent with” an “unambiguous statute”). In the Tailoring Rule, EPA asserts newfound authority to regulate millions of small sources—including retail stores, offices, apartment buildings, shopping centers, schools, and churches—and to decide, on an ongoing basis and without regard for the thresholds prescribed by Congress, how many of those sources to regulate. We are not willing to stand on the dock and wave goodbye as EPA embarks on this multiyear voyage of discovery. We reaffirm the core administrative-law principle that an agency may not rewrite clear statutory terms to suit its own sense of how the statute should operate. EPA therefore lacked authority to “tailor” the Act’s unambiguous numerical thresholds to accommodate its greenhouse-gas-inclusive interpretation of the permitting triggers. Instead, the need to rewrite clear provisions of the statute should have alerted EPA that it had taken a wrong interpretive turn. Agencies are not free to “adopt . . . unreasonable interpretations of statutory provisions and then edit other statu-tory provisions to mitigate the unreasonableness.” App. 175, 2012 WL 6621785, *16 (Kavanaugh, J., dissenting from denial of rehearing en banc). Because the Tailoring Rule cannot save EPA’s interpretation of the triggers, that interpretation was impermissible under Chevron.[8] B. BACT for “Anyway” Sources For the reasons we have given, EPA overstepped its statutory authority when it decided that a source could become subject to PSD or Title V permitting by reason of its greenhouse-gas emissions. But what about “anyway” sources, those that would need permits based on their emissions of more conventional pollutants (such as particulate matter)? We now consider whether EPA reasonably interpreted the Act to require those sources to comply with “best available control technology” emission standards for greenhouse gases. 1 To obtain a PSD permit, a source must be “subject to the best available control technology” for “each pollutant subject to regulation under [the Act]” that it emits. §7475(a)(4). The Act defines BACT as “an emission limitation based on the maximum degree of reduction of each pollutant subject to regulation” that is “achievable . . . through application of production processes and available methods, systems, and techniques, including fuel cleaning, clean fuels, or treatment or innovative fuel combustion techniques.” §7479(3). BACT is determined “on a case-by-case basis, taking into account energy, environmental, and economic impacts and other costs.” Ibid. Some petitioners urge us to hold that EPA may never require BACT for greenhouse gases—even when a source must undergo PSD review based on its emissions of conventional pollutants—because BACT is fundamentally unsuited to greenhouse-gas regulation. BACT, they say, has traditionally been about end-of-stack controls “such as catalytic converters or particle collectors”; but applying it to greenhouse gases will make it more about regulating energy use, which will enable regulators to control “every aspect of a facility’s operation and design,” right down to the “light bulbs in the factory cafeteria.” Brief for Petitioner Energy-Intensive Manufacturers Working Group on Greenhouse Gas Regulation et al. in No. 12–1254, p. 7; see Joint Reply Brief for Petitioners in No. 12–1248 etc., pp. 14–15 (“BACT for [greenhouse gases] becomes an unbounded exercise in command-and-control regulation” of everything from “efficient light bulbs” to “basic industrial processes”). But see Brief for Calpine Corp. as Amicus Curiae 10 (“[I]n Calpine’s experience with ‘anyway’ sources, the [greenhouse-gas] analysis was only a small part of the overall permitting process”). EPA has published a guidance document that lends some credence to petitioners’ fears. It states that at least initially, compulsory improvements in energy efficiency will be the “foundation” of greenhouse-gas BACT, with more traditional end-of-stack controls either not used or “added as they become more available.” PSD and Title V Permitting Guidance for Greenhouse Gases 29 (Mar. 2011) (hereinafter Guidance); see Peloso & Dobbins, Greenhouse Gas PSD Permitting: The Year in Review, 42 Tex. Env. L. J. 233, 247 (2012) (“Because [other controls] tend to prove infeasible, energy efficiency measures dominate the [greenhouse-gas] BACT controls approved by the states and EPA”). But EPA’s guidance also states that BACT analysis should consider options other than energy efficiency, such as “carbon capture and storage.” Guidance 29, 32, 35–36, 42–43. EPA argues that carbon capture is reasonably comparable to more traditional, end-of-stack BACT technologies, id., at 32, n. 86, and petitioners do not dispute that. Moreover, assuming without deciding that BACT may be used to force some improvements in energy efficiency, there are important limitations on BACT that may work to mitigate petitioners’ concerns about “unbounded” regulatory authority. For one, BACT is based on “control technology” for the applicant’s “proposed facility,” §7475(a)(4); therefore, it has long been held that BACT cannot be used to order a fundamental redesign of the facility. See, e.g., Sierra Club v. EPA, 499 F. 3d 653, 654–655 (CA7 2007); In re Pennsauken Cty., N. J., Resource Recovery Facility, 2 E. A. D. 667, 673 (EAB 1988). For another, EPA has long interpreted BACT as required only for pollutants that the source itself emits, see 44 Fed. Reg. 51947 (1979); accordingly, EPA acknowledges that BACT may not be used to require “reductions in a facility’s demand for energy from the electric grid.” Guidance 24. Finally, EPA’s guidance suggests that BACT should not require every conceivable change that could result in minor improvements in energy efficiency, such as the aforementioned light bulbs. Id., at 31. The guidance explains that permitting authorities should instead consider whether a proposed regulatory burden outweighs any reduction in emissions to be achieved, and should concentrate on the facility’s equipment that uses the largest amounts of energy. Ibid. 2 The question before us is whether EPA’s decision to require BACT for greenhouse gases emitted by sources otherwise subject to PSD review is, as a general matter, a permissible interpretation of the statute under Chevron. We conclude that it is. The text of the BACT provision is far less open-ended than the text of the PSD and Title V permitting triggers. It states that BACT is required “for each pollutant subject to regulation under this chapter” (i.e., the entire Act), §7475(a)(4), a phrase that—as the D. C. Circuit wrote 35 years ago—“would not seem readily susceptible [of] misinterpretation.” Alabama Power Co. v. Costle, 636 F. 2d 323, 404 (1979). Whereas the dubious breadth of “any air pollutant” in the permitting triggers suggests a role for agency judgment in identifying the subset of pollutants covered by the particular regulatory program at issue, the more specific phrasing of the BACT provision suggests that the necessary judgment has already been made by Congress. The wider statutory context likewise does not suggest that the BACT provision can bear a narrowing construction: There is no indication that the Act elsewhere uses, or that EPA has interpreted, “each pollutant subject to regulation under this chapter” to mean anything other than what it says. Even if the text were not clear, applying BACT to greenhouse gases is not so disastrously unworkable, and need not result in such a dramatic expansion of agency authority, as to convince us that EPA’s interpretation is unreasonable. We are not talking about extending EPA jurisdiction over millions of previously unregulated entities, but about moderately increasing the demands EPA (or a state permitting authority) can make of entities already subject to its regulation. And it is not yet clear that EPA’s demands will be of a significantly different character from those traditionally associated with PSD review. In short, the record before us does not establish that the BACT provision as written is incapable of being sensibly applied to greenhouse gases. We acknowledge the potential for greenhouse-gas BACT to lead to an unreasonable and unanticipated degree of regulation, and our decision should not be taken as an endorsement of all aspects of EPA’s current approach, nor as a free rein for any future regulatory application of BACT in this distinct context. Our narrow holding is that nothing in the statute categorically prohibits EPA from interpreting the BACT provision to apply to greenhouse gases emitted by “anyway” sources. However, EPA may require an “anyway” source to comply with greenhouse-gas BACT only if the source emits more than a de minimis amount of greenhouse gases. As noted above, the Tailoring Rule applies BACT only if a source emits greenhouse gases in excess of 75,000 tons per year CO2e, but the Rule makes clear that EPA did not arrive at that number by identifying the de minimis level. See nn. 1, 3, supra. EPA may establish an appropriate de minimis threshold below which BACT is not required for a source’s greenhouse-gas emissions. We do not hold that 75,000 tons per year CO2e necessarily exceeds a true de minimis level, only that EPA must justify its selection on proper grounds. Cf. Alabama Power, supra, at 405.[9] * * * To sum up: We hold that EPA exceeded its statutory authority when it interpreted the Clean Air Act to require PSD and Title V permitting for stationary sources based on their greenhouse-gas emissions. Specifically, the Agency may not treat greenhouse gases as a pollutant for pur-poses of defining a “major emitting facility” (or a “modification” thereof) in the PSD context or a “major source” in the Title V context. To the extent its regulations purport to do so, they are invalid. EPA may, however, continue to treat greenhouse gases as a “pollutant subject to regulation under this chapter” for purposes of requiring BACT for “anyway” sources. The judgment of the Court of Appeals is affirmed in part and reversed in part. It is so ordered.Notes 1 Although the statute sets numerical thresholds (100 or 250 tons per year) for emissions that will make a facility “major,” it does not specify by how much a physical or operational change must increase emissions to constitute a permit-requiring “modification.” Nor does it say how much of a given regulated pollutant a “major emitting facility” must emit before it is subject to BACT for that pollutant. EPA, however, has established pollutant-specific numerical thresholds below which a facility’s emissions of a pollutant, and increases therein, are considered for those purposes. See 40 CFR §§51.166(b)(2)(i), (23), (39), (j)(2)–(3), 52.21(b)(2)(i), (23), (40), (j)(2)–(3); see also v. , 636 F. 2d 323, 360–361, 400, 405 (CADC 1979) (recognizing this authority in EPA); cf. v. , (“ . . . is part of the established background of legal principles against which all enactments are adopted”). 2 Comments from other Executive Branch agencies reprinted in the notice echoed those concerns. See, 73 Fed. Reg. 44360 (Departments of Agriculture, Commerce, Transportation, and Energy noting EPA would “exercis[e] de facto zoning authority through control over thousands of what formerly were local or private decisions, impacting the construction of schools, hospitals, and commercial and residential development”); at 44383 (Council of Economic Advisers and Office of Science and Technology Policy stating that “[s]mall manufacturing facilities, schools, and shopping centers” would be subject to “full major source permitting”); at 44385 (Council on Environmental Quality noting “the prospect of essentially automatic and immediate regulation over a vast range of community and business activity”); at 44391 (Small Business Administration finding it “difficult to overemphasize how potentially disruptive and burdensome such a new regulatory regime would be to small entities” such as “office buildings, retail establishments, hotels, . . . schools, prisons, and private hospitals”). 3 EPA stated that its adoption of a 75,000-tons-per-year threshold for emissions requiring BACT and modifications requiring permits was not an exercise of its authority to establish exceptions and that a truly level might be “well below” 75,000 tons per year. Tailoring Rule 31560; cf. n. 1, . 4 The Court of Appeals held that petitioners’ arguments applied only to the PSD program and that petitioners had therefore “forfeited any challenges to EPA’s greenhouse gas-inclusive interpretation of Title V.” 684 F. 3d, at 136. The Solicitor General does not defend the Court of Appeals’ ruling on forfeiture, and he concedes that some of the arguments petitioners have made before this Court apply to Title V as well as the PSD program. See Brief for Federal Respondents 56. We agree, and we are satisfied that those arguments were also made below. See, Brief for State Petitioners et al. in No. 10–1073 etc. (CADC), pp. 59–73; Brief for Non-State Petitioners et al. in No. 10–1073 etc. (CADC), pp. 46–47. 5 Our decision in v. , 564 U. S. ___ (2011), does not suggest otherwise. We there held that the Act’s authorization for EPA to establish performance standards for powerplant greenhouse-gas emissions displaced any federal-common-law right that might otherwise have existed to seek abatement of those emissions. at ___ (slip op., at 10). The authorization to which we referred was that given in the NSPS program of §7411, a part of the Act not at issue here and one that no party in argued was ill suited to accommodating greenhouse gases. 6 During the course of this litigation, several possible limiting constructions for the PSD trigger have been proposed. Judge Kavanaugh argued below that it would be plausible for EPA to read “any air pollutant” in the PSD context as limited to the six NAAQS pollutants. See v. , No. 09–1322 etc. (CADC, Dec. 20, 2012), App. 171–180, 2012 WL 6621785, *15–*18 (opinion dissenting from denial of rehearing en banc). Some petitioners make a slightly different argument: that because PSD permitting is required only for major emitting facilities “in any area to which [the PSD program] applies,” §7475(a), the relevant pollutants are only those NAAQS pollutants for which the area in question is designated attainment or unclassifiable. That approach would bring EPA’s interpretation of the PSD trigger in line with its longstanding interpretation of the permitting requirements for nonattainment areas. Others maintain that “any air pollutant” in the PSD provision should be limited to air pollutants with localized effects on air quality. We do not foreclose EPA or the courts from considering those constructions in the future, but we need not do so today. 7 A few additional points bear mentioning. The Solicitor General conjectures that EPA might eventually alter its longstanding interpretation of “potential to emit” in order to reduce the number of sources required to have permits at the statutory thresholds. But neither he nor the Agency has given us any reason to believe that there exists a plausible reading of “potential to emit” that EPA would willingly adopt and that would eliminate the unreasonableness of EPA’s interpretation. Nor have we been given any information about the ability of other possible “streamlining” techniques alluded to by EPA—such as “general” or “electronic” permitting—to reduce the administrability problems identified above; and in any event, none of those techniques would address the more fundamental problem of EPA’s claiming regulatory authority over millions of small entities that it acknowledges the Act does not seek to regulate. Finally, the Solicitor General suggests that the incompatibility of greenhouse gases with the PSD program and Title V results chiefly from the inclusion of carbon dioxide in the “aggregate pollutant” defined by EPA. We decide these cases on the basis of the pollutant “greenhouse gases” as EPA has defined and regulated it, and we express no view on how our analysis might change were EPA to define it differently. 8 argues, at 10 (opinion concurring in part and dissenting in part), that when the statutory permitting thresholds of 100 or 250 tons per year do not provide a “sensible regulatory line,” EPA is entitled to “read an unwritten exception” into “the particular number used by the statute”—by which he apparently means that the Agency is entitled to substitute a dramatically higher number, such as 100,000. We are aware of no principle of administrative law that would allow an agency to rewrite such a clear statutory term, and we shudder to contemplate the effect that such a principle would have on demo-cratic governance. 9 argues that BACT is “fundamentally incompatible” with greenhouse gases for two reasons. at 4 (opinion concurring in part and dissenting in part). First, BACT requires consideration of “ambient air quality at the proposed site and in areas which may be affected by emissions from [the proposed] facility for each pollutant subject to regulation under this chapter,” §7475(e)(1); see also §7475(e)(3)(B); and it is not obvious how that requirement should apply, or even whether it can apply, to greenhouse gases. , at 4–5. But the possibility that that requirement may be inoperative as to greenhouse gases does not convince us that they must be categorically excluded from BACT even though they are indisputably a “pollutant subject to regulation.” Second, argues that EPA’s guidance on how to implement greenhouse-gas BACT is a recipe for “arbitrary and inconsistent decisionmaking.” at 8. But we are not reviewing EPA’s guidance in these cases, and we cannot say that it is impossible for EPA and state permitting authorities to devise rational ways of complying with the statute’s directive to determine BACT for greenhouse gases “on a case-by-case basis, taking into account energy, environmental, and economic impacts and other costs.” §7479(3).

571.US.277

Petitioner Walden, a Georgia police officer working as a deputized Drug Enforcement Administration agent at a Georgia airport, searched respondents and seized a large amount of cash. Respondents allege that after they returned to their Nevada residence, petitioner helped draft a false probable cause affidavit in support of the funds’ forfeiture and forwarded it to a United States Attorney’s Office in Georgia. In the end, no forfeiture complaint was filed, and respondents’ funds were returned. Respondents filed a tort suit against petitioner in Federal District Court in Nevada. The District Court dismissed the suit, finding that the Georgia search and seizure did not establish a basis to exercise personal jurisdiction in Nevada. The Ninth Circuit reversed, holding that the District Court could properly exercise jurisdiction because petitioner had submitted the false probable cause affidavit with the knowledge that it would affect persons with significant Nevada connections. Held: The District Court lacked personal jurisdiction over petitioner. Pp. 5–14. (a) The Fourteenth Amendment’s Due Process Clause constrains a State’s authority to bind a nonresident defendant to a judgment of its courts, World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 291, and requires that the nonresident have “certain minimum contacts” with the forum State, International Shoe Co. v. Washington, 326 U.S. 310, 316. The inquiry into the “minimum contacts” necessary to create specific jurisdiction focuses “on the relationship among the defendant, the forum, and the litigation.” Keeton v. Hustler Magazine, Inc., 465 U.S. 770, 775. For a State to exercise jurisdiction consistent with due process, that relationship must arise out of contacts that the “defendant himself” creates with the forum, Burger King Corp. v. Rudzewicz, 471 U.S. 462, 475, and must be analyzed with regard to the defendant’s contacts with the forum itself, not with persons residing there, see, e.g., International Shoe, supra, at 319. The plaintiff cannot be the only link between the defendant and the forum. These same principles apply when intentional torts are involved. See Calder v. Jones, 465 U.S. 783, 788–789. Pp. 5–10. (b) Petitioner lacks the “minimal contacts” with Nevada that are a prerequisite to the exercise of jurisdiction over him. No part of petitioner’s course of conduct occurred in Nevada, and he formed no jurisdictionally relevant contacts with that forum. The Ninth Circuit reached its contrary conclusion by improperly shifting the analytical focus from petitioner’s contacts with the forum to his contacts with respondents, obscuring the reality that none of petitioner’s challenged conduct had anything to do with Nevada itself. Respondents emphasize that they suffered the “injury” caused by the delayed return of their funds while residing in Nevada, but Calder made clear that mere injury to a forum resident is not a sufficient connection to the forum. The proper question is whether the defendant’s conduct connects him to the forum in a meaningful way: Here, respondents’ claimed injury does not evince such a connection. The injury occurred in Nevada simply because that is where respondents chose to be when they desired to use the seized funds. Other possible contacts noted by the Ninth Circuit—that respondents’ Nevada attorney contacted petitioner in Georgia, that cash seized in Georgia originated in Nevada, and that funds were returned to respondents in Nevada—are ultimately unavailing. Pp. 11–14. 688 F.3d 558, reversed. Thomas, J., delivered the opinion for a unanimous Court.

This case asks us to decide whether a court in Nevada may exercise personal jurisdiction over a defendant on the basis that he knew his allegedly tortious conduct in Georgia would delay the return of funds to plaintiffs with connections to Nevada. Because the defendant had no other contacts with Nevada, and because a plaintiff’s con-tacts with the forum State cannot be “decisive in determining whether the defendant’s due process rights are violated,” Rush v. Savchuk, 444 U. S. 320, 332 (1980) , we hold that the court in Nevada may not exercise personal jurisdiction under these circumstances. I Petitioner Anthony Walden serves as a police officer for the city of Covington, Georgia. In August 2006, petitioner was working at the Atlanta Hartsfield-Jackson Airport as a deputized agent of the Drug Enforcement Administration (DEA). As part of a task force, petitioner conducted investigative stops and other law enforcement functions in support of the DEA’s airport drug interdiction program. On August 8, 2006, Transportation Security Administration agents searched respondents Gina Fiore and Keith Gipson and their carry-on bags at the San Juan airport in Puerto Rico. They found almost $97,000 in cash. Fiore explained to DEA agents in San Juan that she and Gipson had been gambling at a casino known as the El San Juan, and that they had residences in both California and Nevada (though they provided only California identification). After respondents were cleared for departure, a law enforcement official at the San Juan airport notified petitioner’s task force in Atlanta that respondents had boarded a plane for Atlanta, where they planned to catch a connecting flight to Las Vegas, Nevada. When respondents arrived in Atlanta, petitioner and another DEA agent approached them at the departure gate for their flight to Las Vegas. In response to petitioner’s questioning, Fiore explained that she and Gipson were professional gamblers. Respondents maintained that the cash they were carrying was their gambling “ ‘bank’ ” and winnings. App. 15, 24. After using a drug-sniffing dog to perform a sniff test, petitioner seized the cash.[1] Petitioner advised respondents that their funds would be returned if they later proved a legitimate source for the cash. Respondents then boarded their plane. After respondents departed, petitioner moved the cash to a secure location and the matter was forwarded to DEA headquarters. The next day, petitioner received a phone call from respondents’ attorney in Nevada seeking return of the funds. On two occasions over the next month, petitioner also received documentation from the attorney regarding the legitimacy of the funds. At some point after petitioner seized the cash, he helped draft an affidavit to show probable cause for forfeiture of the funds and forwarded that affidavit to a United States Attorney’s Office in Georgia.[2] According to respondents, the affidavit was false and misleading because petitioner misrepresented the encounter at the airport and omitted exculpatory information regarding the lack of drug evidence and the legitimate source of the funds. In the end, no forfeiture complaint was filed, and the DEA returned the funds to respondents in March 2007. Respondents filed suit against petitioner in the United States District Court for the District of Nevada, seeking money damages under Bivens v. Six Unknown Fed. Narcotics Agents, 403 U. S. 388 (1971) . Respondents alleged that petitioner violated their Fourth Amendment rights by (1) seizing the cash without probable cause; (2) keeping the money after concluding it did not come from drug-related activity; (3) drafting and forwarding a probable cause affidavit to support a forfeiture action while knowing the affidavit contained false statements; (4) willfully seeking forfeiture while withholding exculpatory informa-tion; and (5) withholding that exculpatory information from the United States Attorney’s Office. The District Court granted petitioner’s motion to dismiss. Relying on this Court’s decision in Calder v. Jones, 465 U. S. 783 (1984) , the court determined that petitioner’s search of respondents and his seizure of the cash in Georgia did not establish a basis to exercise personal jurisdiction in Nevada. The court concluded that even if petitioner caused harm to respondents in Nevada while knowing they lived in Nevada, that fact alone did not confer jurisdiction. Because the court dismissed the complaint for lack of personal jurisdiction, it did not determine whether venue was proper. On appeal, a divided panel of the United States Court of Appeals for the Ninth Circuit reversed. The Court of Appeals assumed the District Court had correctly determined that petitioner’s search and seizure in Georgia could not support exercise of jurisdiction in Nevada. The court held, however, that the District Court could properly exercise jurisdiction over “the false probable cause affidavit aspect of the case.” 688 F. 3d 558, 577 (2011). According to the Court of Appeals, petitioner “expressly aimed” his submission of the allegedly false affidavit at Nevada by submitting the affidavit with knowledge that it would affect persons with a “significant connection” to Nevada.[3] Id., at 581. After determining that the delay in returning the funds to respondents caused them “foreseeable harm” in Nevada and that the exercise of personal jurisdiction over petitioner was otherwise reasonable, the court found the District Court’s exercise of personal jurisdiction to be proper.[4] Id., at 582, 585. The Ninth Circuit denied rehearing en banc, with eight judges, in two separate opinions, dissenting. Id., at 562, 568. We granted certiorari to decide whether due process permits a Nevada court to exercise jurisdiction over petitioner. 568 U. S. ___ (2013). We hold that it does not and therefore reverse.[5] II A “Federal courts ordinarily follow state law in determining the bounds of their jurisdiction over persons.” Daimler AG v. Bauman, 571 U. S. ___, ___ (2014) (slip op., at 6). This is because a federal district court’s authority to assert personal jurisdiction in most cases is linked to service of process on a defendant “who is subject to the jurisdiction of a court of general jurisdiction in the state where the district court is located.” Fed. Rule of Civ. Proc. 4(k)(1)(A). Here, Nevada has authorized its courts to exercise jurisdiction over persons “on any basis not inconsistent with . . . the Constitution of the United States.” Nev. Rev. Stat. §14.065 (2011). Thus, in order to determine whetherthe Federal District Court in this case was authorized to exercise jurisdiction over petitioner, we ask whether the exercise of jurisdiction “comports with the limits imposed by federal due process” on the State of Nevada. Daimler, supra, at ___ (slip op., at 6). B 1 The Due Process Clause of the Fourteenth Amendment constrains a State’s authority to bind a nonresidentdefendant to a judgment of its courts. World-Wide Volkswagen Corp. v. Woodson, 444 U. S. 286, 291 (1980) . Although a nonresident’s physical presence within the territorial jurisdiction of the court is not required, the nonresident generally must have “certain minimum contacts . . . such that the maintenance of the suit does not offend ‘traditional notions of fair play and substantial justice.’ ” International Shoe Co. v. Washington, 326 U. S. 310, 316 (1945) (quoting Milliken v. Meyer, 311 U. S. 457, 463 (1940) ). This case addresses the “minimum contacts” necessary to create specific jurisdiction.[6] The inquiry whether a forum State may assert specific jurisdiction over a nonresident defendant “focuses on ‘the relationship among the defendant, the forum, and the litigation.’ ” Keeton v. Hustler Magazine, Inc., 465 U. S. 770, 775 (1984) (quoting Shaffer v. Heitner, 433 U. S. 186, 204 (1977) ). For a State to exercise jurisdiction consistent with due process, the defendant’s suit-related conduct must create a substantial connection with the forum State. Two related aspects of this necessary relationship are relevant in this case. First, the relationship must arise out of contacts that the “defendant himself” creates with the forum State. Burger King Corp. v. Rudzewicz, 471 U. S. 462, 475 (1985) . Due process limits on the State’s adjudicative authority principally protect the liberty of the nonresident defendant—not the convenience of plaintiffs or third parties. See World-Wide Volkswagen Corp., supra, at 291–292. We have consistently rejected attempts to satisfy the defendant-focused “minimum contacts” inquiry by demonstrating contacts between the plaintiff (or third parties) and the forum State. See Helicopteros Nacionales de Colombia, S. A. v. Hall, 466 U. S. 408, 417 (1984) (“[The] unilateral activity of another party or a third person is not an appropriate consideration when determining whether a defendant has sufficient contacts with a forum State to justify an assertion of jurisdiction”). We have thus rejected a plaintiff’s argument that a Florida court could exercise per-sonal jurisdiction over a trustee in Delaware based solely on the contacts of the trust’s settlor, who was domiciled in Florida and had executed powers of appointment there. Hanson v. Denckla, 357 U. S. 235 –254 (1958). We have likewise held that Oklahoma courts could not exercise personal jurisdiction over an automobile distributor that supplies New York, New Jersey, and Connecticut dealers based only on an automobile purchaser’s act of driving it on Oklahoma highways. World-Wide Volks-wagen Corp., supra, at 298. Put simply, however sig-nificant the plaintiff’s contacts with the forum may be, those contacts cannot be “decisive in determining whether the defendant’s due process rights are violated.” Rush, 444 U. S., at 332. Second, our “minimum contacts” analysis looks to the defendant’s contacts with the forum State itself, not the defendant’s contacts with persons who reside there. See, e.g., International Shoe, supra, at 319 (Due process “does not contemplate that a state may make binding a judgment in personam against an individual . . . with which the state has no contacts, ties, or relations”); Hanson, supra, at 251 (“However minimal the burden of defending in a foreign tribunal, a defendant may not be called upon to do so unless he has had the ‘minimal contacts’ with that State that are a prerequisite to its exercise of power over him”). Accordingly, we have upheld the assertion of jurisdiction over defendants who have purposefully “reach[ed] out beyond” their State and into another by, for example, entering a contractual relationship that “envisioned continuing and wide-reaching contacts” in the forum State, Burger King, supra, at 479–480, or by circulating magazines to “deliberately exploi[t]” a market in the forum State, Keeton, supra, at 781. And although physical presence in the forum is not a prerequisite to jurisdiction, Burger King, supra, at 476, physical entry into the State—either by the defendant in person or through an agent, goods, mail, or some other means—is certainly a relevant contact. See, e.g., Keeton, supra, at 773–774. But the plaintiff cannot be the only link between the defendant and the forum. Rather, it is the defendant’s conduct that must form the necessary connection with the forum State that is the basis for its jurisdiction over him. See Burger King, supra, at 478 (“If the question is whether an individual’s contract with an out-of-state party alone can automatically establish sufficient minimum contacts in the other party’s home forum, we believe the answer clearly is that it cannot”); Kulko v. Superior Court of Cal., City and County of San Francisco, 436 U. S. 84, 93 (1978) (declining to “find personal jurisdiction in a State . . . merely because [the plaintiff in a child support action] was residing there”). To be sure, a defendant’s contacts with the forum State may be intertwined with his transactions or interactions with the plaintiff or other parties. But a defendant’s relationship with a plaintiff or third party, standing alone, is an insufficient basis for jurisdiction. See Rush, supra, at 332 (“Naturally, the parties’ relationships with each other may be significant in evaluating their ties to the forum. The requirements of International Shoe, however, must be met as to each defendant over whom a state court exercises jurisdiction”). Due process requires that a defendant be haled into court in a forum State based on his own affiliation with the State, not based on the “random, fortuitous, or attenuated” contacts he makes by interacting with other persons affiliated with the State. Burger King, 471 U. S., at 475 (internal quotation marks omitted). 2 These same principles apply when intentional torts are involved. In that context, it is likewise insufficient to rely on a defendant’s “random, fortuitous, or attenuated contacts” or on the “unilateral activity” of a plaintiff. Ibid. (same). A forum State’s exercise of jurisdiction over an out-of-state intentional tortfeasor must be based on intentional conduct by the defendant that creates the necessary contacts with the forum. Calder v. Jones, 465 U. S. 783 , illustrates the application of these principles. In Calder, a California actress brought a libel suit in California state court against a reporter and an editor, both of whom worked for the National Enquirer at its headquarters in Florida. The plaintiff’s libel claims were based on an article written and edited by the defendants in Florida for publication in the National Enquirer, a national weekly newspaper with a California circulation of roughly 600,000. We held that California’s assertion of jurisdiction over the defendants was consistent with due process. Although we recognized that the defendants’ activities “focus[ed]” on the plaintiff, our jurisdictional inquiry “focuse[d] on ‘the relationship among the defendant, the forum, and the litigation.’ ” Id., at 788 (quoting Shaffer, 433 U. S., at 204). Specifically, we examined the various contacts the defendants had created with California (and not just with the plaintiff) by writing the allegedly libelous story. We found those forum contacts to be ample: The defendants relied on phone calls to “California sources” for the information in their article; they wrote the story about the plaintiff’s activities in California; they caused reputa-tional injury in California by writing an allegedly libelous article that was widely circulated in the State; and the “brunt” of that injury was suffered by the plaintiff in that State. 465 U. S., at 788–789. “In sum, California [wa]s the focal point both of the story and of the harm suffered.” Id., at 789. Jurisdiction over the defendants was “therefore proper in California based on the ‘effects’ of their Florida conduct in California.” Ibid. The crux of Calder was that the reputation-based “effects” of the alleged libel connected the defendants to California, not just to the plaintiff. The strength of that connection was largely a function of the nature of the libel tort. However scandalous a newspaper article might be, it can lead to a loss of reputation only if communicated to (and read and understood by) third persons. See Restatement (Second) of Torts §577, Comment b (1976); see also ibid. (“[R]eputation is the estimation in which one’s character is held by his neighbors or associates”). Accordingly, the reputational injury caused by the defendants’ story would not have occurred but for the fact that the defendants wrote an article for publication in California that was read by a large number of California citizens. Indeed, because publication to third persons is a necessary element of libel, see id., §558, the defendants’ intentional tort actually occurred in California. Keeton, 465 U. S., at 777 (“The tort of libel is generally held to occur wherever the offending material is circulated”). In this way, the “effects” caused by the defendants’ article—i.e., the injury to the plaintiff’s reputation in the estimation of the California public—connected the defendants’ conduct to California, not just to a plaintiff who lived there. That connection, combined with the various facts that gave the article a California focus, sufficed to authorize the California court’s exercise of jurisdiction.[7] III Applying the foregoing principles, we conclude that petitioner lacks the “minimal contacts” with Nevada that are a prerequisite to the exercise of jurisdiction over him. Hanson, 357 U. S., at 251. It is undisputed that no part of petitioner’s course of conduct occurred in Nevada. Petitioner approached, questioned, and searched respondents, and seized the cash at issue, in the Atlanta airport. It is alleged that petitioner later helped draft a “false probable cause affidavit” in Georgia and forwarded that affidavit to a United States Attorney’s Office in Georgia to support a potential action for forfeiture of the seized funds. 688 F. 3d, at 563. Petitioner never traveled to, conducted activities within, contacted anyone in, or sent anything or anyone to Nevada. In short, when viewed through the proper lens—whether the defendant’s actions connect him to the forum—petitioner formed no jurisdictionally relevant contacts with Nevada. The Court of Appeals reached a contrary conclusion by shifting the analytical focus from petitioner’s contacts with the forum to his contacts with respondents. See Rush, 444 U. S., at 332. Rather than assessing petitioner’s own contacts with Nevada, the Court of Appeals looked to petitioner’s knowledge of respondents’ “strong forum connections.” 688 F. 3d, at 577–579, 581. In the court’s view, that knowledge, combined with its conclusion that respondents suffered foreseeable harm in Nevada, satisfied the “minimum contacts” inquiry.[8] Id., at 582. This approach to the “minimum contacts” analysis impermissibly allows a plaintiff’s contacts with the defendant and forum to drive the jurisdictional analysis. Petitioner’s actions in Georgia did not create sufficient contacts with Nevada simply because he allegedly directed his conduct at plaintiffs whom he knew had Nevada connections. Such reasoning improperly attributes a plain-tiff’s forum connections to the defendant and makes those connections “decisive” in the jurisdictional analysis. See Rush, supra, at 332. It also obscures the reality that none of petitioner’s challenged conduct had anything to do with Nevada itself. Relying on Calder, respondents emphasize that they suffered the “injury” caused by petitioner’s allegedly tortious conduct (i.e., the delayed return of their gambling funds) while they were residing in the forum. Brief for Respondents 14. This emphasis is likewise misplaced. As previously noted, Calder made clear that mere injury to a forum resident is not a sufficient connection to the forum. Regardless of where a plaintiff lives or works, an injury is jurisdictionally relevant only insofar as it shows that the defendant has formed a contact with the forum State. The proper question is not where the plaintiff experienced a particular injury or effect but whether the defendant’s conduct connects him to the forum in a meaningful way. Respondents’ claimed injury does not evince a connection between petitioner and Nevada. Even if we consider the continuation of the seizure in Georgia to be a distinct injury, it is not the sort of effect that is tethered to Nevada in any meaningful way. Respondents (and only respondents) lacked access to their funds in Nevada not because anything independently occurred there, but because Nevada is where respondents chose to be at a time when they desired to use the funds seized by petitioner. Respondents would have experienced this same lack of access in California, Mississippi, or wherever else they might have traveled and found themselves wanting more money than they had. Unlike the broad publication of the forum-focused story in Calder, the effects of petitioner’s con-duct on respondents are not connected to the forum State in a way that makes those effects a proper basis forjurisdiction.[9] The Court of Appeals pointed to other possible contacts with Nevada, each ultimately unavailing. Respondents’ Nevada attorney contacted petitioner in Georgia, but that is precisely the sort of “unilateral activity” of a third party that “cannot satisfy the requirement of contact with the forum State.” Hanson, 357 U. S., at 253. Respondents allege that some of the cash seized in Georgia “originated” in Nevada, but that attenuated connection was not created by petitioner, and the cash was in Georgia, not Nevada, when petitioner seized it. Finally, the funds were eventually returned to respondents in Nevada, but petitioner had nothing to do with that return (indeed, it seems likely that it was respondents’ unilateral decision to have their funds sent to Nevada). * * * Well-established principles of personal jurisdiction are sufficient to decide this case. The proper focus of the “minimum contacts” inquiry in intentional-tort cases is “ ‘the relationship among the defendant, the forum, and the litigation.’ ” Calder, 465 U. S., at 788. And it is the defendant, not the plaintiff or third parties, who must create contacts with the forum State. In this case, the application of those principles is clear: Petitioner’s relevant conduct occurred entirely in Georgia, and the mere fact that his conduct affected plaintiffs with connectionsto the forum State does not suffice to authorize jurisdiction. We therefore reverse the judgment of the Court of Appeals. It is so ordered.Notes 1 Respondents allege that the sniff test was “at best, inconclusive,” and there is no indication in the pleadings that drugs or drug residue were ever found on or with the cash. App. 21. 2 The alleged affidavit is not in the record. Because this case comes to us at the motion-to-dismiss stage, we take respondents’ factual allegations as true, including their allegations regarding the existence and content of the affidavit. 3 The allegations in the complaint suggested to the Court of Appeals that petitioner “definitely knew, at some point the seizure but providing the alleged false probable cause affidavit, that [respondents] had a significant connection to Nevada.” 688 F. 3d, at 578. 4 Judge Ikuta dissented. In her view, the “false affidavit/forfeiture proceeding aspect” over which the majority found jurisdiction proper was not raised as a separate claim in the complaint, and she found it “doubtful that such a constitutional tort even exists.” ., at 593. After the court denied rehearing en banc, the majority explained in a postscript that it viewed the filing of the false affidavit, which effected a “continued seizure” of the funds, as a separate violation. at 588–589. Petitioner does not dispute that reading here. 5 We also granted certiorari on the question whether Nevada is a proper venue for the suit under . Because we resolve the case on jurisdictional grounds, we do not decide whether venue was proper in Nevada. 6 “Specific” or “case-linked” jurisdiction “depends on an ‘affiliatio[n] between the forum and the underlying controversy’ ” (, an “activity or an occurrence that takes place in the forum State and is therefore subject to the State’s regulation”). v. , 564 U. S. ___, ___ (2011) (slip op., at 2). This is in contrast to “general” or “all purpose” jurisdiction, which permits a court to assert jurisdiction over a defendant based on a forum connection unrelated to the underlying suit (domicile). Respondents rely on specific jurisdiction only. 7 The defendants in argued that no contacts they had with California were sufficiently purposeful because their employer was responsible for circulation of the article. See v. , . We rejected that argument. Even though the defendants did not circulate the article themselves, they “expressly aimed” “their intentional, and allegedly tortious, actions” at California be-cause they knew the National Enquirer “ha[d] its largest circulation” in California, and that the article would “have a potentially devastating impact” there. , at 789–790 8 Respondents propose a substantially similar analysis. They suggest that “a defendant creates sufficient minimum contacts with a forum when he (1) intentionally targets (2) a known resident of the forum (3) for imposition of an injury (4) to be suffered by the plaintiff while she is residing in the forum state.” Brief for Respondents 26–27. 9 Respondents warn that if we decide petitioner lacks minimum contacts in this case, it will bring about unfairness in cases where intentional torts are committed via the Internet or other electronic means ( fraudulent access of financial accounts or “phishing” schemes). As an initial matter, we reiterate that the “minimum contacts” inquiry principally protects the liberty of the nonresident defendant, not the interests of the plaintiff. v.,444 U. S., 286, 291–292 (1980). In any event, this case does not present the very different questions whether and how a defendant’s virtual “presence” and conduct translate into “contacts” with a particular State. To the contrary, there is no question where the conduct giving rise to this litigation took place: Petitioner seized physical cash from respondents in the Atlanta airport, and he later drafted and forwarded an affidavit in Georgia. We leave questions about virtual contacts for another day.

572.US.415

Respondent pleaded guilty to capital murder, capital kidnaping, and first-degree rape, the statutory aggravating circumstance for the murder. He was sentenced to death after the trial court denied defense counsel’s request to instruct the jury not to draw any adverse inference from respondent’s decision not to testify at the penalty phase. The Kentucky Supreme Court affirmed, finding that the Fifth Amendment’s requirement of a no-adverse-inference instruction to protect a nontestifying defendant at the guilt phase, see Carter v. Kentucky, 450 U.S. 288, is not required at the penalty phase. Subsequently, the Federal District Court granted respondent habeas relief, holding that the trial court’s refusal to give the requested instruction violated respondent’s privilege against self-incrimination. The Sixth Circuit affirmed. Held: Because the Kentucky Supreme Court’s rejection of respondent’s Fifth Amendment claim was not objectively unreasonable, the Sixth Circuit erred in granting the writ. Pp. 3–12. (a) The difficult-to-meet standard of 28 U. S. C. §2254(d) permits a court to grant federal habeas relief on a claim already “adjudicated on the merits in State court” only if that adjudication “resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by [this] Court.” “ ‘[C]learly established Federal law’ ” includes only “ ‘the holdings” of the Court’s decisions,’ ” Howes v. Fields, 565 U. S. ___, ___; and an “unreasonable application of” those holdings must be “ ‘objectively unreasonable,’ ” Lockyer v. Andrade, 538 U.S. 63, 75–76. The state-court ruling must rest on “an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Harrington v. Richter, 562 U. S. ___, ___. Here, the Kentucky Supreme Court’s conclusion was not “contrary to” the Court’s holdings in Carter, supra, which required a no-adverse-inference instruction at the guilt phase; in Estelle v. Smith, 451 U.S. 454, which concerned the introduction at the penalty phase of the results of an involuntary, un-Mirandized pretrial psychiatric examination; or in Mitchell v. United States, 526 U.S. 314, 327–330, which disapproved a trial judge’s drawing of an adverse inference from the defendant’s silence at sentencing “with regard to factual determinations respecting the circumstances and details of the crime.” Nor was the Kentucky Supreme Court’s conclusion an unreasonable application of the holdings in those cases. This Court need not decide whether a no-adverse-inference instruction is required in these circumstances, for the issue before the Kentucky Supreme Court was, at a minimum, not “beyond any possibility for fairminded disagreement,” Harrington, supra, at ___. Mitchell in particular leaves open the possibility that some inferences might permissibly be drawn from a defendant’s penalty-phase silence. Thus, it cannot be read to require the type of blanket no-adverse-inference instruction requested and denied here. Moreover, because respondent’s own admissions of guilt had established every relevant fact on which Kentucky bore the burden of proof, Mitchell’s narrow holding, which implied that it was limited to inferences pertaining to the facts of the crime, does not apply. Pp. 3–9. (b) Respondent contends that the state court was unreasonable in refusing to extend a governing legal principle to a context in which it should have controlled, but this Court has never adopted such a rule. Section 2254(d)(1) provides a remedy for instances in which a state court unreasonably applies this Court’s precedent; it does not require state courts to extend that precedent or license federal courts to treat the failure to do so as error. The appropriate time to consider, as a matter of first impression, whether Carter, Estelle, and Mitchell require a penalty-phase no-adverse-inference instruction would be on direct review, not in a habeas case governed by §2254(d). Pp. 9–12. 685 F.3d 574, reversed and remanded. Scalia, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Alito, and Kagan, JJ., joined. Breyer, J., filed a dissenting opinion, in which Ginsburg and Sotomayor, JJ., joined.

Respondent brutally raped, slashed with a box cutter, and drowned a 16-year-old high-school student. After pleading guilty to murder, rape, and kidnaping, he was sentenced to death. The Kentucky Supreme Court affirmed the sentence, and we denied certiorari. Ten years later, the Court of Appeals for the Sixth Circuit granted respondent’s petition for a writ of habeas corpus on his Fifth Amendment claim. In so doing, it disregarded the limitations of 28 U. S. C. §2254(d)—a provision of law that some federal judges find too confining, but that all federal judges must obey. We reverse. I On the evening of January 25, 1997, Sarah Hansen drove to a convenience store to rent a movie. When she failed to return home several hours later, her family called the police. Officers eventually found the vehicle Hansen had been driving a short distance from the convenience store. They followed a 400- to 500-foot trail of blood from the van to a nearby lake, where Hansen’s unclothed, dead body was found floating in the water. Hansen’s “throat had been slashed twice with each cut approximately 3.5 to 4 inches long,” and “[h]er windpipe was totally severed.” Woodall v. Commonwealth, 63 S. W. 3d 104, 114 (Ky. 2002). Authorities questioned respondent when they learned that he had been in the convenience store on the night of the murder. Respondent gave conflicting statements regarding his whereabouts that evening. Further investigation revealed that respondent’s “fingerprints were on the van the victim was driving,” “[b]lood was found on [respondent’s] front door,” “[b]lood on his clothing and sweatshirt was consistent with the blood of the victim,” and “DNA on . . . vaginal swabs” taken from the victim “was consistent with” respondent’s. Ibid. Faced with overwhelming evidence of his guilt, respondent pleaded guilty to capital murder. He also pleaded guilty to capital kidnaping and first-degree rape, the statutory aggravating circumstance for the murder. See App. 78; Ky. Rev. Stat. Ann. §532.025(2)(a) (West Supp. 2012). At the ensuing penalty-phase trial, respondent called character witnesses but declined to testify himself. Defense counsel asked the trial judge to instruct the jury that “[a] defendant is not compelled to testify and the fact that the defendant did not testify should not prejudice him in any way.” App. 31. The trial judge denied the request, and the Kentucky Supreme Court affirmed that denial. Woodall v. Commonwealth, supra, at 115. While recog-nizing that the Fifth Amendment requires a no-adverse-inference instruction to protect a nontestifying defendant at the guilt phase, see Carter v. Kentucky, 450 U. S. 288 (1981) , the court held that Carter and our subsequent cases did not require such an instruction here. Woodall v. Commonwealth, supra, at 115. We denied respondent’s petition for a writ of certiorari from that direct appeal. Woodall v. Kentucky, 537 U. S. 835 (2002) . In 2006, respondent filed this petition for habeas corpus in Federal District Court. The District Court granted relief, holding, as relevant here, that the trial court’s refusal to issue a no-adverse-inference instruction at the penalty phase violated respondent’s Fifth Amendment privilege against self-incrimination. Woodall v. Simpson, No. 5:06CV–P216–R (WD Ky., Feb. 24, 2009), App. to Pet. for Cert. 58a–61a, 2009 WL 464939, *12. The Court of Appeals affirmed and ordered Kentucky to either resentence respondent within 180 days or release him. Woodall v. Simpson, 685 F. 3d 574, 581 (CA6 2012).[1] Judge Cook dissented. We granted certiorari. 570 U. S. ___ (2013). II A Section 2254(d) of Title 28 provides that “[a]n application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim . . . resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States.” “This standard,” we recently reminded the Sixth Circuit, “is ‘difficult to meet.’ ” Metrish v. Lancaster, 569 U. S. ___, ___ (2013) (slip op., at 4–5). “ ‘[C]learly established Federal law’ ” for purposes of §2254(d)(1) includes only “ ‘the holdings, as opposed to the dicta, of this Court’s decisions.’ ” Howes v. Fields, 565 U. S. ___, ___ (2012) (slip op., at 4) (quoting Williams v. Taylor, 529 U. S. 362, 412 (2000) ). And an “unreasonable application of” those holdings must be “ ‘objectively unrea-sonable,’ ” not merely wrong; even “clear error” will not suffice. Lockyer v. Andrade, 538 U. S. 63 –76 (2003). Rather, “[a]s a condition for obtaining habeas corpus from a federal court, a state prisoner must show that the state court’s ruling on the claim being presented in federal court was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Harrington v. Richter, 562 U. S. ___, ___ (2011) (slip op., at 13). Both the Kentucky Supreme Court and the Court of Appeals identified as the relevant precedents in this area our decisions in Carter, Estelle v. Smith, 451 U. S. 454 (1981) , and Mitchell v. United States, 526 U. S. 314 (1999) . Carter held that a no-adverse-inference instruction is required at the guilt phase. 450 U. S., at 294–295, 300. Estelle concerned the introduction at the penalty phase of the results of an involuntary, un-Mirandized pretrial psychiatric examination. 451 U. S., at 456–457, and n. 1; id., at 461. And Mitchell disapproved a trial judge’s drawing of an adverse inference from the defendant’s silence at sentencing “with regard to factual determinations respecting the circumstances and details of the crime.” 526 U. S., at 327–330. It is clear that the Kentucky Supreme Court’s conclusion is not “contrary to” the actual holding of any of these cases. 28 U. S. C. §2254(d)(1). The Court of Appeals held, however, that the “Kentucky Supreme Court’s denial of this constitutional claim was an unreasonable application of” those cases. 685 F. 3d, at 579. In its view, “reading Carter, Estelle, and Mitchell together, the only reasonable conclusion is that” a no-adverse-inference instruction was required at the penalty phase. Ibid.[2] We need not decide here, and express no view on, whether the conclusion that a no-adverse-inference instruction was required would be correct in a case not reviewed through the lens of §2254(d)(1). For we are satisfied that the issue was, at a minimum, not “beyond any possibility for fairminded disagreement.” Harrington, supra, at ___ (slip op., at 13). We have, it is true, held that the privilege against self-incrimination applies to the penalty phase. See Estelle, supra, at 463; Mitchell, supra, at 328–329. But it is not uncommon for a constitutional rule to apply somewhat differently at the penalty phase than it does at the guilt phase. See, e.g., Bobby v. Mitts, 563 U. S. ___, ___ (2011) (per curiam) (slip op., at 4). We have “never directly held that Carter applies at a sentencing phase where the Fifth Amendment interests of the defendant are different.” United States v. Whitten, 623 F. 3d 125, 131–132, n. 4 (CA2 2010) (Livingston, J., dissenting from denial of rehearing en banc). Indeed, Mitchell itself leaves open the possibility that some inferences might permissibly be drawn from a defendant’s penalty-phase silence. In that case, the District Judge had actually drawn from the defendant’s silence an adverse inference about the drug quantity attributable to the defendant. See 526 U. S., at 317–319. We held that this ran afoul of the defendant’s “right to remain silent at sentencing.” Id., at 325, 327–328 (citing Griffin v. California, 380 U. S. 609, 614 (1965) ). But we framed our holding narrowly, in terms implying that it was limited to inferences pertaining to the facts of the crime: “We decline to adopt an exception for the sentencing phase of a crimi-nal case with regard to factual determinations respecting the circumstances and details of the crime.” Mitchell, 526 U. S., at 328 (emphasis added). “The Government retains,” we said, “the burden of proving facts relevant to the crime . . . and cannot enlist the defendant in this process at the expense of the self-incrimination privilege.” Id., at 330 (emphasis added). And Mitchell included an express reservation of direct relevance here: “Whether silence bears upon the determination of a lack of remorse, or upon acceptance of responsibility for purposes of the downward adjustment provided in §3E1.1 of the United States Sentencing Guidelines (1998), is a separate question. It is not before us, and we express no view on it.” Ibid.[3] Mitchell’s reservation is relevant here for two reasons. First, if Mitchell suggests that some actual inferences might be permissible at the penalty phase, it certainly cannot be read to require a blanket no-adverse-inference instruction at every penalty-phase trial. And it was a blanket instruction that was requested and denied in this case; respondent’s requested instruction would have informed the jury that “[a] defendant is not compelled to testify and the fact that the defendant did not testify should not prejudice him in any way.” App. 31 (emphasis added). Counsel for respondent conceded at oral argument that remorse was at issue during the penalty-phase trial, see Tr. of Oral Arg. 39; see also Brief for Respondent 18, yet the proposed instruction would have precluded the jury from considering respondent’s silence as indicative of his lack of remorse. Indeed, the trial judge declined to give the no-adverse-inference instruction precisely because he was “aware of no case law that precludes the jury from considering the defendant’s lack of expression of remorse . . . in sentencing.” App. 36. This alone suffices to establish that the Kentucky Supreme Court’s conclusion was not “objectively unreasonable.” Andrade, 538 U. S., at 76. Second, regardless of the scope of respondent’s proposed instruction, any inferences that could have been drawn from respondent’s silence would arguably fall within the class of inferences as to which Mitchell leaves the door open. Respondent pleaded guilty to all of the charges he faced, including the applicable aggravating circumstances. Thus, Kentucky could not have shifted to respondent its “burden of proving facts relevant to the crime,” 526 U. S., at 330: Respondent’s own admissions had already established every relevant fact on which Kentucky bore the burden of proof. There are reasonable arguments that the logic of Mitchell does not apply to such cases. See, e.g., United States v. Ronquillo, 508 F. 3d 744, 749 (CA5 2007) (“Mitchell is inapplicable to the sentencing decision in this case because ‘the facts of the offense’ were based entirely on Ronquillo’s admissions, not on any adverse inference . . . . Ronquillo, unlike the defendant in Mitchell, admitted all the predicate facts of his offenses”). The dissent insists that Mitchell is irrelevant because it merely declined to create an exception to the “normal rule,” supposedly established by Estelle, “that a defendant is entitled to a requested no-adverse-inference instruction” at sentencing. Post, at 5 (opinion of Breyer, J.). That argument disregards perfectly reasonable interpretations of Estelle and Mitchell and hence contravenes §2254(d)’s deferential standard of review. Estelle did not involve an adverse inference based on the defendant’s silence or a corresponding jury instruction. See 451 U. S., at 461–469. Thus, whatever Estelle said about the Fifth Amendment, its holding[4]—the only aspect of the decision relevant here—does not “requir[e]” the categorical rule the dissent ascribes to it. Carey v. Musladin, 549 U. S. 70, 76 (2006) . Likewise, fairminded jurists could conclude that Mitchell’s reservation regarding remorse and acceptance of responsibility would have served no meaningful purpose if Estelle had created an across-the-board rule against adverse inferences; we are, after all, hardly in the habit of reserving “separate question[s],” Mitchell, supra, at 330, that have already been definitively answered. In these circumstances, where the “ ‘precise contours’ ” of the right remain “ ‘unclear,’ ” state courts enjoy “broad discretion” in their adjudication of a prisoner’s claims. Lockyer, 538 U. S., at 76 (quoting Harmelin v. Michigan, 501 U. S. 957, 998 (1991) (Kennedy, J., concurring in part and in judgment)). B In arguing for a contrary result, respondent leans heavily on the notion that a state-court “ ‘determination may be set aside . . . if, under clearly established federal law, the state court was unreasonable in refusing to extend the governing legal principle to a context in which the principle should have controlled.’ ” Brief for Respondent 21 (quoting Ramdass v. Angelone, 530 U. S. 156, 166 (2000) (plurality opinion)). The Court of Appeals and District Court relied on the same proposition in sustaining respondent’s Fifth Amendment claim. See 685 F. 3d, at 579; App. to Pet. for Cert. 37a–39a, 2009 WL 464939, *4. The unreasonable-refusal-to-extend concept originated in a Fourth Circuit opinion we discussed at length in Williams, our first in-depth analysis of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). See 529 U. S., at 407–409 (citing Green v. French, 143 F. 3d 865, 869–870 (1998)). We described the Fourth Circuit’s interpretation of §2254(d)(1)’s “unreasonable application” clause as “generally correct,” 529 U. S., at 407, and approved its conclusion that “a state-court decision involves an unreasonable application of this Court’s precedent if the state court identifies the correct governing legal rule . . . but unreasonably applies it to the facts of the particular state prisoner’s case,” id., at 407–408 (citing Green, supra, at 869–870). But we took no position on the Fourth Circuit’s further conclusion that a state court commits AEDPA error if it “unreasonably refuse[s] to extend a legal principle to a new context where it should apply.” 529 U. S., at 408–409 (citing Green, supra, at 869–870). We chose not “to decide how such ‘extension of legal principle’ cases should be treated under §2254(d)(1)” because the Fourth Circuit’s proposed rule for resolving them presented several “problems of precision.” 529 U. S., at 408–409. Two months later, a plurality paraphrased and applied the unreasonable-refusal-to-extend concept in Ramdass. See 530 U. S., at 166–170. It did not, however, grant the habeas petitioner relief on that basis, finding that there was no unreasonable refusal to extend. Moreover, Justice O’Connor, whose vote was necessary to form a majority, cited Williams and made no mention of the unreasonable-refusal-to-extend concept in her separate opinion concurring in the judgment. See 530 U. S., at 178–181. Ramdass therefore did not alter the interpretation of §2254(d)(1) set forth in Williams. Aside from one opinion criticizing the unreasonable-refusal-to-extend doctrine, see Yarborough v. Alvarado, 541 U. S. 652, 666 (2004) , we have not revisited the issue since Williams and Ramdass. During that same 14-year stretch, however, we have repeatedly restated our “hold[ing]” in Williams, supra, at 409, that a state-court decision is an unreasonable application of our clearly established precedent if it correctly identifies the governing legal rule but applies that rule unreasonably to the facts of a particular prisoner’s case, see, e.g., Cullen v. Pinholster, 563 U. S. ___, ___ (2011) (slip op., at 10); Rompilla v. Beard, 545 U. S. 374, 380 (2005) ; Yarborough, supra, at 663; Penry v. Johnson, 532 U. S. 782, 792 (2001) . Thus, this Court has never adopted the unreasonable-refusal-to-extend rule on which respondent relies. It has not been so much as endorsed in a majority opinion, let alone relied on as a basis for granting habeas relief. To the extent the unreasonable-refusal-to-extend rule differs from the one embraced in Williams and reiterated many times since, we reject it. Section 2254(d)(1) provides a remedy for instances in which a state court unreasonably applies this Court’s precedent; it does not require state courts to extend that precedent or license federal courtsto treat the failure to do so as error. See Scheidegger, Ha-beas Corpus, Relitigation, and the Legislative Power, 98 Colum. L. Rev. 888, 949 (1998). Thus, “if a habeas court must extend a rationale before it can apply to the facts at hand,” then by definition the rationale was not “clearly established at the time of the state-court decision.” Yarborough, 541 U. S., at 666. AEDPA’s carefully constructed framework “would be undermined if habeas courts introduced rules not clearly established under the guise of extensions to existing law.” Ibid. This is not to say that §2254(d)(1) requires an “ ‘identical factual pattern before a legal rule must be applied.’ ” Panetti v. Quarterman, 551 U. S. 930, 953 (2007) . To the contrary, state courts must reasonably apply the rules “squarely established” by this Court’s holdings to the facts of each case. Knowles v. Mirzayance, 556 U. S. 111, 122 (2009) . “[T]he difference between applying a rule and extending it is not always clear,” but “[c]ertain principles are fundamental enough that when new factual permu-tations arise, the necessity to apply the earlier rule willbe beyond doubt.” Yarborough, supra, at 666. The crit-ical point is that relief is available under §2254(d)(1)’s unreasonable-application clause if, and only if, it is so obvious that a clearly established rule applies to a given set of facts that there could be no “fairminded disagreement” on the question, Harrington, 562 U. S., at ___ (slip op., at 13). Perhaps the logical next step from Carter, Estelle, and Mitchell would be to hold that the Fifth Amendment requires a penalty-phase no-adverse-inference instruction in a case like this one; perhaps not. Either way, we have not yet taken that step, and there are reasonable arguments on both sides—which is all Kentucky needs to prevail in this AEDPA case. The appropriate time to consider the question as a matter of first impression would be on direct review, not in a habeas case governed by §2254(d)(1). * * * Because the Kentucky Supreme Court’s rejection of respondent’s Fifth Amendment claim was not objectively unreasonable, the Sixth Circuit erred in granting the writ. We therefore need not reach its further holding that the trial court’s putative error was not harmless. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The Court of Appeals did not reach the alternative ground for the District Court’s decision: respondent’s claim based on v. , . See 685 F. 3d, at 577–578. That claim is not before us here. 2 The Court of Appeals also based its conclusion that respondent “was entitled to receive a no adverse inference instruction” on one of its own cases, v. , 751 F. 2d 858, 863–864 (CA6 1985). 685 F. 3d, at 579 (internal quotation marks omitted). That was improper. As we cautioned the Sixth Circuit two Terms ago, a lower court may not “consul[t] its own precedents, rather than those of this Court, in assessing” a habeas claim governed by §2254. v. , 567 U. S. ___, ___ (2012) () (slip op., at 12). 3 The Courts of Appeals have recognized that left this unresolved; their diverging approaches to the question illustrate the possibility of fairminded disagreement. Compare v. , 597 F. 3d 608, 629–630 (CA4 2010) (direct appeal) (noting that “reserved the question of whether silence bears upon lack of remorse,” but reasoning that “and together suggest that the may well prohibit considering a defendant’s silence regarding the nonstatutory aggravating factor of lack of remorse”), with v. , 546 F. 3d 828, 832 (CA7 2008) (habeas) (while the right to remain silent persists at sentencing, “silence can be consistent not only with exercising one’s constitutional right, but also with a lack of remorse,” which “is properly considered at sentencing” (citing , 526 U. S., at 326–327)); v. , 451 F. 3d 598, 605, n. 3 (CA10 2006) (habeas) (“[T]he circuit courts have readily confined to its stated holding, and have allowed sentencing courts to rely on, or draw inferences from, a defendant’s exercise of his rights for purposes other than determining the facts of the offense of conviction”). 4 The dissent says “held that ‘so far as the protection of the is concerned,’ it could ‘discern no basis to distinguish between the guilt and penalty phases of a defendant’s ‘capital murder trial.’ ” at 2 (quoting , 451 U. S., at 462–463). Of course, it did not “hold” that. Rather, it held that the defendant’s “rights were abridged by the State’s introduction of ” a pretrial psychiatric evaluation that was administered without the preliminary warning required by v. , . 451 U. S., at 473. In any event, even ’s dictum did not assume an entitlement to a blanket no-adverse-inference instruction. The quoted language is reasonably read as referring to theof the privilege at sentencing rather than thepreciseofthat privilege when applied in the sentencing context. Indeed, it appears in a passage responding to the State’s argument that the defendant “was not entitled to the protection of the ” in the first place. at 462.

572.US.744

While campaigning for a second term, President George W. Bush was scheduled to spend the night at a Jacksonville, Oregon, cottage. Local law enforcement officials permitted a group of Bush supporters and a group of protesters to assemble on opposite sides of a street along the President’s motorcade route. When the President made a last-minute decision to have dinner at the outdoor patio area of the Jacksonville Inn’s restaurant before resuming the drive to the cottage, the protesters moved to an area in front of the Inn, which placed them within weapons range of the President. The supporters remained in their original location, where a two-story building blocked sight of, and weapons access to, the patio. At the direction of two Secret Service agents responsible for the President’s security, petitioners here (the agents), local police cleared the area where the protesters had gathered, eventually moving them two blocks away to a street beyond weapons reach of the President. The agents did not require the guests already inside the Inn to leave, stay clear of the patio, or go through a security screening. After the President dined, his motorcade passed the supporters, but the protesters, now two blocks from the motorcade’s route, were beyond his sight and hearing. The protesters sued the agents for damages, alleging that the agents engaged in viewpoint discrimination in violation of the First Amendment when they moved the protesters away from the Inn but allowed the supporters to remain in their original location. The District Court denied the agents’ motion to dismiss the suit for failure to state a claim and on qualified immunity grounds, but on interlocutory appeal, the Ninth Circuit reversed. The court held that the protesters had failed to state a First Amendment claim under the plead-ing standards of Bell Atlantic Corp. v. Twombly, 550 U.S. 544, and Ashcroft v. Iqbal, 556 U.S. 662. Because those decisions were rendered after the protesters commenced suit, the Court of Appeals granted leave to amend the complaint. On remand, the protesters supplemented the complaint with allegations that the agents acted pursuant to an unwritten Secret Service policy of working with the Bush White House to inhibit the expression of disfavored views at presidential appearances. The District Court denied the agents’ renewed motion to dismiss. This time, the Ninth Circuit affirmed, concluding that viewpoint-driven conduct on the agents’ part could be inferred from the absence of a legitimate security rationale for the different treatment accorded the two groups of demonstrators. The Court of Appeals further held that the agents were not entitled to qualified immunity because this Court’s precedent made clear that the Government may not regulate speech based on its content. Held: The agents are entitled to qualified immunity. Pp. 11–18. (a) Government officials may not exclude from public places persons engaged in peaceful expressive activity solely because the government actor fears, dislikes, or disagrees with the views expressed. See, e.g., Police Dept. of Chicago v. Mosley, 408 U.S. 92, 96. The fundamental right to speak, however, does not leave people at liberty to publicize their views “ ‘whenever and however and wherever they please.’ ” United States v. Grace, 461 U.S. 171, 177. In deciding whether the protesters have alleged violation of a clearly established First Amendment right, this Court assumes without deciding that Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, which involved alleged Fourth Amendment violations, extends to First Amendment claims, see, e.g., Iqbal, 556 U. S., at 675. The doctrine of qualified immunity protects government officials from liability for civil damages “unless a plaintiff pleads facts showing (1) that the official violated a statutory or constitutional right, and (2) that the right was ‘clearly established’ at the time of the challenged conduct.” Ashcroft v. al-Kidd, 563 U. S. ___, ___. The “dispositive inquiry . . . is whether it would [have been] clear to a reasonable officer” in the agents’ position “that [their] conduct was unlawful in the situation [they] confronted.” Saucier v. Katz, 533 U.S. 194, 202. At the time of the Jacksonville incident, this Court had addressed a constitutional challenge to Secret Service actions only once. In Hunter v. Bryant, 502 U.S. 224, the plaintiff challenged the lawfulness of his arrest by two Secret Service agents for writing and delivering a letter about a plot to assassinate President Reagan. Holding that the agents were shielded by qualified immunity, the Court stated that “accommodation for reasonable error . . . is nowhere more important than when the specter of Presidential assassination is raised.” Id., at 229. This Court has recognized the overwhelming importance of safeguarding the President in other contexts as well. See Watts v. United States, 394 U.S. 705, 707. Mindful that officers may be faced with unanticipated security situations, the key question addressed is whether it should have been clear to the agents that the security perimeter they established violated the First Amendment. Pp. 11–13. (b) The protesters assert, and the Ninth Circuit agreed, that the agents violated clearly established federal law by denying them “equal access to the President.” No decision of which the Court is aware, however, would alert Secret Service agents engaged in crowd control that they bear a First Amendment obligation to make sure that groups with conflicting views are at all times in equivalent positions. Nor would the maintenance of equal access make sense in the situation the agents here confronted, where only the protesters, not the supporters, had a direct line of sight to the patio where the President was dining. The protesters suggest that the agents could have moved the supporters out of the motorcade’s range as well, but there would have been no security rationale for such a move. Pp. 13–15. (c) The protesters allege that, in directing their displacement, the agents acted not to ensure the President’s safety, but to insulate the President from their message. These allegations are undermined by a map of the area, which shows that, because of the protesters’ location, they posed a potential security risk to the President, while the supporters, because of their location, did not. The protesters’ counterarguments are unavailing. They urge that, had the agents’ professed interest in the President’s safety been sincere, the agents would have screened or removed from the premises persons already at the Inn when the President arrived. But staff, other diners, and Inn guests were on the premises before the agents knew of the President’s plans, and thus could not have anticipated seeing the President, no less causing harm to him. The agents also could keep a close watch on the relatively small number of people already inside the Inn, surveillance that would have been impossible for the hundreds of people outside the Inn. A White House manual directs the President’s advance team to “work with the Secret Service . . . to designate a protest area . . . preferably not in view of the event site or motorcade route.” The manual guides the conduct of the political advance team, not the Secret Service, whose own written guides explicitly prohibit “agents from discriminating between anti-government and pro-government demonstrators.” Even assuming, as the protesters maintain, that other agents, at other times and places, have assisted in shielding the President from political speech, this case is scarcely one in which the agents lacked a valid security reason for their ac-tions. Moreover, because individual government officials “cannot be held liable” in a Bivens suit “unless they themselves acted [unconstitutionally],” Iqbal, 556 U. S., at 683, this Court declines to infer from alleged instances of misconduct on the part of particular agents an unwritten Secret Service policy to suppress disfavored expression, and then attribute that supposed policy to all field-level operatives. Pp. 15–18. 711 F.3d 941, reversed. Ginsburg, J., delivered the opinion for a unanimous Court.

This case concerns a charge that two Secret Service agents, in carrying out their responsibility to protect the President, engaged in unconstitutional viewpoint-based discrimination. The episode in suit occurred in Jacksonville, Oregon, on the evening of October 14, 2004. President George W. Bush, campaigning in the area for a second term, was scheduled to spend the evening at a cottage in Jacksonville. With permission from local law enforcement officials, two groups assembled on opposite sides of the street on which the President’s motorcade was to travel to reach the cottage. One group supported the President, the other opposed him. The President made a last-minute decision to stop in town for dinner before completing the drive to the cottage. His motorcade therefore turned from the planned route and proceeded to the outdoor patio dining area of the Jacksonville Inn’s restaurant. Learning of the route change, the protesters moved down the sidewalk to the area in front of the Inn. The President’s supporters remained across the street and about a half block away from the Inn. At the direction of the Secret Service agents, state and local police cleared the block on which the Inn was located and moved the protesters some two blocks away to a street beyond handgun or explosive reach of the President. The move placed the protesters a block farther away from the Inn than the supporters. Officials are sheltered from suit, under a doctrine known as qualified immunity, when their conduct “does not violate clearly established . . . constitutional rights” a reasonable official, similarly situated, would have comprehended. Harlow v. Fitzgerald, 457 U. S. 800, 818 (1982) . The First Amendment, our precedent makes plain, disfavors viewpoint-based discrimination. See Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819, 828 (1995) . But safeguarding the President is also of overwhelming importance in our constitutional system. See Watts v. United States, 394 U. S. 705, 707 (1969) (per curiam). Faced with the President’s sudden decision to stop for dinner, the Secret Service agents had to cope with a security situation not earlier anticipated. No decision of this Court so much as hinted that their on-the-spot action was unlawful because they failed to keep the protesters and supporters, throughout the episode, equidistant from the President. The United States Court of Appeals for the Ninth Circuit ruled otherwise. It found dispositive of the agents’ motion to dismiss “the considerable disparity in the distance each group was allowed to stand from the Presiden[t].” Moss v. United States Secret Serv., 711 F. 3d 941, 946 (2013). Because no “clearly established law” so controlled the agents’ response to the motorcade’s detour, we reverse the Ninth Circuit’s judgment. I A On October 14, 2004, after a nearby campaign appear-ance, President George W. Bush was scheduled to spend the night at a cottage in Jacksonville, Oregon. Anticipating the visit, a group of individuals, including respondents (the protesters), organized a demonstration to express their opposition to the President and his policies. At around 6:00 p.m. on the evening the President’s motorcade was expected to pass through the town, between 200 and 300 protesters gathered in Jacksonville, on California Street between Third and Fourth Streets. See infra, at 4 (map depicting the relevant area in Jacksonville). The gathering had been precleared with local law enforcement authorities. On the opposite side of Third Street, a similarly sized group of individuals (the supporters) assembled to show their support for the President. If, as planned, the motorcade had traveled down Third Street to reach the cottage, with no stops along the way, the protesters and supporters would have had equal access to the President throughout in delivering their respective messages. This situation was unsettled when President Bush made a spur-of-the-moment decision to stop for dinner at the Jacksonville Inn before proceeding to the cottage. The Inn stands on the north side of California Street, on the block where the protesters had assembled. Learning of the President’s change in plans, the protesters moved along the block to face the Inn. The respective positions of the protesters and supporters at the time the President arrived at the Inn are shown on the following map, which the protesters attached as an exhibit to their complaint:[1] As the map indicates, the protesters massed on the sidewalk directly in front of the Inn, while the supporters remained assembled on the block west of Third Street, some distance from the Inn. The map also shows an alley running along the east side of the Inn (the California Street alley) leading to an outdoor patio used by the Inn’s restaurant as a dining area. A six-foot high wooden fence surrounded the patio. At the location where the President’s supporters gathered, a large two-story building, the U. S. Hotel, extended north around the corner of California and Third Streets. That structure blocked sight of, and weapons access to, the patio from points on California Street west of the Inn. Petitioners are two Secret Service agents (the agents) responsible for the President’s security during the Jacksonville visit. Shortly after 7:00 p.m. on the evening in question, the agents enlisted the aid of local police officers to secure the area for the President’s unexpected stop at the Inn. Following the agents’ instructions, the local officers first cleared the alley running from Third Street to the patio (the Third Street alley), which the President’s motorcade would use to access the Inn. The officers then cleared Third Street north of California Street, as well as the California Street alley. At around 7:15 p.m., the President arrived at the Inn. As the motorcade entered the Third Street alley, both sets of demonstrators were equally within the President’s sight and hearing. When the President reached the outdoor patio dining area, the protesters stood on the sidewalk directly in front of the California Street alley, exhibiting signs and chanting slogans critical of the President and his policies. In view of the short distance between California Street and the patio, the protesters no longer contest that they were then within weapons range of the President. See Tr. of Oral Arg. 3–4, 35, 39–40; Brief for Petitioners 44. Approximately 15 minutes later, the agents directed the officers to clear the protesters from the block in front of the Inn and move them to the east side of Fourth Street. From their new location, the protesters were roughly the same distance from the President as the supporters. But unlike the supporters, whose sight and access were obstructed by the U. S. Hotel, only a parking lot separated the protesters from the patio. The protesters thus remained within weapons range of, and had a direct line of sight to, the President’s location. This sight line is illustrated by the broken arrow marked on the map below.[2] After another 15 minutes passed, the agents directed the officers again to move the protesters, this time one block farther away from the Inn, to the east side of Fifth Street. The relocation was necessary, the agents told the local officers, to ensure that no demonstrator would be “within handgun or explosive range of the President.” App. to Pet. for Cert. 177a. The agents, however, did not require the guests already inside the Inn to leave, stay clear of the patio, or go through any security screening. The supporters at all times retained their original location on the west side of Third Street. After the President dined, the motorcade left the Inn by traveling south on Third Street toward the cottage. On its way, the motorcade passed the President’s supporters. The protesters remained on Fifth Street, two blocks away from the motorcade’s route, thus beyond the President’s sight and hearing. B The protesters sued the agents for damages in the U. S. District Court for the District of Oregon. The agents’ actions, the complaint asserted, violated the protesters’ First Amendment rights by the manner in which the agents established a security perimeter around the President during his unscheduled stop for dinner. See Bivens v. Six Unknown Fed. Narcotics Agents, 403 U. S. 388 (1971) (recognizing claim for damages against federal agents for violations of plaintiff’s Fourth Amendment rights).[3] Specifically, the protesters alleged that the agents engaged in viewpoint discrimination when they moved the protesters away from the Inn, while allowing the supporters to remain in their original location. The agents moved to dismiss the complaint on the ground that the protesters’ allegations were insufficient to state a claim for violation of the First Amendment. The agents further maintained that they were sheltered by qualified immunity because the constitutional right alleged by the protesters was not clearly established. The District Court denied the motion, see Moss v. United States Secret Serv., 2007 WL 2915608, *1, 20 (D Ore., Oct. 7, 2007), but on interlocutory appeal,[4] the U. S. Court of Appeals for the Ninth Circuit reversed. See Moss v. United States Secret Serv., 572 F. 3d 962 (2009). The facts alleged in the complaint, the Court of Appeals held, were insufficient to state a First Amendment claim under the pleading standards prescribed in Bell Atlantic Corp. v. Twombly, 550 U. S. 544 (2007) , and Ashcroft v. Iqbal, 556 U. S. 662 (2009) . 572 F. 3d, at 974–975.[5] Because Twombly and Iqbal were decided after the protesters filed their complaint, however, the Ninth Circuit instructed the District Court to grant the protesters leave to amend. 572 F. 3d, at 972. On remand, the protesters supplemented their complaint with allegations that the agents acted pursuant to an “actual but unwritten” Secret Service policy of “work[ing] with the White House under President Bush to eliminate dissent and protest from presidential appearances.” App. to Pet. for Cert. 184a. Relying on published media reports, the protesters’ amended complaint cited several instances in which other Secret Service agents allegedly engaged in conduct designed to suppress expression critical of President Bush at his public appearances. The amended complaint also included an excerpt from a White House manual instructing the President’s advance team to “work with the Secret Service and have them ask the local police department to designate a protest area where demonstrators can be placed; preferably not in view of the event site or motorcade route.” Id., at 219a. See also id., at 183a. The agents renewed their motion to dismiss the suit for failure to state a claim and on qualified immunity grounds. The District Court denied the motion, holding that the complaint adequately alleged a violation of the First Amendment, and that the constitutional right asserted was clearly established. Moss v. United States Secret Serv., 750 F. Supp. 2d 1197, 1216–1228 (Ore. 2010). The agents again sought an interlocutory appeal. This time, the Ninth Circuit affirmed, 711 F. 3d 941, satisfied that the amended pleading plausibly alleged that the agents “sought to suppress [the protesters’] political speech” based on the viewpoint they expressed, id., at 958. Viewpoint-driven conduct, the Court of Appeals maintained, could be inferred from the absence of a legitimate security rationale for “the differential treatment” accorded the two groups of demonstrators. See id., at 946. The Court of Appeals further held that the agents were not entitled to qualified immunity because this Court’s precedent “make[s] clear . . . ‘that the government may not regulate speech based on its substantive content or the message it conveys.’ ” Id., at 963 (quoting Rosenberger, 515 U. S., at 828). The agents petitioned for rehearing and rehearing en banc, urging that the panel erred in finding the alleged constitutional violation clearly established. Over the dissent of eight judges, the Ninth Circuit denied the en banc petition. See 711 F. 3d, at 947 (O’Scannlain, J., dissenting from denial of rehearing en banc). We granted certiorari. 571 U. S. ___ (2013). II A It is uncontested and uncontestable that government officials may not exclude from public places persons engaged in peaceful expressive activity solely because the government actor fears, dislikes, or disagrees with the views those persons express. See, e.g., Police Dept. of Chicago v. Mosley, 408 U. S. 92, 96 (1972) . It is equally plain that the fundamental right to speak secured by the First Amendment does not leave people at liberty to pub-licize their views “ ‘whenever and however and wher-ever they please.’ ” United States v. Grace, 461 U. S. 171 –178 (1983) (quoting Adderly v. Florida, 385 U. S.39, 48 (1966)). Our decision in this case starts from those premises. The particular question before us is whether the protesters have alleged violation of a clearly established First Amendment right based on the agents’ decision to order the protesters moved from their original location in front of the Inn, first to the block just east of the Inn, and then another block farther. We note, initially, an antecedent issue: Does the First Amendment give rise to an implied right of action for damages against federal officers who violate that Amendment’s guarantees? In Bivens, cited supra, at 8, we recognized an implied right of action against federal officers for violations of the Fourth Amendment. Thereafter, we have several times assumed without deciding that Bivens extends to First Amendment claims. See, e.g., Iqbal, 556 U. S., at 675. We do so again in this case. See Tr. of Oral Arg. 10–11 (counsel for petitioners observed that the implication of a right to sue derived from the First Amendment itself was an issue “not preserved below” and therefore “not presented” in this Court). The doctrine of qualified immunity protects government officials from liability for civil damages “unless a plaintiff pleads facts showing (1) that the official violated a statutory or constitutional right, and (2) that the right was ‘clearly established’ at the time of the challenged conduct.” Ashcroft v. al-Kidd, 563 U. S. ___, ___ (2011) (slip op., at 3). And under the governing pleading standard, the “complaint must contain sufficient factual matter, accepted as true, to state a claim to relief that is plausible on its face.” Iqbal, 556 U. S., at 678 (internal quotation marks omitted). Requiring the alleged violation of law to be “clearly established” “balances . . . the need to hold public officials accountable when they exercise power irresponsibly and the need to shield officials from harassment, distraction, and liability when they perform their duties reasonably.” Pearson v. Callahan, 555 U. S. 223, 231 (2009) . The “dispositive inquiry,” we have said, “is whether it would [have been] clear to a reasonable officer” in the agents’ position “that [their] conduct was unlawful in the situation [they] confronted.” Saucier v. Katz, 533 U. S. 194, 202 (2001) . At the time of the Jacksonville incident, this Court had addressed a constitutional challenge to Secret Service actions on only one occasion.[6] In Hunter v. Bryant, 502 U. S. 224 (1991) (per curiam), the plaintiff sued two Secret Service agents alleging that they arrested him without probable cause for writing and delivering to two Univer-sity of Southern California offices a letter referring to a plot to assassinate President Ronald Reagan. We held that qualified immunity shielded the agents from claims that the arrest violated the plaintiff’s rights under the Fourth, Fifth, Sixth, and Fourteenth Amendments. “[N]owhere,” we stated, is “accommodation for reasonable error . . . more important than when the specter of Presidential assassination is raised.” Id., at 229. In other contexts, we have similarly recognized the Nation’s “valid, even . . . overwhelming, interest in protecting the safety of its Chief Executive.” Watts, 394 U. S., at 707. See also Rubin v. United States, 525 U. S. 990 –991 (1998) (Breyer, J., dissenting from denial of certio-rari) (“The physical security of the President of the United States has a special legal role to play in our constitutional system.”). Mindful that “[o]fficers assigned to protect public officials must make singularly swift, on the spot, decisions whether the safety of the person they are guarding is in jeopardy,” Reichle v. Howards, 566 U. S. ___, ___ (2012) (Ginsburg, J., concurring in judgment) (slip op., at 2), we address the key question: Should it have been clear to the agents that the security perimeter they established violated the First Amendment? B The protesters assert that it violated clearly established First Amendment law to deny them “equal access to the President,” App. Pet. for Cert. 175a, during his dinner at the Inn and subsequent drive to the cottage, id., at 185a.[7] The Court of Appeals agreed, holding that the agents violated clearly established law by moving the protesters to a location that “was in relevant ways not comparable to the place where the pro-Bush group was allowed to remain.” 711 F. 3d, at 946 (internal quotation marks and ellipsis omitted). The Ninth Circuit did not deny that security concerns justified “mov[ing] the anti-Bush pro-testers somewhere.” Ibid. But, the court determined, no reason was shown for “the considerable disparity in the distance each group was allowed to stand from the Presidential party.” Ibid. The agents thus offended the First Amendment, in the Court of Appeals’ view, because their directions to the local officers placed the protesters at a “comparativ[e] disadvantag[e] in expressing their views” to the President. Ibid. No decision of which we are aware, however, would alert Secret Service agents engaged in crowd control that they bear a First Amendment obligation “to ensure that groups with different viewpoints are at comparable locations at all times.” Id., at 952 (O’Scannlain, J., dissenting from denial of rehearing en banc). Nor would the maintenance of equal access make sense in the situation the agents confronted. Recall that at the protesters’ location on the north side of California Street, see supra, at 4, they faced an alley giving them a direct line of sight to the outdoor patio where the President stopped to dine. The first move, to the corner of Fourth and California Streets, proved no solution, for there, only a parking lot stood between the protesters and the patio. True, at both locations, a six-foot wooden fence and an unspecified number of local police officers impeded access to the President. Even so, 200 to 300 protesters were within weapons range, and had a largely unobstructed view, of the President’s location. See Tr. of Oral Arg. 41 (counsel for respondents acknowledged that “in hindsight, you could . . . conclude” that “proximity [of the protesters to the President] alone . . . is enough to create a security [risk]”). See also Eggen & Fletcher, FBI: Grenade Was a Threat to Bush, Washington Post, May 19, 2005, p. A1 (reporting that a live grenade thrown at President Bush in 2005, had it detonated, could have injured him from 100 feet away). The protesters suggest that the agents could have moved the President’s supporters further to the west so that they would not be in range of the President when the motorcade drove from the Inn to the cottage where the President would stay overnight. See App. Pet. for Cert. 178a. As earlier explained, however, see supra, at 4–5, there would have been no security rationale for such a move. In contrast to the open alley and parking lot on the east side of the Inn, to the west of the Inn where the supporters stood, a large, two-story building blocked sight of, or weapons access to, the patio the agents endeavored to secure.[8] No clearly established law, we agree, required the Secret Service “to interfere with even more speech than security concerns would require in an attempt to keep opposing groups at roughly equal distances from the President.” Brief for Petitioners 32. And surely no such law required the agents to attempt to maintain equal dis-tances by “prevail[ing] upon the President not to dine at the Inn.” Oral Arg. Audio in No. 10–36152 (CA9) 42:22 to 43:36 (argument by protesters’ counsel), available at http://www.ca9.uscourts.gov/media/view.php?pk_id=0000008129.(as visited May 19, 2014, and in Clerk of Court’s case file) (argument tendered by protesters’ counsel). III The protesters allege that, when the agents directed their displacement, the agents acted not to ensure the President’s safety from handguns or explosive devices. Instead, the protesters urge, the agents had them moved solely to insulate the President from their message, thereby giving the President’s supporters greater visibility and audibility. See Tr. of Oral Arg. 35–36. The Ninth Circuit found sufficient the protesters’ allegations that the agents “acted with the sole intent to discriminate against [the protesters] because of their viewpoint”. 711 F. 3d, at 964. Accordingly, the Court of Appeals “allow[ed] the protestors’ claim of viewpoint discrimination to proceed.” Id., at 962. It may be, the agents acknowledged, that clearly established law proscribed the Secret Service from disadvantaging one group of speakers in comparison to another if the agents had “no objectively reasonable security rationale” for their conduct, but acted solely to inhibit the expression of disfavored views. See Tr. of Oral Arg. 28–29; Brief for Petitioners 52 (entitlement to relief might have been established if, for example, “the pro-Bush group had . . . been allowed to move into the nearer location that the anti-Bush had vacated”). We agree with the agents, however, that the map itself, reproduced supra, at 4, undermines the protesters’ allegations of viewpoint discrimination as the sole reason for the agents’ directions. The map corroborates that, because of their location, the protesters posed a potential security risk to the President, while the supporters, because of their location, did not. The protesters make three arguments to shore up their charge that the agents’ asserted security concerns are disingenuous. First, the protesters urge that, had the agents’ professed interest in the President’s safety been sincere, the agents would have directed all persons pres-ent at the Inn to be screened or removed from the prem-ises. See Brief for Respondents 27. But staff, other diners, and Inn guests were there even before the agents themselves knew that the President would dine at the Inn. See Brief for Petitioners 47. Those already at the Inn “could not have had any expectation that they would see the President that evening or any opportunity to premeditate a plan to cause him harm.” Reply Brief 16. The Secret Service, moreover, could take measures to ensure that the relatively small number of people already inside the Inn were kept under close watch; no similar surveillance would have been possible for 200 to 300 people congregating in front of the Inn. See ibid. The protesters also point to a White House manual, which states that the President’s advance team should “work with the Secret Service . . . to designate a protest area . . . preferably not in view of the event site or motorcade route.” App. to Pet. for Cert. 219a. This manual guides the conduct of the President’s political advance team. See id., at 220a (distinguishing between the political role of the advance team and the security mission of the Secret Service).[9] As the complaint acknowledges, the Secret Service has its own “written guidelines, directives, instructions and rules.” Id., at 184a. Those guides explicitly “prohibit Secret Service agents from discriminating between anti-government and pro-government demonstrators.” Ibid. The protesters maintain that the Secret Service does not adhere to its own written guides. They recite several instances in which Secret Service agents allegedly engaged in viewpoint discrimination. See id., at 189a–194a. Even accepting as true the submission that Secret Service agents, at times, have assisted in shielding the President from political speech, this case is scarcely one in which the agents acted “without a valid security reason.” Brief for Respondents 40. We emphasize, again, that the protesters were at least as close to the President as were the supporters when the motorcade arrived at the Jacksonville Inn. See supra, at 5. And as the map attached to the complaint shows, see supra, at 4, when the President reached the patio to dine, the protesters, but not the supporters, were within weapons range of his location. See supra, at 14. Given that situation, the protesters cannot plausibly urge that the agents “had no valid security reason to request or order the[ir] eviction.” App. to Pet. for Cert. 186a. We note, moreover, that individual government officials “cannot be held liable” in a Bivens suit “unless they themselves acted [unconstitutionally].” Iqbal, 556 U. S., at 683. We therefore decline to infer from alleged instances of misconduct on the part of particular agents an unwritten policy of the Secret Service to suppress disfavored expression, and then to attribute that supposed policy to all field-level operatives. See Reply Brief 20. * * * This case comes to us on the agents’ petition to review the Ninth Circuit’s denial of their qualified immunity defense. See Tr. of Oral Arg. 10 (petitioners’ briefing on appeal trained on the issue of qualified immunity). Limiting our decision to that question, we hold, for the reasons stated, that the agents are entitled to qualified immunity. Accordingly, we reverse the judgment of the Court of Appeals. It is so ordered.Notes 1 App. to Brief for Petitioners (Diagram A). 2 This map appears as an appendix to the agents’ opening brief. See App. to Brief for Petitioners (Diagram B). Except for the arrow, Diagram B is identical to the map included in the protesters’ complaint. 3 The protesters’ complaint also asserted claims against local police officers for using excessive force in violation of the . Those claims were dismissed for failure to state a claim, see v. ., 711 F. 3d 941, 954 (CA9 2013), and are not at issue here. 4 We have repeatedly “stressed the importance of resolving immunity questions at the earliest possible stage [of the] litigation,” v., (). 5 In ruling on a motion to dismiss, we have instructed, courts “must take all of the factual allegations in the complaint as true,” but “are not bound to accept as true a legal conclusion couched as a factual allegation.” v. (internal quotation marks omitted). 6 Subsequent to the incident at issue here, we held in v. , 566 U. S. ___, ___ (2012) (slip op., at 1), that two Secret Service agents were “immune from suit for allegedly arresting a suspect in retaliation for [negative comments he made about Vice President Cheney], when the agents had probable cause to arrest the suspect for committing a federal crime.” 7 The protesters, however, do not maintain that “the entitled them to be returned to their original location after the President’s dinner and before his motorcade departed.” Brief for Respondents 39–40, n. 7. They urge only that “it was constitutionally improper to move them in the first place.” , at 40, n. 7; see Tr. of Oral Arg. 50 (same). 8 Neither side contends that the presence of demonstrators along the President’s motorcade route posed an unmanageable security risk, or that there would have been a legitimate security rationale for removing the protesters, but not the supporters, from the motorcade route. The President’s detour for dinner, however, set the two groups apart. “[T]he security concerns arising from the presence of a large group of people near the open-air patio where the President was dining were plainly different from those associated with permitting a group . . . to remain along Third Street while the President’s [armored limousine] traveled by.” Brief for Petitioners 46. 9 “An ‘advance man’ is ‘[o]ne who arranges for publicity, protocol, transportation, speaking schedules, conferences with local government officials, and minute details of a visit, smoothing the way for a political figure.’ ” See 711 F. 3d, at 950, n. 2 (O’Scannlain, J., dissenting from denial of rehearing en banc) (quoting W. Safire, Safire’s Political Dictionary 8 (5th ed. 2008)).

575.US.2014_13-553

Alabama imposes sales and use taxes on railroads when they purchase or consume diesel fuel, but exempts from those taxes trucking transport companies (motor carriers) and companies that transport goods interstate through navigable waters (water carriers), both railroad competitors. Motor carriers pay an alternative fuel-excise tax on diesel, but water carriers pay neither the sales tax nor the excise tax. Respondent (CSX), an interstate rail carrier that operates in Alabama, sought to enjoin state officers from collecting sales tax on its diesel fuel purchases, claiming that the State’s asymmetrical tax treatment “discriminates against a rail carrier” in violation of the Railroad Revitalization and Regulation Reform Act of 1976, or 4–R Act, 49 U. S. C. §11501(b)(4). This Court held that a tax “discriminates” under subsection (b)(4) when it treats “groups [that] are similarly situated” differently without sufficient “justification for the difference in treatment,” CSX Transp. v. Ala. Dept. of Revenue, 562 U. S. 277 (CSX I). On remand, the District Court rejected CSX’s claim. Reversing, the Eleventh Circuit held that CSX could establish discrimination by showing that Alabama taxed rail carriers differently than their competitors, but rejected Alabama’s argument that imposing a fuel-excise tax on motor carriers, but not rail carriers, justified imposing the sales tax on rail carriers, but not motor carriers. Held: 1. The Eleventh Circuit properly concluded that CSX’s competitors are an appropriate comparison class for its subsection (b)(4) claim. All general and commercial taxpayers may be an appropriate comparison class for a subsection (b)(4) claim, but it is not the only one. Nothing in the ordinary meaning of the word “discrimination” suggests that it occurs only when the victim is singled out relative to the population at large. Context confirms this reading. The 4–R Act is an “asymmetrical statute.” CSX I, supra, at 296. In subsections (b)(1) to (b)(3)—which specify prohibitions directed toward property taxes—the comparison class is limited to commercial and industrial property in the same assessment jurisdiction. But subsection (b)(4) contains no such limitation, so the comparison class is to be determined based on the theory of discrimination alleged in the claim. Thus, when a railroad alleges that a tax disadvantages it compared to its transportation industry competitors, its competitors in that jurisdiction are the comparison class. Because subsection (b)(4) requires a showing of discrimination, however, the comparison class must consist of individuals similarly situated to the claimant. Subsection (b)(4) would be deprived of all real-world effect if “similarly situated” were given the same narrow construction the concept has in the Equal Protection Clause context, where it would be permissible for a State to tax a rail carrier more than a motor carrier, despite their seemingly similar lines of business. The category of “similarly situated” (b)(4) comparison classes must at least include the commercial and industrial taxpayers specified in the other subsections. But it also can include a railroad’s competitors. Discrimination in favor of that class both falls within the ordinary meaning of “discrimination” and frustrates the 4–R Act’s purpose of “restor[ing] the financial stability of the [Nation’s] railway system” while “foster[ing] competition among all carriers by railroad and other modes of transportation,” 90Stat. 33. Contrary to Alabama’s argument, normal rules of interpretation would say that the explicit limitation to “commercial and industrial” in the first three provisions, and its absence in the fourth, suggests that no such limitation applies to the fourth. Alabama’s additional arguments are also unavailing. Pp. 4–8. 2. The Eleventh Circuit erred in refusing to consider whether Alabama could justify its decision to exempt motor carriers from its sales and use taxes through its decision to subject motor carriers to a fuel-excise tax. It does not accord with ordinary English usage to say that a tax discriminates against a rail carrier if a rival who is exempt from that tax must pay another comparable tax from which the rail carrier is exempt, since both competitors could then claim to be discriminated against relative to each other. The Court’s negative Commerce Clause cases endorse the proposition that an additional tax on third parties may justify an otherwise discriminatory tax. Gregg Dyeing Co. v. Query, 286 U. S. 472 –480. Similarly, an alternative, roughly equivalent tax is one possible justification that renders a tax disparity non-discriminatory. CSX’s counterarguments are rejected. On remand, the Eleventh Circuit is to consider whether Alabama’s fuel-excise tax is the rough equivalent of Alabama’s sales tax as applied to diesel fuel, and therefore justifies the motor carrier sales-tax exemption. Although the State cannot offer a similar defense with respect to its water carrier exemption, the court should also examine whether any of the State’s alternative rationales justify that exemption. Pp. 8–10. 720 F. 3d 863, reversed and remanded. Scalia, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Ginsburg, J., joined.

Federal law prohibits States from imposing taxes that “discriminat[e] against a rail carrier.”49 U. S. C. §11501(b)(4). We are asked to decide whether a State violates this prohibition by taxing diesel fuel purchases made by a rail carrier while exempting similar purchases made by its competitors; and if so, whether the violation is eliminated when other tax provisions offset the challenged treatment of railroads.I Alabama taxes businesses and individuals for the purchase or use of personal property. Ala. Code §§40–23–2(1), 40–23–61(a) (2011). Alabama law sets the general tax rate at 4% of the value of the property purchased or used. Ibid. The State applies the tax, at the usual 4% rate, to railroads’ purchase or use of diesel fuel for their rail operations. But it exempts from the tax purchases and uses of diesel fuel made by trucking transport companies (whom we will call motor carriers) and companies that transport goods interstate through navigable waters (water carriers). Motor carriers instead pay a 19-cent-per-gallon fuel-excise tax on diesel; water carriers pay neither the sales nor fuel-excise tax on their diesel. §40–17–325(a)(2), and (b); §40–23–4(a)(10) (2014 Cum. Supp.). The parties stipulate that rail carriers, motor carriers, and water carriers compete. Respondent CSX Transportation, a rail carrier operating in Alabama and other States, believes this asymmetrical tax treatment “discriminates against a rail carrier” in violation of the alliterative Railroad Revitalization and Regulation Reform Act of 1976, or 4–R Act.49 U. S. C. §11501(b)(4). It sought to enjoin petitioners, the Alabama Department of Revenue and its Commissioner (Alabama or State), from collecting sales tax on its diesel fuelpurchases. At first, the District Court and Eleventh Circuit both rejected CSX’s complaint. CSX Transp., Inc. v. Alabama Dept. of Revenue, 350 Fed. Appx. 318 (2009). On this lawsuit’s first trip here, we reversed. We rejected the State’s argument that sales-and-use tax exemptions cannot “discriminate” within the meaning of subsection (b)(4), and remanded the case for further proceedings. CSX Transp., Inc. v. Alabama Dept. of Revenue,562 U. S. 277–297 (2011) (CSX I). On remand, the District Court rejected CSX’s claim after a trial. 892 F. Supp. 2d 1300 (ND Ala. 2012). The Eleventh Circuit reversed. 720 F. 3d 863 (2013). It held that, on CSX’s challenge, CSX could establish discrimination by showing the State taxed rail carriers differently than their competitors—which, by stipulation, included motor carriers and water carriers. But it rejected Alabama’s argument that the fuel-excise taxes offset the sales taxes—in other words, that because it imposed its fuel-excise tax on motor carriers, but not rail carriers, it was justified in imposing the sales tax on rail carriers, but not motor carriers. Ibid. We granted certiorari to resolve whether the Eleventh Circuit properly regarded CSX’s competitors as an appropriate comparison class for its subsection (b)(4) claim. 573 U. S. ___ (2014). We also directed the parties to address whether, when resolving a claim of unlawful tax discrimination, a court should consider aspects of a State’s tax scheme apart from the challenged provision. Ibid.II The 4–R Act provides: “(b) The following acts unreasonably burden and discriminate against interstate commerce, and a State, subdivision of a State, or authority acting for a State or subdivision of a State may not do any of them: “(1) Assess rail transportation property at a value that has a higher ratio to the true market value of the rail transportation property than the ratio that the assessed value of other commercial and industrial property in the same assessment jurisdiction has to the true market value of the other commercial and industrial property. “(2) Levy or collect a tax that may not be made under paragraph (1) of this subsection. “(3) Levy or collect an ad valorem property tax at a tax rate that exceeds the tax rate applicable to commercial and industrial property in the same assessment jurisdiction. “(4) Impose another tax that discriminates against a rail carrier providing transportation subject to thejurisdiction of the Board under this part.” §11501(b)(1)–(4). In our last opinion in this case, we held that “discriminates” in subsection (b)(4) carries its ordinary meaning, and that a tax discriminates under subsection (b)(4) when it treats “groups [that] are similarly situated” differently without sufficient “justification for the difference in treatment.” CSX I, supra, at 287. Here, we address the meaning of these two quoted phrases.A The first question in this case is who is the “comparison class” for purposes of a subsection (b)(4) claim. Alabama argues that the only appropriate comparison class for a subsection (b)(4) claim is all general commercial and industrial taxpayers. We disagree. While all general and commercial taxpayers is an appropriate comparison class, it is not the only one. Nothing in the ordinary meaning of the word “discrimination” suggests that it occurs only when the victim is singled out relative to the population at large. If, for example, a State offers free college education to all returning combat veterans, but arbitrarily excepts those who served in the Marines, we would say that Marines have experienced discrimination. That would remain the case even though the Marines are treated the same way as members of the general public, who have to pay for their education. Context confirms that the comparison class for subsection (b)(4) is not limited as Alabama suggests. The 4–R Act is an “asymmetrical statute.” Id., at 296. Subsections (b)(1) to (b)(3) contain three specific prohibitions directed towards property taxes. Each requires comparison of railroad property to commercial and industrial property in the same assessment jurisdiction. The Act therefore limits the comparison class for challenges under those provisions. Even if the jurisdiction treats railroads less favor-ably than residential property, no violation of these subsec-tions has occurred. Subsection (b)(4) contains no such limitation, leaving the comparison class to be determined as it is normally determined with respect to discrimination claims. And we think that depends on the theory of discrimination alleged in the claim. When a railroad alleges that a tax targets it for worse treatment than local businesses, all other commercial and industrial taxpayers are the comparison class. When a railroad alleges that a tax disadvantages it compared to its competitors in the transportation industry, the railroad’s competitors in that jurisdiction are the comparison class. So, picking a comparison class is extraordinarily easy. Unlike under subsections (b)(1)–(3), the railroad is not limited to all commercial and industrial taxpayers; all the world, or at least all the world within the taxing jurisdiction, is its comparison-class oyster. But that is not as generous a concession as might seem. What subsection (b)(4) requires, and subsections (b)(1)–(3) do not, is a showing of discrimination—of a failure to treat similarly situated persons alike. A comparison class will thus support a discrimination claim only if it consists of individuals similarly situated to the claimant. That raises the question of when a proposed comparison class qualifies as similarly situated. In the Equal Protection Clause context, very few taxpayers are regarded as similarly situated and thus entitled to equal treatment. There, a State may tax different lines of businesses differently with near-impunity, even if they are apparently similar. We have upheld or approved of distinctions between utilities—including a railroad—and other corporations, New York Rapid Transit Corp. v. City of New York,303 U. S. 573,579 (1938), between wholesalers and retailers in goods, Caskey Baking Co. v. Virginia,313 U. S. 117–121 (1941), between chain retail stores and independent retail stores, State Bd. of Tax Comm’rs of Ind. v. Jackson,283 U. S. 527–542 (1931), between anthracite coal mines and bituminous coal mines, Heisler v. Thomas Colliery Co.,260 U. S. 245,254,257 (1922), and between sellers of coal oil and sellers of coal, Southwestern Oil Co. v. Texas,217 U. S. 114,121 (1910). As one treatise has observed, we recognize a “wide latitude state legislatures enjoy in drawing tax classifications under the Equal Protection Clause.” 1 J. Hellerstein & W. Hellerstein, State Taxation ¶3.03[1], p. 3–5 (3d ed. 2001–2005). This includes the power to impose “widely different taxes on various trades or professions.” Id., at 3–5 to 3–6. It would be permissible—as far as the Equal Protection Clause is concerned—for a State to tax a rail carrier more than a motor carrier, despite the seeming similarity in their lines of business. The concept of “similarly situated” individuals cannot be so narrow here. That would deprive subsection (b)(4) of all real-world effect, providing protection that the Equal Protection Clause already provides. Moreover, the cate-gory of “similarly situated” (b)(4) comparison classes must include commercial and industrial taxpayers. There is no conceivable reason why the statute would forbid property taxes higher than what that class enjoys (or suffers), but permit other taxes that discriminate in favor of that class vis-à-vis railroads. And we think the competitors of railroads can be another “similarly situated” comparison class, since discrimination in favor of that class most obviously frustrates the purpose of the 4–R Act, which was to “restore the financial stability of the railway system of the United States,” §101(a),90Stat.33, while “foster[ing] competition among all carriers by railroad and other modes of transportation,” §101(b)(2). We need not, and thus do not, express any opinion on what other comparison classes may qualify. Sufficient unto the day is the evil thereof. Alabama claims that because subsections (b)(1) and (b)(3) (and (b)(2) through reference to (b)(1)) establish a comparison class of “commercial and industrial property,” subsection (b)(4) must establish a comparison class of “general commercial and industrial taxpayers.” This inverts normal rules of interpretation, which would say that the explicit limitation to “commercial and industrial” in the first three provisions, and the absence of such a limitation in the fourth, suggests that no such limitation applies to the fourth. Moreover, Alabama’s interpretation would require us to dragoon the modifier “commercial and industrial”—but not the noun “property”—from the first three provisions, append “general” in front of it and “taxpayers” after, both words foreign to the preceding subsections. We might also have to strip away the restrictions in the definition of “commercial and industrial property,” which excludes land primarily used for agricultural purposes and timber growing.49 U. S. C. §11501(a)(4). This is not our concept of fidelity to a statute’s text. Alabama responds that the introductory clause of §11501(b)—which declares that the “following acts unreasonably burden and discriminate against interstate commerce,”—“binds its four subsections together,” Brief for Petitioners 23 (emphasis deleted), and gives them a common object and scope. The last time this case appeared before us, Alabama made a similar argument in support of the claim that, because subsections (b)(1)–(3) cover only property taxes, so too does subsection (b)(4). See Brief for Respondents in CSX Transp., Inc. v. Alabama Dept. of Revenue, O. T. 2010, No. 09–520, p. 25–26. We rejected this argument then, and we reject it again now. Alabama persists that a case-specific inquiry allows a railroad to “hand-pick [its] comparison class,” Brief for Petitioners 41, which would be unfair—a “windfall” to railroads. Ibid. As we have described above, picking a class is easy, but it is not easy to establish that the selected class is “similarly situated” for purposes of discrimina-tion in taxation. The Eleventh Circuit properly concluded that, in light of CSX Transportation’s complaint and the parties’ stipulation, a comparison class of competitors consisting of motor carriers and water carriers was appropriate, and differential treatment vis-à-vis that class would constitute discrimination. We therefore turn tothe court’s refusal to consider Alabama’s alternative tax justifications.B A State’s tax discriminates only where the State cannot sufficiently justify differences in treatment between similarly situated taxpayers. As we have discussed above, a rail carrier and its competitors can be considered similarly situated for purposes of this provision. But what about the claim that those competitors are subject to other taxes that the railroads avoid? We think Alabama can justify its decision to exempt motor carriers from its sales and use tax through its decision to subject motor carriers to a fuel-excise tax. It does not accord with ordinary English usage to say that a tax discriminates against a rail carrier if a rival who is exempt from that tax must pay another comparable tax from which the rail carrier is exempt. If that were true, both competitors could claim to be disfavored—discriminated against—relative to each other. Our negative Commerce Clause cases endorse the proposition that an additional tax on third parties may justify an otherwise discriminatory tax. Gregg Dyeing Co. v. Query,286 U. S. 472–480 (1932). We think that an alternative, roughly equivalent tax is one possible justification that rendersa tax disparity nondiscriminatory. CSX claims that because the statutory prohibition forbids “impos[ing] another tax that discriminates against a rail carrier,”49 U. S. C. §11501(b)(4)—“tax” in the singular—the appropriate inquiry is whether the challenged tax discriminates, not whether the tax code as a whole does so. It is undoubtedly correct that the “tax” (singular) must discriminate—but it does not discriminate unless it treats railroads differently from other similarly situated taxpayers without sufficient justification. A comparable tax levied on a competitor may justify not extending that competitor’s exemption from a general tax to a railroad. It is easy to display the error of CSX’s single-tax-provision approach. Under that model, the following tax would violate the 4–R Act: “(1) All railroads shall pay a 4% sales tax. (2) All other individuals shall also pay a 4% sales tax.” CSX would undoubtedly object that not every case will be so easy, and that federal courts are ill qualified to explore the vagaries of state tax law. We are inclined to agree, but that cannot carry the day. Congress assigned this task to the courts by drafting an antidiscrimination command in such sweeping terms. There is simply no discrimination when there are roughly comparable taxes. If the task of determining when that is so is “Sisyphean,” as the Eleventh Circuit called it, 720 F. 3d, at 871, it is a Sisyphean task that the statute imposes. We therefore cannot approve of the Eleventh Circuit’s refusal to consider Alabama’s tax-based justification, and remand for that court to consider whether Alabama’s fuel-excise tax is the rough equivalent of Alabama’s sales tax as applied to diesel fuel, and therefore justifies the motor carrier sales-tax exemption.C While the State argues that the existence of a fuel-excise tax justifies its decision to exempt motor carriers from the sales and use tax, it cannot offer a similar defense with respect to its exemption for water carriers. Water carriers pay neither tax. The State, however, offers other justifications for the water carrier exemption—for example, that such an exemption is compelled by federal law. The Eleventh Circuit failed to examine these justifications, asserting that the water carriers were the beneficiaries of a discriminatory tax regime. We do not consider whether Alabama’s alternative rationales justify its exemption, but leave that question for the Eleventh Circuit on remand.* * * The judgment of the Eleventh Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion.It is so ordered.

575.US.2014_13-352

Respondent Hargis Industries, Inc. (Hargis), tried to register its trademark for SEALTITE with the United States Patent and Trademark Office pursuant to the Lanham Act. Petitioner, B&B Hardware, Inc. (B&B), however, opposed registration, claiming that SEALTITE is too similar to B&B’s own SEALTIGHT trademark. The Trademark Trial and Appeal Board (TTAB) concluded that SEALTITE should not be registered because of the likelihood of confusion. Hargis did not seek judicial review of that decision. Later, in an infringement suit before the District Court, B&B argued that Hargis was precluded from contesting the likelihood of confusion because of the TTAB’s decision. The District Court disagreed. The Eighth Circuit affirmed, holding that preclusion was unwarranted because the TTAB and the court used different factors to evaluate likelihood of confusion, the TTAB placed too much emphasis on the appearance and sound of the two marks, and Hargis bore the burden of persuasion before the TTAB while B&B bore it before the District Court. Held: So long as the other ordinary elements of issue preclusion are met, when the usages adjudicated by the TTAB are materially the same as those before a district court, issue preclusion should apply. Pp. 8–22. (a) An agency decision can ground issue preclusion. The Court’s cases establish that when Congress authorizes agencies to resolve disputes, “courts may take it as given that Congress has legislated with the expectation that [issue preclusion] will apply except when a statutory purpose to the contrary is evident.” Astoria Fed. Sav. & Loan Assn. v. Solimino, 501 U. S. 104 . Constitutional avoidance does not compel a different conclusion. Pp. 8–12. (b) Neither the Lanham Act’s text nor its structure rebuts the “presumption” in favor of giving preclusive effect to TTAB decisions where the ordinary elements of issue preclusion are met. Astoria, 501 U. S., at 108. This case is unlike Astoria. There, where exhausting the administrative process was a prerequisite to suit in court, giving preclusive effect to the agency’s determination in that very administrative process could have rendered the judicial suit “strictly pro forma.” Id., at 111. By contrast, registration involves a separate proceeding to decide separate rights. Pp. 12–14. (c) There is no categorical reason why registration decisions can never meet the ordinary elements of issue preclusion. That many registrations will not satisfy those ordinary elements does not mean that none will. Pp. 15–22. (1) Contrary to the Eighth Circuit’s conclusion, the same likelihood-of-confusion standard applies to both registration and infringement. The factors that the TTAB and the Eighth Circuit use to assess likelihood of confusion are not fundamentally different, and, more important, the operative language of each statute is essentially the same. Hargis claims that the standards are different, noting that the registration provision asks whether the marks “resemble” each other, 15 U. S. C. §1052(d), while the infringement provision is directed towards the “use in commerce” of the marks, §1114(1). That the TTAB and a district court do not always consider the same usages, however, does not mean that the TTAB applies a different standard to the usages it does consider. If a mark owner uses its mark in materially the same ways as the usages included in its registration application, then the TTAB is deciding the same likelihood-of-confusion issue as a district court in infringement litigation. For a similar reason, the Eighth Circuit erred in holding that issue preclusion could not apply because the TTAB relied too heavily on “appearance and sound.” Pp. 15–19. (2) The fact that the TTAB and district courts use different procedures suggests only that sometimes issue preclusion might be inappropriate, not that it always is. Here, there is no categorical “reason to doubt the quality, extensiveness, or fairness,” Montana v. United States, 440 U. S. 147 , 164, n. 11, of the agency’s procedures. In large part they are exactly the same as in federal court. Also contrary to the Eighth Circuit’s conclusion, B&B, the party opposing registration, not Hargis, bore the burden of persuasion before the TTAB, just as it did in the infringement suit. Pp. 19–21. (3) Hargis is also wrong that the stakes for registration are always too low for issue preclusion in later infringement litigation. When registration is opposed, there is good reason to think that both sides will take the matter seriously. Congress’ creation of an elaborate registration scheme, with many important rights attached and backed up by plenary review, confirms that registration decisions can be weighty enough to ground issue preclusion. Pp. 21–22. 716 F. 3d 1020, reversed and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Ginsburg, J., filed a concurring opinion. Thomas, J., filed a dissenting opinion, in which Scalia, J., joined.

Sometimes two different tribunals are asked to decide the same issue. When that happens, the decision of the first tribunal usually must be followed by the second, at least if the issue is really the same. Allowing the same issue to be decided more than once wastes litigants’ resources and adjudicators’ time, and it encourages parties who lose before one tribunal to shop around for another. The doctrine of collateral estoppel or issue preclusion is designed to prevent this from occurring. This case concerns the application of issue preclusion in the context of trademark law. Petitioner, B&B Hardware, Inc. (B&B), and respondent Hargis Industries, Inc. (Hargis), both use similar trademarks; B&B owns SEALTIGHT while Hargis owns SEALTITE. Under the Lanham Act,60Stat.427, as amended,15 U. S. C. §1051 et seq., an applicant can seek to register a trademark through an administrative process within the United States Patent and Trademark Office (PTO). But if another party believes that the PTO should not register a mark because it is too similar to its own, that party can oppose registration before the Trademark Trial and Appeal Board (TTAB). Here, Hargis tried to register the mark SEALTITE, but B&B opposed SEALTITE’s registration. After a lengthy proceeding, the TTAB agreed with B&B that SEALTITE should not be registered. In addition to permitting a party to object to the registration of a mark, the Lanham Act allows a mark owner to sue for trademark infringement. Both a registration proceeding and a suit for trademark infringement, more-over, can occur at the same time. In this case, while the TTAB was deciding whether SEALTITE should be registered, B&B and Hargis were also litigating the SEALTIGHT versus SEALTITE dispute in federal court. In both registration proceedings and infringement litigation, the tribunal asks whether a likelihood of confusion exists between the mark sought to be protected (here, SEALTIGHT) and the other mark (SEALTITE). The question before this Court is whether the District Court in this case should have applied issue preclusion to the TTAB’s decision that SEALTITE is confusingly similar to SEALTIGHT. Here, the Eighth Circuit rejected issue preclusion for reasons that would make it difficult for the doctrine ever to apply in trademark disputes. We disagree with that narrow understanding of issue preclusion. Instead, consistent with principles of law that apply in innumerable contexts, we hold that a court should give preclusive effect to TTAB decisions if the ordinary elements of issue preclusion are met. We therefore reverse the judgment of the Eighth Circuit and remand for further proceedings.IA Trademark law has a long history, going back at least to Roman times. See Restatement (Third) of Unfair Competition §9, Comment b (1993). The principle underlying trademark protection is that distinctive marks—words, names, symbols, and the like—can help distinguish a particular artisan’s goods from those of others. Ibid. One who first uses a distinct mark in commerce thus acquires rights to that mark. See 2 J. McCarthy, Trademarks and Unfair Competition §16:1 (4th ed. 2014) (hereinafter McCarthy). Those rights include preventing others from using the mark. See 1 A. LaLonde, Gilson on Trademarks §3.02[8] (2014) (hereinafter Gilson). Though federal law does not create trademarks, see, e.g., Trade-Mark Cases,100 U. S. 82,92 (1879), Congress has long played a role in protecting them. In 1946, Congress enacted the Lanham Act, the current federal trademark scheme. As relevant here, the Lanham Act creates at least two adjudicative mechanisms to help protect marks. First, a trademark owner can register its mark with the PTO. Second, a mark owner can bring a suit for infringement in federal court. Registration is significant. The Lanham Act confers “important legal rights and benefits” on trademark owners who register their marks. 3 McCarthy §19:3, at 19–21 see also id., §19:9, at 19–34 (listing seven of the “procedural and substantive legal advantages” of registration). Registration, for instance, serves as “constructive notice of the registrant’s claim of ownership” of the mark.15 U. S. C. §1072. It also is “prima facie evidence of the validity of the registered mark and of the registration of the mark, of the owner’s ownership of the mark, and of the owner’s exclusive right to use the registered mark in commerce on or in connection with the goods or services specified in the certificate.” §1057(b). And once a mark has been registered for five years, it can become “incontestable.” §§1065, 1115(b) To obtain the benefits of registration, a mark owner files an application with the PTO. §1051. The application must include, among other things, “the date of the applicant’s first use of the mark, the date of the applicant’s first use of the mark in commerce, the goods in connection with which the mark is used, and a drawing of the mark.” §1051(a)(2). The usages listed in the application—i.e., those goods on which the mark appears along with, if applicable, their channels of distribution—are critical. See, e.g., 3 McCarthy §20:24, at 20–83 (“[T]he applicant’s right to register must be made on the basis of the goods described in the application”); id., §20:15, at 20–85 (explaining that if an “application does not delimit any spe-cific trade channels of distribution, no limitation will be” applied). The PTO generally cannot register a mark which “so resembles” another mark “as to be likely, when used on or in connection with the goods of the applicant, to cause confusion, or to cause mistake, or to deceive.”15 U. S. C. §1052(d). If a trademark examiner believes that registration is warranted, the mark is published in the Official Gazette of the PTO. §1062. At that point, “[a]ny person who believes that he would be damaged by the registration” may “file an opposition.” §1063(a). Opposition proceedings occur before the TTAB (or panels thereof). §1067(a). The TTAB consists of administrative trademark judges and high-ranking PTO officials, including the Director of the PTO and the Commissioner of Trademarks. §1067(b). Opposition proceedings before the TTAB are in many ways “similar to a civil action in a federal district court.” TTAB Manual of Procedure §102.03 (2014) (hereinafter TTAB Manual), online at http://www.uspto.gov (as visited Mar. 20, 2015, and available in Clerk of Court’s case file). These proceedings, for instance, are largely governed by the Federal Rules of Civil Procedure and Evidence. See 37 CFR §§2.116(a), 2.122(a) (2014). The TTAB also allows discovery and depositions. See §§2.120, 2.123(a). The party opposing registration bears the burden of proof, see §2.116(b), and if that burden cannot be met, the opposed mark must be registered, see15 U. S. C. §1063(b). The primary way in which TTAB proceedings differ from ordinary civil litigation is that “proceedings before the Board are conducted in writing, and the Board’s actions in a particular case are based upon the written record therein.” TTAB Manual §102.03. In other words, there is no live testimony. Even so, the TTAB allows parties to submit transcribed testimony, taken under oath and subject to cross-examination, and to request oral argument. See 37 CFR §§2.123, 2.129. When a party opposes registration because it believes the mark proposed to be registered is too similar to its own, the TTAB evaluates likelihood of confusion by applying some or all of the 13 factors set out in In re E. I. DuPont DeNemours & Co., 476 F. 2d 1357 (CCPA 1973). After the TTAB decides whether to register the mark, a party can seek review in the U. S. Court of Appeals for the Federal Circuit, or it can file a new action in district court. See15 U. S. C. §1071. In district court, the parties can conduct additional discovery and the judge resolves registration de novo. §1071(b); see also 3 McCarthy §21:20 (explaining differences between the forums); cf. Kappos v. Hyatt, 566 U. S. ___ (2012) (de novo review for analogous scheme in patent law). The Lanham Act, of course, also creates a federal cause of action for trademark infringement. The owner of a mark, whether registered or not, can bring suit in federal court if another is using a mark that too closely resembles the plaintiff’s. The court must decide whether the defendant’s use of a mark in commerce “is likely to cause confusion, or to cause mistake, or to deceive” with regards to the plaintiff’s mark. See15 U. S. C. §1114(1)(a) (registered marks); §1125(a)(1)(A) (unregistered marks). In infringement litigation, the district court considers the full range of a mark’s usages, not just those in the application.B Petitioner B&B and respondent Hargis both manufacture metal fasteners. B&B manufactures fasteners for the aerospace industry, while Hargis manufactures fasteners for use in the construction trade. Although there are obvious differences between space shuttles and A-frame buildings, both aerospace and construction engineers prefer fasteners that seal things tightly. Accordingly, both B&B and Hargis want their wares associated with tight seals. A feud of nearly two decades has sprung from this seemingly commonplace set of facts. In 1993 B&B registered SEALTIGHT for “threaded or unthreaded metal fasteners and other related hardwar[e]; namely, self-sealing nuts, bolts, screws, rivets and washers, all having a captive o-ring, for use in the aerospace industry.” App. 223a (capitalization omitted). In 1996, Hargis sought to register SEALTITE for “self-piercing and self-drilling metal screws for use in the manufacture of metal and post-frame buildings.” App. 70a (capitalization omitted). B&B opposed Hargis’ registration because, although the two companies sell different products, it believes that SEALTITE is confusingly similar to SEALTIGHT. The twists and turns in the SEALTIGHT versus SEALTITE controversy are labyrinthine. The question whether either of these marks should be registered, and if so, which one, has bounced around within the PTO for about two decades; related infringement litigation has been before the Eighth Circuit three times; and two separate juries have been empaneled and returned verdicts. The full story could fill a long, unhappy book. For purposes here, we pick up the story in 2002, when the PTO published SEALTITE in the Official Gazette. This prompted opposition proceedings before the TTAB, complete with discovery, including depositions. B&B ar-gued that SEALTITE could not be registered because itis confusingly similar to SEALTIGHT. B&B explained, for instance, that both companies have an online presence, the largest distributor of fasteners sells both companies’ products, and consumers sometimes call the wrong company to place orders. Hargis rejoined that the companies sell different products, for different uses, to different types of consumers, through different channels of trade. Invoking a number of the DuPont factors, the TTAB sided with B&B. The Board considered, for instance, whether SEALTIGHT is famous (it’s not, said the Board), how the two products are used (differently), how much the marks resemble each other (very much), and whether customers are actually confused (perhaps sometimes). See App. to Pet. for Cert. 55a–71a. Concluding that “the most critical factors in [its] likelihood of confusion analysis are the similarities of the marks and the similarity of the goods,” id., at 70a, the TTAB determined that SEALTITE—when “used in connection with ‘self-piercing and self-drilling metal screws for use in the manufacture of metal and post-frame buildings’ ”—could not be registered because it “so resembles” SEALTIGHT when “used in connection with fasteners that provide leakproof protection from liquids and gases, fasteners that have a captive o-ring, and ‘threaded or unthreaded metal fastners and other related hardware . . . for use in the aerospace industry’ as to be likely to cause confusion,” id., at 71a. Despite a right to do so, Hargis did not seek judicial review in either the Federal Circuit or District Court. All the while, B&B had sued Hargis for infringement. Before the District Court ruled on likelihood of confusion, however, the TTAB announced its decision. After a series of proceedings not relevant here, B&B argued to the District Court that Hargis could not contest likelihood of confusion because of the preclusive effect of the TTAB decision. The District Court disagreed, reasoning that the TTAB is not an Article III court. The jury returned a verdict for Hargis, finding no likelihood of confusion. B&B appealed to the Eighth Circuit. Though accepting for the sake of argument that agency decisions can ground issue preclusion, the panel majority affirmed for three reasons: first, because the TTAB uses different factors than the Eighth Circuit to evaluate likelihood of confusion; second, because the TTAB placed too much emphasis on the appearance and sound of the two marks; and third, because Hargis bore the burden of persuasion before the TTAB, while B&B bore it before the District Court. 716 F. 3d 1020 (2013). Judge Colloton dissented, concluding that issue preclusion should apply. After calling for the views of the Solicitor General, we granted certiorari. 573 U. S. ___ (2014).II The first question that we must address is whether an agency decision can ever ground issue preclusion. The District Court rejected issue preclusion because agencies are not Article III courts. The Eighth Circuit did not adopt that view, and, given this Court’s cases, it was right to take that course. This Court has long recognized that “the determination of a question directly involved in one action is conclusive as to that question in a second suit.” Cromwell v. County of Sac,94 U. S. 351,354 (1877). The idea is straightforward: Once a court has decided an issue, it is “forever settled as between the parties,” Baldwin v. Iowa State Traveling Men’s Assn.283 U. S. 522,525 (1931), thereby “protect[ing]” against “the expense and vexation attending multiple lawsuits, conserv[ing] judicial resources, and foster[ing] reliance on judicial action by minimizing the possibility of inconsistent verdicts,” Montana v. United States,440 U. S. 147–154 (1979). In short, “a losing litigant deserves no rematch after a defeat fairly suffered.” Astoria Fed. Sav. & Loan Assn. v. Solimino,501 U. S. 104,107 (1991). Although the idea of issue preclusion is straightforward, it can be challenging to implement. The Court, therefore, regularly turns to the Restatement (Second) of Judgments for a statement of the ordinary elements of issue preclusion. See, e.g., Bobby v. Bies,556 U. S. 825,834 (2009); New Hampshire v. Maine,532 U. S. 742–749 (2001); Baker v. General Motors Corp.,522 U. S. 222,233, n.5 (1998). The Restatement explains that subject to certain well-known exceptions, the general rule is that “[w]hen an issue of fact or law is actually litigated and determined by a valid and final judgment, and the determination is essential to the judgment, the determination is conclusive in a subsequent action between the parties, whether on the same or a different claim.” Restatement (Second) of Judgments §27, p. 250 (1980); see also id., §28, at 273 (listing exceptions such as whether appellate review was available or whether there were “differences in the quality or extensiveness of the procedures followed”). Both this Court’s cases and the Restatement make clear that issue preclusion is not limited to those situations in which the same issue is before two courts. Rather, where a single issue is before a court and an administrative agency, preclusion also often applies. Indeed, this Court has explained that because the principle of issue preclusion was so “well established” at common law, in those situations in which Congress has authorized agencies to resolve disputes, “courts may take it as given that Congress has legislated with the expectation that the principle [of issue preclusion] will apply except when a statutory purpose to the contrary is evident.” Astoria, supra, at 108. This reflects the Court’s longstanding view that “ ‘[w]hen an administrative agency is acting in a judicial capacity and resolves disputed issues of fact properly before it which the parties have had an adequate opportunity to litigate, the courts have not hesitated to apply res judicata to enforce repose.’ ” University of Tenn. v. Elliott,478 U. S. 788–798 (1986) (quoting United States v. UtahConstr. & Mining Co.,384 U. S. 394,422 (1966)); see also Hayfield Northern R. Co. v. Chicago & North Western Transp. Co.,467 U. S. 622,636, n.15 (1984) (noting Utah Construction); Kremer v. Chemical Constr. Corp.,456 U. S. 461–485, n. 26 (1982) (characterizing Utah Construction’s discussion of administrative preclusion as a holding); Restatement (Second) of Judgments §83(1), at 266 (explaining that, with some limits, “a valid and final adjudicative determination by an administrative tribunal has the same effects under the rules of res judicata, subject to the same exceptions and qualifications, as a judgment of a court”). Although apparently accepting Astoria and Utah Construction,[1] Hargis argues that we should not read the Lanham Act (or, presumably, many other federal statutes) as authorizing issue preclusion. Otherwise, Hargis warns, the Court would have to confront “ ‘grave and doubtful questions’ as to the Lanham Act’s consistency with the Seventh Amendment and Article III of the Constitution.” Brief for Respondent 38 (quoting United States ex rel. Attorney General v. Delaware & Hudson Co.,213 U. S. 366,408 (1909)). We are not persuaded. At the outset, we note that Hargis does not argue that giving issue preclusive effect to the TTAB’s decision would be unconstitutional. Instead, Hargis contends only that we should read the Lanham Act narrowly because a broad reading might be unconstitutional. See, e.g., Brief for Respondent 37, 39, 40, 41–42. The likely reason that Hargis has not directly advanced a constitutional argument is that, at least as to a jury trial right, Hargis did not even list the Seventh Amendment as an authority in its appellee brief to the Eighth Circuit. Moreover, although Hargis pressed an Article III argument below, in its opposition to certiorari in this Court, Hargis seemingly con-ceded that TTAB decisions can sometimes ground issue preclusion, though it now protests otherwise. SeeSupplemental Brief in Opposition 2. To the extent, if any, that there could be a meritorious constitutional objection, it is not before us. See Plaut v. Spendthrift Farm, Inc.,514 U. S. 211–232 (1995). We reject Hargis’ statutory argument that we should jettison administrative preclusion in whole or in part to avoid potential constitutional concerns. As to the Seventh Amendment, for instance, the Court has already held that the right to a jury trial does not negate the issue-preclusive effect of a judgment, even if that judgment was entered by a juryless tribunal. See Parklane Hosiery Co. v. Shore,439 U. S. 322,337 (1979). It would seem to follow naturally that although the Seventh Amendment creates a jury trial right in suits for trademark damages, see Dairy Queen, Inc. v. Wood,369 U. S. 469–480 (1962), TTAB decisions still can have preclusive effect in such suits. Hargis disputes this reasoning even though it admits that in 1791 “ ‘a party was not entitled to have a jury determine issues that had been previously adjudi-cated by a chancellor in equity.’ ” Brief for Respondent 39 (quoting Parklane Hosiery, supra, at 333). Instead, Hargis contends that issue preclusion should not apply to TTAB registration decisions because there were no agencies at common law. But our precedent holds that the Seventh Amendment does not strip competent tribunals of the power to issue judgments with preclusive effect; that logic would not seem to turn on the nature of the competent tribunal. And at the same time, adopting Hargis’ view would dramatically undercut agency preclusion, despite what the Court has already said to the contrary. Nothing in Hargis’ avoidance argument is weighty enough to overcome these weaknesses. The claim that we should read the Lanham Act narrowly to avoid Article III concerns is equally unavailing—andfor similar reasons. Hargis argues that because it might violate Article III if an agency could make a decision with preclusive effect in a later proceeding before a federal court, we should conclude, as a statutory matter, that issue preclusion is unavailable. Such a holding would not fit with our precedent. For instance, in Elliott, the Court, relying on Utah Construction, explained that absent a contrary indication, Congress presumptively intends that an agency’s determination (there, a state agency) has preclusive effect. 478 U. S., at 796–799; see also Astoria, 501 U. S., at 110 (recognizing the “presumption”). To be sure, the Court has never addressed whether such preclusion offends Article III. But because this Court’s cases are so clear, there is no ambiguity for this Court to sidestep through constitutional avoidance.[2]III The next question is whether there is an “evident” reason why Congress would not want TTAB decisions to receive preclusive effect, even in those cases in which the ordinary elements of issue preclusion are met. Astoria, supra, at 108. We conclude that nothing in the Lanham Act bars the application of issue preclusion in such cases. The Lanham Act’s text certainly does not forbid issue preclusion. Nor does the Act’s structure. Granted, one can seek judicial review of a TTAB registration decision in a de novo district court action, and some courts have concluded from this that Congress does not want unreviewed TTAB decisions to ground issue preclusion. See, e.g., American Heritage Life Ins. Co. v. Heritage Life Ins. Co., 494 F. 2d 3, 9–10 (CA5 1974). But that conclusion does not follow. Ordinary preclusion law teaches that if a party to a court proceeding does not challenge an adverse decision, that decision can have preclusive effect in other cases, even if it would have been reviewed de novo. See Restatement (Second) of Judgments §28, Comment a and Illustration 1 (explaining that the failure to pursue an appeal does not undermine issue preclusion and including an example of an apparently unappealed district court’s dismissal for failure to state a claim); cf. Federated Department Stores, Inc. v. Moitie,452 U. S. 394,398 (1981) (noting “the res judicata consequences of a final, unappealed judgment on the merits”). This case is also unlike Astoria, where a plaintiff claiming discrimination first went to an agency and then sued in court about the same alleged conduct. See 501 U. S., at 111. The Court concluded, quite sensibly, that the structure of that scheme indicated that the agency decision could not ground issue preclusion. When exhausting an administrative process is a prerequisite to suit in court, giving preclusive effect to the agency’s determination in that very administrative process could render the judicial suit “strictly pro forma.” Ibid.; see also Elliott, supra, at 795–796 (similar analysis). Here, if a party urged a district court reviewing a TTAB registration decision to give preclusive effect to the very TTAB decision under review, Astoria would apply. But that is not this case. What matters here is that registration is not a prerequisite to an infringement action. Rather, it is a separate proceeding to decide separate rights. Neither is issue preclusion a one-way street. When a district court, as part of its judgment, decides an issue that overlaps with part of the TTAB’s analysis, the TTAB gives preclusive effect to the court’s judgment. See App. to Pet. for Cert. 54a–55a (giving preclusive effect to the District Court’s ear-lier decision regarding SEALTIGHT’s distinctiveness be-cause the issue “was actually litigated and necessarily determined”). Hargis also argues that allowing TTAB decisions to have issue-preclusive effect will adversely affect the registration process. Because of the TTAB’s “ ‘limited jurisdiction’ ” and “ ‘the narrowness of the issues’ ” before it, Hargis contends, the Court should infer that TTAB proceedings are supposed to be more streamlined than infringement litigation. See Brief for Respondent 30 (quoting TTAB Manual §402.01). But, the argument goes, if TTAB decisions can have issue-preclusive effect in infringement litigation, parties may spend more time and energy before the TTAB, thus bogging down the registration process. This concern does not change our conclusion. Issue preclusion is available unless it is “evident,” Astoria, supra, at 108, that Congress does not want it. Here, if a streamlined process in all registration matters was particularly dear to Congress, it would not have authorized de novo challenges for those “dissatisfied” with TTAB decisions.15 U. S. C. §1071(b). Plenary review serves many functions, but ensuring a streamlined process is not one of them. Moreover, as explained below, for a great many registration decisions issue preclusion obviously will not apply because the ordinary elements will not be met. For those registrations, nothing we say today is relevant.IV At last we turn to whether there is a categorical reason why registration decisions can never meet the ordinary elements of issue preclusion, e.g., those elements set out in §27 of the Restatement (Second) of Judgments. Although many registrations will not satisfy those ordinary elements, that does not mean that none will. We agree with Professor McCarthy that issue preclusion applies where “the issues in the two cases are indeed identical and the other rules of collateral estoppel are carefully observed.” 6 McCarthy §32:99, at 32–244; see also 3 Gilson §11.08[4][i][iii][B], p. 11–319 (“Ultimately, Board decisions on likelihood of confusion . . . should be given preclusive effect on a case-by-case basis”).A The Eighth Circuit’s primary objection to issue preclusion was that the TTAB considers different factors than it does. Whereas the TTAB employs some or all of the DuPont factors to assess likelihood of confusion, the Eighth Circuit looks to similar, but not identical, factors identified in SquirtCo v. Seven-Up Co., 628 F. 2d 1086, 1091 (CA8 1980). The court’s instinct was sound: “[I]ssues are not identical if the second action involves application of a different legal standard, even though the factual setting of both suits may be the same.” 18 C. Wright, A. Miller, & E. Cooper, Federal Practice & Procedure §4417, p. 449 (2d ed. 2002) (hereinafter Wright & Miller). Here, however, the same likelihood-of-confusion standard applies to both registration and infringement. To begin with, it does not matter that registration and infringement are governed by different statutory provisions. Often a single standard is placed in different statutes; that does not foreclose issue preclusion. See, e.g., Smith v. Bayer Corp., 564 U. S. ___, ___ (2011) (slip op., at 7). Neither does it matter that the TTAB and the Eighth Circuit use different factors to assess likelihood of confusion. For one thing, the factors are not fundamentally different, and “[m]inor variations in the application of what is in essence the same legal standard do not defeat preclusion.” Id., at ___, n. 9 (slip op., at 12, n. 9). More important, if federal law provides a single standard, parties cannot escape preclusion simply by litigating anew in tribunals that apply that one standard differently. A contrary rule would encourage the very evils that issue preclusion helps to prevent. The real question, therefore, is whether likelihood of confusion for purposes of registration is the same standard as likelihood of confusion for purposes of infringement. We conclude it is, for at least three reasons. First, the operative language is essentially the same; the fact that the registration provision separates “likely” from “to cause confusion, or to cause mistake, or to deceive” does not change that reality.[3] See 2 Gilson §5.01[2][a], at 5–17 (explaining that “the same statutory test” applies). Second, the likelihood-of-confusion language that Congress used in these Lanham Act provisions has been central to trademark registration since at least 1881. See Act of Mar. 3, 1881, ch. 138, §3,21Stat.503 (using a “likely to cause confusion” standard for registration). That could hardly have been by accident. And third, district courts can cancel registrations during infringement litigation, just as they can adjudicate infringement in suits seeking judicial review of registration decisions. See15 U. S. C. §1119; 3 McCarthy §21:20. There is no reason to think that the same district judge in the same case should apply two separate standards of likelihood of confusion. Hargis responds that the text is not actually the same because the registration provision asks whether the marks “resemble” each other,15 U. S. C. §1052(d), while the infringement provision is directed towards the “use in commerce” of the marks, §1114(1). Indeed, according to Hargis, the distinction between “resembl[ance]” and “use” has been key to trademark law for over a century. There is some force to this argument. It is true that “a party opposing an application to register a mark before the Board often relies only on its federal registration, not on any common-law rights in usages not encompassed by its registration,” and “the Board typically analyzes the marks, goods, and channels of trade only as set forth in the application and in the opposer’s registration, regardless of whether the actual usage of the marks by either party differs.” Brief for United States as Amicus Curiae 23; see also id., at 5 (explaining that “the Board typically reviews only the usages encompassed by the registration”) (citing 3 Gilson §9.03[2][a][ii]); 3 McCarthy §20:15, at 20–45 (explaining that for registration “it is the mark as shown in the application and as used on the goods described in the application which must be considered, not the mark as actually used”). This means that unlike in infringement litigation, “[t]he Board’s determination that a likelihood of confusion does or does not exist will not resolve the confusion issue with respect to non-disclosed usages.” Brief for United States as Amicus Curiae 23. Hargis’ argument falls short, however, because it mistakes a reason not to apply issue preclusion in some or even many cases as a reason never to apply issue preclusion. Just because the TTAB does not always consider the same usages as a district court does, it does not follow that the Board applies a different standard to the usages it does consider.[4] If a mark owner uses its mark in ways that are materially the same as the usages included in its registration application, then the TTAB is deciding the same likelihood-of-confusion issue as a district court in infringement litigation. By contrast, if a mark owner uses its mark in ways that are materially unlike the usages in its application, then the TTAB is not deciding the same issue. Thus, if the TTAB does not consider the marketplace usage of the parties’ marks, the TTAB’s decision should “have no later preclusive effect in a suit where actual usage in the marketplace is the paramount issue.” 6 McCarthy §32:101, at 32–246. Materiality, of course, is essential—trivial variations between the usages set out in an application and the use of a mark in the marketplace do not create different “issues,” just as trivial variations do not create different “marks.” See generally 4 id., §23:50, at 23–265 (explaining that “adding descriptive or non-distinctive” elements to another’s mark generally will not negate confusion). Otherwise, a party could escape the preclusive effect of an adverse judgment simply by adding an immaterial feature to its mark. That is not the law. See, e.g., Restatement (Second) of Judgments §27, Comment c, at 252–253 (explaining that “issue” must be understood broadly enough “to prevent repetitious litigation of what is essentially the same dispute”); United States v. Stauffer Chemical Co.,464 U. S. 165,172 (1984) (applying issue preclusion where a party sought to “litigate twice . . . an issue arising . . . from virtually identical facts” because the “factual differences” were “of no legal significance”). A fortiori, if the TTAB considers a different mark altogether, issue preclusion would not apply. Needless to say, moreover, if the TTAB has not decided the same issue as that before the district court, there is no reason why any deference would be warranted. For a similar reason, the Eighth Circuit erred in holding that issue preclusion could not apply here because the TTAB relied too heavily on “appearance and sound.” App. to Pet. for Cert. 10a. Undoubtedly there are cases in which the TTAB places more weight on certain factors than it should. When that happens, an aggrieved party should seek judicial review. The fact that the TTAB may have erred, however, does not prevent preclusion. As Judge Colloton observed in dissent, “ ‘issue preclusion prevent[s] relitigation of wrong decisions just as much as right ones.’ ” 716 F. 3d, at 1029 (quoting Clark v. Clark, 984 F. 2d 272, 273 (CA8 1993)); see also Restatement (Second) of Judgments §28, Comment j, at 284 (explaining that “refusal to give the first judgment preclusive effect should not . . . be based simply on a conclusion that [it] was patently erroneous”).B Hargis also argues that registration is categorically incompatible with issue preclusion because the TTAB uses procedures that differ from those used by district courts. Granted, “[r]edetermination of issues is warranted if there is reason to doubt the quality, extensiveness, or fairness of procedures followed in prior litigation.” Montana, 440 U. S., at 164, n. 11; see also Parklane Hosiery, 439 U. S., at 331, and n. 15 (similar). But again, this only suggests that sometimes issue preclusion might be inappropriate, not that it always is. No one disputes that the TTAB and district courts use different procedures. Most notably, district courts feature live witnesses. Procedural differences, by themselves, however, do not defeat issue preclusion. Equity courts used different procedures than did law courts, but that did not bar issue preclusion. See id., at 333. Nor is there reason to think that the state agency in Elliott used procedures identical to those in federal court; nonetheless, the Court held that preclusion could apply. See 478 U. S., at 796–799. Rather than focusing on whether procedural differences exist—they often will—the correct inquiry is whether the procedures used in the first proceeding were fundamentally poor, cursory, or unfair. See Montana, 440 U. S., at 164, n. 11. Here, there is no categorical “reason to doubt the quality, extensiveness, or fairness,” ibid., of the agency’s procedures. In large part they are exactly the same as in fed-eral court. See 37 CFR §§2.116(a), 2.122(a). For instance, although the “[t]he scope of discovery in Board proceedings . . . . is generally narrower than in court proceedings”—reflecting the fact that there are often fewer usages at issue—the TTAB has adopted almost the whole of Federal Rule of Civil Procedure 26. TTAB Manual §402.01; see also id., §401. It is conceivable, of course, that the TTAB’s procedures may prove ill-suited for a particular issue in a particular case, e.g., a party may have tried to introduce material evidence but was prevented by the TTAB from doing so, or the TTAB’s bar on live testimony may materially prejudice a party’s ability to present its case. The ordinary law of issue preclusion, however, already accounts for those “rare” cases where a “compelling showing of unfairness” can be made. Restatement (Second) of Judgments §28, Comments g and j, at 283–284. The Eighth Circuit likewise erred by concluding that Hargis bore the burden of persuasion before the TTAB. B&B, the party opposing registration, bore the burden, see 37 CFR §2.116(b); TTAB Manual §702.04(a), just as it did in the infringement action. Hargis does not defend the decision below on this ground.C Hargis also contends that the stakes for registration are so much lower than for infringement that issue preclusion should never apply to TTAB decisions. Issue preclusion may be inapt if “the amount in controversy in the first action [was] so small in relation to the amount in controversy in the second that preclusion would be plainly unfair.” Restatement (Second) of Judgments §28, Comment j, at 283–284. After all, “[f]ew . . . litigants would spend $50,000 to defend a $5,000 claim.” Wright & Miller §4423, at 612. Hargis is wrong, however, that this exception to issue preclusion applies to every registration. Tothe contrary: When registration is opposed, there isgood reason to think that both sides will take the matter seriously. The benefits of registration are substantial. Registration is “prima facie evidence of the validity of the registered mark,”15 U. S. C. §1057(b), and is a precondition for a mark to become “incontestable,” §1065. Incontestability is a powerful protection. See, e.g., Park ’N Fly, Inc. v. Dollar Park & Fly, Inc.,469 U. S. 189,194 (1985) (holding that an incontestable mark cannot be challenged as merely descriptive); see also id., at 193 (explaining that “Con-gress determined that . . . ‘trademarks should receive nationally the greatest protection that can be given them’ ” and that “[a]mong the new protections created by the Lanham Act were the statutory provisions that allow a federally registered mark to become incontestable” (quoting S. Rep. No. 1333, 79th Cong., 2d Sess., 6 (1946))). The importance of registration is undoubtedly why Congress provided for de novo review of TTAB decisions in district court. It is incredible to think that a district court’s adjudication of particular usages would not have preclusive effect in another district court. Why would unchallenged TTAB decisions be different? Congress’ creation of this elaborate registration scheme, with so many important rights attached and backed up by plenary review, confirms that registration decisions can be weighty enough to ground issue preclusion.V For these reasons, the Eighth Circuit erred in this case. On remand, the court should apply the following rule: So long as the other ordinary elements of issue preclusion are met, when the usages adjudicated by the TTAB are materially the same as those before the district court, issue preclusion should apply. The judgment of the United States Court of Appeals for the Eighth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion.It is so ordered.Notes1 See Brief for Respondent 28 (acknowledging that administrative “[p]reclusion’s status as part of the common-law backdrop means that courts may presume its application” absent contrary indication from Congress) (citing Astoria 501 U. S., at 110); Brief for Respondent 34 (explaining that Utah Construction determined that “an administrative board’s factfinding . . . could . . . have preclusive effect in an Article III suit raising damages claims over which the board had no jurisdiction”).2 Our dissenting colleagues argue that Utah Construction’s conclu-sion that courts “have not hesitated” to apply administrative preclusion, 384 U. S., at 422, was mistaken and certainly should not be applied to statutes—such as the Lanham Act—enacted prior to 1966. We do not decide who reads the history better. The Court has repeat-edly endorsed Utah Construction and, importantly, neither party chal-lenges its historical accuracy. For the same reason, we do not decide whether such preclusion is unconstitutional because the issue is not before us. 3 Compare15 U. S. C. §1114(1) (“Any person who shall . . . use in commerce any . . . mark in connection with the sale, offering for sale, distribution, or advertising of any goods or services on or in connection with which such use is likely to cause confusion, or to cause mistake, or to deceive . . . shall be liable in a civil action by the registrant for the remedies hereinafter provided” (emphasis added)) with §1052(d) (“No trademark . . . shall be refused registration . . . unless it . . . [c]onsists of or comprises a mark which so resembles a mark registered in the Patent and Trademark Office . . . as to be likely, when used on or in connection with the goods of the applicant, to cause confusion, or to cause mistake, or to deceive . . .” (emphasis added)).4 The parties dispute whether and how often the TTAB considers usages beyond those listed in the application and registration. We do not resolve that dispute here. Suffice it to say that when the TTAB adjudicates a usage within its authority, that adjudication can ground issue preclusion. See Restatement (Second) of Judgments §11 (1980).

576.US.2014_14-103

Respondent ASARCO LLC hired petitioner law firms pursuant to §327(a) of the Bankruptcy Code to assist it in carrying out its duties as a Chapter 11 debtor in possession. See 11 U. S. C. §327(a). When ASARCO emerged from bankruptcy, the law firms filed fee applications requesting fees under §330(a)(1), which permits bankruptcy courts to “award . . . reasonable compensation for actual, necessary services rendered by” §327(a) professionals. ASARCO challenged the applications, but the Bankruptcy Court rejected ASARCO’s objections and awarded the law firms fees for time spent defending the applications. ASARCO appealed to the District Court, which held that the law firms could be awarded fees for defending their fee applications. The Fifth Circuit reversed, holding that §330(a)(1) did not authorize fee awards for defending fee applications. Held: Section §330(a)(1) does not permit bankruptcy courts to award fees to §327(a) professionals for defending fee applications. Pp. 3–13. (a) The American Rule provides the “ ‘basic point of reference’ ” for awards of attorney’s fees: “ ‘Each litigant pays his own attorney’s fees, win or lose, unless a statute or contract provides otherwise.’ ” Hardt v. Reliance Standard Life Ins. Co., 560 U. S. 242 –253. Because the rule is deeply rooted in the common law, see, e.g., Arcambel v. Wiseman, 3 Dall. 306, this Court will not deviate from it “ ‘absent explicit statutory authority,’ ” Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health and Human Resources, 532 U. S. 598 . Departures from the American Rule have been recognized only in “specific and explicit provisions,” Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240 , usually containing language that authorizes the award of “a reasonable attorney’s fee,” “fees,” or “litigation costs,” and referring to a “prevailing party” in the context of an adversarial “action,” see generally Hardt, supra, at 253, and nn. 3–7. Pp. 3–4. (b) Congress did not depart from the American Rule in §330(a)(1) for fee-defense litigation. Section 327(a) professionals are hired to serve an estate’s administrator for the benefit of the estate, and §330(a)(1) authorizes “reasonable compensation for actual, necessary services rendered.” The word “services” ordinarily refers to “labor performed for another,” Webster’s New International Dictionary 2288. Thus, the phrase “ ‘reasonable compensation for services rendered’ necessarily implies loyal and disinterested service in the interest of” a client, Woods v. City Nat. Bank & Trust Co. of Chicago, 312 U. S. 262 . Time spent litigating a fee application against the bankruptcy estate’s administrator cannot be fairly described as “labor performed for”—let alone “disinterested service to”—that administrator. Had Congress wished to shift the burdens of fee-defense litigation under §330(a)(1), it could have done so, as it has done in other Bankruptcy Code provisions, e.g., §110(i)(1)(C). Pp. 4–7. (c) Neither the law firms nor the United States, as amicus curiae, offers a persuasive theory for why §330(a)(1) should override the American Rule in this context. Pp. 7–13. (1) The law firms’ view—that fee-defense litigation is part of the “services rendered” to the estate administrator—not only suffers from an unnatural interpretation of the term “services rendered,” but would require a particularly unusual deviation from the American Rule, as it would permit attorneys to be awarded fees for unsuccessfully defending fee applications when most fee-shifting provisions permit awards only to “a ‘prevailing party,’ ” Hardt, supra, at 253. Pp. 7–8. (2) The Government’s argument is also unpersuasive. Its theory—that fees for fee-defense litigation must be understood as a component of the “reasonable compensation for [the underlying] services rendered” so that compensation for the “actual . . . services rendered” will not be diluted by unpaid time spent litigating fees—cannot be reconciled with the relevant text. Section 330(a)(1) does not authorize courts to award “reasonable compensation,” but “reasonable compensation for actual, necessary services rendered,” and the Government properly concedes that litigation in defense of a fee application is not a “service.” And §330(a)(6), which presupposes compensation “for the preparation of a fee application,” does not suggest that time spent defending a fee application must also be compensable. Commissioner v. Jean, 496 U. S. 154 , distinguished. The Government’s theory ultimately rests on the flawed policy argument that a “judicial exception” is needed to compensate fee-defense litigation and safeguard Congress’ aim of ensuring that talented attorneys take on bankruptcy work. But since no attorneys are entitled to such fees absent express statutory authorization, requiring bankruptcy attorneys to bear the costs of their fee-defense litigation under §330(a)(1) creates no disincentive to bankruptcy practice. And even if this Court believed that uncompensated fee-defense litigation would fall particularly hard on the bankruptcy bar, it has no “roving authority . . . to allow counsel fees . . . whenever [it] might deem them warranted,” Alyeska Pipeline, supra, at 260. Pp. 8–13. 751 F. 3d 291, affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, and Alito, JJ., joined, and in which Sotomayor, J., joined as to all but Part III–B–2. Sotomayor, J., filed an opinion concurring in part and concurring in the judgment. Breyer, J., filed a dissenting opinion, in which Ginsburg and Kagan, JJ., joined.

Section 327(a) of the Bankruptcy Code allows bank-ruptcy trustees to hire attorneys, accountants, and other professionals to assist them in carrying out their statutory duties. 11 U. S. C. §327(a). Another provision, §330(a)(1), states that a bankruptcy court “may award . . . reasonable compensation for actual, necessary services rendered by” those professionals. The question before us is whether §330(a)(1) permits a bankruptcy court to award attorney’s fees for work performed in defending a fee application in court. We hold that it does not and therefore affirm the judgment of the Court of Appeals. I In 2005, respondent ASARCO LLC, a copper mining, smelting, and refining company, found itself in financial trouble. Faced with falling copper prices, debt, cash flow deficiencies, environmental liabilities, and a striking work force, ASARCO filed for Chapter 11 bankruptcy. As in many Chapter 11 bankruptcies, no trustee was appointed and ASARCO—the “ ‘debtor in possession’ ”—administered the bankruptcy estate as a fiduciary for the estate’s creditors. §§1101(1), 1107(a). Relying on §327(a) of the Bankruptcy Code, which permits trustees to employ attorneys and other professionals to assist them in their duties, ASARCO obtained the Bankruptcy Court’s permission to hire two law firms, petitioners Baker Botts L.L.P. and Jordan, Hyden, Womble, Culbreth & Holzer, P.C., to provide legal representation during the bankruptcy.[1] Among other services, the firms prosecuted fraudulent-transfer claims against ASARCO’s parent company and ultimately obtained a judgment against it worth between $7 and $10 billion. This judgment contributed to a successful reorganization in which all of ASARCO’s creditors were paid in full. After over four years in bankruptcy, ASARCO emerged in 2009 with $1.4 billion in cash, little debt, and resolution of its environmental liabilities. The law firms sought compensation under §330(a)(1), which provides that a bankruptcy court “may award . . . reasonable compensation for actual, necessary services rendered by” professionals hired under §327(a). As required by the bankruptcy rules, the two firms filed fee applications. Fed. Rule Bkrtcy. Proc. 2016(a). ASARCO, controlled once again by its parent company, challenged the compensation requested in the applications. After extensive discovery and a 6-day trial on fees, the Bankruptcy Court rejected ASARCO’s objections and awarded the firms approximately $120 million for their work in the bankruptcy proceeding plus a $4.1 million enhancement for exceptional performance. The court also awarded the firms over $5 million for time spent litigating in defense of their fee applications. ASARCO appealed various aspects of the award to the District Court. As relevant here, the court held that the firms could recover fees for defending their fee application. The Court of Appeals for the Fifth Circuit reversed. It reasoned that the American Rule—the rule that each side must pay its own attorney’s fees—“applies absent explicit statutory . . . authority” to the contrary and that “the Code contains no statutory provision for the recovery of attor-ney fees for defending a fee application.” In re ASARCO, L.L.C., 751 F. 3d 291, 301 (2014) (internal quotation marks omitted). It observed that §330(a)(1) provides “that professional services are compensable only if they are likely to benefit a debtor’s estate or are necessary to case administration.” Id., at 299. Because “[t]he primary beneficiary of a professional fee application, of course, is the professional,” compensation for litigation defending that application does not fall within §330(a)(1). Ibid. We granted certiorari, 573 U. S. ___ (2014), and now affirm. II A “Our basic point of reference when considering the award of attorney’s fees is the bedrock principle known as the American Rule: Each litigant pays his own attorney’s fees, win or lose, unless a statute or contract provides otherwise.” Hardt v. Reliance Standard Life Ins. Co., 560 U. S. 242 –253 (2010) (internal quotation marks omitted). The American Rule has roots in our common law reaching back to at least the 18th century, see Arcambel v. Wiseman, 3 Dall. 306 (1796), and “[s]tatutes which invade the common law are to be read with a presumption favoring the retention of long-established and familiar [legal] principles,” Fogerty v. Fantasy, Inc., 510 U. S. 517, 534 (1994) (internal quotation marks and ellipsis omitted). We consequently will not deviate from the American Rule “ ‘absent explicit statutory authority.’ ” Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health and Human Resources, 532 U. S. 598, 602 (2001) (quoting Key Tronic Corp. v. United States, 511 U. S. 809, 814 (1994) ). We have recognized departures from the American Rule only in “specific and explicit provisions for the allowance of attorneys’ fees under selected statutes.” Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240, 260 (1975) . Although these “[s]tatutory changes to [the American Rule] take various forms,” Hardt, supra, at 253, they tend to authorize the award of “a reasonable attorney’s fee,” “fees,” or “litigation costs,” and usually refer to a “prevailing party” in the context of an adversarial “action,” see, e.g., 28 U. S. C. §2412(d)(1)(A); 42 U. S. C. §§1988(b), 2000e–5(k); see generally Hardt, supra, at 253, and nn. 3–7 (collecting examples). The attorney’s fees provision of the Equal Access to Justice Act offers a good example of the clarity we have required to deviate from the American Rule. See 28 U. S. C. §2412(d)(1)(A). That section provides that “a court shall award to a prevailing party other than the United States fees and other expenses . . . incurred by that party in any civil action (other than cases sounding in tort) . . . brought by or against the United States” under certain conditions. Ibid. As our decision in Commissioner v. Jean, 496 U. S. 154 (1990) , reveals, there could be little dispute that this provision—which mentions “fees,” a “prevailing party,” and a “civil action”—is a “fee-shifting statut[e]” that trumps the American Rule, id., at 161. B Congress did not expressly depart from the American Rule to permit compensation for fee-defense litigation by professionals hired to assist trustees in bankruptcy proceedings. Section 327(a) authorizes the employment of such professionals, providing that a “trustee, with the court’s approval, may employ one or more attorneys, accountants, appraisers, auctioneers, or other professional persons, that do not hold or represent an interest adverse to the estate, and that are disinterested persons, to represent or assist [him] in carrying out [his] duties.” In other words, §327(a) professionals are hired to serve the administrator of the estate for the benefit of the estate. Section 330(a)(1) in turn authorizes compensation for these professionals as follows: “After notice to the parties in interest and the United States Trustee and a hearing, and subject to sec-tions 326, 328, and 329, the court may award to a trustee, a consumer privacy ombudsman appointed under section 332, an examiner, an ombudsman appointed under section 333, or a professional person employed under section 327 or 1103— “(A) reasonable compensation for actual, necessary services rendered by the trustee, examiner, ombudsman, professional person, or attorney and by any paraprofessional person employed by any such person; and “(B) reimbursement for actual, necessary expenses.” (Emphasis added.) This text cannot displace the American Rule with respect to fee-defense litigation. To be sure, the phrase “reason-able compensation for actual, necessary services rendered” permits courts to award fees to attorneys for work done to assist the administrator of the estate, as the Bankruptcy Court did here when it ordered ASARCO to pay roughly $120 million for the firms’ work in the bankruptcy proceeding. No one disputes that §330(a)(1) authorizes an award of attorney’s fees for that kind of work. See Alyeska Pipeline, supra, at 260, and n. 33 (listing §330(a)(1)’s predecessor as an example of a provision authorizing attorney’s fees). But the phrase “reasonable compensation for actual, necessary services rendered” neither specifi-cally nor explicitly authorizes courts to shift the costs ofadversarial litigation from one side to the other—in this case, from the attorneys seeking fees to the administrator of the estate—as most statutes that displace the American Rule do. Instead, §330(a)(1) provides compensation for all §327(a) professionals—whether accountant, attorney, or auctioneer—for all manner of work done in service of the estate administrator. More specifically, §330(a)(1) allows “reasonable compensation” only for “actual, necessary services rendered.” (Emphasis added.) That qualification is significant. The word “services” ordinarily refers to “labor performed for another.” Webster’s New International Dictionary 2288 (def. 4) (2d ed. 1934); see also Black’s Law Dictionary 1607 (3d ed. 1933) (“duty or labor to be rendered by one person to another”); Oxford English Dictionary 517 (def. 19) (1933) (“action of serving, helping or benefiting; conduct tending to the welfare or advantage of another”).[2] Thus, in a case addressing §330(a)’s predecessor, this Court concluded that the phrase “ ‘reasonable compensation for services rendered’ necessarily implies loyal and disinterested service in the interest of” a client. Woods v. City Nat. Bank & Trust Co. of Chicago, 312 U. S. 262, 268 (1941) ; accord, American United Mut. Life Ins. Co. v. Avon Park, 311 U. S. 138, 147 (1940) . Time spent litigating a fee application against the administrator of a bankruptcy estate cannot be fairly described as “labor performed for”—let alone “disinterested service to”—that administrator. This legislative decision to limit “compensation” to “services rendered” is particularly telling given that other provisions of the Bankruptcy Code expressly transfer the costs of litigation from one adversarial party to the other. Section 110(i), for instance, provides that “[i]f a bank-ruptcy petition preparer . . . commits any act that the court finds to be fraudulent, unfair, or deceptive, on the motion of the debtor, trustee, United States trustee (or the bankruptcy administrator, if any),” the bankruptcy court must “order the bankruptcy petition preparer to pay the debtor . . . reasonable attorneys’ fees and costs in moving for damages under this subsection.” §110(i)(1)(C). Had Congress wished to shift the burdens of fee-defense litigation under §330(a)(1) in a similar manner, it easily could have done so. We accordingly refuse “to invade the legislature’s province by redistributing litigation costs” here. Alyeska Pipeline, 421 U. S., at 271. III The law firms, the United States as amicus curiae, and the dissent resist this straightforward interpretation of the statute. The law firms and the Government each offer a theory for why §330(a)(1) expressly overrides the American Rule in the context of litigation in defense of a fee application, and the dissent embraces the latter. Neither theory is persuasive. A We begin with the law firms’ approach. According to the firms, fee-defense litigation is part of the “services rendered” to the estate administrator under §330(a)(1). See Brief for Petitioners 23–30. As explained above, that reading is untenable. The term “services” in this provision cannot be read to encompass adversarial fee-defense litigation. See Part II–B, supra. Even the dissent agrees on this point. See post, at 1 (opinion of Breyer, J.). Indeed, reading “services” in this manner could end up compensating attorneys for the unsuccessful defense of a fee application. The firms insist that “estates do benefit from fee defenses”—and thus receive a “service” under §330(a)(1)—because “the estate has an interest in obtaining a just determination of the amount it should pay its professionals.” Brief for Petitioners 25–26 (internal quotation marks omitted). But that alleged interest—and hence the supposed provision of a “service”—exists whether or not a §327(a) professional prevails in his fee dispute. We decline to adopt a reading of §330(a)(1) that would allow courts to pay professionals for arguing for fees they were found never to have been entitled to in the first place. Such a result would not only require an unnatural interpretation of the term “services rendered,” but a particu-larly unusual deviation from the American Rule as well, as“[m]ost fee-shifting provisions permit a court to award attorney’s fees only to a ‘prevailing party,’ ” a “ ‘substantially prevailing’ party,” or “a ‘successful’ litigant,” Hardt, 560 U. S., at 253 (footnote omitted). There is no indication that Congress departed from the American Rule in §330(a)(1) with respect to fee-defense litigation, let alone that it did so in such an unusual manner. B The Government’s theory, embraced by the dissent, fares no better. Although the United States agrees that “the defense of a fee application does not itself qualify as an independently compensable service,” it nonetheless contends that “compensation for such work is properly viewed as part of the compensation for the underlying services in the bankruptcy proceeding.” Brief for United States as Amicus Curiae 25. According to the Government, if an attorney is not repaid for his time spent successfully litigating fees, his compensation for his actual “services rendered” to the estate administrator in the underlying proceeding will be diluted. Id., at 18. The United States thus urges us to treat fees for fee-defense work “as a component of ‘reasonable compensation.’ ” Id., at 33; accord, post, at 1 (Breyer, J., dissenting). We refuse to do so for several reasons. 1 First and foremost, the Government’s theory cannot be reconciled with the relevant text. Section 330(a)(1) does not authorize courts to award “reasonable compensation” simpliciter, but “reasonable compensation for actual, necessary services rendered by” the §327(a) professional. §330(a)(1)(A) (emphasis added). Here, the contested award was tied to the firms’ work on the fee-defense litigation and is correctly understood only as compensation for that work. The Government and the dissent properly concede that litigation in defense of a fee application is not a “service” within the meaning of §330(a)(1); it follows that the contested award was not “compensation” for a “service.” Thus, the only way to reach their reading of the statute would be to excise the phrase “for actual, necessary services rendered” from the statute.[3] Contrary to the Government’s assertion, §330(a)(6) does not presuppose that courts are free to award compensation based on work that does not qualify as a service to the estate administrator. That provision specifies that “[a]ny compensation awarded for the preparation of a fee application shall be based on the level and skill reasonably required to prepare the application.” The Government argues that because time spent preparing a fee application is compensable, time spent defending it must be too. But the provision cuts the other way. A §327(a) professional’s preparation of a fee application is best understood as a “servic[e] rendered” to the estate administrator under §330(a)(1), whereas a professional’s defense of that application is not. By way of analogy, it would be natural to describe a car mechanic’s preparation of an itemized bill as part of his “services” to the customer because it allows a customer to understand—and, if necessary, dispute—his expenses. But it would be less natural to describe a subsequent court battle over the bill as part of the “services rendered” to the customer. The Government used to understand that time spent preparing a fee application was different from time spent defending one for the purposes of §330(a)(1). Just a few years ago, the U. S. Trustee explained that “[r]easonable charges for preparing . . . fee applications . . . are compensable . . . because the preparation of a fee application is not required for lawyers practicing in areas other than bankruptcy as a condition to getting paid.” 78 Fed. Reg. 36250 (2013) (emphasis deleted). By contrast, “time spent . . . defending . . . fee applications” is ordinarily “not compensable,” the Trustee observed, as such time can be “properly characterized as work that is for the benefit of the professional and not the estate.” Ibid. To support its broader interpretation of §330(a)(6), the Government, echoed by the dissent, relies on our remark in Jean that “[w]e find no textual or logical argument for treating so differently a party’s preparation of a fee application and its ensuing efforts to support that same application.” 496 U. S., at 162; see post, at 7. But that use of Jean begs the question. Jean addressed a statutory provision that everyone agreed authorized court-awarded fees for fee-defense litigation. 496 U. S., at 162. The “only dispute” in that context was over what “finding [was] necessary to support such an award.” Ibid. In resolving that issue, the Court declined to treat fee-application and fee-litigation work differently given that the relevant statutory text—“a court shall award to a prevailing party . . . fees and other expenses . . . incurred by that party in any civil action”—could not support such a distinction. Id., at 158. Here, by contrast, the operative language—“reasonable compensation for actual, necessary services rendered”—reaches only the fee-application work. The fact that the provision at issue in Jean “did not mention fee-defense work,” post, at 5, is thus irrelevant. In any event, the Government’s textual foothold for its argument is too insubstantial to support a deviation from the American Rule. The open-ended phrase “reasonable compensation,” standing alone, is not the sort of “specific and explicit provisio[n]” that Congress must provide in order to alter this default rule. Alyeska Pipeline, 421 U. S., at 260. 2 Ultimately, the Government’s theory rests on a flawed and irrelevant policy argument. The United States contends that awarding fees for fee-defense litigation is a “judicial exception” necessary to the proper functioning of the Bankruptcy Code. Brief for United States as Amicus Curiae 15, n. 7 (internal quotation marks omitted). Absent this exception, it warns, fee-defense litigation will dilute attorney’s fees and result in bankruptcy lawyers receiving less compensation than nonbankruptcy lawyers, thereby undermining the congressional aim of ensuring that talented attorneys will take on bankruptcy work. Accord, post, at 3. As an initial matter, we find this policy argument unconvincing. In our legal system, no attorneys, regardless of whether they practice in bankruptcy, are entitled to receive fees for fee-defense litigation absent express statutory authorization. Requiring bankruptcy attorneys to pay for the defense of their fees thus will not result inany disparity between bankruptcy and nonbankruptcy lawyers.[4] The United States nonetheless contends that uncompensated fee litigation in bankruptcy will be particularly costly because multiple parties in interest may object to fee applications, whereas nonbankruptcy fee litigation typically involves just a lawyer and his client. But this argument rests on unsupported predictions of how the statutory scheme will operate in practice, and the Government’s conduct in this case reveals the perils associ-ated with relying on such prognostications to interpretstatutes: The United States took the opposite view below, asserting that “requiring a professional to bear the normal litigation costs of litigating a contested request for payment . . . dilutes a bankruptcy fee award no more than any litigation over professional fees.” Reply Brief for Appellant United States Trustee in No. 11–290 (SD Tex.), p. 15. The speed with which the Government has changed its tune offers a good argument against substituting policy-oriented predictions for statutory text. More importantly, we would lack the authority to rewrite the statute even if we believed that uncompensated fee litigation would fall particularly hard on the bank-ruptcy bar. “Our unwillingness to soften the import of Con-gress’ chosen words even if we believe the words lead to a harsh outcome is longstanding,” and that is no less true in bankruptcy than it is elsewhere. Lamie v. United States Trustee, 540 U. S. 526, 538 (2004) . Whether or not the Government’s theory is desirable as a matter of policy, Congress has not granted us “roving authority . . . to allow counsel fees . . . whenever [we] might deem them warranted.” Alyeska Pipeline, supra, at 260. Our job is to followthe text even if doing so will supposedly “undercut a basic objective of the statute,” post, at 3. Section 330(a)(1) itself does not authorize the award of fees for defending a fee application, and that is the end of the matter. * * * As we long ago observed, “The general practice of the United States is in opposition” to forcing one side to pay the other’s attorney’s fees, and “even if that practice [is] not strictly correct in principle, it is entitled to the respect of the court, till it is changed, or modified, by statute.” Arcambel, 3 Dall., at 306 (emphasis deleted). We follow that approach today. Because §330(a)(1) does not explic-itly override the American Rule with respect to fee-defense litigation, it does not permit bankruptcy courts to award compensation for such litigation. We therefore affirm the judgment of the Court of Appeals. It is so ordered.Notes 1 Although §327(a) directly applies only to trustees, §1107(a) gives Chapter 11 debtors in possession the same authority as trustees to retain §327(a) professionals. For the sake of simplicity, we refer to §327(a) alone throughout this opinion. 2 Congress added the phrase “reasonable compensation for the services rendered” to federal bankruptcy law in 1934. Act of June 7, 1934, §77B(c)(9), 48Stat. 917. We look to the ordinary meaning of those words at that time. 3 The dissent’s focus on reasonable compensation is therefore a red herring. See post, at 5–6. The question is not whether an award for fee-defense work would be “reasonable,” but whether such work is compensable in the first place. 4 To the extent the United States harbors any concern about the possibility of frivolous objections to fee applications, we note that “Federal Rule of Bankruptcy Procedure 9011—bankruptcy’s analogue to Civil Rule 11—authorizes the court to impose sanctions for bad-faith litigation conduct, which may include ‘an order directing payment . . . of some or all of the reasonable attorneys’ fees and other expenses incurred as a direct result of the violation.’ ” Law v. Siegel, 571 U. S. ___, ___ (2014) (slip op., at 12).

576.US.2014_13-1175

Petitioner, the city of Los Angeles (City), requires hotel operators to record and keep specific information about their guests on the premises for a 90-day period. Los Angeles Municipal Code §41.49. These records “shall be made available to any officer of the Los Angeles Police Department for inspection . . . at a time and in a manner that minimizes any interference with the operation of the business,” §41.49(3)(a), and a hotel operator’s failure to make the records available is a criminal misdemeanor, §11.00(m). Respondents, a group of motel operators and a lodging association, brought a facial challenge to §41.49(3)(a) on Fourth Amendment grounds. The District Court entered judgment for the City, finding that respondents lacked a reasonable expectation of privacy in their records. The Ninth Circuit subsequently reversed, determining that inspections under §41.49(3)(a) are Fourth Amendment searches and that such searches are unreasonable under the Fourth Amendment because hotel owners are subjected to punishment for failure to turn over their records without first being afforded the opportunity for precompliance review. Held: 1. Facial challenges under the Fourth Amendment are not categorically barred or especially disfavored. Pp. 4–8. (a) Facial challenges to statutes—as opposed to challenges to particular applications of statutes—have been permitted to proceed under a diverse array of constitutional provisions. See, e.g., Sorrell v. IMS Health Inc., 564 U. S. ___ ( First Amendment); District of Columbia v. Heller, 554 U. S. 570 ( Second Amendment). The Fourth Amendment is no exception. Sibron v. New York, 392 U. S. 40 , distinguished. This Court has entertained facial challenges to statutes authorizing warrantless searches, declaring them, on several occasions, facially invalid, see, e.g., Chandler v. Miller, 520 U. S. 305 –309. Pp. 4–7. (b) Petitioner contends that facial challenges to statutes authorizing warrantless searches must fail because they will never be unconstitutional in all applications, but this Court’s precedents demonstrate that such challenges can be brought, and can succeed. Under the proper facial-challenge analysis, only applications of a statute in which the statute actually authorizes or prohibits conduct are considered. See, e.g., Planned Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833 . When addressing a facial challenge to a statute authorizing warrantless searches, the proper focus is on searches that the law actually authorizes and not those that could proceed irrespective of whether they are authorized by the statute, e.g., where exigent circumstances, a warrant, or consent to search exists. Pp. 7–8. 2. Section 41.49(3)(a) is facially unconstitutional because it fails to provide hotel operators with an opportunity for precompliance review. Pp. 9–17. (a) “ ‘[S]earches conducted outside the judicial process . . . are per se unreasonable under the Fourth Amendment—subject only to a few . . . exceptions.’ ” Arizona v. Gant, 556 U. S. 332 . One exception is for administrative searches. See Camara v. Municipal Court of City and County of San Francisco, 387 U. S. 523 . To be constitutional, the subject of an administrative search must, among other things, be afforded an opportunity to obtain precompliance review before a neutral decisionmaker. See See v. Seattle, 387 U. S. 541 . Assuming the administrative search exception otherwise applies here, §41.49 is facially invalid because it fails to afford hotel operators any opportunity for precompliance review. To be clear, a hotel owner must only be afforded an opportunity for precompliance review; actual review need occur only when a hotel operator objects to turning over the records. This opportunity can be provided without imposing onerous burdens on law enforcement. For instance, officers in the field can issue administrative subpoenas without probable cause that a regulation is being infringed. This narrow holding does not call into question those parts of §41.49 requiring hotel operators to keep records nor does it prevent police from obtaining access to those records where a hotel operator consents to the search, where the officer has a proper administrative warrant, or where some other exception to the warrant requirement applies. Pp. 9–13. (b) Petitioner’s argument that the ordinance is facially valid under the more relaxed standard for closely regulated industries is rejected. See Marshall v. Barlow’s, Inc., 436 U. S. 307 . This Court has only recognized four such industries, and nothing inherent in the operation of hotels poses a comparable clear and significant risk to the public welfare. Additionally, because the majority of regulations applicable to hotels apply to many businesses, to classify hotels as closely regulated would permit what has always been a narrow exception to swallow the rule. But even if hotels were closely regulated, §41.49 would still contravene the Fourth Amendment as it fails to satisfy the additional criteria that must be met for searches of closely regulated industries to be reasonable. See New York v. Burger, 482 U. S. 691 –703. Pp. 13–17. 738 F. 3d 1058, affirmed. Sotomayor, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined. Scalia, J., filed a dissenting opinion, in which Roberts, C. J., and Thomas, J., joined. Alito, J., filed a dissenting opinion, in which Thomas, J., joined.

Respondents brought a Fourth Amendment challenge to a provision of the Los Angeles Municipal Code that compels “[e]very operator of a hotel to keep a record” containing specified information concerning guests and to make this record “available to any officer of the Los Angeles Police Department for inspection” on demand. Los Angeles Municipal Code §§41.49(2), (3)(a), (4) (2015). The questions presented are whether facial challenges to statutes can be brought under the Fourth Amendment and, if so, whether this provision of the Los Angeles Municipal Code is facially invalid. We hold facial challenges can be brought under the Fourth Amendment. We further hold that the provision of the Los Angeles Municipal Code that requires hotel operators to make their registries available to the police on demand is facially unconstitutional because it penalizes them for declining to turn over their records without affording them any opportunity for precompliance review. I A Los Angeles Municipal Code (LAMC) §41.49 requires hotel operators to record information about their guests, including: the guest’s name and address; the number of people in each guest’s party; the make, model, and license plate number of any guest’s vehicle parked on hotel property; the guest’s date and time of arrival and scheduled departure date; the room number assigned to the guest; the rate charged and amount collected for the room; and the method of payment. §41.49(2). Guests without reservations, those who pay for their rooms with cash, and any guests who rent a room for less than 12 hours must present photographic identification at the time of check-in, and hotel operators are required to record the number and expiration date of that document. §41.49(4). For those guests who check in using an electronic kiosk, the hotel’s records must also contain the guest’s credit card information. §41.49(2)(b). This information can be maintained in either electronic or paper form, but it must be “kept on the hotel premises in the guest reception or guest check-in area or in an office adjacent” thereto for a period of 90 days. §41.49(3)(a). Section 41.49(3)(a)—the only provision at issue here—states, in pertinent part, that hotel guest records “shall be made available to any officer of the Los Angeles Police Department for inspection,” provided that “[w]henever possible, the inspection shall be conducted at a time and in a manner that minimizes any interference with the operation of the business.” A hotel operator’s failure to make his or her guest records available for police inspection is a misdemeanor punishable by up to six months in jail and a $1,000 fine. §11.00(m) (general provision applicable to entire LAMC). B In 2003, respondents, a group of motel operators along with a lodging association, sued the city of Los Angeles (City or petitioner) in three consolidated cases challenging the constitutionality of §41.49(3)(a). They sought declaratory and injunctive relief. The parties “agree[d] that the sole issue in the . . . action [would be] a facial constitu-tional challenge” to §41.49(3)(a) under the Fourth Amend-ment. App. 195. They further stipulated that respondents have been subjected to mandatory record inspections under the ordinance without consent or a warrant. Id., at 194–195. Following a bench trial, the District Court entered judgment in favor of the City, holding that respondents’ facial challenge failed because they lacked a reasonable expectation of privacy in the records subject to inspection. A divided panel of the Ninth Circuit affirmed on the same grounds. 686 F. 3d 1085 (2012). On rehearing en banc, however, the Court of Appeals reversed. 738 F. 3d 1058, 1065 (2013). The en banc court first determined that a police officer’s nonconsensual inspection of hotel records under §41.49 is a Fourth Amendment “search” because “[t]he business records covered by §41.49 are the hotel’s private property” and the hotel therefore “has the right to exclude others from prying into the[ir] contents.” Id., at 1061. Next, the court assessed “whether the searches authorized by §41.49 are reasonable.” Id., at 1063. Relying on Donovan v. Lone Steer, Inc., 464 U. S. 408 (1984) , and See v. Seattle, 387 U. S. 541 (1967) , the court held that §41.49 is facially unconstitutional “as it authorizes inspections” of hotel records “without affording an opportunity to ‘obtain judicial review of the reasonableness of the demand prior to suffering penalties for refusing to comply.’ ” 738 F. 3d, at 1065 (quoting See, 387 U. S., at 545). Two dissenting opinions were filed. The first dissent argued that facial relief should rarely be available for Fourth Amendment challenges, and was inappropriate here because the ordinance would be constitutional in those circumstances where police officers demand access to hotel records with a warrant in hand or exigent circumstances justify the search. 738 F. 3d, at 1065–1070 (opinion of Tallman, J.). The second dissent conceded that inspections under §41.49 constitute Fourth Amendment searches, but faulted the majority for assessing the reasonableness of these searches without accounting for the weakness of the hotel operators’ privacy interest in the content of their guest registries. Id., at 1070–1074 (opinion of Clifton, J.). We granted certiorari, 574 U. S. ___ (2014), and now affirm. II We first clarify that facial challenges under the Fourth Amendment are not categorically barred or especially disfavored. A A facial challenge is an attack on a statute itself as opposed to a particular application. While such challenges are “the most difficult . . . to mount successfully,” United States v. Salerno, 481 U. S. 739, 745 (1987) , the Court has have never held that these claims cannot be brought under any otherwise enforceable provision of the Constitution. Cf. Fallon, Fact and Fiction About Facial Chal-lenges, 99 Cal. L. Rev. 915, 918 (2011) (pointing to several Terms in which “the Court adjudicated more facial challenges on the merits than it did as-applied challenges”). Instead, the Court has allowed such challenges to proceed under a diverse array of constitutional provisions. See, e.g., Sorrell v. IMS Health Inc., 564 U. S. ___ (2011) ( First Amendment); District of Columbia v. Heller, 554 U. S. 570 (2008) ( Second Amendment); Chicago v. Morales, 527 U. S. 41 (1999) (Due Process Clause of the Fourteenth Amendment); Kraft Gen. Foods, Inc. v. Iowa Dept. of Revenue and Finance, 505 U. S. 71 (1992) (Foreign Commerce Clause). Fourth Amendment challenges to statutes authorizing warrantless searches are no exception. Any claim to the contrary reflects a misunderstanding of our decision in Sibron v. New York, 392 U. S. 40 (1968) . In Sibron, two criminal defendants challenged the constitutionality of a statute authorizing police to, among other things, “ ‘stop any person abroad in a public place whom [they] reason-ably suspec[t] is committing, has committed or is about to commit a felony.” Id., at 43 (quoting then N. Y. Code Crim. Proc. §180–a). The Court held that the search of one of the defendants under the statute violated the Fourth Amendment, 392 U. S., at 59, 62, but refused to opine more broadly on the statute’s validity, stating that “[t]he constitutional validity of a warrantless search is pre-eminently the sort of question which can only be decided in the concrete factual context of the individual case.” Id., at 59. This statement from Sibron—which on its face might suggest an intent to foreclose all facial challenges to statutes authorizing warrantless searches—must be understood in the broader context of that case. In the same section of the opinion, the Court emphasized that the “operative categories” of the New York law at issue were “susceptible of a wide variety of interpretations,” id., at 60, and that “[the law] was passed too recently for the State’s highest court to have ruled upon many of the questions involving potential intersections with federal constitutional guarantees,” id., at 60, n. 20. Sibron thus stands for the simple proposition that claims for facial relief under the Fourth Amendment are unlikely to succeed when there is substantial ambiguity as to what conduct a statute authorizes: Where a statute consists of “extraordinarily elastic categories,” it may be “impossible to tell” whether and to what extent it deviates from the requirements of the Fourth Amendment. Id., at 59, 61, n. 20. This reading of Sibron is confirmed by subsequent precedents. Since Sibron, the Court has entertained facial challenges under the Fourth Amendment to statutes authorizing warrantless searches. See, e.g., Vernonia School District 47J v. Acton, 515 U. S. 646, 648 (1995) (“We granted certiorari to decide whether” petitioner’s student athlete drug testing policy “violates the Fourth and Fourteenth Amendments to the United States Constitution”); Skinner v. Railway Labor Executives’ Assn., 489 U. S. 602, 633, n. 10 (1989) (“[R]espondents have challenged the administrative scheme on its face. We deal therefore with whether the [drug] tests contemplated by the regulation can ever be conducted”); cf. Illinois v. Krull, 480 U. S. 340, 354 (1987) (“[A] person subject to a statute authorizing searches without a warrant or probable cause may bring an action seeking a declaration that the statute is unconstitutional and an injunction barring its implementation”). Perhaps more importantly, the Court has on numerous occasions declared statutes facially invalid under the Fourth Amendment. For instance, in Chandler v. Miller, 520 U. S. 305 –309 (1997), the Court struck down a Georgia statute requiring candidates for certain state offices to take and pass a drug test, concluding that this “requirement . . . [did] not fit within the closely guarded category of constitutionally permissible suspicionless searches.” Similar examples abound. See, e.g., Ferguson v. Charleston, 532 U. S. 67, 86 (2001) (holding that a hospital policy authorizing “nonconsensual, warrantless, and suspicionless searches” contravened the Fourth Amendment); Payton v. New York, 445 U. S. 573, 574, 576 (1980) (holding that a New York statute “authoriz[ing] police officers to enter a private residence without a warrant and with force, if necessary, to make a routine felony arrest” was “not consistent with the Fourth Amendment”); Torres v. Puerto Rico, 442 U. S. 465, 466, 471 (1979) (holding that a Puerto Rico statute authorizing “police to search the luggage of any person arriving in Puerto Rico from the United States” was unconstitutional because it failed to require either probable cause or a warrant). B Petitioner principally contends that facial challenges to statutes authorizing warrantless searches must fail because such searches will never be unconstitutional in all applications. Cf. Salerno, 481 U. S., at 745 (to obtain facial relief the party seeking it “must establish that no set of circumstances exists under which the [statute] would be valid”). In particular, the City points to situations where police are responding to an emergency, where the subject of the search consents to the intrusion, and where police are acting under a court-ordered warrant. See Brief for Petitioner 19–20. While petitioner frames this argument as an objection to respondents’ challenge in this case, its logic would preclude facial relief in every Fourth Amendment challenge to a statute authorizing warrantless searches. For this reason alone, the City’s argument must fail: The Court’s precedents demonstrate not only that facial challenges to statutes authorizing warrantless searches can be brought, but also that they can succeed. See Part II–A, supra. Moreover, the City’s argument misunderstands how courts analyze facial challenges. Under the most exacting standard the Court has prescribed for facial challenges, a plaintiff must establish that a “law is unconstitutional in all of its applications.” Washington State Grange v. Washington State Republican Party, 552 U. S. 442, 449 (2008) . But when assessing whether a statute meets this standard, the Court has considered only applications of the statute in which it actually authorizes or prohibits conduct. For instance, in Planned Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833 (1992) , the Court struck down a provision of Pennsylvania’s abortion law that required a woman to notify her husband before obtaining an abortion. Those defending the statute argued that facial relief was inappropriate because most women voluntarily notify their husbands about a planned abortion and for them the law would not impose an undue burden. The Court rejected this argument, explaining: The “[l]egislation is measured for consistency with the Constitution by its impact on those whose conduct it affects. . . . The proper focus of the constitutional inquiry is the group for whom the law is a restriction, not the group for whom the law is irrelevant.” Id., at 894. Similarly, when addressing a facial challenge to a statute authorizing warrantless searches, the proper focus of the constitutional inquiry is searches that the law actually authorizes, not those for which it is irrelevant. If exigency or a warrant justifies an officer’s search, the subject of the search must permit it to proceed irrespective of whether it is authorized by statute. Statutes authorizing warrantless searches also do no work where the subject of a search has consented. Accordingly, the constitutional “applications” that petitioner claims prevent facial relief here are irrelevant to our analysis because they do not involve actual applications of the statute.[1] III Turning to the merits of the particular claim before us, we hold that §41.49(3)(a) is facially unconstitutional because it fails to provide hotel operators with an opportu-nity for precompliance review. A The Fourth Amendment protects “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” It further provides that “no Warrants shall issue, but upon probable cause.” Based on this constitutional text, the Court has repeatedly held that “ ‘searches conducted outside the judicial process, without prior approval by [a] judge or [a] magistrate [judge], are per se unreasonable . . . subject only to a few specifically established and well-delineated exceptions.’ ” Arizona v. Gant, 556 U. S. 332, 338 (2009) (quoting Katz v. United States, 389 U. S. 347, 357 (1967) ). This rule “applies to commercial premises as well as to homes.” Marshall v. Barlow’s, Inc., 436 U. S. 307, 312 (1978) . Search regimes where no warrant is ever required may be reasonable where “ ‘special needs . . . make the warrant and probable-cause requirement impracticable,’ ” Skinner, 489 U. S., at 619 (quoting Griffin v. Wisconsin, 483 U. S. 868, 873 (1987) (some internal quotation marks omitted)), and where the “primary purpose” of the searches is “[d]istinguishable from the general interest in crime control,” Indianapolis v. Edmond, 531 U. S. 32, 44 (2000) . Here, we assume that the searches authorized by §41.49 serve a “special need” other than conducting criminal investigations: They ensure compliance with the recordkeeping requirement, which in turn deters criminals from operating on the hotels’ premises.[2] The Court has referred to this kind of search as an “administrative searc[h].” Camara v. Municipal Court of City and County of San Francisco, 387 U. S. 523, 534 (1967) . Thus, we consider whether §41.49 falls within the administrative search exception to the warrant requirement. The Court has held that absent consent, exigent circumstances, or the like, in order for an administrative search to be constitutional, the subject of the search must be afforded an opportunity to obtain precompliance review before a neutral decisionmaker. See See, 387 U. S., at 545; Lone Steer, 464 U. S., at 415 (noting that an administrative search may proceed with only a subpoena where the subpoenaed party is sufficiently protected by the opportunity to “question the reasonableness of the subpoena, before suffering any penalties for refusing to comply with it, by raising objections in an action in district court”). And, we see no reason why this minimal requirement is inapplicable here. While the Court has never attempted to prescribe the exact form an opportunity for precompliance review must take, the City does not even attempt to argue that §41.49(3)(a) affords hotel operators any opportunity whatsoever. Section 41.49(3)(a) is, therefore, facially invalid. A hotel owner who refuses to give an officer access to his or her registry can be arrested on the spot. The Court has held that business owners cannot reasonably be put to this kind of choice. Camara, 387 U. S., at 533 (holding that “broad statutory safeguards are no substitute for individualized review, particularly when those safeguards may only be invoked at the risk of a criminal penalty”). Absent an opportunity for precompliance review, the ordinance creates an intolerable risk that searches authorized by it will exceed statutory limits, or be used as a pretext to harass hotel operators and their guests. Even if a hotel has been searched 10 times a day, every day, for three months, without any violation being found, the operator can only refuse to comply with an officer’s demand to turn over the registry at his or her own peril. To be clear, we hold only that a hotel owner must be afforded an opportunity to have a neutral decisionmaker review an officer’s demand to search the registry before he or she faces penalties for failing to comply. Actual review need only occur in those rare instances where a hotel operator objects to turning over the registry. Moreover, this opportunity can be provided without imposing onerous burdens on those charged with an administrative scheme’s enforcement. For instance, respondents accept that the searches authorized by §41.49(3)(a) would be constitutional if they were performed pursuant to an administrative subpoena. Tr. of Oral Arg. 36–37. These subpoenas, which are typically a simple form, can be issued by the individual seeking the record—here, officers in the field—without probable cause that a regulation is being infringed. See See, 387 U. S., at 544 (“[T]he demand to inspect may be issued by the agency”). Issuing a subpoena will usually be the full extent of an officer’s burden because “the great majority of businessmen can be expected in normal course to consent to inspection without warrant.” Barlow’s, Inc., 436 U. S., at 316. Indeed, the City has cited no evidence suggesting that without an ordinance authorizing on-demand searches, hotel operators would regularly refuse to cooperate with the police. In those instances, however, where a subpoenaed hotel operator believes that an attempted search is motivated by illicit purposes, respondents suggest it would be sufficient if he or she could move to quash the subpoena before any search takes place. Tr. of Oral Arg. 38–39. A neutral decisionmaker, including an administrative law judge, would then review the subpoenaed party’s objections before deciding whether the subpoena is enforceable. Given the limited grounds on which a motion to quash can be granted, such challenges will likely be rare. And, in the even rarer event that an officer reasonably suspects that a hotel operator may tamper with the registry while the motion to quash is pending, he or she can guard the registry until the required hearing can occur, which ought not take long. Riley v. California, 573 U. S. ___ (2014) (slip op., at 12) (police may seize and hold a cell phone “to prevent destruction of evidence while seeking a warrant”); Illinois v. McArthur, 531 U. S. 326, 334 (2001) (citing cases upholding the constitutionality of “temporary restraints where [they are] needed to preserve evidence until police could obtain a warrant”). Cf. Missouri v. McNeely, 569 U. S. ___ (2013) (slip op., at 12) (noting that many States have procedures in place for considering warrant applications telephonically).[3] Procedures along these lines are ubiquitous. A 2002 report by the Department of Justice “identifiedapproximately 335 existing administrative subpoena authorities held by various [federal] executive branch entities.” Office of Legal Policy, Report to Congresson the Use of Administrative Subpoena Authorities by Executive Branch Agencies and Entities 3, onlineat http://www.justice.gov/archive/olp/rpt_to_congress.htm(All Internet materials as visited June 19, 2015, andavailable in Clerk of Court’s case file). Their prevalenceconfirms what common sense alone would otherwise lead us to conclude: In most contexts, business owners can be afforded at least an opportunity to contest an administrative search’s propriety without unduly compromising the government’s ability to achieve its regulatory aims. Of course administrative subpoenas are only one way in which an opportunity for precompliance review can be made available. But whatever the precise form, the availability of precompliance review alters the dynamic between the officer and the hotel to be searched, and reduces the risk that officers will use these administrative searches as a pretext to harass business owners. Finally, we underscore the narrow nature of our holding. Respondents have not challenged and nothing in our opinion calls into question those parts of §41.49 that require hotel operators to maintain guest registries containing certain information. And, even absent legislative action to create a procedure along the lines discussed above, see supra, at 11, police will not be prevented from obtaining access to these documents. As they often do, hotel operators remain free to consent to searches of their registries and police can compel them to turn them overif they have a proper administrative warrant—including one that was issued ex parte—or if some other exceptionto the warrant requirement applies, including exigent circumstances.[4] B Rather than arguing that §41.49(3)(a) is constitutional under the general administrative search doctrine, the City and Justice Scalia contend that hotels are “closely regulated,” and that the ordinance is facially valid under the more relaxed standard that applies to searches of this category of businesses. Brief for Petitioner 28–47; post, at 5. They are wrong on both counts. Over the past 45 years, the Court has identified only four industries that “have such a history of government oversight that no reasonable expectation of privacy . . . could exist for a proprietor over the stock of such an enterprise,” Barlow’s, Inc., 436 U. S., 313. Simply listing these industries refutes petitioner’s argument that hotels should be counted among them. Unlike liquor sales, Colonnade Catering Corp. v. United States, 397 U. S. 72 (1970) , firearms dealing, United States v. Biswell, 406 U. S. 311 –312 (1972), mining, Donovan v. Dewey, 452 U. S. 594 (1981) , or running an automobile junkyard, New York v. Burger, 482 U. S. 691 (1987) , nothing inherent in the operation of hotels poses a clear and significant risk to the public welfare. See, e.g., id., at 709 (“Automobile junkyards and vehicle dismantlers provide the major market for stolen vehicles and vehicle parts”); Dewey, 452 U. S., at 602 (describing the mining industry as “among the most hazardous in the country”).[5] Moreover, “[t]he clear import of our cases is that the closely regulated industry . . . is the exception.” Barlow’s, Inc., 436 U. S., at 313. To classify hotels as pervasively regulated would permit what has always been a narrow exception to swallow the rule. The City wisely refrains from arguing that §41.49 itself renders hotels closely regulated. Nor do any of the other regulations on which petitioner and Justice Scalia rely—regulations requiring hotels to, inter alia, maintain a license, collect taxes, conspicuously post their rates, and meet certain sanitary standards—establish a comprehensive scheme of regulation that distinguishes hotels from numerous other businesses. See Brief for Petitioner 33–34 (citing regulations); post, at 7 (same). All businesses in Los Angeles need a license to operate. LAMC §§21.03(a), 21.09(a). While some regulations apply to a smaller set of businesses, see e.g. Cal. Code Regs., tit. 25, §40 (2015) (requiring linensto be changed between rental guests), online at http://www.oal.ca.gov/ccr.htm, these can hardly be said to have created a “ ‘comprehensive’ ” scheme that puts hotel owners on notice that their “ ‘property will be subject to periodic inspections undertaken for specific purposes,’ ” Burger, 482 U. S., at 705, n. 16 (quoting Dewey, 452 U. S., at 600). Instead, they are more akin to the widely applicable minimum wage and maximum hour rules that the Court rejected as a basis for deeming “the entirety of American interstate commerce” to be closely regulated in Barlow’s, Inc. 436 U. S., at 314. If such general regulations were sufficient to invoke the closely regulated industry exception, it would be hard to imagine a type of business that would not qualify. See Brief for Google Inc. as Amicus Curiae 16–17; Brief for the Chamber of Commerce of United States of America as Amicus Curiae 12–13. Petitioner attempts to recast this hodgepodge of reg-ulations as a comprehensive scheme by referring to a “centuries-old tradition” of warrantless searches of hotels. Brief for Petitioner 34–36. History is relevant when deter-mining whether an industry is closely regulated. See,e.g., Burger, 482 U. S., at 707. The historical record here, however, is not as clear as petitioner suggests. The City and Justice Scalia principally point to evidence that hotels were treated as public accommodations. Brief for Petitioner 34–36; post, at 5–6, and n. 1. For instance, the Commonwealth of Massachusetts required innkeepers to “ ‘furnish[ ] . . . suitable provisions and lodging, for the refreshment and entertainment of strangers and travellers, pasturing and stable room, hay and provender . . . for their horses and cattle.’ ” Brief for Petitioner 35 (quoting An Act For The Due Regulation Of Licensed Houses (1786), reprinted in Acts and Laws of the Commonwealth of Massachusetts 209 (1893)). But laws obligating inns to provide suitable lodging to all paying guests are not the same as laws subjecting inns to warrantless searches. Petitioner also asserts that “[f]or a long time, [hotel] owners left their registers open to widespread inspection.” Brief for Petitioner 51. Setting aside that modern hotel registries contain sensitive information, such as driver’s licenses and credit card numbers for which there is no historic analog, the fact that some hotels chose to make registries accessible to the public has little bearing on whether government authorities could have viewed these documents on demand without a hotel’s consent. Even if we were to find that hotels are pervasively regulated, §41.49 would need to satisfy three additional criteria to be reasonable under the Fourth Amendment: (1) “[T]here must be a ‘substantial’ government interest that informs the regulatory scheme pursuant to which the inspection is made”; (2) “the warrantless inspections must be ‘necessary’ to further [the] regulatory scheme”; and (3) “the statute’s inspection program, in terms of the certainty and regularity of its application, [must] provid[e] a constitutionally adequate substitute for a warrant.” Burger, 482 U. S., at 702–703 (internal quotation marks omitted). We assume petitioner’s interest in ensuring that hotels maintain accurate and complete registries might fulfill the first of these requirements, but conclude that §41.49 fails the second and third prongs of this test. The City claims that affording hotel operators any opportunity for precompliance review would fatally undermine the scheme’s efficacy by giving operators a chance to falsify their records. Brief for Petitioner 41–42. The Court has previously rejected this exact argument, which could be made regarding any recordkeeping requirement. See Barlow’s, Inc., 436 U. S., at 320 (“[It is not] apparent why the advantages of surprise would be lost if, after being refused entry, procedures were available for the [Labor] Secretary to seek an ex parte warrant to reappear at the premises without further notice to the establishment being inspected”); cf. Lone Steer, 464 U. S., at 411, 415 (affirming use of administrative subpoena which provided an opportunity for precompliance review as a means for obtaining “payroll and sales records”). We see no reason to accept it here. As explained above, nothing in our decision today precludes an officer from conducting a surprise inspection by obtaining an ex parte warrant or, where an officer reasonably suspects the registry would be altered, from guarding the registry pending a hearing on a motion to quash. See Barlow’s, Inc., 436 U. S., at 319–321; Riley, 573 U. S., at ___ (slip op., at 12). Justice Scalia’s claim that these procedures will prove unworkable given the large number of hotels in Los Angeles is a red herring. See post, at 11. While there are approximately 2,000 hotels in Los Angeles, ibid., there is no basis to believe that resort to such measures will be needed to conduct spot checks in the vast majority of them. See supra, at 11. Section 41.49 is also constitutionally deficient under the “certainty and regularity” prong of the closely regulated industries test because it fails sufficiently to constrain police officers’ discretion as to which hotels to search and under what circumstances. While the Court has upheld inspection schemes of closely regulated industries that called for searches at least four times a year, Dewey, 452 U. S., at 604, or on a “regular basis,” Burger, 482 U. S., at 711, §41.49 imposes no comparable standard. * * * For the foregoing reasons, we agree with the Ninth Circuit that §41.49(3)(a) is facially invalid insofar as it fails to provide any opportunity for precompliance review before a hotel must give its guest registry to the police for inspection. Accordingly, the judgment of the Ninth Circuit is affirmed. It is so ordered.Notes 1 Relatedly, the United States claims that a statute authorizing warrantless searches may still have independent force if it imposes a penalty for failing to cooperate in a search conducted under a warrant or in an exigency. See Brief for United States as Amicus Curiae 19. This argument gets things backwards. An otherwise facially unconstitutional statute cannot be saved from invalidation based solely on the existence of a penalty provision that applies when searches are not actually authorized by the statute. This argument is especially unconvincing where, as here, an independent obstruction of justice statute imposes a penalty for “willfully, resist[ing], delay[ing], or obstruct[ing] any public officer . . . in the discharge or attempt to discharge any duty of his or her office of employment.” Cal. Penal Code Ann. §148(a)(1) (West 2014). 2 Respondents contend that §41.49’s principal purpose instead is to facilitate criminal investigation. Brief for Respondents 44–47. Because we find that the searches authorized by §41.49 are unconstitutional even if they serve the City’s asserted purpose, we decline to address this argument. 3 Justice Scalia professes to be baffled at the idea that we could suggest that in certain circumstances, police officers may seize something that they cannot immediately search. Post, at 10–11 (dissenting opinion). But that is what this Court’s cases have explicitly endorsed, including Riley just last Term. 4 In suggesting that our holding today will somehow impede law enforcement from achieving its important aims, Justice Scalia relies on instances where hotels were used as “prisons for migrants smuggled across the border and held for ransom” or as “rendezvous sites where child sex workers meet their clients on threat of violence from their procurers.” See post, at 2. It is hard to imagine circumstances more exigent than these. 5 Justice Scalia’s effort to depict hotels as raising a comparable degree of risk rings hollow. See post, at 1, 14. Hotels—like practically all commercial premises or services—can be put to use for nefarious ends. But unlike the industries that the Court has found to be closely regulated, hotels are not intrinsically dangerous.

575.US.2014_13-1333

Ordinarily, a federal litigant who is too poor to pay court fees may proceed in forma pauperis. This means that the litigant may commence a civil action without prepaying fees or paying certain expenses. See 28 U. S. C. §1915(a). But a special “three strikes” provision prevents a court from affording in forma pauperis status to a prisoner who “has, on 3 or more prior occasions, while incarcerated . . . , brought an action or appeal in a court of the United States that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted.” §1915(g). Petitioner Coleman, a state prisoner, filed three federal lawsuits that were dismissed on grounds enumerated in §1915(g). While the third dismissal was pending on appeal, he filed four additional federal lawsuits, moving to proceed in forma pauperis in each. The District Court refused to permit him to proceed in forma pauperis in any of those lawsuits, holding that a prior dismissal is a strike under §1915(g) even if it is pending on appeal. The Sixth Circuit agreed with the District Court. Held: A prior dismissal on one of §1915(g)’s statutorily enumerated grounds counts as a strike, even if the dismissal is the subject of an ongoing appeal. Pp. 4–9. (a) Coleman suggests that that a dismissal should count as a strike only once appellate review is complete. But the word “dismissed” does not normally include subsequent appellate activity. See, e.g., Heintz v. Jenkins, 514 U. S. 291 . And §1915 itself describes dismissal as an action taken by a single court, not as a sequence of events involving multiple courts. See §1915(e). Coleman further contends that the phrase “prior occasions” creates ambiguity. But nothing about that phrase transforms a dismissal into a dismissal-plus-appellate-review. In the context of §1915(g), a “prior occasion” merely means a previous instance in which a “prisoner has . . . brought an action or appeal . . . that was dismissed on” statutorily enumerated grounds. A literal reading of the “three strikes” provision is consistent with the statute’s treatment of the trial and appellate states of litigation as distinct. See §§1915(a)(2), (a)(3), (b)(1), (e)(2), (g). It is also supported by the way in which the law ordinarily treats trial court judgments, i.e., a judgment normally takes effect despite a pending appeal, see Fed. Rule Civ. Proc. 62; Fed. Rule App. Proc. 8(a), and its preclusive effect is generally immediate, notwithstanding any appeal, see Clay v. United States, 537 U. S. 522 . Finally, the statute’s purpose favors this Court’s interpretation. The “three strikes” provision was “designed to filter out the bad claims and facilitate consideration of the good,” Jones v. Bock, 549 U. S. 199 . To refuse to count a prior dismissal because of a pending appeal would produce a leaky filter, because a prisoner could file many new lawsuits before reaching the end of the often lengthy appellate process. By contrast, the Court perceives no great risk that an erroneous trial court dismissal might wrongly deprive a prisoner of in forma pauperis status in a subsequent lawsuit. Pp. 4–8. (b) Coleman also argues that if the dismissal of a third complaint counts as a third strike, a litigant will lose the ability to appeal in forma pauperis from that strike itself. He believes this is a result that Congress could not possibly have intended. Because Coleman is not appealing from a third-strike trial-court dismissal here, the Court declines to address that question. Pp. 8–9. 733 F. 3d 175, affirmed. Breyer, J., delivered the opinion for a unanimous Court. Notes 1 * Together with Coleman, aka Coleman-Bey v. Bowerman et al.; Coleman, aka Coleman-Bey v. Dykehouse et al., and Coleman, aka Coleman-Bey v. Vroman et al. (see this Court’s Rule 12.4), also on certiorari to the same court.

Ordinarily, a federal litigant who is too poor to pay court fees may proceed in forma pauperis. This means that the litigant may commence a civil action without prepaying fees or paying certain expenses. See 28 U. S. C. §1915. But a special “three strikes” provision prevents a court from affording in forma pauperis status where the litigant is a prisoner and he or she “has, on 3 or more prior occasions, while incarcerated . . . , brought an action or appeal in a court of the United States that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted.” §1915(g). Prior to this litigation, a Federal District Court had dismissed on those grounds three actions brought by a state prisoner. While the third dismissal was pending on appeal, the prisoner sought to bring several additional actions in the federal courts. The question before us is whether the prisoner may litigate his new actions in forma pauperis. Where an appeals court has not yet decided whether a prior dismissal is legally proper, should courts count, or should they ignore, that dismissal when calcu-lating how many qualifying dismissals the litigant has suffered? We conclude that the courts must count the dismissal even though it remains pending on appeal. The litigant here has accumulated three prior dismissals on statutorily enumerated grounds. Consequently, a court may not afford him in forma pauperis status with respect to his additional civil actions. I A Congress first enacted an in forma pauperis statute in 1892. See Act of July 20, ch. 209, 27Stat. 252. Congress recognized that “no citizen sh[ould] be denied an opportunity to commence, prosecute, or defend an action, civil or criminal, in any court of the United States, solely because his poverty makes it impossible for him to pay or secure the costs.” Adkins v. E. I. DuPont de Nemours & Co., 335 U. S. 331, 342 (1948) (internal quotation marks omitted). It therefore permitted a citizen to “commence and prosecute to conclusion any such . . . action without being required to prepay fees or costs, or give security therefor before or after bringing suit.” §1, 27Stat. 252. The current statute permits an individual to litigate a federal action in forma pauperis if the individual files an affidavit stating, among other things, that he or she is unable to prepay fees “or give security therefor.” 28 U. S. C. §1915(a)(1). Even in 1892, “Congress recognized . . . that a litigant whose filing fees and court costs are assumed by the public, unlike a paying litigant, lacks an economic incentive to refrain from filing frivolous, malicious, or repetitive lawsuits.” Neitzke v. Williams, 490 U. S. 319, 324 (1989) . And as the years passed, Congress came to see that prisoner suits in particular represented a disproportionate share of federal filings. Jones v. Bock, 549 U. S. 199 –203 (2007). It responded by “enact[ing] a variety of reforms designed to filter out the bad claims [filed by prisoners] and facilitate consideration of the good.” Id., at 204. Among those reforms was the “three strikes” rule here at issue. The rule, which applies to in forma pauperis status, reads in its entirety as follows: “In no event shall a prisoner bring a civil action or appeal a judgment in a civil action or proceeding [in forma pauperis] if the prisoner has, on 3 or more prior occasions, while incarcerated or detained in any facil-ity, brought an action or appeal in a court of the United States that was dismissed on the grounds that it is frivolous, malicious, or fails to state a claim upon which relief may be granted, unless the prisoner is under imminent danger of serious physical injury.” §1915(g). B The petitioner, André Lee Coleman, is incarcerated at the Baraga Correctional Facility in Michigan. By 2010, three federal lawsuits filed by Coleman during his incarceration had been dismissed as frivolous (or on other grounds enumerated in §1915(g)). Nonetheless, when Coleman filed four new federal lawsuits between April 2010 and January 2011, he moved to proceed in forma pauperis in each. He denied that his third dismissed lawsuit counted as a strike under §1915(g). That is because he had appealed the dismissal, and the appeals court had not yet ruled. Thus, in Coleman’s view, he had fewer than three qualifying dismissals, and was eligible for in forma pauperis status under the statute. The District Court rejected Coleman’s argument. It held that “a dismissal counts as a strike even if it is pending on appeal at the time that the plaintiff files his new action.” No. 10–cv–337 (WD Mich., Apr. 12, 2011), App. to Pet. for Cert. 21a, 24a. It thus refused to permit Coleman to proceed in forma pauperis in any of his four suits. On appeal, a divided panel of the Sixth Circuit agreed with the District Court. 733 F. 3d 175 (2013). It resolved the four cases using slightly different procedures. In one of the four cases, the Sixth Circuit affirmed the District Court’s judgment. In the remaining three cases, it denied Coleman’s request to proceed in forma pauperis on appeal. It subsequently dismissed the three cases for want of prosecution after Coleman failed to pay the appellate filing fees. In contrast to the Sixth Circuit, the vast majority of the other Courts of Appeals have held that a prior dismissal on a statutorily enumerated ground does not count as a strike while an appeal of that dismissal remains pending. See Henslee v. Keller, 681 F. 3d 538, 541 (CA4 2012) (listing, and joining, courts that have adopted the majority view). In light of the division of opinion among the Circuits, we granted Coleman’s petition for certiorari. II A In our view, the Sixth Circuit majority correctly applied §1915(g). A prior dismissal on a statutorily enumerated ground counts as a strike even if the dismissal is the subject of an appeal. That, after all, is what the statute literally says. The “three strikes” provision applies where a prisoner “has, on 3 or more prior occasions . . . brought an action or appeal . . . that was dismissed on” certain grounds. §1915(g) (emphasis added). Coleman believes that we should read the statute as if it referred to an “affirmed dismissal,” as if it considered a trial court dismissal to be provisional, or as if it meant that a dismissal falls within the statute’s scope only when the litigant has no further chance to secure a reversal. But the statute itself says none of these things. Instead, the statute refers to whether an action or appeal “was dismissed.” §1915(g). The linguistic term “dismiss,” taken alone, does not normally include subsequent appellate activity. See, e.g., Heintz v. Jenkins, 514 U. S. 291, 294 (1995) (“[T]he District Court dismissed [the] lawsuit for failure to state a claim. . . . However, the Court of Appeals for the Seventh Circuit reversed the District Court’s judgment”); Gray v. Netherland, 518 U. S. 152, 158 (1996) (“The Suffolk Circuit Court dismissed petitioner’s state petition for a writ of habeas corpus. The Virginia Supreme Court affirmed the dismissal”). Indeed, §1915 itself describes dismissal as an action taken by a single court, not as a sequence of events involving multiple courts. See §1915(e)(2) (“[T]he court shall dismiss the case at any time if the court determines that—(A) the allegation of poverty is untrue; or (B) the action or appeal—(i) is frivolous or malicious; [or] (ii) fails to state a claim on which relief may be granted” (emphasis added)). Coleman insists that §1915(g) is not so clear. Even if the term “dismissed” is unambiguous, contends Coleman, the phrase “prior occasions” creates ambiguity. Coleman observes that the phrase “ ‘may refer to a single moment or to a continuing event: to an appeal, independent of the underlying action, or to the continuing claim, inclusive of both the action and its appeal.’ ” Brief for Petitioner 17 (quoting Henslee, supra, at 542). Coleman believes that a “prior occasion” in the context of §1915(g) may therefore include both a dismissal on an enumerated ground and any subsequent appeal. We find it difficult to agree. Linguistically speaking, we see nothing about the phrase “prior occasions” that would transform a dismissal into a dismissal-plus-appellate-review. An “occasion” is “a particular occurrence,” a “happening,” or an “incident.” Webster’s Third New International Dictionary 1560 (3d ed. 1993). And the statute provides the content of that occurrence, happening, or incident: It is an instance in which a “prisoner has . . . brought an action or appeal in a court of the United States that was dismissed on” statutorily enumerated grounds. §1915(g). Under the plain language of the statute, when Coleman filed the suits at issue here, he had already experienced three such “prior occasions.” Our literal reading of the phrases “prior occasions” and “was dismissed” is consistent with the statute’s discussion of actions and appeals. The in forma pauperis statute repeatedly treats the trial and appellate stages of litigation as distinct. See §§1915(a)(2), (a)(3), (b)(1), (e)(2), (g). Related provisions reflect a congressional focus upon trial court dismissal as an important separate element of the statutory scheme. See §1915A (requiring a district court to screen certain prisoner complaints “as soon as practicable” and to dismiss any portion of the complaint that “is frivolous, malicious, or fails to state a claim upon which relief may be granted”); 42 U. S. C. §1997e(c)(1) (similar). We have found nothing in these provisions indicating that Congress considered a trial court dismissal and an appellate court decision as if they were a single entity—or that Congress intended the former to take effect only when affirmed by the latter. Our literal reading of the “three strikes” provision also is supported by the way in which the law ordinarily treats trial court judgments. Unless a court issues a stay, a trial court’s judgment (say, dismissing a case) normally takes effect despite a pending appeal. See Fed. Rule Civ. Proc. 62; Fed. Rule App. Proc. 8(a). And a judgment’s preclusive effect is generally immediate, notwithstanding any appeal. See Clay v. United States, 537 U. S. 522, 527 (2003) (“Typically, a federal judgment becomes final for . . . claim preclusion purposes when the district court disassociates itself from the case, leaving nothing to be done at the court of first instance save execution of the judgment”). The ordinary rules of civil procedure thus provide additional support for our interpretation of the statute. See Jones, 549 U. S., at 211–216 (applying the ordinary rules of civil procedure where the procedural requirements for prison litigation do not call for an alternative). Finally, the statute’s purpose favors our interpretation. The “three strikes” provision was “designed to filter out the bad claims and facilitate consideration of the good.” Id., at 204. To refuse to count a prior dismissal because of a pending appeal would produce a leaky filter. Appeals take time. During that time, a prisoner could file many lawsuits, including additional lawsuits that are frivolous, malicious, or fail to state a claim upon which relief may be granted. Indeed, Coleman filed these four cases after he suffered his third qualifying dismissal, in October 2009, and before the affirmance of that order, in March 2011. We recognize that our interpretation of the statute may create a different risk: An erroneous trial court dismissal might wrongly deprive a prisoner of in forma pauperis status with respect to lawsuits filed after a dismissal but before its reversal on appeal. But that risk does not seem great. For one thing, the Solicitor General informs us that he has been able to identify only two instances in which a Court of Appeals has reversed a District Court’s issuance of a third strike. Brief for United States as Amicus Curiae 22, n. 5. For another, where a court of appeals reverses a third strike, in some instances the prisoner will be able to refile his or her lawsuit after the reversal, seeking in forma pauperis status at that time. Further, if the statute of limitations governing that lawsuit has run out before the court of appeals reverses the third strike, the Solicitor General assures us that prisoners will find relief in Federal Rule of Civil Procedure 60(b). According to the Solicitor General, a prisoner may move to reopen his or her interim lawsuits (reinstating the cases as of the dates originally filed) and may then seek in forma pauperis status anew. In any event, we believe our interpretation of the statute hews more closely to its meaning and objective than does Coleman’s alternative. B Coleman makes an additional argument. He poses a hypothetical: What if this case had involved an attempt to appeal from the trial court’s dismissal of his third complaint instead of an attempt to file several additional complaints? If the dismissal were counted as his third strike, Coleman asserts, he would lose the ability to appeal in forma pauperis from that strike itself. He believes that this result, which potentially could deprive him of appellate review, would be unfair. He further believes that it would be such a departure from the federal courts’ normal appellate practice that Congress could not possibly have intended it. The Solicitor General, while subscribing to our interpretation of the statute, supports Coleman on this point. The Solicitor General says that we can and should read the statute to afford a prisoner in forma pauperis status with respect to an appeal from a third qualifying dismissal—even if it does not allow a prisoner to file a fourth case during that time. He believes that the statute, in referring to dismissals “on 3 or more prior occasions,” 28 U. S. C. §1915(g) (emphasis added), means that a trial court dismissal qualifies as a strike only if it occurred in a prior, different, lawsuit. We need not, and do not, now decide whether the Solicitor General’s interpretation (or some other interpretation with the same result) is correct. That is because Coleman is not here appealing from a third-strike trial-court dismissal. He is appealing from the denial of in forma pauperis status with respect to several separate suits filed after the trial court dismissed his earlier third-strike suit. With respect to those suits, the earlier dismissals certainly took place on “prior occasions.” If and when the situation that Coleman hypothesizes does arise, the courts can consider the problem in context. * * * For the reasons stated, we hold that a prisoner who has accumulated three prior qualifying dismissals under §1915(g) may not file an additional suit in forma pauperis while his appeal of one such dismissal is pending. The judgments of the Court of Appeals are Affirmed.

575.US.2014_13-485

Maryland’s personal income tax on state residents consists of a “state” income tax, Md. Tax-Gen. Code Ann. §10–105(a), and a “county” income tax, §§10–103, 10–106. Residents who pay income tax to another jurisdiction for income earned in that other jurisdiction are allowed a credit against the “state” tax but not the “county” tax. §10–703. Nonresidents who earn income from sources within Maryland must pay the “state” income tax, §§10–105(d), 10–210, and nonresidents not subject to the county tax must pay a “special nonresident tax” in lieu of the “county” tax, §10–106.1. Respondents, Maryland residents, earned pass-through income from a Subchapter S corporation that earned income in several States. Respondents claimed an income tax credit on their 2006 Maryland income tax return for taxes paid to other States. The Mary-land State Comptroller of the Treasury, petitioner here, allowed respondents a credit against their “state” income tax but not against their “county” income tax and assessed a tax deficiency. That decision was affirmed by the Hearings and Appeals Section of the Comptroller’s Office and by the Maryland Tax Court, but the Circuit Court for Howard County reversed on the ground that Maryland’s tax system violated the Commerce Clause of the Federal Constitution. The Court of Appeals of Maryland affirmed and held that the tax unconstitutionally discriminated against interstate commerce. Held: Maryland’s personal income tax scheme violates the dormant Commerce Clause. Pp. 4–28. (a) The Commerce Clause, which grants Congress power to “regulate Commerce . . . among the several States,” Art I, §8, cl. 3, also has “a further, negative command, known as the dormant Commerce Clause,” Oklahoma Tax Comm’n v. Jefferson Lines, Inc., 514 U. S. 175, 179, which precludes States from “discriminat[ing] between transactions on the basis of some interstate element,” Boston Stock Exchange v. State Tax Comm’n, 429 U. S. 318 , n. 12. Thus, inter alia, a State “may not tax a transaction or incident more heavily when it crosses state lines than when it occurs entirely within the State,” Armco Inc. v. Hardesty, 467 U. S. 638 , or “impose a tax which discriminates against interstate commerce either by providing a direct commercial advantage to local business, or by subjecting interstate commerce to the burden of ‘multiple taxation,’ ” Northwestern States Portland Cement Co. v. Minnesota, 358 U. S. 450 . Pp. 4–6. (b) The result in this case is all but dictated by this Court’s dormant Commerce Clause cases, particularly J. D. Adams Mfg. Co. v. Storen, 304 U. S. 307 , Gwin, White & Prince, Inc. v. Henneford, 305 U. S. 434 , and Central Greyhound Lines, Inc. v. Mealey, 334 U. S. 653 , which all invalidated state tax schemes that might lead to double taxation of out-of-state income and that discriminated in favor of intrastate over interstate economic activity. Pp. 6–7. (c) This conclusion is not affected by the fact that these three cases involved a tax on gross receipts rather than net income, and a tax on corporations rather than individuals. This Court’s decisions have previously rejected the formal distinction between gross receipts and net income taxes. And there is no reason the dormant Commerce Clause should treat individuals less favorably than corporations; in addition, the taxes invalidated in J. D. Adams and Gwin, White applied to the income of both individuals and corporations. Nor does the right of the individual to vote in political elections justify disparate treatment of corporate and personal income. Thus the Court has previously entertained and even sustained dormant Commerce Clause challenges by individual residents of the State that imposed the alleged burden on interstate commerce. See Department of Revenue of Ky. v. Davis, 553 U. S. 328 ; Granholm v. Heald, 544 U. S. 460, 469 (2005) . Pp. 7–12. (d) Maryland’s tax scheme is not immune from dormant Commerce Clause scrutiny simply because Maryland has the jurisdictional power under the Due Process Clause to impose the tax. “[W]hile a state may, consistent with the Due Process Clause, have the authority to tax a particular taxpayer, imposition of the tax may nonetheless violate the Commerce Clause.” Quill Corp. v. North Dakota, 504 U. S. 298 . Pp. 12–15. (e) Maryland’s income tax scheme discriminates against interstate commerce. The “internal consistency” test, which helps courts identify tax schemes that discriminate against interstate commerce, as-sumes that every State has the same tax structure. Maryland’s income tax scheme fails the internal consistency test because if every State adopted Maryland’s tax structure, interstate commerce would be taxed at a higher rate than intrastate commerce. Maryland’s tax scheme is inherently discriminatory and operates as a tariff, which is fatal because tariffs are “[t]he paradigmatic example of a law discriminating against interstate commerce.” West Lynn Creamery, Inc. v. Healy, 512 U. S. 186 . Petitioner emphasizes that by offering residents who earn income in interstate commerce a credit against the “state” portion of the income tax, Maryland actually receives less tax revenue from residents who earn income from interstate commerce rather than intrastate commerce, but this argument is a red herring. The critical point is that the total tax burden on interstate commerce is higher. Pp. 18–26. 431 Md. 147, 64 A. 3d 453, affirmed. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, and Sotomayor, JJ., joined. Scalia, J., filed a dissenting opinion, in which Thomas, J., joined as to Parts I and II. Thomas, J., filed a dissenting opinion, in which Scalia, J., joined except as to the first paragraph. Ginsburg, J., filed a dissenting opinion, in which Scalia and Kagan, JJ., joined.

This case involves the constitutionality of an unusual feature of Maryland’s personal income tax scheme. Like many other States, Maryland taxes the income its residents earn both within and outside the State, as well as the income that nonresidents earn from sources within Maryland. But unlike most other States, Maryland does not offer its residents a full credit against the income taxes that they pay to other States. The effect of this scheme is that some of the income earned by Maryland residents outside the State is taxed twice. Maryland’s scheme creates an incentive for taxpayers to opt for intrastate rather than interstate economic activity. We have long held that States cannot subject corporate income to tax schemes similar to Maryland’s, and we see no reason why income earned by individuals should be treated less favorably. Maryland admits that its law has the same economic effect as a state tariff, the quintessential evil targeted by the dormant Commerce Clause. We therefore affirm the decision of Maryland’s highest court and hold that this feature of the State’s tax scheme vio-lates the Federal Constitution. I Maryland, like most States, raises revenue in part by levying a personal income tax. The income tax that Maryland imposes upon its own residents has two parts: a “state” income tax, which is set at a graduated rate, Md. Tax-Gen. Code Ann. §10–105(a) (Supp. 2014), and a so-called “county” income tax, which is set at a rate that varies by county but is capped at 3.2%, §§10–103, 10–106 (2010). Despite the names that Maryland has assigned to these taxes, both are State taxes, and both are collected by the State’s Comptroller of the Treasury. Frey v. Comptroller of Treasury, 422 Md. 111, 125, 141–142, 29 A. 3d 475, 483, 492 (2011). Of course, some Maryland residents earn income in other States, and some of those States also tax this income. If Maryland residents pay income tax to another jurisdiction for income earned there, Maryland allows them a credit against the “state” tax but not the “county” tax. §10–703; 431 Md. 147, 156–157, 64 A. 3d 453, 458 (2013) (case below). As a result, part of the income that a Maryland resident earns outside the State may be taxed twice. Maryland also taxes the income of nonresidents. This tax has two parts. First, nonresidents must pay the “state” income tax on all the income that they earn from sources within Maryland. §§10–105(d) (Supp. 2014), 10–210 (2010). Second, nonresidents not subject to the county tax must pay a “special nonresident tax” in lieu of the “county” tax. §10–106.1; Frey, supra, at 125–126, 29 A. 3d, at 483. The “special nonresident tax” is levied on income earned from sources within Maryland, and its rate is “equal to the lowest county income tax rate set by any Maryland county.” §10–106.1. Maryland does not tax the income that nonresidents earn from sources outside Maryland. See §10–210. Respondents Brian and Karen Wynne are Maryland residents. In 2006, the relevant tax year, Brian Wynne owned stock in Maxim Healthcare Services, Inc., a Subchapter S corporation.[1] That year, Maxim earned income in States other than Maryland, and it filed state income tax returns in 39 States. The Wynnes earned income passed through to them from Maxim. On their 2006 Mary-land tax return, the Wynnes claimed an income tax credit for income taxes paid to other States. Petitioner, the Maryland State Comptroller of the Treasury, denied this claim and assessed a tax deficiency. In accordance with Maryland law, the Comptroller allowed the Wynnes a credit against their Maryland “state” income tax but not against their “county” income tax. The Hearings and Appeals Section of the Comptroller’s Office slightly modified the assessment but otherwise affirmed. The Maryland Tax Court also affirmed, but the Circuit Court for Howard County reversed on the ground that Maryland’s tax system violated the Commerce Clause. The Court of Appeals of Maryland affirmed. 431 Md. 147, 64 A. 3d 453. That court evaluated the tax under the four-part test of Complete Auto Transit, Inc. v. Brady, 430 U. S. 274 (1977), which asks whether a “tax is applied to an activity with a substantial nexus with the taxing State, is fairly apportioned, does not discriminate against interstate commerce, and is fairly related to the services provided by the State.” Id., at 279. The Court of Appeals held that the tax failed both the fair apportionment and nondiscrimination parts of the Complete Auto test. With respect to fair apportionment, the court first held that the tax failed the “internal consistency” test because if every State adopted Maryland’s tax scheme, interstate commerce would be taxed at a higher rate than intrastate commerce. It then held that the tax failed the “external consistency” test because it created a risk of multiple taxation. With respect to nondiscrimination, the court held that the tax discriminated against interstate commerce because it denied residents a credit on income taxes paid to other States and so taxed income earned interstate at a rate higher than income earned intrastate. The court thus concluded that Maryland’s tax scheme was unconstitutional insofar as it denied the Wynnes a credit against the “county” tax for income taxes they paid to other States. Two judges dissented and argued that the tax did not violate the Commerce Clause. The Court of Appeals later issued a brief clarification that “[a] state may avoid discrimination against interstate commerce by providing a tax credit, or some other method of apportionment, to avoid discriminating against interstate commerce in violation of the dormant Commerce Clause.” 431 Md., at 189, 64 A. 3d at 478. We granted certiorari. 572 U. S. ___ (2014). II A The Commerce Clause grants Congress power to “regulate Commerce . . . among the several States.” Art. I, § 8, cl. 3. These “few simple words . . . reflected a central concern of the Framers that was an immediate reason for calling the Constitutional Convention: the conviction that in order to succeed, the new Union would have to avoid the tendencies toward economic Balkanization that had plagued relations among the Colonies and later among the States under the Articles of Confederation.” Hughes v. Oklahoma, 441 U. S. 322 –326 (1979). Although the Clause is framed as a positive grant of power to Congress, “we have consistently held this language to contain a further, negative command, known as the dormant Commerce Clause, prohibiting certain state taxation even when Congress has failed to legislate on the subject.” Oklahoma Tax Comm’n v. Jefferson Lines, Inc., 514 U. S. 175, 179 (1995) . This interpretation of the Commerce Clause has been disputed. See Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U. S. 564 –620 (1997) (Thomas, J., dissenting); Tyler Pipe Industries, Inc. v. Washington State Dept. of Revenue, 483 U. S. 232 –265 (1987) (Scalia, J., concurring in part and dissenting in part); License Cases, 5 How. 504, 578–579 (1847) (Taney, C. J.). But it also has deep roots. See, e.g., Case of the State Freight Tax, 15 Wall. 232, 279–280 (1873); Cooley v. Board of Wardens of Port of Philadelphia ex rel. Soc. for Relief of Distressed Pilots, 12 How. 299, 318–319 (1852); Gibbons v. Ogden, 9 Wheat. 1, 209 (1824) (Marshall, C. J.). By prohibiting States from discriminating against or imposing excessive burdens on interstate commerce without congressional approval, it strikes at one of the chief evils that led to the adoption of the Constitution, namely, state tariffs and other laws that burdened interstate commerce. Fulton Corp. v. Faulkner, 516 U. S. 325 –331 (1996); Hughes, supra, at 325; Welton v. Missouri, 91 U. S. 275, 280 (1876) ; see also The Federalist Nos. 7, 11 (A. Hamilton), and 42 (J. Madison). Under our precedents, the dormant Commerce Clause precludes States from “discriminat[ing] between transactions on the basis of some interstate element.” Boston Stock Exchange v. State Tax Comm’n, 429 U. S. 318 , n. 12 (1977). This means, among other things, that a State “may not tax a transaction or incident more heavily when it crosses state lines than when it occurs entirely within the State.” Armco Inc. v. Hardesty, 467 U. S. 638, 642 (1984) . “Nor may a State impose a tax which discriminates against interstate commerce either by providing a direct commercial advantage to local business, or by subjecting interstate commerce to the burden of ‘multiple taxation.’ ” Northwestern States Portland Cement Co. v. Minnesota, 358 U. S. 450, 458 (1959) (citations omitted). B Our existing dormant Commerce Clause cases all but dictate the result reached in this case by Maryland’s highest court. Three cases involving the taxation of the income of domestic corporations are particularly instructive. In J. D. Adams Mfg. Co. v. Storen, 304 U. S. 307 (1938) , Indiana taxed the income of every Indiana resident (including individuals) and the income that every nonresident derived from sources within Indiana. Id., at 308. The State levied the tax on income earned by the plaintiff Indiana corporation on sales made out of the State. Id., at 309. Holding that this scheme violated the dormant Commerce Clause, we explained that the “vice of the statute” was that it taxed, “without apportionment, receipts derived from activities in interstate commerce.” Id., at 311. If these receipts were also taxed by the States in which the sales occurred, we warned, interstate commerce would be subjected “to the risk of a double tax burden to which intrastate commerce is not exposed, and which the commerce clause forbids.” Ibid. The next year, in Gwin, White & Prince, Inc. v. Henneford, 305 U. S. 434 (1939) , we reached a similar result. In that case, the State of Washington taxed all the income of persons doing business in the State. Id., at 435. Washington levied that tax on income that the plaintiff Washington corporation earned in shipping fruit from Washington to other States and foreign countries. Id., at 436–437. This tax, we wrote, “discriminates against interstate commerce, since it imposes upon it, merely because interstate commerce is being done, the risk of a multiple burden to which local commerce is not exposed.” Id., at 439. In the third of these cases involving the taxation of a domestic corporation, Central Greyhound Lines, Inc. v. Mealey, 334 U. S. 653 (1948) , New York sought to tax the portion of a domiciliary bus company’s gross receipts that were derived from services provided in neighboring States. Id., at 660; see also id., at 665 (Murphy, J., dissenting) (stating that the plaintiff was a New York corporation). Noting that these other States might also attempt to tax this portion of the company’s gross receipts, the Court held that the New York scheme violated the dormant Commerce Clause because it imposed an “unfair burden” on interstate commerce. Id., at 662 (majority opinion). In all three of these cases, the Court struck down a state tax scheme that might have resulted in the double taxation of income earned out of the State and that discriminated in favor of intrastate over interstate economic activity. As we will explain, see Part II–F, infra, Maryland’s tax scheme is unconstitutional for similar reasons. C The principal dissent distinguishes these cases on the sole ground that they involved a tax on gross receipts rather than net income. We see no reason why the distinction between gross receipts and net income should matter, particularly in light of the admonition that we must consider “not the formal language of the tax statute but rather its practical effect.” Complete Auto, 430 U. S., at 279. The principal dissent claims, post, at 13 (opinion of Ginsburg, J.), that “[t]he Court, historically, has taken the position that the difference between taxes on net income and taxes on gross receipts from interstate commerce warrants different results.” 2 C. Trost & P. Hartman, Federal Limitations on State and Local Taxation 2d §10:1, p. 251 (2003) (emphasis added) (hereinafter Trost). But this historical point is irrelevant. As the principal dissent seems to acknowledge, our cases rejected this formal distinction some time ago. And the distinction between gross receipts and net income taxes was not the basis for our decisions in J. D. Adams, Gwin, White, and Central Greyhound, which turned instead on the threat of multiple taxation. The discarded distinction between taxes on gross receipts and net income was based on the notion, endorsed in some early cases, that a tax on gross receipts is an impermissible “direct and immediate burden” on interstate commerce, whereas a tax on net income is merely an “indirect and incidental” burden. United States Glue Co. v. Town of Oak Creek, 247 U. S. 321 –329 (1918); see also Shaffer v. Carter, 252 U. S. 37, 57 (1920) . This arid distinction between direct and indirect burdens allowed “very little coherent, trustworthy guidance as to tax valid-ity.” 2 Trost §9:1, at 212. And so, beginning with Justice Stone’s seminal opinion in Western Live Stock v. Bureau of Revenue, 303 U. S. 250 (1938) , and continuing through cases like J. D. Adams and Gwin, White, the direct-indirect burdens test was replaced with a more practical approach that looked to the economic impact of the tax. These cases worked “a substantial judicial reinterpretation of the power of the States to levy taxes on gross income from interstate commerce.” 1 Trost §2:20, at 175. After a temporary reversion to our earlier formalism, see Spector Motor Service, Inc. v. O’Connor, 340 U. S. 602 (1951), “the gross receipts judicial pendulum has swung in a wide arc, recently reaching the place where taxation of gross receipts from interstate commerce is placed on an equal footing with receipts from local business, in Complete Auto Transit Inc. v. Brady,” 2 Trost §9:1, at 212. And we have now squarely rejected the argument that the Commerce Clause distinguishes between taxes on net and gross income. See Jefferson Lines, 514 U. S., at 190 (explaining that the Court in Central Greyhound “understood the gross receipts tax to be simply a variety of tax on income”); Moorman Mfg. Co. v. Bair, 437 U. S. 267, 280 (1978) (rejecting a suggestion that the Commerce Clause distinguishes between gross receipts taxes and net income taxes); id., at 281 (Brennan, J., dissenting) (“I agree with the Court that, for purposes of constitutional review, there is no distinction between a corporate income tax and a gross-receipts tax”); Complete Auto, supra, at 280 (upholding a gross receipts tax and rejecting the notion that the Commerce Clause places “a blanket prohibition against any state taxation imposed directly on an interstate transaction”).[2] For its part, petitioner distinguishes J. D. Adams, Gwin, White, and Central Greyhound on the ground that they concerned the taxation of corporations, not individuals. But it is hard to see why the dormant Commerce Clause should treat individuals less favorably than corporations. See Camps Newfound, 520 U. S., at 574 (“A tax on real estate, like any other tax, may impermissibly burden interstate commerce” (emphasis added)). In addition, the distinction between individuals and corporations cannot stand because the taxes invalidated in J. D. Adams and Gwin, White applied to the income of both individuals and corporations. See Ind. Stat. Ann., ch. 26, §64–2602 (Burns 1933) (tax in J. D. Adams); 1935 Wash. Sess. Laws ch. 180, Tit. II, §4(e), pp. 710–711 (tax in Gwin, White). Attempting to explain why the dormant Commerce Clause should provide less protection for natural persons than for corporations, petitioner and the Solicitor General argue that States should have a free hand to tax their residents’ out-of-state income because States provide their residents with many services. As the Solicitor General puts it, individuals “reap the benefits of local roads, local police and fire protection, local public schools, [and] local health and welfare benefits.” Brief for United States as Amicus Curiae 30. This argument fails because corporations also benefit heavily from state and local services. Trucks hauling a corporation’s supplies and goods, and vehicles transporting its employees, use local roads. Corporations call upon local police and fire departments to protect their facilities. Corporations rely on local schools to educate prospective employees, and the availability of good schools and other government services are features that may aid a corporation in attracting and retaining employees. Thus, disparate treatment of corporate and personal income cannot be justified based on the state services enjoyed by these two groups of taxpayers. The sole remaining attribute that, in the view of petitioner, distinguishes a corporation from an individual for present purposes is the right of the individual to vote. The principal dissent also emphasizes that residents can vote to change Maryland’s discriminatory tax law. Post, at 3–4. The argument is that this Court need not be concerned about state laws that burden the interstate activities of individuals because those individuals can lobby and vote against legislators who support such measures. But if a State’s tax unconstitutionally discriminates against interstate commerce, it is invalid regardless of whether the plaintiff is a resident voter or nonresident of the State. This Court has thus entertained and even sustained dormant Commerce Clause challenges by individual residents of the State that imposed the alleged burden on interstate commerce, Department of Revenue of Ky. v. Davis, 553 U. S. 328, 336 (2008) ; Granholm v. Heald, 544 U. S. 460, 469 (2005) , and we have also sustained such a challenge to a tax whose burden was borne by in-state consumers, Bacchus Imports, Ltd. v. Dias, 468 U. S. 263, 272 (1984) .[3] The principal dissent and Justice Scalia respond to these holdings by relying on dictum in Goldberg v. Sweet, 488 U. S. 252, 266 (1989) , that it is not the purpose of the dormant Commerce Clause “ ‘to protect state residents from their own state taxes.’ ” Post, at 3 (Ginsburg, J., dissenting); post, at 5 (Scalia, J., dissenting). But we repudiated that dictum in West Lynn Creamery, Inc. v. Healy, 512 U. S. 186 (1994) , where we stated that “[s]tate taxes are ordinarily paid by in-state businesses and consumers, yet if they discriminate against out-of-state products, they are unconstitutional.” Id., at 203. And, of course, the dictum must bow to the holdings of our many cases entertaining Commerce Clause challenges brought by residents. We find the dissents’ reliance on Goldberg’s dictum particularly inappropriate since they do not find themselves similarly bound by the rule of that case, which applied the internal consistency test to determine whether the tax at issue violated the dormant Commerce Clause. 488 U. S., at 261. In addition, the notion that the victims of such discrimination have a complete remedy at the polls is fanciful. It is likely that only a distinct minority of a State’s residents earns income out of State. Schemes that discriminate against income earned in other States may be attractiveto legislators and a majority of their constituents for precisely this reason. It is even more farfetched to suggest that natural persons with out-of-state income are better able to influence state lawmakers than large corporations headquartered in the State. In short, petitioner’s argument would leave no security where the majority of voters prefer protectionism at the expense of the few who earn income interstate. It would be particularly incongruous in the present case to disregard our prior decisions regarding the taxation of corporate income because the income at issue here is a type of corporate income, namely, the income of a Subchapter S corporation. Only small businesses may incorporate under Subchapter S, and thus acceptance of petitioner’s submission would provide greater protection for income earned by large Subchapter C corporations than small businesses incorporated under Subchapter S. D In attempting to justify Maryland’s unusual tax scheme, the principal dissent argues that the Commerce Clause imposes no limit on Maryland’s ability to tax the income of its residents, no matter where that income is earned. It argues that Maryland has the sovereign power to tax all of the income of its residents, wherever earned, and it there-fore reasons that the dormant Commerce Clause cannot constrain Maryland’s ability to expose its residents (and nonresidents) to the threat of double taxation. This argument confuses what a State may do without violating the Due Process Clause of the Fourteenth Amendment with what it may do without violating the Commerce Clause. The Due Process Clause allows a State to tax “all the income of its residents, even income earned outside the taxing jurisdiction.” Oklahoma Tax Comm’n v. Chickasaw Nation, 515 U. S. 450 –463 (1995). But “while a State may, consistent with the Due Process Clause, have the authority to tax a particular taxpayer, imposition of the tax may nonetheless violate the Commerce Clause.” Quill Corp. v. North Dakota, 504 U. S. 298, 305 (1992) (rejecting a due process challenge to a tax before sustaining a Commerce Clause challenge to that tax). Our decision in Camps Newfound illustrates the point. There, we held that the Commerce Clause prohibited Maine from granting more favorable tax treatment to charities that operated principally for the benefit of Maine residents. 520 U. S., at 580–583. Because the plaintiff charity in that case was a Maine nonprofit corporation, there is no question that Maine had the raw jurisdictional power to tax the charity. See Chickasaw Nation, supra, at 462–463. Nonetheless, the tax failed scrutiny under the Commerce Clause. Camps Newfound, supra, at 580–581. Similarly, Maryland’s raw power to tax its residents’ out-of-state income does not insulate its tax scheme from scrutiny under the dormant Commerce Clause. Although the principal dissent claims the mantle of precedent, it is unable to identify a single case that endorses its essential premise, namely, that the Commerce Clause places no constraint on a State’s power to tax the income of its residents wherever earned. This is unsurprising. As cases like Quill Corp. and Camps Newfound recognize, the fact that a State has the jurisdictional power to impose a tax says nothing about whether that tax violates the Commerce Clause. See also, e.g., Barclays Bank PLC v. Franchise Tax Bd. of Cal., 512 U. S. 298 (1994) (separately addressing due process and Commerce Clause challenges to a tax); Moorman, 437 U. S. 267 (same); Standard Pressed Steel Co. v. Department of Revenue of Wash., 419 U. S. 560 (1975) (same); Lawrence v. State Tax Comm’n of Miss., 286 U. S. 276 (1932) (separately addressing due process and equal protection challengesto a tax); Travis v. Yale & Towne Mfg. Co., 252 U. S. 60 (1920) (separately addressing due process and privileges-and-immunities challenges to a tax). One good reason why we have never accepted the principal dissent’s logic is that it would lead to plainly untenable results. Imagine that Maryland taxed the income that its residents earned in other States but exempted income earned out of State from any business that primarily served Maryland residents. Such a tax would violate the dormant Commerce Clause, see Camps Newfound, supra, and it cannot be saved by the principal dissent’s admonition that Maryland has the power to tax all the income of its residents. There is no principled difference between that hypothetical Commerce Clause challenge and this one. The principal dissent, if accepted, would work a sea change in our Commerce Clause jurisprudence. Legion are the cases in which we have considered and even upheld dormant Commerce Clause challenges brought by residents to taxes that the State had the jurisdictional power to impose. See, e.g., Davis, 553 U. S. 328 ; Camps Newfound, 520 U. S. 564 ; Fulton Corp., 516 U. S. 325 ; Bacchus Imports, 468 U. S. 263 ; Central Greyhound, 334 U. S. 653 ; Gwin, White, 305 U. S. 434 ; J. D. Adams, 304 U. S. 307 . If the principal dissent were to prevail, all of these cases would be thrown into doubt. After all, in those cases, as here, the State’s decision to tax in a way that allegedly discriminates against interstate commerce could be justified by the argument that a State may tax its residents without any Commerce Clause constraints. E While the principal dissent claims that we are departing from principles that have been accepted for “a century” and have been “repeatedly acknowledged by this Court,” see post, at 1, 2, 19, when it comes to providing supporting authority for this assertion, it cites exactly two Commerce Clause decisions that are supposedly inconsistent with our decision today. One is a summary affirmance, West Publishing Co. v. McColgan, 328 U. S. 823 (1946) , and neither actually supports the principal dissent’s argument. In the first of these cases, Shaffer v. Carter, 252 U. S. 37 , a resident of Illinois who earned income from oil in Oklahoma unsuccessfully argued that his Oklahoma income tax assessment violated several provisions of the Federal Constitution. His main argument was based on due process, but he also raised a dormant Commerce Clause challenge. Although the principal dissent relies on Shaffer for the proposition that a State may tax the income of its residents wherever earned, Shaffer did not reject the Commerce Clause challenge on that basis. The dormant Commerce Clause challenge in Shaffer was nothing like the Wynnes’ challenge here. The tax-payer in Shaffer argued that “[i]f the tax is considered an excise tax on business, rather than an income tax proper,” it unconstitutionally burdened interstate commerce. Brief for Appellant, O. T. 1919, No. 531, p. 166. The taxpayer did not argue that this burden occurred because he was subject to double taxation; instead, he argued that the tax was an impermissible direct “tax on interstate business.” Ibid. That argument was based on the notion that States may not impose a tax “directly” on interstate commerce. See supra, at 8–9. After assuming that the taxpayer’s business was engaged in interstate commerce, we held that “it is sufficient to say that the tax is imposed not upon the gross receipts, but only upon the net proceeds, and is plainly sustainable, even if it includes net gains from interstate commerce. [United States Glue Co. v. Town of Oak Creek], 247 U. S. 321 .” Shaffer, supra, at 57 (citation omitted). Shaffer thus did not adjudicate anything like the double taxation argument that was accepted in later cases and is before us today. And the principal dissent’s suggestion that Shaffer allows States to levy discriminatory net income taxes is refuted by a case decided that same day. In Travis, a Connecticut corporation challenged New York’s net income tax, which allowed residents, but not nonresidents, certain tax exemptions. The Court first rejected the taxpayer’s due process argument as “settled by our decision in Shaffer.” 252 U. S., at 75. But that due process inquiry was not the end of the matter: the Court then separately considered—and sustained—the argument that the net income tax’s disparate treatment of residents and nonresidents violated the Privileges and Immunities Clause. Id., at 79–80. The second case on which the principal dissent relies, West Publishing, is a summary affirmance and thus has “considerably less precedential value than an opinion on the merits.” Illinois Bd. of Elections v. Socialist Workers Party, 440 U. S. 173 –181 (1979). A summary affirmance “ ‘is not to be read as a renunciation by this Court of doctrines previously announced in our opinions after full argument.’ ” Mandel v. Bradley, 432 U. S. 173, 176 (1977) (per curiam) (quoting Fusari v. Steinberg, 419 U. S. 379, 392 (1975) (Burger, C. J., concurring)). The principal dissent’s reliance on the state-court decision below in that case is particularly inappropriate because “a summary affirmance is an affirmance of the judgment only,” and “the rationale of the affirmance may not be gleaned solely from the opinion below.” 432 U. S., at 176. Moreover, we do not disagree with the result of West Publishing. The tax in that case was levied only on “ ‘the net income of every corporation derived from sources within this State,’ ” and thus was an internally consistent and nondiscriminatory tax scheme. See West Publishing Co. v. McColgan, 27 Cal. 2d 705, 707, n., 166 P. 2d 861, 862, n. (1946) (emphasis added). Moreover, even if we did disagree with the result, the citation in our summary affirmance to United States Glue Co. suggests that our decision was based on the since-discarded distinction between net income and gross receipts taxes. West Publishing did not—indeed, it could not—repudiate the double taxation cases upon which we rely. The principal dissent also finds it significant that, when States first enacted modern income taxes in the early 1900’s, some States had tax schemes similar to Maryland’s. This practice, however, was by no means universal. A great many States—such as Alabama, Colorado, Georgia, Kentucky, and Maryland—had early income tax schemes that allowed their residents a credit against taxes paid to other States. See Ala. Code, Tit. 51, ch. 17, §390 (1940); Colo. Stat. Ann., ch. 84A, §38 (Cum. Supp. 1951); Ga. Code Ann. §92–3111 (1974); Carroll’s Ky. Stat. Ann., ch. 108, Art. XX, §4281b–15 (Baldwin rev. 1936); Md. Ann. Code, Art. 81, ch. 277, §231 (1939). Other States also adopted internally consistent tax schemes. For example, Massachusetts and Utah taxed only the income of residents, not nonresidents. See Mass. Gen. Laws, ch. 62 (1932); Utah Rev. Stat. §80–14–1 et seq. (1933). In any event, it is hardly surprising that these early state ventures into the taxation of income included some protectionist regimes that favored the local economy over interstate commerce. What is much more significant is that over the next century, as our Commerce Clause juris-prudence developed, the States have almost entirely abandoned that approach, perhaps in recognition of their doubtful constitutionality. Today, the near-universal state practice is to provide credits against personal income taxes for such taxes paid to other States. See 2 J. Hellerstein & W. Hellerstein, State Taxation, ¶20.10, pp. 20–163 to 20–164 (3d ed. 2003).[4] F 1 As previously noted, the tax schemes held to be unconstitutional in J. D. Adams, Gwin, White, and Central Greyhound, had the potential to result in the discriminatory double taxation of income earned out of state and created a powerful incentive to engage in intrastate rather than interstate economic activity. Although we did not use the term in those cases, we held that those schemes could be cured by taxes that satisfy what we have subsequently labeled the “internal consistency” test. See Jefferson Lines, 514 U. S., at 185 (citing Gwin, White as a case requiring internal consistency); see also 1 Trost §2:19, at 122–123, and n. 160 (explaining that the internal consistency test has its origins in Western Live Stock, J. D. Adams, and Gwin, White). This test, which helps courts identify tax schemes that discriminate against interstate commerce, “looks to the structure of the tax at issue to see whether its identical application by every State in the Union would place interstate commerce at a disadvantage as compared with commerce intrastate.” 514 U. S., at 185. See also, e.g., Tyler Pipe, 483 U. S., at 246–248; Armco, 467 U. S., at 644–645; Container Corp. of America v. Franchise Tax Bd., 463 U. S. 159, 169 (1983) . By hypothetically assuming that every State has the same tax structure, the internal consistency test allows courts to isolate the effect of a defendant State’s tax scheme. This is a virtue of the test because it allows courts to distinguish between (1) tax schemes that inherently discriminate against interstate commerce without regard to the tax policies of other States, and (2) tax schemes that create disparate incentives to engage in interstate commerce (and sometimes result in double taxation) only as a result of the interaction of two different but nondiscriminatory and internally consistent schemes. See Armco, supra, at 645–646; Moorman, 437 U. S., at 277, n. 12; Brief for Tax Economists as Amici Curiae 23–24 (hereinafter Brief for Tax Economists); Brief for Michael S. Knoll & Ruth Mason as Amici Curiae 18–23 (hereinafter Brief for Knoll & Mason). The first category of taxes is typically unconstitutional; the second is not.[5] See Armco, supra, at 644–646; Moorman, supra, at 277, and n. 12. Tax schemes that fail the internal consistency test will fall into the first category, not the second: “[A]ny cross-border tax disadvantage that remains after application of the [test] cannot be due to tax disparities”[6] but is instead attributable to the taxing State’s discriminatory policies alone. Neither petitioner nor the principal dissent questions the economic bona fides of the internal consistency test. And despite its professed adherence to precedent, the principal dissent ignores the numerous cases in which we have applied the internal consistency test in the past. The internal consistency test was formally introduced more than three decades ago, see Container Corp., supra, and it has been invoked in no fewer than seven cases, invalidating the tax in three of those cases. See American Trucking Assns., Inc. v. Michigan Pub. Serv. Comm’n, 545 U. S. 429 (2005) ;[7] Jefferson Lines, Inc., 514 U. S. 175 ; Goldberg, 488 U. S. 252; American Trucking Assns., Inc. v. Scheiner, 483 U. S. 266 (1987) ; Tyler Pipe, 483 U. S. 232 ; Armco, 467 U. S. 638 ; Container Corp., supra. 2 Maryland’s income tax scheme fails the internal consistency test.[8] A simple example illustrates the point. Assume that every State imposed the following taxes, which are similar to Maryland’s “county” and “special nonresident” taxes: (1) a 1.25% tax on income that residents earn in State, (2) a 1.25% tax on income that residents earn in other jurisdictions, and (3) a 1.25% tax on income that nonresidents earn in State. Assume further that two taxpayers, April and Bob, both live in State A, but that April earns her income in State A whereas Bob earns his income in State B. In this circumstance, Bob will pay more income tax than April solely because he earns income interstate. Specifically, April will have to pay a 1.25% tax only once, to State A. But Bob will have to pay a 1.25% tax twice: once to State A, where he resides, and once to State B, where he earns the income. Critically—and this dispels a central argument made by petitioner and the principal dissent—the Maryland scheme’s discriminatory treatment of interstate commerce is not simply the result of its interaction with the taxing schemes of other States. Instead, the internal consistency test reveals what the undisputed economic analysis shows: Maryland’s tax scheme is inherently discriminatory and operates as a tariff. See Brief for Tax Economists 4, 9; Brief for Knoll & Mason 2. This identity between Maryland’s tax and a tariff is fatal because tariffs are “[t]he paradigmatic example of a law discriminating against interstate commerce.” West Lynn, 512 U. S., at 193. Indeed, when asked about the foregoing analysis made by amici Tax Economists and Knoll & Mason, counsel for Maryland responded, “I don’t dispute the mathematics. They lose me when they switch from tariffs to income taxes.” Tr. of Oral Arg. 9. But Maryland has offered no reason why our analysis should change because we deal with an income tax rather than a formal tariff, and we see none. After all, “tariffs against the products of other States are so patently unconstitutional that our cases reveal not a single attempt by any State to enact one. Instead, the cases are filled with state laws that aspire to reap some of the benefits of tariffs by other means.” West Lynn, supra, at 193. None of our dissenting colleagues dispute this economic analysis. The principal dissent focuses instead on a supposed “oddity” with our analysis: The principal dissent can envision other tax schemes that result in double taxation but do not violate the internal consistency test. This would happen, the principal dissent points out, if State A taxed only based on residence and State B taxed only based on source. Post, at 17 (Ginsburg, J., dissenting); see also post, at 7 (Scalia, J., dissenting). Our prior decisions have already considered and rejected this precise argument—and for good reason. For example, in Armco, we struck down an internally inconsistent tax that posed a risk of double taxation even though we recognized that there might be other permissible arrangements that would result in double taxation. Such schemes would be constitutional, we explained, because “such a result would not arise from impermissible discrimination against interstate commerce.” 467 U. S., at 645. The principal dissent’s protest that our distinction is “entirely circular,” post, at 17–18, n. 10, misunderstands the critical distinction, recognized in cases like Armco, between discriminatory tax schemes and double taxation that results only from the interaction of two different but nondiscriminatory tax schemes. See also Moorman, 437 U. S., at 277, n. 12 (distinguishing “the potential consequences of the use of different formulas by the two States,” which is not prohibited by the Commerce Clause, from discrimination that “inhere[s] in either State’s formula,” which is prohibited). Petitioner and the Solicitor General argue that Maryland’s tax is neutral, not discriminatory, because the same tax applies to all three categories of income. Specifically, they point out that the same tax is levied on (1) residents who earn income in State, (2) residents who earn income out of State, and (3) nonresidents who earn income in State. But the fact that the tax might have “ ‘the advantage of appearing nondiscriminatory’ does not save it from invalidation.” Tyler Pipe, 483 U. S., at 248 (quoting General Motors Corp. v. Washington, 377 U. S. 436, 460 (1964) (Goldberg, J., dissenting)). See also American Trucking Assns., Inc. v. Scheiner, 483 U. S. at, 281 (dormant Commerce Clause applies to state taxes even when they “do not allocate tax burdens between insiders and outsiders in a manner that is facially discriminatory”); Maine v. Taylor, 477 U. S. 131, 138 (1986) (a state law may discriminate against interstate commerce “ ‘either on its face or in practical effect’ ” (quoting Hughes, 441 U. S., at 336)). In this case, the internal consistency test and economic analysis—indeed, petitioner’s own concession—confirm that the tax scheme operates as a tariff and discriminates against interstate commerce, and so the scheme is invalid. Petitioner and the principal dissent, post, at 6, also note that by offering residents who earn income in interstate commerce a credit against the “state” portion of the income tax, Maryland actually receives less tax revenue from residents who earn income from interstate commerce rather than intrastate commerce. This argument is a red herring. The critical point is that the total tax burden on interstate commerce is higher, not that Maryland may receive more or less tax revenue from a particular tax-payer. See Armco, supra, at 642–645. Maryland’s tax un-constitutionally discriminates against interstate commerce, and it is thus invalid regardless of how much a particular taxpayer must pay to the taxing State. Once again, a simple hypothetical illustrates the point. Assume that State A imposes a 5% tax on the income that its residents earn in-state but a 10% tax on income they earn in other jurisdictions. Assume also that State A happens to grant a credit against income taxes paid to other States. Such a scheme discriminates against interstate commerce because it taxes income earned interstate at a higher rate than income earned intrastate. This is so despite the fact that, in certain circumstances, a resident of State A who earns income interstate may pay less tax to State A than a neighbor who earns income intrastate. For example, if Bob lives in State A but earns his income in State B, which has a 6% income tax rate, Bob would pay a total tax of 10% on his income, though 6% would go to State B and (because of the credit) only 4% would go to State A. Bob would thus pay less to State A than his neighbor, April, who lives in State A and earns all of her income there, because April would pay a 5% tax to State A. But Bob’s tax burden to State A is irrelevant; his total tax burden is what matters. The principal dissent is left with two arguments against the internal consistency test. These arguments are inconsistent with each other and with our precedents. First, the principal dissent claims that the analysis outlined above requires a State taxing based on residence to “recede” to a State taxing based on source. Post, at 1–2. We establish no such rule of priority. To be sure, Maryland could remedy the infirmity in its tax scheme by offering, as most States do, a credit against income taxes paid to other States. See Tyler Pipe, supra, at 245–246, and n. 13. If it did, Maryland’s tax scheme would survive the internal consistency test and would not be inherently discriminatory. Tweak our first hypothetical, supra, at 21–22, and assume that all States impose a 1.25% tax on all three categories of income but also allow a credit against income taxes that residents pay to other jurisdictions. In that circumstance, April (who lives and works in State A) and Bob (who lives in State A but works in State B) would pay the same tax. Specifically, April would pay a 1.25% tax only once (to State A), and Bob would pay a 1.25% tax only once (to State B, because State A would give him a credit against the tax he paid to State B). But while Maryland could cure the problem with its current system by granting a credit for taxes paid to other States, we do not foreclose the possibility that it could comply with the Commerce Clause in some other way. See Brief for Tax Economists 32; Brief for Knoll & Mason 28–30. Of course, we do not decide the constitutionality of a hypothetical tax scheme that Maryland might adopt because such a scheme is not before us. That Maryland’s existing tax unconstitutionally discriminates against interstate commerce is enough to decide this case. Second, the principal dissent finds a “deep flaw” with the possibility that “Maryland could eliminate the inconsistency [with its tax scheme] by terminating the special nonresident tax—a measure that would not help the Wynnes at all.” Post, at 16. This second objection refutes the first. By positing that Maryland could remedy the unconstitutionality of its tax scheme by eliminating the special nonresident tax, the principal dissent accepts that Maryland’s desire to tax based on residence need not “recede” to another State’s desire to tax based on source. Moreover, the principal dissent’s supposed flaw is simply a truism about every case under the dormant Commerce Clause (not to mention the Equal Protection Clause): Whenever government impermissibly treats like cases differently, it can cure the violation by either “leveling up” or “leveling down.” Whenever a State impermissibly taxes interstate commerce at a higher rate than intrastate commerce, that infirmity could be cured by lowering the higher rate, raising the lower rate, or a combination of the two. For this reason, we have concluded that “a State found to have imposed an impermissibly discriminatory tax retains flexibility in responding to this determination.” McKesson Corp. v. Division of Alcoholic Beverages and Tobacco, Fla. Dept. of Business Regulation, 496 U. S. 18 –40 (1990). See also Associated Industries of Mo. v. Lohman, 511 U. S. 641, 656 (1994) ; Fulton Corp., 516 U. S., at 346–347. If every claim that suffers from this “flaw” cannot succeed, no dormant Commerce Clause or equal protection claim could ever succeed. G Justice Scalia would uphold the constitutionality of the Maryland tax scheme because the dormant Commerce Clause, in his words, is “a judicial fraud.” Post, at 2. That was not the view of the Court in Gibbons v. Ogden, 9 Wheat, at 209, where Chief Justice Marshall wrote that there was “great force” in the argument that the Commerce Clause by itself limits the power of the States to enact laws regulating interstate commerce. Since that time, this supposedly fraudulent doctrine has been applied in dozens of our opinions, joined by dozens of Justices. Perhaps for this reason, petitioner in this case, while challenging the interpretation and application of that doctrine by the court below, did not ask us to reconsider the doctrine’s validity. Justice Scalia does not dispute the fact that State tariffs were among the principal problems that led to the adoption of the Constitution. See post, at 3. Nor does he dispute the fact that the Maryland tax scheme is tantamount to a tariff on work done out of State. He argues, however, that the Constitution addresses the problem of state tariffs by prohibiting States from imposing “ ‘Imposts or Duties on Imports or Exports.’ ” Ibid. (quoting Art. I, §10, cl. 2). But he does not explain why, under his interpretation of the Constitution, the Import-Export Clause would not lead to the same result that we reach under the dormant Commerce Clause. Our cases have noted the close relationship between the two provisions. See, e.g., State Tonnage Tax Cases, 12 Wall. 204, 214 (1871). Justice Thomas also refuses to accept the dormant Commerce Clause doctrine, and he suggests that the Constitution was ratified on the understanding that it would not prevent a State from doing what Maryland has done here. He notes that some States imposed income taxes at the time of the adoption of the Constitution, and he observes that “[t]here is no indication that those early state income tax schemes provided credits for income taxes paid elsewhere.” Post, at 2 (dissenting opinion). “It seems highly implausible,” he writes, “that those who ratified the Commerce Clause understood it to conflict with the income tax laws of their States and nonetheless adopted it without a word of concern.” Ibid. This argument is plainly unsound. First, because of the difficulty of interstate travel, the number of individuals who earned income out of State in 1787 was surely very small. (We are unaware of records showing, for example, that it was common in 1787 for workers to commute to Manhattan from New Jersey by rowboat or from Connecticut by stagecoach.) Second, Justice Thomas has not shown that the small number of individuals who earned income out of State were taxed twice on that income. A number of Founding-era income tax schemes appear to have taxed only the income of residents, not nonresidents. For example, in his report to Congress on direct taxes, Oliver Wolcott, Jr., Secretary of Treasury, describes Delaware’s income tax as being imposed only on “the inhabitants of this State,” and he makes no mention of the taxation of nonresidents’ income. Report to 4th Cong., 2d Sess. (1796), concerning Direct Taxes, in 1 American State Papers, Finance 429 (1832). Justice Thomas likewise understands that the Massachusetts and Delaware income taxes were imposed only on residents. Post, at 2, n. These tax schemes, of course, pass the internal consistency test. Moreover, the difficulty of administering an income tax on nonresidents would have diminished the likelihood of double taxation. See R. Blakey, State Income Taxation 1 (1930). Third, even if some persons were taxed twice, it is unlikely that this was a matter of such common knowledge that it must have been known by the delegates to the State ratifying conventions who voted to adopt theConstitution. * * * For these reasons, the judgment of the Court of Appeals of Maryland is affirmed. It is so ordered.Notes 1 Under federal law, S corporations permit shareholders “to elect a ‘pass-through’ taxation system under which income is subjected to only one level of taxation. The corporation’s profits pass through directly to its shareholders on a pro rata basis and are reported on the shareholders’ individual tax returns.” Gitlitz v. Commissioner, 531 U. S. 206, 209 (2001) (citation omitted). Maryland affords similar pass-through treatment to the income of an S corporation. 431 Md. 147, 158, 64 A. 3d 453, 459 (2013). By contrast, C corporations—organized under Subchapter C rather than S of Chapter 1 of the Internal Revenue Code—must pay their own taxes because they are considered to be separate tax entities from their shareholders. 14A W. Fletcher, Cyclopedia of the Law of Corporations §§6971, 6973 (rev. ed. 2008 and Cum. Supp. 2014–2015). Because of limitations on the number and type of shareholders they may have, S corporations tend to be smaller, more closely held corporations. Id., §§7025.50, 7026. 2 The principal dissent mischaracterizes the import of the Court’s statement in Moorman that a gross receipts tax is “ ‘more burdensome’ ” than a net income tax. Post, at 13. This was a statement about the relative economic impact of the taxes (a gross receipts tax applies regardless of whether the corporation makes a profit). It was not, as Justice Brennan confirmed in dissent, a suggestion that net income taxes are subject to lesser constitutional scrutiny than gross receipts taxes. Indeed, we noted in Moorman that “the actual burden on interstate commerce would have been the same had Iowa imposed a plainly valid gross-receipts tax instead of the challenged [net] income tax.” Moorman Mfg. Co. v. Bair, 437 U. S. 267 –281 (1978). 3 Similarly, we have sustained dormant Commerce Clause challenges by corporate residents of the State that imposed the burden on interstate commerce. See, e.g., Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U. S. 564, 567 (1997) ; Fulton Corp. v. Faulkner, 516 U. S. 325, 328 (1996) ; Central Greyhound Lines, Inc. v. Mealey, 334 U. S. 653, 654 (1948) ; Gwin, White & Prince, Inc. v. Henneford, 305 U. S. 434, 435 (1939) ; J. D. Adams Mfg. Co. v. Storen, 304 U. S. 307, 308 (1938) . 4 There is no merit to petitioner’s argument that Maryland is free to adopt any tax scheme that is not actually intended to discriminate against interstate commerce. Reply Brief 7. The Commerce Clause regulates effects, not motives, and it does not require courts to inquire into voters’ or legislators’ reasons for enacting a law that has a discriminatory effect. See, e.g., Associated Industries of Mo. v. Lohman, 511 U. S. 641, 653 (1994) ; Philadelphia v. New Jersey, 437 U. S. 617 –627 (1978); Hunt v. Washington State Apple Advertising Comm’n, 432 U. S. 333 –353 (1977). 5 Our cases have held that tax schemes may be invalid under the dormant Commerce Clause even absent a showing of actual double taxation. Mobil Oil Corp. v. Commissioner of Taxes of Vt., 445 U. S. 425, 444 (1980) ; Gwin, White, 305 U. S., at 439. We note, however, that petitioner does not dispute that respondents have been subject to actual multiple taxation in this case. 6 Mason, Made in America for European Tax: The Internal Consistency Test, 49 Boston College L. Rev. 1277, 1310 (2008). 7 The principal dissent and Justice Scalia inaccurately state that the Court in American Trucking “conceded that a trucking tax ‘fail[ed] the “internal consistency” test,’ but upheld the tax anyway.” Post, at 5 (Scalia, J., dissenting); see also post, at 14–15 (Ginsburg, J., dissenting). The Court did not say that the tax in question “failed the ‘internal consistency test.’ ” The Court wrote that this is what petitioner argued. See American Trucking, 545 U. S., at 437. And the Court did not concede that this was true. The tax in that case was a flat tax on any truck that made point-to-point deliveries in Michigan. The tax therefore fell on all trucks that made solely intrastate deliveries and some that made interstate deliveries, namely, those that also made some intrastate deliveries. What the Court “concede[d]” was that “if all States [adopted a similar tax], an interstate truck would have to pay fees totaling several hundred dollars, or even several thousand dollars, were it to ‘top off’ its business by carrying local loads in many (or even all) other States.” Id., at 438 (emphasis added). But that was not the same as a concession that the tax violated the internal consistency test. 8 In order to apply the internal consistency test in this case, we must evaluate the Maryland income tax scheme as a whole. That scheme taxes three separate categories of income: (1) the “county tax” on income that Maryland residents earn in Maryland; (2) the “county tax” on income that Maryland residents earn in other States; and (3) the “special nonresident tax” on income that nonresidents earn in Maryland. For Commerce Clause purposes, it is immaterial that Maryland assigns different labels (i.e., “county tax” and “special nonresident tax”) to these taxes. In applying the dormant Commerce Clause, they must be considered as one. Cf. Oregon Waste Systems, Inc. v. Department of Environmental Quality of Ore., 511 U. S. 93 –103 (1994) (independent taxes on intrastate and interstate commerce are “compensatory” if they are rough equivalents imposed upon substantially similar events). If state labels controlled, a State would always be free to tax domestic, inbound, and outbound income at discriminatory rates simply by attaching different labels.

574.US.81

A defendant seeking to remove a case from state to federal court must file in the federal forum a notice of removal “containing a short and plain statement of the grounds for removal.” 28 U. S. C. §1446(a). Respondent Owens filed a putative class action in Kansas state court, seeking compensation for damages class members allegedly sustained when petitioners (collectively, Dart) underpaid royalties due under certain oil and gas leases. Dart removed the case to the Federal District Court, invoking the Class Action Fairness Act of 2005 (CAFA), which gives federal courts jurisdiction over class actions if the amount in controversy exceeds $5 million, 28 U. S. C. §1332(d)(2). Dart’s notice of removal alleged that the purported underpayments totaled over $8.2 million. Owens moved to remand the case to state court, asserting that the removal notice was “deficient as a matter of law” because it included “no evidence” proving that the amount in controversy exceeded $5 million. In response, Dart submitted an executive’s detailed declaration supporting an amount in controversy in excess of $11 million. The District Court granted Owens’ remand motion, reading Tenth Circuit precedent to require proof of the amount in controversy in the notice of removal itself. Dart petitioned the Tenth Circuit for permission to appeal, see §1453(c)(1), but that court denied review and rehearing en banc. Held: 1. As specified in §1446(a), a defendant’s notice of removal need include only a plausible allegation that the amount in controversy exceeds the jurisdictional threshold; the notice need not contain evidentiary submissions. Section 1446(a) tracks the general pleading requirement stated in Rule 8(a) of the Federal Rules of Civil Procedure. By borrowing Rule 8(a)’s “short and plain statement” standard, corroborative history indicates, Congress intended to clarify that courts should “apply the same liberal rules [to removal allegations as] to other matters of pleading.” H. R. Rep. No. 100–889, p. 71. The amount-in-controversy allegation of a plaintiff invoking federal-court jurisdiction is accepted if made in good faith. See, e.g., Mt. Healthy City Bd. of Ed. v. Doyle, 429 U.S. 274, 276. Similarly, the amount-in-controversy allegation of a defendant seeking federal-court adjudication should be accepted when not contested by the plaintiff or questioned by the court. In the event that the plaintiff does contest the defendant’s allegations, both sides submit proof and the court decides, by a preponderance of the evidence, whether the amount-in-controversy requirement has been satisfied, see §1446(c)(2)(B). In remanding the case to state court, the District Court relied, in part, on a purported “presumption” against removal, but no antiremoval presumption attends cases invoking CAFA, a statute Congress enacted to facilitate adjudication of certain class actions in federal court. See Standard Fire Ins. Co. v. Knowles, 568 U. S. ___, ___. Pp. 4–7. 2. The District Court erred in remanding this case for want of an evidentiary submission in the notice of removal, and the Tenth Circuit abused its discretion in denying review of that decision. Pp. 7–14. (a) This Court concludes that no jurisdictional barrier impedes settlement of the question presented: whether evidence supporting the amount in controversy must be included in a notice of removal. The case was “in” the Tenth Circuit because of Dart’s application for leave to appeal, and the Court has jurisdiction to review what the Court of Appeals did with that application. See 28 U. S. C. §1254; Hohn v. United States, 524 U.S. 236, 248. Pp. 7–8. (b) While appellate review of a remand order is discretionary, exercise of that discretion is not rudderless, see Highmark Inc. v. Allcare Health Management System, Inc., 572 U. S. ___, ___, and a court “would necessarily abuse its discretion if it based its ruling on an erroneous view of the law,” Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 405. The Tenth Circuit had previously stated considerations bearing on the intelligent exercise of discretion under §1453(c)(1). One of those considerations is particularly relevant here: a court of appeals should inquire whether, if a district court’s remand order remains undisturbed, the case will “leave the ambit of the federal courts for good, precluding any other opportunity for [the defendant] to vindicate its claimed legal entitlement [under CAFA] . . . to have a federal tribunal adjudicate the merits.” BP America, Inc. v. Oklahoma ex rel. Edmondson, 613 F.3d 1029, 1035. Thus the Tenth Circuit’s own guide weighed heavily in favor of accepting Dart’s appeal. In practical effect, the Court of Appeals’ denial of review established the law—the requirement of proof of the amount in controversy in the removal notice—not simply for this case, but for future CAFA removals sought by defendants in the Tenth Circuit, leaving those defendants with no realistic opportunity to resist making the evidentiary submission. The District Court, driven by its conscientious endeavor to follow Circuit precedent, erred in ruling that Dart’s amount-in-controversy allegation failed for want of proof. It was an abuse of discretion for the Tenth Circuit to deny Dart’s request for review, for that disposition fastened on district courts within the Circuit an erroneous view of the law. Contrary to the law the District Court derived from Tenth Circuit precedent, a removal notice need only plausibly allege, not detail proof of, the amount in controversy.Pp. 8–14. Vacated and remanded. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Breyer, Alito, and Sotomayor, JJ., joined. Scalia, J., filed a dissenting opinion, in which Kennedy and Kagan, JJ., joined, and in which Thomas, J., joined as to all but the final sentence. Thomas, J., filed a dissenting opinion.

To remove a case from a state court to a federal court, a defendant must file in the federal forum a notice of removal “containing a short and plain statement of the grounds for removal.”28 U. S. C. §1446(a). When re-moval is based on diversity of citizenship, an amount-in-controversy requirement must be met. Ordinarily, “the matter in controversy [must] excee[d] the sum or value of $75,000.” §1332(a). In class actions for which the requirement of diversity of citizenship is relaxed, §1332(d)(2)(A)–(C), “the matter in controversy [must] excee[d]the sum or value of $5,000,000,” §1332(d)(2). If theplaintiff’s complaint, filed in state court, demands mon-etary relief of a stated sum, that sum, if asserted ingood faith, is “deemed to be the amount in controversy.” §1446(c)(2). When the plaintiff’s complaint does not state the amount in controversy, the defendant’s notice of removal may do so. §1446(c)(2)(A). To assert the amount in controversy adequately in the removal notice, does it suffice to allege the requisite amount plausibly, or must the defendant incorporate into the notice of removal evidence supporting the allegation? That is the single question argued here and below by the parties and the issue on which we granted review. The answer, we hold, is supplied by the removal statute itself. A statement “short and plain” need not contain evidentiary submissions.I Brandon W. Owens, plaintiff below and respondent here, filed a putative class action in Kansas state court alleging that defendants Dart Cherokee Basin Operating Company, LLC, and Cherokee Basin Pipeline, LLC (collectively, Dart), underpaid royalties owed to putative class members under certain oil and gas leases. The complaint sought “a fair and reasonable amount” to compensate putative class members for “damages” they sustained due to the alleged underpayments. App. to Pet. for Cert. 34a, 35a. Invoking federal jurisdiction under the Class Action Fairness Act of 2005 (CAFA), Dart removed the case to the U. S. District Court for the District of Kansas. CAFA gives federal courts jurisdiction over certain class actions, defined in §1332(d)(1), if the class has more than 100 members, the parties are minimally diverse, and the amount in controversy exceeds $5 million. §1332(d)(2), (5)(B); see Standard Fire Ins. Co. v. Knowles, 568 U. S. ___, ___ (2013) (slip op., at 3). Dart’s notice of removal alleged that all three requirements were satisfied. With respect to the amount in controversy, Dart stated that the purported underpayments to putative class members totaled more than $8.2 million. Owens moved to remand the case to state court. The notice of removal was “deficient as a matter of law,”Owens asserted, because it included “no evidence” proving that the amount in controversy exceeded $5 million. App. to Pet. for Cert. 46a, 53a. In response, Dart submitted a declaration by one of its executive officers. The declaration included a detailed damages calculation indicating that the amount in controversy, sans interest, exceeded $11 million. Without challenging Dart’s calculation,Owens urged that Dart’s amount-in-controversy submissioncame too late. “[The] legally deficient [notice of removal],” Owens maintained, could not be cured by “post-removal evidence about the amount in controversy.” Id., at 100a. Reading Tenth Circuit precedent to require proof ofthe amount in controversy in the notice of removal itself, the District Court granted Owens’ remand motion. Dart’sdeclaration, the District Court held, could not serve to keep the case in federal court. The Tenth Circuit, as the District Court read Circuit precedent, “has consistently held that reference to factual allegations or evidence out-side of the petition and notice of removal is not permitted to determine the amount in controversy.” App. to Pet.for Cert. 26a, and n. 37 (citing Laughlin v. Kmart Corp., 50 F. 3d 871, 873 (1995); Martin v. Franklin Capital Corp., 251 F. 3d 1284, 1291, n. 4 (2001); Oklahoma Farm Bureau Mut. Ins. Co. v. JSSJ Corp., 149 Fed. Appx. 775 (2005)). Ordinarily, remand orders “[are] not reviewable on appeal or otherwise.” §1447(d). There is an exception, however, for cases invoking CAFA. §1453(c)(1). In such cases, “a court of appeals may accept an appeal from an order of a district court granting or denying a motion to remand.” Ibid. Citing this exception, Dart petitioned the Tenth Circuit for permission to appeal. “Upon careful consideration of the parties’ submissions, as well as the applicable law,” the Tenth Circuit panel, dividing two-to-one, denied review. App. to Pet. for Cert. 13a–14a. An evenly divided court denied Dart’s petition for en banc review. Dissenting from the denial of rehearing en banc, Judge Hartz observed that the Tenth Circuit “[had] let stand a district-court decision that will in effect impose in this circuit requirements for notices of removal that are even more onerous than the code pleading requirements that . . . federal courts abandoned long ago.” 730 F. 3d 1234 (2013). The Tenth Circuit was duty-bound to grant Dart’s petition for rehearing en banc, Judge Hartz urged, because the opportunity “to correct the law in our circuit” likely would not arise again. Id., at 1235. Henceforth, Judge Hartz explained, “any diligent attorney . . . would submit to the evidentiary burden rather than take a chance on remand to state court.” Ibid. Dart filed a petition for certiorari in this Court requesting resolution of the following question: “Whether a defendant seeking removal to federal court is required to include evidence supporting federal jurisdiction in the notice of removal, or is alleging the required ‘short and plain statement of the grounds for removal’ enough?” Pet. for Cert. i. Owens’ brief in opposition raised no impediment to this Court’s review. (Nor, later, did Owens’ merits brief suggest any barrier to our consideration of Dart’s petition.) We granted certiorari to resolve a division among the Circuits on the question presented. 572 U. S. ___ (2014). Compare Ellenburg v. Spartan Motors Chassis, Inc., 519 F. 3d 192, 200 (CA4 2008) (a removing party’s notice of removal need not “meet a higher pleading standard than the one imposed on a plaintiff in drafting an initial complaint”), and Spivey v. Vertrue, Inc., 528 F. 3d 982, 986 (CA7 2008) (similar), with Laughlin, 50 F. 3d, at 873 (“the requisite amount in controversy . . . must be affirmatively established on the face of either the petition or the removal notice”).II As noted above, a defendant seeking to remove a case to a federal court must file in the federal forum a notice of removal “containing a short and plain statement of the grounds for removal.” §1446(a). By design, §1446(a) tracks the general pleading requirement stated in Rule 8(a) of the Federal Rules of Civil Procedure. See 14C C. Wright, A. Miller, E. Cooper, & J. Steinman, Federal Practice and Procedure §3733, pp. 639–641 (4th ed. 2009) (“Section 1446(a) requires only that the grounds for removal be stated in ‘a short and plain statement’—terms borrowed from the pleading requirement set forth in Federal Rule of Civil Procedure 8(a).”). The legislative history of §1446(a) is corroborative. Congress, by borrowing the familiar “short and plain statement” standard from Rule 8(a), intended to “simplify the ‘pleading’ requirements for removal” and to clarify that courts should “apply the same liberal rules [to removal allegations] that are applied to other matters of pleading.” H. R. Rep. No. 100–889, p. 71 (1988). See also ibid. (disapproving decisions requiring “detailed pleading”). When a plaintiff invokes federal-court jurisdiction, the plaintiff’s amount-in-controversy allegation is accepted if made in good faith. See, e.g., Mt. Healthy City Bd. of Ed. v. Doyle,429 U. S. 274,276 (1977) (“ ‘[T]he sum claimed by the plaintiff controls if the claim is apparently made in good faith.’ ”) (quoting St. Paul Mercury Indemnity Co. v. Red Cab Co.,303 U. S. 283,288 (1938); alteration in original). Similarly, when a defendant seeks federal-court adjudication, the defendant’s amount-in-controversy allegation should be accepted when not contested by the plaintiff or questioned by the court. Indeed, the Tenth Circuit, although not disturbing prior decisions demanding proof together with the removal notice, recognized that it was anomalous to treat commencing plaintiffs and removing defendants differently with regard to the amount in controversy. See McPhail v. Deere & Co., 529 F. 3d 947, 953 (2008) (requiring proof by defendant but not by plaintiff “bears no evident logical relationship either to the purpose of diversity jurisdiction, or to the principle that those who seek to invoke federal jurisdiction must establish its prerequisites”). If the plaintiff contests the defendant’s allegation, §1446(c)(2)(B) instructs: “[R]emoval . . . is proper on the basis of an amount in controversy asserted” by the defendant “if the district court finds, by the preponderance of the evidence, that the amount in controversy exceeds” the jurisdictional threshold.[1] This provision, added to §1446 as part of the Federal Courts Jurisdiction and Venue Clarification Act of 2011 (JVCA), clarifies the procedure in order when a defendant’s assertion of the amount in controversy is challenged. In such a case, both sides submit proof and the court decides, by a preponderance of the evidence, whether the amount-in-controversy requirement has been satisfied. As the House Judiciary Committee Report on the JVCA observed:“[D]efendants do not need to prove to a legal certainty that the amount in controversy requirement has been met. Rather, defendants may simply allege or assert that the jurisdictional threshold has been met. Discovery may be taken with regard to that question. In case of a dispute, the district court must make findings of jurisdictional fact to which the preponderance standard applies.” H. R. Rep. No. 112–10, p. 16 (2011).Of course, a dispute about a defendant’s jurisdictional allegations cannot arise until after the defendant files a notice of removal containing those allegations. Brief for Dart 14. In remanding the case to state court, the District Court relied, in part, on a purported “presumption” against removal. App. to Pet. for Cert. 28a. See, e.g., Laughlin, 50 F. 3d, at 873 (“[T]here is a presumption against removal jurisdiction.”). We need not here decide whether such a presumption is proper in mine-run diversity cases. It suffices to point out that no antiremoval presumption attends cases invoking CAFA, which Congress enacted to facilitate adjudication of certain class actions in federal court. See Standard Fire Ins. Co., 568 U. S., at ___ (slip op., at 6) (“CAFA’s primary objective” is to “ensur[e] ‘Federal court consideration of interstate cases of national importance.’ ” (quoting §2(b)(2),119Stat.5)); S. Rep. No. 109–14, p. 43 (2005) (CAFA’s “provisions should be read broadly, with a strong preference that interstate class actions should be heard in a federal court if properly removed by any defendant.”). In sum, as specified in §1446(a), a defendant’s notice of removal need include only a plausible allegation that the amount in controversy exceeds the jurisdictional threshold. Evidence establishing the amount is required by §1446(c)(2)(B) only when the plaintiff contests, or the court questions, the defendant’s allegation.III As in Standard Fire Ins. Co., 568 U. S., at ___–___ (slip op., at 2–3), we granted review in this case after the Court of Appeals declined to hear an appeal from a remand order. Neither party in that case or in this one questioned our review authority under28 U. S. C. §1254(1) (“Cases in the courts of appeals may be reviewed . . . [b]y writ of certiorari upon the petition of any party . . . before or after rendition of judgment.”).[2] An amicus brief filed in support of Owens by Public Citizen, Inc., however, raised a jurisdictional impediment. Section 1453(c)(1), Public Citizen noted, provides that “a court of appeals may accept an appeal from an order ofa district court granting or denying a motion to remand a class action to the State court from which it was removed[.]” (Emphasis added.) Because court of appeals review of a remand order is discretionary, see supra, at 3, and the Tenth Circuit exercised its discretion to deny review, Public Citizen urged, “[b]oth parties ask this Court to decide an issue that is not properly before it.” Brief for Public Citizen 6. “Absent grounds for reversing the court of appeals’ decision to deny permission to appeal,” Public Citizen asserted, “the merits of the district court’s decision are not before any appellate court, including this one.” Ibid. Satisfied that there are indeed “grounds for reversing the [Tenth Circuit’s] decision to deny permission to appeal,” we find no jurisdictional barrier to our settlement of the question presented. The case was “in” the Court of Appeals because of Dart’s leave-to-appeal application, and we have jurisdiction to review what the Court of Appeals did with that application. See28 U. S. C. §1254; Hohn v. United States,524 U. S. 236,248 (1998). Owens, we reiterate, did not contest the scope of our review. Discretion to review a remand order is not rudderless. See Highmark Inc. v. Allcare Health Management System, Inc., 572 U. S. ___, ___ (2014) (slip op., at 4) (“matters of discretion are reviewable for abuse of discretion” (internal quotation marks omitted)). A court “would necessarily abuse its discretion if it based its ruling on an erroneous view of the law.” Cooter & Gell v. Hartmarx Corp.,496 U. S. 384,405 (1990). This case fits that bill.[3] There are many signals that the Tenth Circuit relied on the legally erroneous premise that the District Court’s decision was correct. In an earlier case, the Tenth Circuit, following the First Circuit’s lead, stated considerations that it regards as relevant to the intelligent exercise of discretion under §1453(c)(1). BP America, Inc. v. Okla-homa ex rel. Edmondson, 613 F. 3d 1029, 1034–1035 (2010)(adopting factors set out in College of Dental Surgeons of Puerto Rico v. Connecticut Gen. Life Ins. Co., 585 F. 3d 33, 38–39 (CA1 2009)).[4] When the CAFA-related question presented in an appeal from a remand order is “important, unsettled, and recurrent,” the First Circuit instructed, a court of appeals should inquire: “Absent an interlocutory appeal, [will the question] in all probability escape meaningful appellate review.” Id., at 39. Or, as phrased by the Tenth Circuit, if a district court’s remand order remains undisturbed, will the case “leave the ambit of the federal courts for good, precluding any other opportunity for [the defendant] to vindicate its claimed legal entitlement [under CAFA] . . . to have a federal tribunal adjudicate the merits.” BP America, 613 F. 3d, at 1035. See also Coffey v. Freeport McMoran Copper & Gold, 581 F. 3d 1240, 1247 (CA10 2009) (noting that “the purpose of §1453(c)(1) isto develop a body of appellate law interpreting CAFA”(brackets and internal quotation marks omitted)). Thus, the Tenth Circuit’s own guide weighed heavily in favor of accepting Dart’s appeal. That the Court of Appeals, instead, rejected Dart’s appeal strongly suggests that the panel thought the District Court got it right in requiring proof of the amount in controversy in the removal notice. In practical effect, the Court of Appeals’ denial of review established the law not simply for this case, but for future CAFA removals sought by defendants in the Tenth Circuit. The likelihood is slim that a later case will arise in which the Tenth Circuit will face a plea to retract the rule that both Owens and the District Court ascribed to decisions of the Court of Appeals: Defendants seeking to remove under CAFA must be sent back to state court unless they submit with the notice of removal evidence proving the alleged amount in controversy. See supra, at 3. On this point, Judge Hartz’s observation, dissenting from the Tenth Circuit’s denial of rehearing en banc, see supra, at 4, bears recounting in full:“After today’s decision any diligent attorney (and one can assume that an attorney representing a defendant in a case involving at least $5 million—the threshold for removal under CAFA—would have substantial incentive to be diligent) would submit to the evidentiary burden rather than take a chance on remand to state court.” 730 F. 3d, at 1235.With no responsible attorney likely to renew the fray, Judge Hartz anticipated, “the issue will not arise again.” Ibid. Consequently, the law applied by the District Court—demanding that the notice of removal contain evidence documenting the amount in controversy—will be frozen in place for all venues within the Tenth Circuit.[5] Recall that the Court of Appeals denied Dart’s petition for review “[u]pon careful consideration of the parties’ submissions, as well as the applicable law.” App. to Pet. for Cert. 13a. What did the parties submit to the Tenth Circuit? Their presentations urged conflicting views on whether a removing defendant must tender prima facie proof of the amount in controversy as part of the removal notice. And what was “the applicable law” other than the rule recited by the Tenth Circuit in Laughlin and follow-on decisions, i.e., to remove successfully, a defendant must present with the notice of removal evidence proving the amount in controversy.[6] From all signals one can discern then, the Tenth Circuit’s denial of Dart’s request for review of the remand order was infected by legal error. The District Court erred in ruling that Dart’s amount-in-controversy allegation failed for want of proof, but that error was driven by the District Court’s conscientious endeavor to follow Circuit precedent. The parties trained their arguments in the Tenth Circuit, as they did here, on the question whether Dart could successfully remove without detailing in the removal notice evidence of the amount in controversy. See Tr. of Oral Arg. 47 (acknowledgment by Owens’ counsel that “the issues . . . provided to . . . the Tenth Circuit were very similar to what you see in this Court, with the exception of [the question raised by Public Citizen] whether this Court has jurisdiction”). Dissenting from the denial of rehearing en banc, Judge Hartz explained at length why the Tenth Circuit “owe[d] a duty to the bench and bar” to correct the District Court’s misperception and to state as the Circuit’s law: “[A] defendant seeking removal under CAFA need only allege the jurisdictional amount in its notice of removal and must prove that amount only if the plaintiff challenges the allegation.” 730 F. 3d, at 1234, 1238. In this regard, we note, the Tenth Circuit has cautioned against casual rulings on applications like Dart’s. “The decision whether to grant leave to appeal” under §1453(c), the Tenth Circuit stressed, calls for the exercise of the reviewing court’s correctly “informed discretion.” BP America, 613 F. 3d, at 1035 (emphasis added); see supra, at 8–9. Recall, moreover, that Owens never suggested in his written submissions to this Court that anything other than the question presented accounts for the Court of Appeals’ disposition. If Owens believed that the Tenth Circuit’s denial of leave to appeal rested on some other ground, he might have said so in his brief in opposition or, at least, in his merits brief. See this Court’s Rule 15.2; Granite Rock Co. v. Teamsters,561 U. S. 287,306 (2010). He said nothing of that order, for he, like Dart, antici-pated that the question presented was ripe for this Court’s resolution. In the above-described circumstances, we find it an abuse of discretion for the Tenth Circuit to deny Dart’s request for review. Doing so froze the governing rule in the Circuit for this case and future CAFA removal notices, with no opportunity for defendants in Dart’s position responsibly to resist making the evidentiary submission. That situation would be bizarre for a decisionmaker who did not think that the amount in controversy in diversity cases is a matter a removal notice must demonstrate by evidence, not merely credibly allege.[7] And if the Circuit precedent on which the District Court relied misstated the law, as we hold it did, then the District Court’s order remanding this case to the state court is fatally infected by legal error. Careful inspection thus reveals that the two issues Public Citizen invites us to separate—whether the Tenth Circuit abused its discretion in denying review, and whether the District Court’s remand order was erroneous—do not pose genuinely discrete questions. Instead, resolution of both issues depends on the answer to the very same question: What must the removal notice contain? If the notice need not contain evidence, the Tenth Circuit abused its discretion in effectively making the opposing view the law of the Circuit. By the same token, the District Court erred in remanding the case for want of an evidentiary submission in the removal notice. We no doubt have authority to review for abuse of discretion the Tenth Circuit’s denial of Dart’s appeal from the District Court’s remand order, see supra, at 8, and in doing so, to correct the erroneous view of the law the Tenth Circuit’s decision fastened on district courts within the Circuit’s domain.[8]* * * For the reasons stated, the judgment of the U. S. Court of Appeals for the Tenth Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion.It is so ordered.Notes1 Section 1446(c)(2) applies to removals “sought on the basis of the jurisdiction conferred by section 1332(a),” and §1446(c)(2)(B) provides that “removal of the action is proper . . . [if] the amount in controversy exceeds the [in excess of $75,000] amount specified in section 1332(a)” (emphasis added). We assume, without deciding, a point the parties do not dispute: Sections 1446(c)(2) and 1446(c)(2)(B) apply to cases removed under §1332(d)(2), and removal is proper if the amount in controversy exceeds $5 million, the amount specified in §1332(d)(2). See Frederick v. Hartford Underwriters Ins. Co., 683 F. 3d 1242, 1247 (CA10 2012) (“[T]here is no logical reason why we should demand more from a CAFA defendant than other parties invoking federal jurisdiction.” (internal quotation marks omitted)).2 Today’s dissenters joined the opinion in Standard Fire Ins. Co. v. Knowles, 568 U. S. ___ (2013), without suggesting any lack of jurisdiction to reach the merits.3 Justice Scalia’s dissent (hereafter dissent) faults Dart for asserting, late in the day, that the Tenth Circuit abused its discretion, observing that Dart did so only in its reply brief. Post, at 6. But Public Citizen teed up that issue after the parties filed their merits briefs. In view of this Court’s decision in Standard Fire Ins. Co., 568 U. S. ___, see supra, at 7–8, the parties had no cause to address the matter earlier.4 Neither court stated the listed considerations as an inflexible test. We have no occasion in this case to review each of the factors identified by the First and Tenth Circuits.5 The dissent suggests that the Tenth Circuit may have another opportunity to set Circuit precedent straight: A lawyer may be irresponsible or fail to learn from Dart’s experience; or perhaps a lawyer will put in evidence the district court deems insufficient, and then have a go at arguing that the evidence was sufficient and, in any event, “no evidence is required at all.” Post, at 5–6. That such a case will occur, and that the Tenth Circuit would then seize the very opportunity it passed up in Dart’s case, is hardly probable.6 The dissent posits that “the applicable law” might have been something other than the law governing the parties’ submissions. Post, at 3, 4. That is a strained reading of the Tenth Circuit’s expression. Perhaps the Tenth Circuit found this case a “poor vehicle,” the dissent suggests, post, at 2, but no potential vehicle concerns were urged by Owens, and the dissent identifies none. Or the Tenth Circuit might have doubted its “ability to quickly resolve the issue” within the 60-day time limit provided in §1453(c)(2)–(3). Ibid.; see also post, at 4. Section 1453(c)’s timing provision, however, was designed to promote expedition, not to discourage Courts of Appeals from acting on petitions for appeal. As a third “maybe,” the dissent observes that proof of the amount in controversy in removal notices is not “a question unique to [CAFA].” Post, at 3. True, the Tenth Circuit demands such proof in ordinary diversity cases. See Laughlin v. Kmart Corp., 50 F. 3d 871, 873 (1995). But that does not make the imposition one whit less in CAFA cases.7 Caution is in order when attributing a basis to an unreasoned decision. But we have not insisted upon absolute certainty when that basis is fairly inferred from the record. See Taylor v. McKeithen,407 U. S. 191, n. 2 (1972) (per curiam) (rejecting “possible, but unlikely” basis for unreasoned decision); Nixon v. Fitzgerald,457 U. S. 731–743 (1982) (facing an unreasoned Court of Appeals decision, we projected what the Court of Appeals “appears to have” reasoned); Tr. of OralArg. 18–19 (observing that an appellate court often assumes that a first instance court based its unexplained discretionary decision on the ground the prevailing party presented).8 Our disposition does not preclude the Tenth Circuit from asserting and explaining on remand that a permissible ground underlies its decision to decline Dart’s appeal.

576.US.2014_13-1428

During jury selection in respondent Ayala’s murder trial, Ayala, who is Hispanic, objected that seven of the prosecution’s peremptory challenges were impermissibly race-based under Batson v. Kentucky, 476 U. S. 79 . The judge permitted the prosecution to disclose its reasons for the strikes outside the presence of the defense and concluded that the prosecution had valid, race-neutral reasons for the strikes. Ayala was eventually convicted and sentenced to death. On appeal, the California Supreme Court analyzed Ayala’s challenge under both Batson and its state-law analogue, concluding that it was error, as a matter of state law, to exclude Ayala from the hearings. The court held, however, that the error was harmless under state law and that, if a federal error occurred, it too was harmless beyond a reasonable doubt under Chapman v. California, 386 U. S. 18 . Ayala subsequently pressed his claims in federal court. There, the District Court held that even if the ex parte proceedings violated federal law, the state court’s harmlessness finding could not be overturned because it was not contrary to or an unreasonable application of clearly established federal law under 28 U. S. C. §2254(d). A divided panel of the Ninth Circuit disagreed and granted Ayala habeas relief. The panel majority held that the ex parte proceedings violated Ayala’s federal constitutional rights and that the error was not harmless under Brecht v. Abrahamson, 507 U. S. 619 , as to at least three of the seven prospective jurors. Held: Any federal constitutional error that may have occurred by excluding Ayala’s attorney from part of the Batson hearing was harmless. Pp. 9–29. (a) Even assuming that Ayala’s federal rights were violated, he is entitled to habeas relief only if the prosecution cannot demonstrate harmlessness. Glebe v. Frost, 574 U. S. ___, ___. Under Brecht, federal habeas petitioners “are not entitled to habeas relief based on trial error unless they can establish that it resulted in ‘actual prejudice.’ ” 507 U. S., at 637. Because Ayala seeks federal habeas corpus relief, he must meet the Brecht standard, but that does not mean, as the Ninth Circuit thought, that a state court’s harmlessness determination has no significance under Brecht. The Brecht standard subsumes the requirements that §2254(d) imposes when a federal habeas petitioner contests a state court’s determination that a constitutional error was harmless under Chapman. Fry v. Pliler, 551 U. S. 112 . But Brecht did not abrogate the limitation on federal habeas relief that the Antiterrorism and Effective Death Penalty Act of 1996 plainly sets out. There is no dispute that the California Supreme Court held that any federal error was harmless under Chapman, and this decision was an “adjudication on the merits” of Ayala’s claim. Accordingly, a federal court cannot grant Ayala relief unless the state court’s rejection of his claim was contrary to or involved an unreasonable application of clearly established federal law as determined by the Supreme Court, or was based on an unreasonable determination of the facts. Pp. 9–12. (b) Any federal constitutional error was harmless with respect to all seven prospective jurors. Pp. 12–28. (1) The prosecution stated that it struck Olanders D., an African-American man, because it was concerned that he could not impose the death penalty and because of the poor quality of his responses. As the trial court and State Supreme Court found, the record amply supports the prosecution’s concerns, and Ayala cannot establish that the ex parte hearing prejudiced him. The Ninth Circuit misunderstood the role of a federal court in a habeas case. That role is not to conduct de novo review of factual findings and substitute the federal court’s own opinions for the determination made on the scene by the trial judge. Pp. 14–18. (2) The prosecution stated that it struck Gerardo O., a Hispanic man, because he had a poor grasp of English, his answers suggested an unwillingness to impose the death penalty, and he did not appear to get along with other jurors. Each of these reasons was amply supported by the record, and there is no basis for finding that the absence of defense counsel affected the trial judge’s evaluation of the strike. Ayala cannot establish that the ex parte hearing actually prejudiced him or that no fairminded jurist could agree with the state court’s application of Chapman. Once again, the Ninth Circuit’s decision was based on a misapplication of basic rules regarding harmless error. The inquiry is not whether the federal habeas court could definitively say that the defense could make no winning arguments, but whether the evidence in the record raised “grave doubt[s]” about whether the trial judge would have ruled differently. O’Neal v. McAninch, 513 U. S. 432 . That standard was not met in this case. Pp. 18–24. (3) The prosecution stated that it struck Robert M., a Hispanic man, because it was concerned that he could not impose the death penalty and because he had followed a controversial murder trial. Not only was the Ninth Circuit incorrect to suppose that the presence of Ayala’s counsel at the hearing would have made a difference in the trial court’s evaluation of the strike, but the Ninth Circuit failed to mention that defense counsel specifically addressed the issue during voir dire and reminded the judge that Robert M. also made several statements favorable to the death penalty. Thus, the trial judge heard counsel’s arguments and concluded that the record supplied a legitimate basis for the prosecution’s concern. That defense counsel did not have the opportunity to repeat that argument does not create grave doubt about whether the trial court would have decided the issue differently. Pp. 24–26. (4) With regard to Ayala’s Batson objection about the four remaining prospective jurors who were struck, he does not come close to establishing “actual prejudice” under Brecht or that no fairminded jurist could agree with the California Supreme Court’s decision that excluding counsel was harmless. Pp. 26–28. 756 F. 3d 656, reversed and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, and Thomas, JJ., joined. Kennedy, J., and Thomas, J., filed concurring opinions. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, Breyer, and Kagan, JJ., joined.

A quarter-century after a California jury convicted Hector Ayala of triple murder and sentenced him to death, the Court of Appeals for the Ninth Circuit granted Ayala’s application for a writ of habeas corpus and ordered the State to retry or release him. The Ninth Circuit’s decision was based on the procedure used by the trial judge in ruling on Ayala’s objections under Batson v. Kentucky, 476 U. S. 79 (1986) , to some of the prosecution’s peremptory challenges of prospective jurors. The trial judge allowed the prosecutor to explain the basis for those strikes outside the presence of the defense so as not to disclose trial strategy. On direct appeal, the California Supreme Court found that if this procedure violated any federal constitutional right, the error was harmless beyond a reasonable doubt. The Ninth Circuit, however, held that the error was harmful. The Ninth Circuit’s decision was based on the misapplication of basic rules regarding harmless error. Assuming without deciding that a federal constitutional error occurred, the error was harmless under Brecht v. Abrahamson, 507 U. S. 619 (1993) , and the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), 28 U. S. C. §2254(d). I A Ayala’s conviction resulted from the attempted robbery of an automobile body shop in San Diego, California, in April 1985. The prosecution charged Ayala with three counts of murder, one count of attempted murder, one count of robbery, and three counts of attempted robbery. The prosecution also announced that it would seek the death penalty on the murder counts. Jury selection lasted more than three months, and during this time the court and the parties interviewed the prospective jurors and then called back a subset for general voir dire. As part of the jury selection process, more than 200 potential jurors completed a 77-question, 17-page questionnaire. Potential jurors were then questioned in court regarding their ability to follow the law. Jurors who were not dismissed for cause were called back in groups for voir dire, and the parties exercised their peremptory challenges. Each side was allowed 20 peremptories, and the prosecution used 18 of its allotment. It used seven peremptories to strike all of the African-Americans and Hispanics who were available for service. Ayala, who is Hispanic, raised Batson objections to those challenges. Ayala first objected after the prosecution peremptorily challenged two African-Americans, Olanders D. and Galileo S. The trial judge stated that these two strikes failed to establish a prima facie case of racial discrimination, but he nevertheless required the prosecution to reveal the reasons for the strikes. The prosecutor asked to do this outside the presence of the defense so as not to disclose trial strategy, and over Ayala’s objection, the judgegranted the request. The prosecution then offered several reasons for striking Olanders D., including uncertainty about his willingness to impose the death penalty. The prosecution stated that it dismissed Galileo S. primarily because he had been arrested numerous times and had not informed the court about all his prior arrests. After hearing and evaluating these explanations, the judge concluded that the prosecution had valid, race-neutral reasons for these strikes. Ayala again raised Batson objections when the prosecution used peremptory challenges to dismiss two Hispanics, Gerardo O. and Luis M. As before, the judge found that the defense had not made out a prima facie case, but ordered the prosecution to reveal the reasons for the strikes. This was again done ex parte, but this time the defense did not expressly object. The prosecution explained that it had challenged Gerardo O. and Luis M. in part because it was unsure that they could impose the death penalty. The prosecution also emphasized that Gerardo O.’s English proficiency was limited and that Luis M. had independently investigated the case. The trial court concluded a second time that the prosecution had legitimate race-neutral reasons for the strikes. Ayala raised Batson objections for a third and final time when the prosecution challenged Robert M., who was Hispanic; George S., whose ethnicity was disputed; and Barbara S., who was African-American. At this point, the trial court agreed that Ayala had made a prima facie Batson showing. Ayala’s counsel argued that the strikes were in fact based on race. Ayala’s counsel contended that the challenged jurors were “not significantly different from the white jurors that the prosecution ha[d] chosen to leave on the jury both in terms of their attitudes on the death penalty, their attitudes on the criminal justice system, and their attitudes on the presumption of innocence.” App. 306. Ayala’s counsel then reviewed the questionnaire answers and voir dire testimony of Barbara S. and Robert M., as well as the statements made by three of the prospective jurors who had been the subject of the prior Batson objections, Galileo S., Gerardo O., and Luis M. Counsel argued that their answers showed that they could impose the death penalty. The trial court stated that it would hear the prosecution’s response outside the presence of the jury, and Ayala once more did not object to that ruling. The prosecution then explained that it had dismissed the prospective jurors in question for several race-neutral reasons, including uncertainty that Robert M., George S., or Barbara S. would be open to imposing the death penalty. The prosecution also emphasized (among other points) that Robert M. had followed a controversial trial, that George S. had been a holdout on a prior jury, and that Barbara S. had given the impression during voir dire that she was under the influence of drugs. The trial court concluded, for a third time, that the prosecution’s peremptory challenges were based on race-neutral criteria. In August 1989, the jury convicted Ayala of all the charges except one of the three attempted robberies. With respect to the three murder convictions, the jury found two special circumstances: Ayala committed multiple murders, and he killed during the course of an attempted robbery. The jury returned a verdict of death on all three murder counts, and the trial court entered judgment consistent with that verdict. B Ayala appealed his conviction and sentence, and counsel was appointed to represent him in January 1993. Between 1993 and 1999, Ayala filed 20 applications for an extension of time, 11 of which requested additional time to file his opening brief. After the California Supreme Court eventually ruled that no further extensions would be granted, Ayala filed his opening brief in April 1998, nine years after he was convicted. The State filed its brief in September 1998, and Ayala then asked for four extensions of time to file his reply brief. After the court declared that it would grant him no further extensions, he filed his reply brief in May 1999. In August 2000, the California Supreme Court affirmed Ayala’s conviction and death sentence. People v. Ayala, 24 Cal. 4th 243, 6 P. 3d 193. In an opinion joined by five justices, the State Supreme Court rejected Ayala’s contention that the trial court committed reversible error by excluding the defense from part of the Batson hearing. The court understood Ayala to challenge the peremptory strikes under both Batson and its state-law analogue, People v. Wheeler, 22 Cal. 3d 258, 583 P. 2d 748 (1978). The court first concluded that the prosecution had not offered matters of trial strategy at the ex parte hearing and that, “as a matter of state law, it was [error]” to bar Ayala’s attorney from the hearing. 24 Cal. 4th, at 262, 6 P. 3d, at 203. Turning to the question of prejudice, the court stated: “We have concluded that error occurred under state law, and we have noted [the suggestion in United States v. Thompson, 827 F. 2d 1254 (CA9 1987),] that excluding the defense from a Wheeler-type hearing may amount to a denial of due process. We nonetheless conclude that the error was harmless under state law (People v. Watson (1956) 46 Cal.2d 818, 836), and that, if federal error occurred, it, too, was harmless beyond a reasonable doubt (Chapman v. California (1967) 386 U. S. 18 ) as a matter of federal law. On the record before us, we are confident that the challenged jurors were excluded for proper, race-neutral reasons.” Id., at 264, 6 P. 3d, at 204. The court then reviewed the prosecution’s reasons for striking the seven prospective jurors and found that “[o]n this well-developed record, . . . we are confident that defense counsel could not have argued anything substantial that would have changed the court’s rulings. Accordingly, the error was harmless.” Id., at 268, 6 P. 3d, at 207. The court concluded that the record supported the trial judge’s implicit determination that the prosecution’s justifications were not fabricated and were instead “grounded in fact.” Id., at 267, 6 P. 3d, at 206. And the court emphasized that the “trial court’s rulings in the ex parte hearing indisputably reflect both its familiarity with the record of voir dire of the challenged prospective jurors and its critical assessment of the prosecutor’s proffered justifications.” Id., at 266–267, 6 P. 3d, at 206. The California Supreme Court also rejected Ayala’s argument that his conviction should be vacated because most of the questionnaires filled out by prospective jurors who did not serve had been lost at some point during the decade that had passed since the end of the trial. The court wrote that “the record is sufficiently complete for us to be able to conclude that [the prospective jurors who were the subject of the contested peremptories] were not challenged and excused on the basis of forbidden group bias.” Id., at 270, 6 P. 3d, at 208. And even if the loss of the questionnaires was error under federal or state law, the court held, the error was harmless under Chapman and its state-law analogue. Two justices of the State Supreme Court dissented. We then denied certiorari. 532 U. S. 1029 (2001) . C After the California Supreme Court summarily denied a habeas petition, Ayala turned to federal court. He filed his initial federal habeas petition in 2002, but then went back to state court to exhaust several claims. In December 2004, he filed the operative federal petition and argued, among other things, that the ex parte hearings and loss of the questionnaires violated his rights under the Sixth, Eighth, and Fourteenth Amendments. In 2006, the District Court denied Ayala relief on those claims. The District Court read the decision of the California Supreme Court to mean that the state court had not decided whether the ex parte proceedings violated federal law, and the District Court expressed doubt “whether the trial court’s procedure was constitutionally defective as a matter of clearly established Federal law.” App. to Pet. for Cert. 145a. But even if such a violation occurred, the District Court held, the state court’s finding of harmlessness was not contrary to or an unreasonable application of clearly established law and thus could not be overturned under AEDPA. The District Court also rejected Ayala’s argument about the lost questionnaires, concluding that, even without them, the record was sufficient to resolve Ayala’s other claims. In 2013, a divided panel of the Ninth Circuit granted Ayala federal habeas corpus relief and required California either to release or retry him. Ayala v. Wong, 756 F. 3d 656 (2014). Because Ayala’s federal petition is subject to the requirements of AEDPA, the panel majority began its analysis by inquiring whether the state court had adjudicated Ayala’s claims on the merits. Applying de novo review,[1] the panel held that the ex parte proceedings violated the Federal Constitution, and that the loss of the questionnaires violated Ayala’s federal due process rights if that loss deprived him of “the ability to meaningfully appeal the denial of his Batson claim.” Id., at 671. The panel folded this inquiry into its analysis of the question whether the error regarding the ex parte proceedings was harmless. Turning to the question of harmlessness, the panel identified the applicable standard of review as that set out in Brecht and added: “We apply the Brecht test without regard for the state court’s harmlessness determination.” 756 F. 3d, at 674 (internal quotation marks omitted).[2] The panel used the following complicated formulation to express its understanding of Brecht’s application to Ayala’s claims: “If we cannot say that the exclusion of defense counsel with or without the loss of the questionnaires likely did not prevent Ayala from prevailing on his Batson claim, then we must grant the writ.” 756 F. 3d, at 676. Applying this test, the panel majority found that the error was not harmless, at least with respect to three of the seven prospective jurors. The panel asserted that the absence of Ayala and his counsel had interfered with the trial court’s ability to evaluate the prosecution’s proffered justifications for those strikes and had impeded appellate review, and that the loss of the questionnaires had compounded this impairment. Judge Callahan dissented. She explained that the California Supreme Court’s decision that any federal error was harmless constituted a merits adjudication of Ayala’s federal claims. She then reviewed the prosecution’s explanations for its contested peremptory challenges and concluded that federal habeas relief was barred because “fairminded jurists can concur in the California Supreme Court’s determination of harmless error.” Id., at 706. The Ninth Circuit denied rehearing en banc, but Judge Ikuta wrote a dissent from denial that was joined by seven other judges. Like Judge Callahan, Judge Ikuta concluded that the California Supreme Court adjudicated the merits of Ayala’s federal claims. Instead of the panel’s “de novo review of the record that piles speculation upon speculation,” she would have found that the state court’s harmlessness determination was not an unreasonable application of Chapman. 756 F. 3d, at 723. We granted certiorari. 574 U. S. ___ (2014). II Ayala contends that his federal constitutional rights were violated when the trial court heard the prosecution’s justifications for its strikes outside the presence of the defense, but we find it unnecessary to decide that question. We assume for the sake of argument that Ayala’s federal rights were violated, but that does not necessarily mean that he is entitled to habeas relief. In the absence of “the rare type of error” that requires automatic reversal, relief is appropriate only if the prosecution cannot demonstrate harmlessness. Glebe v. Frost, 574 U. S. ___, ___ (2014) (per curiam) (slip op., at 3). The Ninth Circuit did not hold—and Ayala does not now contend—that the error here falls into that narrow category, and therefore Ayala is entitled to relief only if the error was not harmless. The test for whether a federal constitutional error was harmless depends on the procedural posture of the case. On direct appeal, the harmlessness standard is the one prescribed in Chapman, 386 U. S. 18 : “[B]efore a federal constitutional error can be held harmless, the court must be able to declare a belief that it was harmless beyond a reasonable doubt.” Id., at 24. In a collateral proceeding, the test is different. For reasons of finality, comity, and federalism, habeas petitioners “are not entitled to habeas relief based on trial error unless they can establish that it resulted in ‘actual prejudice.’ ” Brecht, 507 U. S., at 637 (quoting United States v. Lane, 474 U. S. 438, 449 (1986) ). Under this test, relief is proper only if the federal court has “grave doubt about whether a trial error of federal law had ‘substantial and injurious effect or influence in determining the jury’s verdict.’ ” O’Neal v. McAninch, 513 U. S. 432, 436 (1995) . There must be more than a “reasonable possibility” that the error was harmful. Brecht, supra, at 637 (internal quotation marks omitted). The Brecht standard reflects the view that a “State is not to be put to th[e] arduous task [of retrying a defendant] based on mere speculation that the defendant was prejudiced by trial error; the court must find that the defendant was actually prejudiced by the error.” Calderon v. Coleman, 525 U. S. 141, 146 (1998) (per curiam). Because Ayala seeks federal habeas corpus relief, he must meet the Brecht standard, but that does not mean, as the Ninth Circuit thought, that a state court’s harmlessness determination has no significance under Brecht. In Fry v. Pliler, 551 U. S. 112, 120 (2007) , we held that the Brecht standard “subsumes” the requirements that §2254(d) imposes when a federal habeas petitioner contests a state court’s determination that a constitutional error was harmless under Chapman. The Fry Court did not hold—and would have had no possible basis for holding—that Brecht somehow abrogates the limitation on federal habeas relief that §2254(d) plainly sets out. While a federal habeas court need not “formal[ly]” apply both Brecht and “AEDPA/Chapman,” AEDPA nevertheless “sets forth a precondition to the grant of habeas relief.” Fry, supra, at 119–120. Under AEDPA, 28 U. S. C. §2254(d): “An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim— “(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; or “(2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” Section 2254(d) thus demands an inquiry into whether a prisoner’s “claim” has been “adjudicated on the merits” in state court; if it has, AEDPA’s highly deferential standards kick in. Harrington v. Richter, 562 U. S. 86, 103 (2011) . At issue here is Ayala’s claim that the ex parte portion of the Batson hearings violated the Federal Constitution. There is no dispute that the California Supreme Court held that any federal error was harmless beyond a reasonable doubt under Chapman, and this decision undoubtedly constitutes an adjudication of Ayala’s constitutional claim “on the merits.” See, e.g., Mitchell v. Esparza, 540 U. S. 12 –18 (2003) (per curiam). Accordingly, a federal habeas court cannot grant Ayala relief unless the state court’s rejection of his claim (1) was contrary to or involved an unreasonable application of clearly established federal law, or (2) was based on an unreasonable determination of the facts. Because the highly deferential AEDPA standard applies, we may not overturn the California Supreme Court’s decision unless that court applied Chapman “in an ‘objectively unreasonable’ manner.” Id., at 18 (quoting Lockyer v. Andrade, 538 U. S. 63, 75 (2003) ). When a Chapman decision is reviewed under AEDPA, “a federal court may not award habeas relief under §2254 unless the harmlessness determination itself was unreasonable.” Fry, supra, at 119 (emphasis in original). And a state-court decision is not unreasonable if “ ‘fairminded jurists could disagree’ on [its] correctness.” Richter, supra, at 101 (quoting Yarborough v. Alvarado, 541 U. S. 652, 664 (2004) ). Ayala therefore must show that the state court’s decision to reject his claim “was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” 562 U. S., at 103. In sum, a prisoner who seeks federal habeas corpus relief must satisfy Brecht, and if the state court adjudi-cated his claim on the merits, the Brecht test subsumes the limitations imposed by AEDPA. Fry, supra, at 119–120. III With this background in mind, we turn to the question whether Ayala was harmed by the trial court’s decision to receive the prosecution’s explanation for its challenged strikes without the defense present. In order for this argument to succeed, Ayala must show that he was actually prejudiced by this procedure, a standard that he neces-sarily cannot satisfy if a fairminded jurist could agree with the California Supreme Court’s decision that this procedure met the Chapman standard of harmlessness. Evaluation of these questions requires consideration of the trial court’s grounds for rejecting Ayala’s Batson challenges. A Batson held that the Equal Protection Clause of the Fourteenth Amendment prohibits prosecutors from exercising peremptory challenges on the basis of race. 476 U. S., at 89. When adjudicating a Batson claim, trial courts follow a three-step process: “First, a defendant must make a prima facie showing that a peremptory challenge has been exercised on the basis of race; second, if that showing has been made, the prosecution must offer a race-neutral basis for striking the juror in question; and third, in light of the parties’ submissions, the trial court must determine whether the defendant has shown purposeful discrimination.” Snyder v. Louisiana, 552 U. S. 472 –477 (2008) (internal quotation marks and alterations omitted). The opponent of the strike bears the burden of persuasion regarding racial motivation, Purkett v. Elem, 514 U. S. 765, 768 (1995) (per curiam), and a trial court finding regarding the credibility of an attorney’s explanation of the ground for a peremptory challenge is “entitled to ‘great deference,’ ” Felkner v. Jackson, 562 U. S. 594, 598 (2011) (per curiam) (quoting Batson, 476 U. S., at 98, n. 21). On direct appeal, those findings may be reversed only if the trial judge is shown to have committed clear error. Rice v. Collins, 546 U. S. 333, 338 (2006) . Under AEDPA, even more must be shown. A federal habeas court must accept a state-court finding unless it was based on “an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” §2254(d)(2). “State-court factual findings, moreover, are presumed correct; the petitioner has the burden of rebutting the presumption by ‘clear and convincing evidence.’ ” Collins, supra, at 338–339 (quoting §2254(e)(1)). In this case, Ayala challenged seven of the prosecution’s peremptory challenges. As explained above, the Ninth Circuit granted relief based on the dismissal of three potential jurors. The dissent discusses only one, Olanders D. We will devote most of our analysis to the three individuals discussed by the Ninth Circuit, but we hold that any error was harmless with respect to all seven strikes. B 1 Ayala first contests the prosecution’s decision to challenge Olanders D., an African-American man. The prosecution stated that its “primary” reason for striking Olanders D. was uncertainty about whether he could impose the death penalty, and the prosecutor noted that Olanders D. had written on his questionnaire that he did not “believe in the death penalty.” 50 Reporter’s Tr. on Appeal 6185 (hereinafter Tr.). Providing additional reasons for this strike, the prosecutor first stated that Olanders D.’s responses “did not make a lot of sense,” “were not thought out,” and “demonstrate[d] a lack of ability to express himself well.” App. 283. The prosecutor also voiced doubt that Olanders D. “could actively participate in a meaningful way in deliberations with other jurors” and might have lacked the “ability to fit in with a cohesive group of 12 people.” Ibid. The trial court concluded that the strike was race-neutral. The judge stated: “Certainly with reference to whether or not he would get along with 12 people, it may well be that he would get along very well with 12 people. I think the other observations of counsel are accurate and borne out by the record.” 50 Tr. 6186. The California Supreme Court found that the evidence of Olanders D.’s views on the death penalty provided adequate support for the trial judge’s finding that the strike exercised against him was not based on race, and the court further found that defense counsel’s presence would not have affected the outcome of the Batson hearing. The Ninth Circuit reversed, but its decision rested on a misapplication of the applicable harmless-error standards. 2 As the trial court and the State Supreme Court found, Olanders D.’s voir dire responses amply support the prosecution’s concern that he might not have been willing to impose the death penalty. During voir dire, Olanders D. acknowledged that he wrote on his questionnaire, “ ‘I don’t believe in the death penalty,’ ” App. 179, and he agreed that he had at one time “thought that [the death penalty] was completely wrong,” id., at 177. Although he stated during the voir dire that he had reconsidered his views, it was reasonable for the prosecution and the trial court to find that he did not clearly or adequately explain the reason or reasons for this change. When asked about this, Olanders D. gave a vague and rambling reply: “Well, I think it’s—one thing would be the—the—I mean, examining it more closely, I think, and becoming more familiar with the laws and the—and the behavior, I mean, the change in the people, I think. All of those things contributed to the changes.” Id., at 178. The Ninth Circuit reversed because it speculated that defense counsel, if present when the prosecution explained the basis for this strike, “could have pointed to seated white jurors who had expressed similar or greater hesitancy” in imposing the death penalty. 756 F. 3d, at 678. The Ninth Circuit wrote that a seated white juror named Ana L. was “indistinguishable from Olanders D. in this regard” and that she had “made almost precisely the same statement in her questionnaire.” Ibid. The responses of Olanders D. and Ana L., however, were by no means “indistinguishable.” Olanders D. initially voiced unequivocal opposition to the death penalty, stating flatly: “I don’t believe in the death penalty.” He also revealed that he had once thought it was “completely wrong.” Ana L., by contrast, wrote on the questionnaire that she “probably would not be able to vote for the death penalty,” App. 109 (emphasis added), and she then later said at voir dire that she could vote for a verdict of death. In a capital case, it is not surprising for prospective jurors to express varying degrees of hesitancy about voting for a death verdict. Few are likely to have experienced a need to make a comparable decision at any prior time in their lives. As a result, both the prosecution and the defense may be required to make fine judgment calls about which jurors are more or less willing to vote for the ultimate punishment. These judgment calls may involve a comparison of responses that differ in only nuanced respects, as well as a sensitive assessment of jurors’ demeanor. We have previously recognized that peremptory challenges “are often the subjects of instinct,” Miller-El v. Dretke, 545 U. S. 231, 252 (2005) (citing Batson, 476 U. S., at 106 (Marshall, J., concurring)), and that “race-neutral reasons for peremptory challenges often invoke a juror’s demeanor,” Snyder, 552 U. S., at 477. A trial court is best situated to evaluate both the words and the demeanor of jurors who are peremptorily challenged, as well as the credibility of the prosecutor who exercised those strikes. As we have said, “these determinations of credibility and demeanor lie peculiarly within a trial judge’s province,” and “in the absence of exceptional circumstances, we [will] defer to the trial court.” Ibid. (alterations and internal quotation marks omitted). “Appellate judges cannot on the basis of a cold record easily second-guess a trial judge’s decision about likely motivation.” Collins, 546 U. S., at 343 (Breyer, J., concurring). The upshot is that even if “[r]easonable minds reviewing the record might disagree about the prosecutor’s credibility, . . . on habeas review that does not suffice to supersede the trial court’s credibility determination.” Id., at 341–342 (majority opinion). Here, any similarity between the responses of Olanders D. and Ana L. is insufficient to compel an inference of racial discrimination under Brecht or AEDPA. Ayala contends that the presence of defense counsel might have made a difference because defense counsel might have been able to identify white jurors who were not stricken by the prosecution even though they had “expressed similar or greater hesitancy” about the death penalty. We see no basis for this argument. The questionnaires of all the jurors who sat and all the alternates are in the record, and Ana L., whom we just discussed, is apparently the white juror whose answers come the closest to those of Olanders D. Since neither Ayala nor the Ninth Circuit identified a white juror whose statements better support their argument, there is no reason to think that defense counsel could have pointed to a superior comparator at the ex parte proceeding. 3 In rejecting the argument that the prosecutor peremptorily challenged Olanders D. because of his race, the California Supreme Court appears to have interpreted the prosecutor’s explanation of this strike to mean that Olanders D.’s views on the death penalty were alone sufficient to convince him to exercise a strike, see 24 Cal. 4th, at 266, 6 P. 3d, at 206, and this was certainly an interpretation of the record that must be sustained under 28 U. S. C. §2254(d)(2). As a result, it is not necessary for us to consider the prosecutor’s supplementary reason for this strike—the poor quality of Olanders D.’s responses—but in any event, the Ninth Circuit’s evaluation of this reason is also flawed. The Ninth Circuit wrote that its independent “review of the voir dire transcript reveal[ed] nothing that supports the prosecution’s claim: Olanders D.’s answers were responsive and complete.” 756 F. 3d, at 679. The record, however, provides sufficient support for the trial court’s determination. Olanders D.’s incoherent explanation during voir dire of the reasons for his change of opinion about the death penalty was quoted above. He also provided a chronology of the evolution of his views on the subject that did not hold together. He stated that he had been “completely against the death sentence” 10 years earlier but seemed to suggest that his views had changed over the course of the intervening decade. See App. 176–177. However, on the questionnaire, which he had completed just a month before the voir dire, he wrote unequivocally: “I don’t believe in the death penalty.” Id., at 179. And then, at the time of the voir dire, he said that he would be willing to impose the death penalty in some cases. Id., at 180. He explained his answer on the questionnaire as follows: “I answered that kind of fast[.] [N]ormally, I wouldn’t answer that question that way, but I mean, I really went through that kind of fast. I should have done better than that.” Id., at 179–180. These answers during voir dire provide more than sufficient support for the prosecutor’s observation, which the trial court implicitly credited, that Olanders D.’s statements “did not make a lot of sense,” “were not thought out,” and “demonstrate[d] a lack of ability to express himself well.” In ordering federal habeas relief based on their assessment of the responsiveness and completeness of Olanders D.’s answers, the members of the panel majority misunderstood the role of a federal court in a habeas case. The role of a federal habeas court is to “ ‘guard against extreme malfunctions in the state criminal justice systems,’ ” Richter, 562 U. S., at 102–103 (quoting Jackson v. Virginia, 443 U. S. 307 , n. 5 (1979) (Stevens, J., concurring in judgment)), not to apply de novo review of factual findings and to substitute its own opinions for the determination made on the scene by the trial judge. C Ayala next challenges the prosecution’s use of a peremptory challenge to strike Gerardo O., a Hispanic man. The prosecution offered three reasons for this strike: Gerardo O. had a poor grasp of English; his answers during voir dire and on his questionnaire suggested that he might not be willing to impose the death penalty; and he did not appear to get along with the other prospective jurors. The trial judge accepted this explanation, as did the State Supreme Court. The Ninth Circuit, however, rejected the state courts’ determinations based on speculation that defense counsel, if present at the in camera hearing, “likely could have called into question all of the prosecution’s stated reasons for striking Gerardo O.” 756 F. 3d, at 680. The Ninth Circuit thought that it could grant Ayala relief simply because it “[could not] say that Ayala would not have shown that the trial court would or should have determined that the prosecution’s strike of Gerardo O. violated Batson.” Id., at 682. But that is not the test. The inquiry under Brecht is not whether the federal habeas court could definitively say that there were no winning arguments that the defense could have made. Instead, the evidence in the record must raise “grave doubt[s]” about whether the trial judge would have ruled differently. O’Neal, 513 U. S., at 436. This requires much more than a “reasonable possibility” that the result of the hearing would have been different. Brecht, 507 U. S., at 637 (internal quotation marks omitted). And on the record in this case, Ayala cannot establish actual prejudice or that no fairminded jurist could agree with the state court’s application of Chapman. We begin with the prosecution’s explanation that it challenged Gerardo O. because of his limited English proficiency. During voir dire, Gerardo O. acknowledged that someone else had written the answers for him onhis questionnaire “[b]ecause I couldn’t—I cannot read—I cannot spell that well.” App. 163. He added that he “didn’t get” some of the words on the questionnaire. Ibid. Gerardo O.’s testimony also revealed that he might well have been unable to follow what was said at trial. When asked whether he could understand spoken English, he responded: “It depends if you make long words. If you make—if you go—if you say it straight out, then I might understand. If you beat around the bush, I won’t.” Id., at 166. At that point, defense counsel and Gerardo O. engaged in a colloquy that suggests that defense counsel recognized that he lacked the ability to understand words not used in basic everyday speech, “legal words,” and rapid speech in English: “Q: I’ll try not to talk—use any legal words or lawyer talk— “A: Okay. “Q: —and talk regular with you. If you don’t understand anything I say, stop me and tell me, okay? “A: Okay. “Q: If you’re selected as a juror during the trial, and you know you’re serving as a juror and listening to witnesses, can we have your promise that if a witness uses a word you don’t understand, you’ll put your hand up and let us know? “A: Yeah. . . . . . “Q: There’s one more problem that you’re going to have with me, and that is that sometimes . . . I talk real fast . . . .” Id., at 166–167. It is understandable for a prosecutor to strike a potential juror who might have difficulty understanding English.[3] The jurors who were ultimately selected heard many days of testimony, and the instructions at both the guilt and the penalty phases included “legal words” and words not common in everyday speech. The prosecution had an obvious reason to worry that service on this jury would have strained Gerardo O.’s linguistic capability. The Ninth Circuit reached a contrary conclusion by distorting the record and the applicable law. The Ninth Circuit first suggested that Gerardo O.’s English-language deficiencies were limited to reading and writing, 756 F. 3d, at 680, but as the portions of the voir dire quoted above make clear, that was not true; the record shows that his ability to understand spoken English was also limited. The Ninth Circuit then suggested that “[t]he prosecution’s purported reason for striking Gerardo O. . . . was directly related to his status as someone who spoke Spanish as his first language,” ibid., but the prosecutor voiced no concern about Gerardo O.’s ability to speak Spanish or about the fact that Spanish was his first language. The prosecution’s objection concerned Gerardo O.’s limited proficiency in English. The Ninth Circuit quoted the following statement from Hernandez v. New York, 500 U. S. 352, 363 (1991) (plurality opinion): “ ‘[T]he prosecutor’s frank admission that his ground for excusing th[is] juror[ ] related to [his] ability to speak and understand Spanish raised a plausible, though not a necessary, inference that language might be a pretext for what in fact [was a] race-based peremptory challenge[ ].’ ” 756 F. 3d, at 680 (alterations in original). This statement, however, did not concern a peremptory exercised due to a prospective juror’s lack of English proficiency. Instead, it concerned the dismissal of Spanish-speaking members of the venire for fear that, if seated, they might not follow the English translation of testimony given in Spanish. See 500 U. S., at 360. The Ninth Circuit’s decision regarding Gerardo O. was thus based on a misreading of the record and a distortion of our case law. And neither Ayala nor the Ninth Circuit has identified anything that defense counsel might have done at the ex parte hearing to show that the prosecutor’s concern about Gerardo O.’s limited English proficiency was pretextual. The prosecution’s second proffered reason for striking Gerardo O. was concern about his willingness to impose the death penalty, and as the trial court found, this observation was also supported by the record. Indeed, when asked in voir dire how he felt about imposing the death penalty, Gerardo O. responded that he was “[k]ind of shaky about it. . . . I’m not too sure if I can take someone else’s life in my hands and say that; say, you know, ‘death,’ or something.” App. 168. In response to another question about his thoughts on the death penalty, he replied: “I don’t know yet. It’s kind of hard, you know, to pick it up like that and say how I feel about the death penalty.” 15 Tr. 1052. Answering a question about whether his thoughts on the death penalty would affect how he viewed the evidence presented at trial, he responded, “I don’t know, sir, to tell you the truth.” App. 165. And when asked if he had “any feeling that [he] would be unable to vote for the death penalty if [he] thought it was a case that called for it,” Gerardo O. responded once again, “I don’t know.” 15 Tr. 1043. While Gerardo O. did say at one point that he might be willing to impose the death pen-alty, he qualified that statement by adding that he would be comforted by the fact that “there’s eleven more other persons on the jury.” App. 170. What we said above regarding jurors who express doubts about their openness to a death verdict applies as well here. The prosecution’s reluctance to take a chance that Gerardo O. would ultimately be willing to consider the death penalty in accordance with state law did not compel the trial judge to find that the strike of Gerardo O. was based on race. Nor is there a basis for finding that the absence of defense counsel affected the trial judge’s evaluation of the sincerity of this proffered ground for the strike. Defense counsel had a full opportunity during voir dire to create a record regarding Gerardo O.’s openness to the death penalty. And defense counsel had the opportunity prior to the ex parte proceeding on the Gerardo O. strike to compare the minority jurors dismissed by the prosecution with white jurors who were seated. Counsel argued that the answers on the death penalty given by the minority jurors were “not significantly different from [those of] the white jurors that the prosecution ha[d] chosen to leave on the jury.” Id., at 306. The trial judge asked counsel for “particulars,” and counsel discussed Gerardo O., albeit briefly. Id., at 307–308. Thus, there is no reason to believe that counsel could have made a more persuasive argumentat the ex parte proceeding than he made during thisexchange. The prosecution’s final reason for striking Gerardo O. was that he appeared to be “a standoffish type of individ-ual” whose “dress and . . . mannerisms . . . were not in keeping with the other jurors” and who “did not appear to be socializing or mixing with any of the other jurors.” Id., at 298. The trial judge did not dispute that the prosecution’s reflections were borne out by the record. The California Supreme Court affirmed and also emphasized that “the trial court’s rulings in the ex parte hearing indisput-ably reflect both its familiarity with the record of voir dire of the challenged prospective jurors and its critical assessment of the prosecutor’s proffered justifications.” 24 Cal. 4th, at 266–267, 6 P. 3d, at 206. In light of the strength of the prosecution’s first two reasons for striking Gerardo O., it is not at all clear that the prosecution proffered this final reason as an essential factor in its decision to strike, but in any event, there is no support for the suggestion that Ayala’s attorney, if allowed to attend the ex parte hearing, would have been able to convince the judge that this reason was pretextual. The Ninth Circuit, however, was content to speculate about what might have been. Mixing guesswork with armchair sociology, the Ninth Circuit mused that “[i]t is likely that Gerardo O.’s dress and mannerisms were distinctly Hispanic. Perhaps in the late 1980’s Hispanic males in San Diego County were more likely than members of other racial or ethnic groups in the area to wear a particular style or color of shirt, and Gerardo O. was wearing such a shirt.” 756 F. 3d, at 680–681. As for the prosecution’s observation that Gerardo O. did not socialize with other jurors, the Ninth Circuit posited that, “perhaps, unbeknownst to the trial judge, Gerardo O. did ‘socializ[e] or mix[ ]’ with a number of other jurors, and had even organized a dinner for some of them at his favorite Mexican restaurant.” Id., at 681. This is not how habeas review is supposed to work. The record provides no basis for the Ninth Circuit’s flight of fancy. Brecht requires more than speculation about what extrarecord information defense counsel might have mentioned. And speculation of that type is not enough to show that a State Supreme Court’s rejection of the argument regarding Gerardo O. was unreasonable. D The final prospective juror specifically discussed in the Ninth Circuit’s decision was Robert M., who is Hispanic. The prosecution’s primary proffered reason for striking Robert M. was concern that he would not impose the death penalty, though the prosecution added that it was troubled that he had followed the Sagon Penn case, a high-profile prosecution in San Diego in which an alleged murderer was acquitted amid allegations of misconduct by police and prosecutors. In addition, the prosecution also explained to the trial court that Robert M. scored poorly on its 10-point scale for evaluating prospective jurors. The trial court accepted the prosecutor’s explanation of the strike. With respect to the prosecution’s concern that Robert M. might not be willing to impose the death penalty, the Ninth Circuit found that defense counsel, if permitted to attend the in camera proceeding, could have compared Robert M.’s statements about the death penalty to those of other jurors and could have reminded the judge that Robert M. had “repeatedly stated during voir dire that he believed in the death penalty and could personally vote to impose it.” 756 F. 3d, at 682. But as with Olanders D. and Gerardo O., we cannot say that the prosecution had no basis for doubting Robert M.’s willingness to impose the death penalty. For example, when asked at one point whether he could vote for death, Robert M. responded: “Well, I’ve though[t] about that, but it’s a difficult question, and yeah, it is difficult for me to say, you know, one way or the other. I believe in it, but for me to be involved in it is—is hard. It’s hard to accept that aspect of it, do you know what I mean?” App. 149–150. In response to another question, he said: “It would be hard, but I think I could, yes. It’s—it’s hard to say, you know—and I don’t care who the person is—to say that they have to put somebody away, you know. It’s very hard.” Id., at 154. These are hardly answers that would inspire confidence in the minds of prosecutors in a capital case. While the Ninth Circuit argued that defense counsel’s absence at the in camera hearing prejudiced the trial judge’s ability to assess this reason for the strike of Robert M., the Ninth Circuit failed to mention that defense counsel specifically addressed this issue during voir dire. At that time, he pointedly reminded the judge that Robert M. had made several statements during voir dire that were favorable to the death penalty. Id., at 307. The trial judge thus heard defense counsel’s arguments but nevertheless concluded that the record supplied a basis for a legitimate concern about whether Robert M. could impose the death penalty. That Ayala’s attorney did not have the opportunity to repeat this same argument once more at the in camera proceeding does not create grave doubt about whether the trial court would have decided the issue differently. As for the prosecution’s second proffered reason for striking Robert M.—that he had followed the Sagon Penn case[4]—the Ninth Circuit placed great emphasis on the fact that a seated white juror had followed a different murder trial, that of Robert Alton Harris.[5] But the Penn and Harris cases were quite different. Harris was convicted while Penn was acquitted; and since the Harris case was much older, the experience of following it was less likely to have an effect at the time of the trial in this case. E Ayala raised a Batson objection about the prosecution’s use of peremptory challenges on four additional jurors, George S., Barbara S., Galileo S., and Luis M. The Ninth Circuit did not address these prospective jurors at length, and we need not dwell long on them. With respect to all four of these prospective jurors, we conclude that any constitutional error was harmless. Of these four additional jurors, Ayala’s brief in this Court develops an argument with respect to only two, George S. and Barbara S. And while Ayala’s attorney claimed that George S. was Hispanic, the prosecutor said that he thought that George S. was Greek. In any event, the prosecution offered several reasons for striking George S. The prosecutor noted that one of his responses “was essentially, ‘you probably don’t want me to be a juror on this case.’ ” Id., at 312. The prosecutor was also concerned about whether he would vote for death or even a life sentence and whether he would follow the law as opposed to his personal religious beliefs. In addition, the prosecutor noted that George S. had previously been the sole holdout on a jury and that his prior application to be a police officer had been rejected, for reasons that were not clear. The trial court accepted these explanations. Ayala contests only two of these justifications. He quibbles that George S. had not been a “ ‘holdout,’ ” but instead had been the dissenting juror in a civil case on which unanimity was not required. This observation does not render the prosecution’s proffered justification “false or pretextual.” Brief for Respondent 46. The fact that George S. had been willing to dissent from a jury verdict could reasonably give a prosecutor pause in a capital case since a single holdout juror could prevent a guilty verdict or death sentence. The most that Ayala can establish is that reasonable minds can disagree about whether the prosecution’s fears were well founded, but this does not come close to establishing “actual prejudice” under Brecht. Nor does it meet the AEDPA standard. Ayala also points out that a seated white juror, Charles C., had been re-jected by a police force, but George S. admitted that he had applied to law enforcement because he was “trying to get out of the Army,” App. 222, and the reasons for his rejection were not clear. Charles C., by contrast, had received a qualifying score on a law enforcement exam but was not hired because a position was not available. As for Barbara S., the prosecution struck her because, during voir dire, she appeared to be “under the influence of drugs” and disconnected from the proceedings. Id., at 314. The prosecution emphasized that she had “an empty look in her eyes, slow responses, a lack of really being totally in tune with what was going on.” Ibid. It added that she appeared “somewhat angry,” “manifest[ed] a great deal of nervousness,” and seemed like someone who would be unlikely to closely follow the trial. Ibid. The trial judge thought that Barbara S. appeared nervous rather than hostile, but he agreed that she gave incomplete answers that were sometimes “non sequiturs.” Id., at 315. He concluded, “I certainly cannot quarrel . . . with your subjective impression, and the use of your peremp-tory challenge based upon her individual manifestation, as opposed to her ethnicity.” Ibid. Ayala points to the trial court’s disagreement with the prosecutor’s impression that Barbara S. was hostile, but this ruling illustrates the trial judge’s recollection of the demeanor of the prospective jurors and his careful evaluation of each of the prosecutor’s proffered reasons for strikes. And the fact that the trial judge’s impression of Barbara S.’s demeanor was somewhat different from the prosecutor’s hardly shows that the prosecutor’s reasons were pretextual. It is not at all unusual for individuals to come to different conclusions in attempting to read another person’s attitude or mood. IV The pattern of peremptory challenges in this case was sufficient to raise suspicions about the prosecution’s motives and to call for the prosecution to explain its strikes. As we have held, the Fourteenth Amendment prohibits a prosecutor from striking potential jurors based on race. Discrimination in the jury selection process undermines our criminal justice system and poisons public confidence in the evenhanded administration of justice. In Batson, this Court adopted a procedure for ferreting out discrimination in the exercise of peremptory challenges, and this procedure places great responsibility in thehands of the trial judge, who is in the best position to determine whether a peremptory challenge is based on an impermissible factor. This is a difficult determination because of the nature of peremptory challenges: They are often based on subtle impressions and intangible factors. In this case, the conscientious trial judge determined that the strikes at issue were not based on race, and his judgment was entitled to great weight. On appeal, five justices of the California Supreme Court carefully evaluated the record and found no basis to reverse. A Federal District Judge denied federal habeas relief, but a divided panel of the Ninth Circuit reversed the District Court and found that the California Supreme Court had rendered a decision with which no fairminded jurist could agree. For the reasons explained above, it was the Ninth Circuit that erred. The exclusion of Ayala’s attorney from part of the Batson hearing was harmless error. There is no basis for finding that Ayala suffered actual prejudice, and the decision of the California Supreme Court represented an entirely reasonable application of controlling precedent. * * * The judgment of the Court of Appeals for the Ninth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The panel decided this question de novo because it concluded that the California Supreme Court either did not decide whether the ex parte proceedings violated the Federal Constitution or silently decided that question in Ayala’s favor. 756 F. 3d, at 666–670. 2 In a footnote, however, the panel stated: “In holding that Ayala has demonstrated his entitlement to relief under Brecht, we therefore also hold to be an unreasonable application of Chapman the California Supreme Court’s conclusion that Ayala was not prejudiced by the exclusion of the defense.” Id., at 674, n. 13. 3 The California Supreme Court has held that “[i]nsufficient command of the English language to allow full understanding of the words employed in instructions and full participation in deliberations clearly . . . render[s] a juror ‘unable to perform his duty’ ” within the meaning of the California Penal Code. People v. Lomax, 49 Cal. 4th 530, 566, 234 P. 3d 377, 407 (2010) (citation omitted). See also Cal. Code Ann. 4 See Man Acquitted of Killing Officer, N. Y. Times, July 17, 1987, p. B8. 5 See People v. Harris, 28 Cal. 3d 935, 623 P. 2d 240 (1981).

574.US.383

In 2002, Congress enacted the Homeland Security Act, 116Stat. 2135. That Act provides that the Transportation Security Administration (TSA) “shall prescribe regulations prohibiting the disclosure of information . . . if the Under Secretary decides that disclosur[e] would . . . be detrimental to the security of transportation.” 49 U. S. C. §114(r)(1)(C). Around the same time, the TSA promulgated regulations prohibiting the unauthorized disclosure of “sensitive security information,” 67 Fed. Reg. 8351, which included “[s]pecific details of aviation security measures . . . [such as] information concerning specific numbers of Federal Air Marshals, deployments or missions, and the methods involved in such operations,” 49 CFR §1520.7(j). In July 2003, the TSA briefed all federal air marshals—including Robert J. MacLean—about a potential plot to hijack passenger flights. A few days after the briefing, MacLean received from the TSA a text message cancelling all overnight missions from Las Vegas until early August. MacLean, who was stationed in Las Vegas, believed that cancelling those missions during a hijacking alert was dangerous and illegal. He therefore contacted a reporter and told him about the TSA’s decision to cancel the missions. After discovering that MacLean was the source of the disclosure, the TSA fired him for disclosing sensitive security information without authorization. MacLean challenged his firing before the Merit Systems Protection Board. He argued that his disclosure was whistleblowing activity under 5 U. S. C. §2302(b)(8)(A), which protects employees who disclose information that reveals “any violation of any law, rule, or regulation,” or “a substantial and specific danger to public health or safety.” The Board held that MacLean did not qualify for protection under that statute because his disclosure was “specifically prohibited by law,” §2302(b)(8)(A)—namely, by 49 U. S. C. §114(r)(1). The Court of Appeals for the Federal Circuit vacated the Board’s decision, holding that Section 114(r)(1) was not a prohibition. Held: MacLean’s disclosure was not “specifically prohibited by law.” Pp. 5–16. (a) The Government argues that MacLean’s disclosure was “specifically prohibited by law” in two ways: first, by the TSA’s regulations on sensitive security information, and second, by Section 114(r)(1) itself, which authorized the TSA to promulgate those regulations. Pp. 5–14. (i) MacLean’s disclosure was not prohibited by the TSA’s regulations for purposes of Section 2302(b)(8)(A) because regulations do not qualify as “law” under that statute. Throughout Section 2302, Congress repeatedly used the phrase “law, rule, or regulation.” But Congress did not use that phrase in the statutory language at issue here; it used the word “law” standing alone. Congress’s choice to say “specifically prohibited by law,” instead of “specifically prohibited by law, rule, or regulation” suggests that Congress meant to exclude rules and regulations. In addition, Section 2302(b)(8)(A) creates a second exception for disclosures “required by Executive order to be kept secret in the interest of national defense or the conduct of foreign affairs.” That the second exception is limited to actions by the President himself suggests that the first exception does not include action taken by executive agencies. Finally, interpreting the word “law” to include rules and regulations could defeat the purpose of the whistleblower statute. That interpretation would allow an agency to insulate itself from Section 2302(b)(8)(A) simply by promulgating a regulation that “specifically prohibited” all whistleblowing. The Government proposes two alternative interpretations, but neither is persuasive. First, the Government argues that the word “law” includes all regulations that have the “force and effect of law.” The Government bases this argument on the decision in Chrysler Corp. v. Brown, 441 U.S. 281, where this Court held that legislative regulations generally fall within the meaning of the word “law” unless there is a “clear showing of contrary legislative intent.” Id., at 295–296. But Congress’s use of the word “law,” in close connection with the phrase “law, rule, or regulation,” provides the necessary “clear showing” that “law” does not include regulations in this case. Second, the Government argues that the word “law” includes at least those regulations that were “promulgated pursuant to an express congressional directive.” The Government, however, was unable to find a single example of the word “law” being used in that way. Pp. 6–11. (ii) Likewise, MacLean’s disclosure was not prohibited by Section 114(r)(1). That statute does not prohibit anything; instead, it authorizes the TSA to “prescribe regulations.” Thus, by its terms, Section 114(r)(1) did not prohibit the disclosure here. The Government responds that Section 114(r)(1) did prohibit MacLean’s disclosure by imposing a “legislative mandate” on the TSA to promulgate regulations to that effect. But the statute affords substantial discretion to the TSA in deciding whether to prohibit any particular disclosure. Thus, it is the TSA’s regulations—not the statute—that prohibited MacLean’s disclosure, and those regulations do not qualify as “law” under Section 2302(b)(8)(A). Pp. 11–14. (b) The Government argues that providing whistleblower protection to individuals like MacLean would “gravely endanger public safety” by making the confidentiality of sensitive security information depend on the idiosyncratic judgment of each of the TSA’s 60,000 employees. Those concerns are legitimate, but they must be addressed by Congress or the President, rather than by this Court. Pp. 14–15. 714 F. 3d. 1301, affirmed. Roberts, C. J., delivered the opinion of the Court, in which Scalia, Thomas, Ginsburg, Breyer, Alito, and Kagan, JJ., joined. Sotomayor, J., filed a dissenting opinion, in which Kennedy, J., joined.

Federal law generally provides whistleblower protections to an employee who discloses information revealing “any violation of any law, rule, or regulation,” or “a substantial and specific danger to public health or safety.”5 U. S. C. §2302(b)(8)(A). An exception exists, however, for disclosures that are “specifically prohibited by law.” Ibid. Here, a federal air marshal publicly disclosed that the Transportation Security Administration (TSA) had de-cided to cut costs by removing air marshals from certain long-distance flights. The question presented is whether that disclosure was “specifically prohibited by law.”IA In 2002, Congress enacted the Homeland Security Act,116Stat.2135. As relevant here, that Act provides that the TSA “shall prescribe regulations prohibiting the disclosure of information obtained or developed in carrying out security . . . if the Under Secretary decides that disclosing the information would . . . be detrimental to the security of transportation.”49 U. S. C. §114(r)(1)(C). Around the same time, the TSA promulgated regulations prohibiting the unauthorized disclosure of what it called “sensitive security information.” See 67 Fed. Reg. 8351 (2002). The regulations described 18 categories of sensitive security information, including “[s]pecific details of aviation security measures . . . [such as] information concerning specific numbers of Federal Air Marshals, deployments or missions, and the methods involved in such operations.” 49 CFR §1520.7(j) (2002). Sensitive security information is not classified, so the TSA can share it with individuals who do not have a security clearance, such as airport employees. Compare Exec. Order 13526, §4.1, 3 CFR 298, 314–315 (2009 Comp.), with 49 CFR §1520.11(c) (2013).B Robert J. MacLean became a federal air marshal for the TSA in 2001. In that role, MacLean was assigned to protect passenger flights from potential hijackings. See49 U. S. C. §44917(a). On July 26, 2003, the Department of Homeland Security (DHS) issued a confidential advisory about a potential hijacking plot. The advisory said that members of the terrorist group al Qaeda were planning to attack passenger flights, and that they “considered suicide hijackings and bombings as the most promising methods to destroy aircraft in flight, as well as to strike ground targets.” App. 16. The advisory identified a number of potential targets, including the United Kingdom, Italy, Australia, and the east coast of the United States. Finally, the advisory warned that at least one of the attacks “could be executed by the end of the summer 2003.” Ibid. The TSA soon summoned all air marshals (including MacLean) for face-to-face briefings about the hijacking plot. During MacLean’s briefing, a TSA official told him that the hijackers were planning to “smuggle weapons in camera equipment or children’s toys through foreign security,” and then “fly into the United States . . . into an airport that didn’t require them to be screened.” Id., at 92. The hijackers would then board U. S. flights, “overpower the crew or the Air Marshals and . . . fly the planes into East Coast targets.” Id., at 93. A few days after the briefing, MacLean received from the TSA a text message cancelling all overnight missions from Las Vegas until early August. MacLean, who was stationed in Las Vegas, believed that cancelling those missions during a hijacking alert was dangerous. He also believed that the cancellations were illegal, given that federal law required the TSA to put an air marshal on every flight that “present[s] high security risks,”49 U. S. C. §44917(a)(2), and provided that “nonstop, long distance flights, such as those targeted on September 11, 2001, should be a priority,” §44917(b). See App. 95, 99, 101. MacLean therefore asked a supervisor why the TSA had canceled the missions. The supervisor responded that the TSA wanted “to save money on hotel costs because there was no more money in the budget.” Id., at 95. MacLean also called the DHS Inspector General’s Office to report the cancellations. But a special agent in that office told him there was “nothing that could be done.” Id., at 97. Unwilling to accept those responses, MacLean contacted an MSNBC reporter and told him about the canceled missions. In turn, the reporter published a story about the TSA’s decision, titled “Air Marshals pulled from key flights.” Id., at 36. The story reported that air marshals would “no longer be covering cross-country or international flights” because the agency did not want them “to incur the expense of staying overnight in hotels.” Ibid. The story also reported that the cancellations were “particularly disturbing to some” because they “coincide[d] with anew high-level hijacking threat issued by the Department of Homeland Security.” Id., at 37. After MSNBC published the story, several Members of Congress criticized the cancellations. Within 24 hours, the TSA reversed its decision and put air marshals back on the flights. Id., at 50. At first, the TSA did not know that MacLean was the source of the disclosure. In September 2004, however, MacLean appeared on NBC Nightly News to criticize the TSA’s dress code for air marshals, which he believed made them too easy to identify. Although MacLean appeared in disguise, several co-workers recognized his voice, and the TSA began investigating the appearance. During that investigation, MacLean admitted that he had disclosed the text message back in 2003. Consequently, in April 2006, the TSA fired MacLean for disclosing sensitive security information without authorization. MacLean challenged his firing before the Merit Systems Protection Board, arguing in relevant part that his disclosure was protected whistleblowing activity under5 U. S. C. §2302(b)(8)(A). The Board held that MacLean did not qualify for protection under that statute, however, because his disclosure was “specifically prohibited by law.” 116 MSPR 562, 569–572 (2011). The Court of Appeals for the Federal Circuit vacated the Board’s decision. 714 F. 3d 1301 (2013). The parties had agreed that, in order for MacLean’s disclosure to be “specifically prohibited by law,” it must have been “prohibited by a statute rather than by a regulation.” Id., at 1308 (emphasis added). Thus, the issue before the court was whether the statute authorizing the TSA’s regulations—now codified at49 U. S. C. §114(r)(1)—“specifically prohibited” MacLean’s disclosure. 714 F. 3d, at 1308.[1]* The court first held that Section 114(r)(1) was not a prohibition. The statute did “not expressly prohibit employee disclosures,” the court explained, but instead empowered the TSA to “prescribe regulations prohibiting disclosure[s]” if the TSA decided that disclosing the information would harm public safety. Id., at 1309. The court therefore concluded that MacLean’s disclosure was prohibited by a regulation, which the parties had agreed could not be a “law” under Section 2302(b)(8)(A). Ibid. The court then held that, even if Section 114(r)(1) were a prohibition, it was not “sufficiently specific.” Ibid. The court explained that a law is sufficiently specific only if it “requires that matters be withheld from the public as to leave no discretion on the issue, or . . . establishes particular criteria for withholding or refers to particular types of matters to be withheld.” Ibid. (quoting S. Rep. No. 95–969 (1978)). And Section 114(r)(1) did not meet that test because it “provide[d] only general criteria for withholding information and [gave] some discretion to the [TSA] to fashion regulations for prohibiting disclosure.” 714 F. 3d, at 1309. The court accordingly vacated the Board’s decision and remanded for a determination of whether MacLean’s disclosure met the other requirements under Section 2302(b)(8)(A). Id., at 1310–1311. We granted certiorari. 572 U. S. ___ (2014).II Section 2302(b)(8) provides, in relevant part, that a federal agency may not take“a personnel action with respect to any employee or applicant for employment because of“(A) any disclosure of information by an employee or applicant which the employee or applicant reasonably believes evidences“(i) any violation of any law, rule, or regulation, or“(ii) gross mismanagement, a gross waste of funds, an abuse of authority, or a substantial and specific danger to public health or safety,“if such disclosure is not specifically prohibited by law and if such information is not specifically required by Executive order to be kept secret in the interest of national defense or the conduct of foreign affairs.” The Government argues that this whistleblower statute does not protect MacLean because his disclosure regarding the canceled missions was “specifically prohibited by law” in two ways. First, the Government argues that the disclosure was specifically prohibited by the TSA’s regulations on sensitive security information: 49 CFR §§1520.5(a)–(b), 1520.7(j) (2003). Second, the Government argues that the disclosure was specifically prohibited by49 U. S. C. §114(r)(1), which authorized the TSA to pro-mulgate those regulations. We address each argument in turn.A1 In 2003, the TSA’s regulations prohibited the disclosure of “[s]pecific details of aviation security measures . . . [such as] information concerning specific numbers of Federal Air Marshals, deployments or missions, and the methods involved in such operations.” 49 CFR §1520.7(j). MacLean does not dispute before this Court that the TSA’s regulations prohibited his disclosure regarding the canceled missions. Thus, the question here is whether a disclosure that is specifically prohibited by regulation is also “specifically prohibited by law” under Section 2302(b)(8)(A). (Emphasis added.) The answer is no. Throughout Section 2302, Congress repeatedly used the phrase “law, rule, or regulation.” For example, Section 2302(b)(1)(E) prohibits a federal agency from discriminating against an employee “on the basis of marital status or political affiliation, as prohibited under any law, rule, or regulation.” For another example, Section 2302(b)(6) prohibits an agency from “grant[ing] any preference or advantage not authorized by law, rule, or regulation.” And for a third example, Section 2302(b)(9)(A) prohibits an agency from retaliating against an employee for “the exercise of any appeal, complaint, or grievance right granted by any law, rule, or regulation.” In contrast, Congress did not use the phrase “law, rule, or regulation” in the statutory language at issue here; it used the word “law” standing alone. That is significant because Congress generally acts intentionally when it uses particular language in one section of a statute but omits it in another. Russello v. United States,464 U. S. 16,23 (1983). Thus, Congress’s choice to say “specifically prohibited by law” rather than “specifically prohibited by law, rule, or regulation” suggests that Congress meant to exclude rules and regulations. The interpretive canon that Congress acts intentionally when it omits language included elsewhere applies with particular force here for two reasons. First, Congress used “law” and “law, rule, or regulation” in close proximity—indeed, in the same sentence. §2302(b)(8)(A) (protecting the disclosure of “any violation of any law, rule, or regulation . . . if such disclosure is not specifically prohibited by law”). Second, Congress used the broader phrase “law, rule, or regulation” repeatedly—nine times in Section 2302 alone. See §§2302(a)(2)(D)(i), (b)(1)(E), (b)(6), (b)(8)(A)(i), (b)(8)(B)(i), (b)(9)(A), (b)(12), (b)(13), (d)(5). Those two aspects of the whistleblower statute make Con-gress’s choice to use the narrower word “law” seem quite deliberate. We drew the same inference in Department of Treasury, IRS v. FLRA,494 U. S. 922 (1990). There, the Government argued that the word “laws” in one section of the Civil Service Reform Act of 1978 meant the same thing as the phrase “law, rule, or regulation” in another section of the Act. Id., at 931. We rejected that argument as “sim-ply contrary to any reasonable interpretation of the text.” Id., at 932. Indeed, we held that a statute that referred to “laws” in one section and “law, rule, or regulation” in another “cannot, unless we abandon all pretense at precise communication, be deemed to mean the same thing in both places.” Ibid. That inference is even more compelling here, because the statute refers to “law” and “law, rule, or regulation” in the same sentence, rather than several sections apart. Another part of the statutory text points the same way. After creating an exception for disclosures “specifically prohibited by law,” Section 2302(b)(8)(A) goes on to create a second exception for information “specifically required by Executive order to be kept secret in the interest of national defense or the conduct of foreign affairs.” This exception is limited to action taken directly by the President. That suggests that the word “law” in the only other exception is limited to actions by Congress—after all, it would be unusual for the first exception to include action taken by executive agencies, when the second exception requires action by the President himself. In addition, a broad interpretation of the word “law” could defeat the purpose of the whistleblower statute. If “law” included agency rules and regulations, then an agency could insulate itself from the scope of Section 2302(b)(8)(A) merely by promulgating a regulation that “specifically prohibited” whistleblowing. But Congress passed the whistleblower statute precisely because it did not trust agencies to regulate whistleblowers within their ranks. Thus, it is unlikely that Congress meant to include rules and regulations within the word “law.”2 The Government admits that some regulations fall outside the word “law” as used in Section 2302(b)(8)(A). But, the Government says, that does not mean that all regulations are excluded. The Government suggests two interpretations that would distinguish “law” from “law, rule, or regulation,” but would still allow the word “law”to subsume the TSA’s regulations on sensitive security information. First, the Government argues that the word “law” includes all regulations that have the “force and effect of law” (i.e., legislative regulations), while excluding those that do not (e.g., interpretive rules). Brief for Petitioner 19–22. The Government bases this argument on our decision in Chrysler Corp. v. Brown,441 U. S. 281 (1979). There, we held that legislative regulations generally fall within the meaning of the word “law,” and that it would take a “clear showing of contrary legislative intent” before we concluded otherwise. Id., at 295–296. Thus, because the TSA’s regulations have the force and effect of law, the Government says that they should qualify as “law” under the statute. The Government’s description of Chrysler is accurate enough. But Congress’s use of the word “law,” in close connection with the phrase “law, rule, or regulation,” provides the necessary “clear showing” that “law” does not include regulations. Indeed, using “law” and “law, rule, or regulation” in the same sentence would be a very obscure way of drawing the Government’s nuanced distinction between different types of regulations. Had Congress wanted to draw that distinction, there were far easier and clearer ways to do so. For example, at the time Congress passed Section 2302(b)(8)(A), another federal statute defined the words “regulatory order” to include a “rule or regulation, if it has the force and effect of law.”7 U. S. C. §450c(a) (1976 ed.). Likewise, another federal statute defined the words “State law” to include “all laws, decisions, rules, regulations, or other State action having the effect of law.”29 U. S. C. §1144(c)(1) (1976 ed.). As those examples show, Congress knew how to distinguish between regulations that had the force and effect of law and those that did not, but chose not to do so in Section 2302(b)(8)(A). Second, the Government argues that the word “law” includes at least those regulations that were “promulgated pursuant to an express congressional directive.” Brief for Petitioner 21. Outside of this case, however, the Government was unable to find a single example of the word “law” being used in that way. Not a single dictionary definition, not a single statute, not a single case. The Government’s interpretation happens to fit this case precisely, but it needs more than that to recommend it. Although the Government argues here that the word “law” includes rules and regulations, it definitively re-jected that argument in the Court of Appeals. For example, the Government’s brief accepted that the word “law”meant “legislative enactment,” and said that the “only dispute” was whether49 U. S. C. §114(r)(1) “serve[d] as that legislative enactment.” Brief for Respondent in No. 11–3231 (CA Fed.), pp. 46–47. Then, at oral argument, a judge asked the Government’s attorney the following question: “I thought I understood your brief to concede that [the word “law”] can’t be a rule or regulation, it means statute. Am I wrong?” The Government’s attorney responded: “You’re not wrong your honor. I’ll be as clear as I can. ‘Specifically prohibited by law’ here means statute.” Oral Arg. Audio in No. 11–3231, at 22:42–23:03; see also id., at 29:57–30:03 (“Now, as we’ve been discussing here, we’re not saying here that [the word “law”] needs to encompass regulations. We’re saying statute.”). Those concessions reinforce our conclusion that the Government’s proposed interpretations are unpersuasive. In sum, when Congress used the phrase “specifically prohibited by law” instead of “specifically prohibited by law, rule, or regulation,” it meant to exclude rules and regulations. We therefore hold that the TSA’s regulations do not qualify as “law” for purposes of Section 2302(b)(8)(A).B We next consider whether MacLean’s disclosure regarding the canceled missions was “specifically prohibited” by49 U. S. C. §114(r)(1) itself. As relevant here, that statute provides that the TSA “shall prescribe regulations prohibiting the disclosure of information obtained or developed in carrying out security . . . if the Under Secretary decides that disclosing the information would . . . be detrimental to the security of transportation.” §114(r)(1)(C). This statute does not prohibit anything. On the con-trary, it authorizes something—it authorizes the Under Secretary to “prescribe regulations.” Thus, by its terms Section 114(r)(1) did not prohibit the disclosure at issue here. The Government responds that Section 114(r)(1) did prohibit MacLean’s disclosure by imposing a “legislative mandate” on the TSA to promulgate regulations to that effect. See Brief for Petitioner 28, 33; see also post, at 2–3 (Sotomayor, J., dissenting). But the Government pushes the statute too far. Section 114(r)(1) says that the TSA shall prohibit disclosures only “if the Under Secretary decides that disclosing the information would . . . be detrimental to the security of transportation.” §114(r)(1)(C) (emphasis added). That language affords substantial discretion to the TSA in deciding whether to prohibit any particular disclosure. The dissent tries to downplay the scope of that discretion, viewing it as the almost ministerial task of “identifying whether a particular piece of information falls within the scope of Congress’ command.” Post, at 3. But determining which documents meet the statutory standard of “detrimental to the security of transportation” requires the exercise of considerable judgment. For example, the Government says that Section 114(r)(1) requires the Under Secretary to prohibit disclosures like MacLean’s. The Government also says, however, that the statute does not require the Under Secretary to prohibit an employee from disclosing that “federal air marshals will be absent from important flights, but declining to specify which flights.” Reply Brief 23. That fine-grained distinction comes not from Section 114(r)(1) itself, but from the Under Secretary’s exercise of discretion. It is the TSA’s regulations—not the statute—that prohibited MacLean’s disclosure. And as the dissent agrees, a regulation does not count as “law” under the whistleblower statute. See post, at 1. The Government insists, however, that this grant of discretion does not make Section 114(r)(1) any less of a prohibition. In support, the Government relies on Administrator, FAA v. Robertson,422 U. S. 255 (1975). That case involved the Freedom of Information Act (FOIA), which requires federal agencies to disclose information upon request unless, among other things, the information is “specifically exempted from disclosure by statute.”5 U. S. C. §552(b)(3). In Robertson, we held that the Federal Aviation Act of 1958 was one such statute, because it gave the Federal Aviation Administration (FAA) “a broad degree of discretion” in deciding whether to disclose or withhold information. 422 U. S., at 266. The Government tries to analogize that case to this one. In Robertson, the Government says, the FAA’s discretion whether to disclose information did not preclude a finding that the information was “specifically exempted” from disclosure by statute. So too here, the Government says, the TSA’s discretion whether to prohibit disclosure of information does not preclude a finding that the information is “specifically prohibited” from disclosure by Section 114(r)(1). See Brief for Petitioner 30. This analogy fails. FOIA and Section 2302(b)(8)(A) differ in an important way: The provision of FOIA at issue involves information that is “exempted” from disclosure, while Section 2302(b)(8)(A) involves information that is “prohibited” from disclosure. A statute that exempts information from mandatory disclosure may nonetheless give the agency discretion to release that exempt information to the public. In such a case, the agency’s exercise of discretion has no effect on whether the information is “exempted from disclosure by statute”—it remains exempt whatever the agency chooses to do. The situation is different when it comes to a statute giving an agency discretion to prohibit the disclosure of information. The information is not prohibited from disclosure by statute regardless of what the agency does. Itis the agency’s exercise of discretion that determines whether there is a prohibition at all. Thus, when Section 114(r)(1) gave the TSA the discretion to prohibit the disclosure of information, the statute did not create a prohibition—it gave the TSA the power to create one. And because Section 114(r)(1) did not create a prohibition, MacLean’s disclosure was not “prohibited by law” under Section 2302(b)(8)(A), but only by a regulation issued in the TSA’s discretion. In any event, Robertson was a case about FOIA, not Section 2302, and our analysis there depended on two FOIA-specific factors that are not present here. First, we examined the legislative history of FOIA and determined that Congress did not intend that statute to affect laws like the Federal Aviation Act. 422 U. S., at 263–265. In particular, we noted that the Civil Aeronautics Board had expressed its view during congressional hearings that the Federal Aviation Act qualified as an exempting statute under FOIA, and that “no question was raised or challenge made” to the agency’s view. Id., at 264–265. But that legislative history can have no effect on our analysis of Section 2302(b)(8)(A). Second, we said that the Federal Aviation Act could fail to qualify as an exempting statute only if we read FOIA “as repealing by implication all existing statutes which restrict public access to specific Government records.” Id., at 265 (internal quotation marks omitted). Then, relying on the presumption that “repeals by implication are disfavored,” we rejected that interpretation of FOIA. But the presumption against implied repeals has no relevance here. Saying that Section 114(r)(1) is not a prohibition under the whistleblower statute is not the same as saying that the whistleblower statute implicitly repealed Section 114(r)(1). On the contrary, Section 114(r)(1) remains in force by allowing the TSA to deny FOIA requests and prohibit employee disclosures that do not qualify for whistleblower protection under Section 2302(b)(8)(A). Ultimately, FOIA and Section 2302(b)(8)(A) are different statutes—they have different language, different histories, and were enacted in different contexts. Our interpretation of one, therefore, has no impact whatsoever on our interpretation of the other.III Finally, the Government warns that providing whistleblower protection to individuals like MacLean would “gravely endanger public safety.” Brief for Petitioner 38. That protection, the Government argues, would make the confidentiality of sensitive security information depend on the idiosyncratic judgment of each of the TSA’s 60,000 employees. Id., at 37. And those employees will “most likely lack access to all of the information that led the TSA to make particular security decisions.” Id., at 38. Thus, the Government says, we should conclude that Congress did not intend for Section 2302(b)(8)(A) to cover disclosures like MacLean’s. Those concerns are legitimate. But they are concerns that must be addressed by Congress or the President, rather than by this Court. Congress could, for example, amend Section 114(r)(1) so that the TSA’s prohibitions on disclosure override the whistleblower protections in Section 2302(b)(8)(A)—just as those prohibitions currently override FOIA. See §114(r)(1) (authorizing the TSA to prohibit disclosures “[n]otwithstanding section 552 of title 5”); see also10 U. S. C. §2640(h) (“the Secretary of Defense may (notwithstanding any other provision of law) withhold from public disclosure safety-related information that is provided to the Secretary voluntarily by an air carrier for the purposes of this section”). Congress could also exempt the TSA from the requirements of Section 2302(b)(8)(A) entirely, as Congress has already done for the Federal Bureau of Investigation, the Central Intelligence Agency, the Defense Intelligence Agency, the National Geospatial-Intelligence Agency, the National Security Agency, the Office of the Director of National Intelligence, and the National Reconnaissance Office. See5 U. S. C. §2302(a)(2)(C)(ii)(I). Likewise, the President could prohibit the disclosure of sensitive security information by Executive order. Indeed, the Government suggested at oral argument that the President could “entirely duplicate” the regulations that the TSA has issued under Section 114(r)(1). Tr. of Oral Arg. 16–20. Such an action would undoubtedly create an exception to the whistleblower protections found in Section 2302(b)(8)(A). Although Congress and the President each has the power to address the Government’s concerns, neither has done so. It is not our role to do so for them. The judgment of the United States Court of Appeals for the Federal Circuit isAffirmed.Notes1* This statute has a complicated history. It was codified at 49 U. S. C.§40119(b)(1) when the TSA initially promulgated its regulations on sensitive security information. It was codified at §114(s)(1) when MacLean disclosed the text message to MSNBC. And it is now codified at §114(r)(1). The Federal Circuit referred to §40119(b)(1) in its opinion. Because the statute has remained identical in all relevant respects, however, we and the parties refer to the current version.

575.US.2014_13-983

After his wife left him, petitioner Anthony Douglas Elonis, under the pseudonym “Tone Dougie,” used the social networking Web siteFacebook to post self-styled rap lyrics containing graphically violent language and imagery concerning his wife, co-workers, a kindergarten class, and state and federal law enforcement. These posts were often interspersed with disclaimers that the lyrics were “fictitious” and not intended to depict real persons, and with statements that Elonis was exercising his First Amendment rights. Many who knew him saw his posts as threatening, however, including his boss, who fired him for threatening co-workers, and his wife, who sought and was granted a state court protection-from-abuse order against him. When Elonis’s former employer informed the Federal Bureau of Investigation of the posts, the agency began monitoring Elonis’s Face-book activity and eventually arrested him. He was charged with five counts of violating 18 U. S. C. §875(c), which makes it a federal crime to transmit in interstate commerce “any communication containing any threat . . . to injure the person of another.” At trial, Elonis requested a jury instruction that the Government was required to prove that he intended to communicate a “true threat.” Instead, the District Court told the jury that Elonis could be found guilty if a reasonable person would foresee that his statements would be interpreted as a threat. Elonis was convicted on four of the five counts and renewed his jury instruction challenge on appeal. The Third Circuit affirmed, holding that Section 875(c) requires only the intent to communicate words that the defendant understands, and that a reasonable person would view as a threat. Held: The Third Circuit’s instruction, requiring only negligence with respect to the communication of a threat, is not sufficient to support a conviction under Section 875(c). Pp. 7–17. (a) Section 875(c) does not indicate whether the defendant must intend that the communication contain a threat, and the parties can show no indication of a particular mental state requirement in the statute’s text. Elonis claims that the word “threat,” by definition, conveys the intent to inflict harm. But common definitions of “threat” speak to what the statement conveys—not to the author’s mental state. The Government argues that the express “intent to extort” requirements in neighboring Sections 875(b) and (d) should preclude courts from implying an unexpressed “intent to threaten” requirement in Section 875(c). The most that can be concluded from such a comparison, however, is that Congress did not mean to confine Section 875(c) to crimes of extortion, not that it meant to exclude a mental state requirement. Pp. 7–9. (b) The Court does not regard “mere omission from a criminal enactment of any mention of criminal intent” as dispensing with such a requirement. Morissette v. United States, 342 U. S. 246 . This rule of construction reflects the basic principle that “wrongdoing must be conscious to be criminal,” and that a defendant must be “blameworthy in mind” before he can be found guilty. Id., at 252. The “general rule” is that a guilty mind is “a necessary element in the indictment and proof of every crime.” United States v. Balint, 258 U. S. 250 . Thus, criminal statutes are generally interpreted “to include broadly applicable scienter requirements, even where the statute . . . does not contain them.” United States v. X-Citement Video, Inc., 513 U. S. 64 . This does not mean that a defendant must know that his conduct is illegal, but a defendant must have knowledge of “the facts that make his conduct fit the definition of the offense.” Staples v. United States, 511 U. S. 600 , n. 3. Federal criminal statutes that are silent on the required mental state should be read to include “only that mens rea which is necessary to separate” wrongful from innocent conduct. Carter v. United States, 530 U. S. 255 . In some cases, a general requirement that a defendant act knowingly is sufficient, but where such a requirement “would fail to protect the innocent actor,” the statute “would need to be read to require . . . specific intent.” Ibid. Pp. 9–13. (c) The “presumption in favor of a scienter requirement should apply to each of the statutory elements that criminalize otherwise innocent conduct.” X-Citement Video, 513 U. S., at 72. In the context of Section 875(c), that requires proof that a communication was transmitted and that it contained a threat. And because “the crucial element separating legal innocence from wrongful conduct,” id., at 73, is the threatening nature of the communication, the mental state requirement must apply to the fact that the communication contains a threat. Elonis’s conviction was premised solely on how his posts would be viewed by a reasonable person, a standard feature of civil liability in tort law inconsistent with the conventional criminal conduct requirement of “awareness of some wrongdoing,” Staples, 511 U. S., at 606–607. This Court “ha[s] long been reluctant to infer that a negligence standard was intended in criminal statutes.” Rogers v. United States, 422 U. S. 35 (Marshall, J., concurring). And the Government fails to show that the instructions in this case required more than a mental state of negligence. Hamling v. United States, 418 U. S. 87 , distinguished. Section 875(c)’s mental state requirement is satisfied if the defendant transmits a communication for the purpose of issuing a threat or with knowledge that the communication will be viewed as a threat. The Court declines to address whether a mental state of recklessness would also suffice. Given the disposition here, it is unnecessary to consider any First Amendment issues. Pp. 13–17. 730 F. 3d. 321, reversed and remanded. Roberts, C. J., delivered the opinion of the Court, in which Scalia, Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in part and dissenting in part. Thomas, J., filed a dissenting opinion.

Federal law makes it a crime to transmit in interstate commerce “any communication containing any threat . . . to injure the person of another.” 18 U. S. C. §875(c). Petitioner was convicted of violating this provision under instructions that required the jury to find that he communicated what a reasonable person would regard as a threat. The question is whether the statute also requires that the defendant be aware of the threatening nature of the communication, and—if not—whether the First Amendment requires such a showing. I A Anthony Douglas Elonis was an active user of the social networking Web site Facebook. Users of that Web site may post items on their Facebook page that are accessible to other users, including Facebook “friends” who are notified when new content is posted. In May 2010, Elonis’s wife of nearly seven years left him, taking with her their two young children. Elonis began “listening to more violent music” and posting self-styled “rap” lyrics inspired by the music. App. 204, 226. Eventually, Elonis changed the user name on his Facebook page from his actual name to a rap-style nom de plume, “Tone Dougie,” to distinguish himself from his “on-line persona.” Id., at 249, 265. The lyrics Elonis posted as “Tone Dougie” included graphically violent language and imagery. This material was often interspersed with disclaimers that the lyrics were “fictitious,” with no intentional “resemblance to real persons.” Id., at 331, 329. Elonis posted an explanation to another Facebook user that “I’m doing this for me. My writing is therapeutic.” Id., at 329; see also id., at 205 (testifying that it “helps me to deal with the pain”). Elonis’s co-workers and friends viewed the posts in a different light. Around Halloween of 2010, Elonis posted a photograph of himself and a co-worker at a “Halloween Haunt” event at the amusement park where they worked. In the photograph, Elonis was holding a toy knife against his co-worker’s neck, and in the caption Elonis wrote, “I wish.” Id., at 340. Elonis was not Facebook friends with the co-worker and did not “tag” her, a Facebook feature that would have alerted her to the posting. Id., at 175; Brief for Petitioner 6, 9. But the chief of park security was a Facebook “friend” of Elonis, saw the photograph, and fired him. App. 114–116; Brief for Petitioner 9. In response, Elonis posted a new entry on his Facebook page: “Moles! Didn’t I tell y’all I had several? Y’all sayin’ I had access to keys for all the f***in’ gates. That I have sinister plans for all my friends and must have taken home a couple. Y’all think it’s too dark andfoggy to secure your facility from a man as mad as me? You see, even without a paycheck, I’m still the main attraction. Whoever thought the Halloween Haunt could be so f***in’ scary?” App. 332. This post became the basis for Count One of Elonis’ssubsequent indictment, threatening park patrons and employees. Elonis’s posts frequently included crude, degrading, and violent material about his soon-to-be ex-wife. Shortly after he was fired, Elonis posted an adaptation of a satirical sketch that he and his wife had watched together. Id., at 164–165, 207. In the actual sketch, called “It’s Illegal to Say . . . ,” a comedian explains that it is illegal for a person to say he wishes to kill the President, but not illegal to explain that it is illegal for him to say that. When Elonis posted the script of the sketch, however, he substituted his wife for the President. The posting was part of the basis for Count Two of the indictment, threatening his wife: “Hi, I’m Tone Elonis. Did you know that it’s illegal for me to say I want to kill my wife? . . . It’s one of the only sentences that I’m not allowed to say. . . . Now it was okay for me to say it right then because I was just telling you that it’s illegal for me to say I want to kill my wife. . . . Um, but what’s interesting is that it’s very illegal to say I really, really think someone out there should kill my wife. . . . But not illegal to say with a mortar launcher. Because that’s its own sentence. . . . I also found out that it’s incredibly illegal, extremely illegal to go on Facebook and say something like the best place to fire a mortar launcher at her house would be from the cornfield behind it because of easy access to a getaway road and you’d have a clear line of sight through the sun room. . . . Yet even more illegal to show an illustrated diagram. [diagram of the house]. . . .” Id., at 333. The details about the home were accurate. Id., at 154. At the bottom of the post, Elonis included a link to the video of the original skit, and wrote, “Art is about pushing limits. I’m willing to go to jail for my Constitutional rights. Are you?” Id., at 333. After viewing some of Elonis’s posts, his wife felt “extremely afraid for [her] life.” Id., at 156. A state court granted her a three-year protection-from-abuse order against Elonis (essentially, a restraining order). Id., at 148–150. Elonis referred to the order in another post on his “Tone Dougie” page, also included in Count Two of the indictment: “Fold up your [protection-from-abuse order] and put it in your pocket Is it thick enough to stop a bullet? Try to enforce an Order that was improperly granted in the first place Me thinks the Judge needs an education on true threat jurisprudence And prison time’ll add zeros to my settlement . . . And if worse comes to worse I’ve got enough explosives to take care of the State Police and the Sheriff’s Department.” Id., at 334. At the bottom of this post was a link to the Wikipedia article on “Freedom of speech.” Ibid. Elonis’s reference to the police was the basis for Count Three of his indictment, threatening law enforcement officers. That same month, interspersed with posts about a movie Elonis liked and observations on a comedian’s social commentary, id., at 356–358, Elonis posted an entry that gave rise to Count Four of his indictment: “That’s it, I’ve had about enough I’m checking out and making a name for myself Enough elementary schools in a ten mile radius to initiate the most heinous school shooting ever imagined And hell hath no fury like a crazy man in a Kindergarten class The only question is . . . which one?” Id., at 335. Meanwhile, park security had informed both local police and the Federal Bureau of Investigation about Elonis’s posts, and FBI Agent Denise Stevens had created a Facebook account to monitor his online activity. Id., at 49–51, 125. After the post about a school shooting, Agent Stevens and her partner visited Elonis at his house. Id., at 65–66. Following their visit, during which Elonis was polite but uncooperative, Elonis posted another entry on his Facebook page, called “Little Agent Lady,” which led to Count Five: “You know your s***’s ridiculous when you have the FBI knockin’ at yo’ door Little Agent lady stood so close Took all the strength I had not to turn the b**** ghost Pull my knife, flick my wrist, and slit her throat Leave her bleedin’ from her jugular in the arms of her partner [laughter] So the next time you knock, you best be serving a warrant And bring yo’ SWAT and an explosives expert while you’re at it Cause little did y’all know, I was strapped wit’ a bomb Why do you think it took me so long to get dressed with no shoes on? I was jus’ waitin’ for y’all to handcuff me and pat me down Touch the detonator in my pocket and we’re all goin’ [BOOM!] Are all the pieces comin’ together? S***, I’m just a crazy sociopath that gets off playin’ you stupid f***s like a fiddle And if y’all didn’t hear, I’m gonna be famous Cause I’m just an aspiring rapper who likes theattention who happens to be under investigation for terrorism cause y’all think I’m ready to turn the Valley into Fallujah But I ain’t gonna tell you which bridge is gonna fall into which river or road And if you really believe this s*** I’ll have some bridge rubble to sell you tomorrow [BOOM!][BOOM!][BOOM!]” Id., at 336. B A grand jury indicted Elonis for making threats to injure patrons and employees of the park, his estranged wife, police officers, a kindergarten class, and an FBI agent, all in violation of 18 U. S. C. §875(c). App. 14–17. In the District Court, Elonis moved to dismiss the indictment for failing to allege that he had intended to threaten anyone. The District Court denied the motion, holding that Third Circuit precedent required only that Elonis “intentionally made the communication, not that he intended to make a threat.” App. to Pet. for Cert. 51a. At trial, Elonis testified that his posts emulated the rap lyrics of the well-known performer Eminem, some of which involve fantasies about killing his ex-wife. App. 225. In Elonis’s view, he had posted “nothing . . . that hasn’t been said already.” Id., at 205. The Government presented as witnesses Elonis’s wife and co-workers, all of whom said they felt afraid and viewed Elonis’s posts as serious threats. See, e.g., id., at 153, 158. Elonis requested a jury instruction that “the government must prove that he intended to communicate a true threat.” Id., at 21. See also id., at 267–269, 303. The District Court denied that request. The jury instructions instead informed the jury that “A statement is a true threat when a defendant intentionally makes a statement in a context or under such circumstances wherein a reasonable person would foresee that the statement would be interpreted by those to whom the maker communicates the statement as a serious expression of an intention to inflict bodily injury or take the life of an individual.” Id.,at 301. The Government’s closing argument emphasized that it was irrelevant whether Elonis intended the postings to be threats—“it doesn’t matter what he thinks.” Id., at 286. A jury convicted Elonis on four of the five counts against him, acquitting only on the charge of threatening park patrons and employees. Id., at 309. Elonis was sentenced to three years, eight months’ imprisonment and three years’ supervised release. Elonis renewed his challenge to the jury instructions in the Court of Appeals, contending that the jury should have been required to find that he intended his posts to be threats. The Court of Appeals disagreed, holding that the intent required by Section 875(c) is only the intent to communicate words that the defendant understands, and that a reasonable person would view as a threat. 730 F. 3d 321, 332 (CA3 2013). We granted certiorari. 573 U. S. ___ (2014). II A An individual who “transmits in interstate or foreign commerce any communication containing any threat to kidnap any person or any threat to injure the person of another” is guilty of a felony and faces up to five years’ imprisonment. 18 U. S. C. §875(c). This statute requires that a communication be transmitted and that the communication contain a threat. It does not specify that the defendant must have any mental state with respect to these elements. In particular, it does not indicate whether the defendant must intend that his communication contain a threat. Elonis argues that the word “threat” itself in Section 875(c) imposes such a requirement. According to Elonis, every definition of “threat” or “threaten” conveys the notion of an intent to inflict harm. Brief for Petitioner 23. See United States v. Jeffries, 692 F. 3d 473, 483 (CA6 2012) (Sutton, J., dubitante). E.g., 11 Oxford English Dictionary 353 (1933) (“to declare (usually conditionally) one’s intention of inflicting injury upon”); Webster’s New International Dictionary 2633 (2d ed. 1954) (“Law, specif., an expression of an intention to inflict loss or harm on another by illegal means”); Black’s Law Dictionary 1519 (8th ed. 2004) (“A communicated intent to inflict harm or loss on another”). These definitions, however, speak to what the statement conveys—not to the mental state of the author. For example, an anonymous letter that says “I’m going to kill you” is “an expression of an intention to inflict loss or harm” regardless of the author’s intent. A victim who receives that letter in the mail has received a threat, even if the author believes (wrongly) that his message will be taken as a joke. For its part, the Government argues that Section 875(c) should be read in light of its neighboring provisions, Sections 875(b) and 875(d). Those provisions also prohibit certain types of threats, but expressly include a mental state requirement of an “intent to extort.” See 18 U. S. C. §875(b) (proscribing threats to injure or kidnap made “with intent to extort”); §875(d) (proscribing threats to property or reputation made “with intent to extort”). According to the Government, the express “intent to extort” requirements in Sections 875(b) and (d) should preclude courts from implying an unexpressed “intent to threaten” requirement in Section 875(c). See Russello v. United States, 464 U. S. 16, 23 (1983) (“[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.”). The Government takes this expressio unius est exclusio alterius canon too far. The fact that Congress excluded the requirement of an “intent to extort” from Section 875(c) is strong evidence that Congress did not mean to confine Section 875(c) to crimes of extortion. But that does not suggest that Congress, at the same time, also meant to exclude a requirement that a defendant act with a certain mental state in communicating a threat. The most we can conclude from the language of Section 875(c) and its neighboring provisions is that Congress meant to proscribe a broad class of threats in Section 875(c), but did not iden-tify what mental state, if any, a defendant must have to be convicted. In sum, neither Elonis nor the Government has identified any indication of a particular mental state requirement in the text of Section 875(c). B The fact that the statute does not specify any required mental state, however, does not mean that none exists. We have repeatedly held that “mere omission from a criminal enactment of any mention of criminal intent” should not be read “as dispensing with it.” Morissette v. United States, 342 U. S. 246, 250 (1952) . This rule of construction reflects the basic principle that “wrongdoing must be conscious to be criminal.” Id., at 252. As Justice Jackson explained, this principle is “as universal and persistent in mature systems of law as belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil.” Id., at 250. The “central thought” is that a defendant must be “blameworthy in mind” before he can be found guilty, a concept courts have expressed over time through various terms such as mens rea, scienter, malice aforethought, guilty knowledge, and the like. Id., at 252; 1 W. LaFave, Substantive Criminal Law §5.1, pp. 332–333 (2d ed. 2003). Although there are exceptions, the “general rule” is that a guilty mind is “a necessary element in the indictment and proof of every crime.” United States v. Balint, 258 U. S. 250, 251 (1922) . We therefore generally “interpret[ ] criminal statutes to include broadly applicable scienter requirements, even where the statute by its terms does not contain them.” United States v. X-Citement Video, Inc., 513 U. S. 64, 70 (1994) . This is not to say that a defendant must know that his conduct is illegal before he may be found guilty. The familiar maxim “ignorance of the law is no excuse” typi-cally holds true. Instead, our cases have explained that a defendant generally must “know the facts that make his conduct fit the definition of the offense,” Staples v. United States, 511 U. S. 600 , n. 3 (1994), even if he does not know that those facts give rise to a crime. Morissette, for example, involved an individual who had taken spent shell casings from a Government bombing range, believing them to have been abandoned. Dur-ing his trial for “knowingly convert[ing]” property of the United States, the judge instructed the jury that the only question was whether the defendant had knowingly taken the property without authorization. 342 U. S., at 248–249. This Court reversed the defendant’s conviction, ruling that he had to know not only that he was taking the casings, but also that someone else still had property rights in them. He could not be found liable “if he truly believed [the casings] to be abandoned.” Id., at 271; see id., at 276. By the same token, in Liparota v. United States, we considered a statute making it a crime to knowingly possess or use food stamps in an unauthorized manner. 471 U. S. 419, 420 (1985) . The Government’s argument, similar to its position in this case, was that a defendant’s conviction could be upheld if he knowingly possessed or used the food stamps, and in fact his possession or use was unauthorized. Id., at 423. But this Court rejected that interpretation of the statute, because it would have criminalized “a broad range of apparently innocent conduct” and swept in individuals who had no knowledge of the facts that made their conduct blameworthy. Id., at 426. For example, the statute made it illegal to use food stamps at a store that charged higher prices to food stamp customers. Without a mental state requirement in the statute, an individual who unwittingly paid higher prices would be guilty under the Government’s interpretation. Ibid. The Court noted that Congress could have intended to cover such a “broad range of conduct,” but declined “to adopt such a sweeping interpretation” in the absence of a clear indication that Congress intended that result. Id., at 427. The Court instead construed the statute to require knowledge of the facts that made the use of the food stamps unauthorized. Id., at 425. To take another example, in Posters ‘N’ Things, Ltd. v. United States, this Court interpreted a federal statute prohibiting the sale of drug paraphernalia. 511 U. S. 513 (1994) . Whether the items in question qualified as drug paraphernalia was an objective question that did not depend on the defendant’s state of mind. Id., at 517–522. But, we held, an individual could not be convicted of selling such paraphernalia unless he “knew that the items at issue [were] likely to be used with illegal drugs.” Id., at 524. Such a showing was necessary to establish the defendant’s culpable state of mind. And again, in X-Citement Video, we considered a statute criminalizing the distribution of visual depictions of minors engaged in sexually explicit conduct. 513 U. S., at 68. We rejected a reading of the statute which would have required only that a defendant knowingly send the prohibited materials, regardless of whether he knew the age of the performers. Id., at 68–69. We held instead that a defendant must also know that those depicted were minors, because that was “the crucial element separating legal innocence from wrongful conduct.” Id., at 73. See also Staples, 511 U. S., at 619 (defendant must know that his weapon had automatic firing capability to be convicted of possession of such a weapon). When interpreting federal criminal statutes that are silent on the required mental state, we read into the statute “only that mens rea which is necessary to separate wrongful conduct from ‘otherwise innocent conduct.’ ” Carter v. United States, 530 U. S. 255, 269 (2000) (quoting X-Citement Video, 513 U. S., at 72). In some cases, a general requirement that a defendant act knowingly is itself an adequate safeguard. For example, in Carter, we considered whether a conviction under 18 U. S. C. §2113(a), for taking “by force and violence” items of value belonging to or in the care of a bank, requires that a defendant have the intent to steal. 530 U. S., at 261. We held that once the Government proves the defendant forcibly took the money, “the concerns underlying the presumption in favor of scienter are fully satisfied, for a forceful taking—even by a defendant who takes under a good-faith claim of right—falls outside the realm of . . . ‘otherwise innocent’ ” conduct. Id., at 269–270. In other instances, however, requiring only that the defendant act knowingly “would fail to protect the innocent actor.” Id., at 269. A statute similar to Section 2113(a) that did not require a forcible taking or the intent to steal “would run the risk of punishing seemingly innocent conduct in the case of a defendant who peaceably takes money believing it to be his.” Ibid. In such a case, the Court explained, the statute “would need to be read to require . . . that the defendant take the money with ‘intent to steal or purloin.’ ” Ibid. C Section 875(c), as noted, requires proof that a communication was transmitted and that it contained a threat. The “presumption in favor of a scienter requirement should apply to each of the statutory elements that criminalize otherwise innocent conduct.” X-Citement Video, 513 U. S., at 72 (emphasis added). The parties agree that a defendant under Section 875(c) must know that he is transmitting a communication. But communicating something is not what makes the conduct “wrongful.” Here “the crucial element separating legal innocence from wrongful conduct” is the threatening nature of the communication. Id., at 73. The mental state requirement must therefore apply to the fact that the communication contains a threat. Elonis’s conviction, however, was premised solely on how his posts would be understood by a reasonable person. Such a “reasonable person” standard is a familiar feature of civil liability in tort law, but is inconsistent with “the conventional requirement for criminal conduct—awareness of some wrongdoing.” Staples, 511 U. S., at 606–607 (quoting United States v. Dotterweich, 320 U. S. 277, 281 (1943) ; emphasis added). Having liability turn on whether a “reasonable person” regards the communication as a threat—regardless of what the defendant thinks—“reduces culpability on the all-important element of the crime to negligence,” Jeffries, 692 F. 3d, at 484 (Sutton, J., dubitante), and we “have long been reluctant to infer that a negligence standard was intended in criminal statutes,” Rogers v. United States, 422 U. S. 35, 47 (1975) (Marshall, J., concurring) (citing Morissette, 342 U. S. 246 ). See 1 C. Torcia, Wharton’s Criminal Law §27, pp. 171–172 (15th ed. 1993); Cochran v. United States, 157 U. S. 286, 294 (1895) (defendant could face “liability in a civil action for negligence, but he could only be held criminally for an evil intent actually existing in his mind”). Under these principles, “what [Elonis] thinks” does matter. App. 286. The Government is at pains to characterize its position as something other than a negligence standard, emphasizing that its approach would require proof that a defendant “comprehended [the] contents and context” of the communication. Brief for United States 29. The Government gives two examples of individuals who, in its view, would lack this necessary mental state—a “foreigner, ignorant of the English language,” who would not know the meaning of the words at issue, or an individual mailing a sealed envelope without knowing its contents. Ibid. But the fact that the Government would require a defendant to actu-ally know the words of and circumstances surrounding a communication does not amount to a rejection of negligence. Criminal negligence standards often incorporate “the circumstances known” to a defendant. ALI, Model Penal Code §2.02(2)(d) (1985). See id., Comment 4, at 241; 1 LaFave, Substantive Criminal Law §5.4, at 372–373. Courts then ask, however, whether a reasonable person equipped with that knowledge, not the actual defendant, would have recognized the harmfulness of his conduct. That is precisely the Government’s position here: Elonis can be convicted, the Government contends, if he himself knew the contents and context of his posts, and a reason-able person would have recognized that the posts would be read as genuine threats. That is a negligence standard. In support of its position the Government relies most heavily on Hamling v. United States, 418 U. S. 87 (1974) . In that case, the Court rejected the argument that individuals could be convicted of mailing obscene material only if they knew the “legal status of the materials” distributed. Id., at 121. Absolving a defendant of liability because he lacked the knowledge that the materials were legally obscene “would permit the defendant to avoid prosecution by simply claiming that he had not brushed up on the law.” Id., at 123. It was instead enough for liability that “a defendant had knowledge of the contents of the materials he distributed, and that he knew the character and nature of the materials.” Ibid. This holding does not help the Government. In fact, the Court in Hamling approved a state court’s conclusion that requiring a defendant to know the character of the material incorporated a “vital element of scienter” so that “not innocent but calculated purveyance of filth . . . is exorcised.” Id., at 122 (quoting Mishkin v. New York, 383 U. S. 502, 510 (1966) ; internal quotation marks omitted). In this case, “calculated purveyance” of a threat would require that Elonis know the threatening nature of his communication. Put simply, the mental state requirement the Court approved in Hamling turns on whether a defendant knew the character of what was sent, not simply its contents and context. Contrary to the dissent’s suggestion, see post, at 4–5, 9–10 (opinion of Thomas, J.), nothing in Rosen v. United States, 161 U. S. 29 (1896) , undermines this reading. The defendant’s contention in Rosen was that his indictment for mailing obscene material was invalid because it did not allege that he was aware of the contents of the mailing. Id., at 31–33. That is not at issue here; there is no dispute that Elonis knew the words he communicated. The defendant also argued that he could not be convicted of mailing obscene material if he did not know that the material “could be properly or justly characterized as obscene.” Id., at 41. The Court correctly rejected this “ignorance of the law” defense; no such contention is at issue here. See supra, at 10. * * * In light of the foregoing, Elonis’s conviction cannot stand. The jury was instructed that the Government need prove only that a reasonable person would regard Elonis’s communications as threats, and that was error. Federal criminal liability generally does not turn solely on the results of an act without considering the defendant’s mental state. That understanding “took deep and early root in American soil” and Congress left it intact here: Under Section 875(c), “wrongdoing must be conscious to be criminal.” Morissette, 342 U. S., at 252. There is no dispute that the mental state requirement in Section 875(c) is satisfied if the defendant transmits a communication for the purpose of issuing a threat, or with knowledge that the communication will be viewed as a threat. See Tr. of Oral Arg. 25, 56. In response to a question at oral argument, Elonis stated that a finding of recklessness would not be sufficient. See id., at 8–9. Neither Elonis nor the Government has briefed or argued that point, and we accordingly decline to address it. See Department of Treasury, IRS v. FLRA, 494 U. S. 922, 933 (1990) (this Court is “poorly situated” to address an argument the Court of Appeals did not consider, the parties did not brief, and counsel addressed in “only the most cursory fashion at oral argument”). Given our disposition, it is not necessary to consider any First Amendment issues. Both Justice Alito and Justice Thomas complain about our not deciding whether recklessness suffices for liability under Section 875(c). Post, at 1–2 (Alito, J., concurring in part and dissenting in part); post, at 1–2 (opinion of Thomas, J.). Justice Alito contends that each party “argued” this issue, post, at 2, but they did not address it at all until oral argument, and even then only briefly. See Tr. of Oral Arg. at 8, 38–39. Justice Alito also suggests that we have not clarified confusion in the lower courts. That is wrong. Our holding makes clear that negligence is not sufficient to support a conviction under Section 875(c), contrary to the view of nine Courts of Appeals. Pet. for Cert. 17. There was and is no circuit conflict over the question Justice Alito and Justice Thomas would have us decide—whether recklessness suffices for liability under Section 875(c). No Court of Appeals has even addressed that question. We think that is more than sufficient “justification,” post, at 2 (opinion of Alito, J.), for us to decline to be the first appellate tribunal to do so. Such prudence is nothing new. See United States v. Bailey, 444 U. S. 394, 407 (1980) (declining to decide whether mental state of recklessness or negligence could suffice for criminal liability under 18 U. S. C. §751, even though a “court may someday confront a case” presenting issue); Ginsberg v. New York, 390 U. S. 629 –645 (1968) (rejecting defendant’s challenge to obscenity law “makes it unnecessary for us to define further today ‘what sort of mental element is requisite to a constitutionally permissible prosecution’ ”); Smith v. California, 361 U. S. 147, 154 (1959) (overturning conviction because lower court did not require any mental element under statute, but noting that “[w]e need not and most definitely do not pass today on what sort of mental element is requisite to a constitutionally permissible prosecution”); cf. Gulf Oil Co. v. Bernard, 452 U. S. 89 –104 (1981) (finding a lower court’s order impermissible under the First Amendment but not deciding “what standards are mandated by the First Amendment in this kind of case”). We may be “capable of deciding the recklessness issue,” post, at 2 (opinion of Alito, J.), but following our usual practice of awaiting a decision below and hearing from the parties would help ensure that we decide it correctly. The judgment of the United States Court of Appeals for the Third Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

575.US.2014_14-86

Respondent (Abercrombie) refused to hire Samantha Elauf, a practicing Muslim, because the headscarf that she wore pursuant to her religious obligations conflicted with Abercrombie’s employee dress policy. The Equal Employment Opportunity Commission (EEOC) filed suit on Elauf’s behalf, alleging a violation of Title VII of the Civil Rights Act of 1964, which, inter alia, prohibits a prospective employer from refusing to hire an applicant because of the applicant’s religious practice when the practice could be accommodated without undue hardship. The EEOC prevailed in the District Court, but the Tenth Circuit reversed, awarding Abercrombie summary judgment on the ground that failure-to-accommodate liability attaches only when the applicant provides the employer with actual knowledge of his need for an accommodation. Held: To prevail in a disparate-treatment claim, an applicant need show only that his need for an accommodation was a motivating factor in the employer’s decision, not that the employer had knowledge of his need. Title VII’s disparate-treatment provision requires Elauf to show that Abercrombie (1) “fail[ed] . . . to hire” her (2) “because of” (3) “[her] religion” (including a religious practice). 42 U. S. C. §2000e–2(a)(1). And its “because of” standard is understood to mean that the protected characteristic cannot be a “motivating factor” in an employment decision. §2000e–2(m). Thus, rather than imposing a knowledge standard, §2000e–2(a)(1) prohibits certain motives, regardless of the state of the actor’s knowledge: An employer may not make an applicant’s religious practice, confirmed or otherwise, a factor in employment decisions. Title VII contains no knowledge requirement. Furthermore, Title VII’s definition of religion clearly indicates that failure-to-accommodate challenges can be brought as disparate-treatment claims. And Title VII gives favored treatment to religious practices, rather than demanding that religious practices be treated no worse than other practices. Pp. 2–7. 731 F. 3d 1106, reversed and remanded. Scalia, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in the judgment. Thomas, J., filed an opinion concurring in part and dissenting in part.

Title VII of the Civil Rights Act of 1964 prohibits a prospective employer from refusing to hire an applicant in order to avoid accommodating a religious practice that it could accommodate without undue hardship. The question presented is whether this prohibition applies only where an applicant has informed the employer of his need for an accommodation. I We summarize the facts in the light most favorable to the Equal Employment Opportunity Commission (EEOC), against whom the Tenth Circuit granted summary judgment. Respondent Abercrombie & Fitch Stores, Inc., operates several lines of clothing stores, each with its own “style.” Consistent with the image Abercrombie seeks to project for each store, the company imposes a Look Policy that governs its employees’ dress. The Look Policy prohibits “caps”—a term the Policy does not define—as too informal for Abercrombie’s desired image. Samantha Elauf is a practicing Muslim who, consistent with her understanding of her religion’s requirements, wears a headscarf. She applied for a position in an Abercrombie store, and was interviewed by Heather Cooke, the store’s assistant manager. Using Abercrombie’s ordinary system for evaluating applicants, Cooke gave Elauf a rating that qualified her to be hired; Cooke was concerned, however, that Elauf’s headscarf would conflict with the store’s Look Policy. Cooke sought the store manager’s guidance to clarify whether the headscarf was a forbidden “cap.” When this yielded no answer, Cooke turned to Randall Johnson, the district manager. Cooke informed Johnson that she believed Elauf wore her headscarf because of her faith. Johnson told Cooke that Elauf’s headscarf would violate the Look Policy, as would all other headwear, religious or otherwise, and directed Cooke not to hire Elauf. The EEOC sued Abercrombie on Elauf’s behalf, claiming that its refusal to hire Elauf violated Title VII. The District Court granted the EEOC summary judgment on the issue of liability, 798 F. Supp. 2d 1272 (ND Okla. 2011), held a trial on damages, and awarded $20,000. The Tenth Circuit reversed and awarded Abercrombie summary judgment. 731 F. 3d 1106 (2013). It concluded that ordinarily an employer cannot be liable under Title VII for failing to accommodate a religious practice until the applicant (or employee) provides the employer with actual knowledge of his need for an accommodation. Id., at 1131. We granted certiorari. 573 U. S. ___ (2014). II Title VII of the Civil Rights Act of 1964 78Stat. 253, as amended, prohibits two categories of employment prac-tices. It is unlawful for an employer: “(1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin; or (2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race,color, religion, sex, or national origin.” 42 U. S. C. §2000e–2(a). These two proscriptions, often referred to as the “disparate treatment” (or “intentional discrimination”) provision and the “disparate impact” provision, are the only causes of action under Title VII. The word “religion” is defined to “includ[e] all aspects of religious observance and practice, as well as belief, unless an employer demonstrates that he is unable to reasonably accommodate to” a “religious observance or practice without undue hardship on the conduct of the employer’s business.” §2000e( j).[1] Abercrombie’s primary argument is that an applicant cannot show disparate treatment without first showing that an employer has “actual knowledge” of the applicant’s need for an accommodation. We disagree. Instead, an applicant need only show that his need for an accommodation was a motivating factor in the employer’s decision.[2] The disparate-treatment provision forbids employers to: (1) “fail . . . to hire” an applicant (2) “because of” (3) “such individual’s . . . religion” (which includes his religious practice). Here, of course, Abercrombie (1) failed to hire Elauf. The parties concede that (if Elauf sincerely believes that her religion so requires) Elauf’s wearing of a headscarf is (3) a “religious practice.” All that remains is whether she was not hired (2) “because of” her religious practice. The term “because of” appears frequently in antidiscrimination laws. It typically imports, at a minimum, the traditional standard of but-for causation. University of Tex. Southwestern Medical Center v. Nassar, 570 U. S. ___ (2013). Title VII relaxes this standard, however, to prohibit even making a protected characteristic a “motivating factor” in an employment decision. 42 U. S. C. §2000e–2(m). “Because of” in §2000e–2(a)(1) links the forbidden consideration to each of the verbs preceding it; an individual’s actual religious practice may not be a motivating factor in failing to hire, in refusing to hire, and so on. It is significant that §2000e–2(a)(1) does not impose a knowledge requirement. As Abercrombie acknowledges, some antidiscrimination statutes do. For example, the Americans with Disabilities Act of 1990 defines discrimination to include an employer’s failure to make “reason-able accommodations to the known physical or mental limitations” of an applicant. §12112(b)(5)(A) (emphasis added). Title VII contains no such limitation. Instead, the intentional discrimination provision prohibits certain motives, regardless of the state of the actor’s knowledge. Motive and knowledge are separate concepts. An employer who has actual knowledge of the need for an accommodation does not violate Title VII by refusing to hire an applicant if avoiding that accommodation is not his motive. Conversely, an employer who acts with the motive of avoiding accommodation may violate Title VII even if he has no more than an unsubstantiated suspicion that accommodation would be needed. Thus, the rule for disparate-treatment claims based on a failure to accommodate a religious practice is straightforward: An employer may not make an applicant’s religious practice, confirmed or otherwise, a factor in employment decisions. For example, suppose that an employer thinks (though he does not know for certain) that a job applicant may be an orthodox Jew who will observe the Sabbath, and thus be unable to work on Saturdays. If the applicant actually requires an accommodation of that religious practice, and the employer’s desire to avoid the prospective accommodation is a motivating factor in his decision, the employer violates Title VII. Abercrombie urges this Court to adopt the Tenth Circuit’s rule “allocat[ing] the burden of raising a religious conflict.” Brief for Respondent 46. This would require the employer to have actual knowledge of a conflict between an applicant’s religious practice and a work rule. The problem with this approach is the one that inheres in most incorrect interpretations of statutes: It asks us to add words to the law to produce what is thought to be a desirable result. That is Congress’s province. We construe Title VII’s silence as exactly that: silence. Its disparate-treatment provision prohibits actions taken with the motive of avoiding the need for accommodating a religious practice. A request for accommodation, or the employer’s certainty that the practice exists, may make it easier to infer motive, but is not a necessary condition of liability.[3] Abercrombie argues in the alternative that a claim based on a failure to accommodate an applicant’s religious practice must be raised as a disparate-impact claim, not a disparate-treatment claim. We think not. That might have been true if Congress had limited the meaning of “religion” in Title VII to religious belief—so that discriminating against a particular religious practice would not be disparate treatment though it might have disparate impact. In fact, however, Congress defined “religion,” for Title VII’s purposes, as “includ[ing] all aspects of religious observance and practice, as well as belief.” 42 U. S. C. §2000e(j). Thus, religious practice is one of the protected characteristics that cannot be accorded disparate treatment and must be accommodated. Nor does the statute limit disparate-treatment claims to only those employer policies that treat religious practices less favorably than similar secular practices. Abercrombie’s argument that a neutral policy cannot constitute “intentional discrimination” may make sense in other contexts. But Title VII does not demand mere neutrality with regard to religious practices—that they be treated no worse than other practices. Rather, it gives them favored treatment, affirmatively obligating employers not “to fail or refuse to hire or discharge any individual . . . because of such individual’s” “religious observance and practice.” An employer is surely entitled to have, for example, a no-headwear policy as an ordinary matter. But when an applicant requires an accommodation as an “aspec[t] of religious . . . practice,” it is no response that the sub-sequent “fail[ure] . . . to hire” was due to an otherwise-neutral policy. Title VII requires otherwise-neutralpolicies to give way to the need for an accommodation. * * * The Tenth Circuit misinterpreted Title VII’s requirements in granting summary judgment. We reverse its judgment and remand the case for further consideration consistent with this opinion. It is so ordered.Notes 1 For brevity’s sake, we will in the balance of this opinion usuallyomit reference to the §2000e( j) “undue hardship” defense to the accom-modation requirement, discussing the requirement as though it is absolute. 2 The concurrence mysteriously concludes that it is not the plaintiff ’s burden to prove failure to accommodate. Post, at 5. But of course that is the plaintiff’s burden, if failure to hire “because of” the plaintiff’s “religious practice” is the gravamen of the complaint. Failing to hire for that reason is synonymous with refusing to accommodate the religious practice. To accuse the employer of the one is to accuse him of the other. If he is willing to “accommodate”—which means nothing more than allowing the plaintiff to engage in her religious practice despite the employer’s normal rules to the contrary—adverse action “because of” the religious practice is not shown. “The clause that begins with the word ‘unless,’” as the concurrence describes it, ibid., has no function except to place upon the employer the burden of establishing an “undue hardship” defense. The concurrence provides no example, not even an unrealistic hypothetical one, of a claim of failure to hire because of religious practice that does not say the employer refused to permit (“failed to accommodate”) the religious practice. In the nature of things, there cannot be one. 3 While a knowledge requirement cannot be added to the motive requirement, it is arguable that the motive requirement itself is not met unless the employer at least suspects that the practice in question is a religious practice—i.e., that he cannot discriminate “because of” a “religious practice” unless he knows or suspects it to be a religious practice. That issue is not presented in this case, since Abercrombie knew—or at least suspected—that the scarf was worn for religious reasons. The question has therefore not been discussed by either side, in brief or oral argument. It seems to us inappropriate to resolve this unargued point by way of dictum, as the concurrence would do.

574.US.405

Three legal prescriptions figure in this case. Title 28 U. S. C. §1291 gives the courts of appeals jurisdiction over appeals from “all final decisions of the district courts of the United States,” and its core application is to rulings that terminate an action. Federal Rule of Civil Procedure 54(b) permits district courts to authorize immediate appeal of dispositive rulings on separate claims in a civil action raising multiple claims. And 28 U. S. C. §1407 authorizes the Judicial Panel on Multidistrict Litigation (JPML) to transfer civil actions “involving one or more common questions of fact . . . to any district for coordinated or consolidated pretrial proceedings” in order to “promote the just and efficient conduct of such actions,” §1407(a). The London InterBank Offered Rate (LIBOR) is a reference point in determining interest rates for financial instruments in the United States and globally. The JPML established a multidistrict litigation (LIBOR MDL) for cases involving allegations that defendant-banks understated their borrowing costs, thereby depressing LIBOR and enabling the banks to pay lower interest rates on financial instruments sold to investors. Over 60 actions were consolidated for pretrial proceedings in the U. S. District Court for the Southern District of New York, including a class action filed by petitioners Ellen Gelboim and Linda Zacher, who raised the single claim that several banks, acting in concert, had violated federal antitrust law. Determining that no plaintiff could assert a cognizable antitrust injury, the District Court granted the banks’ motion to dismiss all antitrust claims, including the Gelboim-Zacher complaint’s sole claim. The District Court thus dismissed the Gelboim-Zacher complaint, denied leave to amend, and dismissed the case in its entirety. Other cases made part of the LIBOR MDL, however, presented discrete claims which remained before the District Court. Assuming that the Gelboim-Zacher plaintiffs were entitled to an immediate appeal of right under §1291, the District Court granted Rule 54(b) certifications authorizing the plaintiffs in some of the multiple-claim actions to appeal the dismissal of their antitrust claims while their other claims remained pending in the District Court. On its own initiative, the Second Circuit dismissed the Gelboim-Zacher appeal because the order appealed from did not dispose of all of the claims in the consolidated action. The District Court thereafter withdrew its Rule 54(b) certifications. Held: The order dismissing their case in its entirety removed Gelboim and Zacher from the consolidated proceeding, thereby triggering their right to appeal under §1291. Because cases consolidated for MDL pretrial proceedings ordinarily retain their separate identities, an order disposing of one of the discrete cases in its entirety should qualify under §1291 as an appealable final decision. Section 1407 refers to individual “actions” transferrable to a single district court, not to a monolithic multidistrict “action” created by transfer. See Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26, 37. And §1407(a)’s language—“at or before the conclusion of . . . pretrial proceedings,” each transferred action must be remanded to the originating district “unless [the action] shall have been previously terminated”—indicates Congress’ anticipation that, during the pendency of pretrial proceedings, final decisions might be rendered in one or more of the actions consolidated pursuant to §1407. The District Court’s order dismissing the Gelboim-Zacher complaint was a final decision. The District Court completed its adjudication of petitioners’ complaint and terminated their action. Petitioners thus are no longer participants in the consolidated proceedings. Nothing about the initial consolidation of their civil action with other LIBOR MDL cases renders the dismissal of their complaint tentative or incomplete. To hold, as the banks contend, that no appeal of right accrues until a §1407 consolidation ends would leave plaintiffs like Gelboim and Zacher in a quandary about the event that triggers the 30-day period for taking an appeal. The sensible solution to the appeal-clock trigger is to allow an immediate appeal in a case such as this, where the transferee court in an MDL grants a defendant’s dispositive motion on every claim (or the sole claim) in a transferred case. The banks are also concerned about allowing plaintiffs with the weakest cases to appeal because their complaint states only one claim, while leaving those with stronger cases unable to appeal simultaneously because they have other pending claims. But that concern is attended to by Rule 54(b), which authorizes district courts to grant certifications to parties with multiple-claim complaints, thereby enabling plaintiffs in actions that have not been dismissed in their entirety to pursue immediate appellate review of discrete claims. The District Court did that in this very case. Rule 54(b), however, is of no avail to Gelboim and Zacher, who asserted only one claim. See Sears, Roebuck & Co. v. Mackey, 351 U.S. 427, 435. Section 1292(b)—which allows district courts to designate for review certain interlocutory orders—is also inapposite here, for there is nothing “interlocutory” about the dismissal order in the Gelboim-Zacher action. Pp. 6–10. Reversed and remanded. Ginsburg, J., delivered the opinion for a unanimous Court.

An unsuccessful litigant in a federal district court may take an appeal, as a matter of right, from a “final decisio[n] of the district cour[t].”28 U. S. C. §1291. The question here presented: Is the right to appeal secured by §1291 affected when a case is consolidated for pretrial proceedings in multidistrict litigation (or MDL) authorized by28 U. S. C. §1407? Petitioners Ellen Gelboim and Linda Zacher filed in the United States District Court for the Southern District of New York a class-action complaint raising a single claim. They alleged that a number of banks, acting in concert, had violated federal antitrust law. Their case was consolidated for pretrial proceedings together with some 60 other cases, commenced in different districts, raising “one or more common questions of fact,” §1407(a). The defendant banks, respondents here, moved to dismiss the Gelboim-Zacher complaint on the ground that the plaintiffs had suffered no antitrust injury. The District Court granted the motion, denied leave to amend the complaint, and dismissed the case in its entirety. Other cases made part of the multidistrict pretrial proceedings, however, presented discrete claims and remained before the District Court. The Court of Appeals for the Second Circuit, acting on its own motion, dismissed the appeal filed by Gelboim and Zacher for want of appellate jurisdiction. We reverse the Second Circuit’s judgment and hold that the Gelboim-Zacher complaint retained its independent status for purposes of appellate jurisdiction under §1291. Petitioners’ right to appeal ripened when the District Court dismissed their case, not upon eventual completion of multidistrict proceedings in all of the consolidated cases.I Three legal prescriptions figure in this case: Title 28 U. S. C. §§1291 and 1407, and Federal Rule of Civil Procedure 54(b). Section 1291 gives the courts of appeals jurisdiction over appeals from “all final decisions of the district courts of the United States.” A “final decision” is one “by which a district court disassociates itself from a case.” Swint v. Chambers County Comm’n,514 U. S. 35,42 (1995). While decisions of this Court have accorded §1291 a “practical rather than a technical construction,” Mohawk Industries, Inc. v. Carpenter,558 U. S. 100,106 (2009) (quoting Cohen v. Beneficial Industrial Loan Corp.,337 U. S. 541,546 (1949)), the statute’s core application is to rulings that terminate an action, see Catlin v. United States,324 U. S. 229,233 (1945) (final decision is “one which ends the litigation on the merits and leaves nothing for the court to do but execute the judgment”). Rule 54(b) permits district courts to authorize immediate appeal of dispositive rulings on separate claims in a civil action raising multiple claims:“When an action presents more than one claim for relief . . . or when multiple parties are involved, the court may direct entry of a final judgment as to one or more, but fewer than all, claims or parties only if the court expressly determines that there is no just reason for delay.”[1]Rule 54(b) relaxes “the former general practice that, in multiple claims actions, all the claims had to be finally decided before an appeal could be entertained from a final decision upon any of them.” Sears, Roebuck & Co. v. Mackey,351 U. S. 427,434 (1956). The Federal Rules allow a plaintiff to “state [in one complaint] as many separate claims . . . as it has.” Rule 8(d)(3). Rule 54(b) was adopted in view of the breadth of the “civil action” the Rules allow, specifically “to avoid the possible injustice” of “delay[ing] judgment o[n] a distinctly separate claim [pending] adjudication of the entire case.” Report of Advisory Committee on Proposed Amendments to Rules of Civil Procedure 70 (1946) (explaining that Rule 54(b) was recast in 1946 to avoid confusion and misapplication); see Dickinson v. Petroleum Conversion Corp.,338 U. S. 507,511 (1950) (Rule 54(b) responded to liberalized joinder of claims and parties under the Federal Rules, which “increased the danger of hardship and denial of justice through delay if each issue must await the determination of all issues as to all parties before a final judgment can be had”). The Rule thus aimed to augment, not diminish, appeal opportunity. Section 1407 is of more recent vintage. Enacted in 1968 in response to a growing number of complex but related cases filed in multiple districts, §1407 authorizes the Judicial Panel on Multidistrict Litigation (JPML) to transfer civil actions “involving one or more common questions of fact . . . to any district for coordinated or consolidated pretrial proceedings” in order to “promote the just and efficient conduct of such actions.” §1407(a); see H. R. Rep. No. 1130, 90th Cong., 2d Sess., 2 (1968) (§1407 codified procedures used in the early 1960’s to resolve more than 1,800 separate actions filed against electrical equipment manufacturers in 33 District Courts, all of the actions seeking damages for antitrust law violations). Transfer under §1407 aims to “eliminate duplication in discovery, avoid conflicting rulings and schedules, reduce litigation cost, and save the time and effort of the parties, the attorneys, the witnesses, and the courts.” Manual for Complex Litigation §20.131, p. 220 (4th ed. 2004). “Each action” transferred pursuant to §1407, the provision instructs, “shall be remanded by the panel at or before the conclusion of . . . pretrial proceedings to the district from which it was transferred unless it shall have been previously terminated.” §1407(a).II The London InterBank Offered Rate (LIBOR) is a benchmark interest rate disseminated by the British Bankers’ Association based on the rate at which certain banks predict they can borrow funds. LIBOR is a reference point in determining interest rates for financial instruments in the United States and globally. In August 2011, the JPML established MDL No. 2262 (LIBOR MDL) for cases involving allegations that the banks named as defendants understated their borrowing costs, thereby depressing LIBOR and enabling the banks to pay lower interest rates on financial instruments sold to investors. In re Libor-Based Financial Instruments Antitrust Litigation, 802 F. Supp. 2d 1380 (JPML 2011). Composing the LIBOR MDL, over 60 actions, commenced in California, Illinois, Iowa, Kansas, Massachusetts, Minnesota, New Jersey, New York, Ohio, Pennsylvania, Texas, Virginia, and Wisconsin, were coordinated or consolidated for pretrial proceedings in the United States District Court for the Southern District of New York. In June 2012, the District Court entertained a motion to dismiss four categories of cases included in the MDL. The first three categories involved putative class actions, each with a single lead case: (1) the Gelboim-Zacher action, filed on behalf of purchasers of bonds with LIBOR-linked interest rates; (2) an action filed on behalf of purchasers of over-the-counter LIBOR-based instruments (OTC plaintiffs); (3) an action filed on behalf of purchasers of LIBOR-based instruments on exchanges (Exchange plaintiffs). The fourth category, not relevant here, comprised a set of individual actions filed by Charles Schwab Corporation and related entities. The Gelboim-Zacher complaint asserted a federal antitrust claim under §1 of the Sherman Act, 15 U. S. C. §1, and that claim only, while the complaints in the other actions asserted a federal antitrust claim in addition to other differently based federal and state claims. Determining that no plaintiff could assert a cognizable antitrust injury, the District Court granted the banks’ motion to dismiss plaintiffs’ antitrust claims—the sole claim raised in the Gelboim-Zacher complaint. Assuming that the Gelboim-Zacher plaintiffs were entitled to an immediate appeal of right under §1291 because their suit had been “dismissed in [its] entirety,” App. to Pet. for Cert. 219a, the District Court granted Rule 54(b) certifications to the OTC and Exchange plaintiffs authorizing them to appeal the dismissal of their antitrust claims while their other claims remained pending in the District Court. On its own initiative, the Second Circuit dismissed the Gelboim-Zacher appeal because the “orde[r] appealed from did not dispose of all claims in the consolidated action.” Id., at 2a.[2] The District Court thereafter withdrew its Rule 54(b) certifications in the OTC and Exchange plaintiffs’ actions and, “given the reaction of the Second Circuit,” App. 294, denied petitioners’ request for a Rule 54(b) certification. We granted review of the Second Circuit’s judgment dismissing the Gelboim-Zacher appeal. 573 U. S. ___ (2014). Before this Court, petitioners Gelboim and Zacher contend that the order dismissing their case in its entirety removed them from the consolidated proceeding, thereby triggering their right to appeal under §1291. Respondent banks urge that consolidated cases proceed as one unit for the duration of the consolidation. Consequently, they maintain, there is no appeal of right from an order dismissing fewer than all consolidated claims, thus the sole avenue for appeal while the consolidation continues is Rule 54(b). Agreeing with Gelboim and Zacher, we reverse the Court of Appeals’ judgment.III Cases consolidated for MDL pretrial proceedings ordinarily retain their separate identities,[3] so an order disposing of one of the discrete cases in its entirety should qual-ify under §1291 as an appealable final decision. Section 1407 refers to individual “actions” which may be transferred to a single district court, not to any monolithic multidistrict “action” created by transfer. See Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach,523 U. S. 26,37 (1998) (§1407 does not “imbu[e] transferred actions with some new and distinctive . . . character”).[4] And Congress anticipated that, during the pendency of pretrial proceedings, final decisions might be rendered in one or more of the actions consolidated pursuant to §1407. It specified that “at or before the conclusion of . . . pretrial proceedings,” each of the transferred actions must be remanded to the originating district “unless [the action] shall have been previously terminated.” §1407(a) (emphasis added). The District Court’s order dismissing the Gelboim-Zacher complaint for lack of antitrust injury, without leave to amend, had the hallmarks of a final decision. Ruling on the merits of the case, the District Court completed its adjudication of petitioners’ complaint and terminated their action. As a result of the District Court’s disposition, petitioners are no longer participants in the consolidated proceedings. Nothing about the initial consolidation of their civil action with other cases in the LIBOR MDL renders the dismissal of their complaint in any way tentative or incomplete. As is ordinarily the case, the §1407 consolidation offered convenience for the parties and promoted efficient judicial administration, but did not meld the Gelboim-Zacher action and others in the MDL into a single unit. Cf. supra, at 6, n. 3.[5] The banks’ view that, in a §1407 consolidation, no appeal of right accrues until the consolidation ends would leave plaintiffs like Gelboim and Zacher in a quandary about the proper timing of their appeals. Under Federal Rule of Appellate Procedure 4, which this Court has called “jurisdictional,” Bowles v. Russell,551 U. S. 205,209 (2007), a notice of appeal in a civil case must be filed “within 30 days after entry of the judgment or order appealed from,” Rule 4(a)(1)(A). If plaintiffs whose actions have been dismissed with prejudice by a district court must await the termination of pretrial proceedings in all consolidated cases, what event or order would start the 30-day clock? When pretrial consolidation concludes, there may be no occasion for the entry of any judgment. Orders may issue returning cases to their originating courts,[6] but an order of that genre would not qualify as the dispositive ruling Gelboim and Zacher seek to overturn on appeal. And surely would-be appellants need not await final disposition of all cases in their originating districts, long after pretrial consolidation under §1407 could even arguably justify treating the cases as a judicial unit. The sensible solution to the appeal-clock trigger is evident: When the transferee court overseeing pretrial proceedings in multidistrict litigation grants a defendant’s dispositive motion “on all issues in some transferred cases, [those cases] become immediately appealable . . . while cases where other issues remain would not be appealable at that time.” D. Herr, Multidistrict Litigation Manual §9:21, p. 312 (2014). The banks express concern that plaintiffs with the weakest cases may be positioned to appeal because their complaint states only one claim, while plaintiffs with stronger cases will be unable to appeal simultaneously because they have other claims still pending. Brief for Respondents 46–47. Rule 54(b) attends to this concern. District courts may grant certifications under that Rule, thereby enabling plaintiffs in actions that have not been dismissed in their entirety to pursue immediate appellate review. That is just what happened in this very case. The District Court granted Rule 54(b) certifications to the OTC and Exchange plaintiffs so they could appeal at the same time Gelboim and Zacher could. See supra, at 5. And if the MDL court believes that further proceedings might be relevant to a claim a defendant moves to dismiss, the court ordinarily can defer ruling on the motion, thus allowing all plaintiffs to participate in the ongoing MDL proceedings. While Rule 54(b) can aid parties with multiple-claim complaints—here, the OTC and Exchange plaintiffs, supra, at 5—the Rule, properly read, is of no avail to Gelboim and Zacher. Rule 54(b) addresses orders finally adjudicating fewer than all claims presented in a civil action complaint. It “does not apply to a single claim action nor to a multiple claims action in which all of the claims have been finally decided.” Mackey, 351 U. S., at 435; see Liberty Mut. Ins. Co. v. Wetzel,424 U. S. 737–743 (1976) (Rule 54(b) inapplicable where “complaint advanced a single legal theory which was applied to only one set of facts”). In short, Rule 54(b) is designed to permit acceleration of appeals in multiple-claim cases, not to retard appeals in single-claim cases.[7] Section 1292(b), the banks conceded at argument, see Tr. of Oral Arg. 40–41, is inapposite here. It allows district courts to designate for review interlocutory orders “not otherwise appealable,” where immediate appeal “may materially advance the ultimate termination of the litigation.” §1292(b). The designation may be accepted or rejected in the discretion of the court of appeals. Ibid. See generally Solimine, Revitalizing Interlocutory Appeals in the Federal Courts, 58 Geo. Wash. L. Rev. 1165 (1990); Note, Interlocutory Appeals in the Federal Courts Under 28 U. S. C. §1292(b), 88 Harv. L. Rev. 607 (1975). It suffices to note that there is nothing “interlocutory” about the dismissal order in the Gelboim-Zacher action.* * * For the reasons stated, we reverse the judgment of the U. S. Court of Appeals for the Second Circuit deeming the District Court’s dismissal of the Gelboim-Zacher complaint unripe for appellate review, and we remand the case for further proceedings consistent with this opinion.It is so ordered.Notes1 Compare Rule 54(b), which lodges discretion to authorize appeals in district courts, with Federal Rule of Civil Procedure 23(f), which authorizes courts of appeals to permit an immediate appeal from a district court order granting or denying class-action certification.2 The Second Circuit “strong[ly] presum[es]” that a “judgment in a consolidated [proceeding] that does not dispose of all [consolidated] claims . . . is not appealable absent Rule 54(b) certification.” Hageman v. City Investing Co., 851 F. 2d 69, 71 (1988). In this regard, the Circuit does not differentiate between all-purpose consolidations, see ibid. (actions “could originally have been brought as one action” and there was “no indication that the cases were consolidated only for limited purposes”); Houbigant, Inc. v. IMG Fragrance Brands, LLC, 627 F. 3d 497, 498 (2010) (actions consolidated “for all purposes”), and, as this case illustrates, §1407 consolidations for pretrial proceedings only. The presumption may be overcome in “highly unusual circumstances,” Hageman, 851 F. 2d, at 71, but the Second Circuit has not elaborated on what those circumstances might be.3 Parties may elect to file a “master complaint” and a corresponding “consolidated answer,” which supersede prior individual pleadings. In such a case, the transferee court may treat the master pleadings as merging the discrete actions for the duration of the MDL pretrial proceedings. In re Refrigerant Compressors Antitrust Litigation, 731 F. 3d 586, 590–592 (CA6 2013). No merger occurs, however, when “the master complaint is not meant to be a pleading with legal effect but only an administrative summary of the claims brought by all the plaintiffs.” Id., at 590.4 We express no opinion on whether an order deciding one of multiple cases combined in an all-purpose consolidation qualifies under §1291 as a final decision appealable of right. See Brown v. United States, 976 F. 2d 1104, 1107 (CA7 1992) (cases consolidated for all purposes “become a single judicial unit,” but where the consolidation is for limited purposes only, “a decision disposing of all the claims in only one of the cases is a final decision subject to immediate appeal”).5 In delineating the narrow scope of the “collateral-order” doctrine, we have cautioned against permitting piecemeal, prejudgment appeals. Those admonitions, cited by the banks, Brief for Respondents 18–19, are not pertinent here. Under the collateral order doctrine, an order may be deemed “final” if it disposes of a matter “separable from, and collateral to” the merits of the main proceeding, “too important to be denied review,” and “too independent of the cause itself to require that appellate consideration be deferred until the whole case is adjudicated.” Cohen v. Beneficial Industrial Loan Corp.,337 U. S. 541,546 (1949). The order dismissing the Gelboim-Zacher complaint in its entirety was in no sense “collateral,” i.e., “independent of the cause itself.” Scarcely a prejudgment ruling, the dismissal order left nothing still pending decision in the District Court.6 In fact, “[f]ew cases [consolidated pursuant to §1407] are remanded for trial; most multidistrict litigation is settled in the transferee court.” Manual for Complex Litigation §20.132, p. 223 (4th ed. 2004).7 We need not decide whether or how Rule 54(b) applies to cases consolidated for all purposes involving closely related issues, actions that could have been brought under the umbrella of one complaint. Cf. supra, at 7, n. 4. The Rule surely was not designed to apply to numerous actions that the MDL panel combines for efficient pretrial proceedings because they have in common “one or more questions of fact,” but otherwise vary in character.

576.US.2014_14-7955

Because capital punishment is constitutional, there must be a constitutional means of carrying it out. After Oklahoma adopted lethal injection as its method of execution, it settled on a three-drug protocol of (1) sodium thiopental (a barbiturate) to induce a state of unconsciousness, (2) a paralytic agent to inhibit all muscular-skeletal movements, and (3) potassium chloride to induce cardiac arrest. In Baze v. Rees, 553 U. S. 35 , the Court held that this protocol does not violate the Eighth Amendment’s prohibition against cruel and unusual punishments. Anti-death-penalty advocates then pressured pharmaceutical companies to prevent sodium thiopental (and, later, another barbiturate called pentobarbital) from being used in executions. Unable to obtain either sodium thiopental or pentobarbital, Oklahoma decided to use a 500-milligram dose of midazolam, a sedative, as the first drug in its three-drug protocol. Oklahoma death-row inmates filed a 42 U. S. C. §1983 action claiming that the use of midazolam violates the Eighth Amendment. Four of those inmates filed a motion for a preliminary injunction and argued that a 500-milligram dose of midazolam will not render them unable to feel pain associated with administration of the second and third drugs. After a three-day evidentiary hearing, the District Court denied the motion. It held that the prisoners failed to identify a known and available alternative method of execution that presented a substantially less severe risk of pain. It also held that the prisoners failed to establish a likelihood of showing that the use of midazolam created a demonstrated risk of severe pain. The Tenth Circuit affirmed. Held: Petitioners have failed to establish a likelihood of success on the merits of their claim that the use of midazolam violates the Eighth Amendment. Pp. 11–29. (a) To obtain a preliminary injunction, petitioners must establish, among other things, a likelihood of success on the merits of their claim. See Winter v. Natural Resources Defense Council, Inc., 555 U. S. 7 . To succeed on an Eighth Amendment method-of-execution claim, a prisoner must establish that the method creates a demonstrated risk of severe pain and that the risk is substantial when compared to the known and available alternatives. Baze, supra, at 61 (plurality opinion). Pp. 11–13. (b) Petitioners failed to establish that any risk of harm was substantial when compared to a known and available alternative method of execution. Petitioners have suggested that Oklahoma could execute them using sodium thiopental or pentobarbital, but the District Court did not commit a clear error when it found that those drugs are unavailable to the State. Petitioners argue that the Eighth Amendment does not require them to identify such an alternative, but their argument is inconsistent with the controlling opinion in Baze, which imposed a requirement that the Court now follows. Petitioners also argue that the requirement to identify an alternative is inconsistent with the Court’s pre-Baze decision in Hill v. McDonough, 547 U. S. 573 , but they misread that decision. Hill concerned a question of civil procedure, not a substantive Eighth Amendment question. That case held that §1983 alone does not require an inmate asserting a method-of-execution claim to plead an acceptable alternative. Baze, on the other hand, made clear that the Eighth Amendment requires a prisoner to plead and prove a known and available alternative. Pp. 13–16. (c) The District Court did not commit clear error when it found that midazolam is likely to render a person unable to feel pain associated with administration of the paralytic agent and potassium chloride. Pp. 16–29. (1) Several initial considerations bear emphasis. First, the District Court’s factual findings are reviewed under the deferential “clear error” standard. Second, petitioners have the burden of persuasion on the question whether midazolam is effective. Third, the fact that numerous courts have concluded that midazolam is likely to render an inmate insensate to pain during execution heightens the deference owed to the District Court’s findings. Finally, challenges to lethal injection protocols test the boundaries of the authority and competency of federal courts, which should not embroil themselves in ongoing scientific controversies beyond their expertise. Baze, supra, at 51. Pp. 16–18. (2) The State’s expert presented persuasive testimony that a 500-milligram dose of midazolam would make it a virtual certainty that an inmate will not feel pain associated with the second and third drugs, and petitioners’ experts acknowledged that they had no contrary scientific proof. Expert testimony presented by both sides lends support to the District Court’s conclusion. Evidence suggested that a 500-milligram dose of midazolam will induce a coma, and even one of petitioners’ experts agreed that as the dose of midazolam increases, it is expected to produce a lack of response to pain. It is not dispositive that midazolam is not recommended or approved for use as the sole anesthetic during painful surgery. First, the 500-milligram dose at issue here is many times higher than a normal therapeutic dose. Second, the fact that a low dose of midazolam is not the best drug for maintaining unconsciousness says little about whether a 500-milligram dose is constitutionally adequate to conduct an execution. Finally, the District Court did not err in concluding that the safeguards adopted by Oklahoma to ensure proper administration of midazolam serve to minimize any risk that the drug will not operate as intended. Pp. 18–22. (3) Petitioners’ speculative evidence regarding midazolam’s “ceiling effect” does not establish that the District Court’s findings were clearly erroneous. The mere fact that midazolam has a ceiling above which an increase in dosage produces no effect cannot be dispositive, and petitioners provided little probative evidence on the relevant question, i.e., whether midazolam’s ceiling effect occurs below the level of a 500-milligram dose and at a point at which the drug does not have the effect of rendering a person insensate to pain caused by the second and third drugs. Petitioners attempt to deflect attention from their failure of proof on this point by criticizing the testimony of the State’s expert. They emphasize an apparent conflict between the State’s expert and their own expert regarding the biological process that produces midazolam’s ceiling effect. But even if petitioners’ expert is correct regarding that biological process, it is largely beside the point. What matters for present purposes is the dosage at which the ceiling effect kicks in, not the biological process that produces the effect. Pp. 22–25. (4) Petitioners’ remaining arguments—that an expert report presented in the District Court should have been rejected because it referenced unreliable sources and contained an alleged mathematical error, that only four States have used midazolam in an execution, and that difficulties during two recent executions suggest that midazolam is ineffective—all lack merit. Pp. 26–29. 776 F. 3d 721, affirmed. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, and Thomas, JJ., joined. Scalia, J., filed a concurring opinion, in which Thomas, J., joined. Thomas, J., filed a concurring opinion, in which Scalia, J., joined. Breyer, J., filed a dissenting opinion, in which Ginsburg, J., joined. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, Breyer, and Kagan, JJ., joined.

Prisoners sentenced to death in the State of Oklahoma filed an action in federal court under Rev. Stat. §1979, 42 U. S. C. §1983, contending that the method of execution now used by the State violates the Eighth Amendment because it creates an unacceptable risk of severe pain. They argue that midazolam, the first drug employed in the State’s current three-drug protocol, fails to render a person insensate to pain. After holding an evidentiary hearing, the District Court denied four prisoners’ application for a preliminary injunction, finding that they had failed to prove that midazolam is ineffective. The Court of Appeals for the Tenth Circuit affirmed and accepted the District Court’s finding of fact regarding midazolam’s efficacy. For two independent reasons, we also affirm. First, the prisoners failed to identify a known and available alternative method of execution that entails a lesser risk of pain, a requirement of all Eighth Amendment method-of-execution claims. See Baze v. Rees, 553 U. S. 35, 61 (2008) (plurality opinion). Second, the District Court did not commit clear error when it found that the prisoners failed to establish that Oklahoma’s use of a massive dose of midazolam in its execution protocol entails a substantial risk of severe pain. I A The death penalty was an accepted punishment at the time of the adoption of the Constitution and the Bill of Rights. In that era, death sentences were usually carried out by hanging. The Death Penalty in America: Current Controversies 4 (H. Bedau ed. 1997). Hanging remained the standard method of execution through much of the 19th century, but that began to change in the century’s later years. See Baze, supra, at 41–42. In the 1880’s, the Legislature of the State of New York appointed a commission to find “ ‘the most humane and practical method known to modern science of carrying into effect the sentence of death in capital cases.’ ” In re Kemmler, 136 U. S. 436, 444 (1890) . The commission recommended electrocution, and in 1888, the Legislature enacted a law providing for this method of execution. Id., at 444–445. In subsequent years, other States followed New York’s lead in the “ ‘belief that electrocution is less painful and more humane than hanging.’ ” Baze, 553 U. S., at 42 (quoting Malloy v. South Carolina, 237 U. S. 180, 185 (1915) ). In 1921, the Nevada Legislature adopted another new method of execution, lethal gas, after concluding that this was “the most humane manner known to modern science.” State v. Jon, 46 Nev. 418, 437, 211 P. 676, 682 (1923). The Nevada Supreme Court rejected the argument that the use of lethal gas was unconstitutional, id., at 435–437, 211 P., at 681–682, and other States followed Nevada’s lead, see, e.g., Ariz. Const., Art. XXII, §22 (1933); 1937 Cal. Stats. ch. 172, §1; 1933 Colo. Sess. Laws ch. 61, §1; 1955 Md. Laws ch. 625, §1, p. 1017; 1937 Mo. Laws p. 222, §1. Nevertheless, hanging and the firing squad were retained in some States, see, e.g., 1961 Del. Laws ch. 309, §2 (hanging); 1935 Kan. Sess. Laws ch. 155, §1 (hanging); Utah Code Crim. Proc. §105–37–16 (1933) (hanging or firing squad), and electrocution remained the predominant method of execution until the 9-year hiatus in executions that ended with our judgment in Gregg v. Georgia, 428 U. S. 153 (1976) . See Baze, supra, at 42. After Gregg reaffirmed that the death penalty does not violate the Constitution, some States once again sought a more humane way to carry out death sentences. They eventually adopted lethal injection, which today is “by far the most prevalent method of execution in the United States.” Baze, supra, at 42. Oklahoma adopted lethal injection in 1977, see 1977 Okla. Sess. Laws p. 89, and it eventually settled on a protocol that called for the use of three drugs: (1) sodium thiopental, “a fast-acting barbiturate sedative that induces a deep, comalike unconsciousness when given in the amounts used for lethal injection,” (2) a paralytic agent, which “inhibits all muscular-skeletal movements and, by paralyzing the diaphragm, stops respiration,” and (3) potassium chloride, which “interferes with the electrical signals that stimulate the contractions of the heart, inducing cardiac arrest.” Baze, supra, at 44; see also Brief for Respondents 9. By 2008, at least 30 of the 36 States that used lethal injection employed that particular three-drug protocol. 553 U. S., at 44. While methods of execution have changed over the years, “[t]his Court has never invalidated a State’s chosen procedure for carrying out a sentence of death as the infliction of cruel and unusual punishment.” Id., at 48. In Wilkerson v. Utah, 99 U. S. 130 –135 (1879), the Court upheld a sentence of death by firing squad. In In re Kemmler, supra, at 447–449, the Court rejected a challenge to the use of the electric chair. And the Court did not retreat from that holding even when presented with a case in which a State’s initial attempt to execute a pris-oner by electrocution was unsuccessful. Louisiana ex rel. Francis v. Resweber, 329 U. S. 459 –464 (1947) (plurality opinion). Most recently, in Baze, supra, seven Justices agreed that the three-drug protocol just discussed does not violate the Eighth Amendment. Our decisions in this area have been animated in part by the recognition that because it is settled that capital punishment is constitutional, “[i]t necessarily follows that there must be a [constitutional] means of carrying it out.” Id., at 47. And because some risk of pain is inherent in any method of execution, we have held that the Constitution does not require the avoidance of all risk of pain. Ibid. After all, while most humans wish to die a painless death, many do not have that good fortune. Holding that the Eighth Amendment demands the elimination of essentially all risk of pain would effectively outlaw the death penalty altogether. B Baze cleared any legal obstacle to use of the most common three-drug protocol that had enabled States to carry out the death penalty in a quick and painless fashion. But a practical obstacle soon emerged, as anti-death-penalty advocates pressured pharmaceutical companies to refuse to supply the drugs used to carry out death sentences. The sole American manufacturer of sodium thiopental, the first drug used in the standard three-drug protocol, was persuaded to cease production of the drug. After suspending domestic production in 2009, the company planned to resume production in Italy. Koppel, Execution Drug Halt Raises Ire of Doctors, Wall Street Journal, Jan. 25, 2011, p. A6. Activists then pressured both the company and the Italian Government to stop the sale of sodium thiopental for use in lethal injections in this country. Bonner, Letter from Europe: Drug Company in Cross Hairs of Death Penalty Opponents, N. Y. Times, Mar. 30, 2011; Koppel, Drug Halt Hinders Executions in the U. S., Wall Street Journal, Jan. 22, 2011, p. A1. That effort proved successful, and in January 2011, the company announced that it would exit the sodium thiopental market entirely. See Hospira, Press Release, Hospira Statement Regarding PentothalTM (sodium thiopental) Market Exit (Jan. 21, 2011). After other efforts to procure sodium thiopental proved unsuccessful, States sought an alternative, and they eventually replaced sodium thiopental with pentobarbital, another barbiturate. In December 2010, Oklahoma became the first State to execute an inmate using pentobarbital. See Reuters, Chicago Tribune, New Drug Mix Used in Oklahoma Execution, Dec. 17 2010, p. 41. That execution occurred without incident, and States gradually shifted to pentobarbital as their supplies of sodium thiopentalran out. It is reported that pentobarbital was used in all of the 43 executions carried out in 2012. The Death Penalty Institute, Execution List 2012, online at www.deathpenaltyinfo.org/execution-list-2012 (all Internet materials as visited June 26, 2015, and available in Clerk of Court’s case file). Petitioners concede that pentobarbital, like sodium thiopental, can “reliably induce and maintain a comalike state that renders a person insensate to pain” caused by administration of the second and third drugs in the protocol. Brief for Petitioners 2. And courts across the country have held that the use of pentobarbital in executions does not violate the Eighth Amendment. See, e.g., Jackson v. Danberg, 656 F. 3d 157 (CA3 2011); Beaty v. Brewer, 649 F. 3d 1071 (CA9 2011); DeYoung v. Owens, 646 F. 3d 1319 (CA11 2011); Pavatt v. Jones, 627 F. 3d 1336 (CA10 2010). Before long, however, pentobarbital also became unavailable. Anti-death-penalty advocates lobbied the Danish manufacturer of the drug to stop selling it for use in executions. See Bonner, supra. That manufacturer opposed the death penalty and took steps to block the shipment of pentobarbital for use in executions in the United States. Stein, New Obstacle to Death Penalty in U. S., Washington Post, July 3, 2011, p. A4. Oklahoma eventually became unable to acquire the drug through any means. The District Court below found that both sodium thiopental and pentobarbital are now unavailable to Oklahoma. App. 67–68. C Unable to acquire either sodium thiopental or pentobarbital, some States have turned to midazolam, a sedative in the benzodiazepine family of drugs. In October 2013, Florida became the first State to substitute midazolam for pentobarbital as part of a three-drug lethal injection protocol. Fernandez, Executions Stall As States Seek Different Drugs, N. Y. Times, Nov. 9, 2013, p. A1. To date, Florida has conducted 11 executions using that protocol, which calls for midazolam followed by a paralytic agent and potassium chloride. See Brief for State of Florida as Amicus Curiae 2–3; Chavez v. Florida SP Warden, 742 F. 3d 1267, 1269 (CA11 2014). In 2014, Oklahoma also substituted midazolam for pentobarbital as part of its three-drug protocol. Oklahoma has already used this three-drug protocol twice: to execute Clayton Lockett in April 2014 and Charles Warner in January 2015. (Warner was one of the four inmates who moved for a preliminary injunction in this case.) The Lockett execution caused Oklahoma to implement new safety precautions as part of its lethal injection protocol. When Oklahoma executed Lockett, its protocol called for the administration of 100 milligrams of midazolam, as compared to the 500 milligrams that are currently required. On the morning of his execution, Lockett cut himself twice at “ ‘the bend of the elbow.’ ” App. 50. That evening, the execution team spent nearly an hour making at least one dozen attempts to establish intravenous (IV) access to Lockett’s cardiovascular system, including at his arms and elsewhere on his body. The team eventually believed that it had established intravenous access through Lockett’s right femoral vein, and it covered the injection access point with a sheet, in part to preserve Lockett’s dignity during the execution. After the team administered the midazolam and a physician determined that Lockett was unconscious, the team next administered the paralytic agent (vecuronium bromide) and most of the potassium chloride. Lockett began to move and speak, at which point the physician lifted the sheet and determined that the IV had “infiltrated,” which means that “the IV fluid, rather than entering Lockett’s blood stream, had leaked into the tissue surrounding the IV access point.” Warner v. Gross, 776 F. 3d 721, 725 (CA10 2015) (case below). The execution team stopped administering the remaining potassium chloride and terminated the execution about 33 minutes after the midazolam was first injected. About 10 minutes later, Lockett was pronounced dead. An investigation into the Lockett execution concluded that “the viability of the IV access point was the single greatest factor that contributed to the difficulty in administering the execution drugs.” App. 398. The investigation, which took five months to complete, recommended several changes to Oklahoma’s execution protocol, and Oklahoma adopted a new protocol with an effective date of September 30, 2014. That protocol allows the Oklahoma Department of Corrections to choose among four different drug combinations. The option that Oklahoma plans to use to execute petitioners calls for the administration of 500 milligrams of midazolam followed by a paralytic agent and potassium chloride.[1] The paralytic agent may be pancuronium bromide, vecuronium bromide, or rocuronium bromide, three drugs that, all agree, are functionally equivalent for purposes of this case. The protocol also includes procedural safeguards to help ensure that an inmate remains insensate to any pain caused by the administration of the paralytic agent and potassium chloride. Those safeguards include: (1) the insertion of both a primary and backup IV catheter, (2) procedures to confirm the viability of the IV site, (3) the option to postpone an execution if viable IV sites cannot be established within an hour, (4) a mandatory pause between administration of the first and second drugs, (5) numerous procedures for monitoring the offender’s consciousness, including the use of an electrocardiograph and direct observation, and (6) detailed provisions with respect to the training and preparation of the execution team. In January of this year, Oklahoma executed Warner using these revised procedures and the combination of midazolam, a paralytic agent, and potassium chloride. II A In June 2014, after Oklahoma switched from pentobarbital to midazolam and executed Lockett, 21 Oklahoma death row inmates filed an action under 42 U. S. C. §1983 challenging the State’s new lethal injection protocol. The complaint alleged that Oklahoma’s use of midazolam violates the Eighth Amendment’s prohibition of cruel and unusual punishment. In November 2014, four of those plaintiffs—Richard Glossip, Benjamin Cole, John Grant, and Warner—filed a motion for a preliminary injunction. All four men had been convicted of murder and sentenced to death by Oklahoma juries. Glossip hired Justin Sneed to kill his employer, Barry Van Treese. Sneed entered a room where Van Treese was sleeping and beat him to death with a baseball bat. See Glossip v. State, 2007 OK CR 12, 157 P. 3d 143, 147–149. Cole murdered his 9-month-old daughter after she would not stop crying. Cole bent her body backwards until he snapped her spine in half. After the child died, Cole played video games. See Cole v. State, 2007 OK CR 27, 164 P. 3d 1089, 1092–1093. Grant, while serving terms of imprisonment totaling 130 years, killed Gay Carter, a prison food service supervisor, by pulling her into a mop closet and stabbing her numerous times with a shank. See Grant v. State, 2002 OK CR 36, 58 P. 3d 783, 789. Warner anally raped and murdered an 11-month-old girl. The child’s injuries included two skull fractures, internal brain injuries, two fractures to her jaw, a lacerated liver, and a bruised spleen and lungs. See Warner v. State, 2006 OK CR 40, 144 P. 3d 838, 856–857. The Oklahoma Court of Criminal Appeals affirmed the murder conviction and death sentence of each offender. Each of the men then unsuccessfully sought both state postconviction and federal habeas corpus relief. Having exhausted the avenues for challenging their convictions and sentences, they moved for a preliminary injunction against Oklahoma’s lethal injection protocol. B In December 2014, after discovery, the District Court held a 3-day evidentiary hearing on the preliminary injunction motion. The District Court heard testimony from 17 witnesses and reviewed numerous exhibits. Dr. David Lubarsky, an anesthesiologist, and Dr. Larry Sasich, a doctor of pharmacy, provided expert testimony about midazolam for petitioners, and Dr. Roswell Evans, adoctor of pharmacy, provided expert testimony forrespondents. After reviewing the evidence, the District Court issued an oral ruling denying the motion for a preliminary injunction. The District Court first rejected petitioners’ challenge under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U. S. 579 (1993) , to the testimony of Dr. Evans. It concluded that Dr. Evans, the Dean of Auburn University’s School of Pharmacy, was well qualified to testify about midazolam’s properties and that he offered reliable testimony. The District Court then held that petitioners failed to establish a likelihood of success on the merits of their claim that the use of midazolam violates the Eighth Amendment. The court provided two independent reasons for this conclusion. First, the court held that petitioners failed to identify a known and available method of execution that presented a substantially less severe risk of pain than the method that the State proposed to use. Second, the court found that petitioners failed to prove that Oklahoma’s protocol “presents a risk that is ‘sure or very likely to cause serious illness and needless suffering,’ amounting to ‘an objectively intolerable risk of harm.’ ” App. 96 (quoting Baze, 553 U. S., at 50). The court emphasized that the Oklahoma protocol featured numerous safeguards, including the establishment of two IV access sites, confirmation of the viability of those sites, and monitoring of the offender’s level of consciousness throughout the procedure. The District Court supported its decision with findings of fact about midazolam. It found that a 500-milligram dose of midazolam “would make it a virtual certainty that any individual will be at a sufficient level of unconsciousness to resist the noxious stimuli which could occur from the application of the second and third drugs.” App. 77. Indeed, it found that a 500-milligram dose alone would likely cause death by respiratory arrest within 30 minutes or an hour. The Court of Appeals for the Tenth Circuit affirmed. 776 F. 3d 721. The Court of Appeals explained that our decision in Baze requires a plaintiff challenging a lethal injection protocol to demonstrate that the risk of severe pain presented by an execution protocol is substantial “ ‘when compared to the known and available alternatives.’ ” Id., at 732 (quoting Baze, supra, at 61). And it agreed with the District Court that petitioners had not identified any such alternative. The Court of Appeals added, however, that this holding was “not outcome-determinative in this case” because petitioners additionally failed to establish that the use of midazolam creates a demonstrated risk of severe pain. 776 F. 3d, at 732. The Court of Appeals found that the District Court did not abuse its discretion by relying on Dr. Evans’ testimony, and it concluded that the District Court’s factual findings about midazolam were not clearly erroneous. It also held that alleged errors in Dr. Evans’ testimony did not render his testimony unreliable or the District Court’s findings clearly erroneous. Oklahoma executed Warner on January 15, 2015, but we subsequently voted to grant review and then stayed the executions of Glossip, Cole, and Grant pending the resolution of this case. 574 U. S. ___ (2015). III “A plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.” Winter v. Natural Resources Defense Council, Inc., 555 U. S. 7, 20 (2008) . The parties agree that this case turns on whether petitioners are able to establish a likelihood of success on the merits. The Eighth Amendment, made applicable to the States through the Fourteenth Amendment, prohibits the infliction of “cruel and unusual punishments.” The controlling opinion in Baze outlined what a prisoner must establish to succeed on an Eighth Amendment method-of-execution claim. Baze involved a challenge by Kentucky death row inmates to that State’s three-drug lethal injection protocol of sodium thiopental, pancuronium bromide, and potassium chloride. The inmates conceded that the protocol, if properly administered, would result in a humane and constitutional execution because sodium thiopental would render an inmate oblivious to any pain caused by the second and third drugs. 553 U. S., at 49. But they argued that there was an unacceptable risk that sodium thiopental would not be properly administered. Ibid. The inmates also maintained that a significant risk of harm could be eliminated if Kentucky adopted a one-drug protocol and additional monitoring by trained personnel. Id., at 51. The controlling opinion in Baze first concluded that prisoners cannot successfully challenge a method of execution unless they establish that the method presents a risk that is “ ‘sure or very likely to cause serious illness and needless suffering,’ and give rise to ‘sufficiently imminent dangers.’ ” Id., at 50 (quoting Helling v. McKinney, 509 U. S. 25 –35 (1993)). To prevail on such a claim, “there must be a ‘substantial risk of serious harm,’ an ‘objectively intolerable risk of harm’ that prevents prison officials from pleading that they were ‘subjectively blameless for purposes of the Eighth Amendment.’ ” 553 U. S., at 50 (quoting Farmer v. Brennan, 511 U. S. 825 , and n. 9 (1994)). The controlling opinion also stated that prisoners “cannot successfully challenge a State’s method of execution merely by showing a slightly or marginally safer alternative.” 553 U. S., at 51. Instead, prisoners must identify an alternative that is “feasible, readily implemented, and in fact significantly reduce[s] a substantial risk of severe pain.” Id., at 52. The controlling opinion summarized the requirements of an Eighth Amendment method-of-execution claim as follows: “A stay of execution may not be granted on grounds such as those asserted here unless the condemned prisoner establishes that the State’s lethal injection protocol creates a demonstrated risk of severe pain. [And] [h]e must show that the risk is substantial when compared to the known and available alternatives.” Id., at 61. The preliminary injunction posture of the present case thus requires petitioners to establish a likelihood that they can establish both that Oklahoma’s lethal injection protocol creates a demonstrated risk of severe pain and that the risk is substantial when compared to the known and available alternatives. The challenge in Baze failed both because the Kentucky inmates did not show that the risks they identified were substantial and imminent, id., at 56, and because they did not establish the existence of a known and available alternative method of execution that would entail a significantly less severe risk, id., at 57–60. Petitioners’ argumentshere fail for similar reasons. First, petitioners have not proved that any risk posed by midazolam is substantial when compared to known and available alternative methods of execution. Second, they have failed to establish that the District Court committed clear error when it found that the use of midazolam will not result in severe pain and suffering. We address each reason in turn. IV Our first ground for affirmance is based on petitioners’ failure to satisfy their burden of establishing that any risk of harm was substantial when compared to a known and available alternative method of execution. In their amended complaint, petitioners proffered that the State could use sodium thiopental as part of a single-drug protocol. They have since suggested that it might also be constitutional for Oklahoma to use pentobarbital. But the District Court found that both sodium thiopental and pentobarbital are now unavailable to Oklahoma’s Department of Corrections. The Court of Appeals affirmed that finding, and it is not clearly erroneous. On the contrary, the record shows that Oklahoma has been unable to procure those drugs despite a good-faith effort to do so. Petitioners do not seriously contest this factual finding, and they have not identified any available drug or drugs that could be used in place of those that Oklahoma is now unable to obtain. Nor have they shown a risk of pain so great that other acceptable, available methods must be used. Instead, they argue that they need not identify a known and available method of execution that presents less risk. But this argument is inconsistent with the controlling opinion in Baze, 553 U. S., at 61, which imposed a requirement that the Court now follows.[2] Petitioners contend that the requirement to identify an alternative method of execution contravenes our pre-Baze decision in Hill v. McDonough, 547 U. S. 573 (2006) , but they misread that decision. The portion of the opinion in Hill on which they rely concerned a question of civil procedure, not a substantive Eighth Amendment question. In Hill, the issue was whether a challenge to a method of execution must be brought by means of an application for a writ of habeas corpus or a civil action under §1983. Id., at 576. We held that a method-of-execution claim must be brought under §1983 because such a claim does not attack the validity of the prisoner’s conviction or death sentence. Id., at 579–580. The United States as amicus curiae argued that we should adopt a special pleading requirement to stop inmates from using §1983 actions to attack, not just a particular means of execution, but the death penalty itself. To achieve this end, the United States proposed that an inmate asserting a method-of-execution claim should be required to plead an acceptable alternative method of execution. Id., at 582. We rejected that argument because “[s]pecific pleading requirements are mandated by the Federal Rules of Civil Procedure, and not, as a general rule, through case-by-case determinations of the federal courts.” Ibid. Hill thus held that §1983 alone does not impose a heightened pleading requirement. Baze, on the other hand, addressed the substantive elements of an Eighth Amendment method-of-execution claim, and it made clear that the Eighth Amendment requires a prisoner to plead and prove a known and available alternative. Because petitioners failed to do this, the District Court properly held that they did not establish a likelihood of success on their Eighth Amendment claim. Readers can judge for themselves how much distance there is between the principal dissent’s argument against requiring prisoners to identify an alternative and the view, now announced by Justices Breyer and Ginsburg, that the death penalty is categorically unconstitutional. Post, p. ___ (Breyer, J., dissenting). The principal dissent goes out of its way to suggest that a State would violate the Eighth Amendment if it used one of the methods of execution employed before the advent of lethal injection. Post, at 30–31. And the principal dissent makes this suggestion even though the Court held in Wilkerson that this method (the firing squad) is constitutional and even though, in the words of the principal dissent, “there is some reason to think that it is relatively quick and painless.” Post, at 30. Tellingly silent about the methods of execution most commonly used before States switched to lethal injection (the electric chair and gas chamber), the principal dissent implies that it would be unconstitutional to use a method that “could be seen as a devolution to a more primitive era.” Ibid. If States cannot return to any of the “more primitive” methods used in the past and if no drug that meets with the principal dissent’s approval is available for use in carrying out a death sentence, the logical conclusion is clear. But we have time and again reaffirmed that capital punishment is not per se unconstitutional. See, e.g., Baze, 553 U. S., at 47; id., at 87–88 (Scalia, J., concurring in judgment); Gregg, 428 U. S., at 187 (joint opinion of Stewart, Powell, and Stevens, JJ.); id., at 226 (White, J., concurring in judgment); Resweber, 329 U. S., at 464; In re Kemmler, 136 U. S., at 447; Wilkerson, 99 U. S., at 134–135. We decline to effectively overrule these decisions. V We also affirm for a second reason: The District Court did not commit clear error when it found that midazolam is highly likely to render a person unable to feel pain during an execution. We emphasize four points at the outset of our analysis. First, we review the District Court’s factual findings under the deferential “clear error” standard. This standard does not entitle us to overturn a finding “simply because [we are] convinced that [we] would have decided the case differently.” Anderson v. Bessemer City, 470 U. S. 564, 573 (1985) . Second, petitioners bear the burden of persuasion on this issue. Baze, supra, at 41. Although petitionersexpend great effort attacking peripheral aspects of Dr. Evans’ testimony, they make little attempt to prove what is critical, i.e., that the evidence they presented to the District Court establishes that the use of midazolam is sure or very likely to result in needless suffering. Third, numerous courts have concluded that the use of midazolam as the first drug in a three-drug protocol is likely to render an inmate insensate to pain that might result from administration of the paralytic agent and potassium chloride. See, e.g., 776 F. 3d 721 (case below affirming the District Court); Chavez v. Florida SP Warden, 742 F. 3d 1267 (affirming the District Court); Banks v. State, 150 So. 3d 797 (Fla. 2014) (affirming the lower court); Howell v. State, 133 So. 3d 511 (Fla. 2014) (same); Muhammad v. State, 132 So. 3d 176 (Fla. 2013) (same). (It is noteworthy that one or both of the two key witnesses in this case—Dr. Lubarsky for petitioners and Dr. Evans for respondents—were witnesses in the Chavez, Howell, and Muhammad cases.) “Where an intermediate court reviews, and affirms, a trial court’s factual findings, this Court will not ‘lightly overturn’ the concurrent findings of the two lower courts.” Easley v. Cromartie, 532 U. S. 234, 242 (2001) . Our review is even more deferential where, as here, multiple trial courts have reached the same finding, and multiple appellate courts have affirmed those findings. Cf. Exxon Co., U. S. A. v. Sofec, Inc., 517 U. S. 830, 841 (1996) (explaining that this Court “ ‘cannot undertake to review concurrent findings of fact by two courts below in the absence of a very obvious and exceptional showing of error’ ” (quoting Graver Tank & Mfg. Co. v. Linde Air Products Co., 336 U. S. 271, 275 (1949) )). Fourth, challenges to lethal injection protocols test the boundaries of the authority and competency of federal courts. Although we must invalidate a lethal injection protocol if it violates the Eighth Amendment, federal courts should not “embroil [themselves] in ongoing scientific controversies beyond their expertise.” Baze, supra, at 51. Accordingly, an inmate challenging a protocol bears the burden to show, based on evidence presented to the court, that there is a substantial risk of severe pain. A Petitioners attack the District Court’s findings of fact on two main grounds.[3] First, they argue that even if midazolam is powerful enough to induce unconsciousness, it is too weak to maintain unconsciousness and insensitivity to pain once the second and third drugs are administered. Second, while conceding that the 500-milligram dose of midazolam is much higher than the normal therapeutic dose, they contend that this fact is irrelevant because midazolam has a “ceiling effect”—that is, at a certain point, an increase in the dose administered will not have any greater effect on the inmate. Neither argumentsucceeds. The District Court found that midazolam is capable of placing a person “at a sufficient level of unconsciousness to resist the noxious stimuli which could occur from the application of the second and third drugs.” App. 77. This conclusion was not clearly erroneous. Respondents’ expert, Dr. Evans, testified that the proper administration of a 500-milligram dose of midazolam would make it “a virtual certainty” that any individual would be “at a sufficient level of unconsciousness to resist the noxious stimuli which could occur from application of the 2nd and 3rd drugs” used in the Oklahoma protocol. Id., at 302; see also id., at 322. And petitioners’ experts acknowledged that they had no contrary scientific proof. See id., at 243–244 (Dr. Sasich stating that the ability of midazolam to render a person insensate to the second and third drugs “has not been subjected to scientific testing”); id., at 176 (Dr. Lubarksy stating that “there is no scientific literature addressing the use of midazolam as a manner to administer lethal injections in humans”). In an effort to explain this dearth of evidence, Dr. Sasich testified that “[i]t’s not my responsibility or the [Food and Drug Administration’s] responsibility to prove that the drug doesn’t work or is not safe.” Tr. of Preliminary Injunction Hearing 357 (Tr.). Instead, he stated, “it’s the responsibility of the proponent to show that the drug is safe and effective.” Ibid. Dr. Sasich confused the standard imposed on a drug manufacturer seeking approval of a therapeutic drug with the standard that must be borne by a party challenging a State’s lethal injection protocol. When a method of execution is authorized under state law, a party contending that this method violates the Eighth Amendment bears the burden of showing that the method creates an unacceptable risk of pain. Here, petitioners’ own experts effectively conceded that they lacked evidence to prove their case beyond dispute. Petitioners attempt to avoid this deficiency by criticizing respondents’ expert. They argue that the District Court should not have credited Dr. Evans’ testimony because he admitted that his findings were based on “ ‘extrapolat[ions]’ ” from studies done about much lower therapeutic doses of midazolam. See Brief for Petitioners 34 (citing Tr. 667–668; emphasis deleted). But because a 500-milligram dose is never administered for a therapeutic purpose, extrapolation was reasonable. And the conclusions of petitioners’ experts were also based on extrapolations and assumptions. For example, Dr. Lubarsky relied on “extrapolation of the ceiling effect data.” App. 177. Based on the evidence that the parties presented to the District Court, we must affirm. Testimony from both sides supports the District Court’s conclusion that midazolam can render a person insensate to pain. Dr. Evans testified that although midazolam is not an analgesic, it can nonetheless “render the person unconscious and ‘insensate’ during the remainder of the procedure.” Id., at 294. In his discussion about the ceiling effect, Dr. Sasich agreed that as the dose of midazolam increases, it is “expected to produce sedation, amnesia, and finally lack of response to stimuli such as pain (unconsciousness).” Id., at 243. Petitioners argue that midazolam is not powerful enough to keep a person insensate to pain after the administration of the second and third drugs, but Dr. Evans presented creditable testimony to the contrary. See, e.g., Tr. 661 (testifying that a 500-milligram dose of midazolam will induce a coma).[4] Indeed, low doses of midazolam are sufficient to induce unconsciousness and are even some- times used as the sole relevant drug in certain medical procedures. Dr. Sasich conceded, for example, that midazolam might be used for medical procedures like colonoscopies and gastroscopies. App. 267–268; see also Brief for Respondents 6–8.[5] Petitioners emphasize that midazolam is not recommended or approved for use as the sole anesthetic during painful surgery, but there are two reasons why this is not dispositive. First, as the District Court found, the 500-milligram dose at issue here “is many times higher than a normal therapeutic dose of midazolam.” App. 76. The effect of a small dose of midazolam has minimal probative value about the effect of a 500-milligram dose. Second, the fact that a low dose of midazolam is not the best drug for maintaining unconsciousness during surgery says little about whether a 500-milligram dose of midazolam is constitutionally adequate for purposes of conducting an execution. We recognized this point in Baze, where we concluded that although the medical standard of care might require the use of a blood pressure cuff and an electrocardiogram during surgeries, this does not mean those procedures are required for an execution to pass Eighth Amendment scrutiny. 553 U. S., at 60. Oklahoma has also adopted important safeguards to ensure that midazolam is properly administered. The District Court emphasized three requirements in particular: The execution team must secure both a primary and backup IV access site, it must confirm the viability of the IV sites, and it must continuously monitor the offender’s level of consciousness. The District Court did not commit clear error in concluding that these safeguards help to minimize any risk that might occur in the event that midazolam does not operate as intended. Indeed, we concluded in Baze that many of the safeguards that Oklahoma employs—including the establishment of a primary and backup IV and the presence of personnel to monitor an inmate—help in significantly reducing the risk that an execution protocol will violate the Eighth Amendment. Id., at 55–56. And many other safeguards that Oklahoma has adopted mirror those that the dissent in Baze complained were absent from Kentucky’s protocol in that case. For example, the dissent argued that because a consciousness check before injection of the second drug “can reduce a risk of dreadful pain,” Kentucky’s failure to include that step in its procedure was unconstitutional. Id., at 119 (opinion of Ginsburg, J.). The dissent also complained that Kentucky did not monitor the effectiveness of the first drug or pause between injection of the first and second drugs. Id., at 120–121. Oklahoma has accommodated each of those concerns. B Petitioners assert that midazolam’s “ceiling effect” undermines the District Court’s finding about the effectiveness of the huge dose administered in the Oklahoma protocol. Petitioners argue that midazolam has a “ceiling” above which any increase in dosage produces no effect. As a result, they maintain, it is wrong to assume that a 500-milligram dose has a much greater effect than a therapeutic dose of about 5 milligrams. But the mere fact that midazolam has such a ceiling cannot be dispositive. Dr. Sasich testified that “all drugs essentially have a ceiling effect.” Tr. 343. The relevant question here is whether midazolam’s ceiling effect occurs below the level of a 500-milligram dose and at a point at which the drug does not have the effect of rendering a person insensate to pain caused by the second and third drugs. Petitioners provided little probative evidence on this point, and the speculative evidence that they did present to the District Court does not come close to establishing that its factual findings were clearly erroneous. Dr. Sasich stated in his expert report that the literature “indicates” that midazolam has a ceiling effect, but he conceded that he “was unable to determine the midazolam dose for a ceiling effect on unconsciousness because there is no literature in which such testing has been done.” App. 243–244. Dr. Lubarsky’s report was similar, id., at 171–172, and the testimony of petitioners’ experts at the hearing was no more compelling. Dr. Sasich frankly admitted that he did a “search to try and determine at what dose of midazolam you would get a ceiling effect,” but concluded: “I could not find one.” Tr. 344. The closest petitioners came was Dr. Lubarsky’s suggestion that the ceiling effect occurs “[p]robably after about . . . 40 to 50 milligrams,” but he added that he had not actually done the relevant calculations, and he admitted: “I can’t tell you right now” at what dose the ceiling effect occurs. App. 225. We cannot conclude that the District Court committed clear error in declining to find, based on such speculative evidence, that the ceiling effect negates midazolam’s ability to render an inmate insensate to pain caused by the second and third drugs in the protocol. The principal dissent discusses the ceiling effect at length, but it studiously avoids suggesting that petitioners presented probative evidence about the dose at which the ceiling effect occurs or about whether the effect occurs before a person becomes insensate to pain. The principal dissent avoids these critical issues by suggesting that such evidence is “irrelevant if there is no dose at which the drug can . . . render a person ‘insensate to pain.’ ” Post, at 17. But the District Court heard evidence that the drug can render a person insensate to pain, and not just from Dr. Evans: Dr. Sasich (one of petitioners’ own experts) testified that higher doses of midazolam are “expected to produce . . . lack of response to stimuli such as pain.” App. 243.[6] In their brief, petitioners attempt to deflect attention from their failure of proof regarding midazolam’s ceiling effect by criticizing Dr. Evans’ testimony. But it was petitioners’ burden to establish that midazolam’s ceiling occurred at a dosage below the massive 500-milligram dose employed in the Oklahoma protocol and at a point at which the drug failed to render the recipient insensate to pain. They did not meet that burden, and their criticisms do not undermine Dr. Evans’ central point, which the District Court credited, that a properly administered 500-milligram dose of midazolam will render the recipient unable to feel pain. One of petitioners’ criticisms of Dr. Evans’ testimony is little more than a quibble about the wording chosen by Dr. Evans at one point in his oral testimony. Petitioners’ expert, Dr. Lubarsky, stated in his report that midazolam “increases effective binding of [gamma-aminobutyric acid (GABA)] to its receptor to induce unconsciousness.”[7] App. 172. Dr. Evans’ report provided a similar explanation of the way in which midazolam works, see id., at 293–294, and Dr. Lubarsky did not dispute the accuracy of that explanation when he testified at the hearing. Petitioners contend, however, that Dr. Evans erred when he said at the hearing that “[m]idazolam attaches to GABA receptors, inhibiting GABA.” Id., at 312 (emphasis added). Petitioners contend that this statement was incorrect because “far from inhibiting GABA, midazolam facilitates its binding to GABA receptors.” Brief for Petitioners 38. In making this argument, petitioners are simply quarrelling with the words that Dr. Evans used during oral testimony in an effort to explain how midazolam works in terms understandable to a layman. Petitioners do not suggest that the discussion of midazolam in Dr. Evans’ expert report was inaccurate, and as for Dr. Evans’ passing use of the term “inhibiting,” Dr. Lubarksy’s own expert report states that GABA’s “inhibition of brain activity is accentuated by midazolam.” App. 232 (emphasis added). Dr. Evans’ oral use of the word “inhibiting”—particularly in light of his written testimony—does not invalidate the District Court’s decision to rely on his testimony. Petitioners also point to an apparent conflict between Dr. Evans’ testimony and a declaration by Dr. Lubarsky (submitted after the District Court ruled) regarding the biological process that produces midazolam’s ceiling effect. But even if Dr. Lubarsky’s declaration is correct, it is largely beside the point. What matters for present purposes is the dosage at which the ceiling effect kicks in, not the biological process that produces the effect. And Dr. Lubarsky’s declaration does not render the District Court’s findings clearly erroneous with respect to that critical issue. C Petitioners’ remaining arguments about midazolam all lack merit. First, we are not persuaded by petitioners’ argument that Dr. Evans’ testimony should have been rejected because of some of the sources listed in his report. Petitioners criticize two of the “selected references” that Dr. Evans cited in his expert report: the Web site drugs.com and a material safety data sheet (MSDS) about midazolam. Petitioners’ argument is more of a Daubert challenge to Dr. Evans’ testimony than an argument that the District Court’s findings were clearly erroneous. The District Court concluded that Dr. Evans was “well-qualified to give the expert testimony that he gave” and that “his testimony was the product of reliable principles and methods reliably applied to the facts of this case.” App. 75–76. To the extent that the reliability of Dr.Evans’ testimony is even before us, the District Court’s con-clusion that his testimony was based on reliable sources is reviewed under the deferential “abuse-of-discretion” standard. General Elec. Co. v. Joiner, 522 U. S. 136 –143 (1997). Dr. Evans relied on multiple sources and his own expertise, and his testimony may not be disqualified simply because one source (drugs.com) warns that it “ ‘is not intended for medical advice’ ” and another (the MSDS) states that its information is provided “ ‘without any warranty, express or implied, regarding its correctness.’ ” Brief for Petitioners 36. Medical journals that both parties rely upon typically contain similar disclaimers. See, e.g., Anesthesiology, Terms and Conditions of Use, online at http://anesthesiology.pubs.asahq.org/ss/terms.aspx (“None of the information on this Site shall be used to diagnose or treat any health problem or disease”). Dr. Lubarsky—petitioners’ own expert—relied on an MSDS to argue that midazolam has a ceiling effect. And petitioners do not identify any incorrect statements from drugs.com on which Dr. Evans relied. In fact, although Dr. Sasich submitted a declaration to the Court of Appeals criticizing Dr. Evans’ reference to drugs.com, that declaration does not identify a single fact from that site’s discussion of midazolam that was materially inaccurate. Second, petitioners argue that Dr. Evans’ expert report contained a mathematical error, but we find this argument insignificant. Dr. Evans stated in his expert report that the lowest dose of midazolam resulting in human deaths, according to an MSDS, is 0.071 mg/kg delivered intravenously. App. 294. Dr. Lubarsky agreed with this statement. Specifically, he testified that fatalities have occurred in doses ranging from 0.04 to 0.07 mg/kg, and he stated that Dr. Evans’ testimony to that effect was “a true statement” (though he added those fatalities occurred among the elderly). Id., at 217. We do not understand petitioners to dispute the testimony of Dr. Evans and their own expert that 0.071 mg/kg is a potentially fatal dose of midazolam. Instead, they make much of the fact that the MSDS attached to Dr. Evans’ report apparently contained a typographical error and reported the lowest toxic dose as 71 mg/kg. That Dr. Evans did not repeat that incorrect figure but instead reported the correct dose supports rather than undermines his testimony. In any event, the alleged error in the MSDS is irrelevant because the District Court expressly stated that it did not rely on the figure in the MSDS. See id., at 75. Third, petitioners argue that there is no consensus among the States regarding midazolam’s efficacy because only four States (Oklahoma, Arizona, Florida, and Ohio) have used midazolam as part of an execution. Petitioners rely on the plurality’s statement in Baze that “it is difficult to regard a practice as ‘objectively intolerable’ when it is in fact widely tolerated,” and the plurality’s emphasis on the fact that 36 States had adopted lethal injection and 30 States used the particular three-drug protocol at issue in that case. 553 U. S., at 53. But while the near-universal use of the particular protocol at issue in Baze supported our conclusion that this protocol did not violate the Eighth Amendment, we did not say that the converse was true, i.e., that other protocols or methods of execution are of doubtful constitutionality. That argument, if accepted, would hamper the adoption of new and potentially more humane methods of execution and would prevent States from adapting to changes in the availability of suitable drugs. Fourth, petitioners argue that difficulties with Oklahoma’s execution of Lockett and Arizona’s July 2014 execution of Joseph Wood establish that midazolam is sure or very likely to cause serious pain. We are not persuaded. Aside from the Lockett execution, 12 other executions have been conducted using the three-drug protocol at issue here, and those appear to have been conducted without any significant problems. See Brief for Respondents 32; Brief for State of Florida as Amicus Curiae 1. Moreover, Lockett was administered only 100 milligrams of midazolam, and Oklahoma’s investigation into that execution concluded that the difficulties were due primarily to the execution team’s inability to obtain an IV access site. And the Wood execution did not involve the protocol at issue here. Wood did not receive a single dose of 500 milligrams of midazolam; instead, he received fifteen 50-milligram doses over the span of two hours.[8] Brief for Respondents 12, n. 9. And Arizona used a different two-drug protocol that paired midazolam with hydromorphone, a drug that is not at issue in this case. Ibid. When all of the circumstances are considered, the Lockett and Wood executions have little probative value for present purposes. Finally, we find it appropriate to respond to the principal dissent’s groundless suggestion that our decision is tantamount to allowing prisoners to be “drawn and quartered, slowly tortured to death, or actually burned at the stake.” Post, at 28. That is simply not true, and the principal dissent’s resort to this outlandish rhetoric reveals the weakness of its legal arguments. VI For these reasons, the judgment of the Court of Appeals for the Tenth Circuit is affirmed. It is so ordered.Notes 1 The three other drug combinations that Oklahoma may admin-ister are: (1) a single dose of pentobarbital, (2) a single dose ofsodium thiopental, and (3) a dose of midazolam followed by a dose of hydromorphone. 2 Justice Sotomayor’s dissent (hereinafter principal dissent), post, at 24–25, inexplicably refuses to recognize that The Chief Justice’s opinion in Baze sets out the holding of the case. In Baze, the opinion of The Chief Justice was joined by two other Justices. Justices Scalia and Thomas took the broader position that a method of execution is consistent with the Eighth Amendment unless it is deliberately designed to inflict pain. 553 U. S., at 94 (Thomas, J. concurring in judgment). Thus, as explained in Marks v. United States, 430 U. S. 188, 193 (1977) , The Chief Justice’s opinion sets out the holding of the case. It is for this reason that petitioners base their argument on the rule set out in that opinion. See Brief for Petitioners 25, 28. 3 Drs. Lubarsky and Sasich, petitioners’ key witnesses, both testified that midazolam is inappropriate for a third reason, namely, that it creates a risk of “paradoxical reactions” such as agitation, hyperactiv-ity, and combativeness. App. 175 (expert report of Dr. Lubarsky); id., at 242, 244 (expert report of Dr. Sasich). The District Court found, however, that the frequency with which a paradoxical reaction occurs “is speculative” and that the risk “occurs with the highest frequency in low therapeutic doses.” Id., at 78. Indeed, Dr. Sasich conceded that the incidence or risk of paradoxical reactions with midazolam “is unknown” and that reports estimate the risk to vary only “from 1% to above 10%.” Id., at 244. Moreover, the mere fact that a method of execution might result in some unintended side effects does not amount to an Eighth Amendment violation. “[T]he Constitution does not demand the avoidance of all risk of pain.” Baze, 553 U. S., at 47 (plurality opinion). 4 The principal dissent misunderstands the record when it bizarrely suggests that midazolam is about as dangerous as a peanut. Post, at 15. Dr. Evans and Dr. Lubarsky agreed that midazolam has caused fatalities in doses as low as 0.04 to 0.07 milligrams per kilogram. App. 217, 294. Even if death from such low doses is a “rare, unfortunate side effec[t],” post, at 15, the District Court found that a massive 500-milligram dose—many times higher than the lowest dose reported to have produced death—will likely cause death in under an hour. App. 76–77. 5 Petitioners’ experts also declined to testify that a 500-milligram dose of midazolam is always insufficient to place a person in a coma and render him insensate to pain. Dr. Lubarsky argued only that the 500-milligram dose cannot “reliably” produce a coma. Id., 228. And when Dr. Sasich was asked whether he could say to a reasonable degree of certainty that a 500-milligram dose of midazolam would not render someone unconscious, he replied that he could not. Id., at 271–272. A product label for midazolam that Dr. Sasich attached to his expert report also acknowledged that an overdose of midazolam can cause a coma. See Expert Report of Larry D. Sasich, in No. 14–6244 (CA10), p. 34. 6 The principal dissent emphasizes Dr. Lubarsky’s supposedly contrary testimony, but the District Court was entitled to credit Dr. Evans (and Dr. Sasich) instead of Dr. Lubarsky on this point. And the District Court had strong reasons not to credit Dr. Lubarsky, who even argued that a protocol that includes sodium thiopental is “constructed to produce egregious harm and suffering.” App. 227. 7 GABA is “an amino acid that functions as an inhibitory neurotransmitter in the brain and spinal cord.” Mosby’s Medical Dictionary 8 The principal dissent emphasizes Dr. Lubarsky’s testimony that it is irrelevant that Wood was administered the drug over a 2-hour period. Post, at 20. But Dr. Evans disagreed and testified that if a 750-milligram dose “was spread out over a long period of time,” such as one hour (i.e., half the time at issue in the Wood execution), the drug might not be as effective as if it were administered all at once. Tr. 667. The principal dissent states that this “pronouncement was entirely unsupported,” post, at 20, n. 6, but it was supported by Dr. Evans’ expertise and decades of experience. And it would be unusual for an expert testifying on the stand to punctuate each sentence with citation to a

574.US.418

Petitioner, Hana Financial, Inc., and respondent Hana Bank both provide financial services to individuals in the United States. When Hana Financial sued Hana Bank for trademark infringement, Hana Bank invoked in defense the tacking doctrine, under which lower courts have provided that a trademark user may make certain modifications to its mark over time while, in limited circumstances, retaining its priority position. Petitioner’s claim was tried before a jury, and the District Court adopted in substantial part the jury instruction on tacking proposed by petitioner. The jury returned a verdict in respondent’s favor. Affirming, the Ninth Circuit explained that the tacking inquiry was an exceptionally limited and highly fact-sensitive matter reserved for juries, not judges. Held: Whether two trademarks may be tacked for purposes of determining priority is a question for the jury. Pp. 3–8. (a) Lower courts have held that two marks may be tacked when they are considered to be “legal equivalents,” i.e., they “create the same, continuing commercial impression.” Van Dyne-Crotty, Inc. v. Wear-Guard Corp. 926 F.2d 1156, 1159. And “commercial impression” “must be viewed through the eyes of a consumer.” DuoProSS Meditech Corp. v. Inviro Medical Devices, Ltd., 695 F.3d 1247, 1253. When the relevant question is how an ordinary person or community would make an assessment, the jury is generally the decisionmaker that ought to provide the fact-intensive answer. See, e.g., United States v. Gaudin, 515 U.S. 506, 512. Pp. 3–5. (b) Each of petitioner’s four arguments in support of its view that tacking is a question of law to be resolved by a judge is unpersuasive. First, it may be true that the “legal equivalents” test involves a legal standard, but such “ ‘mixed question[s] of law and fact,’ [have] typi-cally been resolved by juries.” Gaudin, 515 U. S., at 512. And any concern that a jury may improperly apply the relevant legal standard can be remedied by crafting careful jury instructions. Second, petitioner offers no support for its claim that tacking determinations create new law in a unique way that requires those determinations to be reserved for judges. Third, petitioner worries that the predictability required for a functioning trademark system will be absent if tacking questions are assigned to juries, but offers no reason why trademark tacking should be treated differently from the tort, contract, and criminal justice systems, where juries answer often-dispositive factual questions or make dispositive applications of legal standards to facts. Finally, in arguing that judges have historically resolved tacking disputes, petitioner points to cases arising in the contexts of bench trials, summary judgment, and the like, in which it is undisputed that judges may resolve tacking disputes. Pp. 5–8. 735 F. 3d. 1158, affirmed. Sotomayor, J., delivered the opinion for a unanimous Court.

Rights in a trademark are determined by the date of the mark’s first use in commerce. The party who first uses a mark in commerce is said to have priority over other users. Recognizing that trademark users ought to be permitted to make certain modifications to their marks over time without losing priority, lower courts have provided that, in limited circumstances, a party may clothe a new mark with the priority position of an older mark. This doctrine is called “tacking,” and lower courts have found tacking to be available when the original and revised marks are “legal equivalents” in that they create the same, continuing commercial impression. The question presented here is whether a judge or a jury should determine whether tacking is available in a given case. Because the tacking inquiry operates from the perspective of an ordinary purchaser or consumer, we hold that a jury should make this determination.I Petitioner, Hana Financial, and respondent Hana Bank, a subsidiary of respondent Hana Financial Group, both provide financial services to individuals in the United States. Hana Bank (hereinafter respondent) was established in 1971 as a Korean entity called Korea Investment Finance Corporation. In 1991, that entity changed its name to “Hana Bank” and began using this name in Korea. In 1994, it established a service called Hana Overseas Korean Club to provide financial services to Korean expatriates, and specifically advertised that service in the United States. Those advertisements used the name “Hana Overseas Korean Club” in both English and Korean, and included the name “Hana Bank” in Korean and respondent’s “dancing man” logo. See App. 206. In 2000, respondent changed the name of the Hana Overseas Ko-rean Club to “Hana World Center.” In 2002, respondent began operating a bank in the United States under the name “Hana Bank.” This enterprise amounted to respondent’s first physical presence in the United States. Petitioner was established in 1994 as a California corporation called Hana Financial. It began using that name and an associated trademark in commerce in 1995. In 1996, it obtained a federal trademark registration for a pyramid logo with the name “Hana Financial” for use in connection with financial services. In 2007, petitioner sued respondent, alleging infringement of its “Hana Financial” mark. As relevant here, respondent denied infringement by invoking the tacking doctrine and claiming that it had priority. The District Court initially granted summary judgment to respondent on the infringement claim, but the Court of Appeals for the Ninth Circuit reversed, holding that there were genuine issues of material fact as to priority. On remand, the infringement claim was tried before a jury. The District Court adopted in substantial part the jury instruction proposed by petitioner, and, without objection from petitioner, instructed the jury as follows:“A party may claim priority in a mark based on the first use date of a similar but technically distinct mark where the previously used mark is the legal equivalent of the mark in question or indistinguish-able therefrom such that consumers consider both as the same mark. This is called ‘tacking.’ The marks must create the same, continuing commercial impression, and the later mark should not materially differ from or alter the character of the mark attemptedto be tacked.” App. 173; see id., at 140 (proposedinstruction).The jury returned a verdict in favor of respondent, and the District Court denied petitioner’s motion for judgment as a matter of law. The Court of Appeals for the Ninth Circuit affirmed. The court explained that, although tacking applies only in “exceptionally narrow circumstances,” 735 F. 3d 1158, 1160 (2013) (internal quotation marks omitted), it “ ‘requires a highly fact-sensitive inquiry’ ” that is “reserved for the jury,” ibid. (quoting One Industries, LLC v. Jim O’Neal Distributing, Inc., 578 F. 3d 1154, 1160 (CA9 2009)). The court acknowledged, however, that whether tacking should be decided by juries or judges “is the subject of a circuit split.” 735 F. 3d, at 1164, n. 5 (noting that the Federal and Sixth Circuits “evaluate tacking as a question of law”); see Van Dyne-Crotty, Inc. v. Wear-Guard Corp., 926 F. 2d 1156, 1159 (CA Fed. 1991); Data Concepts, Inc. v. Digital Consulting, Inc., 150 F. 3d 620, 623 (CA6 1998). We granted certiorari, 573 U. S. ___ (2014), and now affirm.II As discussed above, the general rule adopted by lower courts has been that two marks may be tacked when the original and revised marks are “legal equivalents.” This term refers to two marks that “create the same, continuing commercial impression” so that consumers “consider both as the same mark.”[1] Van Dyne-Crotty, Inc., 926 F. 2d, at 1159 (internal quotation marks omitted); see, e.g., George & Co., LLC v. Imagination Entertainment Ltd., 575 F. 3d 383, 402 (CA4 2009); Brookfield Communications, Inc. v. West Coast Entertainment Corp., 174 F. 3d 1036, 1047–1048 (CA9 1999); Data Concepts, Inc., 150 F. 3d, at 623. “The commercial impression that a mark conveys must be viewed through the eyes of a consumer.” DuoProSS Medi-tech Corp. v. Inviro Medical Devices, Ltd., 695 F. 3d 1247, 1253 (CA Fed. 2012); see 3 J. McCarthy, Trademarks and Unfair Competition §17:26, p. 17–71 (4th ed. 2014) (“ ‘Commercial impression,’ like most issues in trademark law, should be determined from the perspective of the ordinary purchaser of these kinds of goods or services”). Application of a test that relies upon an ordinary consumer’s understanding of the impression that a mark conveys falls comfortably within the ken of a jury. Indeed, we have long recognized across a variety of doctrinal contexts that, when the relevant question is how an ordinary person or community would make an assessment, the jury is generally the decisionmaker that ought to provide the fact-intensive answer. See, e.g., United States v. Gaudin,515 U. S. 506,512 (1995) (recognizing that “ ‘delicate assessments of the inferences a ‘reasonable [decisionmaker]’ would draw . . . [are] peculiarly one[s] for the trier of fact’ ” (quoting TSC Industries, Inc. v. Northway, Inc.,426 U. S. 438,450 (1976); first alteration in original); id., at 450, n. 12 (observing that the jury has a “unique competence in applying the ‘reasonable man’ standard”); Hamling v. United States,418 U. S. 87–105 (1974) (emphasizing “the ability of the juror to ascertain the sense of the ‘average person’ ” by drawing upon “his own knowledge of the views of the average person in thecommunity or vicinage from which he comes” and his “knowledge of the propensities of a ‘reasonable’ person”); Railroad Co. v. Stout, 17 Wall. 657, 664 (1874) (“It is assumed that twelve men know more of the common affairs of life than does one man, [and] that they can draw wiser and safer conclusions from admitted facts thus occurring than can a single judge”). This is certainly not to say that a judge may never determine whether two marks may be tacked. If the facts warrant it, a judge may decide a tacking question on a motion for summary judgment or for judgment as a matter of law. See Fed. Rules Civ. Proc. 50, 56(a). And if the parties have opted to try their case before a judge, the judge may of course decide a tacking question in his or her factfinding capacity. We hold only that, when a jury trial has been requested and when the facts do not warrant entry of summary judgment or judgment as a matter of law, the question whether tacking is warranted must be decided by a jury.III Attempting to overcome our conclusion, petitioner offers four reasons why, in its view, tacking is a question of law that should be resolved by a judge. None persuades us. Petitioner first observes that the “legal equivalents” test involves the application of a legal standard. See Brief for Petitioner 20. True enough, but “the application-of-legal-standard-to-fact sort of question . . . , commonly called a ‘mixed question of law and fact,’ has typically been resolved by juries.” Gaudin, 515 U. S., at 512; see id., at 514 (“[T]he jury’s constitutional responsibility is not merely to determine the facts, but to apply the law to those facts and draw the ultimate conclusion . . .”); Miller v. Fenton,474 U. S. 104,113 (1985) (“[A]n issue does not lose its factual character merely because its resolution is dispositive of the ultimate . . . question”). The “mixed” analysis that takes place during the tacking inquiry is no different. And insofar as petitioner is concerned that a jury may improperly apply the relevant legal standard, the solution is to craft careful jury instructions that make that standard clear. Here, however, petitioner can hardly criticize the instruction the District Court gave the jury, as it was essentially the instruction petitioner proposed. Second, petitioner argues that tacking determinations will “create new law that will guide future tacking disputes”—a task reserved for judges. Brief for Petitioner 21. It is not at all clear, however, why a tacking determination in a particular case will “create new law” any more than will a jury verdict in a tort case, a contract dispute, or a criminal proceeding. Petitioner insists that tacking questions “have to be” resolved by comparing two marks in a given case “against those addressed in other tacking cases,” id., at 22, but we do not agree. Of course, in deciding summary judgment motions, or in making rulings in bench trials, judges may look to past cases holding that trademark owners either were or were not entitled to tacking as a matter of law. But petitioner offers no support for the claim that tacking cases “have to be” resolved by reliance on precedent. Indeed, in many of the cases petitioner cites in support of this argument, the courts in question relied on precedent only to define the relevant legal standard. See, e.g., Specht v. Google Inc., 758 F. Supp. 2d 570, 583–585 (ND Ill. 2010), aff’d, 747 F. 3d 929 (CA7 2014); Children’s Legal Servs. PLLC v. Kresch, 2008 WL 1901245, *1–*2 (ED Mich., Apr. 25, 2008), aff ’d sub nom. Children’s Legal Servs., P. L. L. C. v. Saiontz, Kirk & Miles, P. A., 2009 WL 1868809 (CA6, June 18, 2009).[2] Third, and related, petitioner worries that the predict-ability required for a functioning trademark system will be absent if tacking questions are assigned to juries. See Brief for Petitioner 25–27. But, again, the same could be said about the tort, contract, and criminal justice systems: In all of these areas, juries answer often-dispositive fac-tual questions or make dispositive applications of legal standards to facts. The fact that another jury, hearing the same case, might reach a different conclusion may make the system “unpredictable,” but it has never stopped us from employing juries in these analogous contexts. Petitioner has offered no reason why trademark tacking ought to be treated differently. Moreover, decisionmaking in fact-intensive disputes necessarily requires judgment calls. Regardless of whether those judgment calls are made by juries or judges, they necessarily involve some degree of uncertainty, particularly when they have to do with how reasonable persons would behave. Finally, petitioner argues that, as a historical matter, judges have resolved tacking disputes. See Brief for Petitioner 30–35. But petitioner relies on cases in which judges have resolved tacking disputes in bench trials, at summary judgment, or the like. See, e.g., Drexel Enterprises, Inc. v. Richardson, 312 F. 2d 525, 526 (CA10 1962) (“[This action] was tried without a jury”); Perfectform Corp. v. Perfect Brassiere Co., 256 F. 2d 736, 738 (CA3 1958) (“The district court dismissed the complaint”); John Morrell & Co. v. Hauser Packing Co., 20 F. 2d 713 (CA9 1927) (“In the court below, there was a dismissal of both the bill and of defendant’s counterclaim”); Beech-Nut Packing Co. v. P. Lorillard Co., 299 F. 834, 835 (NJ 1924) (equitable claims tried solely before a judge). As we have noted, it is undisputed that judges may resolve tacking disputes in those contexts. But recognizing as much does not gainsay our conclusion that, when a jury is to be empaneled and when the facts warrant neither summary judgment nor judgment as a matter of law, tacking is a question for the jury.* * * The Ninth Circuit correctly held that whether two marks may be tacked for purposes of determining priority is a question for the jury. Accordingly, the judgment of the Ninth Circuit is affirmed.It is so ordered.Notes1 The parties do not question the existence of the tacking doctrine or this substantive standard.2 Our decision in Markman v. Westview Instruments, Inc.,517 U. S. 370 (1996), is not to the contrary. In Markman, we held that the task of construing patent terms falls to judges and not to juries. We held as much because “[t]he construction of written instruments is one of those things that judges often do and are likely to do better than jurors unburdened by training in exegesis.” Id., at 388; see Teva Pharmaceuticals USA, Inc., v. Sandoz, Inc., ante, at 5. The tacking inquiry, by contrast, involves a factual judgment about whether two marks give the same impression to consumers. Making that kind of judgment is, as discussed above, not “one of those things that judges often do” better than jurors.

574.US.54

Following a suspicious vehicle, Sergeant Matt Darisse noticed that only one of the vehicle’s brake lights was working and pulled the driver over. While issuing a warning ticket for the broken brake light, Darisse became suspicious of the actions of the two occupants and their answers to his questions. Petitioner Nicholas Brady Heien, the car’s owner, gave Darisse consent to search the vehicle. Darisse found cocaine, and Heien was arrested and charged with attempted trafficking. The trial court denied Heien’s motion to suppress the seized evidence on Fourth Amendment grounds, concluding that the vehicle’s faulty brake light gave Darisse reasonable suspicion to initiate the stop. The North Carolina Court of Appeals reversed, holding that the relevant code provision, which requires that a car be “equipped with a stop lamp,” N. C. Gen. Stat. Ann. §20–129(g), requires only a single lamp—which Heien’s vehicle had—and therefore the justification for the stop was objectively unreasonable. Reversing in turn, the State Supreme Court held that, even assuming no violation of the state law had occurred, Darisse’s mistaken understanding of the law was reasonable, and thus the stop was valid. Held: Because Darisse’s mistake of law was reasonable, there was reasonable suspicion justifying the stop under the Fourth Amendment. Pp. 4–13. (a) The Fourth Amendment requires government officials to act reasonably, not perfectly, and gives those officials “fair leeway for enforcing the law,” Brinegar v. United States, 338 U.S. 160, 176. Searches and seizures based on mistakes of fact may be reasonable. See, e.g., Illinois v. Rodriguez, 497 U.S. 177, 183–186. The limiting factor is that “the mistakes must be those of reasonable men.” Brinegar, supra, at 176. Mistakes of law are no less compatible with the concept of reasonable suspicion, which arises from an understanding of both the facts and the relevant law. Whether an officer is reasonably mistaken about the one or the other, the result is the same: the facts are outside the scope of the law. And neither the Fourth Amendment’s text nor this Court’s precedents offer any reason why that result should not be acceptable when reached by a reasonable mistake of law. More than two centuries ago, this Court held that reasonable mistakes of law, like those of fact, could justify a certificate of probable cause. United States v. Riddle, 5 Cranch 311, 313. That holding was reiterated in numerous 19th-century decisions. Although Riddle was not a Fourth Amendment case, it explained the concept of probable cause, which this Court has said carried the same “fixed and well known meaning” in the Fourth Amendment, Brinegar, supra, at 175, and n. 14, and no subsequent decision of this Court has undermined that understanding. The contrary conclusion would be hard to reconcile with the more recent precedent of Michigan v. DeFillippo, 443 U.S. 31, where the Court, addressing the validity of an arrest made under a criminal law later declared unconstitutional, held that the officers’ reasonable assumption that the law was valid gave them “abundant probable cause” to make the arrest, id., at 37. Heien attempts to recast DeFillippo as a case solely about the exclusionary rule, not the Fourth Amendment itself, but DeFillippo’s express holding is that the arrest was constitutionally valid because the officers had probable cause. See id., at 40. Heien misplaces his reliance on cases such as Davis v. United States, 564 U. S. ___, where any consideration of reasonableness was limited to the separate matter of remedy, not whether there was a Fourth Amendment violation in the first place. Heien contends that the rationale that permits reasonable errors of fact does not extend to reasonable errors of law, arguing that officers in the field deserve a margin of error when making factual assessments on the fly. An officer may, however, also be suddenly confronted with a situation requiring application of an unclear statute. This Court’s holding does not discourage officers from learning the law. Because the Fourth Amendment tolerates only objectively reasonable mistakes, cf. Whren v. United States, 517 U.S. 806, 813, an officer can gain no advantage through poor study. Finally, while the maxim “Ignorance of the law is no excuse” correctly implies that the State cannot impose punishment based on a mistake of law, it does not mean a reasonable mistake of law cannot justify an investigatory stop. Pp. 4–12. (b) There is little difficulty in concluding that Officer Darisse’s error of law was reasonable. The North Carolina vehicle code that requires “a stop lamp” also provides that the lamp “may be incorporated into a unit with one or more other rear lamps,” N. C. Gen. Stat. Ann. §20–129(g), and that “all originally equipped rear lamps” must be “in good working order,” §20–129(d). Although the State Court of Appeals held that “rear lamps” do not include brake lights, the word “other,” coupled with the lack of state-court precedent interpreting the provision, made it objectively reasonable to think that a faulty brake light constituted a violation. Pp. 12–13. 367 N. C. 163, 749 S.E.2d 278, affirmed. Roberts, C. J., delivered the opinion of the Court, in which Scalia, Kennedy, Thomas, Ginsburg, Breyer, Alito, and Kagan, JJ., joined. Kagan, J., filed a concurring opinion, in which Ginsburg, J., joined. Sotomayor, J., filed a dissenting opinion.

The Fourth Amendment prohibits “unreasonable searches and seizures.” Under this standard, a search or seizure may be permissible even though the justification for the action includes a reasonable factual mistake. An officer might, for example, stop a motorist for traveling alone in a high-occupancy vehicle lane, only to discover upon approaching the car that two children are slumped over asleep in the back seat. The driver has not violated the law, but neither has the officer violated the Fourth Amendment. But what if the police officer’s reasonable mistake is not one of fact but of law? In this case, an officer stopped a vehicle because one of its two brake lights was out, but a court later determined that a single working brake light was all the law required. The question presented is whether such a mistake of law can nonetheless give rise to the reasonable suspicion necessary to uphold the seizure under the Fourth Amendment. We hold that it can. Because the officer’s mistake about the brake-light law was reasonable, the stop in this case was lawful under the Fourth Amendment. I On the morning of April 29, 2009, Sergeant Matt Da-risse of the Surry County Sheriff’s Department sat in his patrol car near Dobson, North Carolina, observing northbound traffic on Interstate 77. Shortly before 8 a.m., a Ford Escort passed by. Darisse thought the driver looked “very stiff and nervous,” so he pulled onto the interstate and began following the Escort. A few miles down the road, the Escort braked as it approached a slower vehicle, but only the left brake light came on. Noting the faulty right brake light, Darisse activated his vehicle’s lights and pulled the Escort over. App. 4–7, 15–16. Two men were in the car: Maynor Javier Vasquez sat behind the wheel, and petitioner Nicholas Brady Heien lay across the rear seat. Sergeant Darisse explained to Vasquez that as long as his license and registration checked out, he would receive only a warning ticket for the broken brake light. A records check revealed no problems with the documents, and Darisse gave Vasquez the warning ticket. But Darisse had become suspicious during the course of the stop—Vasquez appeared nervous, Heien remained lying down the entire time, and the two gave inconsistent answers about their destination. Darisse asked Vasquez if he would be willing to answer some questions. Vasquez assented, and Darisse asked whether the men were transporting various types of contraband. Told no, Darisse asked whether he could search the Escort. Vasquez said he had no objection, but told Darisse he should ask Heien, because Heien owned the car. Heien gave his consent, and Darisse, aided by a fellow officer who had since arrived, began a thorough search of the vehicle. In the side compartment of a duffle bag, Darisse found a sandwich bag containing cocaine. The officers arrested both men. 366 N. C. 271, 272–273, 737 S. E. 2d 351, 352–353 (2012); App. 5–6, 25, 37. The State charged Heien with attempted trafficking in cocaine. Heien moved to suppress the evidence seized from the car, contending that the stop and search had violated the Fourth Amendment of the United States Constitution. After a hearing at which both officers testified and the State played a video recording of the stop, the trial court denied the suppression motion, concluding that the faulty brake light had given Sergeant Darisse reasonable suspicion to initiate the stop, and that Heien’s subsequent consent to the search was valid. Heien pleaded guilty but reserved his right to appeal the suppression decision. App. 1, 7–10, 12, 29, 43–44. The North Carolina Court of Appeals reversed. 214 N. C. App. 515, 714 S. E. 2d 827 (2011). The initial stop was not valid, the court held, because driving with only one working brake light was not actually a violation of North Carolina law. The relevant provision of the vehicle code provides that a car must be “equipped with a stop lamp on the rear of the vehicle. The stop lamp shall display a red or amber light visible from a distance of not less than 100 feet to the rear in normal sunlight, and shall be actuated upon application of the service (foot) brake. The stop lamp may be incorporated into a unit with one or moreother rear lamps.” N. C. Gen. Stat. Ann. §20–129(g) (2007). Focusing on the statute’s references to “a stop lamp” and “[t]he stop lamp” in the singular, the court concluded that a vehicle is required to have only one working brake light—which Heien’s vehicle indisputably did. The justification for the stop was therefore “objectively unreason-able,” and the stop violated the Fourth Amendment. 214 N. C. App., at 518–522, 714 S. E. 2d, at 829–831. The State appealed, and the North Carolina Supreme Court reversed. 366 N. C. 271, 737 S. E. 2d 351. Noting that the State had chosen not to seek review of the Court of Appeals’ interpretation of the vehicle code, the North Carolina Supreme Court assumed for purposes of its decision that the faulty brake light was not a violation. Id., at 275, 737 S. E. 2d, at 354. But the court concluded that, for several reasons, Sergeant Darisse could have reasonably, even if mistakenly, read the vehicle code to require that both brake lights be in good working order. Most notably, a nearby code provision requires that “all originally equipped rear lamps” be functional. Id., at 282–283, 737 S. E. 2d, at 358–359 (quoting N. C. Gen. Stat. Ann. §20–129(d)). Because Sergeant Darisse’s mistaken understanding of the vehicle code was reasonable, the stop was valid. “An officer may make a mistake, including a mistake of law, yet still act reasonably under the circumstances. . . . [W]hen an officer acts reasonably under the circumstances, he is not violating the Fourth Amendment.” Id., at 279, 737 S. E. 2d, at 356. The North Carolina Supreme Court remanded to the Court of Appeals to address Heien’s other arguments for suppression (which are not at issue here). Id., at 283, 737 S. E. 2d, at 359. The Court of Appeals rejected those arguments and affirmed the trial court’s denial of his motion to suppress. ___ N. C. App. ___, 741 S. E. 2d 1 (2013). The North Carolina Supreme Court affirmed in turn. 367 N. C. 163, 749 S. E. 2d 278 (2013). We granted certiorari. 572 U. S. ___ (2014). II The Fourth Amendment provides: “The right of the people to be secure in their persons, houses, papers, and effects, against unreason-able searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.” A traffic stop for a suspected violation of law is a “seizure” of the occupants of the vehicle and therefore must be conducted in accordance with the Fourth Amendment. Brendlin v. California, 551 U. S. 249 –259 (2007). All parties agree that to justify this type of seizure, officers need only “reasonable suspicion”—that is, “a particularized and objective basis for suspecting the particular person stopped” of breaking the law. Prado Navarette v. California, 572 U. S. ___, ___ (2014) (slip op., at 3) (internal quotation marks omitted). The question here is whether reasonable suspicion can rest on a mistaken understanding of the scope of a legal prohibition. We hold that it can. As the text indicates and we have repeatedly affirmed, “the ultimate touchstone of the Fourth Amendment is ‘reasonableness.’ ” Riley v. California, 573 U. S. ___, ___ (2014) (slip op., at 5) (some internal quotation marks omitted). To be reasonable is not to be perfect, and so the Fourth Amendment allows for some mistakes on the part of government officials, giving them “fair leeway for enforcing the law in the community’s protection.” Brinegar v. United States, 338 U. S. 160, 176 (1949) . We have recognized that searches and seizures based on mistakes of fact can be reasonable. The warrantless search of a home, for instance, is reasonable if undertaken with the consent of a resident, and remains lawful when officers obtain the consent of someone who reasonably appears to be but is not in fact a resident. See Illinois v. Rodriguez, 497 U. S. 177 –186 (1990). By the same token, if officers with probable cause to arrest a suspect mistakenly arrest an individual matching the suspect’s description, neither the seizure nor an accompanying search of the arrestee would be unlawful. See Hill v. California, 401 U. S. 797 –805 (1971). The limit is that “the mistakes must be those of reasonable men.” Brinegar, supra, at 176. But reasonable men make mistakes of law, too, and such mistakes are no less compatible with the concept of reasonable suspicion. Reasonable suspicion arises from the combination of an officer’s understanding of the facts and his understanding of the relevant law. The officer may be reasonably mistaken on either ground. Whether the facts turn out to be not what was thought, or the law turns out to be not what was thought, the result is the same: the facts are outside the scope of the law. There is no reason, under the text of the Fourth Amendment or our precedents, why this same result should be acceptable when reached by way of a reasonable mistake of fact, but not when reached by way of a similarly reasonable mistake of law. The dissent counters that our cases discussing probable cause and reasonable suspicion, most notably Ornelas v. United States, 517 U. S. 690 –697 (1996), have contained “scarcely a peep” about mistakes of law. Post, at 2–3 (opinion of Sotomayor, J.). It would have been surprising, of course, if they had, since none of those cases involved a mistake of law. Although such recent cases did not address mistakes of law, older precedents did. In fact, cases dating back two centuries support treating legal and factual errors alike in this context. Customs statutes enacted by Congress not long after the founding authorized courts to issue certificates indemnifying customs officers against damages suits premised on unlawful seizures. See, e.g., Act of Mar. 2, 1799, ch. 22, §89, 1Stat. 695–696. Courts were to issue such certificates on a showing that the officer had “reasonable cause”—a synonym for “probable cause”—for the challenged seizure. Ibid.; see Stacey v. Emery, 97 U. S. 642, 646 (1878); United States v. Riddle, 5 Cranch 311 (1809). In United States v. Riddle, a customs officer seized goods on the ground that the English shipper had violated the customs laws by preparing an invoice that undervalued the merchandise, even though the American consignee declared the true value to the customs collector. Chief Justice Marshall held that there had been no violation of the customs law because, whatever the shipper’s intention, the consignee had not actually attempted to defraud the Government. Nevertheless, because “the construction of the law was liable to some question,” he affirmed the issuance of a certificate of probable cause: “A doubt as to the true construction of the law is as reasonable a cause for seizure as a doubt respecting the fact.” Id., at 313. This holding—that reasonable mistakes of law, like those of fact, would justify certificates of probable cause—was reiterated in a number of 19th-century decisions. See, e.g., The Friendship, 9 F. Cas. 825, 826 (No. 5,125) (CC Mass. 1812) (Story, J.); United States v. The Reindeer, 27 F. Cas. 758, 768 (No. 16,145) (CC RI 1848); United States v. The Recorder, 27 F. Cas. 723 (No. 16,130) (CC SDNY 1849). By the Civil War, there had been “numerous cases in which [a] captured vessel was in no fault, and had not, under a true construction of the law, presented even ground of suspicion, and yet the captor was exonerated because he acted under an honest mistake of the law.” The La Manche, 14 F. Cas. 965, 972 (No. 8,004) (D Mass. 1863). Riddle and its progeny are not directly on point. Chief Justice Marshall was not construing the Fourth Amendment, and a certificate of probable cause functioned much like a modern-day finding of qualified immunity, which depends on an inquiry distinct from whether an officer has committed a constitutional violation. See, e.g., Carroll v. Carman, ante, at 7 (per curiam). But Chief Justice Marshall was nevertheless explaining the concept of probable cause, which, he noted elsewhere, “in all cases of seizure, has a fixed and well known meaning. It imports a seizure made under circumstances which warrant suspicion.” Locke v. United States, 7 Cranch 339, 348 (1813). We have said the phrase “probable cause” bore this “fixed and well known meaning” in the Fourth Amendment, see Brinegar, supra, at 175, and n. 14, and Riddle illustrates that it encompassed suspicion based on reasonable mistakes of both fact and law. No decision of this Court in the two centuries since has undermined that understanding.[1] The contrary conclusion would be hard to reconcile with a much more recent precedent. In Michigan v. DeFillippo, 443 U. S. 31 (1979) , we addressed the validity of an arrest made under a criminal law later declared unconstitu-tional. A Detroit ordinance that authorized police officers to stop and question individuals suspected of criminal activ-ity also made it an offense for such an individual “to refuse to identify himself and produce evidence of his identity.” Id., at 33. Detroit police officers sent to investigate a report of public intoxication arrested Gary DeFillippo after he failed to identify himself. A search incident to arrest uncovered drugs, and DeFillippo was charged with possession of a controlled substance. The Michigan Court of Appeals ordered the suppression of the drugs, concluding that the identification ordinance was unconstitutionally vague and that DeFillippo’s arrest was therefore invalid. Id., at 34–35. Accepting the unconstitutionality of the ordinance as a given, we nonetheless reversed. At the time the officers arrested DeFillippo, we explained, “there was no controlling precedent that this ordinance was or was not constitutional, and hence the conduct observed violated a presumptively valid ordinance.” Id., at 37. Acknowledging that the outcome might have been different had the ordinance been “grossly and flagrantly unconstitutional,” we concluded that under the circumstances “there was abundant probable cause to satisfy the constitutional prerequisite for an arrest.” Id., at 37–38. The officers were wrong in concluding that DeFillippo was guilty of a criminal offense when he declined to iden-tify himself. That a court only later declared the ordinance unconstitutional does not change the fact that DeFillippo’s conduct was lawful when the officers observed it. See Danforth v. Minnesota, 552 U. S. 264, 271 (2008) . But the officers’ assumption that the law was valid was reason-able, and their observations gave them “abundant probable cause” to arrest DeFillippo. 443 U. S., at 37. Although DeFillippo could not be prosecuted under the identifica-tion ordinance, the search that turned up the drugs was constitutional. Heien struggles to recast DeFillippo as a case solely about the exclusionary rule, not the Fourth Amendment itself. In his view, the officers’ mistake of law resulted in a violation the Fourth Amendment, but suppression of the drugs was not the proper remedy. We did say in a footnote that suppression of the evidence found on DeFillippo would serve none of the purposes of the exclusionary rule. See id., at 38, n. 3. But that literally marginal discussion does not displace our express holding that the arrest was constitutionally valid because the officers had probable cause. See id., at 40. Nor, contrary to Heien’s suggestion, did either United States v. Leon, 468 U. S. 897 (1984) , or Illinois v. Gates, 462 U. S. 213 (1983) , somehow erase that holding and transform DeFillippo into an exclusionary rule decision. See Brief for Petitioner 28–29. In Leon, we said DeFillippo paid “attention to the purposes underlying the exclusionary rule,” but we also clarified that it did “not involv[e] the scope of the rule itself.” 468 U. S., at 911–912. As for Gates, only Justice White’s separate opinion (joined by no other Justice) discussed DeFillippo, and it acknowledged that “DeFillippo did not modify the exclusionary rule itself” but instead “upheld the validity of an arrest.” 462 U. S., at 256, n. 12 (opinion concurring in judgment). Heien is correct that in a number of decisions we have looked to the reasonableness of an officer’s legal error in the course of considering the appropriate remedy for a constitutional violation, instead of whether there was a violation at all. See, e.g., Davis v. United States, 564 U. S. ___, ___ (2011) (slip op., at 11) (exclusionary rule); Illinois v. Krull, 480 U. S. 340 –360 (1987) (exclusionary rule); Wilson v. Layne, 526 U. S. 603, 615 (1999) (qualified immunity); Anderson v. Creighton, 483 U. S. 635, 641 (1987) (qualified immunity). In those cases, however, we had already found or assumed a Fourth Amendment violation. An officer’s mistaken view that the conduct at issue did not give rise to such a violation—no matter how reason-able—could not change that ultimate conclusion. See Brief for Respondent 29–31; Brief for United States as Amicus Curiae 30, n. 3. Any consideration of the reasonableness of an officer’s mistake was therefore limited to the separate matter of remedy. Here, by contrast, the mistake of law relates to the antecedent question of whether it was reasonable for an officer to suspect that the defendant’s conduct was illegal. If so, there was no violation of the Fourth Amendment in the first place. None of the cases Heien or the dissent cites precludes a court from considering a reasonable mistake of law in addressing that question. Cf. Herring v. United States, 555 U. S. 135, 139 (2009) (assuming a Fourth Amendment violation while rejecting application of the exclusionary rule, but noting that “[w]hen a probable-cause determination was based on reasonable but mis-taken assumptions, the person subjected to a search or seizure has not necessarily been the victim of a constitutional violation”). Heien also contends that the reasons the Fourth Amendment allows some errors of fact do not extend to errors of law. Officers in the field must make factual assessments on the fly, Heien notes, and so deserve a margin of error. In Heien’s view, no such margin is appropriate for questions of law: The statute here either requires one working brake light or two, and the answer does not turn on anything “an officer might suddenly confront in the field.” Brief for Petitioner 21. But Heien’s point does not consider the reality that an officer may “suddenly confront” a situation in the field as to which the application of a statute is unclear—however clear it may later become. A law prohibiting “vehicles” in the park either covers Segways or not, see A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 36–38 (2012), but an officer will nevertheless have to make a quick decision on the law the first time one whizzes by. Contrary to the suggestion of Heien and amici, our decision does not discourage officers from learning the law. The Fourth Amendment tolerates only reasonable mistakes, and those mistakes—whether of fact or of law—must be objectively reasonable. We do not examine the subjective understanding of the particular officer involved. Cf. Whren v. United States, 517 U. S. 806, 813 (1996) . And the inquiry is not as forgiving as the one employed in the distinct context of deciding whether an officer is entitled to qualified immunity for a constitutional or statutory violation. Thus, an officer can gain no Fourth Amendment advantage through a sloppy study of the laws he is duty-bound to enforce. Finally, Heien and amici point to the well-known maxim, “Ignorance of the law is no excuse,” and contend that itis fundamentally unfair to let police officers get away with mistakes of law when the citizenry is accorded no such leeway. Though this argument has a certain rhetorical appeal, it misconceives the implication of the maxim. The true symmetry is this: Just as an individual generally cannot escape criminal liability based on a mistaken understanding of the law, so too the government cannot impose criminal liability based on a mistaken understanding of the law. If the law required two working brake lights, Heien could not escape a ticket by claiming he reasonably thought he needed only one; if the law required only one, Sergeant Darisse could not issue a valid ticket by claiming he reasonably thought drivers needed two. But just because mistakes of law cannot justify either the imposition or the avoidance of criminal liability, it does not follow that they cannot justify an investigatory stop. And Heien is not appealing a brake-light ticket; he is appealing a cocaine-trafficking conviction as to which there is no asserted mistake of fact or law. III Here we have little difficulty concluding that the officer’s error of law was reasonable. Although the North Carolina statute at issue refers to “a stop lamp,” suggesting the need for only a single working brake light, it also provides that “[t]he stop lamp may be incorporated into a unit with one or more other rear lamps.” N. C. Gen. Stat. Ann. §20–129(g) (emphasis added). The use of “other” suggests to the everyday reader of English that a “stop lamp” is a type of “rear lamp.” And another subsection of the same provision requires that vehicles “have all originally equipped rear lamps or the equivalent in good working order,” §20–129(d), arguably indicating that if a vehicle has multiple “stop lamp[s],” all must be functional. The North Carolina Court of Appeals concluded that the “rear lamps” discussed in subsection (d) do not include brake lights, but, given the “other,” it would at least have been reasonable to think they did. Both the majority and the dissent in the North Carolina Supreme Court so concluded, and we agree. See 366 N. C., at 282–283, 737 S. E. 2d, at 358–359; id., at 283, 737 S. E. 2d, at 359 (Hudson, J., dissenting) (calling the Court of Appeals’ decision “surprising”). This “stop lamp” provision, moreover, had never been previously construed by North Carolina’s appellate courts. See id., at 283, 737 S. E. 2d, at 359 (majority opinion). It was thus objectively reasonable for an officer in Sergeant Darisse’s position to think that Heien’s faulty right brake light was a violation of North Carolina law. And because the mistake of law was reasonable, there was reasonable suspicion justifying the stop. The judgment of the Supreme Court of North Carolina is Affirmed.Notes 1 The dissent contends that “the tolerance of mistakes of law in cases like Riddle was a result of the specific customs statute that Congress had enacted.” Post, at 8, n. 3 (citing The Apollon, 9 Wheat. 362, 373 (1824) (Story, J.)). The relevant portion of The Apollon, however, addressed “the effect of probable cause,” not what gave rise to it. Id., at 372 (emphasis added); see id., at 376 (finding it “unnecessary” to decide whether probable cause existed because it “would not, under the circumstances of this case, constitute a valid defence”). Justice Story understandably did not cite Riddle or discuss its tolerance of mistakes of law anywhere in The Apollon.

575.US.2014_13-1487

After being charged with the felony offense of distributing marijuana, petitioner Tony Henderson was required as a condition of his bail to turn over firearms that he lawfully owned. Henderson ultimately pleaded guilty, and, as a felon, was prohibited under 18 U. S. C. §922(g) from possessing his (or any other) firearms. Henderson therefore asked the Federal Bureau of Investigation, which had custody of his firearms, to transfer them to his friend. But the agency refused to do so. Henderson then filed a motion in federal district court seeking to transfer his firearms, but the court denied the motion on the ground that Henderson’s requested transfer would give him constructive possession of the firearms in violation of §922(g). The Eleventh Circuit affirmed. Held: A court-ordered transfer of a felon’s lawfully owned firearms from Government custody to a third party is not barred by §922(g) if the court is satisfied that the recipient will not give the felon control over the firearms, so that he could either use them or direct their use. Federal courts have equitable authority to order law enforcement to return property obtained during the course of a criminal proceeding to its rightful owner. Section 922(g), however, bars a court from ordering guns returned to a felon-owner like Henderson, because that would place the owner in violation of the law. And because §922(g) bans constructive as well as actual possession, it also prevents a court from ordering the transfer of a felon’s guns to someone willing to give the felon access to them or to accede to the felon’s instructions about their future use. The Government goes further, arguing that §922(g) prevents all transfers to a third party, no matter how independent of the felon’s influence, unless that recipient is a licensed firearms dealer or other third party who will sell the guns on the open market. But that view conflates possession, which §922(g) prohibits, with an owner’s right merely to alienate his property, which it does not. After all, the Government’s reading of §922(g) would prohibit a felon from disposing of his firearms even when he would lack any control over and thus not possess them before, during, or after the disposition. That reading would also extend §922(g)’s scope far beyond its purpose; preventing a felon like Henderson from disposing of his firearms, even in ways that guarantee he never uses them again, does nothing to advance the statute’s goal of keeping firearms away from felons. Finally, the Government’s insistence that a felon cannot select a third-party recipient over whom he exercises no influence fits poorly with its concession that a felon may select a firearms dealer or third party to sell his guns. The Government’s reading of §922(g) is thus overbroad. Accordingly, a court may approve the transfer of a felon’s guns consistently with §922(g) if, but only if, the recipient will not grant the felon control over those weapons. One way to ensure that result is to order that the guns be turned over to a firearms dealer, himself independent of the felon’s control, for subsequent sale on the open market. But that is not the only option; a court, with proper assurances from the recipient, may also grant a felon’s request to transfer his guns to a person who expects to maintain custody of them. Either way, once a court is satisfied that the transferee will not allow the felon to exert any influence over the firearms, the court has equitable power to accommodate the felon’s transfer request. Pp. 3–8. 555 Fed. Appx. 851, vacated and remanded. Kagan, J., delivered the opinion for a unanimous Court.

Government agencies sometimes come into possession of firearms lawfully owned by individuals facing serious criminal charges. If convicted, such a person cannot recover his guns because a federal statute, 18 U. S. C. §922(g), prohibits any felon from possessing firearms. In this case, we consider what §922(g) allows a court to do when a felon instead seeks the transfer of his guns to either a firearms dealer (for future sale on the open market) or some other third party. We hold that §922(g) does not bar such a transfer unless it would allow the felon to later control the guns, so that he could either use them or direct their use. I The Federal Government charged petitioner Tony Henderson, then a U. S. Border Patrol agent, with the felony offense of distributing marijuana. See 21 U. S. C. §§841(a)(1), (b)(1)(D). A Magistrate Judge required that Henderson surrender all his firearms as a condition of his release on bail. Henderson complied, and the Federal Bureau of Investigation (FBI) took custody of the guns. Soon afterward, Henderson pleaded guilty to the distribution charge; as a result of that conviction, §922(g) prevents him from legally repossessing his firearms. Following his release from prison, Henderson asked the FBI to transfer the guns to Robert Rosier, a friend who had agreed to purchase them for an unspecified price. The FBI denied the request. In a letter to Henderson, it explained that “the release of the firearms to [Rosier] would place you in violation of [§922(g)], as it would amount to constructive possession” of the guns. App. 121. Henderson then returned to the court that had handled his criminal case to seek release of his firearms. Invoking the court’s equitable powers, Henderson asked for an order directing the FBI to transfer the guns either to his wife or to Rosier. The District Court denied the motion, concluding (as the FBI had) that Henderson could not “transfer the firearms or receive money from their sale” without “constructive[ly] possessi[ng]” them in violation of §922(g). No. 3:06–cr–211 (MD Fla., Aug. 8, 2012), App. to Pet. for Cert. 5a–6a, 12a. The Court of Appeals for the Eleventh Circuit affirmed on the same ground, reasoning that granting Henderson’s motion would amount to giving a felon “constructive possession” of his firearms. 555 Fed. Appx. 851, 853 (2014) (per curiam).[1] We granted certiorari, 574 U. S. ___ (2014), to resolve a circuit split over whether, as the courts below held, §922(g) categorically prohibits a court from approving a convicted felon’s request to transfer his firearms to an-other person.[2] We now vacate the decision below. II A federal court has equitable authority, even after a criminal proceeding has ended, to order a law enforcement agency to turn over property it has obtained during the case to the rightful owner or his designee. See, e.g.,United States v. Martinez, 241 F. 3d 1329, 1330–1331 (CA11 2001) (citing numerous appellate decisions to that effect); Tr. of Oral Arg. 41 (Solicitor General agreeing). Congress, however, may cabin that power in various ways. As relevant here, §922(g) makes it unlawful for any person convicted of a felony to “possess in or affecting commerce[ ] any firearm or ammunition.” That provision prevents a court from instructing an agency to return guns in its custody to a felon-owner like Henderson, because that would place him in violation of the law. The question here is how §922(g) affects a court’s authority to instead direct the transfer of such firearms to a third party. Section 922(g) proscribes possession alone, but covers possession in every form. By its terms, §922(g) does not prohibit a felon from owning firearms. Rather, it interferes with a single incident of ownership—one of the proverbial sticks in the bundle of property rights—by preventing the felon from knowingly possessing his (or an-other person’s) guns. But that stick is a thick one, encompassing what the criminal law recognizes as “actual” and “constructive” possession alike. 2A K. O’Malley, J. Grenig, & W. Lee, Federal Jury Practice and Instructions, Criminal §39.12, p. 55 (6th ed. 2009) (hereinafter O’Malley); see National Safe Deposit Co. v. Stead, 232 U. S. 58, 67 (1914) (noting that in “legal terminology” the word “possession” is “interchangeably used to describe” both the actual and the constructive kinds). Actual possession exists when a person has direct physical control over a thing. See Black’s Law Dictionary 1047 (5th ed. 1979) (hereinafter Black’s); 2A O’Malley §39.12, at 55. Constructive possession is established when a person, though lacking such physical custody, still has the power and intent to exercise control over the object. See Black’s 1047; 2A O’Malley §39.12, at 55. Section 922(g) thus prevents a felon not only from holding his firearms himself but also from maintaining control over those guns in the hands of others. That means, as all parties agree, that §922(g) prevents a court from ordering the sale or other transfer of a felon’s guns to someone willing to give the felon access to them or to accede to the felon’s instructions about their future use. See Brief for United States 23; Reply Brief 12. In such a case, the felon would have control over the guns, even while another person kept physical custody. The idea of constructive possession is designed to preclude just that result, “allow[ing] the law to reach beyond puppets to puppeteers.” United States v. Al-Rekabi, 454 F. 3d 1113, 1118 (CA10 2006). A felon cannot evade the strictures of §922(g) by arranging a sham transfer that leaves him in effective control of his guns. And because that is so, a court may no more approve such a transfer than order the return of the firearms to the felon himself. The Government argues that §922(g) prohibits still more—that it bars a felon, except in one circumstance, from transferring his firearms to another person, no matter how independent of the felon’s influence. According to the Government, a felon “exercises his right to control” his firearms, and thus violates §922(g)’s broad ban on possession, merely by “select[ing] the[ir] first recipient,” because that choice “determine[s] who [will] (and who [will] not) next have access to the firearms.” Brief for United States 24. And that remains so even if a felon never retakes physical custody of the guns and needs a court order to approve and effectuate the proposed transfer. The felon (so says the Government) still exerts enough sway over the guns’ disposition to “have constructive possession” of them. Id., at 25. The only time that is not true, the Government claims, is when a felon asks the court to transfer the guns to a licensed dealer or other party who will sell the guns for him on the open market. See id., at 20–22; Tr. of Oral Arg. 18–21. Because the felon then does not control the firearms’ final destination, the Government avers, he does not constructively possess them and a court may approve the transfer. See ibid. But the Government’s theory wrongly conflates the right to possess a gun with another incident of ownership, which §922(g) does not affect: the right merely to sell or otherwise dispose of that item. Cf. Andrus v. Allard, 444 U. S. 51 –66 (1979) (distinguishing between entitlements to possess and sell property). Consider the scenario that the Government claims would violate §922(g). The felon has nothing to do with his guns before, during, or after the transaction in question, except to nominate their recipient. Prior to the transfer, the guns sit in an evidence vault, under the sole custody of law enforcement officers. Assuming the court approves the proposed recipient, FBI agents handle the firearms’ physical conveyance, without the felon’s participation. Afterward, the purchaser or other custodian denies the felon any access to or influence over the guns; the recipient alone decides where to store them, when to loan them out, how to use them, and so on. In short, the arrangement serves only to divest the felon of his firearms—and even that much depends on a court’s approving the designee’s fitness and ordering the transfer to go forward. Such a felon exercises not a possessory interest (whether directly or through another), but instead a naked right of alienation—the capacity to sell or transfer his guns, unaccompanied by any control over them.[3] The Government’s view of what counts as “possession” would also extend §922(g)’s scope far beyond its purpose. Congress enacted that ban to keep firearms away from felons like Henderson, for fear that they would use those guns irresponsibly. See Small v. United States, 544 U. S. 385, 393 (2005) . Yet on the Government’s construction, §922(g) would prevent Henderson from disposing of his firearms even in ways that guarantee he never uses them again, solely because he played a part in selecting their transferee. He could not, for example, place those guns in a secure trust for distribution to his children after his death. He could not sell them to someone halfway around the world. He could not even donate them to a law enforcement agency. See Tr. of Oral Arg. 22. Results of that kind would do nothing to advance §922(g)’s purpose. Finally, the Government’s expansive idea of constructive possession fits poorly with its concession that a felon in Henderson’s position may select a firearms dealer or other third party to sell his guns and give him the proceeds. After all, the felon chooses the guns’ “first recipient” in that case too, deciding who “next ha[s] access to the firearms.” Brief for United States 24; see supra, at 5. If (as the Government argues) that is all it takes to exercise control over and thus constructively possess an item, then (contrary to the Government’s view) the felon would violate §922(g) merely by selecting a dealer to sell his guns. To be sure, that person will predictably convey the firearms to someone whom the felon does not know and cannot control: That is why the Government, as a practical matter, has no worries about the transfer. See Tr. of Oral Arg. 19–21. But that fact merely demonstrates how the Government’s view of §922(g) errs in its focus in a case like this one. What matters here is not whether a felon plays a role in deciding where his firearms should go next: That test would logically prohibit a transfer even whenthe chosen recipient will later sell the guns to someone else. What matters instead is whether the felon will have the ability to use or direct the use of his firearms afterthe transfer. That is what gives the felon constructive possession. Accordingly, a court facing a motion like Henderson’s may approve the transfer of guns consistently with §922(g) if, but only if, that disposition prevents the felon from later exercising control over those weapons, so that he could either use them or tell someone else how to do so. One way to ensure that result, as the Government notes, is to order that the guns be turned over to a firearms dealer, himself independent of the felon’s control, for subsequent sale on the open market. See, e.g., United States v. Zaleski, 686 F. 3d 90, 92–94 (CA2 2012). Indeed, we can see no reason, absent exceptional circumstances, to disapprove a felon’s motion for such a sale, whether or not he has picked the vendor. That option, however, is not the only one available under §922(g). A court may also grant a felon’s request to transfer his guns to a person who expects to maintain custody of them, so long as the recipient will not allow the felon to exert any influence over their use. In considering such a motion, the court may properly seek certain assurances: for example, it may ask the proposed transferee to promise to keep the guns away from the felon, and to acknowledge that allowing him to use them would aid and abet a §922(g) violation. See id., at 94; United States v. Miller, 588 F. 3d 418, 420 (CA7 2009). Even such a pledge, of course, might fail to provide an adequate safeguard, and a court should then disapprove the transfer. See, e.g., State v. Fadness, 363 Mont. 322, 341–342, 268 P. 3d 17, 30 (2012) (upholding a trial court’s finding that the assurances given by a felon’s parents were not credible). But when a court is satisfied that a felon will not retain control over his guns, §922(g) does not apply, and the court has equitable power to accommodate the felon’s request. Neither of the courts below assessed Henderson’s motion for a transfer of his firearms in accord with these principles. We therefore vacate the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 The Court of Appeals added that Henderson’s “equitable argument rings hollow” because a convicted felon has “unclean hands to demand return [or transfer] of his firearms.” 555 Fed. Appx., at 854. That view is wrong, as all parties now agree. See Brief for Petitioner 35–39; Brief for United States 31, n. 8; Tr. of Oral Arg. 33, 42. The unclean hands doctrine proscribes equitable relief when, but only when, an individ-ual’s misconduct has “immediate and necessary relation to the equity that he seeks.” Keystone Driller Co. v. General Excavator Co., 290 U. S. 240, 245 (1933) . The doctrine might apply, for example, if a felon requests the return or transfer of property used in furtherance of his offense. See, e.g., United States v. Kaczynski, 551 F. 3d 1120, 1129–1130 (CA9 2009) (holding that the Unabomber had unclean hands to request the return of bomb-making materials). But Henderson’s felony conviction had nothing to do with his firearms, so the unclean hands rule has no role to play here. 2 Compare 555 Fed. Appx. 851, 853–854 (CA11 2014) (per curiam) (case below) (holding that §922(g) bars any transfer); United States v. Felici, 208 F. 3d 667, 670 (CA8 2000) (same), with United States v. Zaleski, 686 F. 3d 90, 92–94 (CA2 2012) (holding that §922(g) permits some transfers); United States v. Miller, 588 F. 3d 418, 419–420 (CA7 2009) (same). 3 The Government calls our attention to several cases in which courts have found constructive possession of firearms based on evidence that a felon negotiated and arranged a sale of guns while using a third party to make the physical handoff to the buyer. See, e.g., United States v. Nungaray, 697 F. 3d 1114, 1116–1119 (CA9 2012); United States v. Virciglio, 441 F. 2d 1295, 1297–1298 (CA5 1971). But the facts in the cited cases bear no similarity to those here. In each, the defendant-felon controlled the guns’ movement both before and during the transaction at issue (and even was present at the delivery site). As the Government explains, the felon could “make a gun appear” at the time and place of his choosing and decide what would happen to it once it got there. Tr. of Oral Arg. 27. Indeed, he could have chosen to take the firearms for himself or direct them to someone under his influence. The felon’s management of the sale thus exemplified, and served as evidence of, his broader command over the guns’ location and use—the very hallmark of possession. But as just explained, that kind of control is absent when a felon can do no more than nominate an independent recipient for firearms in a federal agency’s custody. The decisions the Government invokes thus have no bearing on this case; nor does our decision here, which addresses only §922(g)’s application to court-supervised transfers of guns, prevent the Government from bringing charges under §922(g) in cases resembling those cited.

574.US.352

Section 3 of the Religious Land Use and Institutionalized Persons Act of 2000 (RLUIPA) provides that “[n]o government shall impose a substantial burden on the religious exercise” of an institutionalized person unless the government demonstrates that the burden “is the least restrictive means of furthering [a] compelling governmental interest.” 42 U. S. C. §2000cc–1(a). Petitioner is an Arkansas inmate and devout Muslim who wishes to grow a ½-inch beard in accordance with his religious beliefs. Respondent Arkansas Department of Correction (Department) prohibits its prisoners from growing beards, with the single exception that inmates with diagnosed skin conditions may grow ¼-inch beards. Petitioner sought an exemption on religious grounds and, although he believes that his faith requires him not to trim his beard at all, he proposed a compromise under which he would be allowed to maintain a ½-inch beard. Prison officials denied his request, and petitioner sued in Federal District Court. At an evidentiary hearing before a Magistrate Judge, Department witnesses testified that beards compromised prison safety because they could be used to hide contraband and because an inmate could quickly shave his beard to disguise his identity. The Magistrate Judge recommended dismissing petitioner’s complaint, emphasizing that prison officials are entitled to deference on security matters and that the prison permitted petitioner to exercise his religion in other ways. The District Court adopted the recommendation in full, and the Eighth Circuit affirmed, holding that the Department had satisfied its burden of showing that the grooming policy was the least restrictive means of furthering its compelling security interests, and reiterating that courts should defer to prison officials on matters of security. Held: The Department’s grooming policy violates RLUIPA insofar as it prevents petitioner from growing a ½-inch beard in accordance with his religious beliefs. Pp. 6–16. (a) Under RLUIPA, the challenging party bears the initial burden of proving that his religious exercise is grounded in a sincerely held religious belief, see Burwell v. Hobby Lobby Stores, Inc., 573 U. S. ___, ___, n. 28, and that the government’s action substantially burdens his religious exercise. Here, petitioner’s sincerity is not in dispute, and he easily satisfies the second obligation. The Department’s policy forces him to choose between “engag[ing] in conduct that seriously violates [his] religious belie[f],” id., at ___, or contravening the grooming policy and risking disciplinary action. In reaching the opposite conclusion, the District Court misunderstood the analysis that RLUIPA demands. First, the District Court erred by concluding that the grooming policy did not substantially burden petitioner’s religious exercise because he could practice his religion in other ways. Second, the District Court erroneously suggested that the burden on petitioner’s religious exercise was slight because petitioner testified that his religion would “credit” him for attempting to follow his religious beliefs, even if that attempt proved unsuccessful. RLUIPA, however, applies to religious exercise regardless of whether it is “compelled.” §2000cc–5(7)(A). Finally, the District Court improperly relied on petitioner’s testimony that not all Muslims believe that men must grow beards. Even if petitioner’s belief were idiosyncratic, RLUIPA’s guarantees are “not limited to beliefs which are shared by all of the members of a religious sect.” Thomas v. Review Bd. of Indiana Employment Security Div., 450 U.S. 707, 715–716. Pp. 6–8. (b) Once the challenging party satisfies his burden, the burden shifts to the government to show that substantially burdening the religious exercise of the “particular claimant” is “the least restrictive means of furthering [a] compelling governmental interest.” Hobby Lobby, supra, at ___; §2000cc–1(a). The Department fails to show that enforcing its beard prohibition against petitioner furthers its compelling interests in preventing prisoners from hiding contraband and disguising their identities. Pp. 8–13. (i) While the Department has a compelling interest in regulating contraband, its argument that this interest is compromised by allowing an inmate to grow a ½-inch beard is unavailing, especially given the difficulty of hiding contraband in such a short beard and the lack of a corresponding policy regulating the length of hair on the head. RLUIPA does not permit the unquestioning deference required to accept the Department’s assessment. See Gonzales v. O Centro Espírita Beneficente União do Vegetal, 546 U.S. 418, 434. Even if the Department could show that denying petitioner a ½-inch beard furthers its interest in rooting out contraband, it would still have to show that its policy is the least restrictive means of furthering that interest, a standard that is “exceptionally demanding” and requires the government to “sho[w] that it lacks other means of achieving its desired goal without imposing a substantial burden on the exercise of religion by the objecting part[y].” Hobby Lobby, supra, at ___. Here, the Department fails to establish that its security concerns cannot be satisfied by simply searching a ½-inch beard. Pp. 9–11. (ii) Even if the Department’s grooming policy furthers its compelling interest in prisoner identification, its policy still violates RLUIPA as applied in the present circumstances. As petitioner argues, re-quiring inmates to be photographed both with and without beards and then periodically thereafter is a less restrictive means of solving the Department’s identification concerns. The Department fails to show why its prison system is so different from the many institutions that allow facial hair that the dual-photo method cannot be employed at its institutions. It also fails to show why the security risk presented by a prisoner shaving a ½-inch beard is so different from the risk of a prisoner shaving a mustache, head hair, or ¼-inch beard. Pp. 11–13. (c) In addition to the Department’s failure to prove that petitioner’s proposed alternatives would not sufficiently serve its security interests, the Department also fails to adequately explain the substantial underinclusiveness of its policy, since it permits ¼-inch beards for prisoners with medical conditions and more than ½ inch of hair on the head. Its failure to pursue its proffered objectives with regard to such “analogous nonreligious conduct” suggests that its interests “could be achieved by narrower ordinances that burdened religion to a far lesser degree.” Church of Lukumi Babalu Aye, Inc. v. Hialeah, 508 U.S. 520, 546. Nor does the Department explain why the vast majority of States and the Federal Government can permit inmates to grow ½-inch beards, either for any reason or for religious reasons, but it cannot. Such evidence requires a prison, at a minimum, to offer persuasive reasons why it believes it must take a different course. See Procunier v. Martinez, 416 U.S. 396, 414, n. 14. Pp. 13–16. 509 Fed. Appx. 561, reversed and remanded. Alito, J., delivered the opinion for a unanimous Court. Ginsburg, J., filed a concurring opinion, in which Sotomayor, J., joined. Sotomayor, J., filed a concurring opinion.

Petitioner Gregory Holt, also known as Abdul Maalik Muhammad, is an Arkansas inmate and a devout Muslim who wishes to grow a 1∕2-inch beard in accordance with his religious beliefs. Petitioner’s objection to shaving his beard clashes with the Arkansas Department of Correction’s grooming policy, which prohibits inmates from growing beards unless they have a particular dermatological condition. We hold that the Department’s policy, as applied in this case, violates the Religious Land Use and Institutionalized Persons Act of 2000 (RLUIPA),114Stat.803,42 U. S. C. §2000cc et seq., which prohibits a stateor local government from taking any action that substantially burdens the religious exercise of an institutionalized person unless the government demonstrates that the action constitutes the least restrictive means of furthering a compelling governmental interest. We conclude in this case that the Department’s policy substantially burdens petitioner’s religious exercise. Although we do not question the importance of the Department’s interests in stopping the flow of contraband and facilitating prisoner identification, we do doubt whether the prohibition against petitioner’s beard furthers its compelling interest about contraband. And we conclude that the Department has failed to show that its policy is the least restrictive means of furthering its compelling interests. We thus reverse the decision of the United States Court of Appeals for the Eighth Circuit.IA Congress enacted RLUIPA and its sister statute, the Religious Freedom Restoration Act of 1993 (RFRA),107Stat.1488,42 U. S. C. §2000bb et seq., “in order to provide very broad protection for religious liberty.” Burwell v. Hobby Lobby Stores, Inc., 573 U. S. ___, ___ (2014) (slip op., at 4). RFRA was enacted three years after our decision in Employment Div., Dept. of Human Resources of Ore. v. Smith,494 U. S. 872 (1990), which held that neutral, generally applicable laws that incidentally burden the exercise of religion usually do not violate the Free Exercise Clause of the First Amendment. Id., at 878–882. Smith largely repudiated the method of analysis used in prior free exercise cases like Wisconsin v. Yoder,406 U. S. 205 (1972), and Sherbert v. Verner,374 U. S. 398 (1963). In those cases, we employed a balancing test that considered whether a challenged government action that substantially burdened the exercise of religion was necessary to further a compelling state interest. See Yoder, supra, at 214, 219; Sherbert, supra, at 403, 406. Following our decision in Smith, Congress enacted RFRA in order to provide greater protection for religious exercise than is available under the First Amendment. See Hobby Lobby, supra, at ___ – ___ (slip op., at 5–6). RFRA provides that “[g]overnment shall not substantially burden a person’s exercise of religion even if the burden results from a rule of general applicability,” unless the government “demonstrates that application of the burden to the person––(1) is in furtherance of a compelling governmental interest; and (2) is the least restrictive means of furthering that compelling governmental interest.” 42 U. S. C. §§2000bb–1(a), (b). In making RFRA applicable to the States and their subdivisions, Congress relied on Section 5 of the Fourteenth Amendment, but in City of Boerne v. Flores,521 U. S. 507 (1997), this Court held that RFRA exceeded Congress’ powers under that provision. Id., at 532–536. Congress responded to City of Boerne by enacting RLUIPA, which applies to the States and their subdivisions and invokes congressional authority under the Spending and Commerce Clauses. See §2000cc–1(b). RLUIPA concerns two areas of government activity: Section 2 governs land-use regulation, §2000cc; and Section 3—the provision at issue in this case—governs religious exercise by institutionalized persons, §2000cc–1. Section 3 mirrors RFRA and provides that “[n]o government shall impose a substantial burden on the religious exercise of a person residing in or confined to an institution . . . even if the burden results from a rule of general applicability, unless the government demonstrates that imposition of the burden on that person––(1) is in furtherance of a compelling governmental interest; and (2) is the least restrictive means of furthering that compelling governmental interest.” §2000cc–1(a). RLUIPA thus allows prisoners “to seek religious accommodations pursuant to the same standard as set forth in RFRA.” Gonzales v. O Centro Espírita Beneficente União do Vegetal,546 U. S. 418,436 (2006). Several provisions of RLUIPA underscore its expansive protection for religious liberty. Congress defined “religious exercise” capaciously to include “any exercise of religion, whether or not compelled by, or central to, a system of religious belief.” §2000cc–5(7)(A). Congress mandated that this concept “shall be construed in favor of a broad protection of religious exercise, to the maximum extent permitted by the terms of this chapter and the Constitution.” §2000cc–3(g). And Congress stated that RLUIPA “may require a government to incur expenses in its own operations to avoid imposing a substantial burden on religious exercise.” §2000cc–3(c). See Hobby Lobby, supra, at ___ – ___, ___ (slip op., at 6–7, 43).B Petitioner, as noted, is in the custody of the Arkansas Department of Correction and he objects on religious grounds to the Department’s grooming policy, which provides that “[n]o inmates will be permitted to wear facial hair other than a neatly trimmed mustache that does not extend beyond the corner of the mouth or over the lip.” App. to Brief for Petitioner 11a. The policy makes no exception for inmates who object on religious grounds, but it does contain an exemption for prisoners with medical needs: “Medical staff may prescribe that inmates with a diagnosed dermatological problem may wear facial hair no longer than one quarter of an inch.” Ibid. The policy provides that “[f]ailure to abide by [the Department’s] grooming standards is grounds for disciplinary action.” Id., at 12a. Petitioner sought permission to grow a beard and, al-though he believes that his faith requires him not to trim his beard at all, he proposed a “compromise” under which he would grow only a 1∕2-inch beard. App. 164. Prison officials denied his request, and the warden told him: “[Y]ou will abide by [Arkansas Department of Correction] policies and if you choose to disobey, you can suffer the consequences.” No. 5:11–cv–00164 (ED Ark., July 21, 2011), Doc. 13, p. 6 (Letter from Gaylon Lay to Gregory Holt (July 19, 2011)). Petitioner filed a pro se complaint in Federal District Court challenging the grooming policy under RLUIPA. We refer to the respondent prison officials collectively as the Department. In October 2011, the District Court granted petitioner a preliminary injunction and remanded to a Magistrate Judge for an evidentiary hearing. At the hearing, the Department called two witnesses. Both expressed the belief that inmates could hide contraband in even a 1∕2-inch beard, but neither pointed to any instances in which this had been done in Arkansas or elsewhere. Both witnesses also acknowledged that inmates could hide items in many other places, such as in the hair on their heads or their clothing. In addition, one of the witnesses—Gaylon Lay, the warden of petitioner’s prison—testified that a prisoner who escaped could change his appearance by shaving his beard, and that a prisoner could shave his beard to disguise himself and enter a restricted area of the prison. Neither witness, however, was able to explain why these problems could not be addressed by taking a photograph of an inmate without a beard, a practice followed in other prison systems. Lay voiced concern that the Department would be unable to monitor the length of a prisoner’s beard to ensure that it did not exceed one-half inch, but he acknowledged that the Department kept track of the length of the beards of those inmates who are allowed to wear a 1∕4-inch beard for medical reasons. As a result of the preliminary injunction, petitioner had a short beard at the time of the hearing, and the Magistrate Judge commented: “I look at your particular circumstance and I say, you know, it’s almost preposterous to think that you could hide contraband in your beard.” App. 155. Nevertheless, the Magistrate Judge recommended that the preliminary injunction be vacated and that petitioner’s complaint be dismissed for failure to state a claim on which relief can be granted. The Magistrate Judge emphasized that “the prison officials are entitled to deference,” id., at 168, and that the grooming policy allowed petitioner to exercise his religion in other ways, such as by praying on a prayer rug, maintaining the diet required by his faith, and observing religious holidays. The District Court adopted the Magistrate Judge’s recommendation in full, and the Court of Appeals for the Eighth Circuit affirmed in a brief per curiam opinion, holding that the Department had satisfied its burden of showing that the grooming policy was the least restrictive means of furthering its compelling security interests. 509 Fed. Appx. 561 (2013). The Court of Appeals stated that “courts should ordinarily defer to [prison officials’] expert judgment” in security matters unless there is substantial evidence that a prison’s response is exaggerated. Id., at 562. And while acknowledging that other prisons allow inmates to maintain facial hair, the Eighth Circuit held that this evidence “does not outweigh deference owed to [the] expert judgment of prison officials who are more familiar with their own institutions.” Ibid. We entered an injunction pending resolution of petitioner’s petition for writ of certiorari, 571 U. S. ___ (2013), and we then granted certiorari, 571 U. S. ___ (2014).II Under RLUIPA, petitioner bore the initial burden of proving that the Department’s grooming policy implicates his religious exercise. RLUIPA protects “any exercise of religion, whether or not compelled by, or central to, a system of religious belief,” §2000cc–5(7)(A), but, of course, a prisoner’s request for an accommodation must be sincerely based on a religious belief and not some other motivation, see Hobby Lobby, 573 U. S., at ___, n. 28 (slip op., at 29, n. 28). Here, the religious exercise at issue is the growing of a beard, which petitioner believes is a dictate of his religious faith, and the Department does not dispute the sincerity of petitioner’s belief. In addition to showing that the relevant exercise of religion is grounded in a sincerely held religious belief, petitioner also bore the burden of proving that the Department’s grooming policy substantially burdened that exercise of religion. Petitioner easily satisfied that obligation. The Department’s grooming policy requires petitioner to shave his beard and thus to “engage in conduct that seriously violates [his] religious beliefs.” Id., at ___ (slip op., at 32). If petitioner contravenes that policy and grows his beard, he will face serious disciplinary action. Because the grooming policy puts petitioner to this choice, it substantially burdens his religious exercise. Indeed, the Department does not argue otherwise. The District Court reached the opposite conclusion, but its reasoning (adopted from the recommendation of the Magistrate Judge) misunderstood the analysis that RLUIPA demands. First, the District Court erred by concluding that the grooming policy did not substantially burden petitioner’s religious exercise because “he had been provided a prayer rug and a list of distributors of Islamic material, he was allowed to correspond with a religious advisor, and was allowed to maintain the required diet and observe religious holidays.” App. 177. In taking this approach, the District Court improperly imported a strand of reasoning from cases involving prisoners’ First Amendment rights. See, e.g., O’Lone v. Estate of Shabazz,482 U. S. 342–352 (1987); see also Turner v. Safley,482 U. S. 78,90 (1987). Under those cases, the availability of alternative means of practicing religion is a relevant consideration, but RLUIPA provides greater protection. RLUIPA’s “substantial burden” inquiry asks whether the government has substantially burdened religious exercise (here, the growing of a 1∕2-inch beard), not whether the RLUIPA claimant is able to engage in other forms of religious exercise. Second, the District Court committed a similar error in suggesting that the burden on petitioner’s religious exercise was slight because, according to petitioner’s testi-mony, his religion would “credit” him for attempting to follow his religious beliefs, even if that attempt provedto be unsuccessful. RLUIPA, however, applies to an exercise of religion regardless of whether it is “compelled.” §2000cc–5(7)(A). Finally, the District Court went astray when it relied on petitioner’s testimony that not all Muslims believe that men must grow beards. Petitioner’s belief is by no means idiosyncratic. See Brief for Islamic Law Scholars as Amici Curiae 2 (“hadith requiring beards . . . are widely followed by observant Muslims across the various schools of Islam”). But even if it were, the protection of RLUIPA, no less than the guarantee of the Free Exercise Clause, is “not limited to beliefs which are shared by all of the members of a religious sect.” Thomas v. Review Bd. of Indiana Employment Security Div.,450 U. S. 707–716 (1981).III Since petitioner met his burden of showing that the Department’s grooming policy substantially burdened his exercise of religion, the burden shifted to the Department to show that its refusal to allow petitioner to grow a 1∕2-inch beard “(1) [was] in furtherance of a compelling governmental interest; and (2) [was] the least restrictive means of furthering that compelling governmental interest.” §2000cc–1(a). The Department argues that its grooming policy represents the least restrictive means of furthering a “ ‘broadly formulated interes[t],’ ” see Hobby Lobby, supra, at ___ (slip op., at 39) (quoting O Centro, 546 U. S., at 431), namely, the Department’s compelling interest in prison safety and security. But RLUIPA, like RFRA, contemplates a “ ‘more focused’ ” inquiry and “ ‘requires the Government to demonstrate that the compelling interest test is satisfied through application of the challenged law “to the person”––the particular claimant whose sincere exercise of religion is being substantially burdened.’ ” Hobby Lobby, 573 U. S., at ___ (slip op., at 39) (quoting O Centro, supra, at 430–431 (quoting §2000bb–1(b))). RLUIPA requires us to “ ‘scrutiniz[e] the asserted harm of granting specific exemptions to particular religious claimants’ ” and “to look to the marginal interest in enforcing” the challenged government action in that particular context. Hobby Lobby, supra, at ___ (slip op., at 39) (quoting O Centro, supra, at 431; alteration in original). In this case, that means the enforcement of the Department’s policy to prevent petitioner from growing a 1∕2-inch beard. The Department contends that enforcing this prohibition is the least restrictive means of furthering prison safety and security in two specific ways.A The Department first claims that the no-beard policy prevents prisoners from hiding contraband. The Department worries that prisoners may use their beards to conceal all manner of prohibited items, including razors, needles, drugs, and cellular phone subscriber identity module (SIM) cards. We readily agree that the Department has a compelling interest in staunching the flow of contraband into and within its facilities, but the argument that this interest would be seriously compromised by allowing an inmate to grow a 1∕2-inch beard is hard to take seriously. As noted, the Magistrate Judge observed that it was “almost preposterous to think that [petitioner] could hide contraband” in the short beard he had grown at the time of the evidentiary hearing. App. 155. An item of contraband would have to be very small indeed to be concealed by a 1∕2-inch beard, and a prisoner seeking to hide an item in such a short beard would have to find a way to prevent the item from falling out. Since the Department does not demand that inmates have shaved heads or short crew cuts, it is hard to see why an inmate would seek to hide contraband in a 1∕2-inch beard rather than in the longer hair on his head. Although the Magistrate Judge dismissed the possibility that contraband could be hidden in a short beard, the Magistrate Judge, the District Court, and the Court of Appeals all thought that they were bound to defer to the Department’s assertion that allowing petitioner to grow such a beard would undermine its interest in suppressing contraband. RLUIPA, however, does not permit such unquestioning deference. RLUIPA, like RFRA, “makes clear that it is the obligation of the courts to consider whether exceptions are required under the test set forth by Congress.” O Centro, supra, at 434. That test requires the Department not merely to explain why it denied the exemption but to prove that denying the exemption is the least restrictive means of furthering a compelling governmental interest. Prison officials are experts in running prisons and evaluating the likely effects of altering prison rules, and courts should respect that expertise. But that respect does not justify the abdication of the responsibility, conferred by Congress, to apply RLUIPA’s rigorous standard. And without a degree of deference that is tantamount to unquestioning acceptance, it is hard to swallow the argument that denying petitioner a 1∕2-inch beard actually furthers the Department’s interest in rooting out contraband. Even if the Department could make that showing, its contraband argument would still fail because the Department cannot show that forbidding very short beards is the least restrictive means of preventing the concealment of contraband. “The least-restrictive-means standard is exceptionally demanding,” and it requires the government to “sho[w] that it lacks other means of achieving its desired goal without imposing a substantial burden on the exercise of religion by the objecting part[y].” Hobby Lobby, supra, at ___ (slip op., at 40). “[I]f a less restrictive means is available for the Government to achieve its goals, the Government must use it.” United States v. Playboy Entertainment Group, Inc.,529 U. S. 803,815 (2000). The Department failed to establish that it could not satisfy its security concerns by simply searching petitioner’s beard. The Department already searches prisoners’ hair and clothing, and it presumably examines the 1∕4-inch beards of inmates with dermatological conditions. It has offered no sound reason why hair, clothing, and 1∕4-inch beards can be searched but 1∕2-inch beards cannot. The Department suggests that requiring guards to search a prisoner’s beard would pose a risk to the physical safety of a guard if a razor or needle was concealed in the beard. But that is no less true for searches of hair, clothing, and 1∕4-inch beards. And the Department has failed to prove that it could not adopt the less restrictive alternative of having the prisoner run a comb through his beard. For all these reasons, the Department’s interest in eliminating contraband cannot sustain its refusal to allow petitioner to grow a 1∕2-inch beard.B The Department contends that its grooming policy is necessary to further an additional compelling interest, i.e., preventing prisoners from disguising their identities. The Department tells us that the no-beard policy allows secu-rity officers to identify prisoners quickly and accurately. It claims that bearded inmates could shave their beards and change their appearance in order to enter restricted areas within the prison, to escape, and to evade apprehension after escaping. We agree that prisons have a compelling interest in the quick and reliable identification of prisoners, and we acknowledge that any alteration in a prisoner’s appearance, such as by shaving a beard, might, in the absence of effective countermeasures, have at least some effect on the ability of guards or others to make a quick identification. But even if we assume for present purposes that the Department’s grooming policy sufficiently furthers its interest in the identification of prisoners, that policy still violates RLUIPA as applied in the circumstances present here. The Department contends that a prisoner who has a beard when he is photographed for identification purposes might confuse guards by shaving his beard. But as petitioner has argued, the Department could largely solve this problem by requiring that all inmates be photographed without beards when first admitted to the facility and, if necessary, periodically thereafter. Once that is done, an inmate like petitioner could be allowed to grow a short beard and could be photographed again when the beard reached the 1∕2-inch limit. Prison guards would then have a bearded and clean-shaven photo to use in making identifications. In fact, the Department (like many other States, see Brief for Petitioner 39) already has a policy of photographing a prisoner both when he enters an institution and when his “appearance changes at any time during [his] incarceration.” Arkansas Department of Correction, Inmate Handbook 3–4 (rev. Jan. 2013). The Department argues that the dual-photo method is inadequate because, even if it might help authorities apprehend a bearded prisoner who escapes and then shaves his beard once outside the prison, this method is unlikely to assist guards when an inmate quickly shaves his beard in order to alter his appearance within the prison. The Department contends that the identification concern is particularly acute at petitioner’s prison, where inmates live in barracks and work in fields. Counsel for the Department suggested at oral argument that a pris-oner could gain entry to a restricted area by shavinghis beard and swapping identification cards with an-other inmate while out in the fields. Tr. of Oral Arg. 28–30, 39–43. We are unpersuaded by these arguments for at least two reasons. First, the Department failed to show, in the face of petitioner’s evidence, that its prison system is so different from the many institutions that allow facial hair that the dual-photo method cannot be employed at its institutions. Second, the Department failed to establish why the risk that a prisoner will shave a 1∕2-inch beard to disguise himself is so great that 1∕2-inch beards cannot be allowed, even though prisoners are allowed to grow mustaches, head hair, or 1∕4-inch beards for medical reasons. All of these could also be shaved off at a moment’s notice, but the Department apparently does not think that this possibility raises a serious security concern.C In addition to its failure to prove that petitioner’s proposed alternatives would not sufficiently serve its security interests, the Department has not provided an adequate response to two additional arguments that implicate the RLUIPA analysis. First, the Department has not adequately demonstrated why its grooming policy is substantially underinclusive in at least two respects. Although the Department denied petitioner’s request to grow a 1∕2-inch beard, it permits prisoners with a dermatological condition to grow 1∕4-inch beards. The Department does this even though both beards pose similar risks. And the Department permits inmates to grow more than a 1∕2-inch of hair on their heads. With respect to hair length, the grooming policy provides only that hair must be worn “above the ear” and “no longer in the back than the middle of the nape of the neck.” App. to Brief for Petitioner 11a. Hair on the head is a more plausible place to hide contraband than a 1∕2-inch beard—and the same is true of an inmate’s clothing and shoes. Nevertheless, the Department does not require inmates to go about bald, barefoot, or naked. Although the Department’s proclaimed objectives are to stop the flow of contraband and to facilitate prisoner identification, “[t]he proffered objectives are not pursued with respect to analogous nonreligious conduct,” which suggests that “those interests could be achieved by narrower ordinances that burdened religion to a far lesser degree.” Church of Lukumi Babalu Aye, Inc. v. Hialeah,508 U. S. 520,546 (1993). In an attempt to demonstrate why its grooming policy is underinclusive in these respects, the Department emphasizes that petitioner’s 1∕2-inch beard is longer than the 1∕4-inch beard allowed for medical reasons. But the Department has failed to establish (and the District Court did not find) that a 1∕4-inch difference in beard length poses a meaningful increase in security risk. The Department also asserts that few inmates require beards for medical reasons while many may request beards for religious reasons. But the Department has not argued that denying petitioner an exemption is necessary to further a compelling interest in cost control or program administration. At bottom, this argument is but another formulation of the “classic rejoinder of bureaucrats throughout history: If I make an exception for you, I’ll have to make one for everybody, so no exceptions.” O Centro, 546 U. S., at 436. We have rejected a similar argument in analogous contexts, see ibid.; Sherbert, 374 U. S., at 407, and we reject it again today. Second, the Department failed to show, in the face of petitioner’s evidence, why the vast majority of States and the Federal Government permit inmates to grow 1∕2-inch beards, either for any reason or for religious reasons, but it cannot. See Brief for Petitioner 24–25; Brief for United States as Amicus Curiae 28–29. “While not necessarily controlling, the policies followed at other well-run institutions would be relevant to a determination of the need for a particular type of restriction.” Procunier v. Martinez,416 U. S. 396, n. 14 (1974). That so many other prisons allow inmates to grow beards while ensuring prison safety and security suggests that the Department could satisfy its security concerns through a means less restrictive than denying petitioner the exemption he seeks. We do not suggest that RLUIPA requires a prison to grant a particular religious exemption as soon as a few other jurisdictions do so. But when so many prisons offer an accommodation, a prison must, at a minimum, offer persuasive reasons why it believes that it must take a different course, and the Department failed to make that showing here. Despite this, the courts below deferred to these prison officials’ mere say-so that they could not accommodate petitioner’s request. RLUIPA, however, demands much more. Courts must hold prisons to their statutory burden, and they must not “assume a plausible, less restrictive alternative would be ineffective.” Playboy Entertainment, 529 U. S., at 824. We emphasize that although RLUIPA provides substantial protection for the religious exercise of institutionalized persons, it also affords prison officials ample ability to maintain security. We highlight three ways in which this is so. First, in applying RLUIPA’s statutory standard, courts should not blind themselves to the fact that the analysis is conducted in the prison setting. Second, if an institution suspects that an inmate is using religious activity to cloak illicit conduct, “prison officials may appropriately question whether a prisoner’s religiosity, asserted as the basis for a requested accommodation, is authentic.” Cutter v. Wilkinson,544 U. S. 709, n. 13 (2005). See also Hobby Lobby, 573 U. S., at ___, n. 28 (slip op., at 29, n. 28). Third, even if a claimant’s religious belief is sincere, an institution might be entitled to withdraw an accommodation if the claimant abuses the exemption in a manner that undermines the prison’s compelling interests.IV In sum, we hold that the Department’s grooming policy violates RLUIPA insofar as it prevents petitioner from growing a 1∕2-inch beard in accordance with his religious beliefs. The judgment of the United States Court of Appeals for the Eighth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion.It is so ordered.

574.US.27

Petitioner Integrity Staffing Solutions, Inc., required its hourly warehouse workers, who retrieved products from warehouse shelves and packaged them for delivery to Amazon.com customers, to undergo a security screening before leaving the warehouse each day. Respondents, former employees, sued the company alleging, as relevant here, that they were entitled to compensation under the Fair Labor Standards Act of 1938 (FLSA) for the roughly 25 minutes each day that they spent waiting to undergo and undergoing those screenings. They also alleged that the company could have reduced that time to a de minimis amount by adding screeners or staggering shift terminations and that the screenings were conducted to prevent employee theft and, thus, for the sole benefit of the employers and their customers. The District Court dismissed the complaint for failure to state a claim, holding that the screenings were not integral and indispensable to the employees’ principal activities but were instead postliminary and noncompensable. The U. S. Court of Appeals for the Ninth Circuit reversed in relevant part, asserting that postshift activities that would ordinarily be classified as noncompensable postliminary activities are compensable as integral and indispensable to an employee’s principal activities if the postshift activities are necessary to the principal work and performed for the employer’s benefit. Held: The time that respondents spent waiting to undergo and undergoing security screenings is not compensable under the FLSA. Pp. 3–9. (a) Congress passed the Portal-to-Portal Act to respond to an economic emergency created by the broad judicial interpretation given to the FLSA’s undefined terms “work” and “workweek.” See 29 U. S. C. §251(a); Tennessee Coal, Iron & R. Co. v. Muscoda Local No. 123, 321 U.S. 590, 598. The Portal-to-Portal Act exempted employers from FLSA liability for claims based on “activities which are preliminary to or postliminary to” the performance of the principal activities that an employee is employed to perform. §254(a)(2). Under this Court’s precedents, the term “principal activities” includes all activities which are an “integral and indispensable part of the principal activities.” Steiner v. Mitchell, 350 U.S. 247, 252–253. An activity is “integral and indispensable” if it is an intrinsic element of the employee’s principal activities and one with which the employee cannot dispense if he is to perform his principal activities. This Court has identified several activities that satisfy this test—see, e.g., id., at 249, 251; Mitchell v. King Packing Co., 350 U.S. 260, 262—and Department of Labor regulations are consistent with this approach, see 29 CFR §§790.8(c), 790.7(g). Pp. 3–7. (b) The security screenings at issue are noncompensable postliminary activities. To begin with, the screenings were not the principal activities the employees were employed to perform—i.e., the workers were employed not to undergo security screenings but to retrieve products from warehouse shelves and package them for shipment. Nor were they “integral and indispensable” to those activities. This view is consistent with a 1951 Department of Labor opinion letter, which found noncompensable under the Portal-to-Portal Act both a preshift screening conducted for employee safety and a postshift search conducted to prevent employee theft. The Ninth Circuit’s test, which focused on whether the particular activity was required by the employer rather than whether it was tied to the productive work that the employee was employed to perform, would sweep into “principal activities” the very activities that the Portal-to-Portal Act was designed to exclude from compensation. See, e.g., IBP, supra, at 41. Finally, respondents’ claim that the screenings are compensable because Integrity Staffing could have reduced the time to a de minimis amount is properly presented at the bargaining table, not to a court in an FLSA claim. Pp. 7–9. 713 F.3d 525, reversed. Thomas, J., delivered the opinion for a unanimous Court. Sotomayor, J., filed a concurring opinion, in which Kagan, J., joined.

The employer in this case required its employees, warehouse workers who retrieved inventory and packaged itfor shipment, to undergo an antitheft security screen-ing before leaving the warehouse each day. The question presented is whether the employees’ time spent waiting to undergo and undergoing those security screenings is compensable under the Fair Labor Standards Act of 1938 (FLSA), 29 U. S. C. §201 et seq., as amended by the Portal-to-Portal Act of 1947, §251 et seq. We hold that the time is not compensable. We therefore reverse the judgment of the United States Court of Appeals for the Ninth Circuit. I Petitioner Integrity Staffing Solutions, Inc., provides warehouse staffing to Amazon.com throughout the United States. Respondents Jesse Busk and Laurie Castro worked as hourly employees of Integrity Staffing at warehouses in Las Vegas and Fenley, Nevada, respectively. As warehouse employees, they retrieved products from the shelves and packaged those products for delivery to Amazon customers. Integrity Staffing required its employees to undergo a security screening before leaving the warehouse at the end of each day. During this screening, employees removed items such as wallets, keys, and belts from their persons and passed through metal detectors. In 2010, Busk and Castro filed a putative class action against Integrity Staffing on behalf of similarly situated employees in the Nevada warehouses for alleged violations of the FLSA and Nevada labor laws. As relevant here, the employees alleged that they were entitled to compensation under the FLSA for the time spent waiting to undergo and actually undergoing the security screenings. They alleged that such time amounted to roughly 25 minutes each day and that it could have been reduced to a de minimis amount by adding more security screeners or by staggering the termination of shifts so that employees could flow through the checkpoint more quickly. They also alleged that the screenings were conducted “to prevent employee theft” and thus occurred “solely for the benefit of the employers and their customers.” App. 19, 21. The District Court dismissed the complaint for failure to state a claim, holding that the time spent waiting for and undergoing the security screenings was not compensable under the FLSA. It explained that, because the screenings occurred after the regular work shift, the employees could state a claim for compensation only if the screenings were an integral and indispensable part of the principal activities they were employed to perform. The District Court held that these screenings were not integral and indispensable but instead fell into a noncompensable category of postliminary activities. The United States Court of Appeals for the Ninth Circuit reversed in relevant part. 713 F. 3d 525 (2013). The Court of Appeals asserted that postshift activities that would ordinarily be classified as noncompensable postliminary activities are nevertheless compensable as integral and indispensable to an employee’s principal activities if those postshift activities are necessary to the principal work performed and done for the benefit of the employer. Id., at 530. Accepting as true the allegation that Integrity Staffing required the security screenings to prevent employee theft, the Court of Appeals concluded that the screenings were “necessary” to the employees’ primary work as warehouse employees and done for Integrity Staffing’s benefit. Id., at 531. We granted certiorari, 571 U. S. ___ (2014), and now reverse. II A Enacted in 1938, the FLSA established a minimum wage and overtime compensation for each hour worked in excess of 40 hours in each workweek. §§6(a)(1), 7(a)(3), 52Stat. 1062–1063. An employer who violated these provisions could be held civilly liable for backpay, liquidated damages, and attorney’s fees. §16, id., at 1069. But the FLSA did not define “work” or “workweek,” and this Court interpreted those terms broadly. It defined “work” as “physical or mental exertion (whether burdensome or not) controlled or required by the employer and pursued necessarily and primarily for the benefit of the employer and his business.” Tennessee Coal, Iron & R. Co. v. Muscoda Local No. 123, 321 U. S. 590, 598 (1944) . Similarly, it defined “the statutory workweek” to “includ[e] all time during which an employee is necessarily required to be on the employer’s premises, on duty or at a prescribed workplace.” Anderson v. Mt. Clemens Pottery Co., 328 U. S. 680 –691 (1946). Applying these expansive definitions, the Court found compensable the time spent traveling between mine portals and underground work areas, Tennessee Coal, supra, at 598, and the time spent walking from timeclocks to work benches, Anderson, supra, at 691–692. These decisions provoked a flood of litigation. In the six months following this Court’s decision in Anderson, unions and employees filed more than 1,500 lawsuits under the FLSA. S. Rep. No. 37, 80th Cong., 1st Sess., pp. 2–3 (1947). These suits sought nearly $6 billion in back pay and liquidated damages for various preshift and postshift activities. Ibid. Congress responded swiftly. It found that the FLSAhad “been interpreted judicially in disregard of long-established customs, practices, and contracts between employers and employees, thereby creating wholly unexpected liabilities, immense in amount and retroactive in operation, upon employers.” 29 U. S. C. §251(a). Declaring the situation to be an “emergency,” Congress found that, if such interpretations “were permitted to stand, . . . the payment of such liabilities would bring about financial ruin of many employers” and “employees would receive windfall payments . . . for activities performed by them without any expectation of reward beyond that included in their agreed rates of pay.” §§251(a)–(b). Congress met this emergency with the Portal-to-Portal Act. The Portal-to-Portal Act exempted employers from liability for future claims based on two categories of work-related activities as follows: “(a) Except as provided in subsection (b) [which covers work compensable by contract or custom], no employer shall be subject to any liability or punishment under the Fair Labor Standards Act of 1938, as amended, . . . on account of the failure of such employer . . .to pay an employee overtime compensation, for or on account of any of the following activities of such employee engaged in on or after the date of the enactment of this Act— “(1) walking, riding, or traveling to and from the ac-tual place of performance of the principal activity or ac-tivities which such employee is employed to perform, and “(2) activities which are preliminary to or postliminary to said principal activity or activities, “which occur either prior to the time on any particular workday at which such employee commences, or subsequent to the time on any particular workday at which he ceases, such principal activity or activities.” §4, 61Stat. 86–87 (codified at 29 U. S. C. §254(a)). At issue here is the exemption for “activities which are preliminary to or postliminary to said principal activity or activities.” B This Court has consistently interpreted “the term ‘principal activity or activities’ [to] embrac[e] all activities which are an ‘integral and indispensable part of the principal activities.’ ” IBP, Inc. v. Alvarez, 546 U. S. 21 –30 (2005) (quoting Steiner v. Mitchell, 350 U. S. 247 –253 (1956)). Our prior opinions used those words in their ordinary sense. The word “integral” means “[b]elonging to or making up an integral whole; constituent, component; spec[ifically] necessary to the completeness or integrity of the whole; forming an intrinsic portion or element, as distinguished from an adjunct or appendage.” 5 Oxford English Dictionary 366 (1933) (OED); accord, Brief for United States as Amicus Curiae 20 (Brief for United States); see also Webster’s New International Dictionary 1290 (2d ed. 1954) (Webster’s Second) (“[e]ssential to completeness; constituent, as a part”). And, when used to describe a duty, “indispensable” means a duty “[t]hat cannot be dispensed with, remitted, set aside, disregarded, or neglected.” 5 OED 219; accord, Brief for United States 19; see also Webster’s Second 1267 (“[n]ot capable of being dispensed with, set aside, neglected, or pronounced nonobligatory”). An activity is therefore integral and indispensable to the principal activities that an employee is employed to perform if it is an intrinsic element of those activities and one with which the employee cannot dispense if he is to perform his principal activities. As we describe below, this definition, as applied in these circumstances, is consistent with the Department of Labor’s regulations. Our precedents have identified several activities that satisfy this test. For example, we have held compensable the time battery-plant employees spent showering and changing clothes because the chemicals in the plant were “toxic to human beings” and the employer conceded that “the clothes-changing and showering activities of the employees [were] indispensable to the performance of their productive work and integrally related thereto.” Steiner, supra, at 249, 251. And we have held compensable the time meatpacker employees spent sharpening their knives because dull knives would “slow down production” on the assembly line, “affect the appearance of the meat as well as the quality of the hides,” “cause waste,” and lead to “accidents.” Mitchell v. King Packing Co., 350 U. S. 260, 262 (1956) . By contrast, we have held noncompensable the time poultry-plant employees spent waiting to don protective gear because such waiting was “two steps removed from the productive activity on the assembly line.” IBP, supra, at 42. The Department of Labor’s regulations are consistent with this approach. See 29 CFR §790.8(b) (2013) (“The term ‘principal activities’ includes all activities which are an integral part of a principal activity”); §790.8(c) (“Among the activities included as an integral part of a principal activity are those closely related activities which are indispensable to its performance”). As an illustration, those regulations explain that the time spent by an employee in a chemical plant changing clothes would be compensable if he “c[ould not] perform his principal activities without putting on certain clothes” but would not be compensable if “changing clothes [were] merely a convenience to the employee and not directly related to his principal activities.” See §790.8(c). As the regulations explain, “when performed under the conditions normally present,” activities including “checking in and out and waiting in line to do so, changing clothes, washing up or showering, and waiting in line to receive pay checks” are “ ‘preliminary’ ” or “ ‘postliminary’ ” activities. §790.7(g). III A The security screenings at issue here are noncompensable postliminary activities. To begin with, the screenings were not the “principal activity or activities which [the] employee is employed to perform.” 29 U. S. C. §254(a)(1). Integrity Staffing did not employ its workers to undergo security screenings, but to retrieve products from warehouse shelves and package those products for shipment to Amazon customers. The security screenings also were not “integral and indispensable” to the employees’ duties as warehouse workers. As explained above, an activity is not integral and indispensable to an employee’s principal activities unless it is an intrinsic element of those activities and one with which the employee cannot dispense if he is to perform those activities. The screenings were not an intrinsic element of retrieving products from warehouse shelves or packaging them for shipment. And Integrity Staffing could have eliminated the screenings altogether without impairing the employees’ ability to complete their work. The Solicitor General, adopting the position of the Department of Labor, agrees that these screenings were noncompensable postliminary activities. See Brief for United States 10. That view is fully consistent with an Opinion Letter the Department issued in 1951. The letter found noncompensable a preshift security search of employees in a rocket-powder plant “ ‘for matches, spark producing devices such as cigarette lighters, and other items which have a direct bearing on the safety of the employees,’ ” as well as a postshift security search of the employees done “ ‘for the purpose of preventing theft.’ ” Opinion Letter from Dept. of Labor, Wage and Hour Div., to Dept. of Army, Office of Chief of Ordnance (Apr. 18, 1951), pp. 1–2 (available in Clerk of Court’s case file). The Department drew no distinction between the searches conducted for the safety of the employees and those conducted for the purpose of preventing theft—neither were compensable under the Portal-to-Portal Act. B The Court of Appeals erred by focusing on whether an employer required a particular activity. The integral and indispensable test is tied to the productive work that the employee is employed to perform. See, e.g., IBP, 546 U. S., at 42; Mitchell, supra, at 262; Steiner, 350 U. S., at 249–251; see also 29 CFR §790.8(a) (explaining that the term “principal activities” was “considered sufficiently broad to embrace within its terms such activities as are indispensable to the performance of productive work” (internal quotation marks omitted; emphasis added)); §790.8(c) (“Among the activities included as an integral part of a principal activity are those closely related activities which are indispensable to its performance” (emphasis added)). If the test could be satisfied merely by the fact that an employer required an activity, it would sweep into “principal activities” the very activities that the Portal-to-Portal Act was designed to address. The employer in Anderson, for instance, required its employees to walk “from a timeclock near the factory gate to a workstation” so that they could “begin their work,” “but it is indisputable that the Portal-to-Portal Act evinces Congress’ intent to repudiate Anderson’s holding that such walking time was compensable under the FLSA.” IBP, supra, at 41. A test that turns on whether the activity is for the benefit of the employer is similarly overbroad. Finally, we reject the employees’ argument that time spent waiting to undergo the security screenings is compensable under the FLSA because Integrity Staffing could have reduced that time to a de minimis amount. The fact that an employer could conceivably reduce the time spent by employees on any preliminary or postliminary activity does not change the nature of the activity or its relationship to the principal activities that an employee is employed to perform. These arguments are properly presented to the employer at the bargaining table, see 29 U. S. C. §254(b)(1), not to a court in an FLSA claim. * * * We hold that an activity is integral and indispensable to the principal activities that an employee is employed to perform—and thus compensable under the FLSA—if it is an intrinsic element of those activities and one with which the employee cannot dispense if he is to perform his principal activities. Because the employees’ time spent waiting to undergo and undergoing Integrity Staffing’s security screenings does not meet these criteria, we reverse the judgment of the Court of Appeals. It is so ordered.

574.US.259

Exactly three years after borrowing money from respondent Countrywide Home Loans, Inc., to refinance their home mortgage, petitioners Larry and Cheryle Jesinoski sent Countrywide and respondent Bank of America Home Loans, which had acquired Countrywide, a letter purporting to rescind the transaction. Bank of America replied, refusing to acknowledge the rescission’s validity. One year and one day later, the Jesinoskis filed suit in federal court, seeking a declaration of rescission and damages. The District Court entered judgment on the pleadings for respondents, concluding that a borrower can exercise the Truth in Lending Act’s right to rescind a loan, see 15 U. S. C. §1635(a), (f), only by filing a lawsuit within three years of the date the loan was consummated. The Jesinoskis’ complaint, filed four years and one day after the loan’s consummation, was ineffective. The Eighth Circuit affirmed. Held: A borrower exercising his right to rescind under the Act need only provide written notice to his lender within the 3-year period, not file suit within that period. Section 1635(a)’s unequivocal terms—a borrower “shall have the right to rescind . . . by notifying the creditor . . . of his intention to do so” (emphasis added)—leave no doubt that rescission is effected when the borrower notifies the creditor of his intention to rescind. This conclusion is not altered by §1635(f), which states when the right to rescind must be exercised, but says nothing about how that right is exercised. Nor does §1635(g)—which states that “in addition to rescission the court may award relief . . . not relating to the right to rescind”—support respondents’ view that rescission is necessarily a consequence of judicial action. And the fact that the Act modified the common-law condition precedent to rescission at law, see §1635(b), hardly implies that the Act thereby codified rescission in equity. Pp. 2–5. 729 F.3d 1092, reversed and remanded. Scalia, J., delivered the opinion for a unanimous Court.

The Truth in Lending Act gives borrowers the right to rescind certain loans for up to three years after the transaction is consummated. The question presented is whether a borrower exercises this right by providing written no-tice to his lender, or whether he must also file a lawsuit before the 3-year period elapses. On February 23, 2007, petitioners Larry and Cheryle Jesinoski refinanced the mortgage on their home by borrowing $611,000 from respondent Countrywide Home Loans, Inc. Exactly three years later, on February 23, 2010, the Jesinoskis mailed respondents a letter purporting to rescind the loan. Respondent Bank of America Home Loans replied on March 12, 2010, refusing to acknowledge the validity of the rescission. On February 24, 2011, the Jesinoskis filed suit in Federal District Court seeking a declaration of rescission and damages. Respondents moved for judgment on the pleadings, which the District Court granted. The court concluded that the Act requires a borrower seeking rescission to file a lawsuit within three years of the transaction’s consummation. Although the Jesinoskis notified respondents of their intention to rescind within that time, they did not file their first complaint until four years and one day after the loan’s consummation. 2012 WL 1365751, *3 (D Minn., Apr. 19, 2012). The Eighth Circuit affirmed. 729 F. 3d 1092, 1093 (2013) (per curiam). Congress passed the Truth in Lending Act,82Stat.146, as amended, to help consumers “avoid the uninformed use of credit, and to protect the consumer against inaccurate and unfair credit billing.”15 U. S. C. §1601(a). To this end, the Act grants borrowers the right to rescind a loan “until midnight of the third business day following the consummation of the transaction or the delivery of the [disclosures required by the Act], whichever is later, by notifying the creditor, in accordance with regulations of the [Federal Reserve] Board, of his intention to do so.” §1635(a) (2006 ed.).[1] This regime grants borrowers an unconditional right to rescind for three days, after which they may rescind only if the lender failed to satisfy the Act’s disclosure requirements. But this conditional right to rescind does not last forever. Even if a lender never makes the required disclosures, the “right of rescission shall expire three years after the date of consummation of the transaction or upon the sale of the property, whichever comes first.” §1635(f). The Eighth Circuit’s affirmance in the present case rested upon its holding in Keiran v. Home Capital, Inc., 720 F. 3d 721, 727–728 (2013) that, unless a borrower has filed a suit for rescission within three years of the transaction’s consummation, §1635(f) extinguishes the right to rescind and bars relief. That was error. Section 1635(a) explains in unequivocal terms how the right to rescind is to be exercised: It provides that a borrower “shall have the right to rescind . . . by notifying the creditor, in accordance with regulations of the Board, of his intention to do so” (emphasis added). The language leaves no doubt that rescission is effected when the borrower notifies the creditor of his intention to rescind. It follows that, so long as the borrower notifies within three years after the transaction is consummated, his rescission is timely. The statute does not also require him to sue within three years. Nothing in §1635(f) changes this conclusion. Although §1635(f) tells us when the right to rescind must be exercised, it says nothing about how that right is exercised. Our observation in Beach v. Ocwen Fed. Bank,523 U. S. 410,417 (1998), that §1635(f) “govern[s] the life of the underlying right” is beside the point. That case concerned a borrower’s attempt to rescind in the course of a foreclosure proceeding initiated six years after the loan’s consummation. We concluded only that there was “no federal right to rescind, defensively or otherwise, after the 3-year period of §1635(f) has run,” id., at 419, not that there was no rescission until a suit is filed. Respondents do not dispute that §1635(a) requires only written notice of rescission. Indeed, they concede that written notice suffices to rescind a loan within the first three days after the transaction is consummated. They further concede that written notice suffices after that period if the parties agree that the lender failed to make the required disclosures. Respondents argue, however, that if the parties dispute the adequacy of the disclosures—and thus the continued availability of the right to rescind—then written notice does not suffice. Section 1635(a) nowhere suggests a distinction between disputed and undisputed rescissions, much less that a lawsuit would be required for the latter. In an effort to sidestep this problem, respondents point to a neighboring provision, §1635(g), which they believe provides support for their interpretation of the Act. Section 1635(g) states merely that, “[i]n any action in which it is determined that a creditor has violated this section, in addition to rescission the court may award relief under section 1640 of this title for violations of this subchapter not relating to the right to rescind.” Respondents argue that the phrase “award relief” “in addition to rescission” confirms that rescission is a consequence of judicial action. But the fact that it can be a consequence of judicial action when §1635(g) is triggered in no way suggests that it can only follow from such action. The Act contemplates various situations in which the question of a lender’s compliance with the Act’s disclosure requirements may arise in a lawsuit—for example, a lender’s foreclosure action in which the borrower raises inadequate disclosure as an affirmative defense. Section 1635(g) makes clear that a court may not only award rescission and thereby relieve the borrower of his financial obligation to the lender, but may also grant any of the remedies available under §1640 (including statutory damages). It has no bearing upon whether and how borrower-rescission under §1635(a) may occur. Finally, respondents invoke the common law. It is true that rescission traditionally required either that the rescinding party return what he received before a rescission could be effected (rescission at law), or else that a court affirmatively decree rescission (rescission in equity). 2 D. Dobbs, Law of Remedies §9.3(3), pp. 585–586 (2d ed. 1993). It is also true that the Act disclaims the common-law condition precedent to rescission at law that the borrower tender the proceeds received under the transaction.15 U. S. C. §1635(b). But the negation of rescission-at-law’s tender requirement hardly implies that the Act codifies rescission in equity. Nothing in our jurisprudence, and no tool of statutory interpretation, requires that a congressional Act must be construed as implementing its closest common-law analogue. Cf. Astoria Fed. Sav. & Loan Assn. v. Solimino,501 U. S. 104–109 (1991). The clear import of §1635(a) is that a borrower need only provide written notice to a lender in order to exercise his right to rescind. To the extent §1635(b) alters the traditional process for unwinding such a unilaterally rescinded transaction, this is simply a case in which statutory law modifies common-law practice.* * * The Jesinoskis mailed respondents written notice of their intention to rescind within three years of their loan’s consummation. Because this is all that a borrower must do in order to exercise his right to rescind under the Act, the court below erred in dismissing the complaint. Accordingly, we reverse the judgment of the Eighth Circuit and remand the case for further proceedings consistent with this opinion.It is so ordered.Notes1 Following the events in this case, Congress transferred the author-ity to promulgate rules implementing the Act to the Consumer Finance Protection Bureau. See Dodd-Frank Wall Street Reform and Consumer Protection Act, §§1061(b)(1), 1100A(2), 1100H,124Stat.2036,2107,2113.

576.US.2014_13-7120

After petitioner Johnson pleaded guilty to being a felon in possession of a firearm, see 18 U. S. C. §922(g), the Government sought an enhanced sentence under the Armed Career Criminal Act, which imposes an increased prison term upon a defendant with three prior convictions for a “violent felony,” §924(e)(1), a term defined by §924(e)(2)(B)’s residual clause to include any felony that “involves conduct that presents a serious potential risk of physical injury to another.” The Government argued that Johnson’s prior conviction for unlawful possession of a short-barreled shotgun met this definition, making the third conviction of a violent felony. This Court had previously pronounced upon the meaning of the residual clause in James v. United States, 550 U. S. 192 ; Begay v. United States, 553 U. S. 137 ; Chambers v. United States, 555 U. S. 122 ; and Sykes v. United States, 564 U. S. 1 , and had rejected suggestions by dissenting Justices in both James and Sykes that the clause is void for vagueness. Here, the District Court held that the residual clause does cover unlawful possession of a short-barreled shotgun, and imposed a 15-year sentence under ACCA. The Eighth Circuit affirmed. Held: Imposing an increased sentence under ACCA’s residual clause violates due process. Pp. 3–15. (a) The Government violates the Due Process Clause when it takes away someone’s life, liberty, or property under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement. Kolender v. Lawson, 461 U. S. 352 –358. Courts must use the “categorical approach” when deciding whether an offense is a violent felony, looking “only to the fact that the defendant has been convicted of crimes falling within certain categories, and not to the facts underlying the prior convictions.” Taylor v. United States, 495 U. S. 575 . Deciding whether the residual clause covers a crime thus requires a court to picture the kind of conduct that the crime involves in “the ordinary case,” and to judge whether that abstraction presents a serious potential risk of physical injury. James, supra, at 208. Pp. 3–5. (b) Two features of the residual clause conspire to make it unconstitutionally vague. By tying the judicial assessment of risk to a judicially imagined “ordinary case” of a crime rather than to real-world facts or statutory elements, the clause leaves grave uncertainty about how to estimate the risk posed by a crime. See James, supra, at 211. At the same time, the residual clause leaves uncertainty about how much risk it takes for a crime to qualify as a violent felony. Taken together, these uncertainties produce more unpredictability and arbitrariness than the Due Process Clause tolerates. This Court’s repeated failure to craft a principled standard out of the residual clause and the lower courts’ persistent inability to apply the clause in a consistent way confirm its hopeless indeterminacy. Pp. 5–10. (c) This Court’s cases squarely contradict the theory that the residual clause is constitutional merely because some underlying crimes may clearly pose a serious potential risk of physical injury to another. See, e.g., United States v. L. Cohen Grocery Co., 255 U. S. 81 . Holding the residual clause void for vagueness does not put other criminal laws that use terms such as “substantial risk” in doubt, because those laws generally require gauging the riskiness of an individual’s conduct on a particular occasion, not the riskiness of an idealized ordinary case of the crime. Pp. 10–13. (d) The doctrine of stare decisis does not require continued adherence to James and Sykes. Experience leaves no doubt about the unavoidable uncertainty and arbitrariness of adjudication under the residual clause. James and Sykes opined about vagueness without full briefing or argument. And continued adherence to those decisions would undermine, rather than promote, the goals of evenhandedness, predictability, and consistency served by stare decisis. Pp. 13–15. 526 Fed. Appx. 708, reversed and remanded. Scalia, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Kennedy, J., and Thomas, J., filed opinions concurring in the judgment. Alito, J., filed a dissenting opinion.

Under the Armed Career Criminal Act of 1984, a defendant convicted of being a felon in possession of a firearm faces more severe punishment if he has three or more previous convictions for a “violent felony,” a term defined to include any felony that “involves conduct that presents a serious potential risk of physical injury to another.” 18 U. S. C. §924(e)(2)(B). We must decide whether this part of the definition of a violent felony survives the Constitution’s prohibition of vague criminal laws. I Federal law forbids certain people—such as convicted felons, persons committed to mental institutions, and drug users—to ship, possess, and receive firearms. §922(g). In general, the law punishes violation of this ban by up to 10 years’ imprisonment. §924(a)(2). But if the violator has three or more earlier convictions for a “serious drug offense” or a “violent felony,” the Armed Career Criminal Act increases his prison term to a minimum of 15 years and a maximum of life. §924(e)(1); Johnson v. United States, 559 U. S. 133, 136 (2010) . The Act defines “violent felony” as follows: “any crime punishable by imprisonment for a term exceeding one year . . . that— “(i) has as an element the use, attempted use, or threatened use of physical force against the person of another; or “(ii) is burglary, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another.” §924(e)(2)(B) (emphasis added). The closing words of this definition, italicized above, have come to be known as the Act’s residual clause. Since 2007, this Court has decided four cases attempting to discern its meaning. We have held that the residual clause (1) covers Florida’s offense of attempted burglary, James v. United States, 550 U. S. 192 (2007) ; (2) does not cover New Mexico’s offense of driving under the influence, Begay v. United States, 553 U. S. 137 (2008) ; (3) does not cover Illinois’ offense of failure to report to a penal institution, Chambers v. United States, 555 U. S. 122 (2009) ; and (4) does cover Indiana’s offense of vehicular flight from a law-enforcement officer, Sykes v. United States, 564 U. S. 1 (2011) . In both James and Sykes, the Court rejected suggestions by dissenting Justices that the residual clause violates the Constitution’s prohibition of vague criminal laws. Compare James, 550 U. S., at 210, n. 6, with id., at 230 (Scalia, J., dissenting); compare Sykes, 564 U. S., at ___ (slip op., at 13–14), with id., at ___ (Scalia, J., dissenting) (slip op., at 6–8). This case involves the application of the residual clause to another crime, Minnesota’s offense of unlawful possession of a short-barreled shotgun. Petitioner Samuel Johnson is a felon with a long criminal record. In 2010, the Federal Bureau of Investigation began to monitor him because of his involvement in a white-supremacist organization that the Bureau suspected was planning to commit acts of terrorism. During the investigation, Johnson disclosed to undercover agents that he had manufactured explosives and that he planned to attack “the Mexican consulate” in Minnesota, “progressive bookstores,” and “ ‘liberals.’ ” Revised Presentence Investigation in No. 0:12CR00104–001 (D. Minn.), p. 15, ¶16. Johnson showed the agents his AK–47 rifle, several semiautomatic firearms, and over 1,000 rounds of ammunition. After his eventual arrest, Johnson pleaded guilty to being a felon in possession of a firearm in violation of §922(g). The Government requested an enhanced sentence under the Armed Career Criminal Act. It argued that three of Johnson’s previous offenses—including unlawful possession of a short-barreled shotgun, see Minn. Stat. §609.67 (2006)—qualified as violent felonies. The District Court agreed and sentenced Johnson to a 15-year prison term under the Act. The Court of Appeals affirmed. 526 Fed. Appx. 708 (CA8 2013) (per curiam). We granted certiorari to decide whether Minnesota’s offense of unlawful possession of a short-barreled shotgun ranks as a violent felony under the residual clause. 572 U. S. ___ (2014). We later asked the parties to present reargument addressing the compatibility of the residual clause with the Constitution’s prohibition of vague criminal laws. 574 U. S. ___ (2015). II The Fifth Amendment provides that “[n]o person shall . . . be deprived of life, liberty, or property, without due process of law.” Our cases establish that the Government violates this guarantee by taking away someone’s life, liberty, or property under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement. Kolender v. Lawson, 461 U. S. 352 –358 (1983). The prohibition of vagueness in criminal statutes “is a well-recognized requirement, consonant alike with ordinary notions of fair play and the settled rules of law,” and a statute that flouts it “violates the first essential of due process.” Connally v. General Constr. Co., 269 U. S. 385, 391 (1926) . These principles apply not only to statutes defining elements of crimes, but also to statutes fixing sentences. United States v. Batchelder, 442 U. S. 114, 123 (1979) . In Taylor v. United States, 495 U. S. 575, 600 (1990) , this Court held that the Armed Career Criminal Act requires courts to use a framework known as the categorical approach when deciding whether an offense “is burglary, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another.” Under the categorical approach, a court assesses whether a crime qualifies as a violent felony “in terms of how the law defines the offense and not in terms of how an individual offender might have committed it on a particular occasion.” Begay, supra, at 141. Deciding whether the residual clause covers a crime thus requires a court to picture the kind of conduct that the crime involves in “the ordinary case,” and to judge whether that abstraction presents a serious potential risk of physical injury. James, supra, at 208. The court’s task goes beyond deciding whether creation of risk is an element of the crime. That is so because, unlike the part of the definition of a violent felony that asks whether the crime “has as an element the use . . . of physical force,” the residual clause asks whether the crime “involves conduct” that presents too much risk of physical injury. What is more, the inclusion of burglary and extortion among the enumerated offenses preceding the residual clause confirms that the court’s task also goes beyond evaluating the chances that the physical acts that make up the crime will injure someone. The act of making an extortionate demand or breaking and entering into someone’s homedoes not, in and of itself, normally cause physical injury. Rather, risk of injury arises because the extortionist might engage in violence after making his demand or because the burglar might confront a resident in the home after breaking and entering. We are convinced that the indeterminacy of the wide-ranging inquiry required by the residual clause both denies fair notice to defendants and invites arbitrary enforcement by judges. Increasing a defendant’s sentence under the clause denies due process of law. A Two features of the residual clause conspire to make it unconstitutionally vague. In the first place, the residual clause leaves grave uncertainty about how to estimate the risk posed by a crime. It ties the judicial assessment of risk to a judicially imagined “ordinary case” of a crime, not to real-world facts or statutory elements. How does one go about deciding what kind of conduct the “ordinary case” of a crime involves? “A statistical analysis of the state reporter? A survey? Expert evidence? Google? Gut instinct?” United States v. Mayer, 560 F. 3d 948, 952 (CA9 2009) (Kozinski, C. J., dissenting from denial of rehearing en banc). To take an example, does the ordinary instance of witness tampering involve offering a witness a bribe? Or threatening a witness with violence? Critically, picturing the criminal’s behavior is not enough; as we have already discussed, assessing “potential risk” seemingly requires the judge to imagine how the idealized ordinary case of the crime subsequently plays out. James illustrates how speculative (and how detached from statutory elements) this enterprise can become. Explaining why attempted burglary poses a serious potential risk of physical injury, the Court said: “An armed would-be burglar may be spotted by a police officer, a private security guard, or a participant in a neighborhood watch program. Or a homeowner . . . may give chase, and a violent encounter may ensue.” 550 U. S., at 211. The dissent, by contrast, asserted that any confrontation that occurs during an attempted burglary “is likely to consist of nothing more than the occupant’s yelling ‘Who’s there?’ from his window, and the burglar’s running away.” Id., at 226 (opinion of Scalia, J.). The residual clause offers no reliable way to choose between these competing accounts of what “ordinary” attempted burglary involves. At the same time, the residual clause leaves uncertainty about how much risk it takes for a crime to qualify as a violent felony. It is one thing to apply an imprecise “serious potential risk” standard to real-world facts; it is quite another to apply it to a judge-imagined abstraction. By asking whether the crime “otherwise involves conduct that presents a serious potential risk,” moreover, the residual clause forces courts to interpret “serious potential risk” in light of the four enumerated crimes—burglary, arson, extortion, and crimes involving the use of explosives. These offenses are “far from clear in respect to the degree of risk each poses.” Begay, 553 U. S., at 143. Does the ordinary burglar invade an occupied home by night or an unoccupied home by day? Does the typical extortionist threaten his victim in person with the use of force, or does he threaten his victim by mail with the revelation of embarrassing personal information? By combining indeterminacy about how to measure the risk posed by a crime with indeterminacy about how much risk it takes for the crime to qualify as a violent felony, the residual clause produces more unpredictability and arbitrariness than the Due Process Clause tolerates. This Court has acknowledged that the failure of “persistent efforts . . . to establish a standard” can provide evidence of vagueness. United States v. L. Cohen Grocery Co., 255 U. S. 81, 91 (1921) . Here, this Court’s repeated attempts and repeated failures to craft a principled and objective standard out of the residual clause confirm its hopeless indeterminacy. Three of the Court’s previous four decisions about the clause concentrated on the level of risk posed by the crime in question, though in each case we found it necessary to resort to a different ad hoc test to guide our inquiry. In James, we asked whether “the risk posed by attempted burglary is comparable to that posed by its closest analog among the enumerated offenses,” namely completed burglary; we concluded that it was. 550 U. S., at 203. That rule takes care of attempted burglary, but offers no help at all with respect to the vast majority of offenses, which have no apparent analog among the enumerated crimes. “Is, for example, driving under the influence of alcohol more analogous to burglary, arson, extortion, or a crime involving use of explosives?” Id., at 215 (Scalia, J., dissenting). Chambers, our next case to focus on risk, relied principally on a statistical report prepared by the Sentencing Commission to conclude that an offender who fails to report to prison is not “significantly more likely than others to attack, or physically to resist, an apprehender, thereby producing a ‘serious potential risk of physical injury.’ ” 555 U. S., at 128–129. So much for failure to report to prison, but what about the tens of thousands of federal and state crimes for which no comparable reports exist? And even those studies that are available might suffer from methodological flaws, be skewed toward rarer forms of the crime, or paint widely divergent pictures of the riskiness of the conduct that the crime involves. See Sykes, 564 U. S., at ___–___ (Scalia, J., dissenting) (slip op., at 4–6); id., at ___, n. 4 (Kagan, J., dissenting) (slip op., at 6, n. 4). Our most recent case, Sykes, also relied on statistics, though only to “confirm the commonsense conclusion that Indiana’s vehicular flight crime is a violent felony.” Id., at ___ (majority opinion) (slip op., at 8). But common sense is a much less useful criterion than it sounds—as Sykes itself illustrates. The Indiana statute involved in that case covered everything from provoking a high-speed car chase to merely failing to stop immediately after seeing a police officer’s signal. See id., at ___ (Kagan, J., dissenting) (slip op., at 3–4). How does common sense help a federal court discern where the “ordinary case” of vehicular flight in Indiana lies along this spectrum? Common sense has not even produced a consistent conception of the degree of risk posed by each of the four enumerated crimes; there is no reason to expect it to fare any better with respect to thousands of unenumerated crimes. All in all, James, Chambers, and Sykes failed to establish any generally appli-cable test that prevents the risk comparison required by the residual clause from devolving into guesswork and intuition. The remaining case, Begay, which preceded Chambers and Sykes, took an entirely different approach. The Court held that in order to qualify as a violent felony under the residual clause, a crime must resemble the enumerated offenses “in kind as well as in degree of risk posed.” 553 U. S., at 143. The Court deemed drunk driving insufficiently similar to the listed crimes, because it typically does not involve “purposeful, violent, and aggressive conduct.” Id., at 144–145 (internal quotation marks omitted). Alas, Begay did not succeed in bringing clarity to the meaning of the residual clause. It did not (and could not) eliminate the need to imagine the kind of conduct typically involved in a crime. In addition, the enumerated crimes are not much more similar to one another in kind than in degree of risk posed, and the concept of “aggressive conduct” is far from clear. Sykes criticized the “purposeful, violent, and aggressive” test as an “addition to the statu-tory text,” explained that “levels of risk” would normally be dispositive, and confined Begay to “strict liability, negligence, and recklessness crimes.” 564 U. S., at ___–___ (slip op., at 10–11). The present case, our fifth about the meaning of the residual clause, opens a new front of uncertainty. When deciding whether unlawful possession of a short-barreled shotgun is a violent felony, do we confine our attention to the risk that the shotgun will go off by accident while in someone’s possession? Or do we also consider the possibility that the person possessing the shotgun will later use it to commit a crime? The inclusion of burglary and extortion among the enumerated offenses suggests that a crime may qualify under the residual clause even if the physical injury is remote from the criminal act. But how remote is too remote? Once again, the residual clause yields no answers. This Court is not the only one that has had trouble making sense of the residual clause. The clause has “created numerous splits among the lower federal courts,” where it has proved “nearly impossible to apply consistently.” Chambers, 555 U. S., at 133 (Alito, J., concurring in judgment). The most telling feature of the lower courts’ decisions is not division about whether the residual clause covers this or that crime (even clear laws produce close cases); it is, rather, pervasive disagreement about the nature of the inquiry one is supposed to conduct and the kinds of factors one is supposed to consider. Some judges have concluded that deciding whether conspiracy is a violent felony requires evaluating only the dangers posed by the “simple act of agreeing [to commit a crime],” United States v. Whitson, 597 F. 3d 1218, 1222 (CA11 2010) (per curiam); others have also considered the probability that the agreement will be carried out, United States v. White, 571 F. 3d 365, 370–371 (CA4 2009). Some judges have assumed that the battery of a police officer (defined to include the slightest touching) could “explode into violence and result in physical injury,” United States v. Williams, 559 F. 3d 1143, 1149 (CA10 2009); others have felt that it “do[es] a great disservice to law enforcement officers” to assume that they would “explod[e] into violence” rather than “rely on their training and experience to determine the best method of responding,” United States v. Carthorne, 726 F. 3d 503, 514 (CA4 2013). Some judges considering whether statutory rape qualifies as a violent felony have concentrated on cases involving a perpetrator much older than the victim, United States v. Daye, 571 F. 3d 225, 230–231 (CA2 2009); others have tried to account for the possibility that “the perpetrator and the victim [might be] close in age,” United States v. McDonald, 592 F. 3d 808, 815 (CA7 2010). Disagreements like these go well beyond disputes over matters of degree. It has been said that the life of the law is experience. Nine years’ experience trying to derive meaning from the residual clause convinces us that we have embarked upon a failed enterprise. Each of the uncertainties in the residual clause may be tolerable in isolation, but “their sum makes a task for us which at best could be only guesswork.” United States v. Evans, 333 U. S. 483, 495 (1948) . Invoking so shapeless a provision to condemn someone to prison for 15 years to life does not comport with the Constitution’s guarantee of due process. B The Government and the dissent claim that there will be straightforward cases under the residual clause, because some crimes clearly pose a serious potential risk of physical injury to another. See post, at 14–15 (opinion of Alito, J.). True enough, though we think many of the cases the Government and the dissent deem easy turn out not to be so easy after all. Consider just one of the Government’s examples, Connecticut’s offense of “rioting at a correctional institution.” See United States v. Johnson, 616 F. 3d 85 (CA2 2010). That certainly sounds like a violent felony—until one realizes that Connecticut defines this offense to include taking part in “any disorder, disturbance, strike, riot or other organized disobedience to the rules and regulations” of the prison. Conn. Gen. Stat. §53a–179b(a) (2012). Who is to say which the ordinary “disorder” most closely resembles—a full-fledged prison riot, a food-fight in the prison cafeteria, or a “passive and nonviolent [act] such as disregarding an order to move,” Johnson, 616 F. 3d, at 95 (Parker, J., dissenting)? In all events, although statements in some of our opinions could be read to suggest otherwise, our holdings squarely contradict the theory that a vague provision is constitutional merely because there is some conduct that clearly falls within the provision’s grasp. For instance, we have deemed a law prohibiting grocers from charging an “unjust or unreasonable rate” void for vagueness—even though charging someone a thousand dollars for a pound of sugar would surely be unjust and unreasonable. L. Cohen Grocery Co., 255 U. S., at 89. We have similarly deemed void for vagueness a law prohibiting people on sidewalks from “conduct[ing] themselves in a manner annoying to persons passing by”—even though spitting in someone’s face would surely be annoying. Coates v. Cincinnati, 402 U. S. 611 (1971) . These decisions refute any suggestion that the existence of some obviously risky crimes establishes the residual clause’s constitutionality. Resisting the force of these decisions, the dissent insists that “a statute is void for vagueness only if it is vague in all its applications.” Post, at 1. It claims that the prohibition of unjust or unreasonable rates in L. Cohen Grocery was “vague in all applications,” even though one can easily envision rates so high that they are unreasonable by any measure. Post, at 16. It seems to us that the dissent’s supposed requirement of vagueness in all applications is not a requirement at all, but a tautology: If we hold a statute to be vague, it is vague in all its applications (and never mind the reality). If the existence of some clearly unreasonable rates would not save the law in L. Cohen Grocery, why should the existence of some clearly risky crimes save the residual clause? The Government and the dissent next point out that dozens of federal and state criminal laws use terms like “substantial risk,” “grave risk,” and “unreasonable risk,” suggesting that to hold the residual clause unconstitutional is to place these provisions in constitutional doubt. See post, at 7–8. Not at all. Almost none of the cited laws links a phrase such as “substantial risk” to a confusing list of examples. “The phrase ‘shades of red,’ standing alone, does not generate confusion or unpredictability; but the phrase ‘fire-engine red, light pink, maroon, navy blue, or colors that otherwise involve shades of red’ assuredly does so.” James, 550 U. S., at 230, n. 7 (Scalia, J., dissenting). More importantly, almost all of the cited laws require gauging the riskiness of conduct in which an individual defendant engages on a particular occasion. As a general matter, we do not doubt the constitutionality of laws that call for the application of a qualitative standard such as “substantial risk” to real-world conduct; “the law is full of instances where a man’s fate depends on his estimating rightly . . . some matter of degree,” Nash v. United States, 229 U. S. 373, 377 (1913) . The residual clause, however, requires application of the “serious potential risk” standard to an idealized ordinary case of the crime. Because “the elements necessary to determine the imaginary ideal are uncertain both in nature and degree of effect,” this abstract inquiry offers significantly less predictability than one “[t]hat deals with the actual, not with an imaginary condition other than the facts.” International Harvester Co. of America v. Kentucky, 234 U. S. 216, 223 (1914) . Finally, the dissent urges us to save the residual clause from vagueness by interpreting it to refer to the risk posed by the particular conduct in which the defendant engaged, not the risk posed by the ordinary case of the defendant’s crime. See post, at 9–13. In other words, the dissent suggests that we jettison for the residual clause (though not for the enumerated crimes) the categorical approach adopted in Taylor, see 495 U. S., at 599–602, and reaffirmed in each of our four residual-clause cases, see James, 550 U. S., at 202; Begay, 553 U. S., at 141; Chambers, 555 U. S., at 125; Sykes, 564 U. S., ___ (slip op., at 5). We decline the dissent’s invitation. In the first place, the Government has not asked us to abandon the categorical approach in residual-clause cases. In addition, Taylor had good reasons to adopt the categorical approach, reasons that apply no less to the residual clause than to the enumerated crimes. Taylor explained that the relevant part of the Armed Career Criminal Act “refers to ‘a person who . . . has three previous convictions’ for—not a person who has committed—three previous violent felonies or drug offenses.” 495 U. S., at 600. This emphasis on convictions indicates that “Congress intended the sentencing court to look only to the fact that the defendant had been convicted of crimes falling within certain categories, and not to the facts underlying the prior convictions.” Ibid. Taylor also pointed out the utter impracticability of requiring a sentencing court to reconstruct, long after the original conviction, the conduct underlying that conviction. For example, if the original conviction rested on a guilty plea, no record of the underlying facts may be available. “[T]he only plausible interpretation” of the law, therefore, requires use of the categorical approach. Id., at 602. C That brings us to stare decisis. This is the first case in which the Court has received briefing and heard argument from the parties about whether the residual clause is void for vagueness. In James, however, the Court stated in a footnote that it was “not persuaded by [the principal dissent’s] suggestion . . . that the residual provision is unconstitutionally vague.” 550 U. S., at 210, n. 6. In Sykes, the Court again rejected a dissenting opinion’s claim of vagueness. 564 U. S., at ___–___ (slip op., at 13–14). The doctrine of stare decisis allows us to revisit an ear-lier decision where experience with its application reveals that it is unworkable. Payne v. Tennessee, 501 U. S. 808, 827 (1991) . Experience is all the more instructive when the decision in question rejected a claim of unconstitu-tional vagueness. Unlike other judicial mistakes that need correction, the error of having rejected a vagueness challenge manifests itself precisely in subsequent judicial decisions: the inability of later opinions to impart the predictability that the earlier opinion forecast. Here, the experience of the federal courts leaves no doubt about the unavoidable uncertainty and arbitrariness of adjudication under the residual clause. Even after Sykes tried to clarify the residual clause’s meaning, the provision remains a “judicial morass that defies systemic solution,” “a black hole of confusion and uncertainty” that frustrates any effort to impart “some sense of order and direction.” United States v. Vann, 660 F. 3d 771, 787 (CA4 2011) (Agee, J., concurring). This Court’s cases make plain that even decisions rendered after full adversarial presentation may have to yield to the lessons of subsequent experience. See, e.g., United States v. Dixon, 509 U. S. 688, 711 (1993) ; Payne, 501 U. S., at 828–830 (1991). But James and Sykes opined about vagueness without full briefing or argument on that issue—a circumstance that leaves us “less constrained to follow precedent,” Hohn v. United States, 524 U. S. 236, 251 (1998) . The brief discussions of vagueness in James and Sykes homed in on the imprecision of the phrase “serious potential risk”; neither opinion evaluated the uncertainty introduced by the need to evaluate the riskiness of an abstract ordinary case of a crime. 550 U. S., at 210, n. 6; 564 U. S., at ___ (slip op., at 13–14). And departing from those decisions does not raise any concerns about upsetting private reliance interests. Although it is a vital rule of judicial self-government, stare decisis does not matter for its own sake. It matters because it “promotes the evenhanded, predictable, and consistent development of legal principles.” Payne, supra, at 827. Decisions under the residual clause have proved to be anything but evenhanded, predictable, or consistent. Standing by James and Sykes would undermine, rather than promote, the goals that stare decisis is meant to serve. * * * We hold that imposing an increased sentence under the residual clause of the Armed Career Criminal Act violates the Constitution’s guarantee of due process. Our contrary holdings in James and Sykes are overruled. Today’s decision does not call into question application of the Act to the four enumerated offenses, or the remainder of the Act’s definition of a violent felony. We reverse the judgment of the Court of Appeals for the Eighth Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered.

575.US.2014_12-1497

Private parties may file civil qui tam actions to enforce the False Claims Act (FCA), which prohibits making “a false or fraudulent claim for payment or approval,” 31 U. S. C. §3729(a)(1), “to . . . the United States,” 3729(b)(2)(A)(i). A qui tam action must generally be brought within six years of a violation, §3731(b), but the Wartime Suspension of Limitations Act (WSLA) suspends “the running of any statute of limitations applicable to any offense” involving fraud against the Federal Government. 18 U. S. C. §3287. Separately, the FCA’s “first-to-file bar” precludes a qui tam suit “based on the facts underlying [a] pending action,” §3730(b)(5). In 2005, respondent worked for one of the petitioners, providing logistical services to the United States military in Iraq. He subsequently filed a qui tam complaint (Carter I), alleging that petitioners, who are defense contractors and related entities, had fraudulently billed the Government for water purification services that were not performed or not performed properly. In 2010, shortly before trial, the Government informed the parties that an earlier-filed qui tam suit (Thorpe) had similar claims, leading the District Court to dismiss Carter I without prejudice under the first-to-file bar. While respondent’s appeal was pending, Thorpe was dismissed for failure to prosecute. Respondent quickly filed a new complaint (Carter II), but the court dismissed it under the first-to-file rule because Carter I’s appeal was pending. Respondent then dismissed that appeal, and in June 2011, more than six years after the alleged fraud, filed the instant complaint (Carter III). The District Court dismissed this complaint with prejudice under the first-to-file rule because of a pending Maryland suit. Further, because the WSLA applies only to criminal charges, the court reasoned, all but one of respondent’s civil actions were untimely. Reversing, the Fourth Circuit concluded that the WSLA applied to civil claims and that the first-to-file bar ceases to apply once a related action is dismissed. Since any pending suits had by then been dismissed, the court held, respondent had the right to refile his case. It thus remanded Carter III with instructions to dismiss without prejudice. Held: 1. As shown by the WSLA’s text, structure, and history, the Act applies only to criminal offenses, not to civil claims like those in this case. Pp. 5–11. (a) The 1921 and 1942 versions of the WSLA were enacted to address war-related fraud during, respectively, the First and Second World Wars. Both extended the statute of limitations for fraud offenses “now indictable under any existing statutes.” Since only crimes are “indictable,” these provisions quite clearly were limited to criminal charges. In 1944, Congress made the WSLA prospectively applicable to future wartime frauds rather than merely applicable to past frauds as earlier versions had been. In doing so, it deleted the phrase “now indictable under any statute,” so that the WSLA now applied to “any offense against the laws of the United States.” Congress made additional changes in 1948 and codified the WSLA in Title 18 U. S. C. Pp. 5–6. (b) Section 3287’s text supports limiting the WSLA to criminal charges. The term “offense” is most commonly used to refer to crimes, especially given the WSLA’s location in Title 18, titled “Crimes and Criminal Procedure,” where no provision appears to employ “offense” to denote a civil violation rather than a civil penalty attached to a criminal offense. And when Title 18 was enacted in 1948, its very first provision classified all offenses as crimes. In similar circumstances, this Court has regarded a provision’s placement as relevant in determining whether its content is civil or criminal. Kansas v. Hendricks, 521 U. S. 346 . The WSLA’s history provides further support for this reading. The term “offenses” in the 1921 and 1942 statutes, the parties agree, applied only to crimes. And it is improbable that the 1944 Act’s removal of the phrase “now indictable under any statute” had the effect of sweeping in civil claims, a fundamental change in scope not typically accomplished with so subtle a move. The more plausible explanation is that Congress removed that phrase in order to change the WSLA from a retroactive measure designed to deal exclusively with past fraud into a permanent measure applicable to future fraud as well. If there were any ambiguity in the WSLA’s use of the term “offense,” that ambiguity should be resolved in favor of a narrower definition. See Bridges v. United States, 346 U. S. 209, 216. Pp. 7–11. 2. The FCA’s first-to-file bar keeps new claims out of court only while related claims are still alive, not in perpetuity. Thus, dismissal with prejudice was not called for in this case. This reading of §3730(b)(5) is in accordance with the ordinary dictionary meaning of the term “pending.” Contrary to petitioners’ reading, the term “pending” cannot be seen as a sort of “short-hand” for first-filed, which is neither a lengthy nor a complex term. Petitioners’ reading would also bar even a suit dismissed for a reason having nothing to do with the merits, such as Thorpe, which was dismissed for failure to prosecute. Pp. 11–13. 710 F. 3d 171, reversed in part, affirmed in part, and remanded. Alito, J., delivered the opinion for a unanimous Court.

Wars have often provided “exceptional opportunities” for fraud on the United States Government. See United States v. Smith, 342 U. S. 225, 228 (1952) . “The False Claims Act was adopted in 1863 and signed into law by President Abraham Lincoln in order to combat rampant fraud in Civil War defense contracts.” S. Rep. No. 99–345, p. 8 (1986). Predecessors of the Wartime Suspension of Limitations Act were enacted to address similar problems that arose during the First and Second World Wars. See Smith, supra, at 228–229. In this case, we must decide two questions regarding those laws: first, whether the Wartime Suspension of Limitations Act applies only to criminal charges or also to civil claims; second, whether the False Claims Act’s first-to-file bar keeps new claims out of court only while related claims are still alive or whether it may bar those claims in perpetuity. I A The False Claims Act (FCA) imposes liability on any person who “knowingly presents . . . a false or fraudulent claim for payment or approval,” 31 U. S. C. §3729(a)(1)(A), “to an officer or employee of the United States,” 3729(b)(2)(A)(i). The FCA may be enforced not just through litigation brought by the Government itself, but also through civil qui tam actions that are filed by private parties, called relators, “in the name of the Government.” §3730(b). In a qui tam suit under the FCA, the relator files a complaint under seal and serves the United States with a copy of the complaint and a disclosure of all material evidence. §3730(b)(2). After reviewing these materials, the United States may “proceed with the action, in which case the action shall be conducted by the Government,” or it may “notify the court that it declines to take over the action, in which case the person bringing the action shall have the right to conduct the action.” §3730(b)(4). Regardless of the option that the United States selects, it retains the right at any time to dismiss the action entirely, §3730(c)(2)(A), or to settle the case, §3730(c)(2)(B). The FCA imposes two restrictions on qui tam suits that are relevant here. One, the “first-to-file” bar, precludes a qui tam suit “based on the facts underlying [a] pending action.” §3730(b)(5) (emphasis added). The other, the FCA’s statute of limitations provision, states that a qui tam action must be brought within six years of a violation or within three years of the date by which the United States should have known about a violation. In no circumstances, however, may a suit be brought more than 10 years after the date of a violation. §3731(b). B The Wartime Suspension of Limitations Act (WSLA) suspends the statute of limitations for “any offense” involving fraud against the Federal Government. 18 U. S. C. §3287. Before 2008, this provision was activated only “[w]hen the United States [was] at war.” Ibid. (2006 ed.). In 2008, however, this provision was made to apply as well whenever Congress has enacted “a specific authorization for the use of the Armed Forces, as described in section 5(b) of the War Powers Resolution (50 U.S.C. 1544(b)).” Ibid. (2012 ed.). II Petitioners are defense contractors and related entities that provided logistical services to the United States military during the armed conflict in Iraq. From January to April 2005, respondent worked in Iraq for one of the petitioners as a water purification operator. He subsequently filed a qui tam complaint against petitioners (Carter I), alleging that they had fraudulently billed the Government for water purification services that were not performed or not performed properly. The Government declined to intervene. In 2010, shortly before trial, the Government informed the parties about an earlier filed qui tam lawsuit, United States ex rel. Thorpe v. Halliburton Co., No. 05–cv–08924 (CD Cal., filed Dec. 23, 2005), that arguably contained similar claims. This initiated a remarkable sequence of dismissals and filings. The District Court held that respondent’s suit was related to Thorpe and thus dismissed his case without prejudice under the first-to-file bar. Respondent appealed, and while his appeal was pending, Thorpe was dismissed for failure to prosecute. Respondent quickly filed a new complaint (Carter II), but the District Court dismissed this second complaint under the first-to-file rule because respondent’s own earlier case was still pending on appeal. Respondent then voluntarily dismissed this appeal, and in June 2011, more than six years after the alleged fraud, he filed yet another complaint (Carter III ), and it is this complaint that is now at issue. Petitioners sought dismissal of this third complaint under the first-to-file rule, pointing to two allegedly related cases, one in Maryland and one in Texas, that had been filed in the interim between the filing of Carter I and Carter III. This time, the court dismissed respondent’s complaint with prejudice. The court held that the latest complaint was barred under the first-to-file rule because the Maryland suit was already pending when that complaint was filed. The court also ruled that the WSLA applies only to criminal charges and thus did not suspend the time for filing respondent’s civil claims. As a result, the court concluded, all but one of those claims were untimely because they were filed more than six years after the alleged wrongdoing. The Fourth Circuit reversed, rejecting the District Court’s analysis of both the WSLA and first-to-file issues. United States ex rel. Carter v. Halliburton Co., 710 F. 3d 171 (2013). Concluding that the WSLA applies to civil claims based on fraud committed during the conflict in Iraq,[1] the Court of Appeals held that respondent’s claims had been filed on time. The Court of Appeals also held that the first-to-file bar ceases to apply once a related action is dismissed. Since the Maryland and Texas cases had been dismissed by the time of the Fourth Circuit’s decision, the court held that respondent had the right to refile his case. The Court of Appeals thus remanded Carter III with instructions to dismiss without prejudice. After this was done, respondent filed Carter IV, but the District Court dismissed Carter IV on the ground that the petition for a writ of certiorari in Carter III (the case now before us) was still pending. We granted that petition, 573 U. S. ___ (2014), and we now reverse in part and affirm in part. III The text, structure, and history of the WSLA show that the Act applies only to criminal offenses. A The WSLA’s roots extend back to the time after the end of World War I. Concerned about war-related frauds, Congress in 1921 enacted a statute that extended the statute of limitations for such offenses. The new law provided as follows: “[I]n offenses involving the defrauding or attempts to defraud the United States or any agency thereof . . . and now indictable under any existing statutes, the period of limitations shall be six years.” Act ofNov. 17, 1921, ch. 124, 42Stat. 220 (emphasis added). Since only crimes are “indictable,” this provision quite clearly was limited to the filing of criminal charges. In 1942, after the United States entered World War II, Congress enacted a similar suspension statute. This law, like its predecessor, applied to fraud “offenses . . . now indictable under any existing statutes,” but this time the law suspended “any” “existing statute of limitations” until the fixed date of June 30, 1945. Act of Aug. 24, 1942, ch. 555, 56Stat. 747–748. As that date approached, Congress decided to adopt a suspension statute which would remain in force for the duration of the war. Congress amended the 1942 WSLA in three important ways. First, Congress deleted the phrase “now indictable under any statute,” so that the WSLA was made to apply simply to “any offense against the laws of the United States.” 58Stat. 667. Second, although previous versions of the WSLA were of definite duration, Congress now suspended the limitations period for the open-ended timeframe of “three years after the termination of hostilities in the present war as proclaimed by the President or by a concurrent resolution of the two Houses of Congress.” Ibid. Third, Congress expanded the statute’s coverage beyond offenses “involving defrauding or attempts to defraud the United States” to include other offenses pertaining to Government contracts and the handling and disposal of Government property. Ibid., and §28, 58Stat. 781. Congress made more changes in 1948. From then until 2008, the WSLA’s relevant language was as follows: “When the United States is at war the running of any statute of limitations applicable to any offense (1) involving fraud or attempted fraud against the United States or any agency thereof in any manner, whether by conspiracy or not . . . shall be suspended until three years after the termination of hostilities as proclaimed by the President or by a concurrent resolution of Congress.” Act of June 25, 1948, §3287, 62Stat. 828. In addition, Congress codified the WSLA in Title 18 ofthe United States Code, titled “Crimes and Criminal Procedure.” Finally, in 2008, Congress once again amended the WSLA, this time in two relevant ways. First, as noted, Congress changed the Act’s triggering event, providing that tolling is available not only “[w]hen the United States is at war,” but also when Congress has enacted a specific authorization for the use of military force. Second, Congress extended the suspension period from three to five years. §855, 122Stat. 4545.[2] B With this background in mind, we turn to the question whether the WSLA applies to civil claims as well as criminal charges. We hold that the Act applies only to the latter. We begin with the WSLA’s text. The WSLA suspends “the running of any statute of limitations applicable to any offense . . . involving fraud or attempted fraud against the United States or any agency thereof.” 18 U. S. C. §3287 (emphasis added). The term “offense” is most commonly used to refer to crimes. At the time of both the 1948 and 2008 amendments to the Act, the primary definition of “offense” in Black’s Law Dictionary referred to crime. Black’s Law Dictionary 1110 (8th ed. 2004) (Black’s) (“A violation of the law; a crime, often a minor one. See crime”); id., at 1232 (4th ed. 1951) (“A crime or misdemeanor; a breach of the criminal laws”); id., at 1282 (3d ed. 1933) (same). The 1942 edition of Webster’s similarly states that “offense” “has no technical legal meaning; but it is sometimes used specifically for an indictable crime . . . and sometimes for a misdemeanor or wrong punishable only by fine or penalty.” Webster’s New International Dictionary 1690 (2d ed.). See also Webster’s Third New International Dictionary 1566 (1976) (Webster’s Third) (“an infraction of law: crime, misdemeanor”); American Heritage Dictionary 1255 (3d ed. 1992) (“A transgression of law; a crime”). It is true that the term “offense” is sometimes used more broadly. For instance, the 1948 edition of Ballentine’s Law Dictionary cautions: “The words ‘crime’ and ‘offense’ are not necessarily synonymous. All crimes are offenses, but some offenses are not crimes.” Ballentine’s Law Dictionary 900. But while the term “offense” is sometimes used in this way, that is not how the word is used in Title 18. Al-though the term appears hundreds of times in Title 18, neither respondent nor the Solicitor General, appearing as an amicus in support of respondent, has been able to find a single provision of that title in which “offense” is employed to denote a civil violation. The Solicitor General cites eight provisions,[3] but not one actually labels a civil wrong as an “offense.” Instead, they all simply attach civil penalties to criminal offenses—as the Deputy Solicitor General acknowledged at oral argument. See Tr. of Oral Arg. 28–29. Not only is this pattern of usage telling, but when Title 18 was enacted in 1948, the very first provision, what was then 18 U.S.C. §1, classified all offenses as crimes. That provision read in pertinent part as follows: “§1. Offenses classified. “Notwithstanding any Act of Congress to thecontrary: “(1) Any offense punishable by death or imprisonment for a term exceeding one year is a felony. “(2) Any other offense is a misdemeanor.” 62Stat. 684 (repealed Oct. 12, 1984). The Solicitor General correctly points out that regulatory provisions outside Title 18 sometimes use the term “offense” to describe a civil violation, see Brief for United States as Amicus Curiae 10 (United States Brief), but it is significant that Congress chose to place the WSLA in Title 18. Although we have cautioned against “plac[ing] too much significance on the location of a statute in the United States Code,” Jones v. R. R. Donnelley & Sons Co., 541 U. S. 369, 376 (2004) , we have in similar circumstances regarded the placement of a provision as relevant in determining whether its content is civil or criminal in nature, see Kansas v. Hendricks, 521 U. S. 346, 361 (1997) . It is also revealing that Congress has used clearer and more specific language when it has wanted to toll the statutes of limitations for civil suits as well as crimes. Only two months after enacting the WSLA, Congress passed a tolling statute for “violations of the antitrust laws . . . now indictable or subject to civil proceedings.” Act of Oct. 10, 1942, ch. 589, 56Stat. 781 (emphasisadded). Congress obviously could have included a similar “civil proceedings” clause in the WSLA, but it did notdo so. The WSLA’s history provides what is perhaps the strongest support for the conclusion that it applies only to criminal charges. The parties do not dispute that the term “offenses” in the 1921 and 1942 suspension statutes applied only to crimes, Brief for Petitioners 23; Brief for Respondent at 24–25, and after 1942, the WSLA continued to use that same term. The retention of the same term in the later laws suggests that no fundamental alteration was intended. Respondent and the Government latch onto the 1944 Act’s removal of the phrase “now indictable under any statute” and argue that this deletion had the effect of sweeping in civil claims, but this argument is most improbable. Simply deleting the phrase “now indictable under the statute,” while leaving the operative term “offense” unchanged would have been an obscure way of substantially expanding the WSLA’s reach. Fundamental changes in the scope of a statute are not typically accomplished with so subtle a move. Converting the WSLA from a provision that suspended the statute of limitations for criminal prosecutions into one that also suspended the time for commencing a civil action would have been a big step. If Congress had meant to make such a change, we would expect it to have used language that made this important modification clear to litigants and courts. Respondent’s and the Government’s interpretation of the significance of the deletion of the phrase “now indict-able” ignores a more plausible explanation, namely, Congress’ decision to make the WSLA applicable, not just to offenses committed in the past during or in the aftermath of particular wars, but also to future offenses committed during future wars. When the phrase “now indictable” first appeared in the 1921 Act, it meant that the statute of limitations was suspended for only those crimes that had already been committed when the Act took effect. This made sense because the 1921 Act was a temporary measure enacted to deal with problems resulting from the First World War. The 1942 Act simply “readopt[ed] the [same] World War I policy” to deal with claims during World War II. Bridges v. United States, 346 U. S. 209, 219 (1953) . The 1944 amendments, however, changed the WSLA from a retroactive measure designed to deal exclusively with past fraud into a measure applicable to future fraud as well. In order to complete this transformation, it was necessary to remove the phrase “now indictable,” which, as noted, limited the applicability of the suspension to offenses committed in the past. Thus, the removal of the “now indictable” provision was more plausibly driven by Congress’ intent to apply the WSLA prospectively, not by any desire to expand the WSLA’s reach to civil suits. For all these reasons, we think it clear that the term “offense” in the WSLA applies solely to crimes. But even if there were some ambiguity in the WSLA’s use of that term, our cases instruct us to resolve that ambiguity in favor of the narrower definition. We have said that the WSLA should be “narrowly construed” and “ ‘interpreted in favor of repose.’ ” Id., at 216 (quoting United States v. Scharton, 285 U. S. 518 –522 (1932). Applying that principle here means that the term “offense” must be construed to refer only to crimes. Because this case involves civil claims, the WSLA does not suspend the applicable statute of limitations under either the 1948 or the 2008 version of the statute.[4] IV Petitioners acknowledge that respondent has raised other arguments that, if successful, could render at least one claim timely on remand. We therefore consider whether respondent’s claims must be dismissed with prejudice under the first-to-file rule. We conclude that dismissal with prejudice was not called for. The first-to-file bar provides that “[w]hen a person brings an action . . . no person other than the Government may intervene or bring a related action based on the facts underlying the pending action.” 31 U. S. C. §3730(b)(5) (emphasis added). The term “pending” means “[r]emaining undecided; awaiting decision.” Black’s 1314 (10th ed. 2014). See also Webster’s Third 1669 (1976) (defining “pending” to mean “not yet decided: in continuance: in suspense”). If the reference to a “pending” action in the FCA is interpreted in this way, an earlier suit bars a later suit while the earlier suit remains undecided but ceases to bar that suit once it is dismissed. We see no reason not to interpret the term “pending” in the FCA in accordance with its ordinary meaning. Petitioners argue that Congress used the term “pending” in a very different—and very peculiar—way. In the FCA, according to petitioners, the term “pending” “is ‘used as a short-hand for the first filed action.’ ” Brief for Petitioners 44. Thus, as petitioners see things, the first-filed action remains “pending” even after it has been dismissed, and it forever bars any subsequent related action. This interpretation does not comport with any known usage of the term “pending.” Under this interpretation, Marbury v. Madison, 1 Cranch 137 (1803), is still “pending.” So is the trial of Socrates. Petitioners say that Congress used the term “pending” in the FCA as a sort of “short-hand,” but a shorthand phrase or term is employed to provide a succinct way of expressing a concept that would otherwise require a lengthy or complex formulation. Here, we are told that “pending” is shorthand for “first-filed,” a term that is neither lengthy nor complex. And if Congress had wanted to adopt the rule that petitioners favor, the task could have been accomplished in other equally economical ways—for example, by replacing “pending,” with “earlier” or “prior.” Not only does petitioners’ argument push the term “pending” far beyond the breaking point, but it wouldlead to strange results that Congress is unlikely to have wanted. Under petitioners’ interpretation, a first-filed suit would bar all subsequent related suits even if that earlier suit was dismissed for a reason having nothing to do with the merits. Here, for example, the Thorpe suit, which provided the ground for the initial invocation of the first-to-file rule, was dismissed for failure to prosecute. Why would Congress want the abandonment of an earlier suit to bar a later potentially successful suit that might result in a large recovery for the Government? Petitioners contend that interpreting “pending” to mean pending would produce practical problems, and there is some merit to their arguments. In particular, as petitioners note, if the first-to-file bar is lifted once the first-filed action ends, defendants may be reluctant to settle such actions for the full amount that they would accept if there were no prospect of subsequent suits asserting the same claims. See Brief for Petitioners at 56–57. Respondent and the United States argue that the doctrine of claim preclusion may protect defendants if the first-filed action is decided on the merits, id., at 60–61; United States Brief 30, but that issue is not before us in this case. The False Claims Act’s qui tam provisions present many interpretive challenges, and it is beyond our ability in this case to make them operate together smoothly like a finely tuned machine. We hold that a qui tam suit under the FCA ceases to be “pending” once it is dismissed. We therefore agree with the Fourth Circuit that the dismissal with prejudice of respondent’s one live claim was error. * * * The judgment of the United States Court of Appeals for the Fourth Circuit is reversed in part and affirmed in part, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The Court of Appeals held that the Authorization for Use of Military Force Against Iraq Resolution of 2002, 116Stat. 1498, note following 50 U. S. C. §1541, p. 312, was sufficient to satisfy the “at war” requirement in the pre-2008 version of the WSLA. The Court of Appeals consequently found it unnecessary to decide whether the pre- or post-2008 version of the WSLA governed respondent’s claims. 2 The claims giving rise to the present suit originated in 2005, but respondent filed the operative complaint in 2011. Resolution of the questions before us in this case does not require us to decide which of these two versions of the WSLA applies to respondent’s claims. 3 18 U. S. C. §§38; 248, 670, 1033(a), 1964, 2292(a), 2339B, 2339C. 4 This holding obviates any need to determine which version of the WSLA applies or whether the term “war” in the 1948 Act applies only when Congress has formally declared war.

576.US.2014_13-1402

Respondent Fauzia Din petitioned to have her husband, Kanishka Berashk, a resident citizen of Afghanistan and former civil servant in the Taliban regime, classified as an “immediate relative” entitled to priority immigration status. Din’s petition was approved, but Berashk’s visa application was ultimately denied. A consular officer informed Berashk that he was inadmissible under §1182(a)(3)(B), which excludes aliens who have engaged in “[t]errorist activities,” but the officer provided no further information. Unable to obtain a more detailed explanation for Berashk’s visa denial, Din filed suit in Federal District Court, which dismissed her complaint. The Ninth Circuit reversed, holding that Din had a protected liberty interest in her marriage that entitled her to review of the denial of Berashk’s visa. It further held that the Government deprived her of that liberty interest without due process when it denied Berashk’s visa application without providing a more detailed explanation of its reasons. Held: The judgment is vacated, and the case is remanded. 718 F. 3d 856, vacated and remanded. Justice Scalia, joined by The Chief Justice and Justice Thomas, concluded that the Government did not deprive Din of any constitutional right entitling her to due process of law. Pp. 3–15. (a) Under a historical understanding of the Due Process Clause, Din cannot possibly claim that the denial of Berashk’s visa application deprived her of life, liberty, or property. Pp. 4–5. (b) Even accepting the textually unsupportable doctrine of implied fundamental rights, nothing in that line of cases establishes a free-floating and categorical liberty interest sufficient to trigger constitutional protection whenever a regulation touches upon any aspect of the marital relationship. Even if those cases could be so broadly construed, the relevant question is not whether the asserted interest “is consistent with this Court’s substantive-due-process line of cases,” but whether it is supported by “this Nation’s history and practice,” Washington v. Glucksberg, 521 U. S. 702 –724. Here, the Government’s long practice of regulating immigration, which has included erecting serious impediments to a person’s ability to bring a spouse into the United States, precludes Din’s claim. And this Court has consistently recognized its lack of “judicial authority to substitute [its] political judgment for that of Congress” with regard to the various distinctions in immigration policy. Fiallo v. Bell, 430 U. S. 787 . Pp. 5–11. Justice Kennedy, joined by Justice Alito, concluded that there is no need to decide whether Din has a protected liberty interest, because, even assuming she does, the notice she received satisfied due process. Pp. 1–6. (a) This conclusion is dictated by the reasoning of Kleindienst v. Mandel, 408 U. S. 753 . There the Court declined to balance the asserted First Amendment interest of college professors seeking a nonimmigrant visa for a revolutionary Marxist speaker against “Congress’ ‘plenary power to make rules for the admission of aliens,’ ” id., at 766, and limited its inquiry to whether the Government had provided a “facially legitimate and bona fide” reason for its action, id., at 770. Mandel’s reasoning has particular force here, where national security is involved. Pp. 2–3. (b) Assuming that Din’s rights were burdened directly by the visa denial, the consular officer’s citation of §1182(a)(3)(B) satisfies Mandel’s “facially legitimate and bona fide” standard. Given Congress’ plenary power to “suppl[y] the conditions of the privilege of entry into the United States,” United States ex rel. Knauff v. Shaughnessy, 338 U. S. 537 , the Government’s decision to exclude Berashk because he did not satisfy a statutory condition for admissibility is facially legitimate. Supporting this conclusion is the fact that, by Din’s own admission, Berashk worked for the Taliban government. These considerations lend to the conclusion that there was a bona fide factual basis for exclusion, absent an affirmative showing of bad faith on the consular officer’s part, which Din has not plausibly alleged. Pp. 4–6. Scalia, J., announced the judgment of the Court and delivered an opinion, in which Roberts, C. J., and Thomas, J., joined. Kennedy, J., filed an opinion concurring in the judgment, in which Alito, J., joined. Breyer, J., filed a dissenting opinion, in which Ginsburg, Sotomayor, and Kagan, JJ., joined.

. The first question that we must ask, then, is whether the denial of Berashk’s visa application deprived Din of any of these interests. Only if we answer in the affirmative must we proceed to consider whether the Government’s explanation afforded sufficient process. A The Due Process Clause has its origin in Magna Carta. As originally drafted, the Great Charter provided that “[n]o freeman shall be taken, or imprisoned, or be disseised of his freehold, or liberties, or free customs, or be outlawed, or exiled, or any otherwise destroyed; nor will we not pass upon him, nor condemn him, but by lawful judgment of his peers, or by the law of the land.” Magna Carta, ch. 29, in 1 E. Coke, The Second Part of the Institutes of the Laws of England 45 (1797) (emphasis added). The Court has recognized that at the time of the Fifth Amendment’s ratification, the words “due process of law” were understood “to convey the same meaning as the words ‘by the law of the land’ ” in Magna Carta. Murray’s Lessee v. Hoboken Land & Improvement Co., 18 How. 272, 276 (1856). Although the terminology associated with the guarantee of due process changed dramatically between 1215 and 1791, the general scope of the underlying rights protected stayed roughly constant. Edward Coke, whose Institutes “were read in the American Colonies by virtually every student of law,” Klopfer v. North Carolina, 386 U. S. 213, 225 (1967) , thoroughly described the scope of the interests that could be deprived only pursuant to “the law of the land.” Magna Carta, he wrote, ensured that, without due process, “no man [may] be taken or imprisoned”; “disseised of his lands, or tenements, or dispossessed of his goods, or chattels”; “put from his livelihood without answer”; “barred to have the benefit of the law”; denied “the franchises, and priviledges, which the subjects have of the gift of the king”; “exiled”; or “fore-judged of life, or limbe, disherited, or put to torture, or death.” 1 Coke, supra, at 46–48. Blackstone’s description of the rights protected by Magna Carta is similar, al-though he discusses them in terms much closer to the “life, liberty, or property” terminology used in the Fifth Amendment. He described first an interest in “personal security,” “consist[ing] in a person’s legal and uninterrupted enjoyment of his life, his limbs, his body, his health,and his reputation.” 1 W. Blackstone, Commentaries on the Laws of England 125 (1769). Second, the “personal liberty of individuals” “consist[ed] in the power of loco-motion, of changing situation, or removing one’s person to whatsoever place one’s own inclination may direct; without imprisonment or restraint.” Id., at 130. And finally, a person’s right to property included “the free use, enjoyment, and disposal of all his acquisitions.” Id., at 134. Din, of course, could not conceivably claim that the denial of Berashk’s visa application deprived her—or for that matter even Berashk—of life or property; and under the above described historical understanding, a claim that it deprived her of liberty is equally absurd. The Government has not “taken or imprisoned” Din, nor has it “confine[d]” her, either by “keeping [her] against h[er] will in a private house, putting h[er] in the stocks, arresting or forcibly detaining h[er] in the street.” Id., at 132. Indeed, not even Berashk has suffered a deprivation of liberty so understood. B Despite this historical evidence, this Court has seen fit on several occasions to expand the meaning of “liberty” under the Due Process Clause to include certain implied “fundamental rights.” (The reasoning presumably goes like this: If you have a right to do something, you are free to do it, and deprivation of freedom is a deprivation of “liberty”—never mind the original meaning of that word in the Due Process Clause.) These implied rights have been given more protection than “life, liberty, or property” properly understood. While one may be dispossessed of property, thrown in jail, or even executed so long as proper procedures are followed, the enjoyment of implied constitutional rights cannot be limited at all, except by provisions that are “narrowly tailored to serve a compelling state interest.” Reno v. Flores, 507 U. S. 292 –302 (1993). Din does not explicitly argue that the Government has violated this absolute prohibition of the substantive component of the Due Process Clause, likely because it is obvious that a law barring aliens engaged in terrorist activities from entering this country is narrowly tailored to serve a compelling state interest. She nevertheless insists that, because enforcement of the law affects her enjoyment of an implied fundamental liberty, the Govern-ment must first provide her a full battery of procedural-due-process protections. I think it worth explaining why, even if one accepts the textually unsupportable doctrine of implied fundamental rights, Din’s arguments would fail. Because “extending constitutional protection to an asserted right or liberty interest . . . place[s] the matter outside the arena of public debate and legislative action,” Washington v. Glucksberg, 521 U. S. 702, 720 (1997) , and because the “guideposts for responsible decisionmaking in this unchartered area are scarce and open-ended,” Collins v. Harker Heights, 503 U. S. 115, 125 (1992) , “[t]he doctrine of judicial self-restraint requires us to exercise the utmost care whenever we are asked to break new ground in this field,” ibid. Accordingly, before conferring constitutional status upon a previously unrecognized “liberty,” we have required “a careful description of the asserted fundamental liberty interest,” as well as a demonstration that the interest is “objectively, deeply rooted in this Nation’s history and tradition, and implicit in the concept of ordered liberty, such that neither liberty nor justice would exist if [it was] sacrificed.” Glucksberg, supra, at 720–721 (citations and internal quotation marks omitted). Din describes the denial of Berashk’s visa application as implicating, alternately, a “liberty interest in her marriage,” Brief for Respondent 28, a “right of association with one’s spouse,” id., at 18, “a liberty interest in being reunited with certain blood relatives,” id., at 22, and “the liberty interest of a U. S. citizen under the Due Process Clause to be free from arbitrary restrictions on his right to live with his spouse,” ibid. To be sure, this Court has at times indulged a propensity for grandiloquence when reviewing the sweep of implied rights, describing them so broadly that they would include not only the interests Din asserts but many others as well. For example: “Without doubt, [the liberty guaranteed by the Due Process Clause] denotes not merely freedom from bodily restraint but also the right of the individual to contract, to engage in anyof the common occupations of life, to acquire useful knowledge, to marry, establish a home and bring up children, [and] to worship God according to the dictates of his own conscience” Meyer v. Nebraska, 262 U. S. 390, 399 (1923) . But this Court is not bound by dicta, especially dicta that have been repudiated by the holdings of our subsequent cases. And the actual holdings of the cases Din relies upon hardly establish the capacious right she now asserts. Unlike the States in Loving v. Virginia, 388 U. S. 1 (1967) , Zablocki v. Redhail, 434 U. S. 374 (1978) , and Turner v. Safley, 482 U. S. 78 (1987) , the Federal Government here has not attempted to forbid a marriage. Al-though Din and the dissent borrow language from those cases invoking a fundamental right to marriage, they both implicitly concede that no such right has been infringed in this case. Din relies on the “associational interests in marriage that necessarily are protected by the right to marry,” and that are “presuppose[d]” by later cases establishing a right to marital privacy. Brief for Respondent 16, 18. The dissent supplements the fundamental right to marriage with a fundamental right to live in the United States in order to find an affected liberty interest. Post, at 2–3 (Breyer, J., dissenting). Attempting to abstract from these cases some liberty interest that might be implicated by Berashk’s visa denial, Din draws on even more inapposite cases. Meyer, for example, invalidated a state statute proscribing the teaching of foreign language to children who had not yet passed the eighth grade, reasoning that it violated the teacher’s “right thus to teach and the right of parents to engage him so to instruct their children.” 262 U. S., at 400. Pierce v. Society of Sisters, 268 U. S. 510 –535 (1925), extended Meyer, finding that a law requiring children to attend public schools “interferes with the liberty of parents and guardians to direct the upbringing and education of children under their control.” Moore v. East Cleveland, 431 U. S. 494 –506 (1977), extended this interest in raising children to caretakers in a child’s extended family, striking down an ordinance that limited occupancy of a single-family house to members of a nuclear family on the ground that “[d]ecisions concerning child rearing . . . long have been shared with grandparents or other relatives.” And Griswold v. Connecticut, 381 U. S. 479, 485 (1965) , concluded that a law criminalizing the use of contraceptives by married couples violated “penumbral rights of ‘privacy and repose’ ” protecting “the sacred precincts of the marital bedroom”—rights which do not plausibly extend into the offices of our consulates abroad. Nothing in the cases Din cites establishes a free-floating and categorical liberty interest in marriage (or any other formulation Din offers) sufficient to trigger constitutional protection whenever a regulation in any way touches upon an aspect of the marital relationship. Even if our cases could be construed so broadly, the relevant question is not whether the asserted interest “is consistent with this Court’s substantive-due-process line of cases,” but whether it is supported by “this Nation’s history and practice.” Glucksberg, 521 U. S., at 723–724 (emphasis deleted). Even if we might “imply” a liberty interest in marriage generally speaking, that must give way when there is a tradition denying the specific application of that general interest. Thus, Glucksberg rejected a claimed liberty interest in “self-sovereignty” and “personal autonomy” that extended to assisted suicide when there was a longstanding tradition of outlawing the practice of suicide. Id., at 724, 727–728 (internal quotation marks omitted). Here, a long practice of regulating spousal immigration precludes Din’s claim that the denial of Berashk’s visa application has deprived her of a fundamental liberty interest. Although immigration was effectively unregu-lated prior to 1875, as soon as Congress began legislating in this area it enacted a complicated web of regulations that erected serious impediments to a person’s ability to bring a spouse into the United States. See Abrams, What Makes the Family Special? 80 U. Chi. L. Rev. 7, 10–16 (2013). Most strikingly, perhaps, the Expatriation Act of 1907 provided that “any American woman who marries a foreigner shall take the nationality of her husband.” Ch. 2534, 34Stat. 1228. Thus, a woman in Din’s position not only lacked a liberty interest that might be affected by the Government’s disposition of her husband’s visa application, she lost her own rights as a citizen upon marriage. When Congress began to impose quotas on immigration by country of origin less than 15 years later, with the Immigration Act of 1921, it omitted fiancés and husbands from the family relations eligible for preferred status in the allocation of quota spots. §2(d), 42Stat. 6. Such relations were similarly excluded from the relations eligible for nonquota status, when that status was expanded three years later. Immigration Act of 1924, §4(a), 43Stat. 155. To be sure, these early regulations were premised on the derivative citizenship of women, a legacy of the law of coverture that was already in decline at the time. C. Bredbenner, A Nationality of Her Own 5 (1998). Modern equal-protection doctrine casts substantial doubt on the permissibility of such asymmetric treatment of women citizens in the immigration context, and modern moral judgment rejects the premises of such a legal order. Never-theless, this all-too-recent practice repudiates any con-tention that Din’s asserted liberty interest is “deeply rooted in this Nation’s history and tradition, and implicit in the concept of ordered liberty.” Glucksberg, supra, at 720 (citations and internal quotations marks omitted). Indeed, the law showed little more solicitude for the marital relationship when it was a male resident or citizen seeking admission for his fiancée or wife. The Immigration Act of 1921 granted nonquota status only to unmarried, minor children of citizens, §2(a), while granting fiancées and wives preferred status within the allocation of quota spots, §2(d). In other words, a citizen could move his spouse forward in the line, but once all the quota spots were filled for the year, the spouse was barred without exception. This was not just a theoretical possibility: As one commentator has observed, “[f]or many immigrants, the family categories did little to help, because the quotas were so small that the number of family members seeking slots far outstripped the number available.” Abrams, supra, at 13. Although Congress has tended to show “a continuing and kindly concern . . . for the unity and the happiness of the immigrant family,” E. Hutchinson, Legislative History of American Immigration Policy 1798–1965, p. 518 (1981), this has been a matter of legislative grace rather than fundamental right. Even where Congress has provided special privileges to promote family immigration, it has also “written in careful checks and qualifications.” Ibid. This Court has consistently recognized that these various distinctions are “policy questions entrusted exclusively to the political branches of our Government, and we have no judicial authority to substitute our political judgment for that of the Congress.” Fiallo v. Bell, 430 U. S. 787, 798 (1977) . Only by diluting the meaning of a fundamental liberty interest and jettisoning our established jurisprudence could we conclude that the denial of Berashk’s visa application implicates any of Din’s fundamental liberty interests. C Justice Breyer suggests that procedural due process rights attach to liberty interests that either are (1) created by nonconstitutional law, such as a statute, or (2) “sufficiently important” so as to “flow ‘implicit[ly]’ from the design, object, and nature of the Due Process Clause.” Post, at 2. The first point is unobjectionable, at least given this Court’s case law. See, e.g., Goldberg v. Kelly, 397 U. S. 254 , and n. 8 (1970); Collins 503 U. S., at 129. But it is unhelpful to Din, who does not argue that a statute confers on her a liberty interest protected by the Due Process Clause. Justice Breyer attempts to make this argument for Din, latching onto language in Wilkinson v. Austin, 545 U. S. 209, 221 (2005) , saying that a liberty interest “may arise from an expectation or interest created by state laws or policies.” Such an “expectation” has been created here, he asserts, because “the law . . . surrounds marriage with a host of legal protections to the point that it creates a strong expectation that government will not deprive married individuals of their freedom to live together without strong reasons and (in individual cases) without fair procedure,” post, at 3. But what Wilkinson meant by an “expectation or interest” was not that sort of judicially unenforceable substantial hope, but a present and legally recognized substantive entitlement.[1]* As sole support for its conclusion that nonconstitutional law can create constitutionally protected liberty interests, Wilkinson cited Wolff v. McDonnell, 418 U. S. 539 –558 (1974), which held that a prisoner could not be deprived of statutory good-time credit without procedural due process. That was not because a prisoner might have “ ‘a strong expectation’ ” that the government would not deprive him of good-time credit “ ‘without strong reasons’ ” or “ ‘fair procedure,’ ” but because “the State itself has not only provided a statutory right to good time [credit] but also specifies that it is to be forfeited only for serious misbehavior,” id., at 557 (emphasis added). The legal benefits afforded to marriages and the preferential treatment accorded to visa applicants with citizen relatives are insufficient to confer on Din a right that can be deprived only pursuant to procedural due process. Justice Breyer’s second point—that procedural due process rights attach even to some nonfundamental liberty interests that have not been created by statute—is much more troubling. He relies on the implied-fundamental-rights cases discussed above to divine a “right of spouses to live together and to raise a family,” along with “a citizen’s right to live within this country.” Post, at 2–3. But perhaps recognizing that our established methodology for identifying fundamental rights cuts against his conclusion, see Part II–B, supra, he argues that the term “liberty” in the Due Process Clause includes implied rights that, although not so fundamental as to deserve substantive-due-process protection, are important enough to deserve procedural-due-process protection. Post, at 2. In other words, there are two categories of implied rights protected by the Due Process Clause: really fundamental rights, which cannot be taken away at all absent a compelling state interest; and not-so-fundamental rights, whichcan be taken away so long as procedural due process is observed. The dissent fails to cite a single case supporting its novel theory of implied nonfundamental rights. It is certainly true that Vitek v. Jones, 445 U. S. 480 (1980) , and Washington v. Harper, 494 U. S. 210 (1990) , do not entail implied fundamental rights, but this is because they do not entail implied rights at all. Vitek concerned the involuntary commitment of a prisoner, deprivation of the expressly protected right of liberty under the original understanding of the term, see Part II–A, supra. “ ‘Among the historic liberties’ protected by the Due Process Clause is the ‘right to be free from, and to obtain judicial relief for, unjustified intrusions on personal security.’ ” Vitek, supra, at 492. The same is true of Harper, which concerned forced administration of psychotropic drugs to an inmate. 494 U. S., at 214. Arguably, Paul v. Davis, 424 U. S. 693 (1976) , also addressed an interest expressly contemplated within the meaning of “liberty.” See 1 W. Blackstone, Commentaries on the Laws of England 125 (“The right of personal security consists in a person’s . . . reputation”). But that case is of no help to the dissent anyway, since it found no liberty interest entitled to the Due Process Clause’s protection. Paul, supra, at 713–714. Finally, the dissent points to Goss v. Lopez, 419 U. S. 565, 574 (1975) , a case that “recognize[d] . . . as a property interest” a student’s right to a public education conferred by Ohio’s express statutory creation of a public school system; and further concluded that the student’s 10-day suspension implicated the constitutionally grounded liberty interest in “ ‘a person’s good name, reputation, honor, or integrity.’ ” Ultimately, the dissent identifies no case holding that there is an implied nonfundamental right protected by procedural due process, and only one case even suggesting that there is. That suggestion, in Smith v. Organization of Foster Families For Equality & Reform, 431 U. S. 816 (1977) , is contained in dictum in a footnote, id., at 842, n. 48. The holding of the case was that “the procedures provided by New York State . . . and by New York Cit[y] . . . are adequate to protect whatever liberty interests appellees may have.” Id., at 856 (emphasis added). The footnoted dictum that Justice Breyer proposes to elevate to constitutional law is a dangerous doctrine. It vastly expands the scope of our implied-rights jurisprudence by setting it free from the requirement that the liberty interest be “objectively, deeply rooted in this Nation’s history and tradition, and implicit in the concept of ordered liberty,” Glucksberg, 521 U. S., at 720–721 (internal quotation marks omitted). Even shallow-rooted liberties would, thanks to this new procedural-rights-only notion of quasi-fundamental rights, qualify for judicially imposed procedural requirements. Moreover, Justice Breyer gives no basis for distinguishing the fundamental rights recognized in the cases he depends on from the nonfundamental right he believes they give rise to in the present case. Neither Din’s right to live with her spouse nor her right to live within this country is implicated here. There is a “simple distinction between government action that di-rectly affects a citizen’s legal rights, or imposes a direct re-straint on his liberty, and action that is directed against a third party and affects the citizen only indirectly or incidentally.” O’Bannon v. Town Court Nursing Center, 447 U. S. 773, 788 (1980) . The Government has not refused to recognize Din’s marriage to Berashk, and Din remains free to live with her husband anywhere in the world that both individuals are permitted to reside. And the Government has not expelled Din from the country. It has simply determined that Kanishka Berashk engaged in terrorist activities within the meaning of the Immigration and Nationality Act, and has therefore denied him admission into the country. This might, indeed, deprive Din of something “important,” post, at 2, but if that is the criterion for Justice Breyer’s new pairing of substantive and procedural due process, we are in for quite a ride. * * * Because Fauzia Din was not deprived of “life, liberty, or property” when the Government denied Kanishka Berashk admission to the United States, there is no process due to her under the Constitution. To the extent that she received any explanation for the Government’s decision, this was more than the Due Process Clause required. The judgment of the Ninth Circuit is vacated, and the case is remanded for further proceedings. It is so ordered.Notes 1 * Justice Breyer characterizes this as a reintroduction of “the rights/privilege distinction that this Court rejected almost five decades ago.” Post, at 3. Not so. All I insist upon (and all that our cases over the past five decades require) is that the privilege be one to which the claimant has been given an entitlement.

576.US.2014_14-114

The Patient Protection and Affordable Care Act grew out of a long history of failed health insurance reform. In the 1990s, several States sought to expand access to coverage by imposing a pair of insurance market regulations—a “guaranteed issue” requirement, which bars insurers from denying coverage to any person because of his health, and a “community rating” requirement, which bars insurers from charging a person higher premiums for the same reason. The reforms achieved the goal of expanding access to coverage, but they also encouraged people to wait until they got sick to buy insurance. The result was an economic “death spiral”: premiums rose, the number of people buying insurance declined, and insurers left the market entirely. In 2006, however, Massachusetts discovered a way to make the guaranteed issue and community rating requirements work—by requiring individuals to buy insurance and by providing tax credits to certain individuals to make insurance more affordable. The combination of these three reforms—insurance market regulations, a coverage mandate, and tax credits—enabled Massachusetts to drastically reduce its uninsured rate. The Affordable Care Act adopts a version of the three key reforms that made the Massachusetts system successful. First, the Act adopts the guaranteed issue and community rating requirements. 42 U. S. C. §§300gg, 300gg–1. Second, the Act generally requires individuals to maintain health insurance coverage or make a payment to the IRS, unless the cost of buying insurance would exceed eight percent of that individual’s income. 26 U. S. C. §5000A. And third, the Act seeks to make insurance more affordable by giving refundable tax credits to individuals with household incomes between 100 percent and 400 percent of the federal poverty line. §36B. In addition to those three reforms, the Act requires the creation of an “Exchange” in each State—basically, a marketplace that allows people to compare and purchase insurance plans. The Act gives each State the opportunity to establish its own Exchange, but provides that the Federal Government will establish “such Exchange” if the State does not. 42 U. S. C. §§18031, 18041. Relatedly, the Act provides that tax credits “shall be allowed” for any “applicable taxpayer,” 26 U. S. C. §36B(a), but only if the taxpayer has enrolled in an insurance plan through “an Exchange established by the State under [ 42 U. S. C. §18031],” §§36B(b)–(c). An IRS regulation interprets that language as making tax credits available on “an Exchange,” 26 CFR §1.36B–2, “regardless of whether the Exchange is established and operated by a State . . . or by HHS,” 45 CFR §155.20. Petitioners are four individuals who live in Virginia, which has a Federal Exchange. They do not wish to purchase health insurance. In their view, Virginia’s Exchange does not qualify as “an Exchange established by the State under [ 42 U. S. C. §18031],” so they should not receive any tax credits. That would make the cost of buying insurance more than eight percent of petitioners’ income, exempting them from the Act’s coverage requirement. As a result of the IRS Rule, however, petitioners would receive tax credits. That would make the cost of buying insurance less than eight percent of their income, which would subject them to the Act’s coverage requirement. Petitioners challenged the IRS Rule in Federal District Court. The District Court dismissed the suit, holding that the Act unambiguously made tax credits available to individuals enrolled through a Federal Exchange. The Court of Appeals for the Fourth Circuit affirmed. The Fourth Circuit viewed the Act as ambiguous, and deferred to the IRS’s interpretation under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 . Held: Section 36B’s tax credits are available to individuals in States that have a Federal Exchange. Pp. 7–21. (a) When analyzing an agency’s interpretation of a statute, this Court often applies the two-step framework announced in Chevron, 467 U. S. 837 . But Chevron does not provide the appropriate framework here. The tax credits are one of the Act’s key reforms and whether they are available on Federal Exchanges is a question of deep “economic and political significance”; had Congress wished to assign that question to an agency, it surely would have done so expressly. And it is especially unlikely that Congress would have delegated this decision to the IRS, which has no expertise in crafting health insurance policy of this sort. It is instead the Court’s task to determine the correct reading of Section 36B. If the statutory language is plain, the Court must enforce it according to its terms. But oftentimes the meaning—or ambiguity—of certain words or phrases may only become evident when placed in context. So when deciding whether the language is plain, the Court must read the words “in their context and with a view to their place in the overall statutory scheme.” FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120 . Pp. 7–9. (b) When read in context, the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” is properly viewed as ambiguous. The phrase may be limited in its reach to State Exchanges. But it could also refer to all Exchanges—both State and Federal—for purposes of the tax credits. If a State chooses not to follow the directive in Section 18031 to establish an Exchange, the Act tells the Secretary of Health and Human Services to establish “such Exchange.” §18041. And by using the words “such Exchange,” the Act indicates that State and Federal Exchanges should be the same. But State and Federal Exchanges would differ in a fundamental way if tax credits were available only on State Exchanges—one type of Exchange would help make insurance more affordable by providing billions of dollars to the States’ citizens; the other type of Exchange would not. Several other provisions in the Act—e.g., Section 18031(i)(3)(B)’s requirement that all Exchanges create outreach programs to “distribute fair and impartial information concerning . . . the availability of premium tax credits under section 36B”—would make little sense if tax credits were not available on Federal Exchanges. The argument that the phrase “established by the State” would be superfluous if Congress meant to extend tax credits to both State and Federal Exchanges is unpersuasive. This Court’s “preference for avoiding surplusage constructions is not absolute.” Lamie v. United States Trustee, 540 U. S. 526 . And rigorous application of that canon does not seem a particularly useful guide to a fair construction of the Affordable Care Act, which contains more than a few examples of inartful drafting. The Court nevertheless must do its best, “bearing in mind the ‘fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.’ ” Utility Air Regulatory Group v. EPA, 573 U. S. ___, ___. Pp. 9–15. (c) Given that the text is ambiguous, the Court must look to the broader structure of the Act to determine whether one of Section 36B’s “permissible meanings produces a substantive effect that is compatible with the rest of the law.” United Sav. Assn. of Tex. v. Timbers of Inwood Forest Associates, Ltd., 484 U. S. 365 . Here, the statutory scheme compels the Court to reject petitioners’ interpretation because it would destabilize the individual insurance market in any State with a Federal Exchange, and likely create the very “death spirals” that Congress designed the Act to avoid. Under petitioners’ reading, the Act would not work in a State with a Federal Exchange. As they see it, one of the Act’s three major reforms—the tax credits—would not apply. And a second major reform—the coverage requirement—would not apply in a meaningful way, because so many individuals would be exempt from the requirement without the tax credits. If petitioners are right, therefore, only one of the Act’s three major reforms would apply in States with a Federal Exchange.The combination of no tax credits and an ineffective coverage requirement could well push a State’s individual insurance market into a death spiral. It is implausible that Congress meant the Act to operate in this manner. Congress made the guaranteed issue and community rating requirements applicable in every State in the Nation, but those requirements only work when combined with the coverage requirement and tax credits. It thus stands to reason that Congress meant for those provisions to apply in every State as well. Pp. 15–19. (d) The structure of Section 36B itself also suggests that tax credits are not limited to State Exchanges. Together, Section 36B(a), which allows tax credits for any “applicable taxpayer,” and Section 36B(c)(1), which defines that term as someone with a household income between 100 percent and 400 percent of the federal poverty line, appear to make anyone in the specified income range eligible for a tax credit. According to petitioners, however, those provisions are an empty promise in States with a Federal Exchange. In their view, an applicable taxpayer in such a State would be eligible for a tax credit, but the amount of that tax credit would always be zero because of two provisions buried deep within the Tax Code. That argument fails because Congress “does not alter the fundamental details of a regulatory scheme in vague terms or ancillary provisions.” Whitman v. American Trucking Assns., Inc., 531 U. S. 457 . Pp. 19–20. (e) Petitioners’ plain-meaning arguments are strong, but the Act’s context and structure compel the conclusion that Section 36B allows tax credits for insurance purchased on any Exchange created under the Act. Those credits are necessary for the Federal Exchanges to function like their State Exchange counterparts, and to avoid the type of calamitous result that Congress plainly meant to avoid. Pp. 20–21. 759 F. 3d 358, affirmed. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Scalia, J., filed a dissenting opinion, in which Thomas and Alito, JJ., joined.

The Patient Protection and Affordable Care Act adopts a series of interlocking reforms designed to expand coverage in the individual health insurance market. First, the Act bars insurers from taking a person’s health into account when deciding whether to sell health insurance or how much to charge. Second, the Act generally requires each person to maintain insurance coverage or make a payment to the Internal Revenue Service. And third, the Act gives tax credits to certain people to make insurance more affordable. In addition to those reforms, the Act requires the creation of an “Exchange” in each State—basically, a marketplace that allows people to compare and purchase insurance plans. The Act gives each State the opportunity to establish its own Exchange, but provides that the Federal Government will establish the Exchange if the State does not. This case is about whether the Act’s interlocking reforms apply equally in each State no matter who establishes the State’s Exchange. Specifically, the question pre-sented is whether the Act’s tax credits are available in States that have a Federal Exchange. I A The Patient Protection and Affordable Care Act, 124Stat. 119, grew out of a long history of failed health insurance reform. In the 1990s, several States began experimenting with ways to expand people’s access to coverage. One common approach was to impose a pair of insurance market regulations—a “guaranteed issue” requirement, which barred insurers from denying coverage to any person because of his health, and a “community rating” requirement, which barred insurers from charging a person higher premiums for the same reason. Together, those requirements were designed to ensure that anyone who wanted to buy health insurance could do so. The guaranteed issue and community rating requirements achieved that goal, but they had an unintended consequence: They encouraged people to wait until they got sick to buy insurance. Why buy insurance coverage when you are healthy, if you can buy the same coverage for the same price when you become ill? This consequence—known as “adverse selection”—led to a second: Insurers were forced to increase premiums to account for the fact that, more and more, it was the sick rather than the healthy who were buying insurance. And that consequence fed back into the first: As the cost of insurance rose, even more people waited until they became ill tobuy it. This led to an economic “death spiral.” As premiums rose higher and higher, and the number of people buying insurance sank lower and lower, insurers began to leave the market entirely. As a result, the number of people without insurance increased dramatically. This cycle happened repeatedly during the 1990s. For example, in 1993, the State of Washington reformed its individual insurance market by adopting the guaranteed issue and community rating requirements. Over the next three years, premiums rose by 78 percent and the number of people enrolled fell by 25 percent. By 1999, 17 of the State’s 19 private insurers had left the market, and the remaining two had announced their intention to do so. Brief for America’s Health Insurance Plans as Amicus Curiae 10–11. For another example, also in 1993, New York adopted the guaranteed issue and community rating requirements. Over the next few years, some major insurers in the individual market raised premiums by roughly 40 percent. By 1996, these reforms had “effectively eliminated the commercial individual indemnity market in New York with the largest individual health insurer exiting the market.” L. Wachenheim & H. Leida, The Impact of Guaranteed Issue and Community Rating Reforms on States’ Individual Insurance Markets 38 (2012). In 1996, Massachusetts adopted the guaranteed issue and community rating requirements and experienced similar results. But in 2006, Massachusetts added two more reforms: The Commonwealth required individuals to buy insurance or pay a penalty, and it gave tax credits to certain individuals to ensure that they could afford the insurance they were required to buy. Brief for Bipartisan Economic Scholars as Amici Curiae 24–25. The combination of these three reforms—insurance market regulations, a coverage mandate, and tax credits—reduced the uninsured rate in Massachusetts to 2.6 percent, by far the lowest in the Nation. Hearing on Examining Individual State Experiences with Health Care Reform Coverage Initiatives in the Context of National Reform before the Senate Committee on Health, Education, Labor, and Pensions, 111th Cong., 1st Sess., 9 (2009). B The Affordable Care Act adopts a version of the three key reforms that made the Massachusetts system successful. First, the Act adopts the guaranteed issue and community rating requirements. The Act provides that “each health insurance issuer that offers health insurance coverage in the individual . . . market in a State must accept every . . . individual in the State that applies for such coverage.” 42 U. S. C. §300gg–1(a). The Act also bars insurers from charging higher premiums on the basis of a person’s health. §300gg. Second, the Act generally requires individuals to maintain health insurance coverage or make a payment to the IRS. 26 U. S. C. §5000A. Congress recognized that, without an incentive, “many individuals would wait to purchase health insurance until they needed care.” 42 U. S. C. §18091(2)(I). So Congress adopted a coverage requirement to “minimize this adverse selection and broaden the health insurance risk pool to include healthy individuals, which will lower health insurance premiums.” Ibid. In Congress’s view, that coverage requirement was “essential to creating effective health insurance markets.” Ibid. Congress also provided an exemption from the coverage requirement for anyone who has to spend more than eight percent of his income on health insurance. 26 U. S. C. §§5000A(e)(1)(A), (e)(1)(B)(ii). Third, the Act seeks to make insurance more affordable by giving refundable tax credits to individuals with household incomes between 100 percent and 400 percent of the federal poverty line. §36B. Individuals who meet the Act’s requirements may purchase insurance with the tax credits, which are provided in advance directly to the individual’s insurer. 42 U. S. C. §§18081, 18082. These three reforms are closely intertwined. As noted, Congress found that the guaranteed issue and community rating requirements would not work without the coverage requirement. §18091(2)(I). And the coverage requirement would not work without the tax credits. The reason is that, without the tax credits, the cost of buying insurance would exceed eight percent of income for a large number of individuals, which would exempt them from the coverage requirement. Given the relationship between these three reforms, the Act provided that they should take effect on the same day—January 1, 2014. See Affordable Care Act, §1253, redesignated §1255, 124Stat. 162, 895; §§1401(e), 1501(d), id., at 220, 249. C In addition to those three reforms, the Act requires the creation of an “Exchange” in each State where peoplecan shop for insurance, usually online. 42 U. S. C. §18031(b)(1). An Exchange may be created in one of two ways. First, the Act provides that “[e]ach State shall . . . establish an American Health Benefit Exchange . . . for the State.” Ibid. Second, if a State nonetheless chooses not to establish its own Exchange, the Act provides that the Secretary of Health and Human Services “shall . . . establish and operate such Exchange within the State.” §18041(c)(1). The issue in this case is whether the Act’s tax credits are available in States that have a Federal Exchange rather than a State Exchange. The Act initially provides that tax credits “shall be allowed” for any “applicable taxpayer.” 26 U. S. C. §36B(a). The Act then provides that the amount of the tax credit depends in part on whether the taxpayer has enrolled in an insurance plan through “an Exchange established by the State under section 1311 of the Patient Protection and Affordable Care Act [hereinafter 42 U. S. C. §18031].” 26 U. S. C. §§36B(b)–(c) (emphasis added). The IRS addressed the availability of tax credits by promulgating a rule that made them available on both State and Federal Exchanges. 77 Fed. Reg. 30378 (2012). As relevant here, the IRS Rule provides that a taxpayer is eligible for a tax credit if he enrolled in an insurance plan through “an Exchange,” 26 CFR §1.36B–2 (2013), which is defined as “an Exchange serving the individual market . . . regardless of whether the Exchange is established and operated by a State . . . or by HHS,” 45 CFR §155.20 (2014). At this point, 16 States and the District of Columbia have established their own Exchanges; the other 34 States have elected to have HHS do so. D Petitioners are four individuals who live in Virginia, which has a Federal Exchange. They do not wish to purchase health insurance. In their view, Virginia’s Exchange does not qualify as “an Exchange established by the State under [ 42 U. S. C. §18031],” so they should not receive any tax credits. That would make the cost of buying insurance more than eight percent of their income, which would exempt them from the Act’s coverage requirement. 26 U. S. C. §5000A(e)(1). Under the IRS Rule, however, Virginia’s Exchange would qualify as “an Exchange established by the State under [ 42 U. S. C. §18031],” so petitioners would receive tax credits. That would make the cost of buying insurance less than eight percent of petitioners’ income, which would subject them to the Act’s coverage requirement. The IRS Rule therefore requires petitioners to either buy health insurance they do not want, or make a payment to the IRS. Petitioners challenged the IRS Rule in Federal District Court. The District Court dismissed the suit, holding that the Act unambiguously made tax credits available to individuals enrolled through a Federal Exchange. King v. Sebelius, 997 F. Supp. 2d 415 (ED Va. 2014). The Court of Appeals for the Fourth Circuit affirmed. 759 F. 3d 358 (2014). The Fourth Circuit viewed the Act as “ambiguous and subject to at least two different interpretations.” Id., at 372. The court therefore deferred to the IRS’s interpretation under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) . 759 F. 3d, at 376. The same day that the Fourth Circuit issued its decision, the Court of Appeals for the District of Columbia Circuit vacated the IRS Rule in a different case, holding that the Act “unambiguously restricts” the tax credits to State Exchanges. Halbig v. Burwell, 758 F. 3d 390, 394 (2014). We granted certiorari in the present case. 574 U. S. ___ (2014). II The Affordable Care Act addresses tax credits in what is now Section 36B of the Internal Revenue Code. That section provides: “In the case of an applicable taxpayer, there shall be allowed as a credit against the tax imposed by this subtitle . . . an amount equal to the premium assistance credit amount.” 26 U. S. C. §36B(a). Section 36B then defines the term “premium assistance credit amount” as “the sum of the premium assistance amounts determined under paragraph (2) with respect to all coverage months of the taxpayer occurring during the taxable year.” §36B(b)(1) (emphasis added). Section 36B goes on to define the two italicized terms—“premium assistance amount” and “coverage month”—in part by referring to an insurance plan that is enrolled in through “an Exchange established by the State under [ 42 U. S. C. §18031].” 26 U. S. C. §§36B(b)(2)(A), (c)(2)(A)(i). The parties dispute whether Section 36B authorizes tax credits for individuals who enroll in an insurance plan through a Federal Exchange. Petitioners argue that a Federal Exchange is not “an Exchange established by the State under [ 42 U. S. C. §18031],” and that the IRS Rule therefore contradicts Section 36B. Brief for Petitioners 18–20. The Government responds that the IRS Rule is lawful because the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” should be read to include Federal Exchanges. Brief for Respondents 20–25. When analyzing an agency’s interpretation of a statute, we often apply the two-step framework announced in Chevron, 467 U. S. 837 . Under that framework, we ask whether the statute is ambiguous and, if so, whether the agency’s interpretation is reasonable. Id., at 842–843. This approach “is premised on the theory that a statute’s ambiguity constitutes an implicit delegation from Congress to the agency to fill in the statutory gaps.” FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120, 159 (2000) . “In extraordinary cases, however, there may be reason to hesitate before concluding that Congress has intended such an implicit delegation.” Ibid. This is one of those cases. The tax credits are among the Act’s key reforms, involving billions of dollars in spending each year and affecting the price of health insurance for millions of people. Whether those credits are available on Federal Exchanges is thus a question of deep “economic and political significance” that is central to this statutory scheme; had Congress wished to assign that question to an agency, it surely would have done so expressly. Utility Air Regulatory Group v. EPA, 573 U. S. ___, ___ (2014) (slip op., at 19) (quoting Brown & Williamson, 529 U. S., at 160). It is especially unlikely that Congress would have delegated this decision to the IRS, which has no expertise in crafting health insurance policy of this sort. See Gonzales v. Oregon, 546 U. S. 243 –267 (2006). This is not a case for the IRS. It is instead our task to determine the correct reading of Section 36B. If the statutory language is plain, we must enforce it according to its terms. Hardt v. Reliance Standard Life Ins. Co., 560 U. S. 242, 251 (2010) . But oftentimes the “meaning—or ambiguity—of certain words or phrases may only become evident when placed in context.” Brown & Williamson, 529 U. S., at 132. So when deciding whether the language is plain, we must read the words “in their context and with a view to their place in the overall statutory scheme.” Id., at 133 (internal quotation marks omitted). Our duty, after all, is “to construe statutes, not isolated provisions.” Graham County Soil and Water Conservation Dist. v. United States ex rel. Wilson, 559 U. S. 280, 290 (2010) (internal quotation marks omitted). A We begin with the text of Section 36B. As relevant here, Section 36B allows an individual to receive tax credits only if the individual enrolls in an insurance plan through “an Exchange established by the State under [ 42 U. S. C. §18031].” In other words, three things must be true: First, the individual must enroll in an insurance plan through “an Exchange.” Second, that Exchange must be “established by the State.” And third, that Exchange must be established “under [ 42 U. S. C. §18031].” We address each requirement in turn. First, all parties agree that a Federal Exchange qualifies as “an Exchange” for purposes of Section 36B. See Brief for Petitioners 22; Brief for Respondents 22. Section 18031 provides that “[e]ach State shall . . . establish an American Health Benefit Exchange . . . for the State.” §18031(b)(1). Although phrased as a requirement, the Act gives the States “flexibility” by allowing them to “elect” whether they want to establish an Exchange. §18041(b). If the State chooses not to do so, Section 18041 provides that the Secretary “shall . . . establish and operatesuch Exchange within the State.” §18041(c)(1) (emphasis added). By using the phrase “such Exchange,” Section 18041 instructs the Secretary to establish and operate the same Exchange that the State was directed to establish under Section 18031. See Black’s Law Dictionary 1661 (10th ed. 2014) (defining “such” as “That or those; having just been mentioned”). In other words, State Exchanges and Fed-eral Exchanges are equivalent—they must meet the same requirements, perform the same functions, and serve the same purposes. Although State and Federal Exchanges are established by different sovereigns, Sections 18031 and 18041 do not suggest that they differ in any meaningful way. A Federal Exchange therefore counts as “an Exchange” under Section 36B. Second, we must determine whether a Federal Exchange is “established by the State” for purposes of Section 36B. At the outset, it might seem that a Federal Exchange cannot fulfill this requirement. After all, the Act defines “State” to mean “each of the 50 States and the District of Columbia”—a definition that does not include the Federal Government. 42 U. S. C. §18024(d). But when read in context, “with a view to [its] place in the overall statutory scheme,” the meaning of the phrase “established by the State” is not so clear. Brown &Williamson, 529 U. S., at 133 (internal quotation marks omitted). After telling each State to establish an Exchange, Section 18031 provides that all Exchanges “shall make available qualified health plans to qualified individuals.” 42 U. S. C. §18031(d)(2)(A). Section 18032 then defines the term “qualified individual” in part as an individual who “resides in the State that established the Exchange.” §18032(f)(1)(A). And that’s a problem: If we give the phrase “the State that established the Exchange” its most natural meaning, there would be no “qualified individuals” on Federal Exchanges. But the Act clearly contemplates that there will be qualified individuals on every Exchange. As we just mentioned, the Act requires all Exchanges to “make available qualified health plans to qualified individuals”—something an Exchange could not do if there were no such individuals. §18031(d)(2)(A). And the Act tells the Exchange, in deciding which health plans to offer, to consider “the interests of qualified individuals . . . in the State or States in which such Exchange operates”—again, something the Exchange could not do if qualified individ-uals did not exist. §18031(e)(1)(B). This problem arises repeatedly throughout the Act. See, e.g., §18031(b)(2) (allowing a State to create “one Exchange . . . for providing . . . services to both qualified individuals and qualified small employers,” rather than creating separate Exchanges for those two groups).[1] These provisions suggest that the Act may not always use the phrase “established by the State” in its most natural sense. Thus, the meaning of that phrase may not be as clear as it appears when read out of context. Third, we must determine whether a Federal Exchange is established “under [ 42 U. S. C. §18031].” This too might seem a requirement that a Federal Exchange cannot fulfill, because it is Section 18041 that tells the Secretary when to “establish and operate such Exchange.” But here again, the way different provisions in the statute interact suggests otherwise. The Act defines the term “Exchange” to mean “an American Health Benefit Exchange established under section 18031.” §300gg–91(d)(21). If we import that definition into Section 18041, the Act tells the Secretary to “establish and operate such ‘American Health Benefit Exchange established under section 18031.’ ” That suggests that Section 18041 authorizes the Secretary to establish an Exchange under Section 18031, not (or not only) under Section 18041. Otherwise, the Federal Exchange, by definition, would not be an “Exchange” at all. See Halbig, 758 F. 3d, at 399–400 (acknowledging that the Secretary establishes Federal Exchanges under Section 18031). This interpretation of “under [ 42 U. S. C. §18031]” fits best with the statutory context. All of the requirements that an Exchange must meet are in Section 18031, so it is sensible to regard all Exchanges as established under that provision. In addition, every time the Act uses the word “Exchange,” the definitional provision requires that we substitute the phrase “Exchange established under section 18031.” If Federal Exchanges were not established under Section 18031, therefore, literally none of the Act’s requirements would apply to them. Finally, the Act repeatedly uses the phrase “established under [ 42 U. S. C. §18031]” in situations where it would make no sense to distinguish between State and Federal Exchanges. See, e.g., 26 U. S. C. §125(f)(3)(A) (2012 ed., Supp. I) (“The term ‘qualified benefit’ shall not include any qualified health plan . . . offered through an Exchange established under [ 42 U. S. C. §18031]”); 26 U. S. C. §6055(b)(1)(B)(iii)(I) (2012 ed.) (requiring insurers to report whether each insurance plan they provided “is a qualified health plan offered through an Exchange established under [ 42 U. S. C. §18031]”). A Federal Exchange may therefore be considered one established “under [ 42 U. S. C. §18031].” The upshot of all this is that the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” is properly viewed as ambiguous. The phrase may be limited in its reach to State Exchanges. But it is also possible that the phrase refers to all Exchanges—both State and Federal—at least for purposes of the tax credits. If a State chooses not to follow the directive in Section 18031 that it establish an Exchange, the Act tells the Secretary to establish “such Exchange.” §18041. And by using the words “such Exchange,” the Act indicates that State and Federal Exchanges should be the same. But State and Federal Exchanges would differ in a fundamental way if tax credits were available only on State Exchanges—one type of Exchange would help make insurance more affordable by providing billions of dollars to the States’ citizens; the other type of Exchange would not.[2] The conclusion that Section 36B is ambiguous is further supported by several provisions that assume tax credits will be available on both State and Federal Exchanges. For example, the Act requires all Exchanges to create outreach programs that must “distribute fair and impartial information concerning . . . the availability of premium tax credits under section 36B.” §18031(i)(3)(B). The Act also requires all Exchanges to “establish and make avail-able by electronic means a calculator to determine the actual cost of coverage after the application of any pre-mium tax credit under section 36B.” §18031(d)(4)(G). And the Act requires all Exchanges to report to the Treasury Secretary information about each health plan they sell, including the “aggregate amount of any advance payment of such credit,” “[a]ny information . . . necessary to determine eligibility for, and the amount of, such credit,” and any “[i]nformation necessary to determine whether a taxpayer has received excess advance payments.” 26 U. S. C. §36B(f)(3). If tax credits were not available on Federal Exchanges, these provisions would make little sense. Petitioners and the dissent respond that the words “established by the State” would be unnecessary if Congress meant to extend tax credits to both State and Fed-eral Exchanges. Brief for Petitioners 20; post, at 4–5. But “our preference for avoiding surplusage constructions is not absolute.” Lamie v. United States Trustee, 540 U. S. 526, 536 (2004) ; see also Marx v. General Revenue Corp., 568 U. S. ___, ___ (2013) (slip op., at 13) (“The canon against surplusage is not an absolute rule”). And specifically with respect to this Act, rigorous application of the canon does not seem a particularly useful guide to a fair construction of the statute. The Affordable Care Act contains more than a few examples of inartful drafting. (To cite just one, the Act creates three separate Section 1563s. See 124Stat. 270, 911, 912.) Several features of the Act’s passage contributed to that unfortunate reality. Congress wrote key partsof the Act behind closed doors, rather than through “the traditional legislative process.” Cannan, A Legislative History of the Affordable Care Act: How Legislative Procedure Shapes Legislative History, 105 L. Lib. J. 131, 163 (2013). And Congress passed much of the Act using a complicated budgetary procedure known as “reconciliation,” which limited opportunities for debate and amendment, and bypassed the Senate’s normal 60-vote filibuster requirement. Id., at 159–167. As a result, the Act does not reflect the type of care and deliberation that one might expect of such significant legislation. Cf. Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. 527, 545 (1947) (describing a cartoon “in which a senator tells his colleagues ‘I admit this new bill is too complicated to understand. We’ll just have to pass it to find out what it means.’ ”). Anyway, we “must do our best, bearing in mind the fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” Util-ity Air Regulatory Group, 573 U. S., at ___ (slip op., at 15) (internal quotation marks omitted). After reading Section 36B along with other related provisions in the Act, we cannot conclude that the phrase “an Exchange established by the State under [Section 18031]” is unambiguous. B Given that the text is ambiguous, we must turn to the broader structure of the Act to determine the meaning of Section 36B. “A provision that may seem ambiguous in isolation is often clarified by the remainder of the statu-tory scheme . . . because only one of the permissible meanings produces a substantive effect that is compatible with the rest of the law.” United Sav. Assn. of Tex. v. Timbers of Inwood Forest Associates, Ltd., 484 U. S. 365, 371 (1988) . Here, the statutory scheme compels us to reject petitioners’ interpretation because it would destabilize the individual insurance market in any State with a Federal Exchange, and likely create the very “death spirals” that Congress designed the Act to avoid. See New York State Dept. of Social Servs. v. Dublino, 413 U. S. 405 –420 (1973) (“We cannot interpret federal statutes to negate their own stated purposes.”).[3] As discussed above, Congress based the Affordable Care Act on three major reforms: first, the guaranteed issue and community rating requirements; second, a requirement that individuals maintain health insurance coverage or make a payment to the IRS; and third, the tax credits for individuals with household incomes between 100 percent and 400 percent of the federal poverty line. In a State that establishes its own Exchange, these three reforms work together to expand insurance coverage. The guaranteed issue and community rating requirements ensure that anyone can buy insurance; the coverage requirement creates an incentive for people to do so before they get sick; and the tax credits—it is hoped—make insurance more affordable. Together, those reforms “minimize . . . adverse selection and broaden the health in-surance risk pool to include healthy individuals, which will lower health insurance premiums.” 42 U. S. C. §18091(2)(I). Under petitioners’ reading, however, the Act would operate quite differently in a State with a Federal Exchange. As they see it, one of the Act’s three major reforms—the tax credits—would not apply. And a second major reform—the coverage requirement—would not apply in a meaningful way. As explained earlier, the coverage requirement applies only when the cost of buying health insurance (minus the amount of the tax credits) is less than eight percent of an individual’s income. 26 U. S. C. §§5000A(e)(1)(A), (e)(1)(B)(ii). So without the tax credits, the coverage requirement would apply to fewer individuals. And it would be a lot fewer. In 2014, approximately 87 percent of people who bought insurance on a Federal Exchange did so with tax credits, and virtually all of those people would become exempt. HHS, A. Burke, A. Misra, & S. Sheingold, Premium Affordability, Competition, and Choice in the Health Insurance Marketplace 5 (2014); Brief for Bipartisan Economic Scholars as Amici Curiae 19–20. If petitioners are right, therefore, only one of the Act’s three major reforms would apply in States with a Federal Exchange. The combination of no tax credits and an ineffective coverage requirement could well push a State’s individual insurance market into a death spiral. One study predicts that premiums would increase by 47 percent and enrollment would decrease by 70 percent. E. Saltzman & C. Eibner, The Effect of Eliminating the Affordable Care Act’s Tax Credits in Federally Facilitated Marketplaces (2015). Another study predicts that premiums would increase by 35 percent and enrollment would decrease by 69 percent. L. Blumberg, M. Buettgens, & J. Holahan, The Implications of a Supreme Court Finding for the Plaintiff in King vs. Burwell: 8.2 Million More Uninsured and 35% Higher Premiums (2015). And those effects would not be limited to individuals who purchase insurance on the Exchanges. Because the Act requires insurers to treat the entire individual market as a single risk pool, 42 U. S. C. §18032(c)(1), premiums outside the Exchange would rise along with those inside the Exchange. Brief for Bipartisan Economic Scholars as Amici Curiae 11–12. It is implausible that Congress meant the Act to operate in this manner. See National Federation of Independent Business v. Sebelius, 567 U. S. ___, ___ (2012) (Scalia, Kennedy, Thomas, and Alito, JJ., dissenting) (slip op., at 60) (“Without the federal subsidies . . . the exchanges would not operate as Congress intended and may not operate at all.”). Congress made the guaranteed issue and community rating requirements applicable in every State in the Nation. But those requirements only work when combined with the coverage requirement and the tax credits. So it stands to reason that Congress meant for those provisions to apply in every State as well.[4] Petitioners respond that Congress was not worried about the effects of withholding tax credits from States with Federal Exchanges because “Congress evidently believed it was offering states a deal they would not refuse.” Brief for Petitioners 36. Congress may have been wrong about the States’ willingness to establish their own Exchanges, petitioners continue, but that does not allow this Court to rewrite the Act to fix that problem. That is particularly true, petitioners conclude, because the States likely would have created their own Exchanges in the absence of the IRS Rule, which eliminated any incentive that the States had to do so. Id., at 36–38. Section 18041 refutes the argument that Congress believed it was offering the States a deal they would not refuse. That section provides that, if a State elects not to establish an Exchange, the Secretary “shall . . . establish and operate such Exchange within the State.” 42 U. S. C. §18041(c)(1)(A). The whole point of that provision is to create a federal fallback in case a State chooses not to establish its own Exchange. Contrary to petitioners’ argument, Congress did not believe it was offering States a deal they would not refuse—it expressly addressed what would happen if a State did refuse the deal. C Finally, the structure of Section 36B itself suggests that tax credits are not limited to State Exchanges. Section 36B(a) initially provides that tax credits “shall be allowed” for any “applicable taxpayer.” Section 36B(c)(1) then defines an “applicable taxpayer” as someone who (among other things) has a household income between 100 percent and 400 percent of the federal poverty line. Together, these two provisions appear to make anyone in the specified income range eligible to receive a tax credit. According to petitioners, however, those provisions are an empty promise in States with a Federal Exchange. In their view, an applicable taxpayer in such a State would be eligible for a tax credit—but the amount of that tax credit would always be zero. And that is because—diving several layers down into the Tax Code—Section 36B says that the amount of the tax credits shall be “an amount equal to the premium assistance credit amount,” §36B(a); and then says that the term “premium assistance credit amount” means “the sum of the premium assistance amounts determined under paragraph (2) with respect to all coverage months of the taxpayer occurring during the taxable year,” §36B(b)(1); and then says that the term “premium assistance amount” is tied to the amount of the monthly premium for insurance purchased on “an Exchange established by the State under [42 U. S. C. §18031],” §36B(b)(2); and then says that the term “coverage month” means any month in which the taxpayer has insurance through “an Exchange established by the State under [ 42 U. S. C. §18031],” §36B(c)(2)(A)(i). We have held that Congress “does not alter the fundamental details of a regulatory scheme in vague terms or ancillary provisions.” Whitman v. American Trucking Assns., Inc., 531 U. S. 457, 468 (2001) . But in petitioners’ view, Congress made the viability of the entire Affordable Care Act turn on the ultimate ancillary provision: a sub-sub-sub section of the Tax Code. We doubt that is what Congress meant to do. Had Congress meant to limit tax credits to State Exchanges, it likely would have done so in the definition of “applicable taxpayer” or in some other prominent manner. It would not have used such a winding path of connect-the-dots provisions about the amount of the credit.[5] D Petitioners’ arguments about the plain meaning of Section 36B are strong. But while the meaning of the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” may seem plain “when viewed in isolation,” such a reading turns out to be “untenable in light of [the statute] as a whole.” Department of Revenue of Ore. v. ACF Industries, Inc., 510 U. S. 332, 343 (1994) . In this instance, the context and structure of the Act compel us to depart from what would otherwise be the most natural reading of the pertinent statutory phrase. Reliance on context and structure in statutory interpretation is a “subtle business, calling for great wariness lest what professes to be mere rendering becomes creation and attempted interpretation of legislation becomes legislation itself.” Palmer v. Massachusetts, 308 U. S. 79, 83 (1939) . For the reasons we have given, however, such reliance is appropriate in this case, and leads us to conclude that Section 36B allows tax credits for insurance purchased on any Exchange created under the Act. Those credits are necessary for the Federal Exchanges to function like their State Exchange counterparts, and to avoid the type of calamitous result that Congress plainly meant to avoid. * * * In a democracy, the power to make the law rests with those chosen by the people. Our role is more confined—“to say what the law is.” Marbury v. Madison, 1 Cranch 137, 177 (1803). That is easier in some cases than in others. But in every case we must respect the role of the Legislature, and take care not to undo what it has done. A fair reading of legislation demands a fair understanding of the legislative plan. Congress passed the Affordable Care Act to improve health insurance markets, not to destroy them. If at all possible, we must interpret the Act in a way that is consistent with the former, and avoids the latter. Section 36B can fairly be read consistent with what we see as Congress’s plan, and that is the reading we adopt. The judgment of the United States Court of Appeals for the Fourth Circuit is Affirmed.Notes 1 The dissent argues that one would “naturally read instructions about qualified individuals to be inapplicable to the extent a particular Exchange has no such individuals.” Post, at 10–11 (Scalia, J., dissenting). But the fact that the dissent’s interpretation would make so many parts of the Act “inapplicable” to Federal Exchanges is precisely what creates the problem. It would be odd indeed for Congress to write such detailed instructions about customers on a State Exchange, while having nothing to say about those on a Federal Exchange. 2 The dissent argues that the phrase “such Exchange” does not suggest that State and Federal Exchanges “are in all respects equivalent.” Post, at 8. In support, it quotes the Constitution’s Elections Clause, which makes the state legislature primarily responsible for prescribing election regulations, but allows Congress to “make or alter such Regulations.” Art. I, §4, cl. 1. No one would say that state and federal election regulations are in all respects equivalent, the dissent contends, so we should not say that State and Federal Exchanges are. But the Elections Clause does not precisely define what an election regulation must look like, so Congress can prescribe regulations that differ from what the State would prescribe. The Affordable Care Act does precisely define what an Exchange must look like, however, so a Federal Exchange cannot differ from a State Exchange. 3 The dissent notes that several other provisions in the Act use the phrase “established by the State,” and argues that our holding applies to each of those provisions. Post, at 5–6. But “the presumption of consistent usage readily yields to context,” and a statutory term may mean different things in different places. Utility Air Regulatory Group v. EPA, 573 U. S. ___, ___ (2014) (slip op., at 15) (internal quotation marks omitted). That is particularly true when, as here, “the Act is far from a chef d’oeuvre of legislative draftsmanship.” Ibid. Because the other provisions cited by the dissent are not at issue here, we do not address them. 4 The dissent argues that our analysis “show[s] only that the statu-tory scheme contains a flaw,” one “that appeared as well in other parts of the Act.” Post, at 14. For support, the dissent notes that the guaranteed issue and community rating requirements might apply in the federal territories, even though the coverage requirement does not. Id., at 14–15. The confusion arises from the fact that the guaranteed issue and community rating requirements were added as amendments to the Public Health Service Act, which contains a definition of the word “State” that includes the territories, 42 U. S. C. §201(f), while the later-enacted Affordable Care Act contains a definition of the word “State” that excludes the territories, §18024(d). The predicate for the dissent’s point is therefore uncertain at best. 5 The dissent cites several provisions that “make[ ] taxpayers of all States eligible for a credit, only to provide later that the amount of the credit may be zero.” Post, at 11 (citing 26 U. S. C. §§24, 32, 35, 36). None of those provisions, however, is crucial to the viability of a comprehensive program like the Affordable Care Act. No one suggests, for example, that the first-time-homebuyer tax credit, §36, is essential to the viability of federal housing regulation.

576.US.2014_14-6368

While petitioner Kingsley was awaiting trial in county jail, officers forcibly removed him from his cell when he refused to comply with their instructions. Kingsley filed a complaint in Federal District Court claiming, as relevant here, that two of the officers used excessive force against him in violation of the Fourteenth Amendment’s Due Process Clause. At the trial’s conclusion, the District Court instructed the jury that Kingsley was required to prove, inter alia, that the officers “recklessly disregarded [Kingsley’s] safety” and “acted with reckless disregard of [his] rights.” The jury found in the officers’ favor. On appeal, Kingsley argued that the jury instruction did not adhere to the proper standard for judging a pretrial detainee’s excessive force claim, namely, objective unreasonableness. The Seventh Circuit disagreed, holding that the law required a subjective inquiry into the officers’ state of mind, i.e., whether the officers actually intended to violate, or recklessly disregarded, Kingsley’s rights. Held: 1. Under 42 U. S. C. §1983, a pretrial detainee must show only that the force purposely or knowingly used against him was objectively unreasonable to prevail on an excessive force claim. Pp. 5–13. (a) This determination must be made from the perspective of a reasonable officer on the scene, including what the officer knew at the time, see Graham v. Connor, 490 U. S. 386 , and must account for the “legitimate interests [stemming from the government’s] need to manage the facility in which the individual is detained,” appropriately deferring to “policies and practices that in th[e] judgment” of jail officials “are needed to preserve internal order and discipline and to maintain institutional security,” Bell v. Wolfish, 441 U. S. 520, 540, 547. Pp. 5–7. (b) Several considerations lead to this conclusion. An objective standard is consistent with precedent. In Bell, for instance, this Court held that a pretrial detainee could prevail on a claim that his due process rights were violated by providing only objective evidence that the challenged governmental action was not rationally related to a legitimate governmental objective or that it was excessive in relation to that purpose. 441 U. S., at 541–543. Cf. Block v. Rutherford, 468 U. S. 576 –586. Experience also suggests that an objective standard is workable. It is consistent with the pattern jury instructions used in several Circuits, and many facilities train officers to interact with detainees as if the officers’ conduct is subject to objective reasonableness. Finally, the use of an objective standard adequately protects an officer who acts in good faith, e.g., by acknowledging that judging the reasonableness of the force used from the perspective and with the knowledge of the defendant officer is an appropriate part of the analysis. Pp. 7–10. (c) None of the cases respondents point to provides significant support for a subjective standard. Whitley v. Albers, 475 U. S. 312 , and Hudson v. McMillian, 503 U. S. 1 , lack relevance in this context because they involved claims brought by convicted prisoners under the Eighth Amendment’s Cruel and Unusual Punishment Clause, not claims brought by pretrial detainees under the Fourteenth Amendment’s Due Process Clause. And in County of Sacramento v. Lewis, 523 U. S. 833 , a statement indicating the need to show “purpose to cause harm,” id., at 854, for due process liability refers not to whether the force intentionally used was excessive, but whether the defendant intended to commit the acts in question, id., at 854, and n. 13. Finally, in Johnson v. Glick, 481 F. 2d 1028 (CA2), a malicious-and-sadistic-purpose-to-cause-harm factor was not suggested as a necessary condition for liability, but as a factor, among others, that might help show that the use of force was excessive. Pp. 10–13. 2. Applying the proper standard, the jury instruction was erroneous. Taken together, the features of that instruction suggested that the jury should weigh respondents’ subjective reasons for using force and subjective views about the excessiveness of that force. Respondents’ claim that, irrespective of this Court’s holding, any error in the instruction was harmless is left to the Seventh Circuit to resolve on remand. Pp. 13–14. 744 F. 3d 443, vacated and remanded. Breyer, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Scalia, J., filed a dissenting opinion, in which Roberts, C. J., and Thomas, J., joined. Alito, J., filed a dissenting opinion.

In this case, an individual detained in a jail prior to trial brought a claim under Rev. Stat. §1979, 42 U. S. C. §1983, against several jail officers, alleging that they used excessive force against him, in violation of the Fourteenth Amendment’s Due Process Clause. The officers concede that they intended to use the force that they used. Butthe parties disagree about whether the force used was excessive. The question before us is whether, to prove an excessive force claim, a pretrial detainee must show that the officers were subjectively aware that their use of force was unreasonable, or only that the officers’ use of that force was objectively unreasonable. We conclude that the latter standard is the correct one. I A Some but not all of the facts are undisputed: Michael Kingsley, the petitioner, was arrested on a drug charge and detained in a Wisconsin county jail prior to trial. On the evening of May 20, 2010, an officer performing a cell check noticed a piece of paper covering the light fixture above Kingsley’s bed. The officer told Kingsley to remove it; Kingsley refused; subsequently other officers told Kingsley to remove the paper; and each time Kingsley refused. The next morning, the jail administrator, Lieutenant Robert Conroy, ordered Kingsley to remove the paper. Kingsley once again refused. Conroy then told Kingsley that officers would remove the paper and that he would be moved to a receiving cell in the interim. Shortly thereafter, four officers, including respondents Sergeant Stan Hendrickson and Deputy Sheriff Fritz Degner, approached the cell and ordered Kingsley to stand, back up to the door, and keep his hands behind him. When Kingsley refused to comply, the officers handcuffed him, forcibly removed him from the cell, carried him to a receiving cell, and placed him face down on a bunk with his hands handcuffed behind his back. The parties’ views about what happened next differ. The officers testified that Kingsley resisted their efforts to remove his handcuffs. Kingsley testified that he did not resist. All agree that Sergeant Hendrickson placed his knee in Kingsley’s back and Kingsley told him in impolite language to get off. Kingsley testified that Hendrickson and Degner then slammed his head into the concrete bunk—an allegation the officers deny. The parties agree, however, about what happened next: Hendrickson directed Degner to stun Kingsley with a Taser; Degner applied a Taser to Kingsley’s back for approximately five seconds; the officers then left the handcuffed Kingsley alone in the receiving cell; and officers returned to the cell 15 minutes later and removed Kingsley’s handcuffs. B Based on these and related events, Kingsley filed a §1983 complaint in Federal District Court claiming (among other things) that Hendrickson and Degner used excessive force against him, in violation of the Fourteenth Amendment’s Due Process Clause. The officers moved for summary judgment, which the District Court denied, stating that “a reasonable jury could conclude that [the officers] acted with malice and intended to harm [Kingsley] when they used force against him.” Kingsley v. Josvai, No. 10–cv–832–bbc (WD Wis., Nov. 16, 2011), App to Pet. for Cert. 66a–67a. Kingsley’s excessive force claim accordingly proceeded to trial. At the conclusion of the trial, the District Court instructed the jury as follows: “Excessive force means force applied recklessly that is unreasonable in light of the facts and circumstances of the time. Thus, to succeed on his claim of excessive use of force, plaintiff must prove each of the following factors by a preponderance of the evidence: “(1) Defendants used force on plaintiff; “(2) Defendants’ use of force was unreasonable in light of the facts and circumstances at the time; “(3) Defendants knew that using force presented a risk of harm to plaintiff, but they recklessly disregarded plaintiff’s safety by failing to take reasonable measures to minimize the risk of harm to plaintiff; and “(4) Defendants’ conduct caused some harm to plaintiff. “In deciding whether one or more defendants used ‘unreasonable’ force against plaintiff, you must consider whether it was unreasonable from the perspective of a reasonable officer facing the same circumstances that defendants faced. You must make this decision based on what defendants knew at the time of the incident, not based on what you know now. “Also, in deciding whether one or more defendants used unreasonable force and acted with reckless disregard of plaintiff’s rights, you may consider factors such as: “• The need to use force; “• The relationship between the need to use force and the amount of force used; “• The extent of plaintiff’s injury; “• Whether defendants reasonably believed there was a threat to the safety of staff or prisoners; and “• Any efforts made by defendants to limit the amount of force used.” App. 277–278 (emphasis added). The jury found in the officers’ favor. On appeal, Kingsley argued that the correct standard for judging a pretrial detainee’s excessive force claim is objective unreasonableness. And, the jury instruction, he said, did not hew to that standard. A panel of the Court of Appeals disagreed, with one judge dissenting. The major-ity held that the law required a “subjective inquiry” into the officer’s state of mind. There must be “ ‘an actual intent to violate [the plaintiff’s] rights or reckless disregard for his rights.’ ” 744 F. 3d 443, 451 (CA7 2014) (quoting Wilson v. Williams, 83 F. 3d 870, 875 (CA7 1996)). The dissent would have used instructions promulgated by the Committee on Pattern Civil Jury Instructions of the Seventh Circuit, which require a pretrial detainee claiming excessive force to show only that the use of force was objectively unreasonable. 744 F. 3d, at 455 (opinion of Hamilton, J.); see Pattern Civ. Jury Instr. §7.08 (2009). The dissent further stated that the District Court’s use of the word “reckless” in the jury instruction added “an unnecessary and confusing element.” 744 F. 3d, at 455. Kingsley filed a petition for certiorari asking us to determine whether the requirements of a §1983 excessive force claim brought by a pretrial detainee must satisfy the subjective standard or only the objective standard. In light of disagreement among the Circuits, we agreed to do so. Compare, e.g., Murray v. Johnson No. 260, 367 Fed. Appx. 196, 198 (CA2 2010); Bozeman v. Orum, 422 F. 3d 1265, 1271 (CA11 2005) (per curiam), with Aldini v. Johnson, 609 F. 3d 858, 865–866 (CA6 2010); Young v. Wolfe, 478 Fed. Appx. 354, 356 (CA9 2012). II A We consider a legally requisite state of mind. In a case like this one, there are, in a sense, two separate state-of-mind questions. The first concerns the defendant’s state of mind with respect to his physical acts—i.e., his state of mind with respect to the bringing about of certain physical consequences in the world. The second question concerns the defendant’s state of mind with respect to whether his use of force was “excessive.” Here, as to the first question, there is no dispute. As to the second, whether to interpret the defendant’s physical acts in the world as involving force that was “excessive,” there is a dispute. We conclude with respect to that question that the relevant standard is objective not subjective. Thus, the defendant’s state of mind is not a matter that a plaintiff is required to prove. Consider the series of physical events that take place in the world—a series of events that might consist, for example, of the swing of a fist that hits a face, a push that leads to a fall, or the shot of a Taser that leads to the stunning of its recipient. No one here denies, and we must assume, that, as to the series of events that have taken place in the world, the defendant must possess a purposeful, a knowing, or possibly a reckless state of mind. That is because, as we have stated, “liability for negligently inflicted harm is categorically beneath the threshold of constitutional due process.” County of Sacramento v. Lewis, 523 U. S. 833, 849 (1998) (emphasis added). See also Daniels v. Williams, 474 U. S. 327, 331 (1986) (“Historically, this guarantee of due process has been applied to deliberate decisions of government officials to deprive a person of life, liberty, or property”). Thus, if an officer’s Taser goes off by accident or if an officer unintentionally trips and falls on a detainee, causing him harm, the pretrial detainee cannot prevail on an excessive force claim. But if the use of force is delib-erate—i.e., purposeful or knowing—the pretrial detainee’s claim may proceed. In the context of a police pursuit of a suspect the Court noted, though without so holding, that recklessness in some cases might suffice as a standard for imposing liability. See Lewis, supra, at 849. Whether that standard might suffice for liability in the case of an alleged mistreatment of a pretrial detainee need not be decided here; for the officers do not dispute that they acted purposefully or knowingly with respect to the force they used against Kingsley. We now consider the question before us here—the defendant’s state of mind with respect to the proper interpretation of the force (a series of events in the world) that the defendant deliberately (not accidentally or negligently) used. In deciding whether the force deliberately used is, constitutionally speaking, “excessive,” should courts use an objective standard only, or instead a subjective standard that takes into account a defendant’s state of mind? It is with respect to this question that we hold that courts must use an objective standard. In short, we agree with the dissenting appeals court judge, the Seventh Circuit’s jury instruction committee, and Kingsley, that a pretrial detainee must show only that the force purposely or knowingly used against him was objectively unreasonable. A court (judge or jury) cannot apply this standard mechanically. See Lewis, supra, at 850. Rather, objective reasonableness turns on the “facts and circumstances of each particular case.” Graham v. Connor, 490 U. S. 386, 396 (1989) . A court must make this determination from the perspective of a reasonable officer on the scene, including what the officer knew at the time, not with the 20/20 vision of hindsight. See ibid. A court must also account for the “legitimate interests that stem from [the government’s] need to manage the facility in which the individual is detained,” appropriately deferring to “policies and practices that in th[e] judgment” of jail officials “are needed to preserve internal order and discipline and to maintain institutional security.” Bell v. Wolfish, 441 U. S. 520, 540, 547 (1979) . Considerations such as the following may bear on the reasonableness or unreasonableness of the force used: the relationship between the need for the use of force and the amount of force used; the extent of the plaintiff’s injury; any effort made by the officer to temper or to limit the amount of force; the severity of the security problem at issue; the threat reasonably perceived by the officer; and whether the plaintiff was actively resisting. See, e.g., Graham, supra, at 396. We do not consider this list to be exclusive. We mention these factors only to illustrate the types of objective circumstances potentially relevant to a determination of excessive force. B Several considerations have led us to conclude that the appropriate standard for a pretrial detainee’s excessive force claim is solely an objective one. For one thing, it is consistent with our precedent. We have said that “the Due Process Clause protects a pretrial detainee from the use of excessive force that amounts to punishment.” Graham, supra, at 395, n. 10. And in Bell, we explained that such “punishment” can consist of actions taken with an “expressed intent to punish.” 441 U. S., at 538. But the Bell Court went on to explain that, in the absence of an expressed intent to punish, a pretrial detainee can nevertheless prevail by showing that the actions are not “rationally related to a legitimate nonpunitive governmental purpose” or that the actions “appear excessive in relation to that purpose.” Id., at 561. The Bell Court applied this latter objective standard to evaluate a variety of prison conditions, including a prison’s practice of double-bunking. In doing so, it did not consider the prison officials’ subjective beliefs about the policy. Id., at 541–543. Rather, the Court examined objective evidence, such as the size of the rooms and available amenities, before concluding that the conditions were reasonably related to the legitimate purpose of holding detainees for trial and did not appear excessive in relation to that purpose. Ibid. Bell’s focus on “punishment” does not mean that proof of intent (or motive) to punish is required for a pretrial detainee to prevail on a claim that his due process rights were violated. Rather, as Bell itself shows (and as our later precedent affirms), a pretrial detainee can prevail by providing only objective evidence that the challenged governmental action is not rationally related to a legitimate governmental objective or that it is excessive in relation to that purpose. Cf. Block v. Rutherford, 468 U. S. 576 –586 (1984) (where there was no suggestion that the purpose of jail policy of denying contact visitation was to punish inmates, the Court need only evaluate whether the policy was “reasonably related to legitimate governmental objectives” and whether it appears excessive in relation to that objective); Schall v. Martin, 467 U. S. 253 –271 (1984) (similar); see also United States v. Salerno, 481 U. S. 739, 747 (1987) (“[T]he punitive/regulatory distinction turns on ‘whether an alternative purpose to which [the restriction] may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned [to it]’ ” (quoting Schall, supra, at 269; emphasis added and some internal quotation marks omitted)). The Court did not suggest in any of these cases, either by its words or its analysis, that its application of Bell’s objective standard should involve subjective considerations. Our standard is also consistent with our use of an objective “excessive force” standard where officers apply force to a person who, like Kingsley, has been accused but not convicted of a crime, but who, unlike Kingsley, is free on bail. See Graham, supra. For another thing, experience suggests that an objective standard is workable. It is consistent with the pattern jury instructions used in several Circuits. We are also told that many facilities, including the facility at issue here, train officers to interact with all detainees as if the officers’ conduct is subject to an objective reasonableness standard. See Brief for Petitioner 26; App. 247–248; Brief for Former Corrections Administrators and Experts as Amici Curiae 8–18. Finally, the use of an objective standard adequately protects an officer who acts in good faith. We recognize that “[r]unning a prison is an inordinately difficult undertaking,” Turner v. Safley, 482 U. S. 78 –85 (1987), and that “safety and order at these institutions requires the expertise of correctional officials, who must have substantial discretion to devise reasonable solutions to the problems they face,” Florence v. Board of Chosen Freeholders of County of Burlington, 566 U. S. ___, ___ (2012) (slip op., at 5). Officers facing disturbances “are often forced to make split-second judgments—in circumstances that are tense, uncertain, and rapidly evolving.” Graham, 490 U. S., at 397. For these reasons, we have stressed that a court must judge the reasonableness of the force used from the perspective and with the knowledge of the defendant officer. We have also explained that a court must take account of the legitimate interests in managing a jail, acknowledging as part of the objective reasonableness analysis that deference to policies and practices needed to maintain order and institutional security is appropriate. See Part II–A, supra. And we have limited liability for excessive force to situations in which the use of force was the result of an intentional and knowing act (though we leave open the possibility of including a “reckless” act as well). Ibid. Additionally, an officer enjoys qualified immunity and is not liable for excessive force unless he has violated a “clearly established” right, such that “it would [have been] clear to a reasonable officer that his conduct was unlawful in the situation he confronted.” Saucier v. Katz, 533 U. S. 194, 202 (2001) ; see also Brief for United States as Amicus Curiae 27–28. It is unlikely (though theoretically possible) that a plaintiff could overcome these hurdles where an officer acted in good faith. C Respondents believe that the relevant legal standard should be subjective, i.e., that the plaintiff must prove that the use of force was not “applied in a good-faith effort to maintain or restore discipline” but, rather, was applied “maliciously and sadistically to cause harm.” Brief for Respondents 27. And they refer to several cases that they believe support their position. See id., at 26–31 (citing Whitley v. Albers, 475 U. S. 312 (1986) ; Hudson v. McMillian, 503 U. S. 1 (1992) ; Lewis, 523 U. S. 833 ; Johnson v. Glick, 481 F. 2d 1028 (CA2 1973)). The first two of these cases, however, concern excessive force claims brought by convicted prisoners under the Eighth Amendment’s Cruel and Unusual Punishment Clause, not claims brought by pretrial detainees under the Fourteenth Amendment’s Due Process Clause. Whitley, supra, at 320; Hudson, supra, at 6–7. The language of the two Clauses differs, and the nature of the claims often differs. And, most importantly, pretrial detainees (unlike convicted prisoners) cannot be punished at all, much less “maliciously and sadistically.” Ingraham v. Wright, 430 U. S. 651 –672, n. 40 (1977); Graham, supra, at 395, n. 10 (1989); see also 4 W. Blackstone, Commentaries *300 (“[I]f the offence be not bailable, or the party cannot find bail, he is to be committed to the county [jail] . . . [b]ut . . . only for safe custody, and not for punishment”). Thus, there is no need here, as there might be in an Eighth Amendment case, to determine when punishment is unconstitutional. Whitley and Hudson are relevant here only insofar as they address the practical importance of taking into account the legitimate safety-related concerns of those who run jails. And, as explained above, we believe we have done so. Lewis does not prove respondents’ point, either. There, the Court considered a claim that a police officer had violated due process by causing a death during a high-speed automobile chase aimed at apprehending a suspect. We wrote that “[j]ust as a purpose to cause harm is needed for Eighth Amendment liability in a [prison] riot case, so it ought to be needed for due process liability in a pursuit case.” 523 U. S., at 854. Respondents contend that this statement shows that the Court embraced a standard for due process claims that requires a showing of subjective intent. Brief for Respondents 30–31. Other portions of the Lewis opinion make clear, however, that this statement referred to the defendant’s intent to commit the acts in question, not to whether the force intentionally used was “excessive.” 523 U. S., at 854, and n. 13. As explained above, the parties here do not dispute that respondents’ use of force was intentional. See Part II–A, supra. Nor does Glick provide respondents with significant support. In that case Judge Friendly, writing for the Second Circuit, considered an excessive force claim brought by a pretrial detainee under the Fourteenth Amendment’s Due Process Clause. Judge Friendly pointed out that the “management by a few guards of large numbers of prisoners” in an institution “may require and justify the occasional use of a degree of intentional force.” 481 F. 2d, at 1033. He added that, in determining whether that intentional use of force “crosse[s]” the “constitutional line,” a court should look: “to such factors as [(1)] the need for the application of force, [(2)] the relationship between the need and the amount of force that was used, [(3)] the extent of in-jury inflicted, and [(4)] whether force was applied in a good faith effort to maintain or restore discipline or maliciously and sadistically for the very purpose of causing harm.” Ibid. This statement does not suggest that the fourth factor (malicious and sadistic purpose to cause harm) is a necessary condition for liability. To the contrary, the words “such . . . as” make clear that the four factors provide examples of some considerations, among others, that might help show that the use of force was excessive. Respondents believe these cases nonetheless help them make a broader point—namely, that a subjective standard “protects against a relative flood of claims,” many of them perhaps unfounded, brought by pretrial detainees. Brief for Respondents 38. But we note that the Prison Litigation Reform Act of 1995, 42 U. S. C. §1997e, which is designed to deter the filing of frivolous litigation against prison officials, applies to both pretrial detainees and convicted prisoners. Nor is there evidence of a rash of unfounded filings in Circuits that use an objective standard. We acknowledge that our view that an objective standard is appropriate in the context of excessive force claims brought by pretrial detainees pursuant to the Fourteenth Amendment may raise questions about the use of a subjective standard in the context of excessive force claims brought by convicted prisoners. We are not confronted with such a claim, however, so we need not address that issue today. III We now consider the lawfulness of the jury instruction given in this case in light of our adoption of an objective standard for pretrial detainees’ excessive force claims. See Part II–A, supra. That jury instruction defined “excessive force” as “force applied recklessly that is unreasonable in light of the facts and circumstances of the time.” App. 277. It required Kingsley to show that the officers “recklessly disregarded [Kingsley’s] safety.” Id., at 278. And it suggested that Kingsley must show the defendants “acted with reckless disregard of [Kingsley’s] rights,” while telling the jury that it could consider several objective factors in making this determination. Ibid. Kingsley argues that the jury instruction is faulty because the word “reckless” suggests a need to prove that respondents acted with a certain subjective state of mind with respect to the excessive or nonexcessive nature of the force used, contrary to what we have just held. Reply Brief 20–22. Respondents argue that irrespective of our holding, any error in the instruction was harmless. Brief for Respondents 57–58. And the Solicitor General suggests that, because the instructions defined “recklessness” with reference to objective factors, those instructions effectively embody our objective standard and did not confuse the jury. Brief for United States as Amicus Curiae 28–32. We agree with Kingsley that the instructions were erroneous. “[R]eckles[s] disregar[d] [of Kingsley’s] safety” was listed as an additional requirement, beyond the need to find that “[respondents’] use of force was unreasonable in light of the facts and circumstances at the time.” App. 278. See also ibid. (Kingsley had to show respondents “used unreasonable force and acted with reckless disregard of [Kingsley’s] rights” (emphasis added)). And in determining whether respondents “acted with reckless disregard of [Kingsley’s] rights,” the jury was instructed to “consider . . . [w]hether [respondents] reasonably believed there was a threat to the safety of staff or prisoners.” Ibid. (emphasis added). Together, these features suggested the jury should weigh respondents’ subjective reasons for using force and subjective views about the excessiveness of the force. As we have just held, that was error. But because the question whether that error was harmless may depend in part on the detailed specifics of this case, we leave that question for the Court of Appeals to resolve in the first instance. The decision of the Court of Appeals is vacated, and the case is remanded for proceedings consistent with this opinion. It is so ordered.

574.US.427

When petitioner M&G Polymers USA, LLC (M&G), purchased the Point Pleasant Polyester Plant in 2000, it entered a collective-bargaining agreement and related Pension, Insurance, and Service Award Agreement (P & I agreement) with respondent union. As relevant here, the P & I agreement provided that certain retirees, along with their surviving spouses and dependents, would “receive a full Company contribution towards the cost of [health care] benefits”; that such benefits would be provided “for the duration of [the] Agreement”; and that the agreement would be subject to renegotiation in three years. Following the expiration of those agreements, M&G announced that it would require retirees to contribute to the cost of their health care benefits. Respondent retirees, on behalf of themselves and others similarly situated, sued M&G and related entities, alleging that the P & I agreement created a vested right to lifetime contribution-free health care benefits. The District Court dismissed the complaint for failure to state a claim, but the Sixth Circuit reversed based on the reasoning of its earlier decision in International Union, United Auto, Aerospace, & Agricultural Implement Workers of Am. v. Yard-Man, Inc., 716 F.2d 1476. On remand, the District Court ruled in favor of the retirees, and the Sixth Circuit affirmed. Held: The Sixth Circuit’s decision rested on principles that are incompatible with ordinary principles of contract law. Pp. 5–14. (a) The Employee Retirement Income Security Act of 1974 (ERISA) governs pension and welfare benefits plans, including those established by collective-bargaining agreements. ERISA establishes minimum funding and vesting standards for pension plans, but exempts welfare benefits plans—which provide the types of benefits at issue here—from those rules. See 29 U. S. C. §§1051(1), 1053, 1081(a)(2), 1083. “[E]mployers have large leeway to design . . . welfare plans as they see fit.” Black & Decker Disability Plan v. Nord, 538 U.S. 822, 833. Pp. 5–7. (b) This Court interprets collective-bargaining agreements, including those establishing ERISA plans, according to ordinary principles of contract law, at least when those principles are not inconsistent with federal labor policy. See Textile Workers v. Lincoln Mills of Ala., 353 U.S. 448, 456–457. When a collective-bargaining agreement is unambiguous, its meaning must be ascertained in accordance with its plainly expressed intent. 11 R. Lord, Williston on Contracts §30:6, p. 108. P. 7. (c) In Yard-Man, the Sixth Circuit found a provision governing retiree insurance benefits ambiguous as to the duration of those benefits; and, looking to other provisions of the agreement, purported to apply ordinary contract law to resolve the ambiguity. First, the court inferred from the existence of termination provisions for other benefits that the absence of a termination provision specifically addressing retiree benefits expressed an intent to vest those benefits for life. The court then purported to apply the rule that contracts should be interpreted to avoid illusory promises, reasoning that, absent vesting, the promise would be illusory for the subset of retirees who would not become eligible for those benefits before the contract expired. Finally, the court relied on “the context” of labor negotiations to resolve the ambiguity, inferring that the parties would have intended such benefits to vest for life because they are not mandatory subjects of collective bargaining; are “typically understood as a form of delayed compensation,” 716 F. 2d, at 1482; and are keyed to the acquisition of retirement status. The court concluded that these contextual clues “outweigh[ed] any contrary implications derived from a routine duration clause.” Id., at 1483. The Sixth Circuit has since extended its Yard-Man analysis in a series of other cases. Pp. 7–10. (d) The inferences applied in Yard-Man and its progeny do not represent ordinary principles of contract law. Yard-Man distorts the attempt to ascertain the intention of the parties by placing a thumb on the scale in favor of vested retiree benefits in all collective-bargaining agreements. Rather than relying on known customs and usages in a particular industry as proven by the parties, the Yard-Man court relied on its own suppositions about the intentions of parties negotiating retiree benefits. It then compounded the error by applying those suppositions indiscriminately across industries. Furthermore, the Sixth Circuit’s refusal to apply general durational clauses to provisions governing retiree benefits distorts an agreement’s text and conflicts with the principle that a written agreement is presumed to encompass the whole agreement of the parties. Perhaps tugged by its inferences, the Sixth Circuit also misapplied the illusory promises doctrine. It construed provisions that admittedly benefited some class of retirees as “illusory” merely because they did not benefit all retirees. That interpretation is a contradiction in terms—a promise that is “partly illusory” is by definition not illusory. And its use of this doctrine is particularly inappropriate in the context of collective-bargaining agreements, which often include provisions inapplicable to some category of employees. The Sixth Circuit also failed even to consider other traditional contract principles, including the rule that courts should not construe ambiguous writings to create lifetime promises and the rule that “contractual obligations will cease, in the ordinary course, upon termination of the bargaining agreement,” Litton Financial Printing Div., Litton Business Systems, Inc. v. NLRB, 501 U.S. 190, 207. Pp. 10–14. (e) Though there is no doubt that Yard-Man and its progeny affected the outcome here, the Sixth Circuit should be the first to review the agreements under ordinary principles of contract law. P. 14. 733 F.3d 589, vacated and remanded. Thomas, J., delivered the opinion for a unanimous Court. Ginsburg, J., filed a concurring opinion, in which Breyer, Sotomayor, and Kagan, JJ., joined.

This case arises out of a disagreement between a group of retired employees and their former employer about the meaning of certain expired collective-bargaining agreements. The retirees (and their former union) claim that these agreements created a right to lifetime contribution-free health care benefits for retirees, their surviving spouses, and their dependents. The employer, for its part, claims that those provisions terminated when the agreements expired. The United States Court of Appeals for the Sixth Circuit sided with the retirees, relying on its conclusion in International Union, United Auto, Aerospace, & Agricultural Implement Workers of Am. v. Yard-Man, Inc., 716 F. 2d 1476, 1479 (1983), that retiree health care benefits are unlikely to be left up to future negotiations. We granted certiorari and now conclude that such reasoning is incompatible with ordinary principles of contract law. We therefore vacate the judgment of the Court of Appeals and remand for it to apply ordinary principles of contract law in the first instance.IA Respondents Hobert Freel Tackett, Woodrow K. Pyles, and Harlan B. Conley worked at (and retired from) the Point Pleasant Polyester Plant in Apple Grove, West Virginia (hereinafter referred to as the Plant). During their employment, respondent United Steel, Paper and Forestry, Rubber, Manufacturing, Energy, Allied Industrial and Service Workers International Union, AFL-CIO-CLC, or its predecessor unions (hereinafter referred to as the Union), represented them in collective bargaining. Tackett and Pyles retired in 1996, and Conley retired in 1998. They represent a class of retired employees from the Plant, along with their surviving spouses and other dependents. Petitioner M&G Polymers USA, LLC, is the current owner of the Plant. When M&G purchased the Plant in 2000, it entered a master collective-bargaining agreement and a Pension, Insurance, and Service Award Agreement (P & I agreement) with the Union, generally similar to agreements the Union had negotiated with M&G’s predecessor. The P & I agreement provided for retiree health care benefits as follows:“Employees who retire on or after January 1, 1996 and who are eligible for and receiving a monthly pension under the 1993 Pension Plan . . . whose full years of attained age and full years of attained continuous service . . . at the time of retirement equals 95 or more points will receive a full Company contribution towards the cost of [health care] benefits described in this Exhibit B–1 . . . . Employees who have less than 95 points at the time of retirement will receive a reduced Company contribution. The Company contribution will be reduced by 2% for every point less than 95. Employees will be required to pay the balance of the health care contribution, as estimated by the Company annually in advance, for the [health care] benefits described in this Exhibit B–1. Failure to pay the required medical contribution will result in cancellation of coverage.” App. 415–416.Exhibit B–1, which described the health care benefits at issue, opened with the following durational clause: “Effective January 1, 1998, and for the duration of this Agreement thereafter, the Employer will provide the following program of hospital benefits, hospital-medical benefits, surgical benefits and prescription drug benefits for eligible employees and their dependents . . . . ” Id., at 377–378 (emphasis deleted). The P & I agreement provided for renegotiation of its terms in three years.[1]B In December 2006, M&G announced that it would begin requiring retirees to contribute to the cost of their health care benefits. Respondent retirees, on behalf of themselves and others similarly situated, sued M&G and re-lated entities, alleging that the decision to require these contributions breached both the collective-bargaining agreement and the P & I agreement, in violation of §301 of the Labor Management Relations Act, 1947 (LMRA) and §502(a)(1)(B) of the Employee Retirement Income Security Act of 1974 (ERISA),88Stat.891.[2] Specifically, the retirees alleged that M&G had promised to provide lifetime contribution-free health care benefits for them, their surviving spouses, and their dependents. They pointed to the language in the 2000 P & I agreement providing that employees with a certain level of seniority “will receive a full Company contribution towards the cost of [health care] benefits described in . . . Exhibit B–1.” The retirees alleged that, with this promise, M&G had created a vested right to such benefits that continued beyond the expiration of the 2000 P & I agreement. The District Court dismissed the complaint for failure to state a claim. 523 F. Supp. 2d 684, 696 (SD Ohio 2007). It concluded that the cited language unambiguously did not create a vested right to retiree benefits. The Court of Appeals reversed based on the reasoning of its earlier decision in Yard-Man. 561 F. 3d 478 (CA6 2009) (Tackett I). Yard-Man involved a similar claim that an employer had breached a collective-bargaining agreement when it terminated retiree benefits. 716 F. 2d, at 1478. Although the court found the text of the provision in that case ambiguous, it relied on the “context” of labor negotiations to resolve that ambiguity in favor of the retirees’ interpretation. Id., at 1482. Specifically, the court inferred that parties to collective bargaining would intend retiree benefits to vest for life because such benefits are “not mandatory” or required to be included in collective-bargaining agreements, are “typically understood as a form of delayed compensation or reward for past services,” and are keyed to the acquisition of retirement status. Ibid. The court concluded that these inferences “outweigh[ed] any contrary implications [about the termination of retiree benefits] derived from” general termination clauses. Id., at 1483. Applying the Yard-Man inferences on review of the District Court’s dismissal of the action, the Court of Appeals concluded that the retirees had stated a plausible claim. Tackett I, 561 F. 3d, at 490. “Keeping in mind the context of the labor-management negotiations identified in Yard-Man,” the court found “it unlikely that [the Union] would agree to language that ensures its members a ‘full Company contribution,’ if the company could unilaterally change the level of contribution.” Ibid. The court construed the language about “employees” contributing to their health care premiums as limited to employees who had not attained the requisite seniority points to be entitled to a full company contribution. Ibid. And it discerned an intent to vest lifetime contribution-free health care benefits from provisions tying eligibility for health care benefits to eligibility for pension benefits. Id., at 490–491. On remand, the District Court conducted a bench trial and ruled in favor of the retirees. It declined to revisit the question whether the P & I agreement created a vested right to retiree benefits, concluding that the Court of Appeals had definitively resolved that issue. It then issued a permanent injunction ordering M&G to reinstate contribution-free health care benefits for the individual respondents and similarly situated retirees. 853 F. Supp. 2d 697 (SD Ohio 2012). The Court of Appeals affirmed, concluding that, al-though the District Court had erred in treating Tackett I as a conclusive resolution of the meaning of the P & I agreement, it had not erred in “presum[ing]” that, “in the absence of extrinsic evidence to the contrary, the agreements indicated an intent to vest lifetime contribution-free benefits.” 733 F. 3d 589, 600 (CA6 2013) (Tackett II). And because the District Court had concluded that the proffered extrinsic evidence was inapplicable, it had not clearly erred in finding that the agreement created those vested rights. We granted certiorari, 572 U. S. ___ (2014), and now vacate and remand.II This case is about the interpretation of collective-bargaining agreements that define rights to welfare benefits plans. The LMRA grants federal courts jurisdiction to resolve disputes between employers and labor unions about collective-bargaining agreements.29 U. S. C. §185. When collective-bargaining agreements create pension or welfare benefits plans, those plans are subject to rules established in ERISA. ERISA defines pension plans as plans, funds, or programs that “provid[e] retirement income to employees” or that “resul[t] in a deferral of income.” §1002(2)(A). It defines welfare benefits plans as plans, funds, or programs established or maintained to provide participants with additional benefits, such as life insurance and disability coverage. §1002(1). ERISA treats these two types of plans differently. Although ERISA imposes elaborate minimum funding and vesting standards for pension plans, §§1053, 1082, 1083, 1084, it explicitly exempts welfare benefits plans from those rules, §§1051(1), 1081(a)(1). Welfare benefits plans must be “established and maintained pursuant to a written instrument,” §1102(a)(1), but “[e]mployers or other plan sponsors are generally free under ERISA, for any reason at any time, to adopt, modify, or terminate welfare plans,” Curtiss-Wright Corp. v. Schoonejongen,514 U. S. 73,78 (1995). As we have previously recognized, “[E]mployers have large leeway to design disability and other welfare plans as they see fit.” Black & Decker Dis-ability Plan v. Nord,538 U. S. 822,833 (2003). And, we have observed, the rule that contractual “provisions ordinarily should be enforced as written is especially appropriate when enforcing an ERISA [welfare benefits] plan.” Heimeshoff v. Hartford Life & Accident Ins. Co., 571 U. S. ___, ___ (2013) (slip op., at 7). That is because the “focus on the written terms of the plan is the linchpin of a system that is not so complex that administrative costs, or litigation expenses, unduly discourage employers from offering [welfare benefits] plans in the first place.” Id., at ___ (slip op., at 8) (internal quotation marks, brackets, and citation omitted). We interpret collective-bargaining agreements, including those establishing ERISA plans, according to ordinary principles of contract law, at least when those principles are not inconsistent with federal labor policy. See Textile Workers v. Lincoln Mills of Ala.,353 U. S. 448–457 (1957). “In this endeavor, as with any other contract, the parties’ intentions control.” Stolt-Nielsen S. A. v. AnimalFeeds Int’l Corp.,559 U. S. 662,682 (2010) (internal quotation marks omitted). “Where the words of a contract in writing are clear and unambiguous, its meaning is to be ascertained in accordance with its plainly expressed intent.” 11 R. Lord, Williston on Contracts §30:6, p. 108 (4th ed. 2012) (Williston) (internal quotation marks omitted). In this case, the Court of Appeals applied the Yard-Man inferences to conclude that, in the absence of extrinsic evidence to the contrary, the provisions of the contract indicated an intent to vest retirees with lifetime benefits. Tackett II, 733 F. 3d, at 599–600. As we now explain, those inferences conflict with ordinary principles of contract law.IIIA1 The Court of Appeals has long insisted that its Yard-Man inferences are drawn from ordinary contract law. In Yard-Man itself, the court purported to apply “traditional rules for contractual interpretation.” 716 F. 2d, at 1479. The court first concluded that the provision governing retiree insurance benefits—which stated only that the employer “will provide” such benefits—was ambiguous as to the duration of those benefits. Id., at 1480. To resolve that ambiguity, it looked to other provisions of the agreement. The agreement included provisions for terminating active employees’ insurance benefits in the case of layoffs and for terminating benefits for a retiree’s spouse and dependents in case of the retiree’s death before the expiration of the collective-bargaining agreement, but no provision specifically addressed the duration of retiree health care benefits. Id., at 1481–1482. From the existence of these termination provisions and the absence of a termination provision specifically addressing retiree benefits, the court inferred an intent to vest those retiree benefits for life. The court then purported to apply the rule that contracts should be interpreted to avoid illusory promises. It noted that the retiree insurance provisions “contain[ed] a promise that the company will pay an early retiree’s insurance upon such retiree reaching age 65 but that the retiree must bear the cost of company insurance until that time.” Id., at 1481. Employees could retire at age 55, but the agreement containing this promise applied only for a 3-year term. Ibid. Thus, retirees between the ages of 55 and 62 would not turn 65 and become eligible for the company contribution before the 3-year agreement expired. In light of this fact, the court reasoned that the promise would be “completely illusory for many early retirees under age 62” if the retiree benefits terminated when the contract expired. Ibid. Finally, the court turned to “the context” of labor negotiations. Id., at 1482. It observed that “[b]enefits for retirees are . . . not mandatory subjects of collective bargaining” and that “employees are presumably aware that the union owes no obligation to bargain for continued benefits for retirees.” Ibid. Based on these observations, the court concluded that “it is unlikely that such benefits . . . would be left to the contingencies of future negotiations.” Ibid. It also asserted that “retiree benefits are in a sense ‘status’ benefits which, as such, carry with them an inference that they continue so long as the prerequisite status is maintained.” Ibid. Although the contract included a general durational clause—meaning that the contract itself would expire at a set time—the court concluded that these contextual clues “outweigh[ed] any contrary implications derived from a routine duration clause.” Id., at 1483.2 Two years after Yard-Man, the court took this analysis even further. In a dispute between retirees and a steel company over retiree health insurance benefits, it construed the language “will continue to provide at its expense, supplemental medicare and major medical benefits for Pensioners aged 65 and over” to “unambiguously confe[r]” lifetime benefits. Policy v. Powell Pressed Steel Co., 770 F. 2d 609, 615 (CA6 1985) (emphasis added). Yet it had interpreted similar language—“will provide insurance benefits equal to the active group”—to be ambiguous in Yard-Man. The court refused to give any weight to provisions that supported a contrary construction—namely, one establishing a fund to pay pension, but not welfare, benefits, and another providing for the continuation of pension, but not welfare, benefits after the agreement expired. Policy, 770 F. 2d, at 615–616. According to the court, a contrary interpretation “would render the Company’s promise [of benefits for retirees aged 65 and over] in substantial part nugatory and illusory” to retirees who were 62 or younger when the 3-year agreement was signed. Ibid. And it faulted the District Court for failing “to give effect” to Yard-Man’s admonition “that retiree benefits normally . . . are interminable.” 770 F. 2d, at 616. The Court of Appeals has continued to extend the reasoning of Yard-Man. Relying on Yard-Man’s statement that context considerations outweigh the effect of a general termination clause, it has concluded that, “ ‘[a]bsent specific durational language referring to retiree benefits themselves,’ a general durational clause says nothing about the vesting of retiree benefits.” Noe v. PolyOne Corp., 520 F. 3d 548, 555 (CA6 2008) (emphasis added). It has also held that a provision that “ties eligibility for retirement-health benefits to eligibility for a pension . . . [leaves] little room for debate that retirees’ health benefits ves[t] upon retirement.” Id., at 558 (internal quotation marks omitted). Commenting on these extensions of Yard-Man, the court has acknowledged that “there is a reasonable argument to be made that, while th[e] court has repeatedly cautioned that Yard-Man does not create a presumption of vesting, [it] ha[s] gone on to apply just such a presumption.” Cole v. ArvinMeritor, Inc., 549 F. 3d 1064, 1074 (CA6 2008).B We disagree with the Court of Appeals’ assessment that the inferences applied in Yard-Man and its progeny represent ordinary principles of contract law. As an initial matter, Yard-Man violates ordinary contract principles by placing a thumb on the scale in favor of vested retiree benefits in all collective-bargaining agreements. That rule has no basis in ordinary principles of contract law. And it distorts the attempt “to ascertain the intention of the parties.” 11 Williston §30:2, at 18 (emphasis added); see also Stolt-Nielsen, 559 U. S., at 682. Yard-Man’s assessment of likely behavior in collective bargaining is too speculative and too far removed from the context of any particular contract to be useful in discerning the parties’ intention. And the Court of Appeals derived its assessment of likely behavior not from record evidence, but instead from its own suppositions about the intentions of employees, unions, and employers negotiating retiree benefits. See Yard-Man, 716 F. 2d, at 1482. For example, it asserted, without any foundation, that, “when . . . parties contract for benefits which accrue upon achievement of retiree status, there is an inference that the parties likely in-tended those benefits to continue as long as the benefi-ciary remains a retiree.” Ibid.; see also ibid. (“[I]t is unlikely that [retiree] benefits . . . would be left to the contingencies of future negotiations”). Although a court may look to known customs or usages in a particular industry to determine the meaning of a contract, the parties must prove those customs or usages using affirmative evidentiary support in a given case. 12 Williston §34:3; accord, Robinson v. United States, 13 Wall. 363, 366 (1872); Oelricks v. Ford, 23 How. 49, 61–62 (1860). Yard-Man relied on no record evidence indicating that employers and unions in that industry customarily vest retiree benefits. Worse, the Court of Appeals has taken the inferences in Yard-Man and applied them indiscriminately across industries. See, e.g., Cole, supra, at 1074 (automobile); Armistead v. Vernitron Corp., 944 F. 2d 1287, 1297 (CA6 1991) (electronics); Policy, supra, at 618 (steel). Because the Court of Appeals did not ground its Yard-Man inferences in any record evidence, it is unsurprising that the inferences rest on a shaky factual foundation. For example, Yard-Man relied in part on the premise that retiree health care benefits are not subjects of mandatory collective bargaining. Parties, however, can and do voluntarily agree to make retiree benefits a subject of mandatory collective bargaining. Indeed, the employer and union in this case entered such an agreement in 2001. App. 435–436. Yard-Man also relied on the premise that re-tiree benefits are a form of deferred compensation, but that characterization is contrary to Congress’ determi-nation otherwise. In ERISA, Congress specifically defined plans that “resul[t] in a deferral of income by employ-ees” as pension plans, §1002(2)(A)(ii), and plans that offer medical benefits as welfare plans, §1002(1)(A). Thus, retiree health care benefits are not a form of deferred compensation. Further compounding this error, the Court of Appeals has refused to apply general durational clauses to provisions governing retiree benefits. Having inferred that parties would not leave retiree benefits to the contingencies of future negotiations, and that retiree benefits generally last as long as the recipient remains a retiree, the court in Yard-Man explicitly concluded that these inferences “outweigh[ed] any contrary implications derived from a routine duration clause terminating the agreement generally.” 716 F. 2d, at 1482–1483. The court’s subsequent decisions went even further, requiring a contract to include a specific durational clause for retiree health care benefits to prevent vesting. E.g., Noe, supra, at 555. These decisions distort the text of the agreement and conflict with the principle of contract law that the written agreement is presumed to encompass the whole agreement of the parties. See 1 W. Story, Law of Contracts §780 (M. Bigelow ed., 5th ed. 1874); see also 11 Williston §31:5. Perhaps tugged by these inferences, the Court of Appeals misapplied other traditional principles of contract law, including the illusory promises doctrine. That doctrine instructs courts to avoid constructions of contracts that would render promises illusory because such promises cannot serve as consideration for a contract. See 3Williston §7:7 (4th ed. 2008). But the Court of Appeals construed provisions that admittedly benefited some class of retirees as “illusory” merely because they did not equally benefit all retirees. See Yard-Man, supra, at 1480–1481. That interpretation is a contradiction in terms—a promise that is “partly” illusory is by definition not illusory. If it benefits some class of retirees, then it may serveas consideration for the union’s promises. And the court’s interpretation is particularly inappropriate in the context of collective-bargaining agreements, which are negotiated on behalf of a broad category of individuals and consequently will often include provisions inapplicable to some category of employees. The Court of Appeals also failed even to consider the traditional principle that courts should not construe ambiguous writings to create lifetime promises. See 3 A. Corbin, Corbin on Contracts §553, p. 216 (1960) (explaining that contracts that are silent as to their duration will ordinarily be treated not as “operative in perpetuity” but as “operative for a reasonable time” (internal quotation marks omitted)). The court recognized that “traditional rules of contractual interpretation require a clear manifestation of intent before conferring a benefit or obligation,” but asserted that “the duration of the benefit once clearly conferred is [not] subject to this stricture.” Yard-Man, supra, at 1481, n. 2. In stark contrast to this assertion, however, the court later applied that very stricture to noncollectively bargained contracts offering retiree benefits. See Sprague v. General Motors Corp., 133 F. 3d 388, 400 (CA6 1998) (“To vest benefits is to render them forever unalterable. Because vesting of welfare plan benefits is not required by law, an employer’s commitment to vest such benefits is not to be inferred lightly; the intent to vest must be found in the plan documents and must be stated in clear and express language” (internal quotation marks omitted)). The different treatment of these two types of employment contracts only underscores Yard-Man’s deviation from ordinary principles of contract law. Similarly, the Court of Appeals failed to consider the traditional principle that “contractual obligations will cease, in the ordinary course, upon termination of the bargaining agreement.” Litton Financial Printing Div., Litton Business Systems, Inc. v. NLRB,501 U. S. 190,207 (1991). That principle does not preclude the conclusion that the parties intended to vest lifetime benefits for retirees. Indeed, we have already recognized that “a collective-bargaining agreement [may] provid[e] in explicit terms that certain benefits continue after the agreement’s expiration.” Ibid. But when a contract is silent as to the duration of retiree benefits, a court may not infer that the parties intended those benefits to vest for life.C There is no doubt that Yard-Man and its progeny af-fected the outcome here. As in its previous decisions, the Court of Appeals here cited the “context of . . . labor-management negotiations” and reasoned that the Union likely would not have agreed to language ensuring its members a “full Company contribution” if the company could change the level of that contribution. Tackett I, 561 F. 3d, at 490 (internal quotation marks omitted). It similarly concluded that the tying of eligibility for health care benefits to receipt of pension benefits suggested an intent to vest health care benefits. Ibid. And it framed its analysis from beginning to end in light of the principles it announced in Yard-Man and its progeny. See 561 F. 3d, at 489; see also Tackett II, 733 F. 3d, at 599–600. We reject the Yard-Man inferences as inconsistent with ordinary principles of contract law. But because “[t]his Court is one of final review, not of first view,” Ford Motor Co. v. United States, 571 U. S. ___, ___ (2013) (per curiam) (slip op., at 2) (internal quotation marks omitted), the Court of Appeals should be the first to review the agreements at issue under the correct legal principles. We vacate the judgment of the Court of Appeals and remand the case for that court to apply ordinary principles of contract law in the first instance.It is so ordered.Notes1 In accordance with this provision, M&G and the Union began bargaining anew in 2003, ultimately reaching a new agreement in 2005. The provisions of the existing agreements remained in effect during the course of those negotiations. See App. to Pet. for Cert. 25, n. 1.2 The Union was a plaintiff in the suit and is a respondent here. For ease of reference, we refer to the respondents collectively as “the retirees.”

575.US.2014_13-1019

Before suing an employer for employment discrimination under Title VII of the Civil Rights Act of 1964, the Equal Employment Opportunity Commission (EEOC or Commission) must first “endeavor to eliminate [the] alleged unlawful employment practice by informal methods of conference, conciliation, and persuasion.” 42 U. S. C. §2000e–5(b). Once the Commission determines that conciliation has failed, it may file suit in federal court. §2000e–5(f)(1). However, “[n]othing said or done during” conciliation may be “used as evidence in a subsequent proceeding without written consent of the persons concerned.” §2000e–5(b). After investigating a sex discrimination charge against petitioner Mach Mining, LLC, respondent EEOC determined that reasonable cause existed to believe that the company had engaged in unlawful hiring practices. The Commission sent a letter inviting Mach Mining and the complainant to participate in informal conciliation proceedings and notifying them that a representative would be contacting them to begin the process. About a year later, the Commission sent Mach Mining another letter stating that it had determined that conciliation efforts had been unsuccessful. The Commission then sued Mach Mining in federal court. In its answer, Mach Mining alleged that the Commission had not attempted to conciliate in good faith. The Commission countered that its conciliation efforts were not subject to judicial review and that, regardless, the two letters it sent to Mach Mining provided adequate proof that it had fulfilled its statutory duty. The District Court agreed that it could review the adequacy of the Commission’s efforts, but granted the Commission leave to immediately appeal. The Seventh Circuit reversed, holding that the Commission’s statutory conciliation obligation was unreviewable. Held: 1. Courts have authority to review whether the EEOC has fulfilled its Title VII duty to attempt conciliation. This Court has recognized a “strong presumption” that Congress means to allow judicial review of administrative action. Bowen v. Michigan Academy of Family Physicians, 476 U. S. 667 . That presumption is rebuttable when a statute’s language or structure demonstrates that Congress intended an agency to police itself. Block v. Community Nutrition Institute, 467 U. S. 340 . But nothing rebuts that presumption here. By its choice of language, Congress imposed a mandatory duty on the EEOC to attempt conciliation and made that duty a precondition to filing a lawsuit. Such compulsory prerequisites are routinely enforced by courts in Title VII litigation. And though Congress gave the EEOC wide latitude to choose which “informal methods” to use, it did not deprive courts of judicially manageable criteria by which to review the conciliation process. By its terms, the statutory obligation to attempt conciliation necessarily entails communication between the parties concerning the alleged unlawful employment practice. The statute therefore requires the EEOC to notify the employer of the claim and give the employer an opportunity to discuss the matter. In enforcing that statutory condition, a court applies a manageable standard. Pp. 4–8. 2. The appropriate scope of judicial review of the EEOC’s conciliation activities is narrow, enforcing only the EEOC’s statutory obligation to give the employer notice and an opportunity to achieve voluntary compliance. This limited review respects the expansive discretion that Title VII gives the EEOC while still ensuring that it follows the law. The Government’s suggestion that review be limited to checking the facial validity of its two letters to Mach Mining falls short of Title VII’s demands. That standard would merely accept the EEOC’s word that it followed the law, whereas the aim of judicial review is to verify that the EEOC actually tried to conciliate a discrimination charge. Citing the standard set out in the National Labor Relations Act, Mach Mining proposes review for whether the EEOC engaged in good-faith negotiation, laying out a number of specific requirements to implement that standard. But the NLRA’s process-based approach provides a poor analogy for Title VII, which ultimately cares about substantive outcomes and eschews any reciprocal duty to negotiate in good faith. Mach Mining’s proposed code of conduct also conflicts with the wide latitude Congress gave the Commission to decide how to conduct and when to end conciliation efforts. And because information obtained during conciliation would be necessary evidence in a good-faith determination proceeding, Mach Mining’s brand of review would violate Title VII’s confidentiality protections. The proper scope of review thus matches the terms of Title VII’s conciliation provision. In order to comply with that provision, the EEOC must inform the employer about the specific discrimination allegation. Such notice must describe what the employer has done and which employees (or class of employees) have suffered. And the EEOC must try to engage the employer in a discussion in order to give the employer a chance to remedy the allegedly discriminatory practice. A sworn affidavit from the EEOC stating that it has performed these obligations should suffice to show that it has met the conciliation requirement. Should the employer present concrete evidence that the EEOC did not provide the requisite information about the charge or attempt to engage in a discussion about conciliating the claim, a court must conduct the factfinding necessary to resolve that limited dispute. Should it find for the employer, the appropriate remedy is to order the EEOC to undertake the mandated conciliation efforts. Pp. 8–14. 738 F. 3d 171, vacated and remanded. Kagan, J., delivered the opinion for a unanimous Court.

Before suing an employer for discrimination, the Equal Employment Opportunity Commission (EEOC or Commission) must try to remedy unlawful workplace practices through informal methods of conciliation. This case requires us to decide whether and how courts may review those efforts. We hold that a court may review whether the EEOC satisfied its statutory obligation to attempt conciliation before filing suit. But we find that the scope of that review is narrow, thus recognizing the EEOC’s extensive discretion to determine the kind and amount of communication with an employer appropriate in any given case.I Title VII of the Civil Rights Act of 1964,78Stat.241,42 U. S. C. §2000e et seq., sets out a detailed, multi-step procedure through which the Commission enforces the statute’s prohibition on employment discrimination. The process generally starts when “a person claiming to be aggrieved” files a charge of an unlawful workplace practice with the EEOC. §2000e–5(b). At that point, the EEOC notifies the employer of the complaint and undertakes an investigation. See ibid. If the Commission finds no “reasonable cause” to think that the allegation has merit, it dismisses the charge and notifies the parties. Ibid. The complainant may then pursue her own lawsuit if she chooses. See §2000e–5(f )(1). If, on the other hand, the Commission finds reasonable cause, it must first “endeavor to eliminate [the] alleged unlawful employment practice by informal methods of conference, conciliation, and persuasion.” §2000e–5(b). To ensure candor in those discussions, the statute limits the disclosure and use of the participants’ statements: “Nothing said or done during and as a part of such informal endeavors” may be publicized by the Commission or “used as evidence in a subsequent proceeding without the written consent of the persons concerned.” Ibid. The statute leaves to the EEOC the ultimate decision whether to accept a settlement or instead to bring a lawsuit. So long as “the Commission has been unable to secure from the respondent a conciliation agreement acceptable to the Commission” itself, the EEOC may sue the employer. §2000e–5(f )(1). This case began when a woman filed a charge with the EEOC claiming that petitioner Mach Mining, LLC, had refused to hire her as a coal miner because of her sex. The Commission investigated the allegation and found reasonable cause to believe that Mach Mining had discriminated against the complainant, along with a class of women who had similarly applied for mining jobs. See App. 15. In a letter announcing that determination, the EEOC invited both the company and the complainant to participate in “informal methods” of dispute resolution, promising that a Commission representative would soon “contact [them] to begin the conciliation process.” Id., at 16. The record does not disclose what happened next. But about a year later, the Commission sent Mach Mining a second letter, stating that “such conciliation efforts as are required by law have occurred and have been unsuccessful” and that any further efforts would be “futile.” Id., at 18–19. The EEOC then sued Mach Mining in federal district court alleging sex discrimination in hiring. The Commission’s complaint maintained that “[a]ll conditions precedent to the institution of this lawsuit”—including an attempt to end the challenged practice through conciliation—“ha[d] been fulfilled.” Id., at 22. In its answer, Mach Mining contested that statement, asserting that the EEOC had failed to “conciliat[e] in good faith” prior to filing suit. Id., at 30. The Commission subsequently moved for summary judgment on that issue, contending that its “conciliation efforts are not subject to judicial review.” Motion for Summary Judgment in No. 3:11–cv–00879 (SD Ill.), p. 1. At most, the Commission argued, the court could inspect the EEOC’s two letters to Mach Mining to confirm that the EEOC had met its duty to attempt conciliation. See id., at 11, 19. Mach Mining responded by urging the court to consider the overall “reasonable[ness]” of the EEOC’s efforts, based on evidence the company would present about the conciliation process. Memorandum in Opposition to Motion for Partial Summary Judgment in No. 3:11–cv–00879 (SD Ill.), p. 20. The trial court agreed with Mach Mining that it should review whether the Commission had made “a sincere and reasonable effort to negotiate.” Civ. No. 11–879 (SD Ill., Jan. 28, 2013), App. to Pet. for Cert. 40a, 2013 WL 319337, *5 (internal quotation marks omitted). At the EEOC’s request, the court then authorized an immediate appeal of its ruling. See Civ. No. 11–879 (SD Ill., May 20, 2013), App. to Pet. for Cert. 52a–55a, 2013 WL 2177770, *5–*6;28 U. S. C. §1292(b). The Court of Appeals for the Seventh Circuit reversed, holding that “the statutory directive to attempt conciliation” is “not subject to judicial review.” 738 F. 3d 171, 177 (2013). According to the court, that provision entrusts conciliation “solely to the EEOC’s expert judgment” and thus provides no “workable standard” of review for courts to apply. Id., at 174, 177. The Seventh Circuit further reasoned that judicial review of the conciliation process would “undermine enforcement of Title VII” by “protract[ing] and complicat[ing]” discrimination suits. Id., at 178–179 (quoting Doe v. Oberweis Diary, 456 F. 3d 704, 710 (CA7 2006)). In its concluding paragraph, however, the court indicated that it had in fact subjected the EEOC’s activities to a smidgen of review: Because the Commission “pled on the face of its complaint that it ha[d] complied with all” prerequisites to suit and because its two letters to Mach Mining were “facially sufficient” to show that conciliation had occurred, the court stated, “our review of [that process] is satisfied.” 738 F. 3d, at 184. Other Courts of Appeals have held that Title VII allows judicial review of the EEOC’s conciliation efforts, but without agreeing on what that review entails.[1] We granted certiorari, 573 U. S. ___ (2014), to address whether andto what extent such an attempt to conciliate is subject to judicial consideration.II Congress rarely intends to prevent courts from enforcing its directives to federal agencies. For that reason, this Court applies a “strong presumption” favoring judicial review of administrative action. Bowen v. Michigan Academy of Family Physicians,476 U. S. 667,670 (1986). That presumption is rebuttable: It fails when a statute’s lan-guage or structure demonstrates that Congress wanted an agency to police its own conduct. See Block v. Community Nutrition Institute,467 U. S. 340,349,351 (1984). But the agency bears a “heavy burden” in attempting to show that Congress “prohibit[ed] all judicial review” of the agency’s compliance with a legislative mandate. Dunlop v. Bachowski,421 U. S. 560,567 (1975). Title VII, as the Government acknowledges, imposes a duty on the EEOC to attempt conciliation of a discrimination charge prior to filing a lawsuit. See Brief for Respondent 20. That obligation is a key component of the statutory scheme. In pursuing the goal of “bring[ing] employment discrimination to an end,” Congress chose “[c]ooperation and voluntary compliance” as its “preferred means.” Ford Motor Co. v. EEOC,458 U. S. 219,228 (1982) (quoting Alexander v. Gardner-Denver Co.,415 U. S. 36,44 (1974)). Accordingly, the statute provides, as earlier noted, that the Commission “shall endeavor to eliminate [an] alleged unlawful employment practice by informal methods of conference, conciliation, and persuasion.” §2000e–5(b); see supra, at 2. That language is mandatory, not precatory. Cf. National Railroad Passenger Corporation v. Morgan,536 U. S. 101,109 (2002) (noting that the word “shall” admits of no discretion). And the duty it imposes serves as a necessary precondition to filing a lawsuit. Only if the Commission is “unable to secure” an acceptable conciliation agreement—that is, only if its attempt to conciliate has failed—may a claim against the employer go forward. §2000e–5(f )(1). Courts routinely enforce such compulsory prerequisites to suit in Title VII litigation (and in many other contexts besides). An employee, for example, may bring a Title VII claim only if she has first filed a timely charge with the EEOC—and a court will usually dismiss a complaint for failure to do so. See, e.g., id., at 104–105, 114–115. Similarly, an employee must obtain a right-to-sue letter before bringing suit—and a court will typically insist on satisfaction of that condition. See, e.g., McDonnell Douglas Corp. v. Green,411 U. S. 792,798 (1973); see also, e.g., Hallstrom v. Tillamook County,493 U. S. 20,26 (1989) (upholding dismissal of an environmental suit for failure to comply with a notice provision serving as a “condition precedent”); United States v. Zucca,351 U. S. 91 (1956) (affirming dismissal of a denaturalization suit because of the Government’s failure to comply with a mandatory prerequisite). That ordinary part of Title VII litigation—see a prerequisite to suit, enforce a prerequisite to suit—supports judicial review of the EEOC’s compliance with the law’s conciliation provision. The Government, reiterating the Seventh Circuit’s view, contests that conclusion, arguing that Title VII provides “no standards by which to judge” the EEOC’s performance of its statutory duty. Brief for Respondent 17. The Government highlights the broad leeway the statute gives the EEOC to decide how to engage in, and when to give up on, conciliation. In granting that discretion, the Government contends, Congress deprived courts of any “judicially manageable” criteria with which to review the EEOC’s efforts. Id., at 36 (quoting Heckler v. Chaney,470 U. S. 821,830 (1985)). And in that way Congress “demonstrate[d] [its] intention to preclude judicial review.” Brief for Respondent 39. But in thus denying that Title VII creates a “reviewable prerequisite to suit,” the Government takes its observation about discretion too far. Id., at 37 (quoting 738 F. 3d, at 175). Yes, the statute provides the EEOC with wide latitude over the conciliation process, and that feature becomes significant when we turn to defining the proper scope of judicial review. See infra, at 10–11. But no, Congress has not left everything to the Commission. Consider if the EEOC declined to make any attempt to conciliate a claim—if, after finding reasonable cause to support a charge, the EEOC took the employer straight to court. In such a case, Title VII would offer a perfectly serviceable standard for judicial review: Without any “endeavor” at all, the EEOC would have failed to satisfy a necessary condition of litigation. Still more, the statute provides certain concrete standards pertaining to what that endeavor must entail. Again, think of how the statute describes the obligatory attempt: “to eliminate [the] alleged unlawful employment practice by informal methods of conference, conciliation, and persuasion.” §2000e–5(b). Those specified methods necessarily involve communication between parties, including the exchange of information and views. As one dictionary variously defines the terms, they involve “consultation or discussion,” an attempt to “reconcile” different positions, and a “means of argument, reasoning, or entreaty.” American Heritage Dictionary 385, 382, 1318 (5th ed. 2011). That communication, moreover, concerns a particular thing: the “alleged unlawful employment practice.” So the EEOC, to meet the statutory condition, must tell the employer about the claim—essentially, what practice has harmed which person or class—and must provide the employer with an opportunity to discuss the matter in an effort to achieve voluntary compliance. See also infra, at 13. If the Commission does not take those specified actions, it has not satisfied Title VII’s requirement to attempt conciliation. And in insisting that the Commission do so, as the statutory language directs, a court applies a manageable standard. Absent such review, the Commission’s compliance with the law would rest in the Commission’s hands alone. We need not doubt the EEOC’s trustworthiness, or its fidelity to law, to shy away from that result. We need only know—and know that Congress knows—that legal lapses and violations occur, and especially so when they have no consequence. That is why this Court has so long applied a strong presumption favoring judicial review of administrative action. See supra, at 4–5. Nothing overcomes that presumption with respect to the EEOC’s duty to attempt conciliation of employment discrimination claims.III That conclusion raises a second dispute between the parties: What is the proper scope of judicial review of the EEOC’s conciliation activities? The Government (once having accepted the necessity for some review) proposes that courts rely solely on facial examination of certain EEOC documents. Mach Mining argues for far more intrusive review, in part analogizing to the way judges superintend bargaining between employers and unions. We accept neither suggestion, because we think neither consistent with the choices Congress made in enacting Title VII. The appropriate scope of review enforces the statute’s requirements as just described—in brief, that the EEOC afford the employer a chance to discuss and rectify a specified discriminatory practice—but goes no further. See supra, at 7; infra, at 13. Such limited review respects the expansive discretion that Title VII gives to the EEOC over the conciliation process, while still ensuring that the Commission follows the law. The Government argues for the most minimalist form of review imaginable. Echoing the final paragraph of the decision below, the Government observes that the EEOC, in line with its standard practice, wrote two letters to Mach Mining. See supra, at 2–3, 4. The first, after announcing the Commission’s finding of reasonable cause, informed the company that “[a] representative of this office will be in contact with each party in the near future to begin the conciliation process.” App. 16. The second, sent about a year later, stated that the legally mandated conciliation attempt had “occurred” and failed. Id., at 18. According to the Government, those “bookend” letters are all a court ever needs for review, because they “establish” that the EEOC met its obligation to attempt conciliation. Brief for Respondent 21. But review of that kind falls short of what Title VII demands because the EEOC’s bookend letters fail to prove what the Government claims. Contrary to its intimation, those letters do not themselves fulfill the conciliation condition: The first declares only that the process will start soon, and the second only that it has concluded. The two letters, to be sure, may provide indirect evidence that conciliation efforts happened in the interim; the later one expressly represents as much. But suppose an employer contests that statement. Let us say the employer files an affidavit alleging that although the EEOC promised to make contact, it in fact did not. In that circumstance, to treat the letters as sufficient—to take them at face value, as the Government wants—is simply to accept the EEOC’s say-so that it complied with the law. And as earlier explained, the point of judicial review is instead to verify the EEOC’s say-so—that is, to determine that the EEOC actually, and not just purportedly, tried to conciliate a discrimination charge. See supra, at 7–8. For that, a court needs more than the two bookend letters the Government proffers. Mach Mining, for its part, would have a court do a deep dive into the conciliation process. Citing the standard set out in the National Labor Relations Act (NLRA), Mach Mining wants a court to consider whether the EEOC has “negotiate[d] in good faith” over a discrimination claim. Brief for Petitioner 37; see29 U. S. C. §158(d) (imposing a duty on employers and unions to bargain “in good faith with respect to . . . terms and conditions of employment”). That good-faith obligation, Mach Mining maintains, here incorporates a number of specific requirements. In every case, the EEOC must let the employer know the “minimum . . . it would take to resolve” the claim—that is, the smallest remedial award the EEOC would accept. Tr. of Oral Arg. 63. The Commission must also lay out “the factual and legal basis for” all its positions, including the calculations underlying any monetary request. Brief for Petitioner 39. And the Commission must refrain from making “take-it-or-leave-it” offers; rather, the EEOC has to go back and forth with the employer, considering and addressing its various counter-offers and giving it sufficient time at each turn “to review and respond.” Id., at 40. The function of judicial review, Mach Mining concludes, is to compel the Commission to abide by these rules. To begin, however, we reject any analogy between the NLRA and Title VII. The NLRA is about process and process alone. It creates a sphere of bargaining—in which both sides have a mutual obligation to deal fairly—without expressing any preference as to the substantive agreements the parties should reach. See §§151, 158(d). By contrast, Title VII ultimately cares about substantive results, while eschewing any reciprocal duties of good-faith negotiation. Its conciliation provision explicitly serves a substantive mission: to “eliminate” unlawful discrimination from the workplace.42 U. S. C. §2000e–5(b). In discussing a claim with an employer, the EEOC must always insist upon legal compliance; and the employer, for its part, has no duty at all to confer or exchange proposals, but only to refrain from any discrimination. Those differences make judicial review of the NLRA’s duty of good-faith bargaining a poor model for review of Title VII’s conciliation requirement. In addressing labor disputes, courts have devised a detailed body of rules to police good-faith dealing divorced from outcomes—and so to protect the NLRA’s core procedural apparatus. But those kinds of rules do not properly apply to a law that treats the conciliation process not as an end in itself, but only as a tool to redress workplace discrimination. More concretely, Mach Mining’s proposed code of con-duct conflicts with the latitude Title VII gives the Commission to pursue voluntary compliance with the law’s commands. Every aspect of Title VII’s conciliation provision smacks of flexibility. To begin with, the EEOC need only “endeavor” to conciliate a claim, without having to devote a set amount of time or resources to that project. §2000e–5(b). Further, the attempt need not involve any specific steps or measures; rather, the Commission may use in each case whatever “informal” means of “conference, conciliation, and persuasion” it deems appropriate. Ibid. And the EEOC alone decides whether in the end to make an agreement or resort to litigation: The Commission may sue whenever “unable to secure” terms “accept-able to the Commission.” §2000e–5(f )(1) (emphasis added). All that leeway respecting how to seek voluntary compliance and when to quit the effort is at odds with Mach Mining’s bargaining checklist. Congress left to the EEOC such strategic decisions as whether to make a bare-minimum offer, to lay all its cards on the table, or to respond to each of an employer’s counter-offers, however far afield. So too Congress granted the EEOC discretion over the pace and duration of conciliation efforts, the plasticity or firmness of its negotiating positions, and the content of its demands for relief. For a court to assess any of those choices—as Mach Mining urges and many courts have done, see n. 1, supra—is not to enforce the law Congress wrote, but to impose extra procedural requirements. Such judicial review extends too far. Mach Mining’s brand of review would also flout Title VII’s protection of the confidentiality of conciliation efforts. The statute, recall, provides that “[n]othing said or done during and as a part of such informal endeavors may be made public by the Commission . . . or used as evidence in a subsequent proceeding without the written consent of the persons concerned”—both the employer and the complainant. §2000e–5(b); see EEOC v. Associated Dry Goods Corp.,449 U. S. 590, and n. 13 (1981). But the judicial inquiry Mach Mining proposes would necessitate the disclosure and use of such information in a later Title VII suit: How else could a court address an allegation that the EEOC failed to comply with all the negotiating rules Mach Mining espouses?[2] The proof is in this very case: The District Court held that it could not strike from the record descriptions of the conciliation process because they spoke to whether the EEOC had made a “sincere and reasonable effort to negotiate.” App. to Pet. for Cert. 40a (internal quotation marks omitted); see supra, at 3. The court thus failed to give effect to the law’s non-disclosure provision. And in so doing, the court undermined the conciliation process itself, because confidentiality promotes candor in discussions and thereby enhances the prospects for agreement. As this Court has explained, “[t]he maximum results from the voluntary approach will be achieved if ” the parties know that statements they make cannot come back to haunt them in litigation. Associated Dry Goods Corp., 449 U. S., at 599, n. 16 (quoting 110 Cong. Rec. 8193 (1964) (remarks of Sen. Dirksen)). And conversely, the minimum results will be achieved if a party can hope to use accounts of those discussions to derail or delay a meritorious claim. By contrast with these flawed proposals, the proper scope of judicial review matches the terms of Title VII’s conciliation provision, as we earlier described them. See supra, at 7. The statute demands, once again, that the EEOC communicate in some way (through “conference, conciliation, and persuasion”) about an “alleged unlawful employment practice” in an “endeavor” to achieve an employer’s voluntary compliance. §2000e–5(b). That means the EEOC must inform the employer about the specific allegation, as the Commission typically does in a letter announcing its determination of “reasonable cause.” Ibid. Such notice properly describes both what the employer has done and which employees (or what class of employees) have suffered as a result. And the EEOC must try to engage the employer in some form of discussion (whether written or oral), so as to give the employer an opportunity to remedy the allegedly discriminatory practice. Judicial review of those requirements (and nothing else) ensures that the Commission complies with the statute. At the same time, that relatively barebones review allows the EEOC to exercise all the expansive discretion Title VII gives it to decide how to conduct conciliation efforts and when to end them. And such review can occur consistent with the statute’s non-disclosure provision, because a court looks only to whether the EEOC attempted to confer about a charge, and not to what happened (i.e., statements made or positions taken) during those discussions. A sworn affidavit from the EEOC stating that it has performed the obligations noted above but that its efforts have failed will usually suffice to show that it has met the conciliation requirement. Cf. United States v. Clarke, 573 U. S. ___, ___ (2014) (slip op., at 6) (“[A]bsent contrary evidence, the [agency] can satisfy [the relevant] standard by submitting a simple affidavit from” the agency representative involved). If, however, the employer provides credible evidence of its own, in the form of an affidavit or otherwise, indicating that the EEOC did not provide the requisite information about the charge or attempt to engage in a discussion about conciliating the claim, a court must conduct the factfinding necessary to decide that limited dispute. Cf. id., at ___–___ (slip op., at 6–7). Should the court find in favor of the employer, the appropriate remedy is to order the EEOC to undertake the mandated efforts to obtain voluntary compliance. See §2000e–5(f )(1) (authorizing a stay of a Title VII action for that purpose).IV Judicial review of administrative action is the norm in our legal system, and nothing in Title VII withdraws the courts’ authority to determine whether the EEOC has fulfilled its duty to attempt conciliation of claims. But the scope of that review is narrow, reflecting the abundant discretion the law gives the EEOC to decide the kind and extent of discussions appropriate in a given case. In addressing a claim like Mach Mining’s, courts may not impinge on that latitude and on the Commission’s concomitant responsibility to eliminate unlawful workplace discrimination. For the reasons stated, we vacate the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion.It is so ordered.Notes1 See, e.g., EEOC v. Asplundh Tree Expert Co., 340 F. 3d 1256, 1259 (CA11 2003) (holding that the EEOC must, among other things, “respond in a reasonable and flexible manner to the reasonable attitudes of the employer”); EEOC v. Keco Industries, Inc., 748 F. 2d 1097, 1102 (CA6 1984) (holding that the EEOC must “make a good faith effort to conciliate”).2 Mach Mining tries to show that broad judicial review is compatible with Title VII’s non-disclosure provision, but fails to do so. The com-pany first contends that the statutory bar is limited to “using what was said or done in a conciliation as evidence going to the merits of the claims.” Brief for Petitioner 27 (emphasis added). But to make that argument, Mach Mining must add many words to the text (those shown here in italics). The actual language refers to “evidence in a subsequent proceeding,” without carving out evidence relating to non-merits issues.42 U. S. C. §2000e–5(b). And in any case, under Mach Mining’s own view of Title VII, compliance with the conciliation mandate is a merits issue, because it is a necessary “element of the [EEOC’s] claim, which the [EEOC] must plead and prove.” Brief for Petitioner 9; see id., at 31. Mach Mining therefore presents a back-up argument: “[T]he confidentiality limitation should be deemed waived” when the employer puts conciliation at issue. Id., at 30. But again, to effect a waiver Title VII requires “the written consent of the persons concerned,” which includes not just the employer but the complainant too. §2000e–5(b); see supra, at 11. And the employer’s decision to contest the EEOC’s conciliation efforts cannot waive, by “deem[ing]” or otherwise, the employee’s statutory rights.

575.US.2014_13-1034

Petitioner Moones Mellouli, a lawful permanent resident, pleaded guilty to a misdemeanor offense under Kansas law, the possession of drug paraphernalia “to . . . store [or] conceal . . . a controlled substance.” Kan. Stat. Ann. §21–5709(b)(2). The sole “paraphernalia” Mellouli was charged with possessing was a sock in which he had placed four unidentified orange tablets. Citing Mellouli’s misdemeanor conviction, an Immigration Judge ordered him deported under 8 U. S. C. §1227(a)(2)(B)(i), which authorizes the deportation (removal) of an alien “convicted of a violation of . . . any law or regulation of a State, the United States, or a foreign country relating to a controlled substance (as defined in section 802 of Title 21).” Section 802, in turn, limits the term “controlled substance” to a “drug or other substance” included in one of five federal schedules. 21 U. S. C. §802(6). Kansas defines “controlled substance” as any drug included on its own schedules, without reference to §802. Kan. Stat. Ann. §21–5701(a). At the time of Mellouli’s conviction, Kansas’ schedules included at least nine substances not on the federal lists. The Board of Immigration Appeals (BIA) affirmed Mellouli’s deportation order, and the Eighth Circuit denied his petition for review. Held: Mellouli’s Kansas conviction for concealing unnamed pills in his sock did not trigger removal under §1227(a)(2)(B)(i). Pp. 5–14. (a) The categorical approach historically taken in determining whether a state conviction renders an alien removable looks to the statutory definition of the offense of conviction, not to the particulars of the alien’s conduct. The state conviction triggers removal only if, by definition, the underlying crime falls within a category of removable offenses defined by federal law. The BIA has long applied the categorical approach to assess whether a state drug conviction triggers removal under successive versions of what is now §1227(a)(2)(B)(i). Matter of Paulus, 11 I. & N. Dec. 274, is illustrative. At the time the BIA decided Paulus, California controlled certain “narcotics” not listed as “narcotic drugs” under federal law. Id., at 275. The BIA concluded that an alien’s California conviction for offering to sell an unidentified “narcotic” was not a deportable offense, for it was possible that the conviction involved a substance controlled only under California, not federal, law. Under the Paulus analysis, Mellouli would not be deportable. The state law involved in Mellouli’s conviction, like the California statute in Paulus, was not confined to federally controlled substances; it also included substances controlled only under state, not federal, law. The BIA, however, announced and applied a different approach to drug-paraphernalia offenses (as distinguished from drug possession and distribution offenses) in Matter of Martinez Espinoza, 25 I. & N. Dec. 118. There, the BIA ranked paraphernalia statutes as relating to “the drug trade in general,” reasoning that a paraphernalia conviction “relates to” any and all controlled substances, whether or not federally listed, with which the paraphernalia can be used. Id., at 120–121. Under this reasoning, there is no need to show that the type of controlled substance involved in a paraphernalia conviction is one defined in §802. The BIA’s disparate approach to drug possession and distribution offenses and paraphernalia possession offenses finds no home in §1227(a)(2)(B)(i)’s text and “leads to consequences Congress could not have intended.” Moncrieffe v. Holder, 569 U. S. ___, ___. That approach has the anomalous result of treating less grave paraphernalia possession misdemeanors more harshly than drug possession and distribution offenses. The incongruous upshot is that an alien is not removable for possessing a substance controlled only under Kansas law, but he is removable for using a sock to contain that substance. Because it makes scant sense, the BIA’s interpretation is owed no deference under the doctrine described in Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 . Pp. 5–11. (b) The Government’s interpretation of the statute is similarly flawed. The Government argues that aliens who commit any drug crime, not just paraphernalia offenses, in States whose drug schedules substantially overlap the federal schedules are deportable, for “state statutes that criminalize hundreds of federally controlled drugs and a handful of similar substances, are laws ‘relating to’ federally controlled substances.” Brief for Respondent 17. While the words “relating to” are broad, the Government’s reading stretches the construction of §1227(a)(2)(B)(i) to the breaking point, reaching state-court convictions, like Mellouli’s, in which “[no] controlled substance (as defined in [§802])” figures as an element of the offense. Construction of §1227(a)(2)(B)(i) must be faithful to the text, which limits the meaning of “controlled substance,” for removal purposes, to the substances controlled under §802. Accordingly, to trigger removal under §1227(a)(2)(B)(i), the Government must connect an element of the alien’s conviction to a drug “defined in [§802].” Pp. 11–14. 719 F. 3d 995, reversed. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Alito, J., joined.

This case requires us to decide how immigration judges should apply a deportation (removal) provision, defined with reference to federal drug laws, to an alien convicted of a state drug-paraphernalia misdemeanor. Lawful permanent resident Moones Mellouli, in 2010, pleaded guilty to a misdemeanor offense under Kansas law, the possession of drug paraphernalia to “store, contain, conceal, inject, ingest, inhale or otherwise introduce a controlled substance into the human body.” Kan. Stat. Ann. §21–5709(b)(2) (2013 Cum. Supp.). The sole “paraphernalia” Mellouli was charged with possessing was a sock in which he had placed four orange tablets. The criminal charge and plea agreement did not identify the controlled substance involved, but Mellouli had acknowledged, prior to the charge and plea, that the tablets were Adderall. Mellouli was sentenced to a suspended term of 359 days and 12 months’ probation. In February 2012, several months after Mellouli successfully completed probation, Immigration and Customs Enforcement officers arrested him as deportable under 8 U. S. C. §1227(a)(2)(B)(i) based on his Kansas misde-meanor conviction. Section 1227(a)(2)(B)(i) authorizes the removal of an alien “convicted of a violation of . . . any law or regulation of a State, the United States, or a foreign country relating to a controlled substance (as defined in section 802 of Title 21).” We hold that Mellouli’s Kansas conviction for concealing unnamed pills in his sock didnot trigger removal under §1227(a)(2)(B)(i). The drug-paraphernalia possession law under which he was convicted, Kan. Stat. Ann. §21–5709(b), by definition, related to a controlled substance: The Kansas statute made it unlawful “to use or possess with intent to use any drug paraphernalia to . . . store [or] conceal . . . a controlled substance.” But it was immaterial under that law whether the substance was defined in 21 U. S. C. §802. Nor didthe State charge, or seek to prove, that Mellouli possessed a substance on the §802 schedules. Federal law (§1227(a)(2)(B)(i)), therefore, did not authorize Mellouli’s removal. I A This case involves the interplay between several federal and state statutes. Section 1227(a)(2)(B)(i), a provision of the Immigration and Nationality Act, 66Stat. 163, as amended, authorizes the removal of an alien “convicted of a violation of . . . any law or regulation of a State, the United States, or a foreign country relating to a controlled substance (as defined in section 802 of Title 21), other than a single offense involving possession for one’s own use of 30 grams or less of marijuana.” Section 1227(a)(2)(B)(i) incorporates 21 U. S. C. §802, which limits the term “controlled substance” to a “drug or other substance” included in one of five federal schedules. §802(6). The statute defining the offense to which Mellouli pleaded guilty, Kan. Stat. Ann. §21–5709(b), proscribes “possess[ion] with intent to use any drug paraphernalia to,” among other things, “store” or “conceal” a “controlled substance.” Kansas defines “controlled substance” as any drug included on its own schedules, and makes no reference to §802 or any other federal law. §21–5701(a).[1] At the time of Mellouli’s conviction, Kansas’ schedules included at least nine substances not included in the federal lists. See §65–4105(d)(30), (31), (33), (34), (36) (2010 Cum. Supp.); §65–4111(g) (2002); §65–4113(d)(1), (e), (f ) (2010 Cum. Supp.); see also Brief for Respondent 9, n. 2. The question presented is whether a Kansas conviction for using drug paraphernalia to store or conceal a controlled substance, §21–5709(b), subjects an alien to deportation under §1227(a)(2)(B)(i), which applies to an alien “convicted of a violation of [a state law] relating to a controlled substance (as defined in [§802]).” B Mellouli, a citizen of Tunisia, entered the United States on a student visa in 2004. He attended U. S. universities, earning a bachelor of arts degree, magna cum laude, as well as master’s degrees in applied mathematics and economics. After completing his education, Mellouli worked as an actuary and taught mathematics at the University of Missouri-Columbia. In 2009, he became a conditional permanent resident and, in 2011, a lawful permanent resident. Since December 2011, Mellouli has been engaged to be married to a U. S. citizen. In 2010, Mellouli was arrested for driving under the influence and driving with a suspended license. During a postarrest search in a Kansas detention facility, deputies discovered four orange tablets hidden in Mellouli’s sock. According to a probable-cause affidavit submitted in the state prosecution, Mellouli acknowledged that the tablets were Adderall and that he did not have a prescription for the drugs. Adderall, the brand name of an amphetamine-based drug typically prescribed to treat attention-deficit hyperactivity disorder,[2] is a controlled substance under both federal and Kansas law. See 21 CFR §1308.12(d)(1) (2014) (listing “amphetamine” and its “salts” and “isomers”); Kan. Stat. Ann. §65–4107(d)(1) (2013 Cum. Supp.) (same). Based on the probable-cause affidavit, a criminal complaint was filed charging Mellouli with trafficking contraband in jail. Ultimately, Mellouli was charged with only the lesser offense of possessing drug paraphernalia, a misdemeanor. The amended complaint alleged that Mellouli had “use[d] or possess[ed] with intent to use drug paraphernalia, to-wit: a sock, to store, contain, conceal, inject, ingest, inhale or otherwise introduce into the human body a controlled substance.” App. 23. The complaint did not identify the substance contained in the sock. Mellouli pleaded guilty to the paraphernalia possession charge; he also pleaded guilty to driving under the influence. For both offenses, Mellouli was sentenced to a suspended term of 359 days and 12 months’ probation. In February 2012, several months after Mellouli successfully completed probation, Immigration and Customs Enforcement officers arrested him as deportable under §1227(a)(2)(B)(i) based on his paraphernalia possession conviction. An Immigration Judge ordered Mellouli deported, and the Board of Immigration Appeals (BIA) affirmed the order. Mellouli was deported in 2012. Under federal law, Mellouli’s concealment of controlled-substance tablets in his sock would not have qualified as a drug-paraphernalia offense. Federal law criminalizes the sale of or commerce in drug paraphernalia, but possession alone is not criminalized at all. See 21 U. S. C. §863(a)–(b). Nor does federal law define drug paraphernalia to include common household or ready-to-wear items like socks; rather, it defines paraphernalia as any “equipment, product, or material” which is “primarily intended or designed for use” in connection with various drug-related activities. §863(d) (emphasis added). In 19 States as well, the conduct for which Mellouli was convicted—use of a sock to conceal a controlled substance—is not a criminal offense. Brief for National Immigrant Justice Center et al. as Amici Curiae 7. At most, it is a low-level infraction, often not attended by a right to counsel. Id., at 9–11. The Eighth Circuit denied Mellouli’s petition for review. 719 F. 3d 995 (2013). We granted certiorari, 573 U. S.___ (2014), and now reverse the judgment of the EighthCircuit. II We address first the rationale offered by the BIA and affirmed by the Eighth Circuit, which differentiates paraphernalia offenses from possession and distribution offenses. Essential background, in evaluating the rationale shared by the BIA and the Eighth Circuit, is the categorical approach historically taken in determining whether a state conviction renders an alien removable under the immigration statute.[3] Because Congress predicated de-portation “on convictions, not conduct,” the approach looks to the statutory definition of the offense of conviction, not to the particulars of an alien’s behavior. Das, The Immigration Penalties of Criminal Convictions: Resurrecting Categorical Analysis in Immigration Law, 86 N. Y. U. L. Rev. 1669, 1701, 1746 (2011). The state conviction triggers removal only if, by definition, the underlying crime falls within a category of removable offenses defined by federal law. Ibid. An alien’s actual conduct is irrelevant to the inquiry, as the adjudicator must “presume that the conviction rested upon nothing more than the least of the acts criminalized” under the state statute. Moncrieffe v. Holder, 569 U. S. ___, ___ (2013) (slip op., at 5) (internal quotation marks and alterations omitted).[4] The categorical approach “has a long pedigree in our Nation’s immigration law.” Id., at ___ (slip op., at 6). As early as 1913, courts examining the federal immigration statute concluded that Congress, by tying immigration penalties to convictions, intended to “limi[t] the immigration adjudicator’s assessment of a past criminal conviction to a legal analysis of the statutory offense,” and to disallow “[examination] of the facts underlying the crime.” Das, supra, at 1688, 1690. Rooted in Congress’ specification of conviction, not conduct, as the trigger for immigration consequences, the categorical approach is suited to the realities of the system. Asking immigration judges in each case to determine the circumstances underlying a state conviction would burden a system in which “large numbers of cases [are resolved by] immigration judges and front-line immigration officers, often years after the convictions.” Koh, The Whole Better than the Sum: A Case for the Categorical Approach to Determining the Immigration Consequences of Crime, 26 Geo. Immigration L. J. 257, 295 (2012). By focusing on the legal question of what a conviction necessarily established, the categorical approach ordinarily works to promote efficiency, fairness, and predictability in the administration of immigration law. See id., at 295–310; Das, supra, at 1725–1742. In particular, the approach enables aliens “to anticipate the immigration consequences of guilty pleas in criminal court,” and to enter “ ‘safe harbor’ guilty pleas [that] do not expose the [alien defendant] to the risk of immigration sanctions.” Koh, supra, at 307. See Das, supra, at 1737–1738.[5] The categorical approach has been applied routinely to assess whether a state drug conviction triggers removal under the immigration statute. As originally enacted, the removal statute specifically listed covered offenses and covered substances. It made deportable, for example, any alien convicted of “import[ing],” “buy[ing],” or “sell[ing]” any “narcotic drug,” defined as “opium, coca leaves, cocaine, or any salt, derivative, or preparation of opium or coca leaves, or cocaine.” Ch. 202, 42Stat. 596–597. Over time, Congress amended the statute to include additional offenses and additional narcotic drugs.[6] Ultimately, the Anti-Drug Abuse Act of 1986 replaced the increasingly long list of controlled substances with the now familiar reference to “a controlled substance (as defined in [§802]).” See §1751, 100Stat. 3207–47. In interpreting successive versions of the removal statute, the BIA inquired whether the state statute under which the alien was convicted covered federally controlled substances and not others.[7] Matter of Paulus, 11 I. & N. Dec. 274 (1965), is illustrative. At the time the BIA decided Paulus, the immigration statute made deportable any alien who had been “convicted of a violation of . . . any law or regulation relating to the illicit possession of or traffic in narcotic drugs or mari-huana.” Id., at 275. California controlled certain “narcotics,” such as peyote, not listed as “narcotic drugs” under federal law. Ibid. The BIA concluded that an alien’s California conviction for offering to sell an unidentified “narcotic” was not a deportable offense, for it was possible that the conviction involved a substance, such as peyote, controlled only under California law. Id., at 275–276. Because the alien’s conviction was not necessarily predicated upon a federally controlled “narcotic drug,” the BIA concluded that the conviction did not establish the alien’s deportability. Id., at 276. Under the Paulus analysis, adhered to as recently as 2014 in Matter of Ferreira, 26 I. & N. Dec. 415 (BIA 2014),[8] Mellouli would not be deportable. Mellouli pleaded guilty to concealing unnamed pills in his sock. At the time of Mellouli’s conviction, Kansas’ schedules of controlled substances included at least nine substances—e.g., salvia and jimson weed—not defined in §802. See Kan. Stat. Ann. §65–4105(d)(30), (31). The state law involved in Mellouli’s conviction, therefore, like the California statute in Paulus, was not confined to federally controlled substances; it required no proof by the prosecutor that Mellouli used his sock to conceal a substance listed under §802, as opposed to a substance controlled only under Kansas law. Under the categorical approach applied in Paulus, Mellouli’s drug-paraphernalia conviction does not render him deportable. In short, the state law under which he was charged categorically “relat[ed] to a controlled substance,” but was not limited to substances “defined in [§802].”[9] The BIA, however, announced and applied a different approach to drug-paraphernalia offenses (as distinguished from drug possession and distribution offenses) in Matter of Martinez Espinoza, 25 I. & N. Dec. 118 (2009). There, the BIA ranked paraphernalia statutes as relating to “the drug trade in general.” Id., at 121. The BIA rejected the argument that a paraphernalia conviction should not count at all because it targeted implements, not controlled substances. Id., at 120. It then reasoned that a paraphernalia conviction “relates to” any and all controlled substances, whether or not federally listed, with which the paraphernalia can be used. Id., at 121. Under this reasoning, there is no need to show that the type of controlled substance involved in a paraphernalia conviction is one defined in §802. The Immigration Judge in this case relied upon Martinez Espinoza in ordering Mellouli’s removal, quoting that decision for the proposition that “ ‘the requirement of a correspondence between the Federal and State controlled substance schedules, embraced by Matter of Paulus . . . has never been extended’ ” to paraphernalia offenses. App. to Pet. for Cert. 32 (quoting Martinez Espinoza, 25 I. & N. Dec., at 121). The BIA affirmed, reasoning that Mellouli’s conviction for possession of drug paraphernalia “involves drug trade in general and, thus, is covered under [§1227(a)(2)(B)(i)].” App. to Pet. for Cert. 18. Denying Mellouli’s petition for review, the Eighth Circuit deferred to the BIA’s decision in Martinez Espinoza, and held that a Kansas paraphernalia conviction “ ‘relates to’ a federal controlled substance because it is a crime . . . ‘associated with the drug trade in general.’ ” 719 F. 3d, at 1000. The disparate approach to state drug convictions, devised by the BIA and applied by the Eighth Circuit, finds no home in the text of §1227(a)(2)(B)(i). The approach, moreover, “leads to consequences Congress could not have intended.” Moncrieffe, 569 U. S., at ___ (slip op., at 15). Statutes should be interpreted “as a symmetrical and coherent regulatory scheme.” FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120, 133 (2000) (internal quotation marks omitted). The BIA, however, has adopted conflicting positions on the meaning of §1227(a)(2)(B)(i), distinguishing drug possession and distribution offenses from offenses involving the drug trade in general, with the anomalous result that minor paraphernalia possession offenses are treated more harshly than drug possession and distribution offenses. Drug possession and distribution convictions trigger removal only if they necessarily involve a federally controlled substance, see Paulus, 11 I. & N. Dec. 274, while convictions for paraphernalia possession, an offense less grave than drug possession and distribution, trigger removal whether or not they necessarily implicate a federally controlled substance, see Martinez Espinoza, 25 I. & N. Dec. 118. The incongruous upshot is that an alien is not removable for possessing a substance controlled only under Kansas law, but he is removable for using a sock to contain that substance. Because it makes scant sense, the BIA’s interpretation, we hold, is owed no deference under the doctrine described in Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837, 843 (1984) . III Offering an addition to the BIA’s rationale, the Eighth Circuit reasoned that a state paraphernalia possession conviction categorically relates to a federally controlled substance so long as there is “nearly a complete overlap” between the drugs controlled under state and federal law. 719 F. 3d, at 1000.[10] The Eighth Circuit’s analysis, however, scarcely explains or ameliorates the BIA’s anomalous separation of paraphernalia possession offenses from drug possession and distribution offenses. Apparently recognizing this problem, the Government urges, as does the dissent, that the overlap between state and federal drug schedules supports the removal of aliens convicted of any drug crime, not just paraphernalia of-fenses. As noted, §1227(a)(2)(B)(i) authorizes the removal of any alien “convicted of a violation of . . . any law or reg-ulation of a State, the United States, or a foreigncountry relating to a controlled substance (as defined in [§802]).” According to the Government, the words “relating to” modify “law or regulation,” rather than “violation.” Brief for Respondent 25–26 (a limiting phrase ordinarily modifies the last antecedent). Therefore, the Government argues, aliens who commit “drug crimes” in States whose drug schedules substantially overlap the federal schedules are removable, for “state statutes that criminalize hundreds of federally controlled drugs and a handful of similar substances, are laws ‘relating to’ federally controlled substances.” Brief for Respondent 17. We do not gainsay that, as the Government urges, the last reasonable referent of “relating to,” as those words appear in §1227(a)(2)(B)(i), is “law or regulation.” The removal provision is thus satisfied when the elements that make up the state crime of conviction relate to a federally controlled substance. As this case illustrates, however, the Government’s construction of the federal removal statute stretches to the breaking point, reaching state-court convictions, like Mellouli’s, in which “[no] controlled substance (as defined in [§802])” figures as an element of the offense. We recognize, too, that the §1227(a)(2)(B)(i) words to which the dissent attaches great weight, i.e., “relating to,” post, at 2–3, are “broad” and “indeterminate.” Maracich v. Spears, 570 U. S. ___, ___ (2013) (slip op., at 9) (internal quotation marks and brackets omitted).[11] As we cautioned in New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U. S. 645, 655 (1995) , those words, “extend[ed] to the furthest stretch of [their] indeterminacy, . . . stop nowhere.” “[C]ontext,” therefore, may “tu[g] . . . in favor of a narrower reading.” Yates v. United States, 574 U. S. ___, ___ (2015) (slip op., at 10). Context does so here. The historical background of §1227(a)(2)(B)(i) demonstrates that Congress and the BIA have long required a direct link between an alien’s crime of conviction and a particular federally controlled drug. Supra, at 8–9. The Government’s position here severs that link by authorizing deportation any time the state statute of conviction bears some general relation to federally controlled drugs. The Government offers no cogent reason why its position is limited to state drug schedules that have a “substantial overlap” with the federal schedules. Brief for Respondent 31. A statute with any overlap would seem to be related to federally controlled drugs. Indeed, the Government’s position might well encompass convictions for offenses related to drug activity more generally, such as gun possession, even if those convictions do not actually involve drugs (let alone federally controlled drugs). The Solicitor General, while resisting this particular example, acknowledged that convictions under statutes “that have some connection to drugs indirectly” might fall within §1227(a)(2)(B)(i). Tr. of Oral Arg. 36. This sweeping interpretation departs so sharply from the statute’s text and history that it cannot be considered a permissible reading. In sum, construction of §1227(a)(2)(B)(i) must be faithful to the text, which limits the meaning of “controlled substance,” for removal purposes, to the substances controlled under §802. We therefore reject the argument that any drug offense renders an alien removable, without regard to the appearance of the drug on a §802 schedule. Instead, to trigger removal under §1227(a)(2)(B)(i), the Government must connect an element of the alien’s conviction to a drug “defined in [§802].” * * * For the reasons stated, the judgment of the U. S. Court of Appeals for the Eighth Circuit is reversed. It is so ordered.Notes 1 At the time of Mellouli’s conviction, Kan. Stat. Ann. §§21–5701(a) and 21–5709(b) (2013 Cum. Supp.) were codified at, respectively, §§21–36a01(a) and 21–36a09(b) (2010 Cum. Supp.). 2 See H. Silverman, The Pill Book 23 (13th ed. 2008). 3 We departed from the categorical approach in Nijhawan v. Holder, 557 U. S. 29 (2009) , based on the atypical cast of the prescription at issue, 8 U. S. C. §1101(a)(43)(M)(i). That provision defines as an “aggravated felony” an offense “involv[ing] fraud or deceit in which the loss to the victim or victims exceeds $10,000.” The following subparagraph, (M)(ii), refers to an offense “described in section 7201 of title 26 (relating to tax evasion) in which the revenue loss to the Government exceeds $10,000.” No offense “described in section 7201 of title 26,” we pointed out, “has a specific loss amount as an element.” 557 U. S., at 38. Similarly, “no widely applicable federal fraud statute . . . contains a relevant monetary loss threshold,” id., at 39, and “[most] States had no major fraud or deceit statute with any relevant monetary threshold,” id., at 40. As categorically interpreted, (M)(ii), the tax evasion provision, would have no application, and (M)(i), the fraud or deceit provision, would apply only in an extraordinarily limited and haphazard manner. Ibid. We therefore concluded that Congress intended the monetary thresholds in subparagraphs (M)(i) and (M)(ii) to apply “to the specific circumstances surrounding an offender’s commission of[the defined] crime on a specific occasion.” Ibid. In the main, §1227(a)(2)(B)(i), the provision at issue here, has no such circumstance-specific thrust; its language refers to crimes generically defined. 4 A version of this approach, known as the “modified categorical approach,” applies to “state statutes that contain several different crimes, each described separately.” Moncrieffe v. Holder, 569 U. S. ___, ___ (2013) (slip op., at 5). In such cases, “a court may determine which particular offense the noncitizen was convicted of by examining the charging document and jury instructions, or in the case of a guilty plea, the plea agreement, plea colloquy, or some comparable judicial record of the factual basis for the plea.” Ibid. (internal quotation marks omitted). Off limits to the adjudicator, however, is any inquiry into the particular facts of the case. Because the Government has not argued that this case falls within the compass of the modified-categorical approach, we need not reach the issue. 5 Mellouli’s plea may be an example. In admitting only paraphernalia possession, Mellouli avoided any identification, in the record of conviction, of the federally controlled substance (Adderall) his sock contained. See supra, at 3–4. 6 The 1956 version of the statute, for example, permitted removal of any alien “who at any time has been convicted of a violation of, or a conspiracy to violate, any law or regulation relating to the illicit possession of or traffic in narcotic drugs, or who has been convicted of a violation of, or a conspiracy to violate, any law or regulation governing or controlling the taxing, manufacture, production, compounding, transportation, sale, exchange, dispensing, giving away, importation, exportation, or the possession for the purpose of the manufacture, production, compounding, transportation, sale, exchange, dispensing, giving away, importation, or exportation of opium, coca leaves, heroin, marihuana, any salt derivative or preparation of opium or coca leaves or isonipecaine or any addiction-forming or addiction-sustaining opiate.” Narcotic Control Act of 1956, §301(b), 70Stat. 575. 7 See, e.g., Matter of Fong, 10 I. & N. Dec. 616, 619 (BIA 1964) (a Pennsylvania conviction for unlawful use of a drug rendered alien removable because “every drug enumerated in the Pennsylvania law [was] found to be a narcotic drug or marijuana within the meaning of [the federal removal statute]”), overruled in part on other grounds, Matter of Sum, 13 I. & N. Dec. 569 (1970). 8 The Government acknowledges that Ferreira “assumed the applicability of [the Paulus] framework.” Brief for Respondent 49. Whether Ferreira applied that framework correctly is not a matter this case calls upon us to decide. 9 The dissent maintains that it is simply following “the statutory text.” Post, at 1. It is evident, however, that the dissent shrinks to the vanishing point the words “as defined in [§802].” If §1227(a)(2)(B)(i) stopped with the words “relating to a controlled substance,” the dissent would make sense. But Congress did not stop there. It qualified “relating to a controlled substance” by adding the limitation “as defined in [§802].” If those words do not confine §1227(a)(2)(B)(i)’s application to drugs defined in §802, one can only wonder why Congress put them there. 10 The BIA posited, but did not rely on, a similar rationale in Martinez Espinoza. See 25 I. & N. Dec., 118, 121 (2009) (basing decision on a “distinction between crimes involving the possession or distribution of a particular drug and those involving other conduct associated with the drug trade in general”). 11 The dissent observes that certain provisions of the immigration statute involving firearms and domestic violence “specif[y] the conduct that subjects an alien to removal” without “the expansive phrase ‘relating to.’ ” Post, at 3. From this statutory context, the dissent infers that Congress must have intended the words “relating to” to have expansive meaning. Post, at 3–4. But the dissent overlooks another contextual clue—i.e., that other provisions of the immigration statute tying immigration consequences to controlled-substance offenses contain no reference to §802. See 8 U. S. C. §1357(d) (allowing detainer of any alien who has been “arrested by a Federal, State, or local law enforcement official for a violation of any law relating to controlled sub-stances”); §1184(d)(3)(B)(iii) (allowing Secretary of Homeland Security to deny certain visa applications when applicant has at least three convictions of crimes “relating to a controlled substance or alcohol not arising from a single act”). These provisions demonstrate that when Congress seeks to capture conduct involving a “controlled substance,” it says just that, not “a controlled substance (as defined in [§802]).”

576.US.2014_14-556

Michigan, Kentucky, Ohio, and Tennessee define marriage as a union between one man and one woman. The petitioners, 14 same-sex couples and two men whose same-sex partners are deceased, filed suits in Federal District Courts in their home States, claiming that respondent state officials violate the Fourteenth Amendment by denying them the right to marry or to have marriages lawfully performed in another State given full recognition. Each District Court ruled in petitioners’ favor, but the Sixth Circuit consolidated the cases and reversed. Held: The Fourteenth Amendment requires a State to license a marriage between two people of the same sex and to recognize a marriage between two people of the same sex when their marriage was lawfully licensed and performed out-of-State. Pp. 3–28. (a) Before turning to the governing principles and precedents, it is appropriate to note the history of the subject now before the Court. Pp. 3–10. (1) The history of marriage as a union between two persons of the opposite sex marks the beginning of these cases. To the respondents, it would demean a timeless institution if marriage were extended to same-sex couples. But the petitioners, far from seeking to devalue marriage, seek it for themselves because of their respect—and need—for its privileges and responsibilities, as illustrated by the petitioners’ own experiences. Pp. 3–6. (2) The history of marriage is one of both continuity and change. Changes, such as the decline of arranged marriages and the abandonment of the law of coverture, have worked deep transformations in the structure of marriage, affecting aspects of marriage once viewed as essential. These new insights have strengthened, not weakened, the institution. Changed understandings of marriage are characteristic of a Nation where new dimensions of freedom become apparent to new generations. This dynamic can be seen in the Nation’s experience with gay and lesbian rights. Well into the 20th century, many States condemned same-sex intimacy as immoral, and homosexuality was treated as an illness. Later in the century, cultural and political developments allowed same-sex couples to lead more open and public lives. Extensive public and private dialogue followed, along with shifts in public attitudes. Questions about the legal treatment of gays and lesbians soon reached the courts, where they could be discussed in the formal discourse of the law. In 2003, this Court overruled its 1986 decision in Bowers v. Hardwick, 478 U. S. 186 , which upheld a Georgia law that criminalized certain homosexual acts, concluding laws making same-sex intimacy a crime “demea[n] the lives of homosexual persons.” Lawrence v. Texas, 539 U. S. 558 . In 2012, the federal Defense of Marriage Act was also struck down. United States v. Windsor, 570 U. S. ___. Numerous same-sex marriage cases reaching the federal courts and state supreme courts have added to the dialogue. Pp. 6–10. (b) The Fourteenth Amendment requires a State to license a marriage between two people of the same sex. Pp. 10–27. (1) The fundamental liberties protected by the Fourteenth Amendment’s Due Process Clause extend to certain personal choices central to individual dignity and autonomy, including intimate choices defining personal identity and beliefs. See, e.g., Eisenstadt v. Baird, 405 U. S. 438 ; Griswold v. Connecticut, 381 U. S. 479 –486. Courts must exercise reasoned judgment in identifying interests of the person so fundamental that the State must accord them its respect. History and tradition guide and discipline the inquiry but do not set its outer boundaries. When new insight reveals discord between the Constitution’s central protections and a received legal stricture, a claim to liberty must be addressed. Applying these tenets, the Court has long held the right to marry is protected by the Constitution. For example, Loving v. Virginia, 388 U. S. 1 , invalidated bans on interracial unions, and Turner v. Safley, 482 U. S. 78 , held that prisoners could not be denied the right to marry. To be sure, these cases presumed a relationship involving opposite-sex partners, as did Baker v. Nelson, 409 U. S. 810 , a one-line summary decision issued in 1972, holding that the exclusion of same-sex couples from marriage did not present a substantial federal question. But other, more instructive precedents have expressed broader principles. See, e.g., Lawrence, supra, at 574. In assessing whether the force and rationale of its cases apply to same-sex couples, the Court must respect the basic reasons why the right to marry has been long protected. See, e.g., Eisenstadt, supra, at 453–454. This analysis compels the conclusion that same-sex couples may exercise the right to marry. Pp. 10–12. (2) Four principles and traditions demonstrate that the reasons marriage is fundamental under the Constitution apply with equal force to same-sex couples. The first premise of this Court’s relevant precedents is that the right to personal choice regarding marriage is inherent in the concept of individual autonomy. This abiding connection between marriage and liberty is why Loving invalidated interracial marriage bans under the Due Process Clause. See 388 U. S., at 12. Decisions about marriage are among the most intimate that an individual can make. See Lawrence, supra, at 574. This is true for all persons, whatever their sexual orientation. A second principle in this Court’s jurisprudence is that the right to marry is fundamental because it supports a two-person union unlike any other in its importance to the committed individuals. The intimate association protected by this right was central to Griswold v. Connecticut, which held the Constitution protects the right of married couples to use contraception, 381 U. S., at 485, and was acknowledged in Turner, supra, at 95. Same-sex couples have the same right as opposite-sex couples to enjoy intimate association, a right extending beyond mere freedom from laws making same-sex intimacy a criminal offense. See Lawrence, supra, at 567. A third basis for protecting the right to marry is that it safeguards children and families and thus draws meaning from related rights of childrearing, procreation, and education. See, e.g., Pierce v. Society of Sisters, 268 U. S. 510 . Without the recognition, stability, and predictability marriage offers, children suffer the stigma of knowing their families are somehow lesser. They also suffer the significant material costs of being raised by unmarried parents, relegated to a more difficult and uncertain family life. The marriage laws at issue thus harm and humiliate the children of same-sex couples. See Windsor, supra, at ___. This does not mean that the right to marry is less meaningful for those who do not or cannot have children. Precedent protects the right of a married couple not to procreate, so the right to marry cannot be conditioned on the capacity or commitment to procreate. Finally, this Court’s cases and the Nation’s traditions make clear that marriage is a keystone of the Nation’s social order. See Maynard v. Hill, 125 U. S. 190 . States have contributed to the fundamental character of marriage by placing it at the center of many facets of the legal and social order. There is no difference between same- and opposite-sex couples with respect to this principle, yet same-sex couples are denied the constellation of benefits that the States have linked to marriage and are consigned to an instability many opposite-sex couples would find intolerable. It is demeaning to lock same-sex couples out of a central institution of the Nation’s society, for they too may aspire to the transcendent purposes of marriage. The limitation of marriage to opposite-sex couples may long have seemed natural and just, but its inconsistency with the central meaning of the fundamental right to marry is now manifest. Pp. 12–18. (3) The right of same-sex couples to marry is also derived from the Fourteenth Amendment’s guarantee of equal protection. The Due Process Clause and the Equal Protection Clause are connected in a profound way. Rights implicit in liberty and rights secured by equal protection may rest on different precepts and are not always co-extensive, yet each may be instructive as to the meaning and reach of the other. This dynamic is reflected in Loving, where the Court invoked both the Equal Protection Clause and the Due Process Clause; and in Zablocki v. Redhail, 434 U. S. 374 , where the Court invalidated a law barring fathers delinquent on child-support payments from marrying. Indeed, recognizing that new insights and societal understandings can reveal unjustified inequality within fundamental institutions that once passed unnoticed and unchallenged, this Court has invoked equal protection principles to invalidate laws imposing sex-based inequality on marriage, see, e.g., Kirchberg v. Feenstra, 450 U. S. 455 –461, and confirmed the relation between liberty and equality, see, e.g., M. L. B. v. S. L. J., 519 U. S. 102 –121. The Court has acknowledged the interlocking nature of these constitutional safeguards in the context of the legal treatment of gays and lesbians. See Lawrence, 539 U. S., at 575. This dynamic also applies to same-sex marriage. The challenged laws burden the liberty of same-sex couples, and they abridge central precepts of equality. The marriage laws at issue are in essence unequal: Same-sex couples are denied benefits afforded opposite-sex couples and are barred from exercising a fundamental right. Especially against a long history of disapproval of their relationships, this denial works a grave and continuing harm, serving to disrespect and subordinate gays and lesbians. Pp. 18–22. (4) The right to marry is a fundamental right inherent in the liberty of the person, and under the Due Process and Equal Protection Clauses of the Fourteenth Amendment couples of the same-sex may not be deprived of that right and that liberty. Same-sex couples may exercise the fundamental right to marry. Baker v. Nelson is overruled. The State laws challenged by the petitioners in these cases are held invalid to the extent they exclude same-sex couples from civil marriage on the same terms and conditions as opposite-sex couples. Pp. 22–23. (5) There may be an initial inclination to await further legislation, litigation, and debate, but referenda, legislative debates, and grassroots campaigns; studies and other writings; and extensive litigation in state and federal courts have led to an enhanced understanding of the issue. While the Constitution contemplates that democracy is the appropriate process for change, individuals who are harmed need not await legislative action before asserting a fundamental right. Bowers, in effect, upheld state action that denied gays and lesbians a fundamental right. Though it was eventually repudiated, men and women suffered pain and humiliation in the interim, and the effects of these injuries no doubt lingered long after Bowers was overruled. A ruling against same-sex couples would have the same effect and would be unjustified under the Fourteenth Amendment. The petitioners’ stories show the urgency of the issue they present to the Court, which has a duty to address these claims and answer these questions. Respondents’ argument that allowing same-sex couples to wed will harm marriage as an institution rests on a counterintuitive view of opposite-sex couples’ decisions about marriage and parenthood. Finally, the First Amendment ensures that religions, those who adhere to religious doctrines, and others have protection as they seek to teach the principles that are so fulfilling and so central to their lives and faiths. Pp. 23–27. (c) The Fourteenth Amendment requires States to recognize same-sex marriages validly performed out of State. Since same-sex couples may now exercise the fundamental right to marry in all States, there is no lawful basis for a State to refuse to recognize a lawful same-sex marriage performed in another State on the ground of its same-sex character. Pp. 27–28. 772 F. 3d 388, reversed. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Roberts, C. J., filed a dissenting opinion, in which Scalia and Thomas, JJ., joined. Scalia, J., filed a dissenting opinion, in which Thomas, J., joined. Thomas, J., filed a dissenting opinion, in which Scalia, J., joined. Alito, J., filed a dissenting opinion, in which Scalia and Thomas, JJ., joined.Notes 1 Together with No. 14–562, Tanco et al. v. Haslam, Governor of Tennessee, et al., No. 14–571, DeBoer et al. v. Snyder, Governor of Michigan, et al., and No. 14–574, Bourke et al. v. Beshear, Governor of Kentucky, also on certiorari to the same court.

The Constitution promises liberty to all within its reach, a liberty that includes certain specific rights that allow persons, within a lawful realm, to define and express their identity. The petitioners in these cases seek to find that liberty by marrying someone of the same sex and having their marriages deemed lawful on the same terms and conditions as marriages between persons of the opposite sex.IThese cases come from Michigan, Kentucky, Ohio, and Tennessee, States that define marriage as a union between one man and one woman. See, e.g., Mich. Const., Art. I, 25; Ky. Const. 233A; Ohio Rev. Code Ann. 3101.01 (Lexis 2008); Tenn. Const., Art. XI, 18. The petitioners are 14 same-sex couples and two men whose same-sex partners are deceased. The respondents are state officials responsible for enforcing the laws in question. The petitioners claim the respondents violate the Fourteenth Amendment by denying them the right to marry or to have their marriages, lawfully performed in another State, given full recognition.Petitioners filed these suits in United States District Courts in their home States. Each District Court ruled in their favor. Citations to those cases are in Appendix A, infra. The respondents appealed the decisions against them to the United States Court of Appeals for the Sixth Circuit. It consolidated the cases and reversed the judgments of the District Courts. DeBoerv. Snyder, 772 F. 3d 388 (2014). The Court of Appeals held that a State has no constitutional obligation to license same-sex marriages or to recognize same-sex marriages performed out of State.The petitioners sought certiorari. This Court granted review, limited to two questions. 574 U. S. ___ (2015). The first, presented by the cases from Michigan and Kentucky, is whether the Fourteenth Amendment requires a State to license a marriage between two people of the same sex. The second, presented by the cases from Ohio, Tennessee, and, again, Kentucky, is whether the Fourteenth Amendment requires a State to recognize a same-sex marriage licensed and performed in a State which does grant that right.IIBefore addressing the principles and precedents that govern these cases, it is appropriate to note the history of the subject now before the Court.AFrom their beginning to their most recent page, the annals of human history reveal the transcendent importance of marriage. The lifelong union of a man and a woman always has promised nobility and dignity to all persons, without regard to their station in life. Marriage is sacred to those who live by their religions and offers unique fulfillment to those who find meaning in the secular realm. Its dynamic allows two people to find a life that could not be found alone, for a marriage becomes greater than just the two persons. Rising from the most basic human needs, marriage is essential to our most profound hopes and aspirations.The centrality of marriage to the human condition makes it unsurprising that the institution has existed for millennia and across civilizations. Since the dawn of history, marriage has transformed strangers into relatives, binding families and societies together. Confucius taught that marriage lies at the foundation of government. 2 Li Chi: Book of Rites 266 (C. Chai & W. Chai eds., J. Legge transl. 1967). This wisdom was echoed centuries later and half a world away by Cicero, who wrote, "The first bond of society is marriage; next, children; and then the family." See De Officiis 57 (W. Miller transl. 1913). There are untold references to the beauty of marriage in religious and philosophical texts spanning time, cultures, and faiths, as well as in art and literature in all their forms. It is fair and necessary to say these references were based on the understanding that marriage is a union between two persons of the opposite sex.That history is the beginning of these cases. The respondents say it should be the end as well. To them, it would demean a timeless institution if the concept and lawful status of marriage were extended to two persons of the same sex. Marriage, in their view, is by its nature a gender-differentiated union of man and woman. This view long has been held and continues to be held in good faith by reasonable and sincere people here and throughout the world.The petitioners acknowledge this history but contend that these cases cannot end there. Were their intent to demean the revered idea and reality of marriage, the petitioners' claims would be of a different order. But that is neither their purpose nor their submission. To the contrary, it is the enduring importance of marriage that underlies the petitioners' contentions. This, they say, is their whole point. Far from seeking to devalue marriage, the petitioners seek it for themselves because of their respect and need for its privileges and responsibilities. And their immutable nature dictates that same-sex marriage is their only real path to this profound commitment.Recounting the circumstances of three of these cases illustrates the urgency of the petitioners' cause from their perspective. Petitioner James Obergefell, a plaintiff in the Ohio case, met John Arthur over two decades ago. They fell in love and started a life together, establishing a lasting, committed relation. In 2011, however, Arthur was diagnosed with amyotrophic lateral sclerosis, or ALS. This debilitating disease is progressive, with no known cure. Two years ago, Obergefell and Arthur decided to commit to one another, resolving to marry before Arthur died. To fulfill their mutual promise, they traveled from Ohio to Maryland, where same-sex marriage was legal. It was difficult for Arthur to move, and so the couple were wed inside a medical transport plane as it remained on the tarmac in Baltimore. Three months later, Arthur died. Ohio law does not permit Obergefell to be listed as the surviving spouse on Arthur's death certificate. By statute, they must remain strangers even in death, a state-imposed separation Obergefell deems "hurtful for the rest of time." App. in No. 14 556 etc., p. 38. He brought suit to be shown as the surviving spouse on Arthur's death certificate.April DeBoer and Jayne Rowse are co-plaintiffs in the case from Michigan. They celebrated a commitment ceremony to honor their permanent relation in 2007. They both work as nurses, DeBoer in a neonatal unit and Rowse in an emergency unit. In 2009, DeBoer and Rowse fostered and then adopted a baby boy. Later that same year, they welcomed another son into their family. The new baby, born prematurely and abandoned by his biological mother, required around-the-clock care. The next year, a baby girl with special needs joined their family. Michigan, however, permits only opposite-sex married couples or single individuals to adopt, so each child can have only one woman as his or her legal parent. If an emergency were to arise, schools and hospitals may treat the three children as if they had only one parent. And, were tragedy to befall either DeBoer or Rowse, the other would have no legal rights over the children she had not been permitted to adopt. This couple seeks relief from the continuing uncertainty their unmarried status creates in their lives.Army Reserve Sergeant First Class Ijpe DeKoe and his partner Thomas Kostura, co-plaintiffs in the Tennessee case, fell in love. In 2011, DeKoe received orders to deploy to Afghanistan. Before leaving, he and Kostura married in New York. A week later, DeKoe began his deployment, which lasted for almost a year. When he returned, the two settled in Tennessee, where DeKoe works full-time for the Army Reserve. Their lawful marriage is stripped from them whenever they reside in Tennessee, returning and disappearing as they travel across state lines. DeKoe, who served this Nation to preserve the freedom the Constitution protects, must endure a substantial burden.The cases now before the Court involve other petitioners as well, each with their own experiences. Their stories reveal that they seek not to denigrate marriage but rather to live their lives, or honor their spouses' memory, joined by its bond.BThe ancient origins of marriage confirm its centrality, but it has not stood in isolation from developments in law and society. The history of marriage is one of both continuity and change. That institution even as confined to opposite-sex relations has evolved over time.For example, marriage was once viewed as an arrangement by the couple's parents based on political, religious, and financial concerns; but by the time of the Nation's founding it was understood to be a voluntary contract between a man and a woman. See N. Cott, Public Vows: A History of Marriage and the Nation 9 17 (2000); S. Coontz, Marriage, A History 15 16 (2005). As the role and status of women changed, the institution further evolved. Under the centuries-old doctrine of coverture, a married man and woman were treated by the State as a single, male-dominated legal entity. See 1 W. Blackstone, Commentaries on the Laws of England 430 (1765). As women gained legal, political, and property rights, and as society began to understand that women have their own equal dignity, the law of coverture was abandoned. See Brief for Historians of Marriage et al. as Amici Curiae 16 19. These and other developments in the institution of marriage over the past centuries were not mere superficial changes. Rather, they worked deep transformations in its structure, affecting aspects of marriage long viewed by many as essential. See generally N. Cott, Public Vows; S. Coontz, Marriage; H. Hartog, Man & Wife in America: A History (2000).These new insights have strengthened, not weakened, the institution of marriage. Indeed, changed understandings of marriage are characteristic of a Nation where new dimensions of freedom become apparent to new generations, often through perspectives that begin in pleas or protests and then are considered in the political sphere and the judicial process.This dynamic can be seen in the Nation's experiences with the rights of gays and lesbians. Until the mid-20th century, same-sex intimacy long had been condemned as immoral by the state itself in most Western nations, a belief often embodied in the criminal law. For this reason, among others, many persons did not deem homosexuals to have dignity in their own distinct identity. A truthful declaration by same-sex couples of what was in their hearts had to remain unspoken. Even when a greater awareness of the humanity and integrity of homosexual persons came in the period after World War II, the argument that gays and lesbians had a just claim to dignity was in conflict with both law and widespread social conventions. Same-sex intimacy remained a crime in many States. Gays and lesbians were prohibited from most government employment, barred from military service, excluded under immigration laws, targeted by police, and burdened in their rights to associate. See Brief for Organization of American Historians as Amicus Curiae 5 28.For much of the 20th century, moreover, homosexuality was treated as an illness. When the American Psychiatric Association published the first Diagnostic and Statistical Manual of Mental Disorders in 1952, homosexuality was classified as a mental disorder, a position adhered to until 1973. See Position Statement on Homosexuality and Civil Rights, 1973, in 131 Am. J. Psychiatry 497 (1974). Only in more recent years have psychiatrists and others recognized that sexual orientation is both a normal expression of human sexuality and immutable. See Brief for American Psychological Association et al. as Amici Curiae 7 17.In the late 20th century, following substantial cultural and political developments, same-sex couples began to lead more open and public lives and to establish families. This development was followed by a quite extensive discussion of the issue in both governmental and private sectors and by a shift in public attitudes toward greater tolerance. As a result, questions about the rights of gays and lesbians soon reached the courts, where the issue could be discussed in the formal discourse of the law.This Court first gave detailed consideration to the legal status of homosexuals in Bowersv. Hardwick, 478 U. S. 186 (1986) . There it upheld the constitutionality of a Georgia law deemed to criminalize certain homosexual acts. Ten years later, in Romerv. Evans, 517 U. S. 620 (1996) , the Court invalidated an amendment to Colorado's Constitution that sought to foreclose any branch or political subdivision of the State from protecting persons against discrimination based on sexual orientation. Then, in 2003, the Court overruled Bowers, holding that laws making same-sex intimacy a crime "demea[n] the lives of homosexual persons." Lawrencev. Texas, 539 U. S. 558 .Against this background, the legal question of same-sex marriage arose. In 1993, the Hawaii Supreme Court held Hawaii's law restricting marriage to opposite-sex couples constituted a classification on the basis of sex and was therefore subject to strict scrutiny under the Hawaii Constitution. Baehrv. Lewin, 74 Haw. 530, 852 P. 2d 44. Although this decision did not mandate that same-sex marriage be allowed, some States were concerned by its implications and reaffirmed in their laws that marriage is defined as a union between opposite-sex partners. So too in 1996, Congress passed the Defense of Marriage Act (DOMA), 110Stat. 2419, defining marriage for all federal-law purposes as "only a legal union between one man and one woman as husband and wife." 1 U. S. C. 7.The new and widespread discussion of the subject led other States to a different conclusion. In 2003, the Supreme Judicial Court of Massachusetts held the State's Constitution guaranteed same-sex couples the right to marry. See Goodridgev. Department of Public Health, 440 Mass. 309, 798 N. E. 2d 941 (2003). After that ruling, some additional States granted marriage rights to same-sex couples, either through judicial or legislative processes. These decisions and statutes are cited in Appendix B, infra. Two Terms ago, in United Statesv. Windsor, 570 U. S. ___ (2013), this Court invalidated DOMA to the extent it barred the Federal Government from treating same-sex marriages as valid even when they were lawful in the State where they were licensed. DOMA, the Court held, impermissibly disparaged those same-sex couples "who wanted to affirm their commitment to one another before their children, their family, their friends, and their community." Id., at ___ (slip op., at 14).Numerous cases about same-sex marriage have reached the United States Courts of Appeals in recent years. In accordance with the judicial duty to base their decisions on principled reasons and neutral discussions, without scornful or disparaging commentary, courts have written a substantial body of law considering all sides of these issues. That case law helps to explain and formulate the underlying principles this Court now must consider. With the exception of the opinion here under review and one other, see Citizens for Equal Protection v. Bruning, 455 F. 3d 859, 864 868 (CAAdd hyphens between digits006), the Courts of Appeals have held that excluding same-sex couples from marriage violates the Constitution. There also have been many thoughtful District Court decisions addressing same-sex marriage and most of them, too, have concluded same-sex couples must be allowed to marry. In addition the highest courts of many States have contributed to this ongoing dialogue in decisions interpreting their own State Constitutions. These state and federal judicial opinions are cited in Appendix A, infra.After years of litigation, legislation, referenda, and the discussions that attended these public acts, the States are now divided on the issue of same-sex marriage. See Office of the Atty. Gen. of Maryland, The State of Marriage Equality in America, State-by-State Supp. (2015).IIIUnder the Due Process Clause of the Fourteenth Amendment, no State shall "deprive any person of life, liberty, or property, without due process of law." The fundamental liberties protected by this Clause include most of the rights enumerated in the Bill of Rights. See Duncanv. Louisiana, 391 U. S. 145 149 (1968). In addition these liberties extend to certain personal choices central to individual dignity and autonomy, including intimate choices that define personal identity and beliefs. See, e.g., Eisenstadtv. Baird, 405 U. S. 438, 453 (1972) ; Griswoldv. Connecticut, 381 U. S. 479 486 (1965).The identification and protection of fundamental rights is an enduring part of the judicial duty to interpret the Constitution. That responsibility, however, "has not been reduced to any formula." Poev. Ullman, 367 U. S. 497, 542 (1961) (Harlan, J., dissenting). Rather, it requires courts to exercise reasoned judgment in identifying interests of the person so fundamental that the State must accord them its respect. See ibid. That process is guided by many of the same considerations relevant to analysis of other constitutional provisions that set forth broad principles rather than specific requirements. History and tradition guide and discipline this inquiry but do not set its outer boundaries. See Lawrence, supra, at 572. That method respects our history and learns from it without allowing the past alone to rule the present.The nature of injustice is that we may not always see it in our own times. The generations that wrote and ratified the Bill of Rights and the Fourteenth Amendment did not presume to know the extent of freedom in all of its dimensions, and so they entrusted to future generations a charter protecting the right of all persons to enjoy liberty as we learn its meaning. When new insight reveals discord between the Constitution's central protections and a received legal stricture, a claim to liberty must be addressed.Applying these established tenets, the Court has long held the right to marry is protected by the Constitution. In Lovingv. Virginia, 388 U. S. 1, 12 (1967) , which invalidated bans on interracial unions, a unanimous Court held marriage is "one of the vital personal rights essential to the orderly pursuit of happiness by free men." The Court reaffirmed that holding in Zablockiv. Redhail, 434 U. S. 374, 384 (1978) , which held the right to marry was burdened by a law prohibiting fathers who were behind on child support from marrying. The Court again applied this principle in Turnerv. Safley, 482 U. S. 78, 95 (1987) , which held the right to marry was abridged by regulations limiting the privilege of prison inmates to marry. Over time and in other contexts, the Court has reiterated that the right to marry is fundamental under the Due Process Clause. See, e.g., M. L. B.v. S. L. J., 519 U. S. 102, 116 (1996) ; Cleveland Bd. of Ed.v. LaFleur, 414 U. S. 632 640 (1974); Griswold, supra, at 486; Skinnerv. Oklahoma ex rel. Williamson, 316 U. S. 535, 541 (1942) ; Meyerv. Nebraska, 262 U. S. 390, 399 (1923) .It cannot be denied that this Court's cases describing the right to marry presumed a relationship involving opposite-sex partners. The Court, like many institutions, has made assumptions defined by the world and time of which it is a part. This was evident in Bakerv. Nelson, 409 U. S. 810 , a one-line summary decision issued in 1972, holding the exclusion of same-sex couples from marriage did not present a substantial federal question.Still, there are other, more instructive precedents. This Court's cases have expressed constitutional principles of broader reach. In defining the right to marry these cases have identified essential attributes of that right based in history, tradition, and other constitutional liberties inherent in this intimate bond. See, e.g., Lawrence, 539 U. S., at 574; Turner, supra, at 95; Zablocki, supra, at 384; Loving, supra, at 12; Griswold, supra, at 486. And in assessing whether the force and rationale of its cases apply to same-sex couples, the Court must respect the basic reasons why the right to marry has been long protected. See, e.g., Eisenstadt, supra, at 453 454; Poe, supra, at 542 553 (Harlan, J., dissenting).This analysis compels the conclusion that same-sex couples may exercise the right to marry. The four principles and traditions to be discussed demonstrate that the reasons marriage is fundamental under the Constitution apply with equal force to same-sex couples.A first premise of the Court's relevant precedents is that the right to personal choice regarding marriage is inherent in the concept of individual autonomy. This abiding connection between marriage and liberty is why Loving invalidated interracial marriage bans under the Due Process Clause. See 388 U. S., at 12; see also Zablocki, supra, at 384 (observing Loving held "the right to marry is of fundamental importance for all individuals"). Like choices concerning contraception, family relationships, procreation, and childrearing, all of which are protected by the Constitution, decisions concerning marriage are among the most intimate that an individual can make. See Lawrence, supra, at 574. Indeed, the Court has noted it would be contradictory "to recognize a right of privacy with respect to other matters of family life and not with respect to the decision to enter the relationship that is the foundation of the family in our society." Zablocki, supra, at 386.Choices about marriage shape an individual's destiny. As the Supreme Judicial Court of Massachusetts has explained, because "it fulfils yearnings for security, safe haven, and connection that express our common human ity, civil marriage is an esteemed institution, and the decision whether and whom to marry is among life's momentous acts of self-definition." Goodridge, 440 Mass., at 322, 798 N. E. 2d, at 955.The nature of marriage is that, through its enduring bond, two persons together can find other freedoms, such as expression, intimacy, and spirituality. This is true for all persons, whatever their sexual orientation. See Windsor, 570 U. S., at ___ ___ (slip op., at 22 23). There is dignity in the bond between two men or two women who seek to marry and in their autonomy to make such profound choices. Cf. Loving, supra, at 12 ("[T]he freedom to marry, or not marry, a person of another race resides with the individual and cannot be infringed by the State").A second principle in this Court's jurisprudence is that the right to marry is fundamental because it supports a two-person union unlike any other in its importance to the committed individuals. This point was central to Griswold v. Connecticut, which held the Constitution protects the right of married couples to use contraception. 381 U. S., at 485. Suggesting that marriage is a right "older than the Bill of Rights," Griswold described marriage this way:"Marriage is a coming together for better or for worse, hopefully enduring, and intimate to the degree of being sacred. It is an association that promotes a way of life, not causes; a harmony in living, not political faiths; a bilateral loyalty, not commercial or social projects. Yet it is an association for as noble a purpose as any involved in our prior decisions. " Id., at 486.And in Turner, the Court again acknowledged the intimate association protected by this right, holding prisoners could not be denied the right to marry because their committed relationships satisfied the basic reasons why marriage is a fundamental right. See 482 U. S., at 95 96. The right to marry thus dignifies couples who "wish to define themselves by their commitment to each other." Windsor, supra, at ___ (slip op., at 14). Marriage responds to the universal fear that a lonely person might call out only to find no one there. It offers the hope of companionship and understanding and assurance that while both still live there will be someone to care for the other.As this Court held in Lawrence, same-sex couples have the same right as opposite-sex couples to enjoy intimate association. Lawrence invalidated laws that made same-sex intimacy a criminal act. And it acknowledged that "[w]hen sexuality finds overt expression in intimate conduct with another person, the conduct can be but one element in a personal bond that is more enduring." 539 U. S., at 567. But while Lawrence confirmed a dimension of freedom that allows individuals to engage in intimate association without criminal liability, it does not follow that freedom stops there. Outlaw to outcast may be a step forward, but it does not achieve the full promise of liberty.A third basis for protecting the right to marry is that it safeguards children and families and thus draws meaning from related rights of childrearing, procreation, and education. See Piercev. Society of Sisters, 268 U. S. 510 (1925) ; Meyer, 262 U. S., at 399. The Court has recognized these connections by describing the varied rights as a unified whole: "[T]he right to 'marry, establish a home and bring up children' is a central part of the liberty protected by the Due Process Clause." Zablocki, 434 U. S., at 384 (quoting Meyer, supra, at 399). Under the laws of the several States, some of marriage's protections for children and families are material. But marriage also confers more profound benefits. By giving recognition and legal structure to their parents' relationship, marriage allows children "to understand the integrity and closeness of their own family and its concord with other families in their community and in their daily lives." Windsor, supra, at ___ (slip op., at 23). Marriage also affords the permanency and stability important to children's best interests. See Brief for Scholars of the Constitutional Rights of Children as Amici Curiae 22 27.As all parties agree, many same-sex couples provide loving and nurturing homes to their children, whether biological or adopted. And hundreds of thousands of children are presently being raised by such couples. See Brief for Gary J. Gates as Amicus Curiae 4. Most States have allowed gays and lesbians to adopt, either as individuals or as couples, and many adopted and foster children have same-sex parents, see id., at 5. This provides powerful confirmation from the law itself that gays and lesbians can create loving, supportive families.Excluding same-sex couples from marriage thus conflicts with a central premise of the right to marry. Without the recognition, stability, and predictability marriage offers, their children suffer the stigma of knowing their families are somehow lesser. They also suffer the significant material costs of being raised by unmarried parents, relegated through no fault of their own to a more difficult and uncertain family life. The marriage laws at issue here thus harm and humiliate the children of same-sex couples. See Windsor, supra, at ___ (slip op., at 23).That is not to say the right to marry is less meaningful for those who do not or cannot have children. An ability, desire, or promise to procreate is not and has not been a prerequisite for a valid marriage in any State. In light of precedent protecting the right of a married couple not to procreate, it cannot be said the Court or the States have conditioned the right to marry on the capacity or commitment to procreate. The constitutional marriage right has many aspects, of which childbearing is only one.Fourth and finally, this Court's cases and the Nation's traditions make clear that marriage is a keystone of our social order. Alexis de Tocqueville recognized this truth on his travels through the United States almost two centuries ago:"There is certainly no country in the world where the tie of marriage is so much respected as in America . . . [W]hen the American retires from the turmoil of public life to the bosom of his family, he finds in it the image of order and of peace . . . . [H]e afterwards carries [that image] with him into public affairs." 1 Democracy in America 309 (H. Reeve transl., rev. ed. 1990).In Maynardv. Hill, 125 U. S. 190, 211 (1888) , the Court echoed de Tocqueville, explaining that marriage is "the foundation of the family and of society, without which there would be neither civilization nor progress." Marriage, the Maynard Court said, has long been " 'a great public institution, giving character to our whole civil polity.' " Id., at 213. This idea has been reiterated even as the institution has evolved in substantial ways over time, superseding rules related to parental consent, gender, and race once thought by many to be essential. See generally N. Cott, Public Vows. Marriage remains a building block of our national community.For that reason, just as a couple vows to support each other, so does society pledge to support the couple, offering symbolic recognition and material benefits to protect and nourish the union. Indeed, while the States are in general free to vary the benefits they confer on all married couples, they have throughout our history made marriage the basis for an expanding list of governmental rights, benefits, and responsibilities. These aspects of marital status include: taxation; inheritance and property rights; rules of intestate succession; spousal privilege in the law of evidence; hospital access; medical decisionmaking authority; adoption rights; the rights and benefits of survivors; birth and death certificates; professional ethics rules; campaign finance restrictions; workers' compensation benefits; health insurance; and child custody, support, and visitation rules. See Brief for United States as Amicus Curiae 6 9; Brief for American Bar Association as Amicus Curiae 8 29. Valid marriage under state law is also a significant status for over a thousand provisions of federal law. See Windsor, 570 U. S., at ___ ___ (slip op., at 15 16). The States have contributed to the fundamental character of the marriage right by placing that institution at the center of so many facets of the legal and social order.There is no difference between same- and opposite-sex couples with respect to this principle. Yet by virtue of their exclusion from that institution, same-sex couples are denied the constellation of benefits that the States have linked to marriage. This harm results in more than just material burdens. Same-sex couples are consigned to an instability many opposite-sex couples would deem intolerable in their own lives. As the State itself makes marriage all the more precious by the significance it attaches to it, exclusion from that status has the effect of teaching that gays and lesbians are unequal in important respects. It demeans gays and lesbians for the State to lock them out of a central institution of the Nation's society. Same-sex couples, too, may aspire to the transcendent purposes of marriage and seek fulfillment in its highest meaning.The limitation of marriage to opposite-sex couples may long have seemed natural and just, but its inconsistency with the central meaning of the fundamental right to marry is now manifest. With that knowledge must come the recognition that laws excluding same-sex couples from the marriage right impose stigma and injury of the kind prohibited by our basic charter.Objecting that this does not reflect an appropriate framing of the issue, the respondents refer to Washingtonv. Glucksberg, 521 U. S. 702, 721 (1997) , which called for a " 'careful description' " of fundamental rights. They assert the petitioners do not seek to exercise the right to marry but rather a new and nonexistent "right to same-sex marriage." Brief for Respondent in No. 14 556, p. 8. Glucksberg did insist that liberty under the Due Process Clause must be defined in a most circumscribed manner, with central reference to specific historical practices. Yet while that approach may have been appropriate for the asserted right there involved (physician-assisted suicide), it is inconsistent with the approach this Court has used in discussing other fundamental rights, including marriage and intimacy. Loving did not ask about a "right to interracial marriage"; Turner did not ask about a "right of inmates to marry"; and Zablocki did not ask about a "right of fathers with unpaid child support duties to marry." Rather, each case inquired about the right to marry in its comprehensive sense, asking if there was a sufficient justification for excluding the relevant class from the right. See also Glucksberg, 521 U. S., at 752 773 (Souter, J., concurring in judgment); id., at 789 792 (Breyer, J., concurring in judgments).That principle applies here. If rights were defined by who exercised them in the past, then received practices could serve as their own continued justification and new groups could not invoke rights once denied. This Court has rejected that approach, both with respect to the right to marry and the rights of gays and lesbians. See Loving 388 U. S., at 12; Lawrence, 539 U. S., at 566 567.The right to marry is fundamental as a matter of history and tradition, but rights come not from ancient sources alone. They rise, too, from a better informed understanding of how constitutional imperatives define a liberty that remains urgent in our own era. Many who deem same-sex marriage to be wrong reach that conclusion based on decent and honorable religious or philosophical premises, and neither they nor their beliefs are disparaged here. But when that sincere, personal opposition becomes enacted law and public policy, the necessary consequence is to put the imprimatur of the State itself on an exclusion that soon demeans or stigmatizes those whose own liberty is then denied. Under the Constitution, same-sex couples seek in marriage the same legal treatment as opposite-sex couples, and it would disparage their choices and diminish their personhood to deny them this right.The right of same-sex couples to marry that is part of the liberty promised by the Fourteenth Amendment is derived, too, from that Amendment's guarantee of the equal protection of the laws. The Due Process Clause and the Equal Protection Clause are connected in a profound way, though they set forth independent principles. Rights implicit in liberty and rights secured by equal protection may rest on different precepts and are not always co-extensive, yet in some instances each may be instructive as to the meaning and reach of the other. In any particular case one Clause may be thought to capture the essence of the right in a more accurate and comprehensive way, even as the two Clauses may converge in the identification and definition of the right. See M. L. B., 519 U. S., at 120 121; id., at 128 129 (Kennedy, J., concurring in judgment); Beardenv. Georgia, 461 U. S. 660, 665 (1983) . This interrelation of the two principles furthers our understanding of what freedom is and must become.The Court's cases touching upon the right to marry reflect this dynamic. In Loving the Court invalidated a prohibition on interracial marriage under both the Equal Protection Clause and the Due Process Clause. The Court first declared the prohibition invalid because of its un-equal treatment of interracial couples. It stated: "There can be no doubt that restricting the freedom to marry solely because of racial classifications violates the central meaning of the Equal Protection Clause." 388 U. S., at 12. With this link to equal protection the Court proceeded to hold the prohibition offended central precepts of liberty: "To deny this fundamental freedom on so unsupportable a basis as the racial classifications embodied in these statutes, classifications so directly subversive of the principle of equality at the heart of the Fourteenth Amendment, is surely to deprive all the State's citizens of liberty without due process of law." Ibid. The reasons why marriage is a fundamental right became more clear and compelling from a full awareness and understanding of the hurt that resulted from laws barring interracial unions.The synergy between the two protections is illustrated further in Zablocki. There the Court invoked the Equal Protection Clause as its basis for invalidating the challenged law, which, as already noted, barred fathers who were behind on child-support payments from marrying without judicial approval. The equal protection analysis depended in central part on the Court's holding that the law burdened a right "of fundamental importance." 434 U. S., at 383. It was the essential nature of the marriage right, discussed at length in Zablocki, see id., at 383 387, that made apparent the law's incompatibility with requirements of equality. Each concept liberty and equal protection leads to a stronger understanding of the other.Indeed, in interpreting the Equal Protection Clause, the Court has recognized that new insights and societal understandings can reveal unjustified inequality within our most fundamental institutions that once passed unnoticed and unchallenged. To take but one period, this occurred with respect to marriage in the 1970's and 1980's. Notwithstanding the gradual erosion of the doctrine of coverture, see supra, at 6, invidious sex-based classifications in marriage remained common through the mid-20th century. See App. to Brief for Appellant in Reedv. Reed, O. T. 1971, No. 70 4, pp. 69 88 (an extensive reference to laws extant as of 1971 treating women as unequal to men in marriage). These classifications denied the equal dignity of men and women. One State's law, for example, provided in 1971 that "the husband is the head of the family and the wife is subject to him; her legal civil existence is merged in the husband, except so far as the law recognizes her separately, either for her own protection, or for her benefit." Ga. Code Ann. 53 501 (1935). Responding to a new awareness, the Court invoked equal protection principles to invalidate laws imposing sex-based inequality on marriage. See, e.g., Kirchbergv. Feenstra, 450 U. S. 455 (1981) ; Wenglerv. Druggists Mut. Ins. Co., 446 U. S. 142 (1980) ; Califanov. Westcott, 443 U. S. 76 (1979) ; Orrv. Orr, 440 U. S. 268 (1979) ; Califanov. Goldfarb, 430 U. S. 199 (1977) (plurality opinion); Weinbergerv. Wiesenfeld, 420 U. S. 636 (1975) ; Frontierov. Richardson, 411 U. S. 677 (1973) . Like Loving and Zablocki, these precedents show the Equal Protection Clause can help to identify and correct inequalities in the institution of marriage, vindicating precepts of liberty and equality under the Constitution.Other cases confirm this relation between liberty and equality. In M. L. B.v. S. L. J., the Court invalidated under due process and equal protection principles a statute requiring indigent mothers to pay a fee in order to appeal the termination of their parental rights. See 519 U. S., at 119 124. In Eisenstadtv. Baird, the Court invoked both principles to invalidate a prohibition on the distribution of contraceptives to unmarried persons but not married persons. See 405 U. S., at 446 454. And in Skinnerv. Oklahoma ex rel. Williamson, the Court invalidated under both principles a law that allowed sterilization of habitual criminals. See 316 U. S., at 538 543.In Lawrence the Court acknowledged the interlocking nature of these constitutional safeguards in the context of the legal treatment of gays and lesbians. See 539 U. S., at 575. Although Lawrence elaborated its holding under the Due Process Clause, it acknowledged, and sought to remedy, the continuing inequality that resulted from laws making intimacy in the lives of gays and lesbians a crime against the State. See ibid. Lawrence therefore drew upon principles of liberty and equality to define and protect the rights of gays and lesbians, holding the State "cannot demean their existence or control their destiny by making their private sexual conduct a crime." Id., at 578.This dynamic also applies to same-sex marriage. It is now clear that the challenged laws burden the liberty of same-sex couples, and it must be further acknowledged that they abridge central precepts of equality. Here the marriage laws enforced by the respondents are in essence unequal: same-sex couples are denied all the benefits afforded to opposite-sex couples and are barred from exercising a fundamental right. Especially against a long history of disapproval of their relationships, this denial to same-sex couples of the right to marry works a grave and continuing harm. The imposition of this disability on gays and lesbians serves to disrespect and subordinate them. And the Equal Protection Clause, like the Due Process Clause, prohibits this unjustified infringement of the fundamental right to marry. See, e.g., Zablocki, supra, at 383 388; Skinner, 316 U. S., at 541.These considerations lead to the conclusion that the right to marry is a fundamental right inherent in the liberty of the person, and under the Due Process and Equal Protection Clauses of the Fourteenth Amendment couples of the same-sex may not be deprived of that right and that liberty. The Court now holds that same-sex couples may exercise the fundamental right to marry. No longer may this liberty be denied to them. Bakerv. Nelson must be and now is overruled, and the State laws challenged by Petitioners in these cases are now held invalid to the extent they exclude same-sex couples from civil marriage on the same terms and conditions as opposite-sex couples.IVThere may be an initial inclination in these cases to proceed with caution to await further legislation, litigation, and debate. The respondents warn there has been insufficient democratic discourse before deciding an issue so basic as the definition of marriage. In its ruling on the cases now before this Court, the majority opinion for the Court of Appeals made a cogent argument that it would be appropriate for the respondents' States to await further public discussion and political measures before licensing same-sex marriages. See DeBoer, 772 F. 3d, at 409.Yet there has been far more deliberation than this argument acknowledges. There have been referenda, legislative debates, and grassroots campaigns, as well as countless studies, papers, books, and other popular and scholarly writings. There has been extensive litigation in state and federal courts. See Appendix A, infra. Judicial opinions addressing the issue have been informed by the contentions of parties and counsel, which, in turn, reflect the more general, societal discussion of same-sex marriage and its meaning that has occurred over the past decades. As more than 100 amici make clear in their filings, many of the central institutions in American life state and local governments, the military, large and small businesses, labor unions, religious organizations, law enforcement, civic groups, professional organizations, and universities have devoted substantial attention to the question. This has led to an enhanced understanding of the issue an understanding reflected in the arguments now presented for resolution as a matter of constitutional law.Of course, the Constitution contemplates that democracy is the appropriate process for change, so long as that process does not abridge fundamental rights. Last Term, a plurality of this Court reaffirmed the importance of the democratic principle in Schuettev. BAMN, 572 U. S. ___ (2014), noting the "right of citizens to debate so they can learn and decide and then, through the political process, act in concert to try to shape the course of their own times." Id., at ___ ___ (slip op., at 15 16). Indeed, it is most often through democracy that liberty is preserved and protected in our lives. But as Schuette also said, "[t]he freedom secured by the Constitution consists, in one of its essential dimensions, of the right of the individual not to be injured by the unlawful exercise of governmental power." Id., at ___ (slip op., at 15). Thus, when the rights of persons are violated, "the Constitution requires redress by the courts," notwithstanding the more general value of democratic decisionmaking. Id., at ___ (slip op., at 17). This holds true even when protecting individual rights affects issues of the utmost importance and sensitivity.The dynamic of our constitutional system is that individuals need not await legislative action before asserting a fundamental right. The Nation's courts are open to injured individuals who come to them to vindicate their own direct, personal stake in our basic charter. An individual can invoke a right to constitutional protection when he or she is harmed, even if the broader public disagrees and even if the legislature refuses to act. The idea of the Constitution "was to withdraw certain subjects from the vicissitudes of political controversy, to place them beyond the reach of majorities and officials and to establish them as legal principles to be applied by the courts." West Virginia Bd. of Ed.v. Barnette, 319 U. S. 624, 638 (1943) . This is why "fundamental rights may not be submitted to a vote; they depend on the outcome of no elections." Ibid. It is of no moment whether advocates of same-sex marriage now enjoy or lack momentum in the democratic process. The issue before the Court here is the legal question whether the Constitution protects the right of same-sex couples to marry.This is not the first time the Court has been asked to adopt a cautious approach to recognizing and protecting fundamental rights. In Bowers, a bare majority upheld a law criminalizing same-sex intimacy. See 478 U. S., at 186, 190 195. That approach might have been viewed as a cautious endorsement of the democratic process, which had only just begun to consider the rights of gays and lesbians. Yet, in effect, Bowers upheld state action that denied gays and lesbians a fundamental right and caused them pain and humiliation. As evidenced by the dissents in that case, the facts and principles necessary to a correct holding were known to the Bowers Court. See id., at 199 (Blackmun, J., joined by Brennan, Marshall, and Stevens, JJ., dissenting); id., at 214 (Stevens, J., joined by Brennan and Marshall, JJ., dissenting). That is why Lawrence held Bowers was "not correct when it was decided." 539 U. S., at 578. Although Bowers was eventually repudiated in Lawrence, men and women were harmed in the interim, and the substantial effects of these injuries no doubt lingered long after Bowers was overruled. Dignitary wounds cannot always be healed with the stroke of a pen.A ruling against same-sex couples would have the same effect and, like Bowers, would be unjustified under the Fourteenth Amendment. The petitioners' stories make clear the urgency of the issue they present to the Court. James Obergefell now asks whether Ohio can erase his marriage to John Arthur for all time. April DeBoer and Jayne Rowse now ask whether Michigan may continue to deny them the certainty and stability all mothers desire to protect their children, and for them and their children the childhood years will pass all too soon. Ijpe DeKoe and Thomas Kostura now ask whether Tennessee can deny to one who has served this Nation the basic dignity of recognizing his New York marriage. Properly presented with the petitioners' cases, the Court has a duty to address these claims and answer these questions.Indeed, faced with a disagreement among the Courts of Appeals a disagreement that caused impermissible geographic variation in the meaning of federal law the Court granted review to determine whether same-sex couples may exercise the right to marry. Were the Court to uphold the challenged laws as constitutional, it would teach the Nation that these laws are in accord with our society's most basic compact. Were the Court to stay its hand to allow slower, case-by-case determination of the required availability of specific public benefits to same-sex couples, it still would deny gays and lesbians many rights and responsibilities intertwined with marriage.The respondents also argue allowing same-sex couples to wed will harm marriage as an institution by leading to fewer opposite-sex marriages. This may occur, the respondents contend, because licensing same-sex marriage severs the connection between natural procreation and marriage. That argument, however, rests on a counterintuitive view of opposite-sex couple's decisionmaking processes regarding marriage and parenthood. Decisions about whether to marry and raise children are based on many personal, romantic, and practical considerations; and it is unrealistic to conclude that an opposite-sex couple would choose not to marry simply because same-sex couples may do so. See Kitchenv. Herbert, 755 F. 3d 1193, 1223 (CA1Add hyphens between digits014) ("[I]t is wholly illogical to believe that state recognition of the love and commitment between same-sex couples will alter the most intimate and personal decisions of opposite-sex couples"). The respondents have not shown a foundation for the conclusion that allowing same-sex marriage will cause the harmful outcomes they describe. Indeed, with respect to this asserted basis for excluding same-sex couples from the right to marry, it is appropriate to observe these cases involve only the rights of two consenting adults whose marriages would pose no risk of harm to themselves or third parties.Finally, it must be emphasized that religions, and those who adhere to religious doctrines, may continue to advocate with utmost, sincere conviction that, by divine precepts, same-sex marriage should not be condoned. The First Amendment ensures that religious organizations and persons are given proper protection as they seek to teach the principles that are so fulfilling and so central to their lives and faiths, and to their own deep aspirations to continue the family structure they have long revered. The same is true of those who oppose same-sex marriage for other reasons. In turn, those who believe allowing same-sex marriage is proper or indeed essential, whether as a matter of religious conviction or secular belief, may engage those who disagree with their view in an open and searching debate. The Constitution, however, does not permit the State to bar same-sex couples from marriage on the same terms as accorded to couples of the opposite sex.VThese cases also present the question whether the Constitution requires States to recognize same-sex marriages validly performed out of State. As made clear by the case of Obergefell and Arthur, and by that of DeKoe and Kostura, the recognition bans inflict substantial and continuing harm on same-sex couples.Being married in one State but having that valid marriage denied in another is one of "the most perplexing and distressing complication[s]" in the law of domestic relations. Williamsv. North Carolina, 317 U. S. 287, 299 (1942) (internal quotation marks omitted). Leaving the current state of affairs in place would maintain and promote instability and uncertainty. For some couples, even an ordinary drive into a neighboring State to visit family or friends risks causing severe hardship in the event of a spouse's hospitalization while across state lines. In light of the fact that many States already allow same-sex marriage and hundreds of thousands of these marriages already have occurred the disruption caused by the recognition bans is significant and ever-growing.As counsel for the respondents acknowledged at argument, if States are required by the Constitution to issue marriage licenses to same-sex couples, the justifications for refusing to recognize those marriages performed elsewhere are undermined. See Tr. of Oral Arg. on Question 2, p. 44. The Court, in this decision, holds same-sex couples may exercise the fundamental right to marry in all States. It follows that the Court also must hold and it now does hold that there is no lawful basis for a State to refuse to recognize a lawful same-sex marriage performed in another State on the ground of its same-sex character.* * *No union is more profound than marriage, for it embodies the highest ideals of love, fidelity, devotion, sacrifice, and family. In forming a marital union, two people become something greater than once they were. As some of the petitioners in these cases demonstrate, marriage embodies a love that may endure even past death. It would misunderstand these men and women to say they disrespect the idea of marriage. Their plea is that they do respect it, respect it so deeply that they seek to find its fulfillment for themselves. Their hope is not to be condemned to live in loneliness, excluded from one of civilization's oldest institutions. They ask for equal dignity in the eyes of the law. The Constitution grants them that right.The judgment of the Court of Appeals for the Sixth Circuit is reversed.It is so ordered.APPENDICESAState and Federal Judicial DecisionsAddressing Same-Sex MarriageUnited States Courts of Appeals DecisionsAdamsv. Howerton, 673 F. 2d 1036 (CAAdd hyphens between digits982)Smeltv. County of Orange, 447 F. 3d 673 (CAAdd hyphens between digits006)Citizens for Equal Protectionv. Bruning, 455 F. 3d 859 (CAAdd hyphens between digits006)Windsorv. United States, 699 F. 3d 169 (CAAdd hyphens between digits012)Massachusettsv. Department of Health and Human Services, 682 F. 3d 1 (CAAdd hyphens between digits012)Perryv. Brown, 671 F. 3d 1052 (CAAdd hyphens between digits012)Lattav. Otter, 771 F. 3d 456 (CAAdd hyphens between digits014)Baskinv. Bogan, 766 F. 3d 648 (CAAdd hyphens between digits014)Bishopv. Smith, 760 F. 3d 1070 (CA1Add hyphens between digits014)Bosticv. Schaefer, 760 F. 3d 352 (CAAdd hyphens between digits014)Kitchenv. Herbert, 755 F. 3d 1193 (CA1Add hyphens between digits014)DeBoerv. Snyder, 772 F. 3d 388 (CAAdd hyphens between digits014)Lattav. Otter, 779 F. 3d 902 (CAAdd hyphens between digits015) (O'Scannlain, J., dissenting from the denial of rehearing en banc) United States District Court DecisionsAdamsv. Howerton, 486 F. Supp. 1119 (CD Cal. 1980) Citizens for Equal Protection, Inc.v. Bruning, 290 F. Supp. 2d 1004 (Neb. 2003)Citizens for Equal Protectionv. Bruning, 368 F. Supp. 2d 980 (Neb. 2005)Wilsonv. Ake, 354 F. Supp. 2d 1298 (MD Fla. 2005) Smeltv. County of Orange, 374 F. Supp. 2d 861 (CD Cal. 2005)Bishopv. Oklahoma ex rel. Edmondson, 447 F. Supp. 2d 1239 (ND Okla. 2006)Massachusettsv. Department of Health and Human Services, 698 F. Supp. 2d 234 (Mass. 2010)Gillv. Office of Personnel Management, 699 F. Supp. 2d 374 (Mass. 2010)Perryv. Schwarzenegger, 704 F. Supp. 2d 921 (ND Cal. 2010) Dragovichv. Department of Treasury, 764 F. Supp. 2d 1178 (ND Cal. 2011) Golinski v. Office of Personnel Management, 824 F. Supp. 2d 968 (ND Cal. 2012)Dragovichv. Department of Treasury, 872 F. Supp. 2d 944 (ND Cal. 2012)Windsorv. United States, 833 F. Supp. 2d 394 (SDNY 2012)Pedersenv. Office of Personnel Management, 881 F. Supp. 2d 294 (Conn. 2012)Jacksonv. Abercrombie, 884 F. Supp. 2d 1065 (Haw. 2012)Sevcikv. Sandoval, 911 F. Supp. 2d 996 (Nev. 2012)Merrittv. Attorney General, 2013 WL 6044329 (MD La., Nov. 14, 2013) Grayv. Orr, 4 F. Supp. 3d 984 (ND Ill. 2013) Leev. Orr, 2013 WL 6490577 (ND Ill., Dec. 10, 2013)Kitchenv. Herbert, 961 F. Supp. 2d 1181 (Utah 2013)Obergefellv. Wymyslo, 962 F. Supp. 2d 968 (SD Ohio 2013)Bishopv. United States ex rel. Holder, 962 F. Supp. 2d 1252 (ND Okla. 2014)Bourkev. Beshear, 996 F. Supp. 2d 542 (WD Ky. 2014)Leev. Orr, 2014 WL 683680 (ND Ill., Feb. 21, 2014)Bosticv. Rainey, 970 F. Supp. 2d 456 (ED Va. 2014)De Leonv. Perry, 975 F. Supp. 2d 632 (WD Tex. 2014)Tancov. Haslam, 7 F. Supp. 3d 759 (MD Tenn. 2014)DeBoerv. Snyder, 973 F. Supp. 2d 757 (ED Mich. 2014)Henryv. Himes, 14 F. Supp. 3d 1036 (SD Ohio 2014)Lattav. Otter, 19 F. Supp. 3d 1054 (Idaho 2014)Geigerv. Kitzhaber, 994 F. Supp. 2d 1128 (Ore. 2014)Evansv. Utah, 21 F. Supp. 3d 1192 (Utah 2014)Whitewoodv. Wolf, 992 F. Supp. 2d 410 (MD Pa. 2014)Wolfv. Walker, 986 F. Supp. 2d 982 (WD Wis. 2014)Baskinv. Bogan, 12 F. Supp. 3d 1144 (SD Ind. 2014)Lovev. Beshear, 989 F. Supp. 2d 536 (WD Ky. 2014)Burnsv. Hickenlooper, 2014 WL 3634834 (Colo., July 23, 2014)Bowlingv. Pence, 39 F. Supp. 3d 1025 (SD Ind. 2014)Brennerv. Scott, 999 F. Supp. 2d 1278 (ND Fla. 2014)Robicheauxv. Caldwell, 2 F. Supp. 3d 910 (ED La. 2014)General Synod of the United Church of Christv. Resinger, 12 F. Supp. 3d 790 (WDNC 2014)Hambyv. Parnell, 56 F. Supp. 3d 1056 (Alaska 2014)Fisher-Bornev. Smith, 14 F. Supp. 3d 695 (MDNC 2014)Majorsv. Horne, 14 F. Supp. 3d 1313 (Ariz. 2014)Connollyv. Jeanes, ___ F. Supp. 3d ___, 2014 WL 5320642 (Ariz., Oct. 17, 2014)Guzzov. Mead, 2014 WL 5317797 (Wyo., Oct. 17, 2014)Conde-Vidalv. Garcia-Padilla, 54 F. Supp. 3d 157 (PR 2014)Mariev. Moser, ___ F. Supp. 3d ___, 2014 WL 5598128 (Kan., Nov. 4, 2014)Lawsonv. Kelly, 58 F. Supp. 3d 923 (WD Mo. 2014)McGeev. Cole, ___ F. Supp. 3d ___, 2014 WL 5802665 (SD W. Va., Nov. 7, 2014)Condonv. Haley, 21 F. Supp. 3d 572 (S. C. 2014)Bradacsv. Haley, 58 F. Supp. 3d 514 (S. C. 2014)Rolandov. Fox, 23 F. Supp. 3d 1227 (Mont. 2014)Jerniganv. Crane, ___ F. Supp. 3d ___, 2014 WL 6685391 (ED Ark., Nov. 25, 2014)Campaign for Southern Equalityv. Bryant, ___ F. Supp. 3d ___, 2014 WL 6680570 (SD Miss., Nov. 25, 2014)Innissv. Aderhold, ___ F. Supp. 3d ___, 2015 WL 300593 (ND Ga., Jan. 8, 2015) Rosenbrahnv. Daugaard, 61 F. Supp. 3d 862 (S. D., 2015)Casparv. Snyder, ___ F. Supp. 3d ___, 2015 WL 224741 (ED Mich., Jan. 15, 2015)Searceyv. Strange, 2015 U. S. Dist. LEXIS 7776 (SD Ala., Jan. 23, 2015)Strawserv. Strange, 44 F. Supp. 3d 1206 (SD Ala. 2015)Watersv. Ricketts, 48 F. Supp. 3d 1271 (Neb. 2015)State Highest Court Decisions Bakerv. Nelson, 291 Minn. 310, 191 N. W. 2d 185 (1971)Jonesv. Hallahan, 501 S. W. 2d 588 (Ky. 1973)Baehrv. Lewin, 74 Haw. 530, 852 P. 2d 44 (1993)Deanv. District of Columbia, 653 A. 2d 307 (D. C. 1995)Bakerv. State, 170 Vt. 194, 744 A. 2d 864 (1999)Brausev. State, 21 P. 3d 357 (Alaska 2001) (ripeness)Goodridgev. Department of Public Health, 440 Mass. 309, 798 N. E. 2d 941 (2003) In re Opinions of the Justices to the Senate, 440 Mass. 1201, 802 N. E. 2d 565 (2004)Liv. State, 338 Or. 376, 110 P. 3d 91 (2005)Cote-Whitacrev. Department of Public Health,446 Mass. 350, 844 N. E. 2d 623 (2006)Lewisv. Harris, 188 N. J. 415, 908 A. 2d 196 (2006)Andersenv. King County, 158 Wash. 2d 1, 138 P. 3d 963 (2006)Hernandezv. Robles, 7 N. Y. 3d 338, 855 N. E. 2d 1 (2006)Conawayv. Deane, 401 Md. 219, 932 A. 2d 571 (2007)In re Marriage Cases, 43 Cal. 4th 757, 183 P. 3d 384 (2008)Kerriganv. Commissioner of Public Health, 289 Conn. 135, 957 A. 2d 407 (2008)Straussv. Horton, 46 Cal. 4th 364, 207 P. 3d 48 (2009)Varnumv. Brien, 763 N. W. 2d 862 (Iowa 2009)Griegov. Oliver, 2014 NMSC 003, ___ N. M. ___, 316 P. 3d 865 (2013)Garden State Equalityv. Dow, 216 N. J. 314, 79 A. 3d 1036 (2013)Ex parte State ex rel. Alabama Policy Institute, ___ So. 3d ___, 2015 WL 892752 (Ala., Mar. 3, 2015)BState Legislation and Judicial DecisionsLegalizing Same-Sex MarriageLegislationDel. Code Ann., Tit. 13, 129 (Cum. Supp. 2014)D. C. Act No. 18 248, 57 D. C. Reg. 27 (2010)Haw. Rev. Stat. 572 1 (2006) and 2013 Cum. Supp.)Ill. Pub. Act No. 98 597Me. Rev. Stat. Ann., Tit. 19, 650 A (Cum. Supp. 2014)2012 Md. Laws p. 92013 Minn Laws p. 4042009 N. H. Laws p. 602011 N. Y Laws p. 7492013 R. I. Laws p. 72009 Vt. Acts & Resolves p. 332012 Wash. Sess. Laws p. 199Judicial DecisionsGoodridgev. Department of Public Health, 440 Mass. 309, 798 N. E. 2d 941 (2003)Kerriganv. Commissioner of Public Health, 289 Conn. 135, 957 A. 2d 407 (2008)Varnumv. Brien, 763 N. W. 2d 862 (Iowa 2009)Griegov. Oliver, 2014 NMSC 003, ___ N. M. ___, 316 P. 3d 865 (2013)Garden State Equalityv. Dow, 216 N. J. 314, 79 A. 3d 1036 (2013)

576.US.2014_13-1352

Respondent Darius Clark sent his girlfriend away to engage in prostitution while he cared for her 3-year-old son L. P. and 18-month-old daughter A. T. When L. P.’s preschool teachers noticed marks on his body, he identified Clark as his abuser. Clark was subsequently tried on multiple counts related to the abuse of both children. At trial, the State introduced L. P.’s statements to his teachers as evidence of Clark’s guilt, but L. P. did not testify. The trial court denied Clark’s motion to exclude the statements under the Sixth Amendment’s Confrontation Clause. A jury convicted Clark on all but one count. The state appellate court reversed the conviction on Confrontation Clause grounds, and the Supreme Court of Ohio affirmed. Held: The introduction of L. P.’s statements at trial did not violate the Confrontation Clause. Pp. 4–12. (a) This Court’s decision in Crawford v. Washington, 541 U. S. 36 , held that the Confrontation Clause generally prohibits the introduction of “testimonial” statements by a nontestifying witness, unless the witness is “unavailable to testify, and the defendant had had a prior opportunity for cross-examination.” A statement qualifies as testimonial if the “primary purpose” of the conversation was to “creat[e] an out-of-court substitute for trial testimony.” Michigan v. Bryant, 562 U. S. 344 . In making that “primary purpose” determination, courts must consider “all of the relevant circumstances.” Ibid. “Where no such primary purpose exists, the admissibility of a statement is the concern of state and federal rules of evidence, not the Confrontation Clause.” Id., at 359. But that does not mean that the Confrontation Clause bars every statement that satisfies the “primary purpose” test. The Court has recognized that the Confrontation Clause does not prohibit the introduction of out-of-court statements that would have been admissible in a criminal case at the time of the founding. See Giles v. California, 554 U. S. 353 –359; Crawford, 541 U. S., at 56, n. 6, 62. Thus, the primary purpose test is a necessary, but not always sufficient, condition for the exclusion of out-of-court statements under the Confrontation Clause. Pp. 4–7. (b) Considering all the relevant circumstances, L. P.’s statements were not testimonial. L. P.’s statements were not made with the primary purpose of creating evidence for Clark’s prosecution. They occurred in the context of an ongoing emergency involving suspected child abuse. L. P.’s teachers asked questions aimed at identifying and ending a threat. They did not inform the child that his answers would be used to arrest or punish his abuser. L. P. never hinted that he intended his statements to be used by the police or prosecutors. And the conversation was informal and spontaneous. L. P.’s age further confirms that the statements in question were not testimonial because statements by very young children will rarely, if ever, implicate the Confrontation Clause. As a historical matter, moreover, there is strong evidence that statements made in circumstances like these were regularly admitted at common law. Finally, although statements to individuals other than law enforcement officers are not categorically outside the Sixth Amendment’s reach, the fact that L. P. was speaking to his teachers is highly relevant. Statements to individuals who are not principally charged with uncovering and prosecuting criminal behavior are significantly less likely to be testimonial than those given to law enforcement officers. Pp. 7–10. (c) Clark’s arguments to the contrary are unpersuasive. Mandatory reporting obligations do not convert a conversation between a concerned teacher and her student into a law enforcement mission aimed at gathering evidence for prosecution. It is irrelevant that the teachers’ questions and their duty to report the matter had the natural tendency to result in Clark’s prosecution. And this Court’s Confrontation Clause decisions do not determine whether a statement is testimonial by examining whether a jury would view the statement as the equivalent of in-court testimony. Instead, the test is whether a statement was given with the “primary purpose of creating an out-of-court substitute for trial testimony.” Bryant, supra, at 358. Here, the answer is clear: L. P.’s statements to his teachers were not testimonial. Pp. 11–12. 137 Ohio St. 3d 346, 2013–Ohio–4731, 999 N. E. 2d 592, reversed and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Scalia, J., filed an opinion concurring in the judgment, in which Ginsburg, J., joined. Thomas, J., filed an opinion concurring in the judgment.

Darius Clark sent his girlfriend hundreds of miles away to engage in prostitution and agreed to care for her two young children while she was out of town. A day later, teachers discovered red marks on her 3-year-old son, and the boy identified Clark as his abuser. The question in this case is whether the Sixth Amendment’s Confrontation Clause prohibited prosecutors from introducing those statements when the child was not available to be cross-examined. Because neither the child nor his teachers had the primary purpose of assisting in Clark’s prosecution, the child’s statements do not implicate the Confrontation Clause and therefore were admissible at trial. I Darius Clark, who went by the nickname “Dee,” lived in Cleveland, Ohio, with his girlfriend, T. T., and her two children: L. P., a 3-year-old boy, and A. T., an 18-month-old girl.[1] Clark was also T. T.’s pimp, and he would regularly send her on trips to Washington, D. C., to work as a prostitute. In March 2010, T. T. went on one such trip, and she left the children in Clark’s care. The next day, Clark took L. P. to preschool. In the lunchroom, one of L. P.’s teachers, Ramona Whitley, observed that L. P.’s left eye appeared bloodshot. She asked him “ ‘[w]hat happened,’ ” and he initially said nothing. 137 Ohio St. 3d 346, 347, 2013–Ohio–4731, 999 N. E. 2d 592, 594. Eventually, however, he told the teacher that he “ ‘fell.’ ” Ibid. When they moved into the brighter lights of a classroom, Whitley noticed “ ‘[r]ed marks, like whips of some sort,’ ” on L. P.’s face. Ibid. She notified the lead teacher, Debra Jones, who asked L. P., “ ‘Who did this? What happened to you?’ ” Id., at 348, 999 N. E. 2d, at 595. According to Jones, L. P. “ ‘seemed kind of bewildered’ ” and “ ‘said something like, Dee, Dee.’ ” Ibid. Jones asked L. P. whether Dee is “big or little,” to which L. P. responded that “Dee is big.” App. 60, 64. Jones then brought L. P.to her supervisor, who lifted the boy’s shirt, revealing more injuries. Whitley called a child abuse hotline to alert authorities about the suspected abuse. When Clark later arrived at the school, he denied responsibility for the injuries and quickly left with L. P. The next day, a social worker found the children at Clark’s mother’s house and took them to a hospital, where a physician discovered additional injuries suggesting child abuse. L. P. had a black eye, belt marks on his back and stomach, and bruises all over his body. A. T. had two black eyes, a swollen hand, and a large burn on her cheek, and two pigtails had been ripped out at the roots of her hair. A grand jury indicted Clark on five counts of felonious assault (four related to A. T. and one related to L. P.), two counts of endangering children (one for each child), and two counts of domestic violence (one for each child). At trial, the State introduced L. P.’s statements to his teachers as evidence of Clark’s guilt, but L. P. did not testify. Under Ohio law, children younger than 10 years old are incompetent to testify if they “appear incapable of receiving just impressions of the facts and transactions respecting which they are examined, or of relating them truly.” Ohio Rule Evid. 601(A) (Lexis 2010). After conducting a hearing, the trial court concluded that L. P. was not competent to testify. But under Ohio Rule of Evidence 807, which allows the admission of reliable hearsay by child abuse victims, the court ruled that L. P.’s statements to his teachers bore sufficient guarantees of trustworthiness to be admitted as evidence. Clark moved to exclude testimony about L. P.’s out-of-court statements under the Confrontation Clause. The trial court denied the motion, ruling that L. P.’s responses were not testimonial statements covered by the Sixth Amendment. The jury found Clark guilty on all counts except for one assault count related to A. T., and it sentenced him to 28 years’ imprisonment. Clark appealed his conviction, and a state appellate court reversed on the ground that the introduction of L. P.’s out-of-court statements violated the Confrontation Clause. In a 4-to-3 decision, the Supreme Court of Ohio affirmed. It held that, under this Court’s Confrontation Clause decisions, L. P.’s statements qualified as testimonial because the primary purpose of the teachers’ questioning “was not to deal with an existing emergency but rather to gather evidence potentially relevant to a subsequent criminal prosecution.” 137 Ohio St. 3d, at 350, 999 N. E. 2d, at 597. The court noted that Ohio has a “mandatory reporting” law that requires certain professionals, including preschool teachers, to report suspected child abuse to government authorities. See id., at 349–350, 999 N. E. 2d, at 596–597. In the court’s view, the teachers acted as agents of the State under the mandatory reporting law and “sought facts concerning past criminal activity to identify the person responsible, eliciting statements that ‘are functionally identical to live, in-court testimony, doing precisely what a witness does on direct examination.’ ” Id., at 355, 999 N. E. 2d, at 600 (quoting Melendez-Diaz v. Massachusetts, 557 U. S. 305 –311 (2009); some internal quotation marks omitted). We granted certiorari, 573 U. S. ___ (2014), and we now reverse. II A The Sixth Amendment’s Confrontation Clause, which is binding on the States through the Fourteenth Amendment, provides: “In all criminal prosecutions, the accused shall enjoy the right . . . to be confronted with the wit-nesses against him.” In Ohio v. Roberts, 448 U. S. 56, 66 (1980) , we interpreted the Clause to permit the admission of out-of-court statements by an unavailable witness, so long as the statements bore “adequate ‘indicia of reliability.’ ” Such indicia are present, we held, if “the evidence falls within a firmly rooted hearsay exception” or bears “particularized guarantees of trustworthiness.” Ibid. In Crawford v. Washington, 541 U. S. 36 (2004) , we adopted a different approach. We explained that “witnesses,” under the Confrontation Clause, are those “who bear testimony,” and we defined “testimony” as “a solemn declaration or affirmation made for the purpose of establishing or proving some fact.” Id., at 51 (internal quotation marks and alteration omitted). The Sixth Amendment, we concluded, prohibits the introduction of testimonial statements by a nontestifying witness, unless the witness is “unavailable to testify, and the defendant had had a prior opportunity for cross-examination.” Id., at 54. Applying that definition to the facts in Crawford, we held that statements by a witness during police questioning at the station house were testimonial and thus could not be admitted. But our decision in Crawford did not offer an exhaustive definition of “testimonial” statements. Instead, Crawford stated that the label “applies at a minimum to prior testimony at a preliminary hearing, before a grand jury, or at a former trial; and to police interrogations.” Id., at 68. Our more recent cases have labored to flesh out what it means for a statement to be “testimonial.” In Davis v. Washington and Hammon v. Indiana, 547 U. S. 813 (2006) , which we decided together, we dealt with statements given to law enforcement officers by the victims of domestic abuse. The victim in Davis made statements to a 911 emergency operator during and shortly after her boyfriend’s violent attack. In Hammon, the victim, after being isolated from her abusive husband, made statements to police that were memorialized in a “ ‘battery affidavit.’ ” Id., at 820. We held that the statements in Hammon were testimonial, while the statements in Davis were not. Announcing what has come to be known as the “primary purpose” test, we explained: “Statements are nontestimonial when made in the course of police interrogation under circumstances objectively indicating that the primary purpose of the interrogation is to enable police assistance to meet an ongoing emergency. They are testimonial when the circumstances objectively indicate that there is no such ongoing emergency, and that the primary purpose of the interrogation is to establish or prove past events poten-tially relevant to later criminal prosecution.” Id., at 822. Because the cases involved statements to law enforcement officers, we reserved the question whether similar statements to individuals other than law enforcement officers would raise similar issues under the Confrontation Clause. See id., at 823, n. 2. In Michigan v. Bryant, 562 U. S. 344 (2011) , we further expounded on the primary purpose test. The inquiry, we emphasized, must consider “all of the relevant circumstances.” Id., at 369. And we reiterated our view in Davis that, when “the primary purpose of an interrogation is to respond to an ‘ongoing emergency,’ its purpose is not to create a record for trial and thus is not within the scope of the [Confrontation] Clause.” 562 U. S., at 358. At the same time, we noted that “there may be other circumstances, aside from ongoing emergencies, when a statement is not procured with a primary purpose of creating an out-of-court substitute for trial testimony.” Ibid. “[T]he existence vel non of an ongoing emergency is not the touchstone of the testimonial inquiry.” Id., at 374. Instead, “whether an ongoing emergency exists is simply one factor . . . that informs the ultimate inquiry regarding the ‘primary purpose’ of an interrogation.” Id., at 366. One additional factor is “the informality of the situation and the interrogation.” Id., at 377. A “formal station-house interrogation,” like the questioning in Crawford, is more likely to provoke testimonial statements, while less formal questioning is less likely to reflect a primary purpose aimed at obtaining testimonial evidence against the accused. Id., at 366, 377. And in determining whether a statement is testimonial, “standard rules of hearsay, designed to identify some statements as reliable, will be relevant.” Id., at 358–359. In the end, the question is whether, in light of all the circumstances, viewed objectively, the “primary purpose” of the conversation was to “creat[e] an out-of-court substitute for trial testimony.” Id., at 358. Applying these principles in Bryant, we held that the statements made by a dying victim about his assailant were not testimonial because the circumstances objectively indicated that the conversation was primarily aimed at quelling an ongoing emergency, not establishing evidence for the prosecution. Because the relevant statements were made to law enforcement officers, we again declined to decide whether the same analysis applies to statements made to individuals other than the police. See id., at 357, n. 3. Thus, under our precedents, a statement cannot fall within the Confrontation Clause unless its primary purpose was testimonial. “Where no such primary purpose exists, the admissibility of a statement is the concern of state and federal rules of evidence, not the Confrontation Clause.” Id., at 359. But that does not mean that the Confrontation Clause bars every statement that satisfies the “primary purpose” test. We have recognized that the Confrontation Clause does not prohibit the introduction of out-of-court statements that would have been admissible in a criminal case at the time of the founding. See Giles v. California, 554 U. S. 353 –359 (2008); Crawford, 541 U. S., at 56, n. 6, 62. Thus, the primary purpose test is a necessary, but not always sufficient, condition for the exclusion of out-of-court statements under the Confrontation Clause. B In this case, we consider statements made to preschool teachers, not the police. We are therefore presented with the question we have repeatedly reserved: whether statements to persons other than law enforcement officers are subject to the Confrontation Clause. Because at least some statements to individuals who are not law enforcement officers could conceivably raise confrontation concerns, we decline to adopt a categorical rule excluding them from the Sixth Amendment’s reach. Nevertheless, such statements are much less likely to be testimonial than statements to law enforcement officers. And considering all the relevant circumstances here, L. P.’s statements clearly were not made with the primary purpose of creating evidence for Clark’s prosecution. Thus, their introduction at trial did not violate the Confrontation Clause. L. P.’s statements occurred in the context of an ongoing emergency involving suspected child abuse. When L. P.’s teachers noticed his injuries, they rightly became worried that the 3-year-old was the victim of serious violence. Because the teachers needed to know whether it was safe to release L. P. to his guardian at the end of the day, they needed to determine who might be abusing the child.[2] Thus, the immediate concern was to protect a vulnerable child who needed help. Our holding in Bryant is instructive. As in Bryant, the emergency in this case was ongoing, and the circumstances were not entirely clear. L. P.’s teachers were not sure who had abused him or how best to secure his safety. Nor were they sure whether any other children might be at risk. As a result, their questions and L. P.’s answers were primarily aimed at identifying and ending the threat. Though not as harried, the conversation here was also similar to the 911 call in Davis. The teachers’ questions were meant to identify the abuser in order to protect the victim from future attacks. Whether the teachers thought that this would be done by apprehending the abuser or by some other means is irrelevant. And the circumstances in this case were unlike the interrogation in Hammon, where the police knew the identity of the assailant and questioned the victim after shielding her from potential harm. There is no indication that the primary purpose of the conversation was to gather evidence for Clark’s prosecution. On the contrary, it is clear that the first objective was to protect L. P. At no point did the teachers inform L. P. that his answers would be used to arrest or punish his abuser. L. P. never hinted that he intended his statements to be used by the police or prosecutors. And the conversation between L. P. and his teachers was informal and spontaneous. The teachers asked L. P. about his injuries immediately upon discovering them, in the informal setting of a preschool lunchroom and classroom, and they did so precisely as any concerned citizen would talk to a child who might be the victim of abuse. This was nothing like the formalized station-house questioning in Crawford or the police interrogation and battery affidavit in Hammon. L. P.’s age fortifies our conclusion that the statements in question were not testimonial. Statements by very young children will rarely, if ever, implicate the Confrontation Clause. Few preschool students understand the details of our criminal justice system. Rather, “[r]esearch on children’s understanding of the legal system finds that” young children “have little understanding of prosecution.” Brief for American Professional Society on the Abuse of Children as Amicus Curiae 7, and n. 5 (collecting sources). And Clark does not dispute those findings. Thus, it is extremely unlikely that a 3-year-old child in L. P.’s position would intend his statements to be a substitute for trial testimony. On the contrary, a young child in these circumstances would simply want the abuse to end, would want to protect other victims, or would have no discernible purpose at all. As a historical matter, moreover, there is strong evidence that statements made in circumstances similar to those facing L. P. and his teachers were admissible at common law. See Lyon & LaMagna, The History of Children’s Hearsay: From Old Bailey to Post-Davis, 82 Ind. L. J. 1029, 1030 (2007); see also id., at 1041–1044 (examining child rape cases from 1687 to 1788); J. Langbein, The Origins of Adversary Criminal Trial 239 (2003) (“The Old Bailey” court in 18th-century London “tolerated flagrant hearsay in rape prosecutions involving a child victim who was not competent to testify because she was too young to appreciate the significance of her oath”). And when 18th-century courts excluded statements of this sort, see, e.g., King v. Brasier, 1 Leach 199, 168 Eng. Rep. 202 (K. B. 1779), they appeared to do so because the child should have been ruled competent to testify, not because the statements were otherwise inadmissible. See Lyon & LaMagna, supra, at 1053–1054. It is thus highly doubtful that statements like L. P.’s ever would have been understood to raise Confrontation Clause concerns. Neither Crawford nor any of the cases that it has produced has mounted evidence that the adoption of the Confrontation Clause was understood to require the exclusion of evidence that was regularly admitted in criminal cases at the time of the founding. Certainly, the statements in this case are nothing like the notorious use of ex parte examination in Sir Walter Raleigh’s trial for treason, which we have frequently identified as “the principal evil at which the Confrontation Clause was directed.” Crawford, 541 U. S., at 50; see also Bryant, 562 U. S., at 358. Finally, although we decline to adopt a rule that statements to individuals who are not law enforcement officers are categorically outside the Sixth Amendment, the fact that L. P. was speaking to his teachers remains highly relevant. Courts must evaluate challenged statements in context, and part of that context is the questioner’s iden-tity. See id., at 369. Statements made to someone who is not principally charged with uncovering and prosecuting criminal behavior are significantly less likely to be testimonial than statements given to law enforcement officers. See, e.g., Giles, 554 U. S., at 376. It is common sense that the relationship between a student and his teacher is very different from that between a citizen and the police. We do not ignore that reality. In light of these circumstances, the Sixth Amendment did not prohibit the State from introducing L. P.’s statements at trial. III Clark’s efforts to avoid this conclusion are all off-base. He emphasizes Ohio’s mandatory reporting obligations, in an attempt to equate L. P.’s teachers with the police and their caring questions with official interrogations. But the comparison is inapt. The teachers’ pressing concern was to protect L. P. and remove him from harm’s way. Like all good teachers, they undoubtedly would have acted with the same purpose whether or not they had a state-law duty to report abuse. And mandatory reporting statutes alone cannot convert a conversation between a concerned teacher and her student into a law enforcement mission aimed primarily at gathering evidence for a prosecution. It is irrelevant that the teachers’ questions and their duty to report the matter had the natural tendency to result in Clark’s prosecution. The statements at issue in Davis and Bryant supported the defendants’ convictions, and the police always have an obligation to ask questions to resolve ongoing emergencies. Yet, we held in those cases that the Confrontation Clause did not prohibit introduction of the statements because they were not primarily intended to be testimonial. Thus, Clark is also wrong to suggest that admitting L. P.’s statements would be fundamentally unfair given that Ohio law does not allow incompetent children to testify. In any Confrontation Clause case, the individual who provided the out-of-court statement is not available as an in-court witness, but the testimony is admissible under an exception to the hearsay rules and is probative of the defendant’s guilt. The fact that the witness is unavailable because of a different rule of evidence does not change our analysis. Finally, Clark asks us to shift our focus from the context of L. P.’s conversation with his teachers to the jury’s perception of those statements. Because, in his view, the “jury treated L. P.’s accusation as the functional equivalent of testimony,” Clark argues that we must prohibit its introduction. Brief for Respondent 42. Our Confrontation Clause decisions, however, do not determine whether a statement is testimonial by examining whether a jury would view the statement as the equivalent of in-court testimony. The logic of this argument, moreover, would lead to the conclusion that virtually all out-of-court statements offered by the prosecution are testimonial. The prosecution is unlikely to offer out-of-court statements unless they tend to support the defendant’s guilt, and all such statements could be viewed as a substitute for in-court testimony. We have never suggested, however, that the Confrontation Clause bars the introduction of all out-of-court statements that support the prosecution’s case. Instead, we ask whether a statement was given with the “primary purpose of creating an out-of-court substitutefor trial testimony.” Bryant, supra, at 358. Here, the an-swer is clear: L. P.’s statements to his teachers were not testimonial. IV We reverse the judgment of the Supreme Court of Ohio and remand the case for further proceedings not inconsistent with this opinion. It is so ordered.Notes 1 Like the Ohio courts, we identify Clark’s victims and their mother by their initials. 2 In fact, the teachers and a social worker who had come to the school were reluctant to release L. P. into Clark’s care after the boy identified Clark as his abuser. But after a brief “stare-down” with the social worker, Clark bolted out the door with L. P., and social services were not able to locate the children until the next day. App. 92–102, 150–151.

575.US.2014_13-271

Respondents, a group of manufacturers, hospitals, and other institutions that buy natural gas directly from interstate pipelines, sued petitioner interstate pipelines, claiming that the pipelines had engaged in behavior that violated state antitrust laws. In particular, respondents alleged that petitioners reported false information to the natural-gas indices on which respondents’ natural-gas contracts were based. The indices affected not only retail natural-gas prices, but also wholesale natural-gas prices. After removing the cases to federal court, the petitioner pipelines sought summary judgment on the ground that the Natural Gas Act pre-empted respondents’ state-law claims. That Act gives the Federal Energy Regulatory Commission (FERC) the authority to determine whether rates charged by natural-gas companies or practices affecting such rates are unreasonable. 15 U. S. C. §717d(a). But it also limits FERC’s jurisdiction to the transportation of natural gas in interstate commerce, the sale in interstate commerce of natural gas for resale, and natural-gas companies engaged in such transportation or sale. §717(b). The Act leaves regulation of other portions of the industry—such as retail sales—to the States. Ibid. The District Court granted petitioners’ motion for summary judgment, reasoning that because petitioners’ challenged practices directly affected wholesale as well as retail prices, they were pre-empted by the Act. The Ninth Circuit reversed. While acknowledging that the pipelines’ index manipulation increased wholesale prices as well as retail prices, it held that the state-law claims were not pre-empted because they were aimed at obtaining damages only for excessively high retail prices. Held: Respondents’ state-law antitrust claims are not within the field of matters pre-empted by the Natural Gas Act. Pp. 10–16. (a) The Act “was drawn with meticulous regard for the continued exercise of state power.” Panhandle Eastern Pipe Line Co. v. Public Serv. Comm’n of Ind., 332 U. S. 507 –518. Where, as here, a practice affects nonjurisdictional as well as jurisdictional sales, pre-emption can be found only where a detailed examination convincingly demonstrates that a matter falls within the pre-empted field as defined by this Court’s precedents. Those precedents emphasize the importance of considering the target at which the state-law claims aim. See, e.g., Northern Natural Gas Co. v. State Corporation Comm’n of Kan., 372 U. S. 84 ; Northwest Central Pipeline Corp. v. State Corporation Comm’n of Kan., 489 U. S. 493 . Here, respondents’ claims are aimed at practices affecting retail prices, a matter “firmly on the States’ side of [the] dividing line.” Id., at 514. Schneidewind v. ANR Pipeline Co., 485 U. S. 293 , is not to the contrary. That opinion explains that the Act does not pre-empt “traditional” state regulation, such as blue sky laws. Id., at 308, n. 11. Antitrust laws, like blue sky laws, are not aimed at natural-gas companies in particular, but rather all businesses in the marketplace. The broad applicability of state antitrust laws supports a finding of no pre-emption here. So, too, does the fact that States have long provided “common-law and statutory remedies against monopolies and unfair business practices,” California v. ARC America Corp., 490 U. S. 93 . As noted earlier, the Act circumscribes FERC’s powers and preserves traditional areas of state authority. §717(b). Pp. 10–14. (b) Neither Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U. S. 354 , nor FPC v. Louisiana Power & Light Co., 406 U. S. 621 , supports petitioners’ position. Mississippi Power is best read as a conflict pre-emption case, not a field pre-emption case. In any event, the state inquiry in Mississippi Power was pre-empted because it was directed at jurisdictional sales in a way that respondents’ state antitrust suits are not. Louisiana Power is also a conflict pre-emption case, and thus does not significantly help petitioners’ field pre-emption argument. Pp. 14–15. (c) Because the parties have not argued conflict pre-emption, questions involving conflicts between state antitrust proceedings and the federal rate-setting process are left for the lower courts to resolve in the first instance. Pp. 15–16. (d) While petitioners and the Government argue that this Court should defer to FERC’s determination that field pre-emption bars respondents’ claims, they fail to point to a specific FERC determination that state antitrust claims fall within the field pre-empted by the Natural Gas Act. Thus, this Court need not consider what legal effect such a determination might have. P. 16. 715 F. 3d 716, affirmed. Breyer, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Alito, Sotomayor, and Kagan, JJ., joined, and in which Thomas, J., joined as to all but Part I–A. Thomas, J., filed an opinion concurring in part and concurring in the judgment. Scalia, J., filed a dissenting opinion, in which Roberts, C. J., joined.

In this case, a group of manufacturers, hospitals, and other institutions that buy natural gas directly from interstate pipelines sued the pipelines, claiming that they engaged in behavior that violated state antitrust laws. The pipelines’ behavior affected both federally regulated wholesale natural-gas prices and nonfederally regulated retail natural-gas prices. The question is whether the federal Natural Gas Act pre-empts these lawsuits. We have said that, in passing the Act, “Congress occupied the field of matters relating to wholesale sales and transportation of natural gas in interstate commerce.” Schneidewind v. ANR Pipeline Co.,485 U. S. 293,305 (1988). Nevertheless, for the reasons given below, we conclude that the Act does not pre-empt the state-law antitrust suits at issue here.IA The Supremacy Clause provides that “the Laws of the United States” (as well as treaties and the Constitution itself ) “shall be the supreme Law of the Land . . . any Thing in the Constitution or Laws of any state to the Contrary notwithstanding.” Art. VI, cl. 2. Congress may consequently pre-empt, i.e., invalidate, a state law through federal legislation. It may do so through express language in a statute. But even where, as here, a statute does not refer expressly to pre-emption, Congress may implicitly pre-empt a state law, rule, or other state action. See Sprietsma v. Mercury Marine,537 U. S. 51,64 (2002). It may do so either through “field” pre-emption or “conflict” pre-emption. As to the former, Congress may have intended “to foreclose any state regulation in the area,” irrespective of whether state law is consistent or inconsistent with “federal standards.” Arizona v. United States, 567 U. S. ___, ___ (2012) (slip op., at 10) (emphasis added). In such situations, Congress has forbidden the State to take action in the field that the federal statute pre-empts. By contrast, conflict pre-emption exists where “compliance with both state and federal law is impossible,” or where “the state law ‘stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.’ ” California v. ARC America Corp.,490 U. S. 93,100,101 (1989). In either situation, federal law must prevail. No one here claims that any relevant federal statute expressly pre-empts state antitrust lawsuits. Nor have the parties argued at any length that these state suits conflict with federal law. Rather, the interstate pipeline companies (petitioners here) argue that Congress implic-itly “ ‘occupied the field of matters relating to wholesale sales and transportation of natural gas in interstate commerce.’ ” Brief for Petitioners 18 (quoting Schneidewind, supra, at 305 (emphasis added)). And they contend that the state antitrust claims advanced by their direct-sales customers (respondents here) fall within that field. The United States, supporting the pipelines, argues similarly. See Brief for United States as Amicus Curiae 15. Since the parties have argued this case almost exclusively in terms of field pre-emption, we consider only the field pre-emption question.B1 Federal regulation of the natural-gas industry began at a time when the industry was divided into three segments. See 1 Regulation of the Natural Gas Industry §1.01 (W. Mogel ed. 2008) (hereinafter Mogel); General Motors Corp. v. Tracy,519 U. S. 278,283 (1997). First, natural-gas producers sunk wells in large oil and gas fields (such as the Permian Basin in Texas and New Mexico). They gathered the gas, brought it to transportation points, and left it to interstate gas pipelines to transport the gas to distant markets. Second, interstate pipelines shipped the gas from the field to cities and towns across the Nation. Third, local gas distributors bought the gas from the interstate pipelines and resold it to business and residential customers within their localities. Originally, the States regulated all three segments of the industry. See 1 Mogel §1.03. But in the early 20th century, this Court held that the Commerce Clause forbids the States to regulate the second part of the business—i.e., the interstate shipment and sale of gas to local distributors for resale. See, e.g., Public Util. Comm’n of R. I. v. Attleboro Steam & Elec. Co.,273 U. S. 83–90 (1927); Missouri ex rel. Barrett v. Kansas Natural Gas Co.,265 U. S. 298–308 (1924). These holdings left a regula-tory gap. Congress enacted the Natural Gas Act,52Stat.821, to fill it. See Phillips Petroleum Co. v. Wisconsin,347 U. S. 672–684, n. 13 (1954) (citing H. R. Rep. No. 709, 75th Cong., 1st Sess., 1–2 (1937); S. Rep. No. 1162, 75th Cong., 1st Sess., 1–2 (1937)). The Act, in §5(a), gives rate-setting authority to the Federal Energy Regulatory Commission (FERC, formerly the Federal Power Commission (FPC)). That authority allows FERC to determine whether “any rate, charge, or classification . . . collected by any natural-gas company in connection with any transportation or sale of natural gas, subject to the jurisdiction of [FERC],” or “any rule, regulation, practice, or contract affecting such rate, charge, or classification is unjust, unreasonable, unduly discriminatory, or preferential.”15 U. S. C. §717d(a) (emphasis added). As the italicized words make clear, §5(a) limits the scope of FERC’s authority to activities “in connection with any transportation or sale of natural gas, subject to the jurisdiction of the Commission.” Ibid. (emphasisadded). And the Act, in §1(b), limits FERC’s “jurisdiction” to (1) “the transportation of natural gas in interstate commerce,” (2) “the sale in interstate commerce of natural gas for resale,” and (3) “natural-gas companies engaged in such transportation or sale.” §717(b). The Act leaves regulation of other portions of the industry—such as production, local distribution facilities, and direct sales—to the States. See Northwest Central Pipeline Corp. v. State Corporation Comm’n of Kan.,489 U. S. 493,507 (1989) (Section 1(b) of the Act “expressly” provides that “States retain jurisdiction over intrastate transportation, local distribution, and distribution facilities, and over ‘the production or gathering of natural gas’ ”). To simplify our discussion, we shall describe the firms that engage in interstate transportation as “jurisdictional sellers” or “interstate pipelines” (though various brokers and others may also fall within the Act’s jurisdictional scope). Similarly, we shall refer to the sales over which FERC has jurisdiction as “jurisdictional sales” or “wholesale sales.”2 Until the 1970’s, natural-gas regulation roughly tracked the industry model we described above. Interstate pipelines would typically buy gas from field producers and resell it to local distribution companies for resale. See Tracy, supra, at 283. FERC (or FPC), acting under the authority of the Natural Gas Act, would set interstate pipeline wholesale rates using classical “cost-of-service” ratemaking methods. See Public Serv. Comm’n of N. Y. v. Mid-Louisiana Gas Co.,463 U. S. 319,328 (1983). That is, FERC would determine a pipeline’s revenue requirement by calculating the costs of providing its services, including operating and maintenance expenses, depreciation expenses, taxes, and a reasonable profit. See FERC, Cost-of-Service Rates Manual 6 (June 1999). FERC would then set wholesale rates at a level designed to meet the pipeline’s revenue requirement. Deregulation of the natural-gas industry, however, brought about changes in FERC’s approach. In the 1950’s, this Court had held that the Natural Gas Act required regulation of prices at the interstate pipelines’ buying end—i.e., the prices at which field producers sold natural gas to interstate pipelines. Phillips Petroleum Co., supra, at 682, 685. By the 1970’s, many in Congress thought that such efforts to regulate field prices had jeopardizednatural-gas supplies in an industry already dependent “on the caprice of nature.” FPC v. Hope Natural Gas Co.,320 U. S. 591,630 (1944) (opinion of Jackson, J.); see id., at 629 (recognizing that “the wealth of Midas and the wit of man cannot produce . . . a natural gas field”). Hoping to avoid future shortages, Congress enacted forms of field price deregulation designed to rely upon competition, rather than regulation, to keep field prices low. See, e.g., Natural Gas Policy Act of 1978,92Stat.3409, codified in part at15 U. S. C. §3301 et seq. (phasing out regulation of wellhead prices charged by producers of natural gas); Natural Gas Wellhead Decontrol Act of 1989,103Stat.157 (removing price controls on wellhead sales as of January 1993). FERC promulgated new regulations designed to further this process of deregulation. See, e.g., Regulation of Natural Gas Pipelines after Partial Wellhead Decontrol, 50 Fed. Reg. 42408 (1985) (allowing “open access” to pipelines so that consumers could pay to ship their own gas). Most important here, FERC adopted an approach that relied on the competitive marketplace, rather than classical regulatory rate-setting, as the main mechanism for keeping wholesale natural-gas rates at a reasonable level. Order No. 636, issued in 1992, allowed FERC to issue blanket certificates that permitted jurisdictional sellers (typically interstate pipelines) to charge market-based rates for gas, provided that FERC had first determined that the sellers lacked market power. See 57 Fed. Reg. 57957–57958 (1992); id., at 13270. After the issuance of this order, FERC’s oversight of the natural-gas market largely consisted of (1) ex ante examinations of jurisdictional sellers’ market power, and (2) the availability of a complaint process under §717d(a). See Brief for United States as Amicus Curiae 4. The new system also led many large gas consumers—such as industrial and commercial users—to buy their own gas directly from gas producers, and to arrange (and often pay separately) for transportation from the field to the place of consumption. See Tracy, 519 U. S., at 284. Insofar as interstate pipelines sold gas to such consumers, they sold it for direct consumption rather than resale.3 The free-market system for setting interstate pipeline rates turned out to be less than perfect. Interstate pipelines, distributing companies, and many of the customers who bought directly from the pipelines found that they had to rely on privately published price indices to determine appropriate prices for their natural-gas contracts. These indices listed the prices at which natural gas was being sold in different (presumably competitive) markets across the country. The information on which these in-dices were based was voluntarily reported by natural-gas traders. In 2003, FERC found that the indices were inaccurate, in part because much of the information that natural-gas traders reported had been false. See FERC, Final Report on Price Manipulation in Western Markets (Mar. 2003), App. 88–89. FERC found that false reporting had involved “inflating the volume of trades, omitting trades, and adjusting the price of trades.” Id., at 88. That is, sometimes those who reported information simply fabricated it. Other times, the information reported reflected “wash trades,” i.e., “prearranged pair[s] of trades of the same good between the same parties, involving no economic risk and no net change in beneficial ownership.” Id., at 215. FERC concluded that these “efforts to manipulate price indices compiled by trade publications” had helped raise “to extraordinary levels” the prices of both jurisdictional sales (that is, interstate pipeline sales for resale) and nonjurisdictional direct sales to ultimate consumers. Id., at 86, 85. After issuing its final report on price manipulation in western markets, FERC issued a Code of Conduct. That code amended all blanket certificates to prohibit jurisdictional sellers “from engaging in actions without a legitimate business purpose that manipulate or attempt to manipulate market conditions, including wash trades and collusion.” 68 Fed. Reg. 66324 (2003). The code also required jurisdictional companies, when they provided information to natural-gas index publishers, to “provide accurate and factual information, and not knowingly submit false or misleading information or omit material information to any such publisher.” Id., at 66337. At the same time, FERC issued a policy statement setting forth “minimum standards for creation and publication of any energy price index,” and “for reporting transaction data to index developers.” Price Discovery in Natural Gas and Elec. Markets, 104 FERC ¶61,121, pp. 61,407, 61,408 (2003). Finally, FERC, after finding that certain jurisdictional sellers had “engaged in wash trading . . . that resulted in the manipulation of [natural-gas] prices,” terminated those sellers’ blanket marketing certificates. Enron Power Marketing, Inc., 103 FERC ¶61,343, p. 62,303 (2003). Congress also took steps to address these problems. In particular, it passed the Energy Policy Act of 2005,119Stat.594, which gives FERC the authority to issue rules and regulations to prevent “any manipulative or deceptive device or contrivance” by “any entity . . . in connection with the purchase or sale of natural gas or the purchase or sale of transportation services subject to the jurisdiction of” FERC,15 U. S. C. §717c–1.C We now turn to the cases before us. Respondents, as we have said, bought large quantities of natural gas directly from interstate pipelines for their own consumption. They believe that they overpaid in these transactions due to the interstate pipelines’ manipulation of the natural-gas indices. Based on this belief, they filed state-law antitrust suits against petitioners in state and federal courts. See App. 244–246 (alleging violations of Wis. Stat. §§133.03, 133.14, 133.18); see also App. 430–433 (same); id., at 519–521 (same); id., at 362–364 (alleging violations of Kansas Restraint of Trade Act, Kan. Stat. Ann. §50–101 et seq.); App. 417–419 (alleging violations of Missouri Antitrust Law, Mo. Rev. Stat. §§416.011–416.161). The pipelines removed all the state cases to federal court, where they were consolidated and sent for pretrial proceedings to the Federal District Court for the District of Nevada. See28 U. S. C. §1407. The pipelines then moved for summary judgment on the ground that the Natural Gas Act pre-empted respondents’ state-law antitrust claims. The District Court granted their motion. It concluded that the pipelines were “jurisdictional sellers,” i.e., “natural gas companies engaged in” the “transportation of natural gas in interstate commerce.” Order in No. 03–cv–1431 (D Nev., July 18, 2011), pp. 4, 11. And it held that respondents’ claims, which were “aimed at” these sellers’ “alleged practices of false price reporting, wash trades, and anticompetitive collusive behavior” were pre-empted because “such practices,” not only affected nonjurisdictional direct-sale prices but also “directly affect[ed]” jurisdictional (i.e., wholesale) rates. Id., at 36–37. The Ninth Circuit reversed. It emphasized that the price-manipulation of which respondents complained affected not only jurisdictional (i.e., wholesale) sales, but also nonjurisdictional (i.e., retail) sales. The court construed the Natural Gas Act’s pre-emptive scope narrowly in light of Congress’ intent—manifested in §1(b) of the Act—to preserve for the States the authority to regulate nonjurisdictional sales. And it held that the Act did not pre-empt state-law claims aimed at obtaining damages for excessively high retail natural-gas prices stemming from interstate pipelines’ price manipulation, even if the manipulation raised wholesale rates as well. See In re Western States Wholesale Natural Gas Antitrust Litigation, 715 F. 3d 716, 729–736 (2013). The pipelines sought certiorari. They asked us to resolve confusion in the lower courts as to whether the Natural Gas Act pre-empts retail customers’ state antitrust law challenges to practices that also affect wholesale rates. Compare id., at 729–736, with Leggett v. Duke Energy Corp., 308 S. W. 3d 843 (Tenn. 2010). We granted the petition.II Petitioners, supported by the United States, argue that their customers’ state antitrust lawsuits are within the field that the Natural Gas Act pre-empts. See Brief for Petitioners 18 (citing Schneidewind, 485 U. S., at 305); Brief for United States as Amicus Curiae 13 (same). They point out that respondents’ antitrust claims target anticompetitive activities that affected wholesale (as well as retail) rates. See Brief for Petitioners 2. They add that the Natural Gas Act expressly grants FERC authority to keep wholesale rates at reasonable levels. See ibid. (citing 15 U. S. C. §§717(b), 717d(a)). In exercising this authority, FERC has prohibited the very kind of anticompetitive conduct that the state actions attack. See Part I–B–3, supra. And, petitioners contend, letting these actions proceed will permit state antitrust courts to reach conclusions about that conduct that differ from those that FERC might reach or has already reached. Accordingly, petitioners argue, respondents’ state-law antitrust suits fall within the pre-empted field.A Petitioners’ arguments are forceful, but we cannot accept their conclusion. As we have repeatedly stressed, the Natural Gas Act “was drawn with meticulous regard for the continued exercise of state power, not to handicap or dilute it in any way.” Panhandle Eastern Pipe Line Co. v. Public Serv. Comm’n of Ind.,332 U. S. 507–518 (1947); see also Northwest Central, 489 U. S., at 511 (the “legislative history of the [Act] is replete with assurances that the Act ‘takes nothing from the State [regulatory] commissions’ ” (quoting 81 Cong. Rec. 6721 (1937))). Accordingly, where (as here) a state law can be applied to nonjurisdictional as well as jurisdictional sales, we must proceed cautiously, finding pre-emption only where detailed examination convinces us that a matter falls within the pre-empted field as defined by our precedents. See Panhandle Eastern, supra, at 516–518; Interstate Natural Gas Co. v. FPC,331 U. S. 682–693 (1947). Those precedents emphasize the importance of considering the target at which the state law aims in determining whether that law is pre-empted. For example, in Northern Natural Gas Co. v. State Corporation Comm’n of Kan.,372 U. S. 84 (1963), the Court said that it had “consistently recognized” that the “significant distinction” for purposes of pre-emption in the natural-gas context is the distinction between “measures aimed directly at interstate purchasers and wholesales for resale, and those aimed at” subjects left to the States to regulate. Id., at 94 (emphasis added). And, in Northwest Central, the Court found that the Natural Gas Act did not pre-empt a state regulation concerning the timing of gas production from a gas field within the State, even though the regulation might have affected the costs of and the prices of interstate wholesale sales, i.e., jurisdictional sales. 489 U. S., at 514. In reaching this conclusion, the Court explained that the state regulation aimed primarily at “protect[ing] producers’ . . . rights—a matter firmly on the States’ side of that dividing line.” Ibid. The Court contrasted this state regulation with the state orders at issue in Northern Natural, which “ ‘inva-lidly invade[d] the federal agency’s exclusive domain’ pre-cisely because” they were “ ‘unmistakably and unambiguously directed at purchasers.’ ” Id., at 513 (quoting Northern Natural, supra, at 92; emphasis added). Here, too, the lawsuits are directed at practices affecting retail rates–which are “firmly on the States’ side of that dividing line.” Petitioners argue that Schneidewind constitutes con-trary authority. In that case, the Court found pre-empted a state law that required public utilities, such as interstate pipelines crossing the State, to obtain state approval before issuing long-term securities. 485 U. S., at 306–309. But the Court there thought that the State’s securities regulation was aimed directly at interstate pipelines. It wrote that the state law was designed to keep “a natural gas company from raising its equity levels above a certain point” in order to keep the company’s revenue requirement low, thereby ensuring lower wholesale rates. Id., at 307–308. Indeed, the Court expressly said that the state law was pre-empted because it was “directed at . . . the control of rates and facilities of natural gas companies,” “precisely the things over which FERC has comprehensive author-ity.” Id., at 308 (emphasis added). The dissent rejects the notion that the proper test for purposes of pre-emption in the natural gas context is whether the challenged measures are “aimed directly at interstate purchasers and wholesales for resale” or not. Northern Natural, supra, at 94. It argues that this approach is “unprecedented,” and that the Court’s focus should be on “what the State seeks to regulate . . . , not why the State seeks to regulate it.” Post, at 6 (opinion of Scalia, J.). But the “target” to which our cases refer must mean more than just the physical activity that a State regulates. After all, a single physical action, such as reporting a price to a specialized journal, could be the subject of many different laws—including tax laws, disclosure laws, and others. To repeat the point we made above, no one could claim that FERC’s regulation of this physical activity for purposes of wholesale rates forecloses every other form of state regulation that affects those rates. Indeed, although the dissent argues that Schneidwind created a definitive test for pre-emption in the natural gas context that turns on whether “the matter on which the State asserts the right to act is in any way regulated by the Federal Act,” post, at 3 (quoting 485 U. S., at 310, n. 13), Schneidewind could not mean this statement as an absolute test. It goes on to explain that the Natural Gas Act does not pre-empt “traditional” state regulation, such as state blue sky laws (which, of course, raise wholesale—as well as retail—investment costs). Id., at 308, n. 11. Antitrust laws, like blue sky laws, are not aimed at natural-gas companies in particular, but rather all businesses in the marketplace. See ibid. They are far broader in their application than, for example, the regulations at issue in Northern Natural, which applied only to entities buying gas from fields within the State. See 372 U. S., at 85–86, n. 1; contra, post, at 5–6 (stating that Northern Natural concerned “background market conditions”). This broad applicability of state antitrust law supports a finding of no pre-emption here. Petitioners and the dissent argue that there is, or should be, a clear division between areas of state and federal authority in natural-gas regulation. See Brief for Petitioners 18; post, at 7. But that Platonic ideal does not describe the natural gas regulatory world. Suppose FERC, when setting wholesale rates in the former cost-of-service rate-making days, had denied cost recovery for pipelines’ failure to recycle. Would that fact deny States the power to enact and apply recycling laws? These state laws might well raise pipelines’ operating costs, and thus the costs of wholesale natural gas transportation. But in Northwest Central we said that “[t]o find field pre-emption of [state] regulation merely because purchasers’ costs and hence rates might be affected would be largely to nullify . . . §1(b).” 489 U. S., at 514. The dissent barely mentions the limitations on FERC’s powers in §1(b), but the enumeration of FERC’s powers in §5(a) is circumscribed by a reference back to the limitations in §1(b). See post, at 1–3. As we explained above, see Part I–B–1, supra, those limits are key to understanding the careful balance between federal and state regulation that Congress struck when it passed the Natural Gas Act. That Act “was drawn with meticulous regard for the continued exercise of state power, not to handicap or dilute it in any way.” Panhandle Eastern, 332 U. S., at 517–518. Contra, post, at 8. States have a “long history of” providing “common-law and statutory remedies against monopolies and unfair business practices.” ARC America, 490 U. S., at 101; see also Watson v. Buck,313 U. S. 387,404 (1941) (noting the States’ “long-recognized power to regulate combinations in restraint of trade”). Respondents’ state-law antitrust suits relied on this well established state power.B Petitioners point to two other cases that they believe support their position. The first is Mississippi Power & Light Co. v. Mississippi ex rel. Moore,487 U. S. 354 (1988). There, the Court held that the Federal Power Act—which gives FERC the authority to determine whether rates charged by public utilities in electric energy sales are “just and reasonable,”16 U. S. C. §824d(a)—pre-empted a state inquiry into the reasonableness of FERC-approved prices for the sale of nuclear power to wholesalers of electricity (which led to higher retail electricity rates). 487 U. S., at 373–377. Petitioners argue that this case shows that state regulation of similar sales here—i.e., by a pipeline to a direct consumer—must also be pre-empted. See Reply Brief 11–12. Mississippi Power, however, is best read as a conflict pre-emption case, not a field pre-emption case. See 487 U. S., at 377 (“[A] state agency’s ‘efforts to regulate commerce must fall when they conflict with or interfere with federal authority over the same activity’ ” (quoting Chicago & North Western Transp. Co. v. Kalo Brick & Tile Co.,450 U. S. 311–319 (1981))). Regardless, the state inquiry in Mississippi Power was pre-empted because it was directed at jurisdictional sales in a way that respondents’ state antitrust lawsuits are not. Mississippi’s inquiry into the reasonableness of FERC-approved purchases was effectively an attempt to “regulate in areas where FERC has properly exercised its jurisdiction to determine just and reasonable wholesale rates.” 487 U. S., at 374. By contrast, respondents’ state antitrust lawsuits do not seek to challenge the reason-ableness of any rates expressly approved by FERC. Rather, they seek to challenge the background marketplace conditions that affected both jurisdictional and nonjurisdic-tional rates. Petitioners additionally point to FPC v. Louisiana Power & Light Co.,406 U. S. 621 (1972). In that case, the Court held that federal law gave FPC the authority to allocate natural gas during shortages by ordering interstate pipelines to curtail gas deliveries to all customers, including retail customers. This latter fact, the pipelines argue, shows that FERC has authority to regulate index manipulation insofar as that manipulation affects retail (as well as wholesale) sales. Brief for Petitioners 26. Accordingly, they contend that state laws that aim at this same subject are pre-empted. This argument, however, makes too much of too little. The Court’s finding of pre-emption in Louisiana Power rested on its belief that the state laws in question con-flicted with federal law. The Court concluded that “FPC has authority to effect orderly curtailment plans involving both direct sales and sales for resale,” 406 U. S., at 631, because otherwise there would be “unavoidable conflict between” state regulation of direct sales and the “uniform federal regulation” that the Natural Gas Act foresees, id., at 633–635. Conflict pre-emption may, of course, invalidate a state law even though field pre-emption does not. Because petitioners have not argued this case as a conflict pre-emption case, Louisiana Power does not offer them significant help.C To the extent any conflicts arise between state antitrust law proceedings and the federal rate-setting process, the doctrine of conflict pre-emption should prove sufficient to address them. But as we have noted, see Part I–A, supra, the parties have not argued conflict pre-emption. See also, e.g., Tr. of Oral Arg. 24 (Solicitor General agrees that he has not “analyzed this [case] under a conflict preemption regime”). We consequently leave conflict pre-emption questions for the lower courts to resolve in the firstinstance.D We note that petitioners and the Solicitor General have argued that we should defer to FERC’s determination that field pre-emption bars the respondents’ claims. See Brief for Petitioners 22 (citing Arlington v. FCC, 569 U. S. ___, ___–___ (2013) (slip op., at 10–14); Brief for United States as Amicus Curiae 32 (same). But they have not pointed to a specific FERC determination that state antitrust claims fall within the field pre-empted by the Natural Gas Act. Rather, they point only to the fact that FERC has promulgated detailed rules governing manipulation of price indices. Because there is no determination by FERC that its regulation pre-empts the field into which respondents’ state-law antitrust suits fall, we need not consider what legal effect such a determination might have. And we conclude that the detailed federal regulations here do not offset the other considerations that weigh against a finding of pre-emption in this context.* * * For these reasons, the judgment of the Court of Appeals for the Ninth Circuit is affirmed.It is so ordered.

575.US.2014_13-1041

The Administrative Procedure Act (APA) establishes the procedures federal administrative agencies use for “rule making,” defined as the process of “formulating, amending, or repealing a rule.” 5 U. S. C. §551(5). The APA distinguishes between two types of rules: So-called “legislative rules” are issued through notice-and-comment rulemaking, see §§553(b), (c), and have the “force and effect of law,” Chrysler Corp. v. Brown, 441 U. S. 281 –303. “Interpretive rules,” by contrast, are “issued . . . to advise the public of the agency’s construction of the statutes and rules which it administers,” Shalala v. Guernsey Memorial Hospital, 514 U. S. 87 , do not require notice-and-comment rulemaking, and “do not have the force and effect of law,” ibid. In 1999 and 2001, the Department of Labor’s Wage and Hour Division issued letters opining that mortgage-loan officers do not qualify for the administrative exemption to overtime pay requirements under the Fair Labor Standards Act of 1938. In 2004, the Department issued new regulations regarding the exemption. Respondent Mortgage Bankers Association (MBA) requested a new interpretation of the revised regulations as they applied to mortgage-loan officers, and in 2006, the Wage and Hour Division issued an opinion letter finding that mortgage-loan officers fell within the administrative exemption under the 2004 regulations. In 2010, the Department again altered its interpretation of the administrative exemption. Without notice or an opportunity for comment, the Department withdrew the 2006 opinion letter and issued an Administrator’s Interpretation concluding that mortgage-loan officers do not qualify for the administrative exemption. MBA filed suit contending, as relevant here, that the Administrator’s Interpretation was procedurally invalid under the D. C. Circuit’s decision in Paralyzed Veterans of Am. v. D. C. Arena L. P., 117 F. 3d 579. The Paralyzed Veterans doctrine holds that an agency must use the APA’s notice-and-comment procedures when it wishes to issue a new interpretation of a regulation that deviates significantly from a previously adopted interpretation. The District Court granted summary judgment to the Department, but the D. C. Circuit applied Paralyzed Veterans and reversed. Held: The Paralyzed Veterans doctrine is contrary to the clear text of the APA’s rulemaking provisions and improperly imposes on agencies an obligation beyond the APA’s maximum procedural requirements. Pp. 6–14. (a) The APA’s categorical exemption of interpretive rules from the notice-and-comment process is fatal to the Paralyzed Veterans doctrine. The D. C. Circuit’s reading of the APA conflates the differing purposes of §§1 and 4 of the Act. Section 1 requires agencies to use the same procedures when they amend or repeal a rule as they used to issue the rule, see 5 U. S. C. §551(5), but it does not say what procedures an agency must use when it engages in rulemaking. That is the purpose of §4. And §4 specifically exempts interpretive rules from notice-and-comment requirements. Because an agency is not required to use notice-and-comment procedures to issue an initial interpretive rule, it is also not required to use those procedures to amend or repeal that rule. Pp. 7–8. (b) This straightforward reading of the APA harmonizes with longstanding principles of this Court’s administrative law jurisprudence, which has consistently held that the APA “sets forth the full extent of judicial authority to review executive agency action for procedural correctness,” FCC v. Fox Television Stations, Inc., 556 U. S. 502 . The APA’s rulemaking provisions are no exception: §4 establishes “the maximum procedural requirements” that courts may impose upon agencies engaged in rulemaking. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U. S. 519 . By mandating notice-and-comment procedures when an agency changes its interpretation of one of the regulations it enforces, Paralyzed Veterans creates a judge-made procedural right that is inconsistent with Congress’ standards. Pp. 8–9. (c) MBA’s reasons for upholding the Paralyzed Veterans doctrine are unpersuasive. Pp. 9–14. (1) MBA asserts that an agency interpretation of a regulation that significantly alters the agency’s prior interpretation effectively amends the underlying regulation. That assertion conflicts with the ordinary meaning of the words “amend” and “interpret,” and it is impossible to reconcile with the longstanding recognition that interpretive rules do not have the force and effect of law. MBA’s theory is particularly odd in light of the limitations of the Paralyzed Veterans doctrine, which applies only when an agency has previously adopted an interpretation of its regulation. MBA fails to explain why its argument regarding revised interpretations should not also extend to the agency’s first interpretation. Christensen v. Harris County, 529 U. S. 576 , and Shalala v. Guernsey Memorial Hospital, 514 U. S. 87 , distinguished. Pp. 9–12. (2) MBA also contends that the Paralyzed Veterans doctrine reinforces the APA’s goal of procedural fairness. But the APA already provides recourse to regulated entities from agency decisions that skirt notice-and-comment provisions by placing a variety of constraints on agency decisionmaking, e.g., the arbitrary and capricious standard. In addition, Congress may include safe-harbor provisions in legislation to shelter regulated entities from liability when they rely on previous agency interpretations. See, e.g., 29 U. S. C. §§259(a), (b)(1). Pp. 12–13. (3) MBA has waived its argument that the 2010 Administrator’s Interpretation should be classified as a legislative rule. From the beginning, this suit has been litigated on the understanding that the Administrator’s Interpretation is an interpretive rule. Neither the District Court nor the Court of Appeals addressed this argument below, and MBA did not raise it here in opposing certiorari. P. 14. 720 F. 3d 966, reversed. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined, and in which Alito, J., joined except for Part III–B. Alito, J., filed an opinion concurring in part and concurring in the judgment. Scalia, J., and Thomas, J., filed opinions concurring in the judgment.Notes 1 Together with No. 13–1052, Nickols et al. v. Mortgage Bankers Association, also on certiorari to the same court.

When a federal administrative agency first issues a rule interpreting one of its regulations, it is generally not required to follow the notice-and-comment rulemaking procedures of the Administrative Procedure Act (APA or Act). See5 U. S. C. §553(b)(A). The United States Court of Appeals for the District of Columbia Circuit has nevertheless held, in a line of cases beginning with Paralyzed Veterans of Am. v. D. C. Arena L. P., 117 F. 3d 579 (1997), that an agency must use the APA’s notice-and-comment procedures when it wishes to issue a new interpretation of a regulation that deviates significantly from one the agency has previously adopted. The question in these cases is whether the rule announced in Paralyzed Veterans is consistent with the APA. We hold that it is not.IA The APA establishes the procedures federal administrative agencies use for “rule making,” defined as the process of “formulating, amending, or repealing a rule.” §551(5). “Rule,” in turn, is defined broadly to include “statement[s] of general or particular applicability and future effect” that are designed to “implement, interpret, or prescribe law or policy.” §551(4). Section 4 of the APA,5 U. S. C. §553, prescribes a three-step procedure for so-called “notice-and-comment rulemaking.” First, the agency must issue a “[g]eneral notice of proposed rule making,” ordinarily by publication in the Federal Register. §553(b). Second, if “notice [is] required,” the agency must “give interested persons an opportunity to participate in the rule making through submission of written data, views, or arguments.” §553(c). An agency must consider and respond to significant comments received during the period for public comment. See Citizens to Preserve Overton Park, Inc. v. Volpe,401 U. S. 402,416 (1971); Thompson v. Clark, 741 F. 2d 401, 408 (CADC 1984). Third, when the agency promulgates the final rule, it must include in the rule’s text “a concise general statement of [its] basis and purpose.” §553(c). Rules issued through the notice-and-comment process are often referred to as “legislative rules” because they have the “force and effect of law.” Chrysler Corp. v. Brown,441 U. S. 281–303 (1979) (internal quotation marks omitted). Not all “rules” must be issued through the notice-and-comment process. Section 4(b)(A) of the APA provides that, unless another statute states otherwise, the notice-and-comment requirement “does not apply” to “interpretative rules, general statements of policy, or rules of agency organization, procedure, or practice.”5 U. S. C. §553(b)(A). The term “interpretative rule,” or “interpretive rule,”[1] is not further defined by the APA, and its precise meaning is the source of much scholarly and judicial debate. See generally Pierce, Distinguishing Legislative Rules From Interpretative Rules, 52 Admin. L. Rev. 547 (2000); Manning, Nonlegislative Rules, 72 Geo. Wash. L. Rev. 893 (2004). We need not, and do not, wade into that debate here. For our purposes, it suffices to say that the critical feature of interpretive rules is that they are “issued by an agency to advise the public of the agency’s construction of the statutes and rules which it administers.” Shalala v. Guernsey Memorial Hospital,514 U. S. 87,99 (1995) (internal quotation marks omitted). The absence of a notice-and-comment obligation makes the process of issuing interpretive rules comparatively easier for agencies than issuing legislative rules. But that convenience comes at a price: Interpretive rules “do not have the force and effect of law and are not accorded that weight in the adjudicatory process.” Ibid.B These cases began as a dispute over efforts by the Department of Labor to determine whether mortgage-loan officers are covered by the Fair Labor Standards Act of 1938 (FLSA),52Stat.1060, as amended,29 U. S. C. §201 et seq. The FLSA “establishe[s] a minimum wage and overtime compensation for each hour worked in excess of 40 hours in each workweek” for many employees. Integ-rity Staffing Solutions, Inc. v. Busk, 574 U. S. ___, ___ (2014) (slip op., at 3). Certain classes of employees, however, are exempt from these provisions. Among these exempt individuals are those “employed in a bona fide executive, administrative, or professional capacity . . . or in the capacity of outside salesman . . . .” §213(a)(1). The exemption for such employees is known as the “administrative” exemption. The FLSA grants the Secretary of Labor authority to “defin[e]” and “delimi[t]” the categories of exempt admin-istrative employees. Ibid. The Secretary’s current regu-lations regarding the administrative exemption were promulgated in 2004 through a notice-and-commentrulemaking. As relevant here, the 2004 regulations differed from the previous regulations in that they contained a new section providing several examples of exempt administrative employees. See 29 CFR §541.203. One of the examples is “[e]mployees in the financial services industry,” who, depending on the nature of their day-to-day work, “generally meet the duties requirements for the administrative exception.” §541.203(b). The financial services example ends with a caveat, noting that “an employee whose primary duty is selling financial products does not qualify for the administrative exemption.” Ibid. In 1999 and again in 2001, the Department’s Wage and Hour Division issued letters opining that mortgage-loan officers do not qualify for the administrative exemption. See Opinion Letter, Loan Officers/Exempt Status, 6A LRR, Wages and Hours Manual 99:8351 (Feb. 16, 2001); Opinion Letter, Mortgage Loan Officers/Exempt Status, id., at 99:8249. (May 17, 1999). In other words, the Department concluded that the FLSA’s minimum wage and maximum hour requirements applied to mortgage-loan officers. When the Department promulgated its current FLSA regulations in 2004, respondent Mortgage Bankers Association (MBA), a national trade association representing real estate finance companies, requested a new opinion interpreting the revised regulations. In 2006, the Department issued an opinion letter finding that mortgage-loan officers fell within the administrative exemption under the 2004 regulations. See App. to Pet. for Cert. in No. 13–1041, pp. 70a–84a. Four years later, however, the Wage and Hour Division again altered its interpretation of the FLSA’s administrative exemption as it applied to mortgage-loan officers. Id., at 49a–69a. Reviewing the provisions of the 2004 regulations and judicial decisions addressing the administrative exemption, the Department’s 2010 Administrator’s Interpretation concluded that mortgage-loan officers “have a primary duty of making sales for their employers, and, therefore, do not qualify” for the administrative exemption. Id., at 49a, 69a. The Department accordingly withdrew its 2006 opinion letter, which it now viewed as relying on “misleading assumption[s] and selective and narrow analysis” of the exemption example in §541.203(b). Id., at 68a. Like the 1999, 2001, and 2006 opinion letters, the 2010 Administrator’s Interpretation was issued without notice or an opportunity for comment.C MBA filed a complaint in Federal District Court challenging the Administrator’s Interpretation. MBA contended that the document was inconsistent with the 2004 regulation it purported to interpret, and thus arbitrary and capricious in violation of §10 of the APA,5 U. S. C. §706. More pertinent to this case, MBA also argued that the Administrator’s Interpretation was procedurally in-valid in light of the D. C. Circuit’s decision in Paralyzed Veterans, 117 F. 3d 579. Under the Paralyzed Veterans doctrine, if “an agency has given its regulation a definitive interpretation, and later significantly revises that interpretation, the agency has in effect amended its rule, something it may not accomplish” under the APA “without notice and comment.” Alaska Professional Hunters Assn., Inc. v. FAA, 177 F. 3d 1030, 1034 (CADC 1999). Three former mortgage-loan officers—Beverly Buck, Ryan Henry, and Jerome Nickols—subsequently intervened in thecase to defend the Administrator’s Interpretation.[2] The District Court granted summary judgment to the Department. Mortgage Bankers Assn. v. Solis, 864 F. Supp. 2d 193 (DC 2012). Though it accepted the parties’ characterization of the Administrator’s Interpretation as an interpretive rule, id., at 203, n. 7, the District Court determined that the Paralyzed Veterans doctrine was inapplicable because MBA had failed to establish its reliance on the contrary interpretation expressed in the Department’s 2006 opinion letter. The Administrator’s Interpretation, the District Court further determined, was fully supported by the text of the 2004 FLSA regulations. The court accordingly held that the 2010 interpretation was not arbitrary or capricious.[3] The D. C. Circuit reversed. Mortgage Bankers Assn. v. Harris, 720 F. 3d 966 (2013). Bound to the rule of Paralyzed Veterans by precedent, the Court of Appeals rejected the Government’s call to abandon the doctrine. 720 F. 3d., at 967, n. 1. In the court’s view, “[t]he only question” properly before it was whether the District Court had erred in requiring MBA to prove that it relied on the Department’s prior interpretation. Id., at 967. Explaining that reliance was not a required element of the Paralyzed Veterans doctrine, and noting the Department’s concession that a prior, conflicting interpretation of the 2004 regulations existed, the D. C. Circuit concluded that the 2010 Administrator’s Interpretation had to be vacated. We granted certiorari, 573 U. S. __ (2014), and now reverse.II The Paralyzed Veterans doctrine is contrary to the clear text of the APA’s rulemaking provisions, and it improperly imposes on agencies an obligation beyond the “maximum procedural requirements” specified in the APA, Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc.,435 U. S. 519,524 (1978).A The text of the APA answers the question presented. Section 4 of the APA provides that “notice of proposed rule making shall be published in the Federal Register.”5 U. S. C. §553(b). When such notice is required by the APA, “the agency shall give interested persons an opportunity to participate in the rule making.” §553(c). But §4 further states that unless “notice or hearing is required by statute,” the Act’s notice-and-comment requirement “does not apply . . . to interpretative rules.” §553(b)(A). This exemption of interpretive rules from the notice-and-comment process is categorical, and it is fatal to the rule announced in Paralyzed Veterans. Rather than examining the exemption for interpretive rules contained in §4(b)(A) of the APA, the D. C. Circuit in Paralyzed Veterans focused its attention on §1 of the Act. That section defines “rule making” to include not only the initial issuance of new rules, but also “repeal[s]” or “amend[ments]” of existing rules. See §551(5). Because notice-and-comment requirements may apply even to these later agency actions, the court reasoned, “allow[ing] an agency to make a fundamental change in its interpretation of a substantive regulation without notice and comment” would undermine the APA’s procedural framework. 117 F. 3d, at 586. This reading of the APA conflates the differing purposes of §§1 and 4 of the Act. Section 1 defines what a rule-making is. It does not, however, say what procedures an agency must use when it engages in rulemaking. That is the purpose of §4. And §4 specifically exempts interpretive rules from the notice-and-comment requirements that apply to legislative rules. So, the D. C. Circuit correctly read §1 of the APA to mandate that agencies use the same procedures when they amend or repeal a rule as they used to issue the rule in the first instance. See FCC v. Fox Television Stations, Inc.,556 U. S. 502,515 (2009) (the APA “make[s] no distinction . . . between initial agency action and subsequent agency action undoing or revising that action”). Where the court went wrong was in failing to apply that accurate understanding of §1 to the exemption for interpretive rules contained in §4: Because an agency is not required to use notice-and-comment procedures to issue an initial interpretive rule, it is also not required to use those procedures when it amends or repeals that interpretive rule.B The straightforward reading of the APA we now adopt harmonizes with longstanding principles of our administrative law jurisprudence. Time and again, we have reiterated that the APA “sets forth the full extent of judicial authority to review executive agency action for procedural correctness.” Fox Television Stations, Inc., 556 U. S., at 513. Beyond the APA’s minimum requirements, courts lack authority “to impose upon [an] agency its own notion of which procedures are ‘best’ or most likely to further some vague, undefined public good.” Vermont Yankee, 435 U. S., at 549. To do otherwise would violate “the very basic tenet of administrative law that agencies should be free to fashion their own rules of procedure.” Id., at 544. These foundational principles apply with equal force to the APA’s procedures for rulemaking. We explained in Vermont Yankee that §4 of the Act “established the maximum procedural requirements which Congress was willing to have the courts impose upon agencies in conducting rulemaking procedures.” Id., at 524. “Agencies are free to grant additional procedural rights in the exercise of their discretion, but reviewing courts are generally not free to impose them if the agencies have not chosen to grant them.” Ibid. The Paralyzed Veterans doctrine creates just such a judge-made procedural right: the right to notice and an opportunity to comment when an agency changes its interpretation of one of the regulations it enforces. That requirement may be wise policy. Or it may not. Regardless, imposing such an obligation is the responsibility of Congress or the administrative agencies, not the courts. We trust that Congress weighed the costs and benefits of placing more rigorous procedural restrictions on the issuance of interpretive rules. See id., at 523 (when Congress enacted the APA, it “settled long-continued and hard-fought contentions, and enact[ed] a formula upon which opposing social and political forces have come to rest” (internal quotation marks omitted)). In the end, Congress decided to adopt standards that permit agencies to promulgate freely such rules—whether or not they are consistent with earlier interpretations. That the D. C. Circuit would have struck the balance differently does not permit that court or this one to overturn Congress’ contrary judgment. Cf. Law v. Siegel, 571 U. S. ___, ___ (2014) (slip op., at 11).III MBA offers several reasons why the Paralyzed Veterans doctrine should be upheld. They are not persuasive.A MBA begins its defense of the Paralyzed Veterans doctrine by attempting to bolster the D. C. Circuit’s reading of the APA. “Paralyzed Veterans,” MBA contends, “simply acknowledges the reality that where an agency significantly alters a prior, definitive interpretation of a regulation, it has effectively amended the regulation itself,” something that under the APA requires use of notice-and-comment procedures. Brief for Respondent 20–21. The act of “amending,” however, in both ordinary parlance and legal usage, has its own meaning separate and apart from the act of “interpreting.” Compare Black’s Law Dictionary 98 (10th ed. 2014) (defining “amend” as “[t]o change the wording of” or “formally alter . . . by striking out, inserting, or substituting words”), with id., at 943 (defining “interpret” as “[t]o ascertain the meaning and significance of thoughts expressed in words”). One would not normally say that a court “amends” a statute when it interprets its text. So too can an agency “interpret” a regulation without “effectively amend[ing]” the underlying source of law. MBA does not explain how, precisely, an interpretive rule changes the regulation it interprets, and its assertion is impossible to reconcile with the longstanding recognition that interpretive rules do not have the force and effect of law. See Chrysler Corp., 441 U. S., at 302, n. 31 (citing Attorney General’s Manual on the Administrative Procedure Act 30, n. 3 (1947)); Skidmore v. Swift & Co.,323 U. S. 134,140 (1944). MBA’s “interpretation-as-amendment” theory is particularly odd in light of the limitations of the Paralyzed Veterans doctrine. Recall that the rule of Paralyzed Veterans applies only when an agency has previously adopted an interpretation of its regulation. Yet in that initial interpretation as much as all that come after, the agency is giving a definite meaning to an ambiguous text—the very act MBA insists requires notice and comment. MBA is unable to say why its arguments regarding revised interpretations should not also extend to the agency’s first interpretation.[4] Next, MBA argues that the Paralyzed Veterans doctrine is more consistent with this Court’s “functional” approach to interpreting the APA. Relying on Christensen v. Harris County,529 U. S. 576 (2000), and Shalala v. Guernsey Memorial Hospital,514 U. S. 87, MBA contends that we have already recognized that an agency may not “avoid notice-and-comment procedures by cloaking its actions in the mantle of mere ‘interpretation.’ ” Brief for Respondent 23–24. Neither of the cases MBA cites supports its argument. Our decision in Christensen did not address a change in agency interpretation. Instead, we there refused to give deference to an agency’s interpretation of an unambiguous regulation, observing that to defer in such a case would allow the agency “to create de facto a new regulation.” 529 U. S., at 588. Put differently, Christensen held that the agency interpretation at issue was substantively invalid because it conflicted with the text of the regulation the agency purported to interpret. That holding is irrelevant to this suit and to the Paralyzed Veterans rule, which assesses whether an agency interpretation is procedurally invalid. As for Guernsey, that case is fully consistent with—indeed, confirms—what the text of the APA makes plain: “Interpretive rules do not require notice and comment.” 514 U. S., at 99. Sidestepping this inconvenient language, MBA instead quotes a portion of the Court’s opinion stating that “APA rulemaking would still be required if [an agency] adopted a new position inconsistent with . . . existing regulations.” Id., at 100. But the statement on which MBA relies is dictum. Worse, it is dictum taken out of context. The “regulations” to which the Court referred were two provisions of the Medicare reimbursement scheme. And it is apparent from the Court’s description of these regulations in Part II of the opinion that they were legislative rules, issued through the notice-and-comment process. See id., at 91–92 (noting that the disputed regulations were codified in the Code of Federal Regulations). Read properly, then, the cited passage from Guernsey merely means that “an agency may only change its interpretation if the revised interpretation is consistent with the underlying regulations.” Brief for Petitioners in No. 13–1052, p. 44.B In the main, MBA attempts to justify the Paralyzed Veterans doctrine on practical and policy grounds. MBA contends that the doctrine reinforces the APA’s goal of “procedural fairness” by preventing agencies from unilaterally and unexpectedly altering their interpretation of important regulations. Brief for Respondent 16. There may be times when an agency’s decision to issue an interpretive rule, rather than a legislative rule, is driven primarily by a desire to skirt notice-and-comment provisions. But regulated entities are not without recourse in such situations. Quite the opposite. The APA contains a variety of constraints on agency decisionmaking—the arbitrary and capricious standard being among the most notable. As we held in Fox Television Stations, and underscore again today, the APA requires an agency to provide more substantial justification when “its new policy rests upon factual findings that contradict those which underlay its prior policy; or when its prior policy has engendered serious reliance interests that must be taken into account. It would be arbitrary and capricious to ignore such matters.” 556 U. S., at 515 (citation omitted); see also id., at 535 (Kennedy, J., concurring in part and concurring in judgment). In addition, Congress is aware that agencies sometimes alter their views in ways that upset settled reliance interests. For that reason, Congress sometimes includes in the statutes it drafts safe-harbor provisions that shelter regulated entities from liability when they act in conformance with previous agency interpretations. The FLSA includes one such provision: As amended by the Portal-to-Portal Act of 1947,29 U. S. C. §251 et seq., the FLSA provides that “no employer shall be subject to any liability” for failing “to pay minimum wages or overtime compensation” if it demonstrates that the “act or omission complained of was in good faith in conformity with and in reliance on any written administrative regulation, order, ruling, approval, or interpretation” of the Administrator of the Department’s Wage and Hour Division, even when the guidance is later “modified or rescinded.” §§259(a), (b)(1). These safe harbors will often protect parties from liability when an agency adopts an interpretation that conflicts with its previous position.[5]C MBA changes direction in the second half of its brief, contending that if the Court overturns the Paralyzed Veterans rule, the D. C. Circuit’s judgment should nonetheless be affirmed. That is so, MBA says, because the agency interpretation at issue—the 2010 Administrator’s Interpretation—should in fact be classified as a legislative rule. We will not address this argument. From the beginning, the parties litigated this suit on the understanding that the Administrator’s Interpretation was—as its name suggests—an interpretive rule. Indeed, if MBA did not think the Administrator’s Interpretation was an interpretive rule, then its decision to invoke the Paralyzed Veterans doctrine in attacking the rule is passing strange. After all, Paralyzed Veterans applied only to interpretive rules. Consequently, neither the District Court nor the D. C. Circuit considered MBA’s current claim that the Administrator’s Interpretation is actually a legislative rule. Beyond that, and more important still, MBA’s brief in opposition to certiorari did not dispute petitioners’ assertions—in their framing of the question presented and in the substance of their petitions—that the Administrator’s Interpretation is an interpretive rule. Thus, even assuming MBA did not waive the argument below, it has done so in this Court. See this Court’s Rule 15.2; Carcieri v. Salazar,555 U. S. 379–396 (2009).* * * For the foregoing reasons, the judgment of the United States Court of Appeals for the District of Columbia Circuit is reversed.It is so ordered.Notes1 The latter is the more common phrasing today, and the one we use throughout this opinion.2 Buck, Henry, and Nickols are petitioners in No. 13–1052 and respondents in No. 13–1041.3 MBA did not challenge this aspect of the District Court’s decision on appeal.4 MBA alternatively suggests that interpretive rules have the force of law because an agency’s interpretation of its own regulations may be entitled to deference under Auer v. Robbins,519 U. S. 452 (1997), and Bowles v. Seminole Rock & Sand Co.,325 U. S. 410 (1945). Even in cases where an agency’s interpretation receives Auer deference, how-ever, it is the court that ultimately decides whether a given regulation means what the agency says. Moreover, Auer deference is not an inexorable command in all cases. See Christopher v. SmithKline Beecham Corp., 567 U. S. ___, ___ (2012) (slip op., at 10) (Auer deference is inappropriate “when the agency’s interpretation is plainly erroneous or inconsistent with the regulation” or “when there is reason to suspect that the agency’s interpretation does not reflect the agency’s fair and considered judgment” (internal quotation marks omitted)); Thomas Jefferson Univ. v. Shalala,512 U. S. 504,515 (1994) (“[A]n agency’s interpretation of a . . . regulation that conflicts with a prior interpretation is entitled to considerably less deference than a consistently held agency view” (internal quotation marks omitted)).5 The United States acknowledged at argument that even in situations where a statute does not contain a safe-harbor provision similar to the one included in the FLSA, an agency’s ability to pursue enforcement actions against regulated entities for conduct in conformance with prior agency interpretations may be limited by principles of retroactiv-ity. See Tr. of Oral Arg. 44–45. We have no occasion to consider how such principles might apply here.

576.US.2014_13-502

Gilbert, Arizona (Town), has a comprehensive code (Sign Code or Code) that prohibits the display of outdoor signs without a permit, but exempts 23 categories of signs, including three relevant here. “Ideological Signs,” defined as signs “communicating a message or ideas” that do not fit in any other Sign Code category, may be up to 20 square feet and have no placement or time restrictions. “Political Signs,” defined as signs “designed to influence the outcome of an election,” may be up to 32 square feet and may only be displayed during an election season. “Temporary Directional Signs,” defined as signs directing the public to a church or other “qualifying event,” have even greater restrictions: No more than four of the signs, limited to six square feet, may be on a single property at any time, and signs may be displayed no more than 12 hours before the “qualifying event” and 1 hour after. Petitioners, Good News Community Church (Church) and its pastor, Clyde Reed, whose Sunday church services are held at various temporary locations in and near the Town, posted signs early each Saturday bearing the Church name and the time and location of the next service and did not remove the signs until around midday Sunday. The Church was cited for exceeding the time limits for displaying temporary directional signs and for failing to include an event date on the signs. Unable to reach an accommodation with the Town, petitioners filed suit, claiming that the Code abridged their freedom of speech. The District Court denied their motion for a preliminary injunction, and the Ninth Circuit affirmed, ultimately concluding that the Code’s sign categories were content neutral, and that the Code satisfied the intermediate scrutiny accorded to content-neutral regulations of speech. Held: The Sign Code’s provisions are content-based regulations of speech that do not survive strict scrutiny. Pp. 6–17. (a) Because content-based laws target speech based on its communicative content, they are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests. E.g., R. A. V. v. St. Paul, 505 U. S. 377 . Speech regulation is content based if a law applies to particular speech because of the topic discussed or the idea or message expressed. E.g., Sorrell v. IMS Health, Inc., 564 U. S. ___, ___–___. And courts are required to consider whether a regulation of speech “on its face” draws distinctions based on the message a speaker conveys. Id., at ___. Whether laws define regulated speech by particular subject matter or by its function or purpose, they are subject to strict scrutiny. The same is true for laws that, though facially content neutral, cannot be “ ‘justified without reference to the content of the regulated speech,’ ” or were adopted by the government “because of disagreement with the message” conveyed. Ward v. Rock Against Racism, 491 U. S. 781 . Pp. 6–7. (b) The Sign Code is content based on its face. It defines the categories of temporary, political, and ideological signs on the basis of their messages and then subjects each category to different restrictions. The restrictions applied thus depend entirely on the sign’s communicative content. Because the Code, on its face, is a content-based regulation of speech, there is no need to consider the government’s justifications or purposes for enacting the Code to determine whether it is subject to strict scrutiny. Pp. 7. (c) None of the Ninth Circuit’s theories for its contrary holding is persuasive. Its conclusion that the Town’s regulation was not based on a disagreement with the message conveyed skips the crucial first step in the content-neutrality analysis: determining whether the law is content neutral on its face. A law that is content based on its face is subject to strict scrutiny regardless of the government’s benign motive, content-neutral justification, or lack of “animus toward the ideas contained” in the regulated speech. Cincinnati v. Discovery Network, Inc., 507 U. S. 410 . Thus, an innocuous justification cannot transform a facially content-based law into one that is content neutral. A court must evaluate each question—whether a law is content based on its face and whether the purpose and justification for the law are content based—before concluding that a law is content neutral. Ward does not require otherwise, for its framework applies only to a content-neutral statute. The Ninth Circuit’s conclusion that the Sign Code does not single out any idea or viewpoint for discrimination conflates two distinct but related limitations that the First Amendment places on government regulation of speech. Government discrimination among viewpoints is a “more blatant” and “egregious form of content discrimination,” Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819 , but “[t]he First Amendment’s hostility to content-based regulation [also] extends . . . to prohibition of public discussion of an entire topic,” Consolidated Edison Co. of N. Y. v. Public Serv. Comm’n of N. Y., 447 U. S. 530 . The Sign Code, a paradigmatic example of content-based discrimination, singles out specific subject matter for differential treatment, even if it does not target viewpoints within that subject matter. The Ninth Circuit also erred in concluding that the Sign Code was not content based because it made only speaker-based and event-based distinctions. The Code’s categories are not speaker-based—the restrictions for political, ideological, and temporary event signs apply equally no matter who sponsors them. And even if the sign categories were speaker based, that would not automatically render the law content neutral. Rather, “laws favoring some speakers over others demand strict scrutiny when the legislature’s speaker preference reflects a content preference.” Turner Broadcasting System, Inc. v. FCC, 512 U. S. 622 . This same analysis applies to event-based distinctions. Pp. 8–14. (d) The Sign Code’s content-based restrictions do not survive strict scrutiny because the Town has not demonstrated that the Code’s differentiation between temporary directional signs and other types of signs furthers a compelling governmental interest and is narrowly tailored to that end. See Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett, 564 U. S. ___, ___. Assuming that the Town has a compelling interest in preserving its aesthetic appeal and traffic safety, the Code’s distinctions are highly underinclusive. The Town cannot claim that placing strict limits on temporary directional signs is necessary to beautify the Town when other types of signs create the same problem. See Discovery Network, supra, at 425. Nor has it shown that temporary directional signs pose a greater threat to public safety than ideological or political signs. Pp. 14–15. (e) This decision will not prevent governments from enacting effective sign laws. The Town has ample content-neutral options available to resolve problems with safety and aesthetics, including regulating size, building materials, lighting, moving parts, and portability. And the Town may be able to forbid postings on public property, so long as it does so in an evenhanded, content-neutral manner. See Members of City Council of Los Angeles v. Taxpayers for Vincent, 466 U. S. 789 . An ordinance narrowly tailored to the challenges of protecting the safety of pedestrians, drivers, and passengers—e.g., warning signs marking hazards on private property or signs directing traffic—might also survive strict scrutiny. Pp. 16–17. 707 F. 3d 1057, reversed and remanded. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Alito, and Sotomayor, JJ., joined. Alito, J., filed a concurring opinion, in which Kennedy and Sotomayor, JJ., joined. Breyer, J., filed an opinion concurring in the judgment. Kagan, J., filed an opinion concurring in the judgment, in which Ginsburg and Breyer, JJ., joined

The town of Gilbert, Arizona (or Town), has adopted a comprehensive code governing the manner in which people may display outdoor signs. Gilbert, Ariz., Land Development Code (Sign Code or Code), ch. 1, §4.402 (2005).[1] The Sign Code identifies various categories of signs based on the type of information they convey, then subjects each category to different restrictions. One of the categories is “Temporary Directional Signs Relating to a Qualifying Event,” loosely defined as signs directing the public to a meeting of a nonprofit group. §4.402(P). The Code imposes more stringent restrictions on these signs than it doeson signs conveying other messages. We hold that these provisions are content-based regulations of speech that cannot survive strict scrutiny. I A The Sign Code prohibits the display of outdoor signs anywhere within the Town without a permit, but it then exempts 23 categories of signs from that requirement. These exemptions include everything from bazaar signs to flying banners. Three categories of exempt signs are particularly relevant here. The first is “Ideological Sign[s].” This category includes any “sign communicating a message or ideas for noncommercial purposes that is not a Construction Sign, Directional Sign, Temporary Directional Sign Relating to a Qualifying Event, Political Sign, Garage Sale Sign, or a sign owned or required by a governmental agency.” Sign Code, Glossary of General Terms (Glossary), p. 23 (emphasis deleted). Of the three categories discussed here, the Code treats ideological signs most favorably, allowing them to be up to 20 square feet in area and to be placed in all “zoning districts” without time limits. §4.402(J). The second category is “Political Sign[s].” This includes any “temporary sign designed to influence the outcome of an election called by a public body.” Glossary 23.[2] The Code treats these signs less favorably than ideological signs. The Code allows the placement of political signs up to 16 square feet on residential property and up to 32 square feet on nonresidential property, undeveloped municipal property, and “rights-of-way.” §4.402(I).[3] These signs may be displayed up to 60 days before a primary election and up to 15 days following a general election. Ibid. The third category is “Temporary Directional Signs Relating to a Qualifying Event.” This includes any “Temporary Sign intended to direct pedestrians, motorists, and other passersby to a ‘qualifying event.’ ” Glossary 25 (emphasis deleted). A “qualifying event” is defined as any “assembly, gathering, activity, or meeting sponsored, arranged, or promoted by a religious, charitable, community service, educational, or other similar non-profit organization.” Ibid. The Code treats temporary directional signs even less favorably than political signs.[4] Temporary directional signs may be no larger than six square feet. §4.402(P). They may be placed on private property or on a public right-of-way, but no more than four signs may be placed on a single property at any time. Ibid. And, they may be displayed no more than 12 hours before the “qualifying event” and no more than 1 hour afterward. Ibid. B Petitioners Good News Community Church (Church) and its pastor, Clyde Reed, wish to advertise the time and location of their Sunday church services. The Church is a small, cash-strapped entity that owns no building, so it holds its services at elementary schools or other locations in or near the Town. In order to inform the public about its services, which are held in a variety of different locations, the Church began placing 15 to 20 temporary signs around the Town, frequently in the public right-of-way abutting the street. The signs typically displayed the Church’s name, along with the time and location of the upcoming service. Church members would post the signs early in the day on Saturday and then remove them around midday on Sunday. The display of these signs requires little money and manpower, and thus has proved to be an economical and effective way for the Church to let the community know where its services are being held each week. This practice caught the attention of the Town’s Sign Code compliance manager, who twice cited the Church for violating the Code. The first citation noted that the Church exceeded the time limits for displaying its temporary directional signs. The second citation referred to the same problem, along with the Church’s failure to include the date of the event on the signs. Town officials even confiscated one of the Church’s signs, which Reed had to retrieve from the municipal offices. Reed contacted the Sign Code Compliance Department in an attempt to reach an accommodation. His efforts proved unsuccessful. The Town’s Code compliance manager informed the Church that there would be “no leni-ency under the Code” and promised to punish any futureviolations. Shortly thereafter, petitioners filed a complaint in the United States District Court for the District of Arizona, arguing that the Sign Code abridged their freedom of speech in violation of the First and Fourteenth Amendments. The District Court denied the petitioners’ motion for a preliminary injunction. The Court of Appeals for the Ninth Circuit affirmed, holding that the Sign Code’s provision regulating temporary directional signs did not regulate speech on the basis of content. 587 F. 3d 966, 979 (2009). It reasoned that, even though an enforcement officer would have to read the sign to determine what provisions of the Sign Code applied to it, the “ ‘kind of cursory examination’ ” that would be necessary for an officer to classify it as a temporary directional sign was “not akin to an officer synthesizing the expressive content of the sign.” Id., at 978. It then remanded for the District Court to determine in the first instance whether the Sign Code’s distinctions among temporary directional signs, political signs, and ideological signs nevertheless constituted a content-based regulation of speech. On remand, the District Court granted summary judgment in favor of the Town. The Court of Appeals again affirmed, holding that the Code’s sign categories were content neutral. The court concluded that “the distinctions between Temporary Directional Signs, Ideological Signs, and Political Signs . . . are based on objective factors relevant to Gilbert’s creation of the specific exemption from the permit requirement and do not otherwise consider the substance of the sign.” 707 F. 3d 1057, 1069 (CA9 2013). Relying on this Court’s decision in Hill v. Colorado, 530 U. S. 703 (2000) , the Court of Appeals concluded that the Sign Code is content neutral. 707 F. 3d, at 1071–1072. As the court explained, “Gilbert did not adopt its regulation of speech because it disagreed with the message conveyed” and its “interests in regulat[ing] temporary signs are unrelated to the content of the sign.” Ibid. Accord-ingly, the court believed that the Code was “content-neutral as that term [has been] defined by the Supreme Court.” Id., at 1071. In light of that determination, it applied a lower level of scrutiny to the Sign Code and concluded that the law did not violate the First Amendment. Id., at 1073–1076. We granted certiorari, 573 U. S. ___ (2014), and now reverse. II A The First Amendment, applicable to the States through the Fourteenth Amendment, prohibits the enactment of laws “abridging the freedom of speech.” U. S. Const., Amdt. 1. Under that Clause, a government, including a municipal government vested with state authority, “has no power to restrict expression because of its message, its ideas, its subject matter, or its content.” Police Dept. of Chicago v. Mosley, 408 U. S. 92, 95 (1972) . Content-based laws—those that target speech based on its communicative content—are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests. R. A. V. v. St. Paul, 505 U. S. 377, 395 (1992) ; Simon & Schuster, Inc. v. Members of N. Y. State Crime Victims Bd., 502 U. S. 105, 115, 118 (1991) . Government regulation of speech is content based if a law applies to particular speech because of the topic discussed or the idea or message expressed. E.g., Sorrell v. IMS Health, Inc., 564 U. S. ___, ___–___ (2011) (slip op., at 8–9); Carey v. Brown, 447 U. S. 455, 462 (1980) ; Mosley, supra, at 95. This commonsense meaning of the phrase “content based” requires a court to consider whether a regulation of speech “on its face” draws distinctions based on the message a speaker conveys. Sorrell, supra, at ___ (slip op., at 8). Some facial distinctions based on a message are obvious, defining regulated speech by particular subject matter, and others are more subtle, defining regulated speech by its function or purpose. Both are distinctions drawn based on the message a speaker conveys, and, therefore, are subject to strict scrutiny. Our precedents have also recognized a separate and additional category of laws that, though facially content neutral, will be considered content-based regulations of speech: laws that cannot be “ ‘justified without reference to the content of the regulated speech,’ ” or that were adopted by the government “because of disagreement with the message [the speech] conveys,” Ward v. Rock Against Racism, 491 U. S. 781, 791 (1989) . Those laws, like those that are content based on their face, must also satisfy strict scrutiny. B The Town’s Sign Code is content based on its face. It defines “Temporary Directional Signs” on the basis of whether a sign conveys the message of directing the public to church or some other “qualifying event.” Glossary 25. It defines “Political Signs” on the basis of whether a sign’s message is “designed to influence the outcome of an election.” Id., at 24. And it defines “Ideological Signs” on the basis of whether a sign “communicat[es] a message or ideas” that do not fit within the Code’s other categories. Id., at 23. It then subjects each of these categories to different restrictions. The restrictions in the Sign Code that apply to any given sign thus depend entirely on the communicative content of the sign. If a sign informs its reader of the time and place a book club will discuss John Locke’s Two Treatises of Government, that sign will be treated differently from a sign expressing the view that one should vote for one of Locke’s followers in an upcoming election, and both signs will be treated differently from a sign expressing an ideological view rooted in Locke’s theory of government. More to the point, the Church’s signs inviting people to attend its worship services are treated differently from signs conveying other types of ideas. On its face, the Sign Code is a content-based regulation of speech. We thus have no need to consider the government’s justifications or purposes for enacting the Code to determine whether it is subject to strict scrutiny. C In reaching the contrary conclusion, the Court of Appeals offered several theories to explain why the Town’s Sign Code should be deemed content neutral. None is persuasive. 1 The Court of Appeals first determined that the Sign Code was content neutral because the Town “did not adopt its regulation of speech [based on] disagree[ment] with the message conveyed,” and its justifications for regulating temporary directional signs were “unrelated to the content of the sign.” 707 F. 3d, at 1071–1072. In its brief to this Court, the United States similarly contends that a sign regulation is content neutral—even if it expressly draws distinctions based on the sign’s communicative content—if those distinctions can be “ ‘justified without reference to the content of the regulated speech.’ ” Brief for United States as Amicus Curiae 20, 24 (quoting Ward, supra, at 791; emphasis deleted). But this analysis skips the crucial first step in thecontent-neutrality analysis: determining whether the law is content neutral on its face. A law that is content based on its face is subject to strict scrutiny regardless of the government’s benign motive, content-neutral justification, or lack of “animus toward the ideas contained” in the regulated speech. Cincinnati v. Discovery Network, Inc., 507 U. S. 410, 429 (1993) . We have thus made clear that “ ‘[i]llicit legislative intent is not the sine qua non of a violation of the First Amendment,’ ” and a party opposing the government “need adduce ‘no evidence of an improper censorial motive.’ ” Simon & Schuster, supra, at 117. Although “a content-based purpose may be sufficient in certain circumstances to show that a regulation is content based, it is not necessary.” Turner Broadcasting System, Inc. v. FCC, 512 U. S. 622, 642 (1994) . In other words, an innocuous justification cannot transform a facially content-based law into one that is content neutral. That is why we have repeatedly considered whether a law is content neutral on its face before turning to the law’s justification or purpose. See, e.g., Sorrell, supra, at ___–___ (slip op., at 8–9) (statute was content based “on its face,” and there was also evidence of an impermissible legislative motive); United States v. Eichman, 496 U. S. 310, 315 (1990) (“Although the [statute] contains no ex-plicit content-based limitation on the scope of prohibited conduct, it is nevertheless clear that the Government’s asserted interest is related to the suppression of free expression” (internal quotation marks omitted)); Members of City Council of Los Angeles v. Taxpayers for Vincent, 466 U. S. 789, 804 (1984) (“The text of the ordinance is neutral,” and “there is not even a hint of bias or censorship in the City’s enactment or enforcement of this ordinance”); Clark v. Community for Creative Non-Violence, 468 U. S. 288, 293 (1984) (requiring that a facially content-neutral ban on camping must be “justified without reference to the content of the regulated speech”); United States v. O’Brien, 391 U. S. 367, 375, 377 (1968) (noting that the statute “on its face deals with conduct having no connection with speech,” but examining whether the “the governmental interest is unrelated to the suppression of free expression”). Because strict scrutiny applies either when a law is content based on its face or when the purpose and justification for the law are content based, a court must evaluate each question before it concludes that the law is content neutral and thus subject to a lower level of scrutiny. The Court of Appeals and the United States misunderstand our decision in Ward as suggesting that a government’s purpose is relevant even when a law is content based on its face. That is incorrect. Ward had nothing to say about facially content-based restrictions because it involved a facially content-neutral ban on the use, in a city-owned music venue, of sound amplification systems not provided by the city. 491 U. S., at 787, and n. 2. In that context, we looked to governmental motive, including whether the government had regulated speech “because of disagreement” with its message, and whether the regulation was “ ‘justified without reference to the content of the speech.’ ” Id., at 791. But Ward’s framework “applies only if a statute is content neutral.” Hill, 530 U. S., at 766 (Kennedy, J., dissenting). Its rules thus operate “to protect speech,” not “to restrict it.” Id., at 765. The First Amendment requires no less. Innocent motives do not eliminate the danger of censorship presented by a facially content-based statute, as future government officials may one day wield such statutes to suppress disfavored speech. That is why the First Amendment expressly targets the operation of the laws—i.e., the “abridg[ement] of speech”—rather than merely the motives of those who enacted them. U. S. Const., Amdt. 1. “ ‘The vice of content-based legislation . . . is not that it is always used for invidious, thought-control purposes, but that it lends itself to use for those purposes.’ ” Hill, supra, at 743 (Scalia, J., dissenting). For instance, in NAACP v. Button, 371 U. S. 415 (1963) , the Court encountered a State’s attempt to use a statute prohibiting “ ‘improper solicitation’ ” by attorneys to outlaw litigation-related speech of the National Association for the Advancement of Colored People. Id., at 438. Although Button predated our more recent formulations of strict scrutiny, the Court rightly rejected the State’s claim that its interest in the “regulation of professional conduct” rendered the statute consistent with the First Amendment, observing that “it is no answer . . . to say . . . that the purpose of these regulations was merely to insure high professional standards and not to curtail free expression.” Id., at 438–439. Likewise, one could easily imagine a Sign Code compliance manager who disliked the Church’s substantive teachings deploying the Sign Code to make it more difficult for the Church to inform the public of the location of its services. Accordingly, we have repeatedly “rejected the argument that ‘discriminatory . . . treatment is suspect under the First Amendment only when the legislature intends to suppress certain ideas.’ ” Discovery Network, 507 U. S., at 429. We do so again today. 2 The Court of Appeals next reasoned that the Sign Code was content neutral because it “does not mention any idea or viewpoint, let alone single one out for differential treatment.” 587 F. 3d, at 977. It reasoned that, for the purpose of the Code provisions, “[i]t makes no difference which candidate is supported, who sponsors the event, or what ideological perspective is asserted.” 707 F. 3d, at 1069. The Town seizes on this reasoning, insisting that “content based” is a term of art that “should be applied flexibly” with the goal of protecting “viewpoints and ideas from government censorship or favoritism.” Brief for Respondents 22. In the Town’s view, a sign regulation that “does not censor or favor particular viewpoints or ideas” cannot be content based. Ibid. The Sign Code allegedly passes this test because its treatment of temporary directional signs does not raise any concerns that the government is “endorsing or suppressing ‘ideas or viewpoints,’ ” id., at 27, and the provisions for political signs and ideological signs “are neutral as to particular ideas or viewpoints” within those categories. Id., at 37. This analysis conflates two distinct but related limitations that the First Amendment places on government regulation of speech. Government discrimination among viewpoints—or the regulation of speech based on “the specific motivating ideology or the opinion or perspective of the speaker”—is a “more blatant” and “egregious form of content discrimination.” Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819, 829 (1995) . But it is well established that “[t]he First Amendment’s hostility to content-based regulation extends not only to restrictions on particular viewpoints, but also to prohibition of public discussion of an entire topic.” Consolidated Edison Co. of N. Y. v. Public Serv. Comm’n of N. Y., 447 U. S. 530, 537 (1980) . Thus, a speech regulation targeted at specific subject matter is content based even if it does not discriminate among viewpoints within that subject matter. Ibid. For example, a law banning the use of sound trucks for political speech—and only political speech—would be a content-based regulation, even if it imposed no limits on the political viewpoints that could be expressed. See Discovery Network, supra, at 428. The Town’s Sign Code likewise singles out specific subject matter for differential treatment, even if it does not target viewpoints within that subject matter. Ideological messages are given more favorable treatment than messages concerning a political candidate, which are themselves given more favorable treatment than messages announcing an assembly of like-minded individuals. That is a paradigmatic example of content-based discrimination. 3 Finally, the Court of Appeals characterized the Sign Code’s distinctions as turning on “ ‘the content-neutral elements of who is speaking through the sign and whether and when an event is occurring.’ ” 707 F. 3d, at 1069. That analysis is mistaken on both factual and legal grounds. To start, the Sign Code’s distinctions are not speaker based. The restrictions for political, ideological, and temporary event signs apply equally no matter who sponsors them. If a local business, for example, sought to put up signs advertising the Church’s meetings, those signs would be subject to the same limitations as such signs placed by the Church. And if Reed had decided to dis-play signs in support of a particular candidate, he could have made those signs far larger—and kept them up for far longer—than signs inviting people to attend hischurch services. If the Code’s distinctions were truly speaker based, both types of signs would receive the same treatment. In any case, the fact that a distinction is speaker based does not, as the Court of Appeals seemed to believe, automatically render the distinction content neutral. Because “[s]peech restrictions based on the identity of the speaker are all too often simply a means to control content,” Citizens United v. Federal Election Comm’n, 558 U. S. 310, 340 (2010) , we have insisted that “laws favoring some speakers over others demand strict scrutiny when the legislature’s speaker preference reflects a content preference,” Turner, 512 U. S., at 658. Thus, a law limiting the content of newspapers, but only newspapers, could not evade strict scrutiny simply because it could be characterized as speaker based. Likewise, a content-based law that restricted the political speech of all corporations would not become content neutral just because it singled out corporations as a class of speakers. See Citizens United, supra, at 340–341. Characterizing a distinction as speaker based is only the beginning—not the end—of the inquiry. Nor do the Sign Code’s distinctions hinge on “whether and when an event is occurring.” The Code does not permit citizens to post signs on any topic whatsoever within a set period leading up to an election, for example. Instead, come election time, it requires Town officials to determine whether a sign is “designed to influence the outcome of an election” (and thus “political”) or merely “communicating a message or ideas for noncommercial purposes” (and thus “ideological”). Glossary 24. That obvious content-based inquiry does not evade strict scrutiny review simply because an event (i.e., an election) is involved. And, just as with speaker-based laws, the fact that a distinction is event based does not render it content neutral. The Court of Appeals cited no precedent from this Court supporting its novel theory of an exception from the content-neutrality requirement for event-based laws. As we have explained, a speech regulation is content based if the law applies to particular speech because of the topic discussed or the idea or message expressed. Supra, at 6. A regulation that targets a sign because it conveys an idea about a specific event is no less content based than a regulation that targets a sign because it conveys some other idea. Here, the Code singles out signs bearing a particular message: the time and location of a specific event. This type of ordinance may seem like a perfectly rational way to regulate signs, but a clear and firm rule governing content neutrality is an essential means of protecting the freedom of speech, even if laws that might seem “entirely reasonable” will sometimes be “struck down because of their content-based nature.” City of Ladue v. Gilleo, 512 U. S. 43, 60 (1994) (O’Connor, J., concurring). III Because the Town’s Sign Code imposes content-based restrictions on speech, those provisions can stand only if they survive strict scrutiny, “ ‘which requires the Government to prove that the restriction furthers a compelling interest and is narrowly tailored to achieve that interest,’ ” Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett, 564 U. S. ___, ___ (2011) (slip op., at 8) (quoting Citizens United, 558 U. S., at 340). Thus, it is the Town’s burden to demonstrate that the Code’s differentiation between temporary directional signs and other types of signs, such as political signs and ideological signs, furthers a compelling governmental interest and is narrowly tailored to that end. See ibid. The Town cannot do so. It has offered only two governmental interests in support of the distinctions the Sign Code draws: preserving the Town’s aesthetic appeal and traffic safety. Assuming for the sake of argument that those are compelling governmental interests, the Code’s distinctions fail as hopelessly underinclusive. Starting with the preservation of aesthetics, temporary directional signs are “no greater an eyesore,” Discovery Network, 507 U. S., at 425, than ideological or political ones. Yet the Code allows unlimited proliferation of larger ideological signs while strictly limiting the number, size, and duration of smaller directional ones. The Town cannot claim that placing strict limits on temporary directional signs is necessary to beautify the Town while at the same time allowing unlimited numbers of other types of signs that create the same problem. The Town similarly has not shown that limiting temporary directional signs is necessary to eliminate threats to traffic safety, but that limiting other types of signs is not. The Town has offered no reason to believe that directional signs pose a greater threat to safety than do ideological or political signs. If anything, a sharply worded ideological sign seems more likely to distract a driver than a sign directing the public to a nearby church meeting. In light of this underinclusiveness, the Town has not met its burden to prove that its Sign Code is narrowly tailored to further a compelling government interest. Because a “ ‘law cannot be regarded as protecting an interest of the highest order, and thus as justifying a restriction on truthful speech, when it leaves appreciable damage to that supposedly vital interest unprohibited,’ ” Republican Party of Minn. v. White, 536 U. S. 765, 780 (2002) , the Sign Code fails strict scrutiny. IV Our decision today will not prevent governments from enacting effective sign laws. The Town asserts that an “ ‘absolutist’ ” content-neutrality rule would render “virtually all distinctions in sign laws . . . subject to strict scrutiny,” Brief for Respondents 34–35, but that is not the case. Not “all distinctions” are subject to strict scrutiny, only content-based ones are. Laws that are content neutral are instead subject to lesser scrutiny. See Clark, 468 U. S., at 295. The Town has ample content-neutral options available to resolve problems with safety and aesthetics. For example, its current Code regulates many aspects of signs that have nothing to do with a sign’s message: size, building materials, lighting, moving parts, and portability. See, e.g., §4.402(R). And on public property, the Town may go a long way toward entirely forbidding the posting of signs, so long as it does so in an evenhanded, content-neutral manner. See Taxpayers for Vincent, 466 U. S., at 817 (upholding content-neutral ban against posting signs on public property). Indeed, some lower courts have long held that similar content-based sign laws receive strict scrutiny, but there is no evidence that towns in those jurisdictions have suffered catastrophic effects. See, e.g., Solantic, LLC v. Neptune Beach, 410 F. 3d 1250, 1264–1269 (CA11 2005) (sign categories similar to the town of Gilbert’s were content based and subject to strict scru-tiny); Matthews v. Needham, 764 F. 2d 58, 59–60 (CA1 1985) (law banning political signs but not commercial signs was content based and subject to strict scrutiny). We acknowledge that a city might reasonably view the general regulation of signs as necessary because signs “take up space and may obstruct views, distract motorists, displace alternative uses for land, and pose other problems that legitimately call for regulation.” City of Ladue, 512 U. S., at 48. At the same time, the presence of certain signs may be essential, both for vehicles and pedestrians, to guide traffic or to identify hazards and ensure safety. A sign ordinance narrowly tailored to the challenges of protecting the safety of pedestrians, drivers, and passengers—such as warning signs marking hazards on private property, signs directing traffic, or street numbers associated with private houses—well might survive strict scrutiny. The signs at issue in this case, including political and ideological signs and signs for events, are far removed from those purposes. As discussed above, they are facially content based and are neither justified by traditional safety concerns nor narrowly tailored. * * * We reverse the judgment of the Court of Appeals and remand the case for proceedings consistent with this opinion. It is so ordered.Notes 1 The Town’s Sign Code is available online at http://www.gilbertaz.gov/departments / development - service / planning - development / land -development-code (as visited June 16, 2015, and available in Clerk of Court’s case file). 2 A “Temporary Sign” is a “sign not permanently attached to the ground, a wall or a building, and not designed or intended for permanent display.” Glossary 25. 3 The Code defines “Right-of-Way” as a “strip of publicly owned land occupied by or planned for a street, utilities, landscaping, sidewalks, trails, and similar facilities.” Id., at 18. 4 The Sign Code has been amended twice during the pendency of this case. When litigation began in 2007, the Code defined the signs at issue as “Religious Assembly Temporary Direction Signs.” App. 75. The Code entirely prohibited placement of those signs in the public right-of-way, and it forbade posting them in any location for more than two hours before the religious assembly or more than one hour afterward. Id., at 75–76. In 2008, the Town redefined the category as “Temporary Directional Signs Related to a Qualifying Event,” and it expanded the time limit to 12 hours before and 1 hour after the “qualifying event.” Ibid. In 2011, the Town amended the Code to authorize placement of temporary directional signs in the public right-of-way. Id., at 89.

576.US.2014_14-185

After petitioner Noel Reyes Mata, an unlawful resident alien, was convicted of assault in a Texas court, an Immigration Judge ordered him removed to Mexico. Mata’s attorney filed a notice of appeal with the Board of Immigration Appeals (BIA or Board), but never filed a brief, and the appeal was dismissed. Acting through different counsel, Mata filed a motion to reopen his removal proceedings, as authorized by statute. See 8 U. S. C. §1229a(c)(7)(A). Acknowledging that he had missed the 90-day deadline for such motions, see §1229a(c)(7)(C)(i), Mata argued that his previous counsel’s ineffective assistance was an exceptional circumstance entitling him to equitable tolling of the time limit. But the BIA disagreed and dismissed the motion as untimely. The BIA also declined to reopen Mata’s removal proceedings sua sponte based on its separate regulatory authority. See 8 CFR §1003.2(a). On appeal, the Fifth Circuit construed Mata’s equitable tolling claim as an invitation for the Board to exercise its regulatory authority to reopen the proceedings sua sponte, and—because circuit precedent forbids the court to review BIA decisions not to exercise that authority—dismissed Mata’s appeal for lack of jurisdiction. Held: The Fifth Circuit erred in declining to take jurisdiction over Mata’s appeal. A court of appeals has jurisdiction to review the BIA’s rejection of an alien’s motion to reopen. Kucana v. Holder, 558 U. S. 233 . Nothing about that jurisdiction changes where the Board rejects a motion as untimely, or when it rejects a motion requesting equitable tolling of the time limit. That jurisdiction likewise remains unchanged if the BIA’s denial also contains a separate decision not to exercise its sua sponte authority. So even assuming the Fifth Circuit is correct that courts of appeals lack jurisdiction to review BIA decisions not to reopen cases sua sponte, that lack of jurisdiction does not affect jurisdiction over the decision on the alien’s motion to reopen. It thus follows that the Fifth Circuit had jurisdiction over this case. The Fifth Circuit’s contrary decision rested on its construing Mata’s motion as an invitation for the Board to exercise its sua sponte discretion. Court-appointed amicus asserts that the Fifth Circuit’s recharacterization was based on the premise that equitable tolling in Mata’s situation is categorically forbidden. In amicus’s view, the court’s construal was therefore an example of the ordinary practice of recharacterizing a doomed request as one for relief that may be available. But even if equitable tolling is prohibited, the Fifth Circuit’s action was not justified. If Mata is not entitled to relief on the merits, then the correct disposition is to take jurisdiction and affirm the BIA’s denial of his motion. For a court retains jurisdiction even if a litigant’s request for relief lacks merit, see Steel Co. v. Citizens for Better Environment, 523 U. S. 83 , and a federal court has a “virtually unflagging obligation,” Colorado River Water Conservation Dist. v. United States, 424 U. S. 800 , to assert jurisdiction where it has that authority. Nor can the established practice of recharacterizing pleadings so as to offer the possibility of relief justify an approach that, as here, renders relief impossible and sidesteps the judicial obligation to assert jurisdiction. Pp. 4–8. 558 Fed. Appx. 366, reversed and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Ginsburg, Breyer, Alito, and Sotomayor, JJ., joined. Thomas, J., filed a dissenting opinion.

An alien ordered to leave the country has a statutory right to file a motion to reopen his removal proceedings. See 8 U. S. C. §1229a(c)(7)(A). If immigration officials deny that motion, a federal court of appeals has jurisdiction to consider a petition to review their decision. See Kucana v. Holder, 558 U. S. 233, 242, 253 (2010) . Notwithstanding that rule, the court below declined to take jurisdiction over such an appeal because the motion to reopen had been denied as untimely. We hold that was error. I The Immigration and Nationality Act (INA), 66Stat. 163, as amended, 8 U. S. C. §1101 et seq., and its implementing regulations set out the process for removing aliens from the country. An immigration judge (IJ) conducts the initial proceedings; if he orders removal, the alien has the opportunity to appeal that decision tothe Board of Immigration Appeals (BIA or Board). §§1229a(a)(1), (c)(5). “[E]very alien ordered removed” also “has a right to file one motion” with the IJ or Boardto “reopen his or her removal proceedings.” Dada v. Mukasey, 554 U. S. 1 –5 (2008); see §1229a(c)(7)(A). Subject to exceptions not relevant here, that motion to reopen “shall be filed within 90 days” of the final removal order. §1229a(c)(7)(C)(i). Finally, the BIA’s regulations provide that, separate and apart from acting on the alien’s motion, the BIA may reopen removal proceedings “on its own motion”—or, in Latin, sua sponte—at any time. 8 CFR §1003.2(a) (2015). Petitioner Noel Reyes Mata is a Mexican citizen who entered the United States unlawfully almost 15 years ago. In 2010, he was convicted of assault under the Texas Penal Code. The federal Department of Homeland Secu-rity (DHS) immediately initiated removal proceedings against him, and in August 2011 an IJ ordered him removed. See App. 6–13. Mata’s lawyer then filed a notice of appeal with the BIA, indicating that he would soon submit a written brief stating grounds for reversing the IJ’s decision. But the attorney never filed the brief, and the BIA dismissed the appeal in September 2012. See App. 4–5. More than a hundred days later, Mata (by then represented by new counsel) filed a motion with the Board to reopen his case. DHS opposed the motion, arguing in part that Mata had failed to file it, as the INA requires, within 90 days of the Board’s decision. Mata responded that the motion was “not time barred” because his first lawyer’s “ineffective assistance” counted as an “exceptional circumstance[ ]” excusing his lateness. Certified Administrative Record in No. 13–60253 (CA5, Aug. 2, 2013), p. 69. In addressing those arguments, the Board reaffirmed prior decisions holding that it had authority to equitably toll the 90-day period in certain cases involving ineffective representation. See App. to Pet. for Cert. 7; see also, e.g., In re Santa Celenia Diaz, 2009 WL 2981747 (BIA, Aug. 21, 2009). But the Board went on to determine that Mata was not entitled to equitable tolling because he could not show prejudice from his attorney’s deficient performance; accordingly, the Board found Mata’s motion untimely. See App. to Pet. for Cert. 7–8. And in closing, the Board decided as well that Mata’s case was not one “that would warrant reopening as an exercise of” its sua sponte authority. Id., at 9 (stating that “the power to reopen on our own motion is not meant to be used as a general cure for filing defects” (internal quotation marks omitted)). Mata petitioned the Court of Appeals for the Fifth Circuit to review the BIA’s denial of his motion to reopen, arguing that he was entitled to equitable tolling. The Fifth Circuit, however, declined to “address the meritsof Mata’s equitable-tolling . . . claim[ ].” Reyes Mata v. Holder, 558 Fed. Appx. 366, 367 (2014) (per curiam). It stated instead that “[i]n this circuit, an alien’s request [to the BIA] for equitable tolling on the basis of ineffective assistance of counsel is construed as an invitation for the BIA to exercise its discretion to reopen the removal proceeding sua sponte.” Ibid. And circuit precedent held that courts have no jurisdiction to review the BIA’s refusal to exercise its sua sponte power to reopen cases. See ibid. The Court of Appeals thus dismissed Mata’s appeal for lack of jurisdiction. Every other Circuit that reviews removal orders has affirmed its jurisdiction to decide an appeal, like Mata’s, that seeks equitable tolling of the statutory time limit to file a motion to reopen a removal proceeding.[1] We granted certiorari to resolve this conflict. 574 U. S. ___ (2015). And because the Federal Government agrees with Mata that the Fifth Circuit had jurisdiction over his appeal, we appointed an amicus curiae to defend the judgment below.[2] We now reverse. II As we held in Kucana v. Holder, circuit courts have jurisdiction when an alien appeals from the Board’s denial of a motion to reopen a removal proceeding. See 558 U. S., at 242, 253. The INA, in combination with a statute cross-referenced there, gives the courts of appeals jurisdiction to review “final order[s] of removal.” 8 U. S. C. §1252(a)(1); 28 U. S. C. §2342. That jurisdiction, as the INA expressly contemplates, encompasses review of decisions refusingto reopen or reconsider such orders. See 8 U. S. C. §1252(b)(6) (“[A]ny review sought of a motion to reopen or reconsider [a removal order] shall be consolidated with the review of the [underlying] order”). Indeed, as we explained in Kucana, courts have reviewed those decisions for nearly a hundred years; and even as Congress curtailed other aspects of courts’ jurisdiction over BIA rulings, it left that authority in place. See 558 U. S., at242–251. Nothing changes when the Board denies a motion to reopen because it is untimely—nor when, in doing so, the Board rejects a request for equitable tolling. Under the INA, as under our century-old practice, the reason for the BIA’s denial makes no difference to the jurisdictional issue. Whether the BIA rejects the alien’s motion to re-open because it comes too late or because it falls short in some other respect, the courts have jurisdiction to review that decision. Similarly, that jurisdiction remains unchanged if the Board, in addition to denying the alien’s statutorily authorized motion, states that it will not exercise its separate sua sponte authority to reopen the case. See supra, at 1–2. In Kucana, we declined to decide whether courts have jurisdiction to review the BIA’s use of that discretionary power. See 558 U. S., at 251, n. 18. Courts of Appeals, including the Fifth Circuit, have held that they generally lack such authority. See, e.g., Enriquez-Alvarado v. Ashcroft, 371 F. 3d 246, 249–250 (CA5 2004); Tamenut v. Mukasey, 521 F. 3d 1000, 1003–1004 (CA8 2008) (en banc) (per curiam) (citing other decisions). Assuming arguendo that is right, it means only that judicial review ends after the court has evaluated the Board’s ruling on the alien’s motion. That courts lack jurisdiction over one matter (the sua sponte decision) does not affect their jurisdiction over another (the decision on the alien’s request). It follows, as the night the day, that the Court of Appeals had jurisdiction over this case. Recall: As authorized by the INA, Mata filed a motion with the Board to reopen his removal proceeding. The Board declined to grant Mata his proposed relief, thus conferring jurisdiction on an appellate court under Kucana. The Board did so for timeliness reasons, holding that Mata had filed his motion after 90 days had elapsed and that he was not entitled to equitable tolling. But as just explained, the reason the Board gave makes no difference: Whenever the Board denies an alien’s statutory motion to reopen a removal case, courts have jurisdiction to review its decision. In addition, the Board determined not to exercise its sua sponte authority to reopen. But once again, that extra ruling does not matter. The Court of Appeals did not lose jurisdiction over the Board’s denial of Mata’s motion just because the Board also declined to reopen his case sua sponte. Nonetheless, the Fifth Circuit dismissed Mata’s appeal for lack of jurisdiction. That decision, as described earlier, hinged on “constru[ing]” Mata’s motion as something it was not: “an invitation for the BIA to exercise” its sua sponte authority. 558 Fed. Appx., at 367; supra, at 3. Amicus’s defense of that approach centrally relies on a merits-based premise: that the INA forbids equitable tolling of the 90-day filing period in any case, no matter how exceptional the circumstances. See Brief for Amicus Curiae by Invitation of the Court 14–35. Given that is so, amicus continues, the court acted permissibly in “recharacteriz[ing]” Mata’s pleadings. Id., at 36. After all, courts often treat a request for “categorically unavailable” relief as instead “seeking relief [that] may be available.” Id., at 35, 38. And here (amicus concludes) that meant construing Mata’s request for equitable tolling as a request for sua sponte reopening—even though that caused the Fifth Circuit to lose its jurisdiction. But that conclusion is wrong even on the assumption—and it is only an assumption—that its core premise about equitable tolling is true.[3] If the INA precludes Mata from getting the relief he seeks, then the right course on appeal is to take jurisdiction over the case, explain why that is so, and affirm the BIA’s decision not to reopen. The jurisdictional question (whether the court has power to decide if tolling is proper) is of course distinct from the merits question (whether tolling is proper). See Steel Co. v. Citizens for Better Environment, 523 U. S. 83, 89 (1998) (“[T]he absence of a valid . . . cause of action does not implicate subject-matter jurisdiction”). The Fifth Circuit thus retains jurisdiction even if Mata’s appeal lacks merit. And when a federal court has jurisdiction, it also has a “virtually unflagging obligation . . . to exercise” that authority. Colorado River Water Conservation Dist. v.United States, 424 U. S. 800, 817 (1976) . Accordingly,the Court of Appeals should have asserted jurisdiction over Mata’s appeal and addressed the equitable tolling question. Contrary to amicus’s view, the practice of recharacterizing pleadings so as to offer the possibility of relief cannot justify the Court of Appeals’ alternative approach. True enough (and a good thing too) that courts sometimes construe one kind of filing as another: If a litigant misbrands a motion, but could get relief under a different label, a court will often make the requisite change. See, e.g., 12 J. Moore, Moore’s Federal Practice, §59.11[4] (3 ed. 2015) (explaining how courts treat untimely Rule 59 motions as Rule 60 motions because the latter have no time limit). But that established practice does not entail sidestepping the judicial obligation to exercise jurisdiction. And it results in identifying a route to relief, not in rendering relief impossible. That makes all the difference between a court’s generously reading pleadings and a court’s construing away adjudicative authority. And if, as amicus argues, that construal rests on an underlying merits decision—that the INA precludes any equitable tolling—then the Court of Appeals has effectively insulated a circuit split from our review. Putting theFifth Circuit to the side, all appellate courts to have addressed the matter have held that the Board may sometimes equitably toll the time limit for an alien’s motion to reopen. See n. 1, supra. Assuming the Fifth Circuit thinks otherwise, that creates the kind of split of authority we typically think we need to resolve. See this Court’s Rule 10(a). But the Fifth Circuit’s practice of recharacterizing appeals like Mata’s as challenges to the Board’s sua sponte decisions and then declining to exercise jurisdiction over them prevents that split from coming to light. Of course, the Court of Appeals may reach whatever conclusion it thinks best as to the availability of equitable tolling; we express no opinion on that matter. See n. 3, supra. What the Fifth Circuit may not do is to wrap such a merits decision in jurisdictional garb so that we cannot address a possible division between that court and every other. For the foregoing reasons, we reverse the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 See, e.g., Da Silva Neves v. Holder, 613 F. 3d 30, 33 (CA1 2010) (per curiam) (exercising jurisdiction over such a petition); Iavorski v. INS, 232 F. 3d 124, 129–134 (CA2 2000) (same); Borges v. Gonzales, 402 F. 3d 398, 406 (CA3 2005) (same); Kuusk v. Holder, 732 F. 3d 302, 305–306 (CA4 2013) (same); Barry v. Mukasey, 524 F. 3d 721, 724–725 (CA6 2008) (same); Pervaiz v. Gonzales, 405 F. 3d 488, 490 (CA7 2005) (same); Hernandez-Moran v. Gonzales, 408 F. 3d 496, 499–500 (CA8 2005) (same); Valeriano v. Gonzales, 474 F. 3d 669, 673 (CA9 2007) (same); Riley v. INS, 310 F. 3d 1253, 1257–1258 (CA10 2002) (same); Avila-Santoyo v. United States Atty. Gen., 713 F. 3d 1357, 1359, 1362–1364 (CA11 2013) (per curiam) (same). Except for Da Silva Neves, which did not resolve the issue, all those decisions also held, on the merits, that the INA allows equitable tolling in certain circumstances. See infra, at 7–8. 2 We appointed William R. Peterson to brief and argue the case, 574 U. S. ___ (2015), and he has ably discharged his responsibilities. 3 We express no opinion as to whether or when the INA allows the Board to equitably toll the 90-day period to file a motion to reopen. Moreover, we are not certain what the Fifth Circuit itself thinks about that question. Perhaps, as amicus asserts, the court believes the INA categorically precludes equitable tolling: It is hard to come up with any other reason why the court construes every argument for tolling as one for sua sponte relief. See Brief for Amicus Curiae by Invitation of the Court 2, 10, 14, n. 2. But the Fifth Circuit has stated that position in only a single sentence in a single unpublished opinion, which (according to the Circuit) has no precedential force. See Lin v. Mukasey, 286 Fed. Appx. 148, 150 (2008) (per curiam); Rule 47.5.4 (2015). And another unpublished decision cuts in the opposite direction, “hold[ing] that the doctrine of equitable tolling applies” when exceptional circumstances excuse an alien’s failure to meet the 90-day reopening deadline. See Torabi v. Gonzales, 165 Fed. Appx. 326, 331 (CA5 2006) (per curiam). So, in the end, it is hard to say.

575.US.2014_13-9972

Officer Struble, a K–9 officer, stopped petitioner Rodriguez for driving on a highway shoulder, a violation of Nebraska law. After Struble attended to everything relating to the stop, including, inter alia, checking the driver’s licenses of Rodriguez and his passenger and issuing a warning for the traffic offense, he asked Rodriguez for permission to walk his dog around the vehicle. When Rodriguez refused, Struble detained him until a second officer arrived. Struble then retrieved his dog, who alerted to the presence of drugs in the vehicle. The ensuing search revealed methamphetamine. Seven or eight minutes elapsed from the time Struble issued the written warning until the dog alerted. Rodriguez was indicted on federal drug charges. He moved to suppress the evidence seized from the vehicle on the ground, among others, that Struble had prolonged the traffic stop without reasonable suspicion in order to conduct the dog sniff. The Magistrate Judge recommended denial of the motion. He found no reasonable suspicion supporting detention once Struble issued the written warning. Under Eighth Circuit precedent, however, he concluded that prolonging the stop by “seven to eight minutes” for the dog sniff was only a de minimis intrusion on Rodriguez’s Fourth Amendment rights and was for that reason permissible. The District Court then denied the motion to suppress. Rodriguez entered a conditional guilty plea and was sentenced to five years in prison. The Eighth Circuit affirmed. Noting that the seven or eight minute delay was an acceptable “de minimis intrusion on Rodriguez’s personal liberty,” the court declined to reach the question whether Struble had reasonable suspicion to continue Rodriguez’s detention after issuing the written warning. Held: 1. Absent reasonable suspicion, police extension of a traffic stop in order to conduct a dog sniff violates the Constitution’s shield against unreasonable seizures. A routine traffic stop is more like a brief stop under Terry v. Ohio, 392 U. S. 1 , than an arrest, see, e.g., Arizona v. Johnson, 555 U. S. 323 . Its tolerable duration is determined by the seizure’s “mission,” which is to address the traffic violation that warranted the stop, Illinois v. Caballes, 543 U. S. 405 and attend to related safety concerns. Authority for the seizure ends when tasks tied to the traffic infraction are—or reasonably should have been—completed. The Fourth Amendment may tolerate certain unrelated investigations that do not lengthen the roadside detention, Johnson, 555 U. S., at 327–328 (questioning); Caballes, 543 U. S., at 406, 408 (dog sniff), but a traffic stop “become[s] unlawful if it is prolonged beyond the time reasonably required to complete th[e] mission” of issuing a warning ticket, id., at 407. Beyond determining whether to issue a traffic ticket, an officer’s mission during a traffic stop typically includes checking the driver’s license, determining whether there are outstanding warrants against the driver, and inspecting the automobile’s registration and proof of insurance. These checks serve the same objective as enforcement of the traffic code: ensuring that vehicles on the road are operated safely and responsibly. See Delaware v. Prouse, 440 U. S. 648 –659. Lacking the same close connection to roadway safety as the ordinary inquiries, a dog sniff is not fairly characterized as part of the officer’s traffic mission. In concluding that the de minimis intrusion here could be offset by the Government’s interest in stopping the flow of illegal drugs, the Eighth Circuit relied on Pennsylvania v. Mimms, 434 U. S. 106 . The Court reasoned in Mimms that the government’s “legitimate and weighty” interest in officer safety outweighed the “de minimis” additional intrusion of requiring a driver, lawfully stopped, to exit a vehicle, id., at 110–111. The officer-safety interest recognized in Mimms, however, stemmed from the danger to the officer associated with the traffic stop itself. On-scene investigation into other crimes, in contrast, detours from the officer’s traffic-control mission and therefore gains no support from Mimms. The Government’s argument that an officer who completes all traffic-related tasks expeditiously should earn extra time to pursue an unrelated criminal investigation is unpersuasive, for a traffic stop “prolonged beyond” the time in fact needed for the officer to complete his traffic-based inquiries is “unlawful,” Caballes, 543 U. S., at 407. The critical question is not whether the dog sniff occurs before or after the officer issues a ticket, but whether conducting the sniff adds time to the stop. Pp. 5–8. 2. The determination adopted by the District Court that detention for the dog sniff was not independently supported by individualized suspicion was not reviewed by the Eighth Circuit. That question therefore remains open for consideration on remand. P. 9. 741 F. 3d 905, vacated and remanded. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Breyer, Sotomayor, and Kagan, JJ., joined. Kennedy, J., filed a dissenting opinion. Thomas, J., filed a dissenting opinion, in which Alito, J., joined, and in which Kennedy, J., joined as to all but Part III. Alito, J., filed a dissenting opinion.

In Illinois v. Caballes,543 U. S. 405 (2005), this Court held that a dog sniff conducted during a lawful traffic stop does not violate the Fourth Amendment’s proscription of unreasonable seizures. This case presents the question whether the Fourth Amendment tolerates a dog sniff conducted after completion of a traffic stop. We hold that a police stop exceeding the time needed to handle the matter for which the stop was made violates the Constitution’s shield against unreasonable seizures. A seizure justified only by a police-observed traffic violation, therefore, “become[s] unlawful if it is prolonged beyond the time reasonably required to complete th[e] mission” of issuing a ticket for the violation. Id., at 407. The Court so recognized in Caballes, and we adhere to the line drawn in that decision.I Just after midnight on March 27, 2012, police officer Morgan Struble observed a Mercury Mountaineer veer slowly onto the shoulder of Nebraska State Highway 275 for one or two seconds and then jerk back onto the road. Nebraska law prohibits driving on highway shoulders, see Neb. Rev. Stat. §60–6,142 (2010), and on that basis, Struble pulled the Mountaineer over at 12:06 a.m. Struble is a K–9 officer with the Valley Police Department in Ne-braska, and his dog Floyd was in his patrol car that night. Two men were in the Mountaineer: the driver, Dennys Rodriguez, and a front-seat passenger, Scott Pollman. Struble approached the Mountaineer on the passenger’s side. After Rodriguez identified himself, Struble asked him why he had driven onto the shoulder. Rodriguez replied that he had swerved to avoid a pothole. Struble then gathered Rodriguez’s license, registration, and proof of insurance, and asked Rodriguez to accompany him to the patrol car. Rodriguez asked if he was required to do so, and Struble answered that he was not. Rodriguez decided to wait in his own vehicle. After running a records check on Rodriguez, Struble returned to the Mountaineer. Struble asked passenger Pollman for his driver’s license and began to question him about where the two men were coming from and where they were going. Pollman replied that they had traveled to Omaha, Nebraska, to look at a Ford Mustang that was for sale and that they were returning to Norfolk, Ne-braska. Struble returned again to his patrol car, where he completed a records check on Pollman, and called for a second officer. Struble then began writing a warning ticket for Rodriguez for driving on the shoulder of the road. Struble returned to Rodriguez’s vehicle a third time to issue the written warning. By 12:27 or 12:28 a.m., Struble had finished explaining the warning to Rodriguez, and had given back to Rodriguez and Pollman the documents obtained from them. As Struble later testified, at that point, Rodriguez and Pollman “had all their documents back and a copy of the written warning. I got all the reason[s] for the stop out of the way[,] . . . took care of all the business.” App. 70. Nevertheless, Struble did not consider Rodriguez “free to leave.” Id., at 69–70. Although justification for the traffic stop was “out of the way,” id., at 70, Struble asked for permission to walk his dog around Rodriguez’s vehicle. Rodriguez said no. Struble then instructed Rodriguez to turn off the ignition, exit the vehicle, and stand in front of the patrol car to wait for the second officer. Rodriguez complied. At 12:33 a.m., a deputy sheriff arrived. Struble retrieved his dog and led him twice around the Mountaineer. The dog alerted to the presence of drugs halfway through Struble’s second pass. All told, seven or eight minutes had elapsed from the time Struble issued the written warning until the dog indicated the presence of drugs. A search of the vehicle revealed a large bag of methamphetamine. Rodriguez was indicted in the United States District Court for the District of Nebraska on one count of possession with intent to distribute 50 grams or more of methamphetamine, in violation of 21 U. S. C. §§841(a)(1) and (b)(1). He moved to suppress the evidence seized from his car on the ground, among others, that Struble had prolonged the traffic stop without reasonable suspicion in order to conduct the dog sniff. After receiving evidence, a Magistrate Judge recommended that the motion be denied. The Magistrate Judge found no probable cause to search the vehicle independent of the dog alert. App. 100 (apart from “information given by the dog,” “Officer Struble had [no]thing other than a rather large hunch”). He further found that no reasonable suspicion supported the detention once Struble issued the written warning. He concluded, however, that under Eighth Circuit precedent, extension of the stop by “seven to eight minutes” for the dog sniff was only a de minimis intrusion on Rodriguez’s Fourth Amendment rights and was therefore permissible. The District Court adopted the Magistrate Judge’s factual findings and legal conclusions and denied Rodriguez’s motion to suppress. The court noted that, in the Eighth Circuit, “dog sniffs that occur within a short time following the completion of a traffic stop are not constitutionally prohibited if they constitute only de minimis intrusions.” App. 114 (quoting United States v. Alexander, 448 F. 3d 1014, 1016 (CA8 2006)). The court thus agreed with the Magistrate Judge that the “7 to 10 minutes” added to the stop by the dog sniff “was not of constitu-tional significance.” App. 114. Impelled by that decision, Rodriguez entered a conditional guilty plea and was sentenced to five years in prison. The Eighth Circuit affirmed. The “seven- or eight-minute delay” in this case, the opinion noted, resembled delays that the court had previously ranked as permissible. 741 F. 3d 905, 907 (2014). The Court of Appeals thus ruled that the delay here constituted an acceptable “de minimis intrusion on Rodriguez’s personal liberty.” Id., at 908. Given that ruling, the court declined to reach the question whether Struble had reasonable suspicion to continue Rodriguez’s detention after issuing the written warning. We granted certiorari to resolve a division among lower courts on the question whether police routinely may extend an otherwise-completed traffic stop, absent reason-able suspicion, in order to conduct a dog sniff. 573 U. S. ___ (2014). Compare, e.g., United States v. Morgan, 270 F. 3d 625, 632 (CA8 2001) (postcompletion delay of “well under ten minutes” permissible), with, e.g., State v. Baker, 2010 UT 18, ¶13, 229 P. 3d 650, 658 (2010) (“[W]ithout additional reasonable suspicion, the officer must allow the seized person to depart once the purpose of the stop has concluded.”).II A seizure for a traffic violation justifies a police investigation of that violation. “[A] relatively brief encounter,” a routine traffic stop is “more analogous to a so-called ‘Terry stop’ . . . than to a formal arrest.” Knowles v. Iowa,525 U. S. 113,117 (1998) (quoting Berkemer v. McCarty,468 U. S. 420,439 (1984), in turn citing Terry v. Ohio,392 U. S. 1 (1968)). See also Arizona v. Johnson,555 U. S. 323,330 (2009). Like a Terry stop, the tolerable duration of police inquiries in the traffic-stop context is determined by the seizure’s “mission”—to address the traffic violation that warranted the stop, Caballes, 543 U. S., at 407, and attend to related safety concerns, infra, at 6–7. See also United States v. Sharpe,470 U. S. 675,685 (1985); Florida v. Royer,460 U. S. 491,500 (1983) (plurality opinion) (“The scope of the detention must be carefully tailored to its underlying justification.”). Because addressing the infraction is the purpose of the stop, it may “last no longer than is necessary to effectuate th[at] purpose.” Ibid. See also Caballes, 543 U. S., at 407. Authority for the seizure thus ends when tasks tied to the traffic infraction are—or reasonably should have been—completed. See Sharpe, 470 U. S., at 686 (in determining the reasonable duration of a stop, “it [is] appropriate to examine whether the police diligently pursued [the] investigation”). Our decisions in Caballes and Johnson heed these constraints. In both cases, we concluded that the Fourth Amendment tolerated certain unrelated investigations that did not lengthen the roadside detention. Johnson, 555 U. S., at 327–328 (questioning); Caballes, 543 U. S., at 406, 408 (dog sniff). In Caballes, however, we cautioned that a traffic stop “can become unlawful if it is prolonged beyond the time reasonably required to complete th[e] mission” of issuing a warning ticket. 543 U. S., at 407. And we repeated that admonition in Johnson: The seizure remains lawful only “so long as [unrelated] inquiries do not measurably extend the duration of the stop.” 555 U. S., at 333. See also Muehler v. Mena,544 U. S. 93,101 (2005) (because unrelated inquiries did not “exten[d] the time [petitioner] was detained[,] . . . no additional Fourth Amendment justification . . . was required”). An officer, in other words, may conduct certain unrelated checks during an otherwise lawful traffic stop. But contrary to Justice Alito’s suggestion, post, at 4, n. 2, he may not do so in a way that prolongs the stop, absent the reasonable suspicion ordinarily demanded to justify detaining an individ-ual. But see post, at 1–2 (Alito, J., dissenting) (premising opinion on the dissent’s own finding of “reasonable suspicion,” although the District Court reached the opposite conclusion, and the Court of Appeals declined to consider the issue). Beyond determining whether to issue a traffic ticket, an officer’s mission includes “ordinary inquiries incident to [the traffic] stop.” Caballes, 543 U. S., at 408. Typically such inquiries involve checking the driver’s license, determining whether there are outstanding warrants against the driver, and inspecting the automobile’s registration and proof of insurance. See Delaware v. Prouse,440 U. S. 648–660 (1979). See also 4 W. LaFave, Search and Seizure §9.3(c), pp. 507–517 (5th ed. 2012). These checks serve the same objective as enforcement of the traffic code: ensuring that vehicles on the road are operated safely and responsibly. See Prouse, 440 U. S., at 658–659; LaFave, Search and Seizure §9.3(c), at 516 (A “warrant check makes it possible to determine whether the apparent traffic violator is wanted for one or more previous traffic offenses.”). A dog sniff, by contrast, is a measure aimed at “detect[ing] evidence of ordinary criminal wrongdoing.” Indianapolis v. Edmond,531 U. S. 32–41 (2000). See also Florida v. Jardines,569 U. S. 1, ___–___ (2013) (slip op., at 7–8). Candidly, the Government acknowledged at oral argument that a dog sniff, unlike the routine measures just mentioned, is not an ordinary incident of a traffic stop. See Tr. of Oral Arg. 33. Lacking the same close connection to roadway safety as the ordinary inquiries, a dog sniff is not fairly characterized as part of the officer’s traffic mission. In advancing its de minimis rule, the Eighth Circuit relied heavily on our decision in Pennsylvania v. Mimms,434 U. S. 106 (1977) (per curiam). See United States v. $404,905.00 in U. S. Currency, 182 F. 3d 643, 649 (CA8 1999). In Mimms, we reasoned that the government’s “legitimate and weighty” interest in officer safety outweighs the “de minimis” additional intrusion of requiring a driver, already lawfully stopped, to exit the vehicle. 434 U. S., at 110–111. See also Maryland v. Wilson,519 U. S. 408–415 (1997) (passengers may be required to exit vehicle stopped for traffic violation). The Eighth Circuit, echoed in Justice Thomas’s dissent, believed that the imposition here similarly could be offset by the Government’s “strong interest in interdicting the flow of illegal drugs along the nation’s highways.” $404,905.00 in U. S. Currency, 182 F. 3d, at 649; see post, at 9. Unlike a general interest in criminal enforcement, however, the government’s officer safety interest stems from the mission of the stop itself. Traffic stops are “especially fraught with danger to police officers,” Johnson, 555 U. S., at 330 (internal quotation marks omitted), so an officer may need to take certain negligibly burdensome precautions in order to complete his mission safely. Cf. United States v. Holt, 264 F. 3d 1215, 1221–1222 (CA10 2001) (en banc) (recognizing officer safety justification for criminal record and outstanding warrant checks), abrogated on other grounds as recognized in United States v. Stewart, 473 F. 3d 1265, 1269 (CA10 2007). On-scene investigation into other crimes, however, detours from that mission. See supra, at 6–7. So too do safety precautions taken in order to facilitate such detours. But cf. post, at 2–3 (Alito, J., dissenting). Thus, even assuming that the imposition here was no more intrusive than the exit order in Mimms, the dog sniff could not be justified on the same basis. Highway and officer safety are interests different in kind from the Government’s endeavor to detect crime in general or drug trafficking in particular. The Government argues that an officer may “incremental[ly]” prolong a stop to conduct a dog sniff so long as the officer is reasonably diligent in pursuing the traffic-related purpose of the stop, and the overall duration of the stop remains reasonable in relation to the duration of other traffic stops involving similar circumstances. Brief for United States 36–39. The Government’s argument, in effect, is that by completing all traffic-related tasks expeditiously, an officer can earn bonus time to pursue an unrelated criminal investigation. See also post, at 2–5 (Thomas, J., dissenting) (embracing the Government’s argument). The reasonableness of a seizure, however, depends on what the police in fact do. See Knowles, 525 U. S., at 115–117. In this regard, the Government acknowledges that “an officer always has to be reasonably diligent.” Tr. of Oral Arg. 49. How could diligence be gauged other than by noting what the officer actually did and how he did it? If an officer can complete traffic-based inquiries expeditiously, then that is the amount of “time reasonably required to complete [the stop’s] mission.” Caballes, 543 U. S., at 407. As we said in Caballes and reiterate today, a traffic stop “prolonged beyond” that point is “unlawful.” Ibid. The critical question, then, is not whether the dog sniff occurs before or after the officer issues a ticket, as Justice Alito supposes, post, at 2–4, but whether conducting the sniff “prolongs”—i.e., adds time to—“the stop,” supra, at 6.III The Magistrate Judge found that detention for the dog sniff in this case was not independently supported by individualized suspicion, see App. 100, and the District Court adopted the Magistrate Judge’s findings, see id., at 112–113. The Court of Appeals, however, did not review that determination. But see post, at 1, 10–12 (Thomas, J., dissenting) (resolving the issue, nevermind that the Court of Appeals left it unaddressed); post, at 1–2 (Alito, J., dissenting) (upbraiding the Court for addressing the sole issue decided by the Court of Appeals and characterizing the Court’s answer as “unnecessary” because the Court, instead, should have decided an issue the Court of Appeals did not decide). The question whether reasonable suspicion of criminal activity justified detaining Rodriguez beyond completion of the traffic infraction investigation, therefore, remains open for Eighth Circuit consideration on remand.* * * For the reasons stated, the judgment of the United States Court of Appeals for the Eighth Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion.It is so ordered.

574.US.293

Respondent Roswell’s city council (Council) held a public hearing to consider an application by petitioner T-Mobile South, LLC, to build a cell phone tower on residential property. During the hearing, several Council members expressed concerns about the tower’s impact on the area. The hearing ended with the Council unanimously passing a motion to deny the application. Two days later, the City’s Planning and Zoning Division informed petitioner by letter that the application had been denied and that minutes from the hearing would be made available. The detailed minutes were published 26 days later. Petitioner filed suit, alleging that the Council’s denial was not supported by substantial evidence in the record. The District Court agreed, concluding that the City, by failing to issue a written decision stating its reasons for denying the application, had violated the Telecommunications Act of 1996, which provides that a locality’s denial “shall be in writing and supported by substantial evidence contained in a written record,” 47 U. S. C. §332(c)(7)(B)(iii). The Eleventh Circuit, following its precedent, found that the Act’s requirements were satisfied here because petitioner had received a denial letter and possessed a transcript of the hearing that it arranged to have recorded. Held: 1. Section 332(c)(7)(B)(iii) requires localities to provide reasons when they deny applications to build cell phone towers. This conclusion follows from the Act’s provisions, which both preserve and specifically limit traditional state and local government authority. It would be considerably difficult for a reviewing court to determine whether a locality’s denial was “supported by substantial evidence contained in a written record,” §332(c)(7)(B)(iii), or whether a locality had “unreasonably discriminate[d] among providers of functionally equivalent services,” §332(c)(7)(B)(i)(I), or regulated siting “on the basis of the environmental effects of radio frequency emissions,” §332(c)(7)(B)(iv), if localities were not obligated to state their reasons for denial. And nothing in the Act suggests that Congress meant to use the phrase “substantial evidence” as anything but an administrative law “term of art” that describes how “an administrative record is to be judged by a reviewing court.” United States v. Carlo Bianchi & Co., 373 U.S. 709, 715. Pp. 6–8. 2. Localities are not required to provide their reasons for denying siting applications in the denial notice itself, but may state those reasons with sufficient clarity in some other written record issued essentially contemporaneously with the denial. Pp. 8–13. (a) Nothing in the Act’s text imposes a requirement that the reasons be given in any particular form, and the Act’s saving clause, §332(c)(7)(A), makes clear that the only limitations imposed on local governments are those enumerated in the statute. Localities comply with their obligation to give written reasons so long as those reasons are stated clearly enough to enable judicial review. Because an adversely affected entity must decide whether to seek judicial review within 30 days from the date of the denial, §332(c)(7)(B)(v), and because a court cannot review the denial without knowing the locality’s reasons, the locality must provide or make available its written reasons at essentially the same time as it communicates its denial. Pp. 8–11. (b) Petitioner’s contrary arguments are unavailing. The statute’s word “decision” does not connote a written document that itself provides all the reasons for a given judgment. The absence of the word “notify” in the provision at issue also does not signal an intention to require communication of more than a judgment. Nor does an obligation to provide reasons in the writing conveying the denial arise from the “substantial evidence” requirement itself or from the requirement of court review “on an expedited basis,” §332(c)(7)(B)(v). It is sufficient that a locality’s reasons be provided in a manner that is clear enough and prompt enough to enable judicial review. Pp. 11–13. 3. The City failed to comply with its statutory obligations under the Act. Although it issued its reasons in writing and did so in an acceptable form, it did not provide its written reasons essentially contemporaneously with its written denial when it issued detailed minutes 26 days after the date of the written denial and 4 days before expiration of petitioner’s time to seek judicial review. P. 14. 731 F.3d 1213, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Scalia, Kennedy, Breyer, Alito, and Kagan, JJ., joined. Alito, J., filed a concurring opinion. Roberts, C. J., filed a dissenting opinion in which Ginsburg, J., joined, and in which Thomas, J., joined as to Part I. Thomas, J., filed a dissenting opinion.

The Telecommunications Act of 1996 provides, in relevant part, that “[a]ny decision by a State or local government or instrumentality thereof to deny a request to place, construct, or modify personal wireless service facilities shall be in writing and supported by substantial evidence contained in a written record.”110Stat.151,47 U. S. C. §332(c)(7)(B)(iii). The question presented is whether, and in what form, localities must provide reasons when they deny telecommunication companies’ applications to construct cell phone towers. We hold that localities must provide or make available their reasons, but that those reasons need not appear in the written denial letter or notice provided by the locality. Instead, the locality’s reasons may appear in some other written record so long as the reasons are sufficiently clear and are provided or made accessible to the applicant essentially contemporaneously with the written denial letter or notice.I In February 2010, petitioner T-Mobile South, LLC, applied to build a new, 108-foot-tall cell phone tower on 2.8 acres of vacant residential property in the city of Ros-well, Georgia (City). Roswell’s city ordinances requirethat any cell phone tower proposed for a residential zoning district must take the form of an “alternative tower structure”—an artificial tree, clock tower, steeple, or light pole—that, in the opinion of the city council (City Council or Council), is “compatible with the natural setting and surrounding structures” and that effectively camouflages the tower. Code of Ordinances §§21.2.2, 21.2.5(a); see App. 68, 75. In accordance with these provisions, petitioner’s application proposed a structure in the shape of an artificial tree or “monopine.” Id., at 42. The City’s Planning and Zoning Division reviewed petitioner’s application, along with a substantial number of letters and petitions opposing it, and ultimately issued a memorandum to the City Council concluding that the application met all of the requirements set out in the City’s ordinances. It recommended that the City Council approve the application on three conditions to which petitioner was prepared to agree. The City Council then held a 2-hour-long public hearing on April 12, 2010, to consider petitioner’s application. Petitioner arranged privately to have the hearing transcribed, and, as discussed below, the City subsequently issued detailed minutes summarizing the proceedings. At the hearing, after the Planning and Zoning Division presented its recommendation and after petitioner’s representatives made a presentation in support of the application, a number of residents raised concerns. Among these were concerns that the tower would lack aesthetic compatibility, that the technology was outdated and unnecessary, and that the tower would be too tall. Petitioner’s representatives responded by reiterating that it had met all of the ordinance’s requirements and by providing testimony from a property appraiser that placement of cell phone towers does not reduce property values. Members of the City Council then commented on the application. One member of the six-person Council was recused, see id., at 111 (hearing transcript); id., at 322 (meeting minutes), leaving five voting members. Member Igleheart said that other carriers had sufficient coverage in the area and that the City did not need to level the playing field for petitioner. Id., at 173–174 (hearing transcript). He also stated that his “[b]ottom line” was that he did not think it was “appropriate for residentially zoned properties to have the cell towers in their location.” Id., at 174 (hearing transcript); id., at 338 (meeting minutes). Member Dippolito found it difficult to believe that the tower would not negatively impact the area and doubted that it would be compatible with the natural setting. Id., at 175–176 (hearing transcript); id., at 339 (meeting minutes). Member Wynn expressed concerns about the lack of a backup generator for emergency services, id., at 172 (hearing transcript), and did not think the tower would be “compatible with this area,” id., at 176 (hearing transcript); id., at 339 (meeting minutes). Member Orlans said only that he was impressed with the information put together by both sides. Id., at 173 (hearing transcript); id., at 337 (meeting minutes). Finally, Member Price, the liaison to the Planning and Zoning Division, made a motion to deny the application. She said that the tower would be aesthetically incompatible with the natural setting, that it would be too tall, and that its proximity to other homes would adversely affect the neighbors and the resale value of their properties. Id., at 176–177 (hearing transcript); id., at 339–340 (meeting minutes). The motion was seconded, and then passed unanimously. Id., at 177 (hearing transcript); id., at 340 (meeting minutes). Two days later, on April 14, 2010, the Planning and Zoning Division sent a letter to petitioner that said in its entirety: “Please be advised the City of Roswell Mayor and City Council denied the request from T-Mobile for a 108’ mono-pine alternative tower structure during their April 12, 2010 hearing. The minutes from the aforementioned hearing may be obtained from the city clerk. Please contact Sue Creel or Betsy Branch at [phone number]. “If you have any additional questions, please contact me at [phone number].” Id., at 278.The detailed written minutes of the hearing, however, were not approved and published by the City until 26 days later, on May 10, 2010. See id., at 321–341 (meeting minutes).[1] On May 13, 2010, 3 days after the detailed minutes were published—and now 29 days after the City denied petitioner’s application—petitioner filed suit in Federal District Court. It alleged that the denial of the application was not supported by substantial evidence in the record, and would effectively prohibit the provision of wireless service in violation of the Telecommunications Act of1996 (Act). The parties filed cross-motions for summary judgment. The District Court granted petitioner’s motion for summary judgment, concluding that the City had violated the Act when it failed to issue a written decision that stated the reasons for denying petitioner’s application. The District Court interpreted the Act to require that a written denial letter or notice describe the reasons for the denial and that those reasons be sufficiently explained to allow a reviewing court to evaluate them against the written record. The Eleventh Circuit reversed. 731 F. 3d 1213 (2013). It explained that, in T-Mobile South, LLC v. Milton, 728 F. 3d 1274 (2013), which was decided after the District Court’s decision in this case, it had held that “to the extent that the decision must contain grounds or reasons or explanations, it is sufficient if those are contained in a different written document or documents that the applicant is given or has access to.” Id., at 1285. The Eleventh Circuit acknowledged that the Courts of Appeals had split on that question, and that it had departed from the majority rule. Compare Southwestern Bell Mobile Systems, Inc. v. Todd, 244 F. 3d 51, 60 (CA1 2001) (requiring that a locality issue a written denial that itself contains a “sufficient explanation of the reasons for the permit denial to allow a reviewing court to evaluate the evidence in the record supporting those reasons”); New Par v. Saginaw, 301 F. 3d 390, 395–396 (CA6 2002); MetroPCS, Inc. v. City and County of San Francisco, 400 F. 3d 715, 723 (CA9 2005), with AT&T Wireless PCS, Inc. v. City Council of Virginia Beach, 155 F. 3d 423, 429 (CA4 1998) (holding that written minutes of a meeting and the word “denied” stamped on a letter describing the application weresufficient). Applying its rule to this case, the Eleventh Circuit found that the requirements of47 U. S. C. §332(c)(7)(B)(iii) were satisfied because petitioner had its own transcript as well as a written letter stating that the application had been denied and informing petitioner that it could obtain access to the minutes of the hearing. 731 F. 3d, at 1221. It did not consider when the City provided its written reasons to petitioner. We granted certiorari, 572 U. S. ___ (2014), and now reverse the judgment of the Eleventh Circuit.IIA The first question we answer is whether the statute requires localities to provide reasons when they deny applications to build cell phone towers. We answer that question in the affirmative. Our conclusion follows from the provisions of the Telecommunications Act. The Act generally preserves “the traditional authority of state and local governments to regulate the location, construction, and modification” of wireless communications facilities like cell phone towers, but imposes “specific limitations” on that authority. Rancho Palos Verdes v. Abrams,544 U. S. 113,115 (2005); see §332(c)(7)(B). One of those limitations is that any decision to deny a request to build a tower “shall be in writing and supported by substantial evidence contained in a written record.” §332(c)(7)(B)(iii). Another is that parties adversely affected by a locality’s decision may seek judicial review. §332(c)(7)(B)(v). In order to determine whether a locality’s denial was supported by substantial evidence, as Congress directed, courts must be able to identify the reason or reasons why the locality denied the application. See Rancho Palos Verdes, 544 U. S., at 128 (Breyer, J., joined by O’Connor, Souter, and Ginsburg, JJ., concurring) (observing that the Act “requires local zoning boards . . . [to] give reasons for [their] denials ‘in writing’ ”). The requirement that localities must provide reasons when they deny applications is further underscored by two of the other limitations on local authority set out in the Act. The Act provides that localities “shall not unreason-ably discriminate among providers of functionally equivalent services,” and may not regulate the construction of personal wireless service facilities “on the basis of the environmental effects of radio frequency emissions to the extent that such facilities comply with the [Federal Communications Commission’s] regulations concerning such emissions.” §§332(c)(7)(B)(i)(I), (iv).[2] Again, it would be considerably more difficult for a reviewing court to determine whether a locality had violated these substantive provisions if the locality were not obligated to state its reasons. This conclusion is not just commonsensical, but flows directly from Congress’ use of the term “substantial evidence.” The statutory phrase “substantial evidence” is a “term of art” in administrative law that describes how “an administrative record is to be judged by a reviewing court.” United States v. Carlo Bianchi & Co.,373 U. S. 709,715 (1963). There is no reason discernible from the text of the Act to think that Congress meant to use the phrase in a different way. See FAA v. Cooper, 566 U. S. ___, ___ (2012) (slip op., at 6) (“[W]hen Congress employs a term of art, it presumably knows and adopts the cluster of ideas that were attached to each borrowed word in the body of learning from which it was taken” (internal quotation marks omitted)). Indeed, for those who consider legislative history relevant, the Conference Report accompanying the Act confirmed as much when it noted that “[t]he phrase ‘substantial evidence contained in a written record’ is the traditional standard used for review of agency actions.” H. R. Conf. Rep. No. 104–458, p. 208 (1996). By employing the term “substantial evidence,” Congress thus invoked, among other things, our recognition that “the orderly functioning of the process of [substantial-evidence] review requires that the grounds upon which the administrative agency acted be clearly disclosed,” and that “courts cannot exercise their duty of [substantial-evidence] review unless they are advised of the considerations underlying the action under review.” SEC v. Chenery Corp.,318 U. S. 80,94 (1943); see also Motor Vehicle Mfrs. Assn. of United States, Inc. v. State Farm Mut. Automobile Ins. Co.,463 U. S. 29,43 (1983) (explaining that an agency must “articulate a satisfactory explanation for its action” to enable substantial-evidence review); Beaumont, S. L. & W. R. Co. v. United States,282 U. S. 74,86 (1930) (“Complete statements by the [agency] showing the grounds upon which its determinations rest are quite as necessary as are opinions of lower courts setting forth the reasons on which they base their decisions . . .”). In response, the City primarily argues that a reason-giving obligation would deprive it of local zoning author-ity. But Congress intended to place “specific limitations on the traditional authority of state and local governments” regarding cell phone tower siting applications. Rancho Palos Verdes, 544 U. S., at 115. One of those “limitations,” §332(c)(7)(B), necessarily implied by the Act’s “substantial evidence” requirement, is that local zoning authorities state their reasons when they deny applications. In short, the statutory text and structure, and the concepts that Congress imported into the statutory framework, all point clearly toward the conclusion that localities must provide reasons when they deny cell phone tower siting applications. We stress, however, that these reasons need not be elaborate or even sophisticated, but rather, as discussed below, simply clear enough to enable judicial review.B The second question we answer is whether these reasons must appear in the same writing that conveys the locality’s denial of an application. We answer that question in the negative. Like our conclusion that localities must provide reasons, our conclusion that the reasons need not appear in a de-nial letter follows from the statutory text. Other than providing that a locality’s reasons must be given in writing, nothing in that text imposes any requirement that the reasons be given in any particular form. The Act’s saving clause makes clear that, other than the enumerated limitations imposed on local governments by the statute itself, “nothing in this chapter shall limit or affect the authority of a State or local government or instrumentality thereof over decisions regarding the placement, construction, and modification of personal wireless service facilities.” §332(c)(7)(A). Given this language, and the system of “cooperative federalism” on which the Act is premised, Rancho Palos Verdes, 544 U. S., at 128 (Breyer, J., concurring), we understand the enumerated limitations to set out an exclusive list. So while the text and structure of the Act render it inescapable that localities must provide reasons in writing when they deny applications, we can locate in the Act no command—either explicit or implicit—that localities must provide those reasons in a specific document. We therefore conclude that Congress imposed no specific requirement on that front, but instead permitted localities to comply with their obligation to give written reasons so long as the locality’s reasons are stated clearly enough to enable judicial review. Although the statute does not require a locality to provide its written reasons in any particular format, and although a locality may rely on detailed meeting minutes as it did here, we agree with the Solicitor General that “the local government may be better served by including a separate statement containing its reasons.” Brief for United States as Amicus Curiae 26; see also id., at 34. If the locality writes a short statement providing its reasons, the locality can likely avoid prolonging the litigation—and adding expense to the taxpayers, the companies, and the legal system—while the parties argue about exactly what the sometimes voluminous record means. Moreover, in that circumstance, the locality need not worry that, upon review of the record, a court will either find that it could not ascertain the locality’s reasons or mistakenly ascribe to the locality a rationale that was not in fact the reason for the locality’s denial. We hasten to add that a locality cannot stymie or burden the judicial review contemplated by the statute by delaying the release of its reasons for a substantial time after it conveys its written denial. The statute provides that an entity adversely affected by a locality’s decision may seek judicial review within 30 days of the decision. §332(c)(7)(B)(v). Because an entity may not be able to make a considered decision whether to seek judicial review without knowing the reasons for the denial of its application, and because a court cannot review the denial without knowing the locality’s reasons, the locality must provide or make available its written reasons at essentially the same time as it communicates its denial.[3] This rule ought not to unduly burden localities given the range of ways in which localities can provide their reasons. Moreover, the denial itself needs only to be issued (or the application otherwise acted upon) “within a reasonable period of time.” §332(c)(7)(B)(ii). In an interpretation we have recently upheld, see Arlington v. FCC, 569 U. S. ___ (2013), the Federal Communications Commission (FCC) has generally interpreted this provision to allow localities 90 days to act on applications to place new antennas on existing towers and 150 days to act on other siting applications. In re Petition for Declaratory Ruling to Clarify Provisions of Section 332(c)(7)(b), 24 FCC Rcd. 13994, 13995, ¶4 (2009). If a locality is not in a position to provide its reasons promptly, the locality can delay the issuance of its denial within this 90- or 150-day window, and instead release it along with its reasons once those reasons are ready to be provided. Only once the denial is issued would the 30-day commencement-of-suit clock begin.[4]III Petitioner offers four reasons why, in its view, our analysis in Part II–B is incorrect. Petitioner argues that the statute requires that a locality’s reasons appear in the writing conveying the denial itself, but none of petitioner’s reasons are persuasive. First, petitioner argues that the word “decision” in the statute—the thing that must be “in writing”—connotes a written document that itself provides all the reasons for a given judgment. See Brief for Petitioner 24 (quoting Black’s Law Dictionary 407 (6th ed. 1990) (a “decision” is a written document providing “ ‘the reasons given for [a] judgment’ ”)). But even petitioner concedes, with its preferred dictionary in hand, that the word “decision” can also mean “something short of a statement of reasons explaining a determination.” Brief for Petitioner 24 (citing Black’s Law Dictionary, at 407).[5] Second, petitioner claims that other provisions in the Act use the word “notify” when the Act means to impose only a requirement that a judgment be communicated.[6] Because the provision at issue here does not use the word “notify,” petitioner argues, it must contemplate something more than a judgment. This does not logically follow. For one thing, the statute at issue here does not use any verb at all to describe the conveying of information from a locality to an applicant; it just says that a denial “shall be in writing and supported by substantial evidence contained in a written record.” §332(c)(7)(B)(iii). But more to the point, “notify” is a verb the use—or nonuse—of which does not reveal what the thing to be notified of or about is. Third, petitioner contends that the “substantial evidence” requirement itself demands that localities identify their reasons in their written denials. See Brief for Petitioner 23. Certainly, as discussed above, the phrase “substantial evidence” requires localities to give reasons, but it says nothing on its own about the document in which those reasons must be stated or presented to a reviewing court. Finally, petitioner invokes the statutory requirement that any adversely affected person shall have their challenge heard by a court “on an expedited basis.” §332(c)(7)(B)(v). See Brief for Petitioner 14–15, 28. As long as the reasons are provided in a written record, however, and as long as they are provided in such a manner that is clear enough and prompt enough to enable judicial review, there is no reason to require that those reasons be provided in the written denial itself. We acknowledge that petitioner, along with those Courts of Appeals that have required a locality’s reasons to appear in its written denial itself, have offered plausible bases for a rule that would require as much. See, e.g., Todd, 244 F. 3d, at 60 (“A written record can create difficulties in determining the rationale behind a board’s decision . . .”). Congress could adopt such a rule if it were so inclined, but it did not do so in this statute. It is not our place to legislate another approach.IV Thus, we hold that the Act requires localities to provide reasons when they deny cell phone tower siting applications, but that the Act does not require localities to provide those reasons in written denial letters or notices themselves. A locality may satisfy its statutory obligations if it states its reasons with sufficient clarity in some other written record issued essentially contemporaneously with the denial. In this case, the City provided its reasons in writing and did so in the acceptable form of detailed minutes of the City Council meeting. The City, however, did not provide its written reasons essentially contemporaneously with its written denial. Instead, the City issued those detailed minutes 26 days after the date of the written denial and just 4 days before petitioner’s time to seek judicial review would have expired.[7] The City therefore did not comply with its statutory obligations. We do not consider questions regarding the applicability of principles of harmless error or questions of remedy, and leave those for the Eleventh Circuit to address on remand.* * * For the foregoing reasons, we reverse the judgment below and remand the case for further proceedings consistent with this opinion.It is so ordered.Notes1 Brief minutes had been adopted on April 19, but these only noted that the motion to deny the application had passed with five members in favor and one member recused. See Council Brief 041210, online at http:// roswell.legistar.com / LegislationDetail.aspx?ID=657578&GUID=08D5297C-0271-41F9-9DAA-E8E3DD6314BD&Options=&Search= (all In-ternet Materials as visited January 12, 2015, and available in Clerk of Court’s case file). According to the meeting calendar for the City Council’s May 10, 2010, meeting, it was on that day that the City Council approved detailed minutes of the April 12 meeting that in-cluded a recitation of each member’s statements during the hearing. See http :/ / roswell.legistar.com / MeetingDetail.aspx?ID = 101786&GUID = 63828B21-EB83-4485-B4EA-10EE65CF48CD&Options=info|&Search=.2 The last “limitation” listed in the Act provides that localities shall act on applications to construct personal wireless service facilities “within a reasonable period of time after the request is duly filed . . . taking into account the nature and scope of such request.” §332(c)(7)(B)(ii).3 The Chief Justice’s dissent rejects this particular requirement, and instead invents a process that turns judicial review on its head. Rather than give effect to a process that would permit an entity seeking to challenge a locality’s decision to see the locality’s written reasons before it files its suit—and the dissent agrees that the statute requires that a locality convey its reasons in writing, see post, at 5—the dissent would fashion a world in which a locality can wait until a lawsuit is commenced and a court orders it to state its reasons. The entity would thus be left to guess at what the locality’s written reasons will be, write a complaint that contains those hypotheses, and risk being sandbagged by the written reasons that the locality subsequently provides in litigation after the challenging entity has shown its cards. The reviewing court would then need to ensure that those reasons are not post hoc rationalizations, see Burlington Truck Lines, Inc. v. United States,371 U. S. 156,168 (1962), but the dissent offers no guidance as to how a reviewing court that has never seen near-contemporaneous reasons would conduct that inquiry.4 The City urges us to hold that the clock does not begin to run until after the reasons are given. We cannot so hold, however, without rewriting the statutory text. The Act provides that a lawsuit may be filed by “[a]ny person adversely affected by any final action or failure to act . . . within 30 days after such action or failure to act.”47 U. S. C. §332(c)(7)(B)(v). The relevant “final action” is the issuance of the written notice of denial, not the subsequent issuance of reasons explaining the denial. See Bennett v. Spear,520 U. S. 154–178 (1997) (agency action is “final” if it “mark[s] the consummation of the agency’s decisionmaking process” and determines “rights or obligations” or triggers “legal consequences” (internal quotation marks omitted)).5 One of petitioner’s amici argues that Congress has used the word “decision” in the context of other communications laws to mean something more than a judgment or verdict. See Brief for Chamber of Commerce of the United States of America (Chamber) et al. 9–13. But while it is true that a word used across “the same act” should be given the same meaning, see Taniguchi v. Kan Pacific Saipan, Ltd., 566 U. S. ___, ___ (2012) (slip op., at 10), the Chamber’s evidence is less persuasive because it arises out of entirely different “acts” and does not involve any term of art. By relying on other parts of Title 47 of the U. S. Code—some enacted in the Communications Act of 1934 decades before the enactment of the Telecommunications Act of 1996 at issue here—the Chamber stretches to invoke this canon of construction beyond its most forceful application. See A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 172–173 (2012).6 For example, petitioner cites §11 (FCC must “notify the partiesconcerned” when it makes a “determination and order” regarding a rail-road or telegraph company’s failure to maintain and operate a telegraph line for public use) and §398(b)(5) (“Whenever the Secretary [of Commerce] makes a final determination . . . that a recipient” of federal funds has engaged in impermissible discrimination, the Secretary shall “notify the recipient in writing of such determination . . .”). Brief for Petitioner 24–25.7 Though petitioner arranged for a transcript of the meeting to be recorded on its own initiative and at its own expense, see App. 109–275, the fact that petitioner took steps to reduce oral statements made at the City Council meeting to writing cannot be said to satisfy the obligation that Congress placed on the City to state clearly its reasons, and to do so in a writing it provides or makes available.

574.US.318

Petitioners, Teva Pharmaceuticals (and related firms), own a patent that covers a manufacturing method for the multiple sclerosis drug Copaxone. When respondents, Sandoz, Inc. (and other firms), tried to market a generic version of the drug, Teva sued them for patent infringement. Sandoz countered that the patent was invalid. Specifically, Sandoz argued that the claim that Copaxone’s active ingredient had “a molecular weight of 5 to 9 kilodaltons” was fatally indefinite, see 35 U. S. C. §112 ¶2, because it did not state which of three methods of calculation—the weight of the most prevalent molecule, the weight as calculated by the average weight of all molecules, or weight as calculated by an average in which heavier molecules count for more—was used to determine that weight. After considering conflicting expert evidence, the District Court concluded that the patent claim was sufficiently definite and the patent was thus valid. As relevant here, it found that in context a skilled artisan would understand that the term “molecular weight” referred to molecular weight as calculated by the first method. In finding the “molecular weight” term indefinite and the patent invalid on appeal, the Federal Circuit reviewed de novo all aspects of the District Court’s claim construction, including the District Court’s determination of subsidiary facts. Held: When reviewing a district court’s resolution of subsidiary factual matters made in the course of its construction of a patent claim, the Federal Circuit must apply a “clear error,” not a de novo, standard of review. Pp. 4–16. (a) Federal Rule of Civil Procedure 52(a)(6) states that a court of appeals “must not . . . set aside” a district court’s “[f]indings of fact” unless they are “clearly erroneous.” It sets out a “clear command,” Anderson v. Bessemer City, 470 U.S. 564, 574, and “does not make exceptions or . . . exclude certain categories of factual findings” from the court of appeals’ obligation, Pullman-Standard v. Swint, 456 U.S. 273, 287. The Rule thus applies to both subsidiary and ultimate facts. Ibid. And the function of an appeals court reviewing the findings of a “ ‘district court sitting without a jury . . . is not to decide factual issues de novo.’ ” Anderson, supra, at 573. Even if exceptions to the Rule were permissible, there is no convincing ground for creating an exception here. Markman v. Westview Instruments, Inc., 517 U.S. 370, neither created, nor argued for, such an exception. There, the Court held that the ultimate question of claim construction is for the judge, not the jury, id., at 372, but it did not thereby create an exception from the ordinary rule governing appellate review of factual matters. Instead, the Court pointed out that a judge, in construing a patent claim, is engaged in much the same task as the judge would be in construing other written instruments, such as deeds, contracts, or tariffs. Id., at 384, 386, 388, 389. Construction of written instruments often presents a “question solely of law,” at least when the words in those instruments are “used in their ordinary meaning.” Great Northern R. Co. v. Merchants Elevator Co., 259 U.S. 285, 291. But if a written instrument uses “technical words or phrases not commonly understood,” id., at 292, those words may give rise to a factual dispute. If so, extrinsic evidence may help to “establish a usage of trade or locality.” Ibid. And in that circumstance, the “determination of the matter of fact” will “preced[e]” the “function of construction.” Ibid. The Markman Court also recognized that courts will sometimes have to resolve subsidiary factual disputes in patent construction; Rule 52 requires appellate courts to review such disputes under the “clearly erroneous” standard. Application of this standard is further supported by precedent and by practical considerations. Clear error review is “particularly” important in patent cases because a district court judge who has presided over, and listened to, the entire proceeding has a comparatively greater opportunity to gain the necessary “familiarity with specific scientific problems and principles,” Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 610, than an appeals court judge who must read a written transcript or perhaps just those portions referenced by the parties. Pp. 4–8. (b) Arguments to the contrary are unavailing. Sandoz claims that separating “factual” from “legal” questions may be difficult and, like the Federal Circuit, posits that it is simpler for the appellate court to review the entirety of the district court’s claim construction de novo than to apply two separate standards. But courts of appeals have long been able to separate factual from legal matters, see, e.g., First Options of Chicago, Inc. v. Kaplan, 514 U.S. 938, 947–948, and the Federal Circuit’s efforts to treat factual findings and legal conclusions similarly have brought with them their own complexities. As for Sandoz’s argument that “clear error” review will bring about less uniformity, neither the Circuit nor Sandoz has shown that divergent claim construction stemming from divergent findings of fact on subsidiary matters should occur more than occasionally. Pp. 8–11. (c) This leaves the question of how the clear error standard should be applied when reviewing subsidiary factfinding in patent claim construction. When the district court reviews only evidence intrinsic to the patent, the judge’s determination is solely a determination of law, and the court of appeals will review that construction de novo. However, where the district court needs to consult extrinsic evidence in order to understand, for example, the background science or the meaning of a term in the relevant art during the relevant time period, and where those subsidiary facts are in dispute, courts will need to make subsidiary factual findings about the extrinsic evidence. The district judge, after deciding the factual dispute, will then interpret the patent claim in light of the facts as he has found them. The ultimate construction of the claim is a legal conclusion that the appellate court can review de novo. But to overturn the judge’s resolution of an underlying factual dispute, the appellate court must find that the judge, in respect to those factual findings, has made a clear error. Pp. 11–14. (d) Here, for example, the District Court made a factual finding, crediting Teva’s expert’s account, and thereby rejecting Sandoz’s expert’s contrary explanation, about how a skilled artisan would understand the way in which a curve created from chromatogram data reflects molecular weights. Based on that factual finding, the District Court reached the legal conclusion that figure 1 did not undermine Teva’s argument that molecular weight referred to the first method of calculating molecular weight. When the Federal Circuit reviewed the District Court’s decision, it did not accept Teva’s expert’s explanation, and it failed to accept that explanation without finding that the District Court’s contrary determination was “clearly erroneous.” The Federal Circuit erred in failing to review this factual finding only for clear error. Teva asserts that there are two additional instances in which the Federal Circuit rejected the District Court’s factual findings without concluding that they were clearly erroneous; those matters are left for the Federal Circuit to consider on remand. Pp. 14–16. 723 F.3d 1363, vacated and remanded. Breyer, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Alito, J., joined.

In Markman v. Westview Instruments, Inc.,517 U. S. 370 (1996), we explained that a patent claim is that “portion of the patent document that defines the scope of the patentee’s rights.” Id., at 372. We held that “the construction of a patent, including terms of art within its claim,” is not for a jury but “exclusively” for “the court” to determine. Ibid. That is so even where the construction of a term of art has “evidentiary underpinnings.” Id., at 390. Today’s case involves claim construction with “evidentiary underpinnings.” See Part III, infra. And, it requires us to determine what standard the Court of Appeals should use when it reviews a trial judge’s resolution of an underlying factual dispute. Should the Court of Appeals review the district court’s factfinding de novo as it would review a question of law? Or, should it review that factfinding as it would review a trial judge’s factfinding in other cases, namely by taking them as correct “unless clearly erroneous?” See Fed. Rule Civ. Proc. 52(a)(6). We hold that the appellate court must apply a “clear error,” not a de novo, standard of review.I The basic dispute in this case concerns the meaning of the words “molecular weight” as those words appear in a patent claim. The petitioners, Teva Pharmaceuticals (along with related firms), own the relevant patent. The patent covers a manufacturing method for Copaxone, a drug used to treat multiple sclerosis. The drug’s active ingredient, called “copolymer-1,” is made up of molecules of varying sizes. App. 1143a. And the relevant claim describes that ingredient as having “a molecular weight of 5 to 9 kilodaltons.” Id., at 1145a. The respondents, Sandoz, Inc. (and several other firms), tried to market a generic version of Copaxone. Teva sued Sandoz for patent infringement. 810 F. Supp. 2d 578, 581 (SDNY 2011). Sandoz defended the suit by arguing that the patent was invalid. Ibid. The Patent Act requires that a claim “particularly poin[t] out and distinctly clai[m] the subject matter which the applicant regards as his invention.”35 U. S. C. §112 ¶2 (2006 ed.); see Nautilus, Inc. v. Biosig Instruments, Inc., 572 U. S. ___, ___, n. 1 (2014) (slip op., at 3, n. 1)). The phrase “molecular weight of 5 to 9 kilodaltons,” said Sandoz, did not satisfy this requirement. The reason that the phrase is fatally indefinite, Sandoz argued, is that, in the context of this patent claim, the term “molecular weight” might mean any one of three different things. The phrase might refer (1) to molecular weight as calculated by the weight of the molecule that is most prevalent in the mix that makes up copolymer-1. (The scientific term for molecular weight so calculated is, we are told, “peak average molecular weight.”) The phrase might refer (2) to molecular weight as calculated by taking all the different-sized molecules in the mix that makes up copolymer-1 and calculating the average weight, i.e., adding up the weight of each molecule and dividing by the number of molecules. (The scientific term for molecular weight so calculated is, we are told, “number average molecular weight.”) Or, the phrase might refer (3) to molecular weight as calculated by taking all the different-sized molecules in the mix that makes up copolymer-1 and calculating their average weight while giving heavier molecules a weight-related bonus when doing so. (The scientific term for molecular weight so calculated, we are told, is “weight average molecular weight.”) See 723 F. 3d 1363, 1367 (CA Fed. 2013); App. 124a. In Sandoz’s view, since Teva’s patent claim does not say which method of calculation should be used, the claim’s phrase “molecular weight” is indefinite, and the claim fails to satisfy the critical patent law requirement. The District Court, after taking evidence from experts, concluded that the patent claim was sufficiently definite. Among other things, it found that in context a skilled artisan would understand that the term “molecular weight” referred to molecular weight as calculated by the first method, i.e., “peak average molecular weight.” 810 F. Supp. 2d, at 596; see Nautilus, supra, at ___ (slip op., at 12) (“[T]he definiteness inquiry trains on the understanding of a skilled artisan at the time of the patent application”). In part for this reason, the District Court held the patent valid. 810 F. Supp. 2d, at 596. On appeal, the Federal Circuit held to the contrary. It found that the term “molecular weight” was indefinite. And it consequently held the patent invalid. 723 F. 3d, at 1369. In reaching this conclusion, the Federal Circuit reviewed de novo all aspects of the District Court’s claim construction, including the District Court’s determination of subsidiary facts. Id., at 1369, 1373; see also Light-ing Ballast Control LLC v. Philips Electronics NorthAm. Corp., 744 F. 3d 1272, 1276–1277 (CA Fed. 2014) (en banc) (reaffirming de novo review of district court claim construction). Teva filed a petition for certiorari. And we granted that petition. The Federal Circuit reviews the claim construction decisions of federal district courts throughout the Nation, and we consequently believe it important to clarify the standard of review that it must apply when doing so.IIA Federal Rule of Civil Procedure 52(a)(6) states that a court of appeals “must not . . . set aside” a district court’s “[f ]indings of fact” unless they are “clearly erroneous.” In our view, this rule and the standard it sets forth must apply when a court of appeals reviews a district court’s resolution of subsidiary factual matters made in the course of its construction of a patent claim. We have made clear that the Rule sets forth a “clear command.” Anderson v. Bessemer City,470 U. S. 564,574 (1985). “It does not make exceptions or purport to exclude certain categories of factual findings from the obligation of a court of appeals to accept a district court’s findings unless clearly erroneous.” Pullman-Standard v. Swint,456 U. S. 273,287 (1982). Accordingly, the Rule applies to both subsidiary and ultimate facts. Ibid. And we have said that, when reviewing the findings of a “ ‘district court sitting without a jury, appellate courts must constantly have in mind that their function is not to decide factual issues de novo.’ ” Anderson, supra, at 573 (quoting Zenith Radio Corp. v. Hazeltine Research, Inc.,395 U. S. 100,123 (1969)). Even if exceptions to the Rule were permissible, we cannot find any convincing ground for creating an exception to that Rule here. The Rules Advisory Committee pointed out that, in general, exceptions “would tend to undermine the legitimacy of the district courts . . . , multiply appeals . . . , and needlessly reallocate judicial author-ity.” Advisory Committee’s 1985 Note on subd. (a) of Fed. Rule Civ. Proc. 52, 28 U. S. C. App., pp. 908–909; see also Anderson, supra, at 574–575 (de novo review of fac-tual findings “would very likely contribute only negligibly” to accuracy “at a huge cost in diversion of judicialresources”). Our opinion in Markman neither created, nor argued for, an exception to Rule 52(a). The question presented in that case was a Seventh Amendment question: Should a jury or a judge construe patent claims? 517 U. S., at 372. We pointed out that history provides no clear answer. Id., at 388. The task primarily involves the construction of written instruments. Id., at 386, 388, 389. And that task is better matched to a judge’s skills. Id., at 388 (“The construction of written instruments is one of those things that judges often do and are likely to do better than jurors unburdened by training in exegesis”). We consequently held that claim construction falls “exclusively within the province of the court,” not that of the jury. Id., at 372. When describing claim construction we concluded that it was proper to treat the ultimate question of the proper construction of the patent as a question of law in the way that we treat document construction as a question of law. Id., at 388–391. But this does not imply an exception to Rule 52(a) for underlying factual disputes. We used the term “question of law” while pointing out that a judge, in construing a patent claim, is engaged in much the same task as the judge would be in construing other written instruments, such as deeds, contracts, or tariffs. Id., at 384, 386, 388, 389; see also Motion Picture Patents Co. v. Universal Film Mfg. Co.,243 U. S. 502,510 (1917) (patent claims are “aptly likened to the description in a deed, which sets the bounds to the grant which it contains”); Goodyear Dental Vulcanite Co. v. Davis,102 U. S. 222,227 (1880) (analogizing patent construction to the construction of other written instruments like contracts). Construction of written instruments often presents a “question solely of law,” at least when the words in those instruments are “used in their ordinary meaning.” Great Northern R. Co. v. Merchants Elevator Co., 259 U. S. 285, 291 (1922). But sometimes, say when a written instrument uses “technical words or phrases not commonly understood,” id., at 292, those words may give rise to a factual dispute. If so, extrinsic evidence may help to “establish a usage of trade or locality.” Ibid. And in that circumstance, the “determination of the matter of fact” will “preced[e]” the “function of construction.” Ibid.; see also 12 R. Lord, Williston on Contracts §§34:1, p. 2, 34:19, p. 174 (4th ed. 2012) (In contract interpretation, the existence of a “usage”—a “practice or method” in the relevant industry—“is a question of fact” (internal quotation marks omitted)). This factual determination, like all other factual determinations, must be reviewed for clear error. See Pullman-Standard, supra at 287 (The Rule does not “exclude certain categories of factual findings” and applies to both “subsidiary” and “ultimate” facts (internal quotation marks omitted)). Accordingly, when we held in Markman that the ultimate question of claim construction is for the judge and not the jury, we did not create an exception from the ordinary rule governing appellate review of factual matters. Markman no more creates an exception to Rule 52(a) than would a holding that judges, not juries, determine equit-able claims, such as requests for injunctions. A conclusion that an issue is for the judge does not indicate that Rule 52(a) is inapplicable. See Fed. Rule Civ. Proc. 52 (setting the standard of review for “[Factual] Findings and Conclusions by the Court” (emphasis added)). While we held in Markman that the ultimate issue of the proper construction of a claim should be treated as a question of law, we also recognized that in patent construction, subsidiary factfinding is sometimes necessary. Indeed, we referred to claim construction as a practice with “evidentiary underpinnings,” a practice that “falls somewhere between a pristine legal standard and a simple historical fact.” 517 U. S., at 378, 388, 390. We added that sometimes courts may have to make “credibility judgments” about witnesses. Id., at 389. In other words, we recognized that courts may have to resolve subsidiary factual disputes. And, as explained above, the Rule requires appellate courts to review all such subsidiary factual findings under the “clearly erroneous” standard. Precedent further supports application of the “clearly erroneous” standard. Before the creation of the Federal Circuit, the Second Circuit explained that in claim construction, the subsidiary “question . . . of how the art un-derstood the term . . . was plainly a question of fact; and unless the [district court’s] finding was ‘clearly erroneous,’ we are to take” it “as controlling.” Harries v. Air King Products, Co., 183 F. 2d 158, 164 (CA2 1950) (L. Hand, C. J.). We have said the same as to subsidiary factual findings concerning other patent law inquiries, including “obviousness.” Dennison Mfg. Co. v. Panduit Corp.,475 U. S. 809,811 (1986) (per curiam) (“subsidiary determinations of the District Court” subject to Rule 52(a)’s clear error standard). Finally, practical considerations favor clear error review. We have previously pointed out that clear error review is “particularly” important where patent law is at issue because patent law is “a field where so much depends upon familiarity with specific scientific problems and principles not usually contained in the general storehouse of knowledge and experience.” Graver Tank & Mfg. Co. v. Linde Air Products Co.,339 U. S. 605,610 (1950). A district court judge who has presided over, and listened to, the entirety of a proceeding has a comparatively greater opportunity to gain that familiarity than an appeals court judge who must read a written transcript or perhaps just those portions to which the parties have referred. Cf. Lighting Ballast, 744 F. 3d, at 1311 (O’Malley, J., dissenting) (Federal Circuit judges “lack the tools that district courts have available to resolve factual disputes fairly and accurately,” such as questioning the experts, examining the invention in operation, or appointing a court-appointed expert); Anderson, 470 U. S., at 574 (“The trial judge’s major role is the determination of fact, and with experience in fulfilling that role comes expertise”).B Sandoz argues that claim construction mostly consists of construing a set of written documents that do not give rise to subsidiary factual disputes. Tr. of Oral Arg. 39. It adds that separating “factual” from “legal” questions is often difficult. And Sandoz, like the Federal Circuit itself, argues that it is simpler for that appellate court to review the entirety of the district court’s claim construction de novo rather than to apply two separate standards. Id., at 38; see also Lighting Ballast, supra, at 1284 (criticizing clear error review in part because of the purportedly difficult task of “disentangling” fact from law). But even were we free to ignore the Federal Rule (which we are not), we would not find this argument convincing. Courts of appeals have long found it possible to separate factual from legal matters. See, e.g., First Options of Chicago, Inc. v. Kaplan,514 U. S. 938–948 (1995) (review of factual findings for clear error and legal conclusions de novo is the “ordinary” standard for courts of appeals). At the same time, the Federal Circuit’s efforts to treat factual findings and legal conclusions similarly have brought with them their own complexities. See e.g., Cybor Corp. v. FAS Technologies, Inc., 138 F. 3d 1448, 1454 (CA Fed. 1998) (en banc) (claim construction does not involve “factual evidentiary findings” (citation and internal quotation marks omitted)); Lighting Ballast, supra, at 1284 (claim construction has “arguably factual aspects”); Dow Jones & Co. v. Ablaise Ltd., 606 F. 3d 1338, 1344–1345 (CA Fed. 2010) (“[T]his court,” while reviewing claim construction without deference, “takes into account the views of the trial judge”); Nazomi Communications Inc., v. Arm Holdings, PLC, 403 F. 3d 1364, 1371 (CA Fed. 2005) (“[C]ommon sense dictates that the trial judge’s view will carry weight” (citation and internal quotation marks omitted)); Lightning Ballast, supra, at 1294 (Lourie, J., concurring) (we should “rarely” overturn district court’s true subsidiary factfinding; “we should, and do, give proper informal deference to the work of judges of a subordinate tribunal”); Cybor, supra, at 1480 (opinion of Newman, J.) (“By continuing the fiction that there are no facts to be found in claim interpretations, we confound rather than ease the litigation process”); see also Anderson, supra, at 575 (the parties “have already been forced to concentrate their energies and resources on persuading the trial judge that their account of the facts is the correct one; requiring them to persuade three more judges at the appellate level is requiring too much”); Brief for Peter S. Menell et al. as Amici Curiae 5 (Federal Circuit overturns district court claim construction at unusually high rate). Finally, the Circuit feared that “clear error” review would bring about less uniformity. Lighting Ballast, supra, at 1280. Neither the Circuit nor Sandoz, however, has shown that (or explained why) divergent claim construction stemming from divergent findings of fact (on subsidiary matters) should occur more than occasionally. After all, the Federal Circuit will continue to review de novo the district court’s ultimate interpretation of the patent claims. And the attorneys will no doubt bring cases construing the same claim to the attention of the trial judge; those prior cases will sometimes be binding because of issue preclusion, see Markman, 517 U. S., at 391, and sometimes will serve as persuasive authority. Moreover, it is always possible to consolidate for discovery different cases that involve construction of the same claims. And, as we said in Markman, subsidiary factfinding is unlikely to loom large in the universe of litigated claim construction. Id., at 389–390.C The dissent argues that claim construction does not involve any “factfinding,” or, if it does, claim construction factfinding is akin to the factfinding that underlies our interpretation of statutes. Post, at 1, 5–7 (opinion of Thomas, J.). Its first, broader contention runs contrary to our recognition in Markman that claim construction has “evidentiary underpinnings” and that courts construing patent claims must sometimes make “credibility judgments” about witnesses. 517 U. S., at 389–390. Indeed, as discussed in Part III, infra, this case provides a perfect example of the factfinding that sometimes underlies claim construction: The parties here presented the District Court with competing fact-related claims by different experts, and the District Court resolved the issues of fact that divided those experts. The dissent’s contention also runs contrary to Sandoz’s concession at oral argument that claim construction will sometimes require subsidiary factfinding. Tr. of Oral Arg. 33–34, 38–40. It is in tension with our interpretation of related areas of patent law, such as the interpretation of “obviousness,” which we have said involves subsidiary factfinding subject to Rule 52(a)’s clear error review. See Dennison, 475 U. S., at 811. And it fights the question presented in this case, which assumes the existence of such factfinding. See Pet. for Cert. i (whether “a district court’s factual finding in support of its construction of a patent claim term may be reviewed de novo, . . . or only for clear error”). Neither do we find factfinding in this context sufficiently similar to the factfinding that underlies statutory interpretation. Statutes, in general, address themselves to the general public; patent claims concern a small portion of that public. Statutes typically (though not always) rest upon congressional consideration of general facts related to a reasonably broad set of social circumstances; patents typically (though not always) rest upon consideration by a few private parties, experts, and administrators of more narrowly circumscribed facts related to specific technical matters. The public, and often an adversarial public, typically considers and discusses the relevant general facts before Congress enacts a statute; only private parties, experts, and administrators likely consider the relevant technical facts before the award of a patent. Given these differences, it is not surprising that this Court has never previously compared patent claim construction in any here relevant way to statutory construction. As discussed supra, at 5, however, the Court has repeatedly compared patent claim construction to the construction of other written instruments such as deeds and contracts. See, e.g., Markman, supra, at 384, 386, 388, 389; Motion Picture Patent Co., 243 U. S., at 510; Goodyear, 102 U. S., at 227.D Now that we have set forth why the Federal Circuit must apply clear error review when reviewing subsidiary factfinding in patent claim construction, it is necessary to explain how the rule must be applied in that context. We recognize that a district court’s construction of a patent claim, like a district court’s interpretation of a written instrument, often requires the judge only to examine and to construe the document’s words without requiring the judge to resolve any underlying factual disputes. As all parties agree, when the district court reviews only evidence intrinsic to the patent (the patent claims and specifications, along with the patent’s prosecution history), the judge’s determination will amount solely to a determination of law, and the Court of Appeals will review that construction de novo. See Brief for Petitioners 27, Re-ply Brief 16; Brief for Respondents 43; see also Brief for United States as Amicus Curiae 12–13. In some cases, however, the district court will need to look beyond the patent’s intrinsic evidence and to consult extrinsic evidence in order to understand, for example, the background science or the meaning of a term in the relevant art during the relevant time period. See, e.g., Seymour v. Osborne, 11 Wall. 516, 546 (1871) (a patent may be “so interspersed with technical terms and terms of art that the testimony of scientific witnesses is indispensable to a correct understanding of its meaning”). In cases where those subsidiary facts are in dispute, courts will need to make subsidiary factual findings about that extrinsic evidence. These are the “evidentiary underpinnings” of claim construction that we discussed in Markman, and this subsidiary factfinding must be reviewed for clear error on appeal. For example, if a district court resolves a dispute between experts and makes a factual finding that, in general, a certain term of art had a particular meaning to a person of ordinary skill in the art at the time of the invention, the district court must then conduct a legal analysis: whether a skilled artisan would ascribe that same meaning to that term in the context of the specific patent claim under review. That is because “[e]xperts may be examined to explain terms of art, and the state of the art, at any given time,” but they cannot be used to prove “the proper or legal construction of any instrument of writing.” Winans v. New York & Erie R. Co., 21 How. 88, 100–101 (1859); see also Markman, supra, at 388 (“ ‘Where technical terms are used, or where the qualities of substances . . . or any similar data necessary to the comprehension of the language of the patent are unknown to the judge, the testimony of witnesses may be received upon these subjects, and any other means of information be employed. But in the actual interpretation of the patent the court proceeds upon its own responsibility, as an arbiter of the law, giving to the patent its true and final character and force’ ” (quoting 2 W. Robinson, Law of Patents §732, pp. 482–483 (1890); emphasis in original)). Accordingly, the question we have answered here concerns review of the district court’s resolution of a subsidiary factual dispute that helps that court determine the proper interpretation of the written patent claim. The district judge, after deciding the factual dispute, will then interpret the patent claim in light of the facts as he has found them. This ultimate interpretation is a legal conclusion. The appellate court can still review the district court’s ultimate construction of the claim de novo. But, to overturn the judge’s resolution of an underlying factual dispute, the Court of Appeals must find that the judge, in respect to those factual findings, has made a clear error. Fed. Rule Civ. Proc. 52(a)(6). In some instances, a factual finding will play only a small role in a judge’s ultimate legal conclusion about the meaning of the patent term. But in some instances, a factual finding may be close to dispositive of the ultimate legal question of the proper meaning of the term in the context of the patent. Nonetheless, the ultimate question of construction will remain a legal question. Simply because a factual finding may be nearly dispositive does not render the subsidiary question a legal one. “[A]n issue does not lose its factual character merely because its resolution is dispositive of the ultimate” legal question. Miller v. Fenton,474 U. S. 104,113 (1985). It is analogous to a judge (sitting without a jury) deciding whether a defendant gave a confession voluntarily. The answer to the legal question about the voluntariness of the confes-sion may turn upon the answer to a subsidiary factual question, say “whether in fact the police engaged in the intimidation tactics alleged by the defendant.” Id., at 112. An appellate court will review the trial judge’s factual determination about the alleged intimidation deferentially (though, after reviewing the factual findings, it will review a judge’s ultimate determination of voluntariness de novo). See id., at 112–118. An appellate court similarly should review for clear error those factual findings that underlie a district court’s claim construction.III We can illustrate our holding by considering an instance in which Teva, with the support of the Solicitor General, argues that the Federal Circuit wrongly reviewed the District Court’s factual finding de novo. See Brief for Petitioners 54–56; Brief for United States as Amicus Curiae 31–32. Recall that Teva’s patent claim specifies an active ingredient with a “molecular weight of about 5 to 9 kilodaltons.” Recall Sandoz’s basic argument, namely that the term “molecular weight” is indefinite or ambiguous. The term might refer to the weight of the most numerous molecule, it might refer to weight as calculated by the average weight of all molecules, or it might refer to weight as calculated by an average in which heavier molecules count for more. The claim, Sandoz argues, does not tell us which way we should calculate weight. See Part I, supra. To illustrate, imagine we have a sample of copolymer-1 (the active ingredient) made up of 10 molecules: 4 weigh 6 kilodaltons each, 3 weigh 8 kilodaltons each, and 3 weigh 9 kilodaltons each. Using the first method of calculation, the “molecular weight” would be 6 kilodaltons, the weight of the most prevalent molecule. Using the second method, the molecular weight would be 7.5 (total weight, 75, di-vided by the number of molecules, 10). Using the third method, the molecular weight would be more than 8, depend-ing upon how much extra weight we gave to the heavier molecules. Teva argued in the District Court that the term “molecular weight” in the patent meant molecular weight calculated in the first way (the weight of the most prevalent molecule, or peak average molecular weight). Sandoz, however, argued that figure 1 of the patent showed that Teva could not be right. 810 F. Supp. 2d, at 590. (We have set forth figure 1 in the Appendix, infra). That figure, said Sandoz, helped to show that the patent term did not refer to the first method of calculation. Figure 1 shows how the weights of a sample’s molecules were distributed in three different samples. The curves indicate the number of molecules of each weight that were present in each of the three. For example, the figure’s legend says that the first sample’s “molecular weight” is 7.7. According to Teva, that should mean that molecules weighing 7.7 kilodaltons were the most prevalent molecules in the sample. But, look at the curve, said Sandoz. It shows that the most prevalent molecule weighed, not 7.7 kilodaltons, but slightly less than 7.7 (about 6.8) kilodaltons. See App. 138a–139a. After all, the peak of the first molecular weight distribution curve (the solid curve in the figure) is not at precisely 7.7 kilodaltons, but at a point just before 7.7. Thus, argued Sandoz, the figure shows that the patent claim term “molecular weight” did not mean molecular weight calculated by the first method. It must mean something else. It is indefinite. 810 F. Supp. 2d, at 590. The District Court did not accept Sandoz’s argument. Teva’s expert testified that a skilled artisan would understand that converting data from a chromatogram to molecular weight distribution curves like those in figure 1 would cause the peak on each curve to shift slightly; this could explain the difference between the value indicated by the peak of the curve (about 6.8) and the value in the figure’s legend (7.7). App. 138a–139a. Sandoz’s expert testified that no such shift would occur. App. 375a–376a. The District Court credited Teva’s expert’s account, thereby rejecting Sandoz’s expert’s explanation. 810 F. Supp. 2d, at 589; Brief for Respondents 61. The District Court’s finding about this matter was a factual finding—about how a skilled artisan would understand the way in which a curve created from chromatogram data reflects molecular weights. Based on that factual finding, the District Court reached the legal conclusion that figure 1 did not undermine Teva’s argument that molecular weight referred to the first method of calculation (peak average molecular weight). 810 F. Supp. 2d, at 590–591. When the Federal Circuit reviewed the District Court’s decision, it recognized that the peak of the curve did not match the 7.7 kilodaltons listed in the legend of figure 1. 723 F. 3d, at 1369. But the Federal Circuit did not accept Teva’s expert’s explanation as to how a skilled artisan would expect the peaks of the curves to shift. And it failed to accept that explanation without finding that the District Court’s contrary determination was “clearly erroneous.” See ibid. The Federal Circuit should have accepted the District Court’s finding unless it was “clearly erroneous.” Our holding today makes clear that, in failing to do so, the Federal Circuit was wrong. Teva claims that there are two additional instances in which the Federal Circuit rejected the District Court’s factual findings without concluding that they were clearly erroneous. We leave these matters for the Federal Circuit to consider on remand in light of today’s opinion. We vacate the Federal Circuit’s judgment, and we remand the case for further proceedings consistent with this opinion.It is so ordered.APPENDIXFIG. 1 (with minor additions to emphasize that the peak of the solid curve does not correspond precisely to 7.7kDa)

576.US.2014_13-1371

The Federal Government provides low-income housing tax credits that are distributed to developers by designated state agencies. In Texas, the Department of Housing and Community Affairs (Department) distributes the credits. The Inclusive Communities Project, Inc. (ICP), a Texas-based nonprofit corporation that assists low-income families in obtaining affordable housing, brought a disparate-impact claim under §§804(a) and 805(a) of the Fair Housing Act (FHA), alleging that the Department and its officers had caused continued segregated housing patterns by allocating too many tax credits to housing in predominantly black inner-city areas and too few in predominantly white suburban neighborhoods. Relying on statistical evidence, the District Court concluded that the ICP had established a prima facie showing of disparate impact. After assuming the Department’s proffered non-discriminatory interests were valid, it found that the Department failed to meet its burden to show that there were no less discriminatory alternatives for allocating the tax credits. While the Department’s appeal was pending, the Secretary of Housing and Urban Development issued a regulation interpreting the FHA to encompass disparate-impact liability and establishing a burden-shifting framework for adjudicating such claims. The Fifth Circuit held that disparate-impact claims are cognizable under the FHA, but reversed and remanded on the merits, concluding that, in light of the new regulation, the District Court had improperly required the Department to prove less discriminatory alternatives. The FHA was adopted shortly after the assassination of Dr. Martin Luther King, Jr. Recognizing that persistent racial segregation had left predominantly black inner cities surrounded by mostly white suburbs, the Act addresses the denial of housing opportunities on the basis of “race, color, religion, or national origin.” In 1988, Congress amended the FHA, and, as relevant here, created certain exemptions from liability. Held: Disparate-impact claims are cognizable under the Fair Housing Act. Pp. 7–24. (a) Two antidiscrimination statutes that preceded the FHA are relevant to its interpretation. Both §703(a)(2) of Title VII of the Civil Rights Act of 1964 and §4(a)(2) of the Age Discrimination in Employment Act of 1967 (ADEA) authorize disparate-impact claims. Under Griggs v. Duke Power Co., 401 U. S. 424 , and Smith v. City of Jackson, 544 U. S. 228 , the cases announcing the rule for Title VII and for the ADEA, respectively, antidiscrimination laws should be construed to encompass disparate-impact claims when their text refers to the consequences of actions and not just to the mindset of actors, and where that interpretation is consistent with statutory purpose. Disparate-impact liability must be limited so employers and other regulated entities are able to make the practical business choices and profit-related decisions that sustain the free-enterprise system. Before rejecting a business justification—or a governmental entity’s analogous public interest—a court must determine that a plaintiff has shown that there is “an available alternative . . . practice that has less disparate impact and serves the [entity’s] legitimate needs.” Ricci v. DeStefano, 557 U. S. 557 . These cases provide essential background and instruction in the case at issue. Pp. 7–10. (b) Under the FHA it is unlawful to “refuse to sell or rent . . . or otherwise make unavailable or deny, a dwelling to a person because of race” or other protected characteristic, §804(a), or “to discriminate against any person in” making certain real-estate transactions “because of race” or other protected characteristic, §805(a). The logic of Griggs and Smith provides strong support for the conclusion that the FHA encompasses disparate-impact claims. The results-oriented phrase “otherwise make unavailable” refers to the consequences of an action rather than the actor’s intent. See United States v. Giles, 300 U. S. 41 . And this phrase is equivalent in function and purpose to Title VII’s and the ADEA’s “otherwise adversely affect” language. In all three statutes the operative text looks to results and plays an identical role: as a catchall phrase, located at the end of a lengthy sentence that begins with prohibitions on disparate treatment. The introductory word “otherwise” also signals a shift in emphasis from an actor’s intent to the consequences of his actions. This similarity in text and structure is even more compelling because Congress passed the FHA only four years after Title VII and four months after the ADEA. Although the FHA does not reiterate Title VII’s exact language, Congress chose words that serve the same purpose and bear the same basic meaning but are consistent with the FHA’s structure and objectives. The FHA contains the phrase “because of race,” but Title VII and the ADEA also contain that wording and this Court nonetheless held that those statutes impose disparate-impact liability. The 1988 amendments signal that Congress ratified such liability. Congress knew that all nine Courts of Appeals to have addressed the question had concluded the FHA encompassed disparate-impact claims, and three exemptions from liability in the 1988 amendments would have been superfluous had Congress assumed that disparate-impact liability did not exist under the FHA. Recognition of disparate-impact claims is also consistent with the central purpose of the FHA, which, like Title VII and the ADEA, was enacted to eradicate discriminatory practices within a sector of the Nation’s economy. Suits targeting unlawful zoning laws and other housing restrictions that unfairly exclude minorities from certain neighborhoods without sufficient justification are at the heartland of disparate-impact liability. See, e.g., Huntington v. Huntington Branch, NAACP, 488 U. S. 15 –18. Recognition of disparate-impact liability under the FHA plays an important role in uncovering discriminatory intent: it permits plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification as disparate treatment. But disparate-impact liability has always been properly limited in key respects to avoid serious constitutional questions that might arise under the FHA, e.g., if such liability were imposed based solely on a showing of a statistical disparity. Here, the underlying dispute involves a novel theory of liability that may, on remand, be seen simply as an attempt to second-guess which of two reasonable approaches a housing authority should follow in allocating tax credits for low-income housing. An important and appropriate means of ensuring that disparate-impact liability is properly limited is to give housing authorities and private developers leeway to state and explain the valid interest their policies serve, an analysis that is analogous to Title VII’s business necessity standard. It would be paradoxical to construe the FHA to impose onerous costs on actors who encourage revitalizing dilapidated housing in the Nation’s cities merely because some other priority might seem preferable. A disparate-impact claim relying on a statistical disparity must fail if the plaintiff cannot point to a defendant’s policy or policies causing that disparity. A robust causality requirement is important in ensuring that defendants do not resort to the use of racial quotas. Courts must therefore examine with care whether a plaintiff has made out a prima facie showing of disparate impact, and prompt resolution of these cases is important. Policies, whether governmental or private, are not contrary to the disparate-impact requirement unless they are “artificial, arbitrary, and unnecessary barriers.” Griggs, 401 U. S., at 431. Courts should avoid interpreting disparate-impact liability to be so expansive as to inject racial considerations into every housing decision. These limitations are also necessary to protect defendants against abusive disparate-impact claims. And when courts do find liability under a disparate-impact theory, their remedial orders must be consistent with the Constitution. Remedial orders in disparate-impact cases should concentrate on the elimination of the offending practice, and courts should strive to design race-neutral remedies. Remedial orders that impose racial targets or quotas might raise difficult constitutional questions. While the automatic or pervasive injection of race into public and private transactions covered by the FHA has special dangers, race may be considered in certain circumstances and in a proper fashion. This Court does not impugn local housing authorities’ race-neutral efforts to encourage revitalization of communities that have long suffered the harsh consequences of segregated housing patterns. These authorities may choose to foster diversity and combat racial isolation with race-neutral tools, and mere awareness of race in attempting to solve the problems facing inner cities does not doom that endeavor at the outset. Pp. 10–23. 747 F. 3d 275, affirmed and remanded. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Scalia and Thomas, JJ., joined.

The underlying dispute in this case concerns where housing for low-income persons should be constructed in Dallas, Texas—that is, whether the housing should be built in the inner city or in the suburbs. This dispute comes to the Court on a disparate-impact theory of liability. In contrast to a disparate-treatment case, where a “plaintiff must establish that the defendant had a discriminatory intent or motive,” a plaintiff bringing a disparate-impact claim challenges practices that have a “disproportionately adverse effect on minorities” and are otherwise unjustified by a legitimate rationale. Ricci v. DeStefano, 557 U. S. 557, 577 (2009) (internal quotation marks omitted). The question presented for the Court’s determination is whether disparate-impact claims are cognizable under the Fair Housing Act (or FHA), 82Stat. 81, as amended, 42 U. S. C. §3601 et seq. I A Before turning to the question presented, it is necessary to discuss a different federal statute that gives rise to this dispute. The Federal Government provides low-income housing tax credits that are distributed to developers through designated state agencies. 26 U. S. C. §42. Congress has directed States to develop plans identifying selection criteria for distributing the credits. §42(m)(1). Those plans must include certain criteria, such as public housing waiting lists, §42(m)(1)(C), as well as certain preferences, including that low-income housing units “contribut[e] to a concerted community revitalization plan” and be built in census tracts populated predominantly by low-income residents. §§42(m)(1)(B)(ii)(III), 42(d)(5)(ii)(I). Federal law thus favors the distribution of these tax credits for the development of housing units in low-income areas. In the State of Texas these federal credits are distrib-uted by the Texas Department of Housing and Community Affairs (Department). Under Texas law, a developer’s application for the tax credits is scored under a point system that gives priority to statutory criteria, such as the financial feasibility of the development project and the income level of tenants. Tex. Govt. Code Ann. §§2306.6710(a)–(b) (West 2008). The Texas Attorney General has interpreted state law to permit the consideration of additional criteria, such as whether the housing units will be built in a neighborhood with good schools. Those criteria cannot be awarded more points than statutorily mandated criteria. Tex. Op. Atty. Gen. No. GA–0208, pp. 2–6 (2004), 2004 WL 1434796, *4–*6. The Inclusive Communities Project, Inc. (ICP), is a Texas-based nonprofit corporation that assists low-income families in obtaining affordable housing. In 2008, the ICP brought this suit against the Department and its officers in the United States District Court for the Northern District of Texas. As relevant here, it brought a disparate-impact claim under §§804(a) and 805(a) of the FHA. The ICP alleged the Department has caused continued segregated housing patterns by its disproportionate allocation of the tax credits, granting too many credits for housing in predominantly black inner-city areas and too few in predominantly white suburban neighborhoods. The ICP contended that the Department must modify its selection criteria in order to encourage the construction of low-income housing in suburban communities. The District Court concluded that the ICP had established a prima facie case of disparate impact. It relied on two pieces of statistical evidence. First, it found “from 1999–2008, [the Department] approved tax credits for 49.7% of proposed non-elderly units in 0% to 9.9% Caucasian areas, but only approved 37.4% of proposed non-elderly units in 90% to 100% Caucasian areas.” 749 F. Supp. 2d 486, 499 (ND Tex. 2010) (footnote omitted). Second, it found “92.29% of [low-income housing tax credit] units in the city of Dallas were located in census tracts with less than 50% Caucasian residents.” Ibid. The District Court then placed the burden on the Department to rebut the ICP’s prima facie showing of disparate impact. 860 F. Supp. 2d 312, 322–323 (2012). After assuming the Department’s proffered interests were legitimate, id., at 326, the District Court held that a defendant—here the Department—must prove “that there are no other less discriminatory alternatives to advancing their proffered interests,” ibid. Because, in its view, the Department “failed to meet [its] burden of proving that there are no less discriminatory alternatives,” the District Court ruled for the ICP. Id., at 331. The District Court’s remedial order required the addition of new selection criteria for the tax credits. For instance, it awarded points for units built in neighborhoods with good schools and disqualified sites that are located adjacent to or near hazardous conditions, such as high crime areas or landfills. See 2012 WL 3201401 (Aug. 7, 2012). The remedial order contained no explicit racial targets or quotas. While the Department’s appeal was pending, the Secretary of Housing and Urban Development (HUD) issued a regulation interpreting the FHA to encompass disparate-impact liability. See Implementation of the Fair Housing Act’s Discriminatory Effects Standard, 78 Fed. Reg. 11460 (2013). The regulation also established a burden-shifting framework for adjudicating disparate-impact claims. Under the regulation, a plaintiff first must make a prima facie showing of disparate impact. That is, the plaintiff “has the burden of proving that a challenged practice caused or predictably will cause a discriminatory effect.” 24 CFR §100.500(c)(1) (2014). If a statistical discrepancy is caused by factors other than the defendant’s policy, a plaintiff cannot establish a prima facie case, and there is no liability. After a plaintiff does establish a prima facie showing of disparate impact, the burden shifts to the defendant to “prov[e] that the challenged practice is necessary to achieve one or more substantial, legitimate, nondiscriminatory interests.” §100.500(c)(2). HUD has clarified that this step of the analysis “is analogous to the Title VII requirement that an employer’s interest in an employment practice with a disparate impact be job related.” 78 Fed. Reg. 11470. Once a defendant has satisfied its burden at step two, a plaintiff may “prevail upon proving that the substantial, legitimate, nondiscriminatory interests supporting the challenged practice could be served by another practice that has a less discriminatory effect.” §100.500(c)(3). The Court of Appeals for the Fifth Circuit held, consistent with its precedent, that disparate-impact claims are cognizable under the FHA. 747 F. 3d 275, 280 (2014). On the merits, however, the Court of Appeals reversed and remanded. Relying on HUD’s regulation, the Court of Appeals held that it was improper for the District Court to have placed the burden on the Department to prove there were no less discriminatory alternatives for allocating low-income housing tax credits. Id., at 282–283. In a concurring opinion, Judge Jones stated that on remand the District Court should reexamine whether the ICP had made out a prima facie case of disparate impact. She suggested the District Court incorrectly relied on bare statistical evidence without engaging in any analysis about causation. She further observed that, if the fed-eral law providing for the distribution of low-income housing tax credits ties the Department’s hands to such an extent that it lacks a meaningful choice, then there is no disparate-impact liability. See id., at 283–284 (specially concurring opinion). The Department filed a petition for a writ of certiorari on the question whether disparate-impact claims are cognizable under the FHA. The question was one of first impression, see Huntington v. Huntington Branch, NAACP, 488 U. S. 15 (1988) (per curiam), and certiorari followed, 573 U. S. ___ (2014). It is now appropriate to provide a brief history of the FHA’s enactment and its later amendment. B De jure residential segregation by race was declared unconstitutional almost a century ago, Buchanan v. Warley, 245 U. S. 60 (1917) , but its vestiges remain today, intertwined with the country’s economic and social life. Some segregated housing patterns can be traced to conditions that arose in the mid-20th century. Rapid urbanization, concomitant with the rise of suburban developments accessible by car, led many white families to leave the inner cities. This often left minority families concentrated in the center of the Nation’s cities. During this time, various practices were followed, sometimes with governmental support, to encourage and maintain the separation of the races: Racially restrictive covenants prevented the conveyance of property to minorities, see Shelley v. Krae-mer, 334 U. S. 1 (1948) ; steering by real-estate agents led potential buyers to consider homes in racially homogenous areas; and discriminatory lending practices, often referred to as redlining, precluded minority families from purchasing homes in affluent areas. See, e.g., M. Klarman, Unfinished Business: Racial Equality in American History 140–141 (2007); Brief for Housing Scholars as Amici Curiae 22–23. By the 1960’s, these policies, practices, and prejudices had created many predominantly black inner cities surrounded by mostly white suburbs. See K. Clark, Dark Ghetto: Dilemmas of Social Power 11, 21–26 (1965). The mid-1960’s was a period of considerable social unrest; and, in response, President Lyndon Johnson established the National Advisory Commission on Civil Disorders, commonly known as the Kerner Commission. Exec. Order No. 11365, 3 CFR 674 (1966–1970 Comp.). After extensive factfinding the Commission identified residential segregation and unequal housing and economic conditions in the inner cities as significant, underlying causes of the social unrest. See Report of the National Advisory Commission on Civil Disorders 91 (1968) (Kerner Commission Report). The Commission found that “[n]early two-thirds of all nonwhite families living in the central cities today live in neighborhoods marked by substandard housing and general urban blight.” Id., at 13. The Commission further found that both open and covert racial discrimination prevented black families from obtaining better housing and moving to integrated communities. Ibid. The Commission concluded that “[o]ur Nation is moving toward two societies, one black, one white—separate and unequal.” Id., at 1. To reverse “[t]his deepening racial division,” ibid., it recommended enactment of “a comprehensive and enforceable open-occupancy law making it an offense to discriminate in the sale or rental of any housing . . . on the basis of race, creed, color, or national origin.” Id., at 263. In April 1968, Dr. Martin Luther King, Jr., was assassinated in Memphis, Tennessee, and the Nation faced a new urgency to resolve the social unrest in the inner cities. Congress responded by adopting the Kerner Commission’s recommendation and passing the Fair Housing Act. The statute addressed the denial of housing opportunities on the basis of “race, color, religion, or national origin.” Civil Rights Act of 1968, §804, 82Stat. 83. Then, in 1988, Congress amended the FHA. Among other provisions, it created certain exemptions from liability and added “familial status” as a protected characteristic. See Fair Housing Amendments Act of 1988, 102Stat. 1619. II The issue here is whether, under a proper interpretation of the FHA, housing decisions with a disparate impact are prohibited. Before turning to the FHA, however, it is necessary to consider two other antidiscrimination statutes that preceded it. The first relevant statute is §703(a) of Title VII of the Civil Rights Act of 1964, 78Stat. 255. The Court addressed the concept of disparate impact under this statute in Griggs v. Duke Power Co., 401 U. S. 424 (1971) . There, the employer had a policy requiring its manual laborers to possess a high school diploma and to obtain satisfactory scores on two intelligence tests. The Court of Appeals held the employer had not adopted these job requirements for a racially discriminatory purpose, and the plaintiffs did not challenge that holding in this Court. Instead, the plaintiffs argued §703(a)(2) covers the discriminatory effect of a practice as well as the motivation behind the practice. Section 703(a), as amended, provides as follows: “It shall be an unlawful employer practice for an employer— “(1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin; or “(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.” 42 U. S. C. §2000e–2(a). The Court did not quote or cite the full statute, but rather relied solely on §703(a)(2). Griggs, 401 U. S., at 426, n. 1. In interpreting §703(a)(2), the Court reasoned that disparate-impact liability furthered the purpose and design of the statute. The Court explained that, in §703(a)(2), Congress “proscribe[d] not only overt discrimination but also practices that are fair in form, but discriminatory in operation.” Id., at 431. For that reason, as the Court noted, “Congress directed the thrust of [§703(a)(2)] to the consequences of employment practices, not simply the motivation.” Id., at 432. In light of the statute’s goal of achieving “equality of employment opportunities and remov[ing] barriers that have operated in the past” to favor some races over others, the Court held §703(a)(2) of Title VII must be interpreted to allow disparate-impact claims. Id., at 429–430. The Court put important limits on its holding: namely, not all employment practices causing a disparate impact impose liability under §703(a)(2). In this respect, the Court held that “business necessity” constitutes a defense to disparate-impact claims. Id., at 431. This rule provides, for example, that in a disparate-impact case, §703(a)(2) does not prohibit hiring criteria with a “manifest relationship” to job performance. Id., at 432; see also Ricci, 557 U. S., at 587–589 (emphasizing the importance of the business necessity defense to disparate-impact liability). On the facts before it, the Court in Griggs found a violation of Title VII because the employer could not establish that high school diplomas and general intelligence tests were related to the job performance of its manual laborers. See 401 U. S., at 431–432. The second relevant statute that bears on the proper interpretation of the FHA is the Age Discrimination in Employment Act of 1967 (ADEA), 81Stat. 602 et seq., as amended. Section 4(a) of the ADEA provides: “It shall be unlawful for an employer— “(1) to fail or refuse to hire or to discharge any individual or otherwise discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s age; “(2) to limit, segregate, or classify his employees in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s age; or “(3) to reduce the wage rate of any employee in order to comply with this chapter.” 29 U. S. C. §623(a). The Court first addressed whether this provision allows disparate-impact claims in Smith v. City of Jackson, 544 U. S. 228 (2005) . There, a group of older employees challenged their employer’s decision to give proportionately greater raises to employees with less than five years of experience. Explaining that Griggs “represented the better reading of [Title VII’s] statutory text,” 544 U. S., at 235, a plurality of the Court concluded that the same reasoning pertained to §4(a)(2) of the ADEA. The Smith plurality emphasized that both §703(a)(2) of Title VII and §4(a)(2) of the ADEA contain language “prohibit[ing] such actions that ‘deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s’ race or age.” 544 U. S., at 235. As the plurality observed, the text of these provisions “focuses on the effects of the action on the employee rather than the motivation for the action of the employer” and therefore compels recognition of disparate-impact liability. Id., at 236. In a separate opinion, Justice Scalia found the ADEA’s text ambiguous and thus deferred under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) , to an Equal Employment Opportunity Commission regulation interpreting the ADEA to imposedisparate-impact liability, see 544 U. S., at 243–247 (opinion concurring in part and concurring in judgment). Together, Griggs holds and the plurality in Smith instructs that antidiscrimination laws must be construed to encompass disparate-impact claims when their text refers to the consequences of actions and not just to the mindset of actors, and where that interpretation is consistent with statutory purpose. These cases also teach that disparate-impact liability must be limited so employers and other regulated entities are able to make the practical business choices and profit-related decisions that sustain a vibrant and dynamic free-enterprise system. And before rejecting a business justification—or, in the case of a governmental entity, an analogous public interest—a court must determine that a plaintiff has shown that there is “an available alternative . . . practice that has less disparate impact and serves the [entity’s] legitimate needs.” Ricci, supra, at 578. The cases interpreting Title VII and the ADEA provide essential background and instruction in the case now before the Court. Turning to the FHA, the ICP relies on two provisions. Section 804(a) provides that it shall be unlawful: “To refuse to sell or rent after the making of a bona fide offer, or to refuse to negotiate for the sale orrental of, or otherwise make unavailable or deny, a dwelling to any person because of race, color, religion, sex, familial status, or national origin.” 42 U. S. C. §3604(a). Here, the phrase “otherwise make unavailable” is of central importance to the analysis that follows. Section 805(a), in turn, provides: “It shall be unlawful for any person or other entity whose business includes engaging in real estate-related transactions to discriminate against any person in making available such a transaction, or in the terms or conditions of such a transaction, because of race, color, religion, sex, handicap, familial status, or national origin.” §3605(a). Applied here, the logic of Griggs and Smith provides strong support for the conclusion that the FHA encompasses disparate-impact claims. Congress’ use of the phrase “otherwise make unavailable” refers to the consequences of an action rather than the actor’s intent. See United States v. Giles, 300 U. S. 41, 48 (1937) (explaining that the “word ‘make’ has many meanings, among them ‘[t]o cause to exist, appear or occur’ ” (quoting Webster’s New International Dictionary 1485 (2d ed. 1934))). This results-oriented language counsels in favor of recognizing disparate-impact liability. See Smith, supra, at 236. The Court has construed statutory language similar to §805(a) to include disparate-impact liability. See, e.g., Board of Ed. of City School Dist. of New York v. Harris, 444 U. S. 130 –141 (1979) (holding the term “discriminat[e]” encompassed disparate-impact liability in the context of a statute’s text, history, purpose, and structure). A comparison to the antidiscrimination statutes examined in Griggs and Smith is useful. Title VII’s and the ADEA’s “otherwise adversely affect” language is equivalent in function and purpose to the FHA’s “otherwise make unavailable” language. In these three statutes the operative text looks to results. The relevant statutory phrases, moreover, play an identical role in the structure common to all three statutes: Located at the end of lengthy sentences that begin with prohibitions on disparate treatment, they serve as catchall phrases looking to consequences, not intent. And all three statutes use the word “otherwise” to introduce the results-oriented phrase. “Otherwise” means “in a different way or manner,” thus signaling a shift in emphasis from an actor’s intent to the consequences of his actions. Webster’s Third New International Dictionary 1598 (1971). This similarity in text and structure is all the more compelling given that Congress passed the FHA in 1968—only four years after passing Title VII and only four months after enacting the ADEA. It is true that Congress did not reiterate Title VII’s exact language in the FHA, but that is because to do so would have made the relevant sentence awkward and unclear. A provision making it unlawful to “refuse to sell[,] . . . or otherwise [adversely affect], a dwelling to any person” because of a protected trait would be grammatically obtuse, difficult to interpret, and far more expansive in scope than Congress likely intended. Congress thus chose words that serve the same purpose and bear the same basic meaning but are consistent with the structure and objectives of the FHA. Emphasizing that the FHA uses the phrase “because of race,” the Department argues this language forecloses disparate-impact liability since “[a]n action is not taken ‘because of race’ unless race is a reason for the action.” Brief for Petitioners 26. Griggs and Smith, however, dispose of this argument. Both Title VII and the ADEA contain identical “because of” language, see 42 U. S. C. §2000e–2(a)(2); 29 U. S. C. §623(a)(2), and the Courtnonetheless held those statutes impose disparate-impact liability. In addition, it is of crucial importance that the existence of disparate-impact liability is supported by amendments to the FHA that Congress enacted in 1988. By that time, all nine Courts of Appeals to have addressed the question had concluded the Fair Housing Act encompassed disparate-impact claims. See Huntington Branch, NAACP v.Huntington, 844 F. 2d 926, 935–936 (CA2 1988); Resident Advisory Bd. v. Rizzo, 564 F. 2d 126, 146 (CA3 1977); Smith v. Clarkton, 682 F. 2d 1055, 1065 (CA4 1982); Hanson v. Veterans Administration, 800 F. 2d 1381, 1386 (CA5 1986); Arthur v. Toledo, 782 F. 2d 565, 574–575 (CA6 1986); Metropolitan Housing Development Corp. v. Arlington Heights, 558 F. 2d 1283, 1290 (CA7 1977); United States v. Black Jack, 508 F. 2d 1179, 1184–1185 (CA8 1974); Halet v. Wend Investment Co., 672 F. 2d 1305, 1311 (CA9 1982); United States v. Marengo Cty. Comm’n, 731 F. 2d 1546, 1559, n. 20 (CA11 1984). When it amended the FHA, Congress was aware of this unanimous precedent. And with that understanding, it made a considered judgment to retain the relevant statutory text. See H. R. Rep. No. 100–711, p. 21, n. 52 (1988) (H. R. Rep.) (discussing suits premised on disparate-impact claims and related judicial precedent); 134 Cong. Rec. 23711 (1988) (statement of Sen. Kennedy) (noting unanimity of Federal Courts of Appeals concerning disparate impact); Fair Housing Amendments Act of 1987: Hearings on S. 558 before the Subcommittee on the Constitution of the Senate Committee on the Judiciary, 100th Cong., 1st Sess., 529 (1987) (testimony of Professor Robert Schwemm) (describing consensus judicial view that the FHA imposed disparate-impact liability). Indeed, Congress rejected a proposed amendment that would have eliminated disparate-impact liability for certain zoning decisions. See H. R. Rep., at 89–93. Against this background understanding in the legal and regulatory system, Congress’ decision in 1988 to amend the FHA while still adhering to the operative language in §§804(a) and 805(a) is convincing support for the conclusion that Congress accepted and ratified the unanimous holdings of the Courts of Appeals finding disparate-impact liability. “If a word or phrase has been . . . given a uniform interpretation by inferior courts . . . , a later version of that act perpetuating the wording is presumed to carry forward that interpretation.” A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 322 (2012); see also Forest Grove School Dist. v. T. A., 557 U. S. 230, 244, n. 11 (2009) (“When Congress amended [the Act] without altering the text of [the relevant provision], it implicitly adopted [this Court’s] construction of the statute”); Manhattan Properties, Inc. v. Irving Trust Co., 291 U. S. 320, 336 (1934) (explaining, where the Courts of Appeals had reached a consensus interpretation of the Bankruptcy Act and Congress had amended the Act without changing the relevant provision, “[t]his is persua-sive that the construction adopted by the [lower federal] courts has been acceptable to the legislative arm of the government”). Further and convincing confirmation of Congress’ understanding that disparate-impact liability exists under the FHA is revealed by the substance of the 1988 amendments. The amendments included three exemptions from liability that assume the existence of disparate-impact claims. The most logical conclusion is that the three amendments were deemed necessary because Congress presupposed disparate impact under the FHA as it had been enacted in 1968. The relevant 1988 amendments were as follows. First, Congress added a clarifying provision: “Nothing in [the FHA] prohibits a person engaged in the business of furnishing appraisals of real property to take into consideration factors other than race, color, religion, national origin, sex, handicap, or familial status.” 42 U. S. C. §3605(c). Second, Congress provided: “Nothing in [the FHA] prohibits conduct against a person because such person has been convicted by any court of competent jurisdiction of the illegal manufacture or distribution of a controlled substance.” §3607(b)(4). And finally, Congress specified: “Nothing in [the FHA] limits the applicability of any reasonable . . . restrictions regarding the maximum number of occupants permitted to occupy a dwelling.” §3607(b)(1). The exemptions embodied in these amendments would be superfluous if Congress had assumed that disparate-impact liability did not exist under the FHA. See Gustafson v. Alloyd Co., 513 U. S. 561, 574 (1995) (“[T]he Court will avoid a reading which renders some words altogether redundant”). Indeed, none of these amendments would make sense if the FHA encompassed only disparate-treatment claims. If that were the sole ground for liability, the amendments merely restate black-letter law. If an actor makes a decision based on reasons other than a protected category, there is no disparate-treatment liability. See, e.g., Texas Dept. of Community Affairs v. Burdine, 450 U. S. 248, 254 (1981) . But the amendments do constrain disparate-impact liability. For instance, certain criminal convictions are correlated with sex and race. See, e.g., Kimbrough v. United States, 552 U. S. 85, 98 (2007) (discussing the racial disparity in convictions for crack cocaine offenses). By adding an exemption from liability for exclusionary practices aimed at individuals with drug convictions, Congress ensured disparate-impact liability would not lie if a landlord excluded tenants with such convictions. The same is true of the provision allowing for reasonable restrictions on occupancy. And the exemption from liability for real-estate appraisers is in the same section as §805(a)’s prohibition of discriminatory practices in real-estate transactions, thus indicating Congress’ recognition that disparate-impact liability arose under §805(a). In short, the 1988 amendments signal that Congress ratified disparate-impact liability. A comparison to Smith’s discussion of the ADEA further demonstrates why the Department’s interpretation would render the 1988 amendments superfluous. Under the ADEA’s reasonable-factor-other-than-age (RFOA) provision, an employer is permitted to take an otherwise prohibited action where “the differentiation is based on reasonable factors other than age.” 29 U. S. C. §623(f )(1). In other words, if an employer makes a decision based on a reasonable factor other than age, it cannot be said to have made a decision on the basis of an employee’s age. According to the Smith plurality, the RFOA provision “plays its principal role” “in cases involving disparate-impact claims” “by precluding liability if the adverse impact was attributable to a nonage factor that was ‘reasonable.’ ” 544 U. S., at 239. The plurality thus reasoned that the RFOA provision would be “simply unnecessary to avoid liability under the ADEA” if liability were limited to disparate-treatment claims. Id., at 238. A similar logic applies here. If a real-estate appraiser took into account a neighborhood’s schools, one could not say the appraiser acted because of race. And by embedding 42 U. S. C. §3605(c)’s exemption in the statutory text, Congress ensured that disparate-impact liability would not be allowed either. Indeed, the inference of disparate-impact liability is even stronger here than it was in Smith. As originally enacted, the ADEA included the RFOA provision, see §4(f)(1), 81Stat. 603, whereas here Congress added the relevant exemptions in the 1988 amendments against the backdrop of the uniform view of the Courts of Appeals that the FHA imposed disparate-impact liability. Recognition of disparate-impact claims is consistent with the FHA’s central purpose. See Smith, supra, at 235 (plurality opinion); Griggs, 401 U. S., at 432. The FHA, like Title VII and the ADEA, was enacted to eradicate discriminatory practices within a sector of our Nation’s economy. See 42 U. S. C. §3601 (“It is the policy of the United States to provide, within constitutional limitations, for fair housing throughout the United States”); H. R. Rep., at 15 (explaining the FHA “provides a clear national policy against discrimination in housing”). These unlawful practices include zoning laws and other housing restrictions that function unfairly to exclude minorities from certain neighborhoods without any sufficient justification. Suits targeting such practices reside at the heartland of disparate-impact liability. See, e.g., Huntington, 488 U. S., at 16–18 (invalidating zoning law preventing construction of multifamily rental units); Black Jack, 508 F. 2d, at 1182–1188 (invalidating ordinance prohibiting construction of new multifamily dwellings); Greater New Orleans Fair Housing Action Center v. St. Bernard Parish, 641 F. Supp. 2d 563, 569, 577–578 (ED La. 2009) (invalidating post-Hurricane Katrina ordinance restricting the rental of housing units to only “ ‘blood relative[s]’ ” in an area of the city that was 88.3% white and 7.6% black); see also Tr. of Oral Arg. 52–53 (discussing these cases). The availability of disparate-impact liability, furthermore, has allowed private developers to vindicate the FHA’s objectives and to protect their prop-erty rights by stopping municipalities from enforcing arbi-trary and, in practice, discriminatory ordinances barring the construction of certain types of housing units. See, e.g., Huntington, supra, at 18. Recognition of disparate-impact liability under the FHA also plays a role in uncovering discriminatory intent: It permits plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification as disparate treatment. In this way disparate-impact liability may prevent segregated housing patterns that might otherwise result from covert and illicit stereotyping. But disparate-impact liability has always been properly limited in key respects that avoid the serious constitutional questions that might arise under the FHA, for instance,if such liability were imposed based solely on a showing of a statistical disparity. Disparate-impact liability mandates the “removal of artificial, arbitrary, and unnecessary barriers,” not the displacement of valid governmental policies. Griggs, supra, at 431. The FHA is not an instrument to force housing authorities to reorder their priorities. Rather, the FHA aims to ensure that those priorities can be achieved without arbitrarily creating discriminatory effects or perpetuating segregation. Unlike the heartland of disparate-impact suits targeting artificial barriers to housing, the underlying dispute in this case involves a novel theory of liability. See Seicsh-naydre, Is Disparate Impact Having Any Impact? An Appellate Analysis of Forty Years of Disparate Impact Claims Under the Fair Housing Act, 63 Am. U. L. Rev. 357, 360–363 (2013) (noting the rarity of this type of claim). This case, on remand, may be seen simply as an attempt to second-guess which of two reasonable approaches a housing authority should follow in the sound exercise of its discretion in allocating tax credits for low-income housing. An important and appropriate means of ensuring that disparate-impact liability is properly limited is to give housing authorities and private developers leeway to state and explain the valid interest served by their policies. This step of the analysis is analogous to the business necessity standard under Title VII and provides a defense against disparate-impact liability. See 78 Fed. Reg. 11470 (explaining that HUD did not use the phrase “business necessity” because that “phrase may not be easily understood to cover the full scope of practices covered by the Fair Housing Act, which applies to individuals, busi-nesses, nonprofit organizations, and public entities”). As the Court explained in Ricci, an entity “could be liable for disparate-impact discrimination only if the [challenged practices] were not job related and consistent with business necessity.” 557 U. S., at 587. Just as an employer may maintain a workplace requirement that causes a disparate impact if that requirement is a “reasonable measure[ment] of job performance,” Griggs, supra, at 436, so too must housing authorities and private developers be allowed to maintain a policy if they can prove it is necessary to achieve a valid interest. To be sure, the Title VII framework may not transfer exactly to the fair-housing context, but the comparison suffices for present purposes. It would be paradoxical to construe the FHA to impose onerous costs on actors who encourage revitalizing dilapidated housing in our Nation’s cities merely because some other priority might seem preferable. Entrepreneurs must be given latitude to consider market factors. Zoning officials, moreover, must often make decisions based on a mix of factors, both objective (such as cost and traffic patterns) and, at least to some extent, subjective (such as preserving historic architecture). These factors contribute to a community’s quality of life and are legitimate concerns for housing authorities. The FHA does not decree a particular vision of urban development; and it does not put housing authorities and private developers in a double bind of liability, subject to suit whether they choose to rejuvenate a city core or to promote new low-income housing in suburban communities. As HUD itself recognized in its recent rulemaking, disparate-impact liability “does not mandate that affordable housing be located in neighborhoods with any particular characteristic.” 78 Fed. Reg. 11476. In a similar vein, a disparate-impact claim that relies on a statistical disparity must fail if the plaintiff cannot point to a defendant’s policy or policies causing that disparity. A robust causality requirement ensures that “[r]acial imbalance . . . does not, without more, establish a prima facie case of disparate impact” and thus protects defendants from being held liable for racial disparities they did not create. Wards Cove Packing Co. v. Atonio, 490 U. S. 642, 653 (1989) , superseded by statute on other grounds, 42 U. S. C. §2000e–2(k). Without adequate safeguards at the prima facie stage, disparate-impact liability might cause race to be used and considered in a pervasive way and “would almost inexorably lead” governmental or private entities to use “numerical quotas,” and serious constitutional questions then could arise. 490 U. S., at 653. The litigation at issue here provides an example. From the standpoint of determining advantage or disadvantage to racial minorities, it seems difficult to say as a general matter that a decision to build low-income housing in a blighted inner-city neighborhood instead of a suburb is discriminatory, or vice versa. If those sorts of judgments are subject to challenge without adequate safeguards, then there is a danger that potential defendants may adopt racial quotas—a circumstance that itself raises serious constitutional concerns. Courts must therefore examine with care whether a plaintiff has made out a prima facie case of disparate impact and prompt resolution of these cases is important. A plaintiff who fails to allege facts at the pleading stage or produce statistical evidence demonstrating a causal connection cannot make out a prima facie case of disparate impact. For instance, a plaintiff challenging the decision of a private developer to construct a new building in one location rather than another will not easily be able to show this is a policy causing a disparate impact because such a one-time decision may not be a policy at all. It may also be difficult to establish causation because of the multiple factors that go into investment decisions about where to construct or renovate housing units. And as Judge Jones observed below, if the ICP cannot show a causal connection between the Department’s policy and a disparate impact—for instance, because federal law substantially limits the Department’s discretion—that should resultin dismissal of this case. 747 F. 3d, at 283–284 (specially concurring opinion). The FHA imposes a command with respect to disparate-impact liability. Here, that command goes to a state entity. In other cases, the command will go to a private person or entity. Governmental or private policies are not contrary to the disparate-impact requirement unless they are “artificial, arbitrary, and unnecessary barriers.” Griggs, 401 U. S., at 431. Difficult questions might arise if disparate-impact liability under the FHA caused race to be used and considered in a pervasive and explicit manner to justify governmental or private actions that, in fact, tend to perpetuate race-based considerations rather than move beyond them. Courts should avoid interpreting disparate-impact liability to be so expansive as to inject racial considerations into every housing decision. The limitations on disparate-impact liability discussed here are also necessary to protect potential defendants against abusive disparate-impact claims. If the specter of disparate-impact litigation causes private developers to no longer construct or renovate housing units for low-income individuals, then the FHA would have undermined its own purpose as well as the free-market system. And as to governmental entities, they must not be prevented from achieving legitimate objectives, such as ensuring compliance with health and safety codes. The Department’s amici, in addition to the well-stated principal dissenting opinion in this case, see post, at 1–2, 29–30 (opinion of Alito, J.), call attention to the decision by the Court of Appeals for the Eighth Circuit in Gallagher v. Magner, 619 F. 3d 823 (2010). Although the Court is reluctant to approve or disapprove a case that is not pending, it should be noted that Magner was decided without the cautionary standards announced in this opinion and, in all events, the case was settled by the parties before an ultimate determination of disparate-impact liability. Were standards for proceeding with disparate-impact suits not to incorporate at least the safeguards discussed here, then disparate-impact liability might displace valid governmental and private priorities, rather than solely “remov[ing] . . . artificial, arbitrary, and unnecessary barriers.” Griggs, 401 U. S., at 431. And that, in turn, would set our Nation back in its quest to reduce the salience of race in our social and economic system. It must be noted further that, even when courts do find liability under a disparate-impact theory, their remedial orders must be consistent with the Constitution. Remedial orders in disparate-impact cases should concentrate onthe elimination of the offending practice that “arbitrar[ily] . . . operate[s] invidiously to discriminate on the basis of rac[e].” Ibid. If additional measures are adopted, courts should strive to design them to eliminate racial disparities through race-neutral means. See Richmond v. J. A. Croson Co., 488 U. S. 469, 510 (1989) (plurality opinion) (“[T]he city has at its disposal a whole array of race-neutral devices to increase the accessibility of city contracting opportunities to small entrepreneurs of all races”). Remedial orders that impose racial targets or quotas might raise more difficult constitutional questions. While the automatic or pervasive injection of race into public and private transactions covered by the FHA has special dangers, it is also true that race may be considered in certain circumstances and in a proper fashion. Cf. Parents Involved in Community Schools v. Seattle School Dist. No. 1, 551 U. S. 701, 789 (2007) (Kennedy, J., concurring in part and concurring in judgment) (“School boards may pursue the goal of bringing together students of diverse backgrounds and races through other means, including strategic site selection of new schools; [and] drawing attendance zones with general recognition of the demographics of neighborhoods”). Just as this Court has not “question[ed] an employer’s affirmative efforts to ensure that all groups have a fair opportunity to apply for promotions and to participate in the [promotion] process,” Ricci, 557 U. S., at 585, it likewise does not impugn housing authorities’ race-neutral efforts to encourage revitalization of communities that have long suffered the harsh consequences of segregated housing patterns. When setting their larger goals, local housing authorities may choose to foster diversity and combat racial isolation with race-neutral tools, and mere awareness of race in attempting to solve the problems facing inner cities does not doom that endeavor at the outset. The Court holds that disparate-impact claims are cognizable under the Fair Housing Act upon considering its results-oriented language, the Court’s interpretation of similar language in Title VII and the ADEA, Congress’ ratification of disparate-impact claims in 1988 against the backdrop of the unanimous view of nine Courts of Appeals, and the statutory purpose. III In light of the longstanding judicial interpretation of the FHA to encompass disparate-impact claims and congressional reaffirmation of that result, residents and policymakers have come to rely on the availability of disparate-impact claims. See Brief for Massachusetts et al. as Amici Curiae 2 (“Without disparate impact claims, States and others will be left with fewer crucial tools to combat the kinds of systemic discrimination that the FHA was intended to address”). Indeed, many of our Nation’s largest cities—entities that are potential defendants in disparate-impact suits—have submitted an amicus brief in this case supporting disparate-impact liability under the FHA. See Brief for City of San Francisco et al. as Amici Curiae 3–6. The existence of disparate-impact liability in the substantial majority of the Courts of Appeals for the last several decades “has not given rise to . . . dire consequences.” Hosanna-Tabor Evangelical Lutheran Church and School v. EEOC, 565 U. S. ___, ___ (2012) (slip op., at 21). Much progress remains to be made in our Nation’s continuing struggle against racial isolation. In striving to achieve our “historic commitment to creating an integrated society,” Parents Involved, supra, at 797 (Kennedy, J., concurring in part and concurring in judgment), we must remain wary of policies that reduce homeowners to nothing more than their race. But since the passage of the Fair Housing Act in 1968 and against the backdrop of disparate-impact liability in nearly every jurisdiction, many cities have become more diverse. The FHA must play an important part in avoiding the Kerner Commission’s grim prophecy that “[o]ur Nation is moving toward two societies, one black, one white—separate and un-equal.” Kerner Commission Report 1. The Court acknowl-edges the Fair Housing Act’s continuing role in moving the Nation toward a more integrated society. The judgment of the Court of Appeals for the Fifth Circuit is affirmed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

575.US.2014_13-550

In 2007, petitioners, beneficiaries of the Edison 401(k) Savings Plan (Plan), sued Plan fiduciaries, respondents Edison International and others, to recover damages for alleged losses suffered by the Plan from alleged breaches of respondents’ fiduciary duties. As relevant here, petitioners argued that respondents violated their fiduciary duties with respect to three mutual funds added to the Plan in 1999 and three mutual funds added to the Plan in 2002. Petitioners argued that respondents acted imprudently by offering six higher priced retail-class mutual funds as Plan investments when materially identical lower priced institutional-class mutual funds were available. Because ERISA requires a breach of fiduciary duty complaint to be filed no more than six years after “the date of the last action which constitutes a part of the breach or violation” or “in the case of an omission the latest date on which the fiduciary could have cured the breach or violation,” 29 U. S. C. §1113, the District Court held that petitioners’ complaint as to the 1999 funds was untimely because they were included in the Plan more than six years before the complaint was filed, and the circumstances had not changed enough within the 6-year statutory period to place respondents under an obligation to review the mutual funds and to convert them to lower priced institutional-class funds. The Ninth Circuit affirmed, concluding that petitioners had not established a change in circumstances that might trigger an obligation to conduct a full due diligence review of the 1999 funds within the 6-year statutory period. Held: The Ninth Circuit erred by applying §1113’s statutory bar to a breach of fiduciary duty claim based on the initial selection of the investments without considering the contours of the alleged breach of fiduciary duty. ERISA’s fiduciary duty is “derived from the common law of trusts,” Central States, Southeast & Southwest Areas Pension Fund v. Central Transport, Inc., 472 U. S. 559 , which provides that a trustee has a continuing duty—separate and apart from the duty to exercise prudence in selecting investments at the outset—to monitor, and remove imprudent, trust investments. So long as a plaintiff’s claim alleging breach of the continuing duty of prudence occurred within six years of suit, the claim is timely. This Court expresses no view on the scope of respondents’ fiduciary duty in this case, e.g., whether a review of the contested mutual funds is required, and, if so, just what kind of review. A fiduciary must discharge his responsibilities “with the care, skill, prudence, and diligence” that a prudent person “acting in a like capacity and familiar with such matters” would use. §1104(a)(1). The case is remanded for the Ninth Circuit to consider petitioners’ claims that respondents breached their duties within the relevant 6-year statutory period under §1113, recognizing the importance of analogous trust law. Pp. 4–8. 729 F. 3d 1110, vacated and remanded. Breyer, J., delivered the opinion for a unanimous Court.

Under the Employee Retirement Income Security Act of 1974 (ERISA), 88Stat. 829 et seq., as amended, a breach of fiduciary duty complaint is timely if filed no more than six years after “the date of the last action which constituted a part of the breach or violation” or “in the case of an omission the latest date on which the fiduciary could have cured the breach or violation.” 29 U. S. C. §1113. The question before us concerns application of this provision to the timeliness of a fiduciary duty complaint. It requires us to consider whether a fiduciary’s allegedly imprudent retention of an investment is an “action” or “omission” that triggers the running of the 6-year limitations period. In 2007, several individual beneficiaries of the Edison 401(k) Savings Plan (Plan) filed a lawsuit on behalf of the Plan and all similarly situated beneficiaries (collectively, petitioners) against Edison International and others (collectively, respondents). Petitioners sought to recover damages for alleged losses suffered by the Plan, in addition to injunctive and other equitable relief based on al-leged breaches of respondents’ fiduciary duties. The Plan is a defined-contribution plan, meaning that participants’ retirement benefits are limited to the value of their own individual investment accounts, which is determined by the market performance of employee and employer contributions, less expenses. Expenses, such as management or administrative fees, can sometimes significantly reduce the value of an account in a defined-contribution plan. As relevant here, petitioners argued that respondents violated their fiduciary duties with respect to three mu-tual funds added to the Plan in 1999 and three mutual funds added to the Plan in 2002. Petitioners argued that respondents acted imprudently by offering six higher priced retail-class mutual funds as Plan investments when materially identical lower priced institutional-class mutual funds were available (the lower price reflects lower administrative costs). Specifically, petitioners claimed that a large institutional investor with billions of dollars, like the Plan, can obtain materially identical lower priced institutional-class mutual funds that are not available to a retail investor. Petitioners asked, how could respondents have acted prudently in offering the six higher priced retail-class mutual funds when respondents could have offered them effectively the same six mutual funds at the lower price offered to institutional investors like the Plan? As to the three funds added to the Plan in 2002, the District Court agreed. It wrote that respondents had “not offered any credible explanation” for offering retail-class, i.e., higher priced mutual funds that “cost the Plan participants wholly unnecessary [administrative] fees,” and it concluded that, with respect to those mutual funds, respondents had failed to exercise “the care, skill, prudence and diligence under the circumstances” that ERISA demands of fiduciaries. No. CV 07–5359 (CD Cal., July 8, 2010), App. to Pet. for Cert. 65, 130, 142, 109. As to the three funds added to the Plan in 1999, how-ever, the District Court held that petitioners’ claims were untimely because, unlike the other contested mutual funds, these mutual funds were included in the Plan more than six years before the complaint was filed in 2007. 639 F. Supp. 2d 1074, 1119–1120 (CD Cal. 2009). As a result, the 6-year statutory period had run. The District Court allowed petitioners to argue that, despite the 1999 selection of the three mutual funds, their complaint was nevertheless timely because these funds underwent significant changes within the 6-year statutory period that should have prompted respondents to undertake a full due-diligence review and convert the higher priced retail-class mutual funds to lower priced institutional-class mutual funds. App. to Pet. for Cert. 142–150. The District Court concluded, however, that petitioners had not met their burden of showing that a prudent fiduciary would have undertaken a full due-diligence review of these funds as a result of the alleged changed circumstances. According to the District Court, the circumstances had not changed enough to place respondents under an obligation to review the mutual funds and to convert them to lower priced institutional-class mutual funds. Ibid. The Ninth Circuit affirmed the District Court as to the six mutual funds. 729 F. 3d 1110 (2013). With respect to the three mutual funds added in 1999, the Ninth Circuit held that petitioners’ claims were untimely because petitioners had not established a change in circumstances that might trigger an obligation to review and to change investments within the 6-year statutory period. Petitioners filed a petition for certiorari asking us to review this latter holding. We agreed to do so. Section 1113 reads, in relevant part, that “[n]o action may be commenced with respect to a fiduciary’s breach of any responsibility, duty, or obligation” after the earlier of “six years after (A) the date of the last action which constituted a part of the breach or violation, or (B) in the case of an omission the latest date on which the fiduciary could have cured the breach or violation.” Both clauses of that provision require only a “breach or violation” to start the 6-year period. Petitioners contend that respondents breached the duty of prudence by offering higher priced retail-class mutual funds when the same investments were available as lower priced institutional-class mutual funds. The Ninth Circuit, without considering the role of the fiduciary’s duty of prudence under trust law, rejected petitioners’ claims as untimely under §1113 on the basis that respondents had selected the three mutual funds more than six years before petitioners brought this action. The Ninth Circuit correctly asked whether the “last action which constituted a part of the breach or violation” of respondents’ duty of prudence occurred within the relevant 6-year period. It focused, however, upon the act of “designating an investment for inclusion” to start the 6-year period. 729 F. 3d, at 1119. The Ninth Circuit stated that “[c]haracterizing the mere continued offering of a plan option, without more, as a subsequent breach would render” the statute meaningless and could even expose present fiduciaries to liability for decisions made decades ago. Id., at 1120. But the Ninth Circuit jumped from this observation to the conclusion that only a significant change in circumstances could engender a new breach of a fiduciary duty, stating that the District Court was “entirely correct” to have entertained the “possibility” that “sig-nificant changes” occurring “within the limitations period” might require “ ‘a full due diligence review of the funds,’ ” equivalent to the diligence review that respondents conduct when adding new funds to the Plan. Ibid. We believe the Ninth Circuit erred by applying a statu-tory bar to a claim of a “breach or violation” of a fiduciary duty without considering the nature of the fiduciary duty. The Ninth Circuit did not recognize that under trust law a fiduciary is required to conduct a regular review of its investment with the nature and timing of the review contingent on the circumstances. Of course, after the Ninth Circuit considers trust-law principles, it is possible that it will conclude that respondents did indeed conduct the sort of review that a prudent fiduciary would have conducted absent a significant change in circumstances. An ERISA fiduciary must discharge his responsibility “with the care, skill, prudence, and diligence” that a prudent person “acting in a like capacity and familiar with such matters” would use. §1104(a)(1); see also Fifth Third Bancorp v. Dudenhoeffer, 573 U. S. ___ (2014). We have often noted that an ERISA fiduciary’s duty is “derived from the common law of trusts.” Central States, Southeast & Southwest Areas Pension Fund v. Central Transport, Inc., 472 U. S. 559, 570 (1985) . In determining the contours of an ERISA fiduciary’s duty, courts often must look to the law of trusts. We are aware of no reason why the Ninth Circuit should not do so here. Under trust law, a trustee has a continuing duty to monitor trust investments and remove imprudent ones. This continuing duty exists separate and apart from the trustee’s duty to exercise prudence in selecting investments at the outset. The Bogert treatise states that “[t]he trustee cannot assume that if investments are legal and proper for retention at the beginning of the trust, or when purchased, they will remain so indefinitely.” A. Hess, G. Bogert, & G. Bogert, Law of Trusts and Trustees §684, pp. 145–146 (3d ed. 2009) (Bogert 3d). Rather, the trustee must “systematic[ally] conside[r] all the investments of the trust at regular intervals” to ensure that they are appropriate. Bogert 3d §684, at 147–148; see also In re Stark’s Estate, 15 N. Y. S. 729, 731 (Surr. Ct. 1891) (stating that a trustee must “exercis[e] a reasonable degree of diligence in looking after the security after the investment had been made”); Johns v. Herbert, 2 App. D. C. 485, 499 (1894) (holding trustee liable for failure to discharge his “duty to watch the investment with reasonable care and diligence”). The Restatement (Third) of Trusts states the following: “[A] trustee’s duties apply not only in making investments but also in monitoring and reviewing investments, which is to be done in a manner that isreasonable and appropriate to the particular investments, courses of action, and strategies involved.” §90, Comment b, p. 295 (2007). The Uniform Prudent Investor Act confirms that “[m]anaging embraces monitoring” and that a trustee has “continuing responsibility for oversight of the suitability of the investments already made.” §2, Comment, 7B U. L. A. 21 (1995) (internal quotation marks omitted). Scott on Trusts implies as much by stating that, “[w]hen the trust estate includes assets that are inappropriate as trust investments, the trustee is ordinarily under a duty to dispose of them within a reasonable time.” 4 A. Scott, W. Fratcher, & M. Ascher, Scott and Ascher on Trusts §19.3.1, p. 1439 (5th ed. 2007). Bogert says the same. Bogert 3d §685, at 156–157 (explaining that if an investment is determined to be imprudent, the trustee “must dispose of it within a reasonable time”); see, e.g., State Street Trust Co. v. DeKalb, 259 Mass. 578, 583, 157 N. E. 334, 336 (1927) (trustee was required to take action to “protect the rights of the beneficiaries” when the value of trust assets declined). In short, under trust law, a fiduciary normally has a continuing duty of some kind to monitor investments and remove imprudent ones. A plaintiff may allege that a fiduciary breached the duty of prudence by failing to properly monitor investments and remove imprudent ones. In such a case, so long as the alleged breach of the continuing duty occurred within six years of suit, the claim is timely. The Ninth Circuit erred by applying a 6-year statutory bar based solely on the initial selection of the three funds without considering the contours of the alleged breach of fiduciary duty. The parties now agree that the duty of prudence involves a continuing duty to monitor investments and remove imprudent ones under trust law. Brief for Petitioners 24 (“Trust law imposes a duty to examine the prudence of existing investments periodically and to remove imprudent investments”); Brief for Respondents 3 (“All agree that a fiduciary has an ongoing duty to monitor trust investments to ensure that they remain prudent”); Brief for United States as Amicus Curiae 7 (“The duty of prudence under ERISA, as under trust law, requires plan fiduciaries with investment responsibility to examine periodically the prudence of existing investments and to remove imprudent investments within a reasonable period of time”). The parties disagree, however, with respect to the scope of that responsibility. Did it require a review of the contested mutual funds here, and if so, just what kind of review did it require? A fiduciary must discharge his responsibilities “with the care, skill, prudence, and diligence” that a prudent person “acting in a like capacity and familiar with such matters” would use. §1104(a)(1). We express no view on the scope of respondents’ fiduciary duty in this case. We remand for the Ninth Circuit to consider petitioners’ claims that respondents breached their duties within the relevant 6-year period under §1113, recognizing the importance of analogous trust law. A final point: Respondents argue that petitioners did not raise the claim below that respondents committed new breaches of the duty of prudence by failing to monitor their investments and remove imprudent ones absent a significant change in circumstances. We leave any questions of forfeiture for the Ninth Circuit on remand. The Ninth Circuit’s judgment is vacated, and the case isremanded for further proceedings consistent with this opinion. It is so ordered.

575.US.2014_13-1080

In 1970, Congress created the National Railroad Passenger Corporation (Amtrak). Congress has given Amtrak priority to use track systems owned by the freight railroads for passenger rail travel, at rates agreed to by the parties or, in case of a dispute, set by the Surface Transportation Board. And in 2008, Congress gave Amtrak and the Federal Railroad Administration (FRA) joint authority to issue “metrics and standards” addressing the performance and scheduling of passenger railroad services, see §207(a), 122Stat. 4907, including Amtrak’s on-time performance and train delays caused by host railroads. Respondent, the Association of American Railroads, sued petitioners—the Department of Transportation, the FRA, and two officials—claiming that the metrics and standards must be invalidated because it is unconstitutional for Congress to allow and direct a private entity like Amtrak to exercise joint authority in their issuance. Its argument rested on the Fifth Amendment Due Process Clause and the constitutional provisions regarding separation of powers. The District Court rejected respondent’s claims, but the District of Columbia Circuit reversed as to the separation of powers claim, reasoning in central part that Amtrak is a private corporation and thus cannot constitutionally be granted regulatory power under §207. Held: For purposes of determining the validity of the metrics and standards, Amtrak is a governmental entity. Pp. 6–12. (a) In concluding otherwise, the Court of Appeals relied on the statutory command that Amtrak “is not a department, agency, or instrumentality of the United States Government,” 49 U. S. C. §24301(a)(3), and the pronouncement that Amtrak “shall be operated and managed as a for profit corporation,” §24301(a)(2). But congressional pronouncements are not dispositive of Amtrak’s status as a governmental entity for purposes of separation of powers analysis under the Constitution, and an independent inquiry reveals the Court of Appeals’ premise that Amtrak is a private entity was flawed. As Amtrak’s ownership and corporate structure show, the political branches control most of Amtrak’s stock and its Board of Directors, most of whom are appointed by the President, §24302(a)(1), confirmed by the Senate, ibid., and understood by the Executive Branch to be removable by the President at will. The political branches also exercise substantial, statutorily mandated supervision over Amtrak’s priorities and operations. See, e.g., §24315. Also of significance, Amtrak is required by statute to pursue broad public objectives, see, e.g., §§24101(b), 24307(a); certain aspects of Amtrak’s day-to-day operations are mandated by Congress, see, e.g., §§24101(c)(6), 24902(b); and Amtrak has been dependent on federal financial support during every year of its existence. Given the combination of these unique features and Amtrak’s significant ties to the Government, Amtrak is not an autonomous private enterprise. Amtrak was created by the Government, is controlled by the Government, and operates for the Government’s benefit. Thus, in jointly issuing the metrics and standards with the FRA, Amtrak acted as a governmental entity for separation of powers purposes. And that exercise of governmental power must be consistent with the Constitution, including those provisions relating to the separation of powers. Pp. 6–10. (b) Respondent’s reliance on congressional statements about Amtrak’s status is misplaced. Lebron v. National Railroad Passenger Corp., 513 U. S. 374 , teaches that, for purposes of Amtrak’s status as a federal actor or instrumentality under the Constitution, the practical reality of federal control and supervision prevails over Congress’ disclaimer of Amtrak’s governmental status. Treating Amtrak as governmental for these purposes, moreover, is not an unbridled grant of authority to an unaccountable actor, for the political branches created Amtrak, control its Board, define its mission, specify many of its day-to-day operations, have imposed substantial transparency and accountability mechanisms, and, for all practical purposes, set and supervise its annual budget. Pp. 10–11. (c) The Court of Appeals may address in the first instance any properly preserved issues respecting the lawfulness of the metrics and standards that may remain in this case, including questions implicating the Constitution’s structural separation of powers and the Appointments Clause. Pp. 11–12. 721 F. 3d 666, vacated and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Ginsburg, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Alito, J., filed a concurring opinion. Thomas, J., filed an opinion concurring in the judgment.

In 1970, Congress created the National Railroad Passenger Corporation, most often known as Amtrak. Later, Congress granted Amtrak and the Federal Railroad Administration (FRA) joint authority to issue metrics and standards that address the performance and scheduling of passenger railroad services. Alleging that the metrics and standards have substantial and adverse effects upon its members freight services, respondent the Association of American Railroads filed this suit to challenge their validity. The defendants below, petitioners here, are the Department of Transportation, the FRA, and two individuals sued in their official capacity. Respondent alleges the metrics and standards must be invalidated on the ground that Amtrak is a private entity and it was therefore unconstitutional for Congress to allow and direct it to exercise joint authority in their issuance. This argument rests on the Fifth Amendment Due Process Clause and the constitutional provisions regarding separation of powers. The District Court rejected both of respondent s claims. The Court of Appeals for the District of Columbia Circuit reversed, finding that, for purposes of this dispute, Amtrak is a private entity and that Congress violated nondelegation principles in its grant of joint authority to Amtrak and the FRA. On that premise the Court of Appeals invalidated the metrics and standards. Having granted the petition for writ of certiorari, 573 U. S. ___ (2014), this Court now holds that, for purposes of determining the validity of the metrics and standards, Amtrak is a governmental entity. Although Amtrak s actions here were governmental, substantial questions respecting the lawfulness of the metrics and standards including questions implicating the Constitution s structural separation of powers and the Appointments Clause, U. S. Const., Art. II, 2, cl. 2 may still remain in the case. As those matters have not yet been passed upon by the Court of Appeals, this case is remanded. I A Amtrak is a corporation established and authorized by a detailed federal statute enacted by Congress for no less a purpose than to preserve passenger services and routes on our Nation s railroads. See Lebron v. National Railroad Passenger Corporation, 513 U. S. 374 384 (1995); National Railroad Passenger Corporation v. Atchison, T. & S. F. R. Co., 470 U. S. 451 457 (1985); see also Rail Passenger Service Act of 1970, 84Stat. 1328. Congress recognized that Amtrak, of necessity, must rely for most of its operations on track systems owned by the freight railroads. So, as a condition of relief from their common-carrier duties, Congress required freight railroads to allow Amtrak to use their tracks and facilities at rates agreed to by the parties or in the event of disagreement to be set by the Interstate Commerce Commission (ICC). See 45 U. S. C. 561, 562 (1970 ed.). The Surface Transportation Board (STB) now occupies the dispute-resolution role originally assigned to the ICC. See 49 U. S. C. 24308(a) (2012 ed.). Since 1973, Amtrak has received a statutory preference over freight transportation in using rail lines, junctions, and crossings. See 24308(c). The metrics and standards at issue here are the result of a further and more recent enactment. Concerned by poor service, unreliability, and delays resulting from freight traffic congestion, Congress passed the Passenger Rail Investment and Improvement Act (PRIIA) in 2008. See 122Stat. 4907. Section 207(a) of the PRIIA provides for the creation of the metrics and standards: Within 180 days after the date of enactment of this Act, the Federal Railroad Administration and Amtrak shall jointly, in consultation with the Surface Transportation Board, rail carriers over whose rail lines Amtrak trains operate, States, Amtrak employees, nonprofit employee organizations representing Amtrak employees, and groups representing Amtrak passengers, as appropriate, develop new or improve existing metrics and minimum standards for measuring the performance and service quality of intercity passenger train operations, including cost recovery, on-time performance and minutes of delay, ridership, on-board services, stations, facilities, equipment, and other services. Id., at 4916. Section 207(d) of the PRIIA further provides: If the development of the metrics and standards is not completed within the 180-day period required by subsection (a), any party involved in the development of those standards may petition the Surface Transportation Board to appoint an arbitrator to assist the parties in resolving their disputes through binding arbitration. Id., at 4917. The PRIIA specifies that the metrics and standards created under 207(a) are to be used for a variety of purposes. Section 207(b) requires the FRA to publish a quarterly report on the performance and service quality of intercity passenger train operations addressing the specific elements to be measured by the metrics and standards. Id., at 4916 4917. Section 207(c) provides that, [t]o the extent practicable, Amtrak and its host rail carriers shall incorporate the metrics and standards developed under subsection (a) into their access and service agreements. Id., at 4917. And 222(a) obliges Amtrak, within one year after the metrics and standards are established, to develop and implement a plan to improve on-board service pursuant to the metrics and standards for such service developed under [ 207(a)]. Id., at 4932. Under 213(a) of the PRIIA, the metrics and standards also may play a role in prompting investigations by the STB and in subsequent enforcement actions. For instance, [i]f the on-time performance of any intercity passenger train averages less than 80 percent for any 2 consecutive calendar quarters, the STB may initiate an investigation to determine whether and to what extent delays . . . are due to causes that could reasonably be addressed . . . by Amtrak or other intercity passenger rail operators. Id., at 4925 4926. While conducting an investigation under 213(a), the STB has authority to review the accuracy of the train performance data and the extent to which scheduling and congestion contribute to delays and shall obtain information from all parties involved and identify reasonable measures and make recommendations to improve the service, quality, and on-time performance of the train. Id., at 4926. Following an investigation, the STB may award damages if it determines that delays or failures to achieve minimum standards . . . are attributable to a rail carrier s failure to provide preference to Amtrak over freight transportation. Ibid. The STB is further empowered to order the host rail carrier to remit damages to Amtrak or to an entity for which Amtrak operates intercity passenger rail service. Ibid. B In March 2009, Amtrak and the FRA published a notice in the Federal Register inviting comments on a draft version of the metrics and standards. App. 75 76. The final version of the metrics and standards was issued jointly by Amtrak and the FRA in May 2010. Id., at 129 144. The metrics and standards address, among other matters, Amtrak s financial performance, its scores on consumer satisfaction surveys, and the percentage of passenger-trips to and from underserved communities. Of most importance for this case, the metrics and standards also address Amtrak s on-time performance and train delays caused by host railroads. The standards associated with the on-time performance metrics require on-time performance by Amtrak trains at least 80% to 95% of the time for each route, depending on the route and year. Id., at 133 135. With respect to host-responsible delays that is to say, delays attributed to the railroads along which Amtrak trains travel the metrics and standards provide that [d]elays must not be more than 900 minutes per 10,000 Train-Miles. Id., at 138. Amtrak conductors determine responsibility for particular delays. Ibid., n. 23. In the District Court for the District of Columbia, respondent alleged injury to its members from being required to modify their rail operations, which mostly involve freight traffic, to satisfy the metrics and standards. Respondent claimed that 207 violates the nondelegation doctrine and the separation of powers principle by placing legislative and rulemaking authority in the hands of a private entity [Amtrak] that participates in the very industry it is supposed to regulate. Id., at 176 177, Complaint 51. Respondent also asserted that 207 violates the Fifth Amendment Due Process Clause by [v]esting the coercive power of the government in Amtrak, an interested private part[y]. Id., at 177, 53 54. In its prayer for relief respondent sought, among other remedies, a declaration of 207 s unconstitutionality and invalidation of the metrics and standards. Id., at 177. The District Court granted summary judgment to petitioners on both claims. See 865 F. Supp. 2d 22 (DC 2012). Without deciding whether Amtrak must be deemed private or governmental, it rejected respondent s nondelegation argument on the ground that the FRA, the STB, and the political branches exercised sufficient control over promulgation and enforcement of the metrics and standards so that 207 is constitutional. See id., at 35. The Court of Appeals for the District of Columbia Circuit reversed the judgment of the District Court as to the nondelegation and separation of powers claim, reasoning in central part that because Amtrak is a private corporation with respect to Congress s power to delegate . . . authority, it cannot constitutionally be granted the regulatory power prescribed in 207. 721 F. 3d 666, 677 (2013). The Court of Appeals did not reach respondent s due process claim. See ibid. II In holding that Congress may not delegate to Amtrak the joint authority to issue the metrics and standards authority it described as regulatory power, ibid. the Court of Appeals concluded Amtrak is a private entity for purposes of determining its status when considering the constitutionality of its actions in the instant dispute. That court s analysis treated as controlling Congress statutory command that Amtrak is not a department, agency, or instrumentality of the United States Government. Id., at 675 (quoting 49 U. S. C. 24301(a)(3)). The Court of Appeals also relied on Congress pronouncement that Amtrak shall be operated and managed as a for-profit corporation. 721 F. 3d, at 675 (quoting 24301(a)(2)); see also id., at 677 ( Though the federal government s involvement in Amtrak is considerable, Congress has both designated it a private corporation and instructed that it be managed so as to maximize profit. In deciding Amtrak s status for purposes of congressional delegations, these declarations are dispositive ). Proceeding from this premise, the Court of Appeals concluded it was impermissible for Congress to delegate regulatory authority to a private entity. Id., at 670; see also ibid. (holding Carter v. Carter Coal Co., 298 U. S. 238 (1936) , prohibits any such delegation of authority). That premise, however, was erroneous. Congressional pronouncements, though instructive as to matters within Congress authority to address, see, e.g., United States ex rel. Totten v. Bombardier Corp., 380 F. 3d 488, 491 492 (CADC 2004) (Roberts, J.), are not dispositive of Amtrak s status as a governmental entity for purposes of separation of powers analysis under the Constitution. And an independent inquiry into Amtrak s status under the Constitution reveals the Court of Appeals premise was flawed. It is appropriate to begin the analysis with Amtrak s ownership and corporate structure. The Secretary of Transportation holds all of Amtrak s preferred stock and most of its common stock. Amtrak s Board of Directors is composed of nine members, one of whom is the Secretary of Transportation. Seven other Board members areappointed by the President and confirmed by the Senate. 49 U. S. C. 24302(a)(1). These eight Board members,in turn, select Amtrak s president. 24302(a)(1)(B); 24303(a). Amtrak s Board members are subject to salary limits set by Congress, 24303(b); and the Executive Branch has concluded that all appointed Board members are removable by the President without cause, see 27 Op. Atty. Gen. 163 (2003). Under further statutory provisions, Amtrak s Board members must possess certain qualifications. Congress has directed that the President make appointments based on an individual s prior experience in the transportation industry, 24302(a)(1)(C), and has provided that not more than five of the seven appointed Board members be from the same political party, 24302(a)(3). In selecting Amtrak s Board members, moreover, the President must consult with leaders of both parties in both Houses of Congress in order to provide adequate and balanced representation of the major geographic regions of the United States served by Amtrak. 24302(a)(2). In addition to controlling Amtrak s stock and Board of Directors the political branches exercise substantial, statutorily mandated supervision over Amtrak s priorities and operations. Amtrak must submit numerous annual reports to Congress and the President, detailing such information as route-specific ridership and on-time performance. 24315. The Freedom of Information Act applies to Amtrak in any year in which it receives a federal subsidy, 5 U. S. C. 552, which thus far has been every year of its existence. Pursuant to its status under the Inspector General Act of 1978 as a designated Federal entity, 5 U. S. C. App. 8G(a)(2), p. 521, Amtrak must maintain an inspector general, much like governmental agencies such as the Federal Communications Commission and the Securities and Exchange Commission. Furthermore, Congress conducts frequent oversight hearings into Amtrak s budget, routes, and prices. See, e.g., Hearing on Reviewing Alternatives to Amtrak s Annual Losses in Food and Beverage Service before the Subcommittee on Government Operations of the House Committee on Oversight and Government Reform, 113th Cong., 1st Sess., 5 (2013) (statement of Thomas J. Hall, chief of customer service, Amtrak); Hearing on Amtrak s Fiscal Year 2014 Budget: The Starting Point for Reauthorization before the Subcommittee on Railroads, Pipelines, and Hazardous Materials of the House Committee on Transportation and Infrastructure, 113th Cong., 1st Sess., p. 6 (2013) (statement of Joseph H. Boardman, president and chief executive officer, Amtrak). It is significant that, rather than advancing its own private economic interests, Amtrak is required to pursue numerous, additional goals defined by statute. To take a few examples: Amtrak must provide efficient and effective intercity passenger rail mobility, 49 U. S. C. 24101(b); minimize Government subsidies, 24101(d); provide reduced fares to the disabled and elderly, 24307(a); and ensure mobility in times of national disaster, 24101(c)(9). In addition to directing Amtrak to serve these broad public objectives, Congress has mandated certain aspects of Amtrak s day-to-day operations. Amtrak must maintain a route between Louisiana and Florida. 24101(c)(6). When making improvements to the Northeast corridor, Amtrak must apply seven considerations in a specified order of priority. 24902(b). And when Amtrak purchases materials worth more than $1 million, these materials must be mined or produced in the United States, or manufactured substantially from components that are mined, produced, or manufactured in the United States, unless the Secretary of Transportation grants an exemption. 24305(f). Finally, Amtrak is also dependent on federal financial support. In its first 43 years of operation, Amtrak has received more than $41 billion in federal subsidies. In recent years these subsidies have exceeded $1 billion annually. See Brief for Petitioners 5, and n. 2, 46. Given the combination of these unique features and its significant ties to the Government, Amtrak is not an autonomous private enterprise. Among other important considerations, its priorities, operations, and decisions are extensively supervised and substantially funded by the political branches. A majority of its Board is appointed by the President and confirmed by the Senate and is understood by the Executive to be removable by the President at will. Amtrak was created by the Government, is controlled by the Government, and operates for the Government s benefit. Thus, in its joint issuance of the metrics and standards with the FRA, Amtrak acted as a governmental entity for purposes of the Constitution s separation of powers provisions. And that exercise of governmental power must be consistent with the design and requirements of the Constitution, including those provisions relating to the separation of powers. Respondent urges that Amtrak cannot be deemed a governmental entity in this respect. Like the Court of Appeals, it relies principally on the statutory directives that Amtrak shall be operated and managed as a for profit corporation and is not a department, agency, or instrumentality of the United States Government. 24301(a)(2) (3). In light of that statutory language, respondent asserts, Amtrak cannot exercise the joint authority entrusted to it and the FRA by 207(a). On that point this Court s decision in Lebron v. National Railroad Passenger Corp., 513 U. S. 374 (1995) , provides necessary instruction. In Lebron, Amtrak prohibited an artist from installing a politically controversial display in New York City s Penn Station. The artist sued Amtrak, alleging a violation of his First Amendment rights. In response Amtrak asserted that it was not a governmental entity, explaining that its charter s disclaimer of agency status prevent[ed] it from being considered a Government entity. Id., at 392. The Court rejected this contention, holding it is not for Congress to make the final determination of Amtrak s status as a Government entity for purposes of determining the constitutional rights of citizens affected by its actions. Ibid. To hold otherwise would allow the Government to evade the most solemn obligations imposed in the Constitution by simply resorting to the corporate form. Id., at 397. Noting that Amtrak is established and organized under federal law for the very purpose of pursuing federal governmental objectives, under the direction and control of federal governmental appointees, id., at 398, and that the Government exerts its control over Amtrak not as a creditor but as a policymaker, the Court held Amtrak is an agency or instrumentality of the United States for the purpose of individual rights guaranteed against the Government by the Constitution. Id., at 394, 399. Lebron teaches that, for purposes of Amtrak s status as a federal actor or instrumentality under the Constitution, the practical reality of federal control and supervision prevails over Congress disclaimer of Amtrak s governmental status. Lebron involved a First Amendment question, while in this case the challenge is to Amtrak s joint authority to issue the metrics and standards. But [t]he structural principles secured by the separation of powers protect the individual as well. Bond v. United States, 564 U. S. ___, ___ (2011) (slip op., at 10). Treating Amtrak as governmental for these purposes, moreover, is not an unbridled grant of authority to an unaccountable actor. The political branches created Amtrak, control its Board, define its mission, specify many of its day-to-day operations, have imposed substantial transparency and accountability mechanisms, and, for all practical purposes, set and supervise its annual budget. Accordingly, the Court holds that Amtrak is a governmental entity, not a private one, for purposes of determining the constitutional issues presented in this case. III Because the Court of Appeals decision was based on the flawed premise that Amtrak should be treated as a private entity, that opinion is now vacated. On remand, the Court of Appeals, after identifying the issues that are properly preserved and before it, will then have the instruction of the analysis set forth here. Respondent argues that the selection of Amtrak s president, who is appointed not by the President . . . but by the other eight Board Members, call[s] into question Amtrak s structure under the Appointments Clause, Brief for Respondent 42; that 207(d) s arbitrator provision is a plain violation of the nondelegation principle and the Appointments Clause requiring invalidation of 207(a), id., at 26; and that Congress violated the Due Process Clause by giv[ing] a federally chartered, nominally private, for-profit corporation regulatory authority over its own industry, id., at 43. Petitioners, in turn, contend that the metrics and standards do not reflect the exercise of rulemaking authority or permit Amtrak to regulate other private entities, and thus do not raise nondelegation concerns. Reply Brief 5 (internal citation omitted). Because [o]urs is a court of final review and not first view, Zivotofsky v. Clinton, 566 U. S. ___, ___ (2012) (slip op., at 12) (internal quotation marks omitted), those issues to the extent they are properly before the Court of Appeals should be addressed in the first instance on remand. The judgment of the Court of Appeals for the District of Columbia Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

575.US.2014_13-1074

The Federal Tort Claims Act (FTCA) provides that a tort claim against the United States “shall be forever barred” unless the claimant meets two deadlines. First, a claim must be presented to the appropriate federal agency for administrative review “within two years after [the] claim accrues.” 28 U. S. C. §2401(b). Second, if the agency denies the claim, the claimant may file suit in federal court “within six months” of the agency’s denial. Ibid. Kwai Fun Wong and Marlene June, respondents in Nos. 13–1074 and 13-1075, respectively, each missed one of those deadlines. Wong failed to file her FTCA claim in federal court within 6 months, but argued that that was only because the District Court had not permitted her to file that claim until after the period expired. June failed to present her FTCA claim to a federal agency within 2 years, but argued that her untimely filing should be excused because the Government had, in her view, concealed facts vital to her claim. In each case, the District Court dismissed the FTCA claim for failure to satisfy §2401(b)’s time bars, holding that, despite any justification for delay, those time bars are jurisdictional and not subject to equitable tolling. The Ninth Circuit reversed in both cases, concluding that §2401(b)’s time bars may be equitably tolled. Held: Section 2401(b)’s time limits are subject to equitable tolling. Pp. 4–18. (a) Irwin v. Department of Veterans Affairs, 498 U. S. 89 , provides the framework for deciding the applicability of equitable tolling to statutes of limitations on suits against the Government. There, the Court adopted a “rebuttable presumption” that such time bars may be equitably tolled. Id., at 95. Irwin’s presumption may, of course, be rebutted. One way to do so—pursued by the Government here—is to demonstrate that the statute of limitations at issue is jurisdictional; if so, the statute cannot be equitably tolled. But this Court will not conclude that a time bar is jurisdictional unless Congress provides a “clear statement” to that effect. Sebelius v. Auburn Regional Medical Center, 568 U. S. ___, ___. And in applying that clear statement rule, this Court has said that most time bars, even if mandatory and emphatic, are nonjurisdictional. See id., at ___. Congress thus must do something special to tag a statute of limitations as jurisdictional and so prohibit a court from tolling it. Pp. 4–7. (b) Congress did no such thing in enacting §2401(b). The text of that provision speaks only to a claim’s timeliness; it does not refer to the jurisdiction of the district courts or address those courts’ authority to hear untimely suits. See Arbaugh v. Y & H Corp., 546 U. S. 500 . Instead, it “reads like an ordinary, run-of-the-mill statute of limitations.” Holland v. Florida, 560 U. S. 631 . Statutory context confirms that reading. Congress’s separation of a filing deadline from a jurisdictional grant often indicates that the deadline is not jurisdictional, and here the FTCA’s jurisdictional grant appears not in §2401(b) but in another section of Title 28, §1346(b)(1). That jurisdictional grant is not expressly conditioned on compliance with §2401(b)’s limitations periods. Finally, assuming it could provide the clear statement that this Court’s cases require, §2401(b)’s legislative history does not clearly demonstrate that Congress intended the provision to impose a jurisdictional bar. Pp. 7–9. (c) The Government’s two principal arguments for treating §2401(b) as jurisdictional are unpersuasive and foreclosed by this Court’s precedents. Pp. 9–17. (1) The Government first points out that §2401(b) includes the same “shall be forever barred” language as the statute of limitations governing Tucker Act claims, which this Court has held to be jurisdictional. See, e.g., Kendall v. United States, 107 U. S. 123 –126. But that phrase was a commonplace in statutes of limitations enacted around the time of the FTCA, and it does not carry talismanic jurisdictional significance. Indeed, this Court has construed the same language to be subject to tolling in the Clayton Act’s statute of limitations. See American Pipe & Constr. Co. v. Utah, 414 U. S. 538 . And in two decisions addressing the Tucker Act’s statute of limitations, the Court has dismissed the idea that that language is jurisdictionally significant. See Irwin, 498 U. S., at 95; John R. Sand & Gravel Co. v. United States, 552 U. S. 130 . The “shall be forever barred” phrase is thus nothing more than an ordinary way to set a statutory deadline. Pp. 9–14. (2) The Government next argues that §2401(b) is jurisdictional because it is a condition on the FTCA’s waiver of sovereign immunity. But that argument is foreclosed by Irwin, which considered an identical objection but concluded that even time limits that condition a waiver of immunity may be equitably tolled. See 498 U. S., at 95–96. The Government’s invocation of sovereign immunity principles is also peculiarly inapt here. Unlike other waivers of sovereign immunity, the FTCA treats the Government much like a private party, and the Court has accordingly declined to construe the Act narrowly merely because it waives the Government’s immunity from suit. There is no reason to do differently here. Pp. 14–17. No. 13-1074, 732 F. 3d 1030, and No. 13-1075, 550 Fed. Appx. 505, affirmed and remanded. Kagan, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Scalia and Thomas, JJ., joined. Notes 1 Together with No. 13-1075, United States v. June, Conservator, also on certiorari to the same court.

The Federal Tort Claims Act (FTCA or Act) provides that a tort claim against the United States “shall be for-ever barred” unless it is presented to the “appropriate Fed-eral agency within two years after such claim accrues” and then brought to federal court “within six months” after the agency acts on the claim.28 U. S. C. §2401(b). In each of the two cases we resolve here, the claimant missed one of those deadlines, but requested equitable tolling on the ground that she had a good reason for filing late. The Government responded that §2401(b)’s time limits are not subject to tolling because they are jurisdictional restrictions. Today, we reject the Government’s argument and conclude that courts may toll both of the FTCA’s limitations periods.I In the first case, respondent Kwai Fun Wong asserts that the Immigration and Naturalization Service (INS) falsely imprisoned her for five days in 1999. As the FTCA requires, Wong first presented that claim to the INS within two years of the alleged unlawful action. See §2401(b); §2675(a). The INS denied the administrative complaint on December 3, 2001. Under the Act, that gave Wong six months, until June 3, 2002, to bring her tort claim in federal court. See §2401(b). Several months prior to the INS’s decision, Wong had filed suit in federal district court asserting various non-FTCA claims against the Government arising out of the same alleged misconduct. Anticipating the INS’s ruling, Wong moved in mid-November 2001 to amend the complaint in that suit by adding her tort claim. On April 5, 2002, a Magistrate Judge recommended granting Wong leave to amend. But the District Court did not finally adopt that proposal until June 25—three weeks after the FTCA’s 6-month deadline. The Government moved to dismiss the tort claim on the ground that it was filed late. The District Court at first rejected the motion. It recognized that Wong had managed to add her FTCA claim only after §2401(b)’s 6-month time period had expired. But the court equitably tolled that period for all the time between the Magistrate Judge’s recommendation and its own order allowing amendment, thus bringing Wong’s FTCA claim within the statutory deadline. Several years later, the Government moved for reconsideration of that ruling based on an intervening Ninth Circuit decision. This time, the District Court dismissed Wong’s claim, reasoning that §2401(b)’s 6-month time bar was jurisdictional and therefore not subject to equitable tolling. On appeal, the Ninth Circuit agreed to hear the case en banc to address an intra-circuit conflict on the issue. The en banc court held that the 6-month limit is not jurisdictional and that equitable tolling is available. Kwai Fun Wong v. Beebe, 732 F. 3d 1030 (2013). It then confirmed the District Court’s prior ruling that the circumstances here justify tolling because Wong “exercis[ed] due diligence” in attempting to amend her complaint before the statutory deadline. Id., at 1052. The second case before us arises from a deadly highway accident. Andrew Booth was killed in 2005 when a car in which he was riding crossed through a cable median barrier and crashed into oncoming traffic. The following year, respondent Marlene June, acting on behalf of Booth’s young son, filed a wrongful death action alleging that the State of Arizona and its contractor had negligently constructed and maintained the median barrier. Years into that state-court litigation, June contends, she discovered that the Federal Highway Administration (FHWA) had approved installation of the barrier knowing it had not been properly crash tested. Relying on that new information, June presented a tort claim to the FHWA in 2010, more than five years after the accident. The FHWA denied the claim, and June promptly filed this action in federal district court. The court dismissed the suit because June had failed to submit her claim to the FHWA within two years of the collision. The FTCA’s 2-year bar, the court ruled, is jurisdictional and therefore not subject to equitable tolling; accordingly, the court did not consider June’s contention that tolling was proper because the Government had concealed its failure to require crash testing. On appeal, the Ninth Circuit reversed in light of its recent decision in Wong, thus holding that §2401(b)’s 2-year deadline, like its 6-month counterpart, is not jurisdictional and may be tolled. 550 Fed. Appx. 505 (2013). We granted certiorari in both cases, 573 U. S. ___ (2014), to resolve a circuit split about whether courts may equitably toll §2401(b)’s two time limits. Compare, e.g., In re FEMA Trailer Formaldehyde Prods. Liability Litigation, 646 F. 3d 185, 190–191 (CA5 2011) (per curiam) (tolling not available), with Arteaga v. United States, 711 F. 3d 828, 832–833 (CA7 2013) (tolling allowed).[1] We now affirm the Court of Appeals’ rulings.II Irwin v. Department of Veterans Affairs,498 U. S. 89,95 (1990), sets out the framework for deciding “the applicability of equitable tolling in suits against the Government.” In Irwin, we recognized that time bars in suits between private parties are presumptively subject to equitable tolling. See id., at 95–96. That means a court usually may pause the running of a limitations statute in private litigation when a party “has pursued his rights diligently but some extraordinary circumstance” prevents him from meeting a deadline. Lozano v. Montoya Alvarez,572 U. S. 1, ___ (2014) (slip op., at 7). We held in Irwin that “the same rebuttable presumption of equitable tolling” should also apply to suits brought against the United States under a statute waiving sovereign immunity. 498 U. S., at 95–96. Our old “ad hoc,” law-by-law approach to determining the availability of tolling in those suits, we reasoned, had produced inconsistency and “unpredictability” without the offsetting virtue of enhanced “fidelity to the intent of Congress.” Id., at 95. Adopting the “general rule” used in private litigation, we stated, would “amount[ ] to little, if any, broadening” of a statutory waiver of immunity. Ibid. Accordingly, we thought such a presumption “likely to be a realistic assessment of legislative intent as well as a practically useful” rule of interpretation. Ibid. A rebuttable presumption, of course, may be rebutted, so Irwin does not end the matter. When enacting a time bar for a suit against the Government (as for one against a private party), Congress may reverse the usual rule if it chooses. See id., at 96. The Government may therefore attempt to establish, through evidence relating to a particular statute of limitations, that Congress opted to forbid equitable tolling. One way to meet that burden—and the way the Government pursues here—is to show that Congress made the time bar at issue jurisdictional.[2] When that is so, a litigant’s failure to comply with the bar deprives a court of all authority to hear a case. Hence, a court must enforce the limitation even if the other party has waived any timeliness objection. See Gonzalez v. Thaler, 565 U. S. ___, ___–___ (2012) (slip op., at 5–6). And, more crucially here, a court must do so even if equitable considerations would support extending the prescribed time period. See John R. Sand & Gravel Co. v. United States,552 U. S. 130–134 (2008).[3] Given those harsh consequences, the Government must clear a high bar to establish that a statute of limitations is jurisdictional. In recent years, we have repeatedly held that procedural rules, including time bars, cabin a court’s power only if Congress has “clearly state[d]” as much. Sebelius v. Auburn Regional Medical Center, 568 U. S. ___, ___ (2013) (slip op., at 6) (quoting Arbaugh v. Y & H Corp.,546 U. S. 500,515 (2006)); see Gonzalez, 565 U. S., at ___–___ (slip op., at 6–7). “[A]bsent such a clear statement, . . . ‘courts should treat the restriction as nonjurisdictional.’ ” Auburn Regional, 568 U. S., at ___–___ (slip op., at 6–7) (quoting Arbaugh, 546 U. S., at 516). That does not mean “Congress must incant magic words.” Auburn Regional, 568 U. S., at ___ (slip op., at 7). But traditional tools of statutory construction must plainly show that Congress imbued a procedural bar with jurisdictional consequences. And in applying that clear statement rule, we have made plain that most time bars are nonjurisdictional. See, e.g., id., at ___ (slip op., at 8) (noting the rarity of jurisdictional time limits). Time and again, we have described filing deadlines as “quintessential claim-processing rules,” which “seek to promote the orderly progress of litigation,” but do not deprive a court of authority to hear a case. Henderson v. Shinseki,562 U. S. 428,435 (2011); see Auburn Regional, 568 U. S., at ___ (slip op., at 8); Scarborough v. Principi,541 U. S. 401,413 (2004). That is so, contrary to the dissent’s suggestion, see post, at 4, 10–11, even when the time limit is important (most are) and even when it is framed in mandatory terms (again, most are); indeed, that is so “however emphatic[ally]” expressed those terms may be, Henderson, 562 U. S., at 439 (quoting Union Pacific R. Co. v. Locomotive Engineers,558 U. S. 67,81 (2009)). Congress must do something special, beyond setting an exception-free deadline, to tag a statute of limitations as jurisdictional and so prohibit a court from tolling it. In enacting the FTCA, Congress did nothing of that kind. It provided no clear statement indicating that §2401(b) is the rare statute of limitations that can deprive a court of jurisdiction. Neither the text nor the context nor the legislative history indicates (much less does so plainly) that Congress meant to enact something other than a standard time bar. Most important, §2401(b)’s text speaks only to a claim’s timeliness, not to a court’s power. It states that “[a] tort claim against the United States shall be forever barred unless it is presented [to the agency] within two years . . . or unless action is begun within six months” of the agency’s denial of the claim. That is mundane statute-of-limitations language, saying only what every time bar,by definition, must: that after a certain time a claimis barred. See infra, at 11, n. 7 (citing many similarly worded limitations statutes). The language is mandatory—“shall” be barred—but (as just noted) that is true of most such statutes, and we have consistently found it of no consequence. See, e.g., Gonzalez, 565 U. S., at ___–___ (slip op., at 10–11). Too, the language might be viewed as emphatic—“forever” barred—but (again) we have often held that not to matter. See, e.g., Henderson, 562 U. S., at 439; Union Pacific, 558 U. S., at 81. What matters instead is that §2401(b) “does not speak in jurisdictional terms or refer in any way to the jurisdiction of the district courts.” Arbaugh, 546 U. S., at 515 (quoting Zipes v. Trans World Airlines, Inc.,455 U. S. 385,394 (1982)). It does not define a federal court’s jurisdiction over tort claims generally, address its authority to hear untimely suits, or in any way cabin its usual equitable powers. Section 2401(b), in short, “reads like an ordinary, run-of-the-mill statute of limitations,” spelling out a litigant’s filing obligations without restricting a court’s authority. Holland v. Flor-ida,560 U. S. 631,647 (2010).[4] Statutory context confirms that reading. This Court has often explained that Congress’s separation of a filing deadline from a jurisdictional grant indicates that the time bar is not jurisdictional. See Henderson, 562 U. S., at 439–440; Reed Elsevier, Inc. v. Muchnick,559 U. S. 154–165 (2010); Arbaugh, 546 U. S., at 515; Zipes, 455 U. S., at 393–394. So too here. Whereas §2401(b) houses the FTCA’s time limitations, a different section of Title 28 confers power on federal district courts to hear FTCA claims. See §1346(b)(1) (“district courts . . . shall have exclusive jurisdiction” over tort claims against the United States). Nothing conditions the jurisdictional grant on the limitations periods, or otherwise links those separate provisions. Treating §2401(b)’s time bars as jurisdictional would thus disregard the structural divide built into the statute. Finally, even assuming legislative history alone could provide a clear statement (which we doubt), none does so here. The report accompanying the FTCA did not discuss whether §2401(b)’s time limits are jurisdictional. See S. Rep. No. 1400, 79th Cong., 2d Sess., 33 (1946). And in amending §2401(b) four times after its enactment, Congress declined again (four times over) to say anything specific about whether the statute of limitations imposes a jurisdictional bar. Congress thus failed to provide anything like the clear statement this Court has demanded before deeming a statute of limitations to curtail a court’s power. And so we wind up back where we started, with Irwin’s “general rule” that equitable tolling is available in suits against the Government. 498 U. S., at 95. The justification the Government offers for departing from that principle fails: Section 2401(b) is not a jurisdictional requirement. The time limits in the FTCA are just time limits, nothing more. Even though they govern litigation against the Government, a court can toll them on equitable grounds.III The Government balks at that straightforward analysis, claiming that it overlooks two reasons for thinking §2401(b) jurisdictional. But neither of those reasons is persuasive. Indeed, our precedents in this area foreclose them both.A The Government principally contends that §2401(b) is jurisdictional because it includes the same language as the statute of limitations governing contract (and some other non-tort) suits brought against the United States under the Tucker Act. See §2501.[5] That statute long provided that such suits “shall be forever barred” if not filed within six years. Act of Mar. 3, 1863, §10,12Stat.767; see Act of Mar. 3, 1911, §156,36Stat.1139.[6] And this Court repeatedly held that 6-year limit to be jurisdictional and thus not subject to equitable tolling. See Kendall v. United States,107 U. S. 123–126 (1883); Finn v. United States,123 U. S. 227,232 (1887); Soriano v. United States,352 U. S. 270–274 (1957). When Congress drafted the FTCA’s time bar, it used the same “shall be forever barred” language (though selecting a shorter limitations period). “In these circumstances,” the Government maintains, “the only reasonable conclusion is that Congress intended the FTCA’s identically worded time limit to be a jurisdictional bar.” Brief for United States in Wong 21–22. According to the Government, Congress wanted the FTCA to serve as “a tort-law analogue to the Tucker Act” and incorporated the words “shall be forever barred” to similarly preclude equitable tolling. Reply Brief in Wong 4. (The dissent relies heavily on the same argument. See post, at 4–8.) But the Government takes too much from Congress’s use in §2401(b) of an utterly unremarkable phrase. The “shall be forever barred” formulation was a commonplace in federal limitations statutes for many decades surrounding Congress’s enactment of the FTCA.[7] And neither this Court nor any other has accorded those words talismanic power to render time bars jurisdictional. To the contrary, we have construed the very same “shall be forever barred” language in15 U. S. C. §15b, the Clayton Act’s statute of limitations, to be subject to tolling; nothing in that provision, we found, “restrict[s] the power of the federal courts” to extend a limitations period when circumstances warrant. American Pipe & Constr. Co. v. Utah,414 U. S. 538,559 (1974); see Hardin v. City Title & Escrow Co., 797 F. 2d 1037, 1040 (CADC 1986) (calling §15(b) “a good example of a non-jurisdictional time limitation” based on its text and separation from the Clayton Act’s jurisdic-tional provisions).[8] As the Government itself has previously acknowledged, referring to the “shall be forever barred” locution: “[T]hat type of language has more to do with the legal rhetoric at the time the statute was passed” than with anything else, and should not “make[ ] a difference” to the jurisdictional analysis. Tr. of Oral Arg. in Irwin, O. T. 1990, No. 89–5867, p. 30. Or, put just a bit differ-ently: Congress’s inclusion of a phrase endemic to limitations statutes of that era, at least some of which allow tolling, cannot provide the requisite clear statement that a time bar curtails a court’s authority. Indeed, in two decisions directly addressing the Tucker Act’s statute of limitations, this Court dismissed the idea that the language the Government relies on here has jurisdictional significance. Twice we described the words in that provision as not meaningfully different from those in a nonjurisdictional statute of limitations. And twice we made clear that the jurisdictional status of the Tucker Act’s time bar has precious little to do with its phrasing. We first did so in Irwin. Using our newly minted presumption, see supra, at 4–5, we decided there that the limitations period governing Title VII suits against the Government,42 U. S. C. §2000e–16(c) (1988 ed.), allowed equitable tolling. In reaching that conclusion, we compared §2000e–16(c)’s text (then stating that an employee “may file a civil action” within 30 days of an agency’s denial of her claim) with the language of the Tucker Act’s time bar. We noted that we had formerly held the Tucker Act’s limitations statute to “jurisdictionally bar[ ]” late claims, and we acknowledged the possibility of justifying that different treatment by characterizing its “language [as] more stringent than” §2000e–16(c)’s. Irwin, 498 U. S., at 94–95. But we rejected that reasoning, instead finding that the two formulations were materially alike. “[W]e are not persuaded,” we stated, “that the difference between them is enough to manifest a different congressional intent with respect to the availability of equitable tolling.” Id., at 95. Leaving for another day the question of what did account for the jurisdictional status of the Tucker Act’s time bar, the Court thus ruled out reliance on its language. In other words, on the core question the Government raises here—whether the phrase “shall be forever barred,” as used in both the Tucker Act and the FTCA, manifests a congressional decision to preclude tolling—Irwin said no. More recently, John R. Sand reaffirmed that conclusion, even as it refused to overturn our century-old view that the Tucker Act’s time bar is jurisdictional. No less than three times, John R. Sand approvingly repeated Irwin’s statement that the textual differences between the Tucker Act’s time bar and §2000e–16(c) were insignificant—i.e., that the language of the two provisions could not explain why the former was jurisdictional and the latter not. See 552 U. S., at 137, 139 (calling the provisions “linguistically similar,” “similar . . . in language,” and “similarly worded”). But if that were so, John R. Sand asked, why not hold that the Tucker Act’s time limit, like §2000e–16(c), is nonjurisdictional? The answer came down to two words: stare decisis. The Tucker Act’s bar was different because it had been the subject of “a definitive earlier interpretation.” Id., at 138; see id., at 137; supra, at 10. And for that reason alone, John R. Sand left in place our prior construction of the Tucker Act’s time limit. See 552 U. S., at 139 (observing, in Justice Brandeis’s words, that “it is more important that” the rule “be settled than that it be settled right” (quoting Burnet v. Coronado Oil & Gas Co.,285 U. S. 393,406 (1932) (dissenting opinion))). What is special about the Tucker Act’s deadline, John R. Sand recognized, comes merely from this Court’s prior rulings, not from Congress’s choice of wording. The Government thus cannot show that the phrase “shall be forever barred” in §2401(b) plainly signifies a jurisdictional statute, as our decisions require. See supra, at 6–7. Unlike in John R. Sand, here stare decisis plays no role: We have not previously considered whether §2401(b) restricts a court’s authority. What we have done is to say, again and again, that the core language in that provision has no jurisdictional significance. It is materi-ally indistinguishable from the language in one nonjurisdic-tional time bar (i.e., §2000e–16(c)). See Irwin, 498 U. S., at 95; John R. Sand, 552 U. S., at 137, 139. And it is identical to the language in another (i.e., 15 U. S. C. §15b). See American Pipe, 414 U. S., at 559. Yes, we have held that the Tucker Act’s time bar, which includes those same words, constrains a court’s power to hear late claims. But as we explained in Irwin, that is not because the phrase itself “manifest[s] a . . . congressional intent with respect to the availability of equitable tolling.” 498 U. S., at 95. The words on which the Government pins its hopes are just the words of a limitations statute of a particular era. And nothing else supports the Government’s claim that Congress, when enacting the FTCA, wanted to incorporate this Court’s view of the Tucker Act’s time bar—much less that Congress expressed that purported intent with the needed clear statement.B The Government next contends that at the time of the FTCA’s enactment, Congress thought that every limitations statute applying to suits against the United States, however framed or worded, cut off a court’s jurisdiction over untimely claims. On that view, the particular language of those statutes makes no difference. All that matters is that such time limits function as conditions on the Government’s waiver of sovereign immunity. In that era—indeed, up until Irwin was decided—those conditions were generally supposed to be “strictly observed.” Soriano, 352 U. S., at 276. That meant, the Government urges, that all time limits on actions against the United States “carr[ied] jurisdictional consequences.” Brief for United States in Wong 34. Accordingly, the Government concludes, Congress “would have expected courts to apply [§2401(b)] as a jurisdictional requirement—just as conditions on waivers of sovereign immunity had always been applied.” Id., at 32. Irwin, however, forecloses that argument. After all, Irwin also considered a pre-Irwin time bar attached to a waiver of sovereign immunity. The Government argued there—anticipating its claim here—that because §2000e–16(c)’s statute of limitations conditioned such a waiver, it must be jurisdictional and not subject to equitable tolling. See Brief for Respondents 6, 10, 14, 19, and Tr. of Oral Arg. 31–37, in Irwin, O. T. 1990, No. 89–5867. But Irwin disagreed, applying the opposite presumption to a time limit passed two decades earlier. See 498 U. S., at 94–96; supra, at 4–5. Justice White protested, much as the Government does now, that at the time of §2000e–16(c)’s enactment, limitations statutes for suits against the Government were “strictly observed” and not amenable to tolling. 498 U. S., at 97 (opinion concurring in part and concurring in judgment) (quoting Soriano, 352 U. S., at 276); see 498 U. S., at 99, n. 2. How could an earlier Congress, Justice White asked, have “had in mind the Court’s present departure from that longstanding rule”? Ibid.; see post, at 9 (asking a variant of the same question). But the Irwin Court was undeterred. The Court noted that it had not applied the former rule so consistently as Justice White suggested. See 498 U. S., at 94. And the Court doubted that the former approach so well reflected congressional intent: On the contrary, because equitable tolling “amounts to little, if any, broadening of the congressional waiver,” we thought that a rule generally allowing tolling is the more “realistic assessment of legislative intent.” Id., at 95; see supra, at 4–5. For those reasons, the Court declined to count time bars as jurisdictional merely because they condition waivers of immunity—even if Congress enacted the deadline when the Court interpreted limitations statutes differently. In the years since, this Court has repeatedly followed Irwin’s lead. We have applied Irwin to pre-Irwin statutes, just as we have to statutes that followed in that decision’s wake. See Scarborough, 541 U. S., at 420–422; Franconia Associates v. United States,536 U. S. 129,145 (2002). To be sure, Irwin’s presumption is rebuttable. But the rebuttal cannot rely on what Irwin itself deemed irrelevant—that Congress passed the statute in an earlier era, when this Court often attached jurisdictional consequence to conditions on waivers of sovereign immunity. Rather, the rebuttal must identify something distinctive about the time limit at issue, whether enacted then or later—a reason for thinking Congress wanted that limitations statute (not all statutes passed in an earlier day) to curtail a court’s jurisdiction. On the Government’s contrary view, Irwin would effectively become only a prospective decision. Nothing could be less consonant with Irwin’s ambition to adopt a “general rule to govern the applicability of equit-able tolling in suits against the Government.” 498 U. S., at 95. And the Government’s claim is peculiarly inapt as applied to §2401(b) because all that is special about the FTCA cuts in favor of allowing equitable tolling. As compared to other waivers of immunity (prominently including the Tucker Act), the FTCA treats the United States more like a commoner than like the Crown. The FTCA’s jurisdictional provision states that courts may hear suits “under circumstances where the United States, if a private person, would be liable to the claimant.”28 U. S. C. §1346(b). And when defining substantive liability for torts, the Act reiterates that the United States is accountable “in the same manner and to the same extent as a private individual.” §2674. In keeping with those provisions, this Court has often rejected the Government’s calls to cabin the FTCA on the ground that it waives sovereign immunity—and indeed, the Court did so in the years immediately after the Act’s passage, even as it was construing other waivers of immunity narrowly. See, e.g., United States v. Aetna Casualty & Surety Co.,338 U. S. 366,383 (1949); Indian Towing Co. v. United States,350 U. S. 61,65 (1955); Rayonier Inc. v. United States,352 U. S. 315–320 (1957). There is no reason to do differently here. As Irwin recognized, treating the Government like a private person means (among other things) permitting equitable tolling. See 498 U. S., at 95–96. So in stressing the Government’s equivalence to a private party, the FTCA goes further than the typical statute waiving sovereign immunity to indicate that its time bar allows a court to hear late claims.IV Our precedents make this a clear-cut case. Irwin requires an affirmative indication from Congress that it intends to preclude equitable tolling in a suit against the Government. See 498 U. S., at 95–96. Congress can provide that signal by making a statute of limitations jurisdictional. But that requires its own plain statement; otherwise, we treat a time bar as a mere claims-processing rule. See Auburn Regional, 568 U. S., at ___, ___ (slip op., at 6, 8). Congress has supplied no such statement here. As this Court has repeatedly stated, nothing about §2401(b)’s core language is special; “shall be forever barred” is an ordinary (albeit old-fashioned) way of setting a deadline, which does not preclude tolling when circumstances warrant. See Irwin, 498 U. S., at 95–96; John R. Sand, 552 U. S., at 137, 139; American Pipe, 414 U. S., at 558–559. And it makes no difference that a time bar conditions a waiver of sovereign immunity, even if Congress enacted the measure when different interpretive conventions applied; that is the very point of this Court’s decision to treat time bars in suits against the Government, whenever passed, the same as in litigation between private parties. See Irwin, 498 U. S., at 95–96; Scarborough, 541 U. S., at 420–422; Franconia, 536 U. S., at 145. Accordingly, we hold that the FTCA’s time bars are nonjurisdictional and subject to equitable tolling. We affirm the judgments of the U. S. Court of Appeals for the Ninth Circuit and remand the cases for further proceedings consistent with this opinion. On remand in June, it is for the District Court to decide whether, on the facts of her case, June is entitled to equitable tolling.It is so ordered.Notes1 Although we did not consolidate these cases, we address them together because everyone agrees that the core arguments for and against equitable tolling apply equally to both of §2401(b)’s deadlines. See, e.g., Brief for United States in June 15 (“Nothing in the text or relevant legislative history . . . suggests that the respective time bars should be interpreted differently with respect to whether they are jurisdictional or subject to equitable tolling”).2 The Government notes, and we agree, that Congress may preclude equitable tolling of even a nonjurisdictional statute of limitations. See Brief for United States in Wong 20; Sebelius v. Auburn Regional Medical Center, 568 U. S. ___, ___–___ (2013) (slip op., at 6–8, 10–11) (finding a nonjurisdictional time limit not amenable to tolling). And the Government contends in passing that even if §2401(b) is nonjurisdictional, it prohibits equitable tolling. See Brief for United States in Wong 20. But the Government makes no independent arguments in support of that position; instead, it relies (and even then implicitly) on the same indicia of congressional intent that, in its view, show that §2401(b)’s time limits are jurisdictional. See infra, at 9–10, 14–15. In addressing the Government’s predominant, jurisdictional claim, we therefore also deal with its subsidiary one.3 The dissent takes issue with the sequence in which we decide the jurisdictional question, contending that we must do so prior to mentioning Irwin’s presumption. See post, at 11–12 (opinion of Alito, J.). We do not understand the point—or more precisely, why the dissent thinks the ordering matters. When Congress makes a time bar in a suit against the Government jurisdictional, one could say (as the dissent does) that Irwin does not apply, or one could say (as we do) that Irwin’s presumption is conclusively rebutted. The bottom line is the same: Tolling is not available. We frame the inquiry as we do in part because that is how the Government presented the issue. See Brief for United States in Wong 19 (“One way to show that [Irwin’s presumption is rebutted] is to establish that the statutory time limit is a ‘jurisdictional’ restriction”). And we think that choice makes especially good sense in these cases because various aspects of Irwin’s reasoning are central to considering the parties’ positions on whether §2401(b) is jurisdictional. See infra, at 12–17.4 The dissent argues that nonjurisdictional time limits typically mention claimants, whereas §2401(b) does not. See post, at 10. But none of our precedents have either said or suggested that such a difference matters—that, for example, a statute barring a “tort claim” is jurisdictional, but one barring a “person’s tort claim” is not. See, e.g., Zipes, 455 U. S., at 394, and n. 10 (concluding that a time limit did “not speak in jurisdictional terms” even though it did not refer to a claimant). Rather, in case after case, we have emphasized another distinction—that jurisdictional statutes speak about jurisdiction, or more generally phrased, about a court’s powers. See Auburn Regional, 568 U. S., at ___ (slip op., at 7); Reed Elsevier, Inc. v. Muchnick,559 U. S. 154–161 (2010); Arbaugh, 546 U. S., at 515.5 The Tucker Act of 1887, ch. 359,24Stat.505, enlarged the Court of Claims’ jurisdiction over contract and other non-tort actions against the Government. The statute of limitations applying to such suits pre-dated the Tucker Act by more than two decades.6 During a recodification occurring in 1948 (two years after passage of the FTCA), Congress omitted the word “forever” from the Tucker Act’s statute of limitations; since then, it has provided simply that untimely claims “shall be barred.”28 U. S. C. §2501; see §2501,62Stat.976. No party contends that change makes any difference to the resolution of these cases.7 See, e.g., §6 of the Portal-to-Portal Act of 1947,61Stat.87,29 U. S. C. §255 (1952 ed.); §3 of the Automobile Dealers’ Day in Court Act,70Stat.1125,15 U. S. C. §1223 (1958 ed.); §111(b) of the National Traffic and Motor Vehicle Safety Act of 1966,80Stat.725,15 U. S. C. §1400(b) (1970 ed.); §7(e) of the Age Discrimination in Employment Act of 1967 (ADEA),81Stat.605,29 U. S. C. §626(e) (1970 ed.); §6(c) of the Agricultural Fair Practices Act of 1967,82Stat.95,7 U. S. C. §2305(c) (1970 ed.); §613(b) of the National Manufactured Housing Construction and Safety Standards Act of 1974,88Stat.707,42 U. S. C. §5412(b) (1976 ed.).8 Even before this Court’s decision in American Pipe, Courts of Appeals had unanimously construed the Clayton Act’s statute of limitations to allow equitable tolling. See General Elec. Co. v. San Antonio, 334 F. 2d 480, 484–485 (CA5 1964) (joining six other Circuits in reaching that conclusion). Similarly, every Court of Appeals to have considered the issue has found that §6 of the Portal-to-Portal Act, which contains the same “shall be forever barred” phrase, permits hearing late claims. See, e.g., Hodgson v. Humphries, 454 F. 2d 1279, 1283–1284 (CA10 1972); Ott v. Midland-Ross Corp., 523 F. 2d 1367, 1370 (CA6 1975); Partlow v. Jewish Orphans’ Home of Southern Cal., Inc., 645 F. 2d 757, 760–761 (CA9 1981), abrogated on other grounds by Hoffmann-La Roche Inc. v. Sperling,493 U. S. 165 (1989). And so too Courts of Appeals unanimously found that the ADEA’s longtime (though not current) time bar containing that language was subject to tolling. See, e.g., Vance v. Whirlpool Corp., 707 F. 2d 483, 489 (CA4 1983); Callowhill v. Allen-Sherman-Hoff Co., 832 F. 2d 269, 273–274 (CA3 1987).

576.US.2014_14-144

Texas offers automobile owners a choice between general-issue and specialty license plates. Those who want the State to issue a particular specialty plate may propose a plate design, comprising a slogan, a graphic, or both. If the Texas Department of Motor Vehicles Board approves the design, the State will make it available for display on vehicles registered in Texas. Here, the Texas Division of the Sons of Confederate Veterans and its officers (collectively SCV) filed suit against the Chairman and members of the Board (collectively Board), arguing that the Board’s rejection of SCV’s proposal for a specialty plate design featuring a Confederate battle flag violated the Free Speech Clause. The District Court entered judgment for the Board, but the Fifth Circuit reversed, holding that Texas’s specialty license plate designs are private speech and that the Board engaged in constitutionally forbidden viewpoint discrimination when it refused to approve SCV’s design. Held: Texas’s specialty license plate designs constitute government speech, and thus Texas was entitled to refuse to issue plates featuring SCV’s proposed design. Pp. 5–18. (a) When government speaks, it is not barred by the Free Speech Clause from determining the content of what it says. Pleasant Grove City v. Summum, 555 U. S. 460 –468. A government is generally entitled to promote a program, espouse a policy, or take a position. Were the Free Speech Clause interpreted otherwise, “it is not easy to imagine how government would function.” Id., at 468. That is not to say that a government’s ability to express itself is without restriction. Constitutional and statutory provisions outside of the Free Speech Clause may limit government speech, and the Free Speech Clause itself may constrain the government’s speech if, for example, the government seeks to compel private persons to convey the government’s speech. Pp. 5–6. (b) This Court’s precedents regarding government speech provide the appropriate framework through which to approach the case. Pp. 6–17. (1) The same analysis the Court used in Summum—to conclude that a city “accepting a privately donated monument and placing it on city property” was engaging in government speech, 555 U. S., at 464—leads to the conclusion that government speech is at issue here. First, history shows that States, including Texas, have long used license plates to convey government speech, e.g., slogans urging action, promoting tourism, and touting local industries. Cf. id., at 470. Second, Texas license plate designs “are often closely identified in the public mind with the [State].” Id., at 472. Each plate is a government article serving the governmental purposes of vehicle registration and identification. The governmental nature of the plates is clear from their faces: the State places the name “TEXAS” in large letters across the top of every plate. Texas also requires Texas vehicle owners to display license plates, issues every Texas plate, and owns all of the designs on its plates. The plates are, essentially, government IDs, and ID issuers “typically do not permit” their IDs to contain “message[s] with which they do not wish to be associated,” id., at 471. Third, Texas maintains direct control over the messages conveyed on its specialty plates, by giving the Board final approval over each design. Like the city government in Summum, Texas “has effectively controlled the messages [conveyed] by exercising final approval authority over their selection.” Id., at 473. These considerations, taken together, show that Texas’s specialty plates are similar enough to the monuments in Summum to call for the same result. Pp. 7–12. (2) Forum analysis, which applies to government restrictions on purely private speech occurring on government property, Cornelius v. NAACP Legal Defense & Ed. Fund, Inc., 473 U. S. 788 , is not appropriate when the State is speaking on its own behalf. The parties agree that Texas’s specialty license plates are not a traditional public forum. Further, Texas’s policies and the nature of its license plates indicate that the State did not intend its specialty plates to serve as either a designated public forum—where “government property . . . not traditionally . . . a public forum is intentionally opened up for that purpose,” Summum, supra, at 469—or a limited public forum—where a government “reserv[es a forum] for certain groups or for the discussion of certain topics,” Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819 . The State exercises final authority over the messages that may be conveyed by its specialty plates, it takes ownership of each specialty plate design, and it has traditionally used its plates for government speech. These features of Texas specialty plates militate against a determination that Texas has created a public forum. Finally, the plates are not a nonpublic forum, where the “government is . . . a proprietor, managing its internal operations.” International Soc. for Krishna Consciousness, Inc. v. Lee, 505 U. S. 672 –679. The fact that private parties take part in the design and propagation of a message does not extinguish the governmental nature of the message or transform the government’s role into that of a mere forum provider. See Summum, supra, at 470–471. Nor does Texas’s requirement that vehicle owners pay annual fees for specialty plates mean that the plates are a forum for private speech. And this case does not resemble other nonpublic forum cases. Perry Ed. Assn. v. Perry Local Educators’ Assn., 460 U. S. 37 –49; Lehman v. Shaker Heights, 418 U. S. 298 ; and Cornelius, supra, at 804–806, distinguished. Pp. 13–17. (c) The determination that Texas’s specialty license plate designs are government speech does not mean that the designs do not also implicate the free speech rights of private persons. The Court has acknowledged that drivers who display a State’s selected license plate designs convey the messages communicated through those designs. See Wooley v. Maynard, 430 U. S. 705 , n. 15. The Court has also recognized that the First Amendment stringently limits a State’s authority to compel a private party to express a view with which the private party disagrees. Just as Texas cannot require SCV to convey “the State’s ideological message,” id., at 715, SCV cannot force Texas to include a Confederate battle flag on its specialty license plates. Pp. 17–18. 759 F. 3d 388, reversed. Breyer, J., delivered the opinion of the Court, in which Thomas, Ginsburg, Sotomayor, and Kagan, JJ., joined. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Scalia and Kennedy, JJ., joined.

Texas offers automobile owners a choice between ordinary and specialty license plates. Those who want the State to issue a particular specialty plate may propose a plate design, comprising a slogan, a graphic, or (most commonly) both. If the Texas Department of Motor Vehicles Board approves the design, the State will make it available for display on vehicles registered in Texas. In this case, the Texas Division of the Sons of Confederate Veterans proposed a specialty license plate design featuring a Confederate battle flag. The Board rejected the proposal. We must decide whether that rejection violated the Constitution’s free speech guarantees. See Amdts. 1, 14. We conclude that it did not. I A Texas law requires all motor vehicles operating on the State’s roads to display valid license plates. See Tex. Transp. Code Ann. §§502.001 (West Supp. 2014), 504.001 (2013), 504.943 (Supp. 2014). And Texas makes available several kinds of plates. Drivers may choose to display the State’s general-issue license plates. See Texas Dept. of Motor Vehicles, Motor Vehicle Registration Manual 9.1 (Apr. 2015). Each of these plates contains the word“Texas,” a license plate number, a silhouette of theState, a graphic of the Lone Star, and the slogan“The Lone Star State.” Texas Dept. of Motor Vehicles, The Texas Classic FAQs (July 16, 2012), online at http://www.txdmv.gov/motorists/license-plates (all Internet materials as visited June 16, 2015, and available in Clerk of Court’s case file). In the alternative, drivers may choose from an assortment of specialty license plates. §504.008(b) (West 2013). Each of these plates contains the word “Texas,” a license plate number, and one of a selection of designs prepared by the State. See ibid.; Specialty License Plates, http://www.txdmv.gov/motorists/license-plates/specialty-license-plates (displaying available Texas specialty plates); Create a Plate: Your Design, http://www.myplates.com/BackgroundOnly (same). Finally, Texas law provides for personalized plates (also knownas vanity plates). 43 Tex. Admin. Code §217.45(c)(7) (2015). Pursuant to the personalization program, a vehicle owner may request a particular alphanumeric pattern for use as a plate number, such as “BOB” or “TEXPL8.” Here we are concerned only with the second category of plates, namely specialty license plates, not with the personalization program. Texas offers vehicle owners a va-riety of specialty plates, generally for an annual fee. See §217.45(b)(2). And Texas selects the designs for specialty plates through three distinct processes. First, the state legislature may specifically call for the development of a specialty license plate. See Tex. Transp. Code §§504.602–504.663 (West 2013 and Supp. 2014). The legislature has enacted statutes authorizing, for example, plates that say “Keep Texas Beautiful” and “Mothers Against Drunk Driving,” plates that “honor” the Texas citrus industry, and plates that feature an image of the World Trade Center towers and the words “Fight Terrorism.” See §§504.602, 504.608, 504.626, 504.647. Second, the Board may approve a specialty plate design proposal that a state-designated private vendor hascreated at the request of an individual or organization. See §§504.6011(a), 504.851(a); 43 Tex. Admin. Code §217.52(b). Among the plates created through the private-vendor process are plates promoting the “Keller Indians” and plates with the slogan “Get it Sold with RE/MAX.” Third, the Board “may create new specialty license plates on its own initiative or on receipt of an application from a” nonprofit entity seeking to sponsor a specialty plate. Tex. Transp. Code Ann. §§504.801(a), (b). A nonprofit must include in its application “a draft design of the specialty license plate.” 43 Tex. Admin. Code §217.45(i)(2)(C). And Texas law vests in the Board authority to approve or to disapprove an application. See §217.45(i)(7). The relevant statute says that the Board “may refuse to create a new specialty license plate” for a number of reasons, for example “if the design might be offensive to any member of the public . . . or for any other reason established by rule.” Tex. Transp. Code Ann. §504.801(c). Specialty plates that the Board has sanctioned through this process include plates featuring the words “The Gator Nation,” together with the Florida Gators logo, and plates featuring the logo of Rotary International and the words “SERVICE ABOVE SELF.” B In 2009, the Sons of Confederate Veterans, Texas Division (a nonprofit entity), applied to sponsor a specialty license plate through this last-mentioned process. SCV’s application included a draft plate design. See Appendix, infra. At the bottom of the proposed plate were the words “SONS OF CONFEDERATE VETERANS.” At the side was the organization’s logo, a square Confederate battle flag framed by the words “Sons of Confederate Veterans 1896.” A faint Confederate battle flag appeared in the back-ground on the lower portion of the plate. Additionally,in the middle of the plate was the license plate number, and at the top was the State’s name and silhouette. The Board’s predecessor denied this application. In 2010, SCV renewed its application before the Board. The Board invited public comment on its website and at an open meeting. After considering the responses, including a number of letters sent by elected officials who opposed the proposal, the Board voted unanimously against issuing the plate. The Board explained that it had found “it necessary to deny th[e] plate design application, specifically the confederate flag portion of the design, because public comments ha[d] shown that many members of the general public find the design offensive, and because such comments are reasonable.” App. 64. The Board added “that a significant portion of the public associate the confederate flag with organizations advocating expressions of hate directed toward people or groups that is demeaning to those people or groups.” Id., at 65. In 2012, SCV and two of its officers (collectively SCV) brought this lawsuit against the chairman and members of the Board (collectively Board). SCV argued that the Board’s decision violated the Free Speech Clause of the First Amendment, and it sought an injunction requiring the Board to approve the proposed plate design. The District Court entered judgment for the Board. A divided panel of the Court of Appeals for the Fifth Circuit reversed. Texas Div., Sons of Confederate Veterans, Inc., v. Vandergriff, 759 F. 3d 388 (2014). It held that Texas’s specialty license plate designs are private speech and that the Board, in refusing to approve SCV’s design, engaged in constitutionally forbidden viewpoint discrimination. The dissenting judge argued that Texas’s specialty license plate designs are government speech, the content of which the State is free to control. We granted the Board’s petition for certiorari, and we now reverse. II When government speaks, it is not barred by the Free Speech Clause from determining the content of what it says. Pleasant Grove City v. Summum, 555 U. S. 460 –468 (2009). That freedom in part reflects the fact that it is the democratic electoral process that first and foremost provides a check on government speech. See Board of Regents of Univ. of Wis. System v. Southworth, 529 U. S. 217, 235 (2000) . Thus, government statements (and government actions and programs that take the form of speech) do not normally trigger the First Amendment rules designed to protect the marketplace of ideas. See Johanns v. Livestock Marketing Assn., 544 U. S. 550, 559 (2005) . Instead, the Free Speech Clause helps produce informed opinions among members of the public, who are then able to influence the choices of a government that, through words and deeds, will reflect its electoral mandate. See Stromberg v. California, 283 U. S. 359, 369 (1931) (observing that “our constitutional system” seeks to maintain “the opportunity for free political discussion to the end that government may be responsive to the will of the people”). Were the Free Speech Clause interpreted otherwise, government would not work. How could a city government create a successful recycling program if officials, when writing householders asking them to recycle cans and bottles, had to include in the letter a long plea from the local trash disposal enterprise demanding the contrary? How could a state government effectively develop programs designed to encourage and provide vaccinations, if officials also had to voice the perspective of those who oppose this type of immunization? “[I]t is not easy to imagine how government could function if it lacked th[e] freedom” to select the messages it wishes to convey. Summum, supra, at 468. We have therefore refused “[t]o hold that the Government unconstitutionally discriminates on the basis of viewpoint when it chooses to fund a program dedicated to advance certain permissible goals, because the program in advancing those goals necessarily discourages alternative goals.” Rust v. Sullivan, 500 U. S. 173, 194 (1991) . We have pointed out that a contrary holding “would render numerous Government programs constitutionally suspect.” Ibid. Cf. Keller v. State Bar of Cal., 496 U. S. 1 –13 (1990) (“If every citizen were to have a right to insist that no one paid by public funds express a view with which he disagreed, debate over issues of great concern to the public would be limited to those in the private sector, and the process of government as we know it radically transformed”). And we have made clear that “the government can speak for itself.” Southworth, supra, at 229. That is not to say that a government’s ability to express itself is without restriction. Constitutional and statutory provisions outside of the Free Speech Clause may limit government speech. Summum, supra, at 468. And the Free Speech Clause itself may constrain the government’s speech if, for example, the government seeks to compel private persons to convey the government’s speech. But, as a general matter, when the government speaks it is entitled to promote a program, to espouse a policy, or to take a position. In doing so, it represents its citizens and it carries out its duties on their behalf. III In our view, specialty license plates issued pursuant to Texas’s statutory scheme convey government speech. Our reasoning rests primarily on our analysis in Summum, a recent case that presented a similar problem. We conclude here, as we did there, that our precedents regarding government speech (and not our precedents regarding forums for private speech) provide the appropriate framework through which to approach the case. See 555 U. S., at 464. A In Summum, we considered a religious organization’s request to erect in a 2.5-acre city park a monument setting forth the organization’s religious tenets. See id., at 464–465. In the park were 15 other permanent displays. Id., at 464. At least 11 of these—including a wishing well, a September 11 monument, a historic granary, the city’s first fire station, and a Ten Commandments monument—had been donated to the city by private entities. Id., at 464–465. The religious organization argued that the Free Speech Clause required the city to display the organization’s proposed monument because, by accepting a broad range of permanent exhibitions at the park, the city had created a forum for private speech in the form of monuments. Brief for Respondent in Pleasant Grove City v. Summum, O. T. 2008, No. 07–665, pp. 2–3, 30–36. This Court rejected the organization’s argument. We held that the city had not “provid[ed] a forum for private speech” with respect to monuments. Summum, 555 U. S., at 470. Rather, the city, even when “accepting a privately donated monument and placing it on city property,” had “engage[d] in expressive conduct.” Id., at 476. The speech at issue, this Court decided, was “best viewed as a form of government speech” and “therefore [was] not subject to scrutiny under the Free Speech Clause.” Id., at 464. We based our conclusion on several factors. First, his-tory shows that “[g]overnments have long used monuments to speak to the public.” Id., at 470. Thus, we observed that “[w]hen a government entity arranges for the construction of a monument, it does so because it wishes to convey some thought or instill some feeling in those who see the structure.” Ibid. Second, we noted that it “is not common for property owners to open up their property for the installation of permanent monuments that convey a message with which they do not wish to be associated.” Id., at 471. As a result, “persons who observe donated monuments routinely—and reasonably—interpret them as conveying some message on the property owner’s behalf.” Ibid. And “observers” of such monuments, as a consequence, ordinarily “appreciate the identity of the speaker.” Ibid. Third, we found relevant the fact that the city maintained control over the selection of monuments. We thought it “fair to say that throughout our Nation’s his-tory, the general government practice with respect to do-nated monuments has been one of selective receptivity.” Ibid. And we observed that the city government in Summum “ ‘effectively controlled’ the messages sent by the monuments in the [p]ark by exercising ‘final approval authority’ over their selection.” Id., at 473. In light of these and a few other relevant considerations, the Court concluded that the expression at issue was government speech. See id., at 470–472. And, in reaching that conclusion, the Court rejected the premise that the involvement of private parties in designing the monuments was sufficient to prevent the government from controlling which monuments it placed in its own public park. See id., at 470–471. Cf. Rust, supra, at 192–196 (upholding a federal regulation limiting speech in aGovernment-funded program where the program was established and administered by private parties). B Our analysis in Summum leads us to the conclusion that here, too, government speech is at issue. First, the history of license plates shows that, insofar as license plates have conveyed more than state names and vehicle identification numbers, they long have communicated messages from the States. Cf. 555 U. S., at 470 (“Governments have long used monuments to speak to the public”). In 1917, Arizona became the first State to display a graphic on its plates. J. Fox, License Plates of the United States 15 (1997) (Fox); J. Minard & T. Stentiford, A Moving History 56 (2004) (Minard). The State presented a depiction of the head of a Hereford steer. Fox 15; Minard 56. In the years since, New Hampshire plates have featured the profile of the “Old Man of the Mountain,” Massachusetts plates have included a representation of the Commonwealth’s famous codfish, and Wyoming plates have displayed a rider atop a bucking bronco. Minard 60, 61, 66. In 1928, Idaho became the first State to include a slogan on its plates. The 1928 Idaho plate proclaimed “Idaho Potatoes” and featured an illustration of a brown potato, onto which the license plate number was superimposed in green. Id., at 61. The brown potato did not catch on, but slogans on license plates did. Over the years, state plates have included the phrases “North to the Future” (Alaska), “Keep Florida Green” (Florida), “Hoosier Hospitality” (Indiana), “The Iodine Products State” (South Carolina), “Green Mountains” (Vermont), and “America’s Dairyland” (Wisconsin). Fox 13, 29, 39, 91, 101, 109. States have used license plate slogans to urge action, to promote tourism, and to tout local industries. Texas, too, has selected various messages to communicate through its license plate designs. By 1919, Texas had begun to display the Lone Star emblem on its plates. Texas Department of Transportation, The History of Texas License Plates 9, 11 (1999). In 1936, the State’s general-issue plates featured the first slogan on Texas license plates: the word “Centennial.” Id., at 20. In 1968, Texas plates promoted a San Antonio event by including the phrase “Hemisfair 68.” Id., at 46. In 1977, Texas replaced the Lone Star with a small silhouette of the State. Id., at 63. And in 1995, Texas plates celebrated “150 Years of Statehood.” Id., at 101. Additionally, the Texas Legislature has specifically authorized specialty plate designs stating, among other things, “Read to Succeed,” “Houston Livestock Show and Rodeo,” “Texans Conquer Cancer,” and “Girl Scouts.” Tex. Transp. Code Ann. §§504.607, 504.613, 504.620, 504.622. This kind of state speech has appeared on Texas plates for decades. Second, Texas license plate designs “are often closely identified in the public mind with the [State].” Summum, supra, at 472. Each Texas license plate is a government article serving the governmental purposes of vehicle registration and identification. The governmental nature of the plates is clear from their faces: The State places the name “TEXAS” in large letters at the top of every plate. More-over, the State requires Texas vehicle owners to display license plates, and every Texas license plate is issued by the State. See §504.943. Texas also owns the designs on its license plates, including the designs that Texas adopts on the basis of proposals made by private individuals and organizations. See §504.002(3). And Texas dictates the manner in which drivers may dispose of unused plates. See §504.901(c). See also §504.008(g) (requiring that vehicle owners return unused specialty plates to the State). Texas license plates are, essentially, government IDs. And issuers of ID “typically do not permit” the placement on their IDs of “message[s] with which they do not wish to be associated.” Summum, 555 U. S., at 471. Consequently, “persons who observe” designs on IDs “routinely—and reasonably—interpret them as conveying some message on the [issuer’s] behalf.” Ibid. Indeed, a person who displays a message on a Texas license plate likely intends to convey to the public that the State has endorsed that message. If not, the individual could simply display the message in question in larger letters on a bumper sticker right next to the plate. But the individual prefers a license plate design to the purely private speech expressed through bumper stickers. That may well be because Texas’s license plate designs convey government agreement with the message displayed. Third, Texas maintains direct control over the messages conveyed on its specialty plates. Texas law provides that the State “has sole control over the design, typeface, color, and alphanumeric pattern for all license plates.” §504.005. The Board must approve every specialty plate design proposal before the design can appear on a Texas plate. 43 Tex. Admin. Code §§217.45(i)(7)–(8), 217.52(b). And the Board and its predecessor have actively exercised this authority. Texas asserts, and SCV concedes, that the State has rejected at least a dozen proposed designs. Reply Brief 10; Tr. of Oral Arg. 49–51. Accordingly, like the city government in Summum, Texas “has ‘effectively controlled’ the messages [conveyed] by exercising ‘final approval authority’ over their selection.” 555 U. S., at 473 (quoting Johanns, 544 U. S., at 560–561). This final approval authority allows Texas to choose how to present itself and its constituency. Thus, Texas offers plates celebrating the many educational institutions attended by its citizens. See Tex. Transp. Code Ann. §504.615. But it need not issue plates deriding schooling. Texas offers plates that pay tribute to the Texas citrus industry. See §504.626. But it need not issue plates praising Florida’s oranges as far better. And Texas offers plates that say “Fight Terrorism.” See §504.647. But it need not issue plates promoting al Qaeda. These considerations, taken together, convince us that the specialty plates here in question are similar enough to the monuments in Summum to call for the same result. That is not to say that every element of our discussion in Summum is relevant here. For instance, in Summum we emphasized that monuments were “permanent” and we observed that “public parks can accommodate only a limited number of permanent monuments.” 555 U. S., at 464, 470, 478. We believed that the speech at issue was government speech rather than private speech in part because we found it “hard to imagine how a public park could be opened up for the installation of permanent monuments by every person or group wishing to engage in that form of expression.” Id., at 479. Here, a State could theoretically offer a much larger number of license plate designs, and those designs need not be available for time immemorial. But those characteristics of the speech at issue in Summum were particularly important because the government speech at issue occurred in public parks, which are traditional public forums for “the delivery of speeches and the holding of marches and demonstrations” by private citizens. Id., at 478. By contrast, license plates are not traditional public forums for private speech. And other features of the designs on Texas’s specialty license plates indicate that the message conveyed by those designs is conveyed on behalf of the government. Texas, through its Board, selects each design featured on the State’s specialty license plates. Texas presents these designs on government-mandated, government-controlled, and government-issued IDs that have traditionally been used as a medium for government speech. And it places the designs directly below the large letters identifying “TEXAS” as the issuer of the IDs. “The [designs] that are accepted, therefore, are meant to convey and have the effect of conveying a government message, and they thus constitute government speech.” Id., at 472. C SCV believes that Texas’s specialty license plate designs are not government speech, at least with respect to the designs (comprising slogans and graphics) that were initially proposed by private parties. According to SCV, the State does not engage in expressive activity through such slogans and graphics, but rather provides a forum for private speech by making license plates available to display the private parties’ designs. We cannot agree. We have previously used what we have called “forum analysis” to evaluate government restrictions on purely private speech that occurs on government property. Cornelius v. NAACP Legal Defense & Ed. Fund, Inc., 473 U. S. 788, 800 (1985) . But forum analysis is misplaced here. Because the State is speaking on its own behalf, the First Amendment strictures that attend the various types of government-established forums do not apply. The parties agree that Texas’s specialty license plates are not a “traditional public forum,” such as a street or a park, “which ha[s] immemorially been held in trust for the use of the public and, time out of mind, ha[s] been used for purposes of assembly, communicating thoughts between citizens, and discussing public questions.” Perry Ed. Assn. v. Perry Local Educators’ Assn., 460 U. S. 37 –46 (1983) (internal quotation marks omitted). “The Court has rejected the view that traditional public forum status extends beyond its historic confines.” Arkansas Ed. Television Comm’n v. Forbes, 523 U. S. 666 , 678 (1998). And state-issued specialty license plates lie far beyond those confines. It is equally clear that Texas’s specialty plates are neither a “ ‘designated public forum,’ ” which exists where “government property that has not traditionally been regarded as a public forum is intentionally opened up for that purpose,” Summum, supra, at 469, nor a “limited public forum,” which exists where a government has “reserv[ed a forum] for certain groups or for the discussion of certain topics,” Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819, 829 (1995) . A government “does not create a public forum by inaction or by permitting limited discourse, but only by intentionally opening a nontraditional forum for public discourse.” Cornelius, 473 U. S., at 802. And in order “to ascertain whether [a government] intended to designate a place not traditionally open to assembly and debate as a public forum,” this Court “has looked to the policy and practice of the government” and to “the nature of the property and its compatibility with expressive activity.” Ibid. Texas’s policies and the nature of its license plates indicate that the State did not intend its specialty license plates to serve as either a designated public forum or a limited public forum. First, the State exercises final authority over each specialty license plate design. This authority militates against a determination that Texas has created a public forum. See id., at 803–804 (explaining that a school mail system was not a public forum because “[t]he practice was to require permission from the individual school principal before access to the system to communicate with teachers was granted”). Second, Texas takes ownership of each specialty plate design, making it particularly untenable that the State intended specialty plates to serve as a forum for public discourse. Finally, Texas license plates have traditionally been used for government speech, are primarily used as a form of government ID, and bear the State’s name. These features of Texas license plates indicate that Texas explicitly associates itself with the speech on its plates. For similar reasons, we conclude that Texas’s specialty license plates are not a “nonpublic for[um],” which exists “[w]here the government is acting as a proprietor, managing its internal operations.” International Soc. for Krishna Consciousness, Inc. v. Lee, 505 U. S. 672 –679 (1992). With respect to specialty license plate designs, Texas is not simply managing government property, but instead is engaging in expressive conduct. As we have described, we reach this conclusion based on the historical context, observers’ reasonable interpretation of the messages conveyed by Texas specialty plates, and the effective control that the State exerts over the design selection process. Texas’s specialty license plate designs “are meant to convey and have the effect of conveying a government message.” Summum, 555 U. S., at 472. They “constitute government speech.” Ibid. The fact that private parties take part in the design and propagation of a message does not extinguish the governmental nature of the message or transform the government’s role into that of a mere forum-provider. In Summum, private entities “financed and donated monuments that the government accept[ed] and display[ed] to the public.” Id., at 470–471. Here, similarly, private parties propose designs that Texas may accept and display on its license plates. In this case, as in Summum, the “government entity may exercise [its] freedom to express its views” even “when it receives assistance from private sources for the purpose of delivering a government-controlled message.” Id., at 468. And in this case, as in Summum, forum analysis is inapposite. See id., at 480. Of course, Texas allows many more license plate designs than the city in Summum allowed monuments. But our holding in Summum was not dependent on the precise number of monuments found within the park. Indeed, we indicated that the permanent displays in New York City’s Central Park also constitute government speech. See id., at 471–472. And an amicus brief had informed us that there were, at the time, 52 such displays. See Brief for City of New York in Pleasant Grove City v. Summum, O. T. 2008, No. 07–665, p. 2. Further, there may well be many more messages that Texas wishes to convey through its license plates than there were messages that the city in Summum wished to convey through its monuments. Texas’s desire to communicate numerous messages does not mean that the messages conveyed are not Texas’s own. Additionally, the fact that Texas vehicle owners pay annual fees in order to display specialty license plates does not imply that the plate designs are merely a forum for private speech. While some nonpublic forums provide governments the opportunity to profit from speech, see, e.g., Lehman v. Shaker Heights, 418 U. S. 298, 299 (1974) (plurality opinion), the existence of government profit alone is insufficient to trigger forum analysis. Thus, if the city in Summum had established a rule that organizations wishing to donate monuments must also pay fees to assist in park maintenance, we do not believe that the result in that case would have been any different. Here, too, we think it sufficiently clear that Texas is speaking through its specialty license plate designs, such that the existence of annual fees does not convince us that the specialty plates are a nonpublic forum. Finally, we note that this case does not resemble other cases in which we have identified a nonpublic forum. This case is not like Perry Ed. Assn., where we found a school district’s internal mail system to be a nonpublic forum for private speech. See 460 U. S., at 48–49. There, it was undisputed that a number of private organizations, including a teachers’ union, had access to the mail system. See id., at 39–40. It was therefore clear that private parties, and not only the government, used the system to communicate. Here, by contrast, each specialty license plate design is formally approved by and stamped with the imprimatur of Texas. Nor is this case like Lehman, where we found the advertising space on city buses to be a nonpublic forum. See R. A. V. v. St. Paul, 505 U. S. 377, 390, n. 6 (1992) (identifying Lehman as a case about a nonpublic forum). There, the messages were located in a context (advertising space) that is traditionally available for private speech. And the advertising space, in contrast to license plates, bore no indicia that the speech was owned or conveyed by the government. Nor is this case like Cornelius, where we determined that a charitable fundraising program directed at federal employees constituted a nonpublic forum. See 473 U. S., at 804–806. That forum lacked the kind of history present here. The fundraising drive had never been a medium for government speech. Instead, it was established “to bring order to [a] solicitation process” which had previously consisted of ad hoc solicitation by individual charitable organizations. Id., at 792, 805. The drive “was designed to minimize . . . disruption to the [federal] workplace,” id., at 805, not to communicate messages from the government. Further, the charitable solicitations did not appear on a government ID under the government’s name. In contrast to the instant case, there was no reason for employees to “interpret [the solicitation] as conveying some message on the [government’s] behalf.” Summum, 555 U. S., at 471. IV Our determination that Texas’s specialty license plate designs are government speech does not mean that the designs do not also implicate the free speech rights of private persons. We have acknowledged that drivers who display a State’s selected license plate designs convey the messages communicated through those designs. See Wooley v. Maynard, 430 U. S. 705 , n. 15, 715 (1977) (observing that a vehicle “is readily associated with its operator” and that drivers displaying license plates “use their private property as a ‘mobile billboard’ for the State’s ideological message”). And we have recognized that the First Amendment stringently limits a State’s authority to compel a private party to express a view with which the private party disagrees. See id., at 715; Hurley v. Irish-American Gay, Lesbian and Bisexual Group of Boston, Inc., 515 U. S. 557, 573 (1995) ; West Virginia Bd. of Ed. v. Barnette, 319 U. S. 624, 642 (1943) . But here, compelled private speech is not at issue. And just as Texas cannot require SCV to convey “the State’s ideological message,” Wooley, supra, at 715, SCV cannot force Texas to include a Confederate battle flag on its specialty license plates. * * * For the reasons stated, we hold that Texas’s specialty license plate designs constitute government speech and that Texas was consequently entitled to refuse to issue plates featuring SCV’s proposed design. Accordingly, the judgment of the United States Court of Appeals for the Fifth Circuit is Reversed.

574.US.40

Petitioner Gregory Warger sued respondent Randy Shauers in federal court for negligence for injuries suffered in a motor vehicle accident. After the jury returned a verdict for Shauers, one of the jurors contacted Warger’s counsel, claiming that Regina Whipple, the jury foreperson, had revealed during deliberations that her daughter had been at fault in a fatal motor vehicle accident, and that a lawsuit would have ruined her daughter’s life. Armed with an affidavit from the juror, Warger moved for a new trial, arguing that Whipple had deliberately lied during voir dire about her impartiality and ability to award damages. The District Court denied Warger’s motion, holding that Federal Rule of Evidence 606(b), which bars evidence “about any statement made . . . during the jury’s deliberations,” barred the affidavit, and that none of the Rule’s three exceptions, see Rule 606(b)(2), were applicable. The Eighth Circuit affirmed. Held: 1. Rule 606(b) applies to juror testimony during a proceeding in which a party seeks to secure a new trial on the ground that a juror lied during voir dire. Pp. 3–10. (a) This reading accords with the plain meaning of Rule 606(b), which applies to “an inquiry into the validity of [the] verdict.” This understanding is also consistent with the underlying common-law rule on which Congress based Rule 606(b). The so-called “federal rule” made jury deliberations evidence inadmissible even if used to demonstrate dishonesty during voir dire. Both the majority of courts and this Court’s pre-Rule606(b) cases, see McDonald v. Pless, 238 U.S. 264, 268 ; Clark v. United States, 289 U.S. 1 , favored this rule over the “Iowa rule,” which permitted the use of such jury deliberations evidence. The federal approach is clearly reflected in the language Congress chose when it enacted Rule 606(b), and legislative history confirms that Congress’ choice was no accident. See Tanner v. United States, 483 U.S. 107, 125 . Pp. 3–8. (b) Warger’s arguments against this straightforward understanding are not persuasive. Pp. 8–10. (1) First, Warger insists that proceedings for a new trial based on voir dire dishonesty do not involve an “inquiry into the validity of the verdict.” His reading would restrict Rule 606(b)’s application to claims of error for which a court must examine the manner in which the jury reached its verdict, but the Rule does not focus on the means by which deliberations evidence might be used to invalidate a verdict. It simply applies during a proceeding in which a verdict may be rendered invalid. Pp. 8–9. (2) Warger also contends that excluding jury deliberations evidence that shows voir dire dishonesty is unnecessary to fulfill Congress’ objectives, but his arguments would apply to all evidence rendered inadmissible by Rule 606(b), and he cannot escape the scope of the Rule merely by asserting that Congress’ concerns were misplaced. P. 9. (3) Finally, Warger invokes the canon of constitutional avoidance, contending that only his interpretation protects the right to an impartial jury. But that canon has no application here, where there is no ambiguity. See United States v. Oakland Cannabis Buyers’ Cooperative, 532 U.S. 483, 494 . Moreover, this Court’s Tanner decision forecloses any claim that Rule 606(b) is unconstitutional. Similar to the right at issue in that case, Warger’s right to an impartial jury remains protected despite Rule 606(b)’s removal of one means of ensuring unbiased jurors. Even if a juror lies to conceal bias, parties may bring to the court’s attention evidence of bias before the verdict is rendered and use nonjuror evidence after the verdict is rendered. Pp. 9–10. 2. The affidavit at issue was not admissible under Rule 606(b)(2)(A)’s exception for evidence of “extraneous prejudicial information.” Generally speaking, extraneous information derives from a source “external” to the jury. See Tanner, 483 U. S., at 117. Here, the excluded affidavit falls on the “internal” side. Warger contends that any information Whipple shared with the other jurors was extraneous because she would have been disqualified from the jury had she disclosed her daughter’s accident. However, such an exception would swallow up much of the rest of the restrictive version of the common-law rule that Congress adopted in enacting Rule 606(b). Pp. 11–13. 721 F.3d 606, affirmed. Sotomayor, J., delivered the opinion for a unanimous Court.

Federal Rule of Evidence 606(b) provides that certain juror testimony regarding what occurred in a jury room is inadmissible “[d]uring an inquiry into the validity of a verdict.” The question presented in this case is whether Rule 606(b) precludes a party seeking a new trial from using one juror’s affidavit of what another juror said in deliberations to demonstrate the other juror’s dishonesty during voir dire. We hold that it does. I Petitioner Gregory Warger was riding his motorcycle on a highway outside Rapid City, South Dakota, when a truck driven by respondent Randy Shauers struck him from behind. Warger claims he was stopped at the time of the accident, while Shauers claims that Warger suddenly pulled out in front of him. Regardless of the cause of the accident, no one disputes its tragic result: Warger sustained serious injuries that ultimately required the amputation of his left leg. Warger sued Shauers for negligence in Federal District Court. During jury selection, counsel for both parties conducted lengthy voir dire of the prospective jurors. Warger’s counsel asked whether any jurors would be unable to award damages for pain and suffering or for future medical expenses, or whether there was any juror who thought, “I don’t think I could be a fair and impartial juror on this kind of case.” App. 105. Prospective juror Regina Whipple, who was later selected as the jury foreperson, answered no to each of these questions. See id., at 83, 89, 105. Trial commenced, and the jury ultimately returned a verdict in favor of Shauers. Shortly thereafter, one of the jurors contacted Warger’s counsel to express concern over juror Whipple’s conduct. The complaining juror subsequently signed an affidavit claiming that Whipple had spoken during deliberations about “a motor vehicle collision in which her daughter was at fault for the collision and a man died,” and had “related that if her daughter had been sued, it would have ruined her life.” App. to Pet. for Cert. 40a–41a. Relying on this affidavit, Warger moved for a new trial. He contended that Whipple had deliberately lied during voir dire about her impartiality and ability to award damages. Thus, he asserted, he had satisfied the requirements of McDonough Power Equipment, Inc. v. Greenwood, 464 U. S. 548 (1984) , which holds that a party may “obtain a new trial” if he “demonstrate[s] that a juror failed to answer honestly a material question on voir dire, and . . . that a correct response would have provided a valid basis for a challenge for cause.” Id., at 556. The District Court refused to grant a new trial, holding that the only evidence that supported Warger’s motion, the complaining juror’s affidavit, was barred by Federal Rule of Evidence 606(b). As relevant here, that Rule provides that “[d]uring an inquiry into the validity of a verdict,” evidence “about any statement made or incident that occurred during the jury’s deliberations” is inadmissible. Rule 606(b)(1). The Rule contains three specific exceptions—allowing testimony “about whether (A) extraneous prejudicial information was improperly brought to the jury’s attention; (B) an outside influence was improperly brought to bear on any juror; or (C) a mistake was made in entering the verdict on the verdict form,” Rule 606(b)(2)—but the District Court found none of these exceptions to be applicable. The Eighth Circuit affirmed. 721 F. 3d 606 (2013). It first held that Warger’s proffered evidence did not fall within the “extraneous prejudicial evidence” exception set forth in Rule 606(b)(2)(A). The court explained that “[j]urors’ personal experiences do not constitute extraneous information; it is unavoidable they will bring such innate experiences into the jury room.” Id., at 611. Next, the court rejected Warger’s alternative argument that Rule 606(b) is wholly inapplicable when a litigant offers evidence to show that a juror was dishonest during voir dire. Acknowledging that there was a split among the Federal Courts of Appeals on this question, the Eighth Circuit joined those Circuits that had held that Rule 606(b) applies to any proceeding in which the jury’s verdict might be invalidated, including efforts to demonstrate that a juror lied during voir dire. Compare id., at 611–612 (citing Williams v. Price, 343 F. 3d 223, 235–237 (CA3 2003), and United States v. Benally, 546 F. 3d 1230, 1235 (CA10 2008)), with Hard v. Burlington N. R. Co., 812 F. 2d 482, 485 (CA9 1987) (“Statements which tend to show deceit during voir dire are not barred by [Rule 606(b)]”), and Maldonado v. Missouri P. R. Co., 798 F. 2d 764, 770 (CA5 1986) (same). We granted certiorari, 571 U. S. ___ (2014), and now affirm. II We hold that Rule 606(b) applies to juror testimony during a proceeding in which a party seeks to secure a new trial on the ground that a juror lied during voir dire. In doing so, we simply accord Rule 606(b)’s terms their plain meaning. The Rule, after all, applies “[d]uring an inquiry into the validity of a verdict.” Rule 606(b)(1). A postverdict motion for a new trial on the ground of voir dire dishonesty plainly entails “an inquiry into the validity of [the] verdict”: If a juror was dishonest during voir dire and an honest response would have provided a valid basis to challenge that juror for cause, the verdict must be invalidated. See McDonough, 464 U. S., at 556. This understanding of the text of Rule 606(b) is consistent with the underlying common-law rule on which it was based. Although some common-law courts would have permitted evidence of jury deliberations to be introduced to demonstrate juror dishonesty during voir dire, the majority would not, and the language of Rule 606(b) reflects Congress’ enactment of the more restrictive version of the common-law rule. Rule 606(b) had its genesis in Vaise v. Delaval, 1 T. R. 11, 99 Eng. Rep. 944 (K. B. 1785), in which Lord Mansfield held inadmissible an affidavit from two jurors claiming that the jury had decided the case through a game of chance. See 8 J. Wigmore, Evidence §2352, p. 696 (J. McNaughton rev. 1961). The rule soon took root in the United States, id., at 696–697, where it was viewed as both promoting the finality of verdicts and insulating the jury from outside influences, see McDonald v. Pless, 238 U. S. 264 –268 (1915). Some versions of the rule were narrower than others. Under what was sometimes known as the “Iowa” approach, juror testimony regarding deliberations was excluded only to the extent that it related to matters that “ ‘inhere[d] in the verdict,’ ” which generally consisted of evidence of the jurors’ subjective intentions and thought processes in reaching a verdict. 3 C. Mueller & L. Kirkpatrick, Federal Evidence §6:16, p. 70 (4th ed. 2013); 8 Wigmore, Evidence §§2353, 2354, at 699–702. [ 1 ] A number of courts adhering to the Iowa rule held that testimony regarding jury deliberations is admissible when used to challenge juror conduct during voir dire. See, e.g., Mathi- sen v. Norton, 187 Wash. 240, 244–246, 60 P. 2d 1, 3–4 (1936); Williams v. Bridges, 140 Cal. App. 537, 538–541, 35 P. 2d 407, 408–409 (1934). But other courts applied a broader version of the anti-impeachment rule. Under this version, sometimes called the “federal” approach, litigants were prohibited from using evidence of jury deliberations unless it was offered to show that an “extraneous matter” had influenced the jury. See 3 Mueller & Kirkpatrick, Federal Evidence §6:16, at 71; Rules of Evidence for United States Courts and Magistrates, 56 F. R. D. 183, 265 (1973). The “great majority” of appellate courts applying this version of the rule held jury deliberations evidence inadmissible even if used to demonstrate dishonesty during voir dire. Wilson v. Wiggins, 54 Ariz. 240, 246, 94 P. 2d 870, 872 (1939); see, e.g., Willis v. Davis, 333 P. 2d 311, 314 (Okla. 1958); Turner v. Hall’s Adm’x, 252 S. W. 2d 30, 34 (Ky. 1952); Hinkel v. Oregon Chair Co., 80 Ore. 404, 406, 156 P. 438, 439 (1916); State v. Cloud, 130 La. 955, 958–960, 58 So. 827, 828–829 (1912); Payne v. Burke, 236 App. Div. 527, 528–530, 260 N. Y. S. 259, 260–262 (1932). This Court occasionally employed language that might have suggested a preference for the Iowa rule. See Hyde v. United States, 225 U. S. 347 –384 (1912) (“[W]e think the rule expressed in Wright v. Illinois & Miss. Tel. Co., 20 Iowa 195 [1866], . . . should apply, that the testimony of jurors should not be received to show matters which essentially inhere in the verdict itself and necessarily depend upon the testimony of the jurors and can receive no corroboration”); Mattox v. United States, 146 U. S. 140 –149 (1892) (quoting at length a Kansas Supreme Court decision setting out the Iowa test). But to the extent that these decisions created any question as to which approach this Court followed, McDonald v. Pless largely settled matters. There, we held that juror affidavits were not admissible to show that jurors had entered a “quotient” verdict, precisely the opposite of the result reached by the Iowa Supreme Court in its decision establishing the Iowa approach. Compare 238 U. S., at 265, 268, with Wright v. Illinois & Miss. Tel. Co., 20 Iowa 195, 211–212 (1866). In doing so, we observed that although decisions in a few States made admissible a “juror’s affidavit as to an overt act of misconduct, which was capable of being controverted by other jurors,” the argument in favor of that approach (i.e., the Iowa rule) had not been generally accepted, because permitting such evidence “would open the door to the most pernicious arts and tampering with jurors.” 238 U. S., at 268 (internal quotation marks omitted). Our subsequent decision in Clark v. United States, 289 U. S. 1 (1933) , was consistent with our apparent rejection of the Iowa approach. In Clark, the Government had prosecuted for contempt a juror who, during voir dire in a prior case, had falsely denied knowing the defendant. Id., at 6–8. We held that the prosecution could introduce evidence of what had occurred during deliberations in the prior case, rejecting the juror’s argument that these communications were privileged. We were careful to explain, however, that nothing in our decision was “at variance with the rule . . . that the testimony of a juror is not admissible for the impeachment of his verdict.” Id., at 18. This was because the verdict in the original case was not at issue, and therefore “the rule against impeachment [was] wholly unrelated to the problem . . . before us.” Ibid.; accord, McDonald, 238 U. S., at 269. Clark thus clarified that the rule against jurors’ impeaching their verdicts applies only in a proceeding actually impeaching that verdict—precisely the line Rule 606(b) draws when it refers to an “inquiry into the validity of a verdict.” In any event, these decisions predated Congress’ enactment of Rule 606(b), and Congress was undoubtedly free to prescribe a broader version of the anti-impeachment rule than we had previously applied. The language of the Rule it adopted clearly reflects the federal approach: As enacted, Rule 606(b) prohibited the use of any evidence of juror deliberations, subject only to the express exceptions for extraneous information and outside influences. [ 2 ] For those who consider legislative history relevant, here it confirms that this choice of language was no accident. Congress rejected a prior version of the Rule that, in accordance with the Iowa approach, would have prohibited juror testimony only as to the “effect of anything upon . . . [any] juror’s mind or emotions . . . or concerning his mental processes.” Committee on Rules of Practice and Procedure of the Judicial Conference of the United States, Revised Draft of Proposed Rules of Evidence for the United States Courts and Magistrates, 51 F. R. D. 315, 387 (1971); see Tanner v. United States, 483 U. S. 107 –125 (1987) (detailing the legislative history of the Rule). Thus Congress “specifically understood, considered, and rejected a version of Rule 606(b)” that would have likely permitted the introduction of evidence of deliberations to show dishonesty during voir dire. Id., at 125. III A Seeking to rebut this straightforward understanding of Rule 606(b), Warger first insists that the proceedings that follow a motion for new trial based on dishonesty during voir dire do not involve an “inquiry into the validity of the verdict.” His argument is as follows: Under McDonough, a party moving for a new trial on the basis of voir dire dishonesty need not show that this dishonesty had an effect on the verdict. See 464 U. S., at 556. Although a successful claim will result in vacatur of the judgment, vacatur is simply the remedy for the McDonough error, just as it may be the remedy for a variety of errors that have nothing to do with the manner in which the jury reached its verdict. See, e.g., United States v. Davila, 569 U. S. ___, ___ (2013) (slip op., at 12) (listing certain “ ‘structural’ ” errors warranting “automatic reversal” of a criminal conviction). Therefore, Warger asserts, the “inquiry begins and ends with what happened during voir dire.” Brief for Petitioner 19–20. We are not persuaded. Warger, it seems, would restrict Rule 606(b)’s application to those claims of error for which a court must examine the manner in which the jury reached its verdict—claims, one might say, involving an inquiry into the jury’s verdict. But the “inquiry” to which the Rule refers is one into the “validity of the verdict,” not into the verdict itself. The Rule does not focus on the means by which deliberations evidence might be used to invalidate a verdict. It does not say “during an inquiry into jury deliberations,” or prohibit the introduction of evidence of deliberations “for use in determining whether an asserted error affected the jury’s verdict.” It simply applies “[d]uring an inquiry into the validity of the verdict”—that is, during a proceeding in which the verdict may be rendered invalid. Whether or not a juror’s alleged misconduct during voir dire had a direct effect on the jury’s verdict, the motion for a new trial requires a court to determine whether the verdict can stand. B Next, Warger contends that excluding jury deliberations evidence tending to show that a juror lied during voir dire is unnecessary to fulfill Congress’ apparent objectives of encouraging full and open debate in the jury room and preventing the harassment of former jurors. He observes that jurors remain free to, and may sometimes be forced to, disclose what happened in the jury room, and that ethical rules limit the ability of parties to harass jurors following trial. But these are arguments against Rule 606(b) generally, not arguments for the particular exception to the Rule that Warger seeks. Congress’ enactment of Rule 606(b) was premised on the concerns that the use of deliberations evidence to challenge verdicts would represent a threat to both jurors and finality in those circumstances not covered by the Rule’s express exceptions. Warger cannot escape the scope of the Rule Congress adopted simply by asserting that its concerns were misplaced. C Nor do we accept Warger’s contention that we must adopt his interpretation of Rule 606(b) so as to avoid constitutional concerns. The Constitution guarantees both criminal and civil litigants a right to an impartial jury. See, e.g., Sheppard v. Maxwell, 384 U. S. 333, 362 (1966) ; Thiel v. Southern Pacific Co., 328 U. S. 217, 220 (1946) . And we have made clear that voir dire can be an essential means of protecting this right. See, e.g., Turner v. Murray, 476 U. S. 28, 36 (1986) (plurality opinion); Ham v. South Carolina, 409 U. S. 524, 527 (1973) . These principles, Warger asserts, require that parties be allowed to use evidence of deliberations to demonstrate that a juror lied during voir dire. Given the clarity of both the text and history of Rule 606(b), however, the canon of constitutional avoidance has no role to play here. The canon “is a tool for choosing between competing plausible interpretations” of a provision. Clark v. Suarez-Martinez, 543 U. S. 371, 381 (2005) . It “has no application in the absence of . . . ambiguity.” United States v. Oakland Cannabis Buyers’ Cooperative, 532 U. S. 483, 494 (2001) . We see none here. Moreover, any claim that Rule 606(b) is unconstitutional in circumstances such as these is foreclosed by our decision in Tanner. In Tanner, we concluded that Rule 606(b) precluded a criminal defendant from introducing evidence that multiple jurors had been intoxicated during trial, rejecting the contention that this exclusion violated the defendant’s Sixth Amendment right to “ ‘a tribunal both impartial and mentally competent to afford a hearing.’ ” 483 U. S., at 126 (quoting Jordan v. Massachusetts, 225 U. S. 167, 176 (1912) ). We reasoned that the defendant’s right to an unimpaired jury was sufficiently protected by voir dire, the observations of court and counsel during trial, and the potential use of “nonjuror evidence” of misconduct. 483 U. S., at 127. Similarly here, a party’s right to an impartial jury remains protected despite Rule 606(b)’s removal of one means of ensuring that jurors are unbiased. Even if jurors lie in voir dire in a way that conceals bias, juror impartiality is adequately assured by the parties’ ability to bring to the court’s attention any evidence of bias before the verdict is rendered, and to employ nonjuror evidence even after the verdict is rendered. [ 3 ] IV We further hold, consonant with the Eighth Circuit, that the affidavit Warger sought to introduce was not admissible under Rule 606(b)(2)(A)’s exception for evidence as to whether “extraneous prejudicial information was improperly brought to the jury’s attention.” Generally speaking, information is deemed “extraneous” if it derives from a source “external” to the jury. See Tanner, 483 U. S., at 117. “External” matters include publicity and information related specifically to the case the jurors are meant to decide, while “internal” matters include the general body of experiences that jurors are understood to bring with them to the jury room. See id., at 117–119; 27 C. Wright & V. Gold, Federal Practice and Procedure: Evidence §6075, pp. 520–521 (2d ed. 2007). Here, the excluded affidavit falls on the “internal” side of the line: Whipple’s daughter’s accident may well have informed her general views about negligence liability for car crashes, but it did not provide either her or the rest of the jury with any specific knowledge regarding Shauers’ collision with Warger. Indeed, Warger does not argue that Whipple’s statements related to “extraneous” information in this sense. Instead, he contends that because Whipple would have been disqualified from the jury had she disclosed her daughter’s accident, any information she shared with the other jurors was extraneous. We cannot agree that whenever a juror should have been excluded from the jury, anything that juror says is necessarily “extraneous” within the meaning of Rule 606(b)(2)(A). Were that correct, parties would find it quite easy to avoid Rule 606(b)’s limitations. As discussed above, Congress adopted the restrictive version of the anti-impeachment rule, one that common-law courts had concluded precludes parties from using deliberations evidence to prove juror dishonesty during voir dire. But if Warger’s understanding of the “extraneous” information exception were accepted, then any time a party could use such evidence to show that a juror’s “correct response [during voir dire] would have provided a valid basis for a challenge”—a prerequisite for relief under McDonough, 464 U. S., at 556—all evidence of what that juror said during deliberations would be admissible. The “extraneous” information exception would swallow much of the rest of Rule 606(b). Even if such a result were not precluded by Congress’ apparent intent to adopt the restrictive federal approach, it is foreclosed by Tanner, which relied upon the doctrine that “treat[s] allegations of the physical or mental incompetence of a juror as ‘internal’ rather than ‘external’ matters.” 483 U. S., at 118. Tanner cited, in particular, cases holding that evidence of jurors’ insanity, inability to understand English, and hearing impairments are all “internal” matters subject to exclusion under Rule 606(b). Id., at 119. Were we to follow Warger’s understanding of the “extraneous information” exception, all these cases, including Tanner, would have been wrongly decided: If the jurors were not able to serve on the jury in the first place, or should have been dismissed for their misconduct during the trial, then what they said or did during deliberations would necessarily be “extraneous” and admissible. Tanner’s implicit rejection of this view easily extends from the sort of juror incompetence considered in that case to the alleged bias considered here. Whether a juror would have been struck from the jury because of incompetence or bias, the mere fact that a juror would have been struck does not make admissible evidence regarding that juror’s conduct and statements during deliberations. For the foregoing reasons, the judgment of the United States Court of Appeals for the Eighth Circuit is affirmed. It is so ordered. Notes 1 The Iowa rule derived from Wright v. Illinois & Miss. Tel. Co., 20 Iowa 195 (1866), in which the Iowa Supreme Court held that a trial court considering a motion for a new trial should have accepted the affidavits of four jurors who claimed that their damages verdict had been determined by taking the average of the sums each juror thought proper (a “quotient” verdict). Id., at 212–213. The Wright court reasoned that, unlike evidence of a juror’s subjective intentions in reaching a verdict, whether the verdict had been obtained in this fashion was an “independent fact” and thus could and should be proved by any avail-able evidence. Id., at 211. 2 The additional exception for mistakes made in entering the verdict on the verdict form was adopted in 2006. See 547 U. S. 1281 . 3 There may be cases of juror bias so extreme that, almost by definition, the jury trial right has been abridged. If and when such a case arises, the Court can consider whether the usual safeguards are or are not sufficient to protect the integrity of the process. We need not consider the question, however, for those facts are not presented here.

575.US.2014_13-935

Respondent Richard Sharif tried to discharge a debt he owed petitioners, Wellness International Network, Ltd., and its owners (collectively Wellness), in his Chapter 7 bankruptcy. Wellness sought, inter alia, a declaratory judgment from the Bankruptcy Court, contending that a trust Sharif claimed to administer was in fact Sharif’s alter-ego, and that its assets were his personal property and part of his bankruptcy estate. The Bankruptcy Court eventually entered a default judgment against Sharif. While Sharif’s appeal was pending in District Court, but before briefing concluded, this Court held that Article III forbids bankruptcy courts to enter a final judgment on claims that seek only to “augment” the bankruptcy estate and would otherwise “exis[t] without regard to any bankruptcy proceeding.” Stern v. Marshall, 564 U. S. ___, ___. After briefing closed, Sharif sought permission to file a supplemental brief raising a Stern objection. The District Court denied the motion, finding it untimely, and affirmed the Bankruptcy Court’s judgment. As relevant here, the Seventh Circuit determined that Sharif’s Stern objection could not be waived because it implicated structural interests and reversed on the alter-ego claim, holding that the Bankruptcy Court lacked constitutional authority to enter final judgment on that claim. Held: 1. Article III permits bankruptcy judges to adjudicate Stern claims with the parties’ knowing and voluntary consent. Pp. 8–17. (a) The foundational case supporting the adjudication of legal disputes by non-Article III judges with the consent of the parties is Commodity Futures Trading Comm’n v. Schor, 478 U. S. 833 . There, the Court held that the right to adjudication before an Article III court is “personal” and therefore “subject to waiver.” Id., at 848. The Court also recognized that if Article III’s structural interests as “ ‘an inseparable element of the constitutional system of checks and balances’ ” are implicated, “the parties cannot by consent cure the constitutional difficulty.” Id., at 850–851. The importance of consent was reiterated in two later cases involving the Federal Magistrates Act’s assignment of non-Article III magistrate judges to supervise voir dire in felony trials. In Gomez v. United States, 490 U. S. 858 , the Court held that a magistrate judge was not permitted to select a jury without the defendant’s consent, id., at 864. But in Peretz v. United States, 501 U. S. 923 , the Court stated that “the defendant’s consent significantly changes the constitutional analysis,” id., at 932. Because an Article III court retained supervisory authority over the process, the Court found “no structural protections . . . implicated” and upheld the Magistrate Judge’s action. Id., at 937. Pp. 8–12. (b) The question whether allowing bankruptcy courts to decide Stern claims by consent would “impermissibly threate[n] the institutional integrity of the Judicial Branch,” Schor, 478 U. S., at 851, must be decided “with an eye to the practical effect that the” practice “will have on the constitutionally assigned role of the federal judiciary,” ibid. For several reasons, this practice does not usurp the constitutional prerogatives of Article III courts. Bankruptcy judges are appointed and may be removed by Article III judges, see 28 U. S. C. §§152(a)(1), (e); “serve as judicial officers of the United States district court,” §151; and collectively “constitute a unit of the district court” for the district in which they serve, §152(a)(1). Bankruptcy courts hear matters solely on a district court’s reference, §157(a), and possess no free-floating authority to decide claims traditionally heard by Article III courts, see Schor, 478 U. S., at 854, 856. “[T]he decision to invoke” the bankruptcy court’s authority “is left entirely to the parties,” id., at 855, and “the power of the federal judiciary to take jurisdiction” remains in place, ibid. Finally, there is no indication that Congress gave bankruptcy courts the ability to decide Stern claims in an effort to aggrandize itself or humble the Judiciary. See, e.g., Peretz, 501 U. S., at 937. Pp. 12–15. (c) Stern does not compel a different result. It turned on the fact that the litigant “did not truly consent to” resolution of the claim against it in a non-Article III forum, 564 U. S., at ___, and thus, does not govern the question whether litigants may validly consent to adjudication by a bankruptcy court. Moreover, expanding Stern to hold that a litigant may not waive the right to an Article III court through consent would be inconsistent with that opinion’s own description of its holding as “a ‘narrow’ one” that did “not change all that much” about the division of labor between district and bankruptcy courts. Id., at ___. Pp. 15–17. 2. Consent to adjudication by a bankruptcy court need not be express, but must be knowing and voluntary. Neither the Constitution nor the relevant statute—which requires “the consent of all parties to the proceeding” to hear a Stern claim, §157(c)(2)—mandates express consent. Such a requirement would be in great tension with this Court’s holding that substantially similar language in §636(c)—which authorizes magistrate judges to conduct proceedings “[u]pon consent of the parties”—permits waiver based on “actions rather than words,” Roell v. Withrow, 538 U. S. 580 . Roell’s implied consent standard supplies the appropriate rule for bankruptcy court adjudications and makes clear that a litigant’s consent—whether express or implied—must be knowing and voluntary. Pp. 18–19. 3. The Seventh Circuit should decide on remand whether Sharif’s actions evinced the requisite knowing and voluntary consent and whether Sharif forfeited his Stern argument below. P. 20. 727 F. 3d 751, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined, and in which Alito, J., joined in part. Alito, J., filed an opinion concurring in part and concurring in the judgment. Roberts, C. J., filed a dissenting opinion, in which Scalia, J., joined, and in which Thomas, J., joined as to Part I. Thomas, J., filed a dissenting opinion.

Article III, §1, of the Constitution provides that “[t]he judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish.” Congress has in turn established 94 District Courts and 13 Courts of Appeals, composed of judges who enjoy the protections of Article III: life tenure and pay that cannot be diminished. Because these protections help to ensure the integrity and independence of the Judiciary, “we have long recognized that, in general, Congress may not withdraw from” the Article III courts “any matter which, from its nature, is the subject of a suit at the common law, or in equity, or in admiralty.” Stern v. Marshall, 564 U. S. ___, ___ (2011) (slip op., at 18) (internal quotation marksomitted). Congress has also authorized the appointment of bankruptcy and magistrate judges, who do not enjoy the protections of Article III, to assist Article III courts in their work. The number of magistrate and bankruptcy judgeships exceeds the number of circuit and district judgeships.[1] And it is no exaggeration to say that without the distinguished service of these judicial colleagues, the work of the federal court system would grind nearly to a halt.[2] Congress’ efforts to align the responsibilities of non-Article III judges with the boundaries set by the Constitution have not always been successful. In Northern Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U. S. 50 (1982) (plurality opinion), and more recently in Stern, this Court held that Congress violated Article III by authorizing bankruptcy judges to decide certain claims for which litigants are constitutionally entitled to an Article III adjudication. This case presents the question whether Article III allows bankruptcy judges to adjudicate such claims with the parties’ consent. We hold that Article III is not violated when the parties knowingly and voluntarily consent to adjudication by a bankruptcy judge. I A Before 1978, district courts typically delegated bankruptcy proceedings to “referees.” Executive Benefits Ins. Agency v. Arkison, 573 U. S. ___, ___ (2014) (slip op., at 4). Under the Bankruptcy Act of 1898, bankruptcy referees had “[s]ummary jurisdiction” over “claims involving ‘property in the actual or constructive possession of the bankruptcy court’ ”—that is, over the apportionment of the bankruptcy estate among creditors. Ibid. (alteration omitted). They could preside over other proceedings—matters implicating the court’s “plenary jurisdiction”—by consent. Id., at ___ (slip op., at 5); see also MacDonald v. Plymouth County Trust Co., 286 U. S. 263 –267 (1932). In 1978, Congress enacted the Bankruptcy Reform Act, which repealed the 1898 Act and gave the newly created bankruptcy courts power “much broader than that exercised under the former referee system.” Northern Pipeline, 458 U. S., at 54. The Act “[e]liminat[ed] the distinction between ‘summary’ and ‘plenary’ jurisdiction” and enabled bankruptcy courts to decide “all ‘civil proceedings arising under title 11 [the Bankruptcy title] or arisingin or related to cases under title 11.’ ” Ibid. (emphasis de-leted). Congress thus vested bankruptcy judges with most of the “ ‘powers of a court of equity, law, and admiralty,’ ” id., at 55, without affording them the benefits of Article III. This Court therefore held parts of the system unconstitutional in Northern Pipeline. Congress responded by enacting the Bankruptcy Amendments and Federal Judgeship Act of 1984. Under that Act, district courts have original jurisdiction over bankruptcy cases and related proceedings. 28 U. S. C. §§1334(a), (b). But “[e]ach district court may provide that any or all” bankruptcy cases and related proceedings “shall be referred to the bankruptcy judges for the district.” §157(a). Bankruptcy judges are “judicial officers of the United States district court,” appointed to 14-year terms by the courts of appeals, and subject to removal for cause. §§152(a)(1), (e). “The district court may withdraw” a reference to the bankruptcy court “on its own motion or on timely motion of any party, for cause shown.” §157(d). When a district court refers a case to a bankruptcy judge, that judge’s statutory authority depends on whether Congress has classified the matter as a “[c]ore proceed-in[g]” or a “[n]on-core proceedin[g],” §§157(b)(2), (4)—much as the authority of bankruptcy referees, before the 1978 Act, depended on whether the proceeding was “summary” or “plenary.” Congress identified as “[c]ore” a nonexclusive list of 16 types of proceedings, §157(b)(2), in which it thought bankruptcy courts could constitutionally enter judgment.[3] Congress gave bankruptcy courts the power to “hear and determine” core proceedings and to “enter appropriate orders and judgments,” subject to appellate review by the district court. §157(b)(1); see §158. But it gave bankruptcy courts more limited author-ity in non-core proceedings: They may “hear and determine” such proceedings, and “enter appropriate orders and judgments,” only “with the consent of all the parties to the proceeding.” §157(c)(2). Absent consent, bankruptcy courts in non-core proceedings may only “submit proposed findings of fact and conclusions of law,” which the district courts review de novo. §157(c)(1). B Petitioner Wellness International Network is a manufacturer of health and nutrition products.[4] Wellness and respondent Sharif entered into a contract under which Sharif would distribute Wellness’ products. The relationship quickly soured, and in 2005, Sharif sued Wellness in the United States District Court for the Northern District of Texas. Sharif repeatedly ignored Wellness’ discovery requests and other litigation obligations, resulting in an entry of default judgment for Wellness. The District Court eventually sanctioned Sharif by awarding Wellness over $650,000 in attorney’s fees. This case arises from Wellness’ long-running—and so far unsuccessful—efforts to collect on that judgment. In February 2009, Sharif filed for Chapter 7 bankruptcy in the Northern District of Illinois. The bankruptcy petition listed Wellness as a creditor. Wellness requested documents concerning Sharif’s assets, which Sharif did not provide. Wellness later obtained a loan application Sharif had filed in 2002, listing more than $5 million in assets. When confronted, Sharif informed Wellness and the Chapter 7 trustee that he had lied on the loan application. The listed assets, Sharif claimed, were actually owned by the Soad Wattar Living Trust (Trust), an entity Sharif said he administered on behalf of his mother, and for the benefit of his sister. Wellness pressed Shariffor information on the Trust, but Sharif again failed to respond. Wellness filed a five-count adversary complaint against Sharif in the Bankruptcy Court. See App. 5–22. Counts I–IV of the complaint objected to the discharge of Sharif’s debts because, among other reasons, Sharif had concealed property by claiming that it was owned by the Trust. Count V of the complaint sought a declaratory judgment that the Trust was Sharif’s alter ego and that its assets should therefore be treated as part of Sharif’s bankruptcy estate. Id., at 21. In his answer, Sharif admitted that the adversary proceeding was a “core proceeding” under 28 U. S. C. §157(b)—i.e., a proceeding in which the Bankruptcy Court could enter final judgment subject to appeal. See §§157(b)(1), (2)(J); App. 24. Indeed, Sharif requested judgment in his favor on all counts of Wellness’ complaint and urged the Bankruptcy Court to “find that the Soad Wattar Living Trust is not property of the [bankruptcy] estate.” Id., at 44. A familiar pattern of discovery evasion ensued. Wellness responded by filing a motion for sanctions, or, in the alternative, to compel discovery. Granting the motion to compel, the Bankruptcy Court warned Sharif that if he did not respond to Wellness’ discovery requests a default judgment would be entered against him. Sharif eventu-ally complied with some discovery obligations, but did not produce any documents related to the Trust. In July 2010, the Bankruptcy Court issued a ruling finding that Sharif had violated the court’s discovery order. See App. to Pet. for Cert. 92a–120a. It accordingly denied Sharif’s request to discharge his debts and entered a default judgment against him in the adversary proceeding. And it declared, as requested by count V of Wellness’ complaint, that the assets supposedly held by the Trust were in fact property of Sharif’s bankruptcy estate because Sharif “treats [the Trust’s] assets as his own prop-erty.” Id., at 119a. Sharif appealed to the District Court. Six weeks before Sharif filed his opening brief in the District Court, this Court decided Stern. In Stern, the Court held that Article III prevents bankruptcy courts from entering final judgment on claims that seek only to “augment” the bankruptcy estate and would otherwise “exis[t] without regard toany bankruptcy proceeding.” 564 U. S., at ___, ___ (slip op., at 27, 34). Sharif did not cite Stern in his opening brief. Rather, after the close of briefing, Sharif moved for leave to file a supplemental brief, arguing that in light of In re Ortiz, 665 F. 3d 906 (CA7 2011)—a recently issued decision interpreting Stern—“the bankruptcy court’s order should only be treated as a report and recommendation.” App. 145. The District Court denied Sharif's motion for supplemental briefing as untimely and affirmed the Bankruptcy Court’s judgment. The Court of Appeals for the Seventh Circuit affirmed in part and reversed in part. 727 F. 3d 751 (2013). The Seventh Circuit acknowledged that ordinarily Sharif’s Stern objection would “not [be] preserved because he waited too long to assert it.” 727 F. 3d, at 767.[5] But the court determined that the ordinary rule did not apply because Sharif’s argument concerned “the allocation of authority between bankruptcy courts and district courts” under Article III, and thus “implicate[d] structural interests.” Id., at 771. Based on those separation-of-powers considerations, the court held that “a litigant may not waive” a Stern objection. Id., at 773. Turning to the merits of Sharif’s contentions, the Seventh Circuit agreed with the Bankruptcy Court’s resolution of counts I–IV of Wellness’ adversary complaint. It further concluded, however, that count V of the complaint alleged a so-called “Stern claim,” that is, “a claim designated for final adjudication in the bankruptcy court as a statutory matter, but prohibited from proceeding in that way as a constitutional matter.” Executive Benefits, 573 U. S., at ___ (slip op., at 4). The Seventh Circuit therefore ruled that the Bankruptcy Court lacked constitutional authority to enter final judgment on count V.[6] We granted certiorari, 573 U. S. ___ (2014), and now reverse the judgment of the Seventh Circuit.[7] II Our precedents make clear that litigants may validly consent to adjudication by bankruptcy courts. A Adjudication by consent is nothing new. Indeed, “[d]uring the early years of the Republic, federal courts, with the consent of the litigants, regularly referred adjudication of entire disputes to non-Article III referees, masters, or arbitrators, for entry of final judgment in accordance with the referee’s report.” Brubaker, The Constitutionality of Litigant Consent to Non-Article III Bankruptcy Adjudications, 32 Bkrtcy. L. Letter No. 12, p. 6 (Dec. 2012); see, e.g., Thornton v. Carson, 7 Cranch 596, 597 (1813) (affirming damages awards in two actions that “were referred, by consent under a rule of Court to arbitrators”); Heckers v. Fowler, 2 Wall. 123, 131 (1865) (observing that the “[p]ractice of referring pending actions under a rule of court, by consent of parties, was well known at common law,” and “is now universally regarded . . . as the proper foundation of judgment”); Newcomb v. Wood, 97 U. S. 581, 583 (1878) (recognizing “[t]he power of a court of justice, with the consent of the parties, to appoint arbitrators and refer a case pending before it”). The foundational case in the modern era is Commodity Futures Trading Comm’n v. Schor, 478 U. S. 833 (1986) . The Commodity Futures Trading Commission (CFTC), which Congress had authorized to hear customer complaints against commodities brokers, issued a regulation allowing itself to hear state-law counterclaims as well. William Schor filed a complaint with the CFTC against his broker, and the broker, which had previously filed claims against Schor in federal court, refiled them as counterclaims in the CFTC proceeding. The CFTC ruled against Schor on the counterclaims. This Court upheld that ruling against both statutory and constitutional challenges. On the constitutional question (the one relevant here) the Court began by holding that Schor had “waived any right he may have possessed to the full trial of [the broker’s] counterclaim before an Article III court.” Id., at 849. The Court then explained why this waiver legitimated the CFTC’s exercise of authority: “[A]s a personal right, Article III’s guarantee of an impartial and independent federal adjudication is subject to waiver, just as are other per-sonal constitutional rights”—such as the right to a jury—“that dictate the procedures by which civil and criminal matters must be tried.” Id., at 848–849. The Court went on to state that a litigant’s waiver of his “personal right” to an Article III court is not always dispositive because Article III “not only preserves to litigants their interest in an impartial and independent federal adjudication of claims . . . , but also serves as ‘an inseparable element of the constitutional system of checks and balances.’ . . . To the extent that this structural principle is implicated in a given case”—but only to that extent—“the parties cannot by consent cure the constitutional difficulty . . . .” Id., at 850–851. Leaning heavily on the importance of Schor’s consent, the Court found no structural concern implicated by the CFTC’s adjudication of the counterclaims against him. While “Congress gave the CFTC the authority to adjudicate such matters,” the Court wrote, “the decision to invoke this forum is left entirely to the parties and the power of the federal judiciary to take jurisdiction of these matters is unaffected. In such circumstances, separation of powers concerns are diminished, for it seems self-evident that just as Congress may encourage parties to settle a dispute out of court or resort to arbitration without impermissible incursions on the separation of powers, Congress may make available a quasi-judicial mechanism through which willing parties may, at their option, elect to resolve their differences.” Id., at 855. The option for parties to submit their disputes to a non-Article III adjudicator was at most a “de minimis” infringement on the prerogative of the federal courts. Id., at 856. A few years after Schor, the Court decided a pair of cases—Gomez v. United States, 490 U. S. 858 (1989) , and Peretz v. United States, 501 U. S. 923 (1991) —that reiterated the importance of consent to the constitutional analysis. Both cases concerned whether the Federal Magistrates Act authorized magistrate judges to preside over jury selection in a felony trial;[8] the difference was that Peretz consented to the practice while Gomez did not. That difference was dispositive. In Gomez, the Court interpreted the statute as not allowing magistrate judges to supervise voir dire without consent, emphasizing the constitutional concerns that might otherwise arise. See 490 U. S., at 864. In Peretz, the Court upheld the Magistrate Judge’s action, stating that “the defendant’s consent significantly changes the constitutional analysis.” 501 U. S., at 932. The Court concluded that allowing a magistrate judge to supervise jury selection—with consent—does not violate Article III, explaining that “litigants may waive their personal right to have an Article III judge preside over a civil trial,” id., at 936 (citing Schor, 478 U. S., at 848), and that “[t]he most basic rights of criminal defendants are similarly subject to waiver,” 501 U. S., at 936. And “[e]ven assuming that a litigant may not waive structural protections provided by Article III,” the Court found “no such structural protections . . . implicated by” a magistrate judge’s supervision of voir dire: “Magistrates are appointed and subject to removal by Article III judges. The ‘ultimate decision’ whether to invoke the magistrate’s assistance is made by the district court, subject to veto by the parties. The decision whether to empanel the jury whose selection a magistrate has supervised also remains entirely with the district court. Because ‘the entire process takes place under the district court’s total control and jurisdiction,’ there is no danger that use of the magistrate involves a ‘congressional attemp[t] “to transfer jurisdiction [to non-Article III tribunals] for the purpose of emasculating” constitutional courts.’ ” Id., at 937 (citations omitted; alteration in original).[9] The lesson of Schor, Peretz, and the history that preceded them is plain: The entitlement to an Article III adjudicator is “a personal right” and thus ordinarily “subject to waiver,” Schor, 478 U. S., at 848. Article III also serves a structural purpose, “barring congressional attempts ‘to transfer jurisdiction [to non-Article III tribunals] for the purpose of emasculating’ constitutional courts and thereby prevent[ing] ‘the encroachment or aggrandizement of one branch at the expense of the other.’ ” Id., at 850 (citations omitted). But allowing Article I adjudicators to decide claims submitted to them by consent does not offend the separation of powers so long as Article III courts retain supervisory authority over the process. B The question here, then, is whether allowing bankruptcy courts to decide Stern claims by consent would “imper-missibly threate[n] the institutional integrity of the Judicial Branch.” Schor, 478 U. S., at 851. And that question must be decided not by “formalistic and unbending rules,” but “with an eye to the practical effect that the” practice “will have on the constitutionally assigned role of the federal judiciary.” Ibid.; see Thomas v. Union Carbide Agricultural Products Co., 473 U. S. 568, 587 (1985) (“[P]ractical attention to substance rather than doctrinaire reliance on formal categories should inform application of Article III”). The Court must weigh “the extent to which the essential attributes of judicial power are reserved to Article III courts, and, con-versely, the extent to which the non-Article III forum exer-cises the range of jurisdiction and powers normally vested only in Article III courts, the origins and importance of the right to be adjudicated, and the concerns that drove Congress to depart from the requirements of Article III.” Schor, 478 U. S., at 851 (internal quotation marks omitted). Applying these factors, we conclude that allowing bankruptcy litigants to waive the right to Article III adjudication of Stern claims does not usurp the constitutional prerogatives of Article III courts. Bankruptcy judges, like magistrate judges, “are appointed and subject to removal by Article III judges,” Peretz, 501 U. S., at 937; see 28 U. S. C. §§152(a)(1), (e). They “serve as judicial officers of the United States district court,” §151, and collectively “constitute a unit of the district court” for that district, §152(a)(1). Just as “[t]he ‘ultimate decision’ whether to invoke [a] magistrate [judge]’s assistance is made by the district court,” Peretz, 501 U. S., at 937, bankruptcy courts hear matters solely on a district court’s reference, §157(a), which the district court may withdraw sua sponte or at the request of a party, §157(d). “[S]eparation of powers concerns are diminished” when, as here, “the decision to invoke [a non-Article III] forum is left entirely to the parties and the power of the federal judiciary to take jurisdiction” remains in place. Schor, 478 U. S., at 855. Furthermore, like the CFTC in Schor, bankruptcy courts possess no free-floating authority to decide claims traditionally heard by Article III courts. Their ability to resolve such matters is limited to “a narrow class of common law claims as an incident to the [bankruptcy courts’] primary, and unchallenged, adjudicative function.” Id., at 854. “In such circumstances, the magnitude of any intrusion on the Judicial Branch can only be termed de minimis.” Id., at 856. Finally, there is no indication that Congress gave bankruptcy courts the ability to decide Stern claims in an effort to aggrandize itself or humble the Judiciary. As in Peretz, “[b]ecause ‘the entire process takes place under the district court’s total control and jurisdiction,’ there is no danger that use of the [bankruptcy court] involves a ‘congres-sional attemp[t] “to transfer jurisdiction [to non-Article III tribunals] for the purpose of emasculating” constitutional courts.’ ” 501 U. S., at 937 (citation omitted); see also Schor, 478 U. S., at 855 (allowing CFTC’s adjudication of counterclaims because of “the degree of judicial control saved to the federal courts, as well as the congressional purpose behind the jurisdictional delegation, the demonstrated need for the delegation, and the limited nature of the delegation” (citation omitted)); Pacemaker Diagnostic Clinic of America, Inc. v. Instromedix, Inc., 725 F. 2d 537, 544 (CA9 1984) (en banc) (Kennedy, J.) (magistrate judges may adjudicate civil cases by consent because the Federal Magistrates Act “invests the Article III judiciary with extensive administrative control over the management, composition, and operation of the magistrate system”).[10] Congress could choose to rest the full share of the Judiciary’s labor on the shoulders of Article III judges. But doing so would require a substantial increase in the number of district judgeships. Instead, Congress has supplemented the capacity of district courts through the able assistance of bankruptcy judges. So long as those judges are subject to control by the Article III courts, their work poses no threat to the separation of powers. C Our recent decision in Stern, on which Sharif and the principal dissent rely heavily, does not compel a different result. That is because Stern—like its predecessor, Northern Pipeline—turned on the fact that the litigant “did not truly consent to” resolution of the claim against it in a non-Article III forum. 564 U. S., at ___ (slip op., at 27). To understand Stern, it is necessary to first understand Northern Pipeline. There, the Court considered whether bankruptcy judges “could ‘constitutionally be vested with jurisdiction to decide [a] state-law contract claim’ against an entity that was not otherwise part of the bankruptcy proceedings.” 564 U. S., at ___ (slip op., at 19). In answering that question in the negative, both the plurality and then-Justice Rehnquist, concurring in the judgment, noted that the entity in question did not consent to the bankruptcy court’s adjudication of the claim. See 458 U. S., at 80, n. 31 (plurality opinion); id., at 91 (opinion of Rehnquist, J.). The Court confirmed in two later cases that Northern Pipeline turned on the lack of consent. See Schor, 478 U. S., at 849 (“[I]n Northern Pipeline, . . . the absence of consent to an initial adjudication before a non-Article III tribunal was relied on as a significant factor in determining that Article III forbade such adjudication”); Thomas, 473 U. S., at 584. Stern presented the same scenario. The majority cited the dissent’s observation that Northern Pipeline “establish[ed] only that Congress may not vest in a non-Article III court the power to adjudicate, render final judgment, and issue binding orders in a traditional contract action arising under state law, without consent of the litigants, and subject only to ordinary appellate review,” 564 U. S., at ___ (slip op., at 28–29) (emphasis added; internal quotation marks omitted). To which the majority responded, “Just so: Substitute ‘tort’ for ‘contract,’ and that statement directly covers this case.” Id., at ___ (slip op., at 29); see also id., at ___ (slip op., at 27) (defendant litigated in the Bankruptcy Court because he “had nowhere else to go” to pursue his claim). Because Stern was premised on nonconsent to adjudication by the Bankruptcy Court, the “constitutional bar” it announced, see post, at 14 (Roberts, C. J., dissenting), simply does not govern the question whether litigants may validly consent to adjudication by a bankruptcy court. An expansive reading of Stern, moreover, would be inconsistent with the opinion’s own description of its holding. The Court in Stern took pains to note that the question before it was “a ‘narrow’ one,” and that its answer did “not change all that much” about the division of labor between district courts and bankruptcy courts. Id., at ___ (slip op., at 37); see also id., at ___ (slip op., at 38) (stating that Congress had exceeded the limitations of Article III “in one isolated respect”). That could not have been a fair characterization of the decision if it meant that bank-ruptcy judges could no longer exercise their longstanding authority to resolve claims submitted to them by consent. Interpreting Stern to bar consensual adjudications by bankruptcy courts would “meaningfully chang[e] the division of labor” in our judicial system, contra, id., at ___ (slip op., at 37).[11] In sum, the cases in which this Court has found a violation of a litigant’s right to an Article III decisionmaker have involved an objecting defendant forced to litigate involuntarily before a non-Article III court. The Court has never done what Sharif and the principal dissent would have us do—hold that a litigant who has the right to an Article III court may not waive that right through his consent. D The principal dissent warns darkly of the consequences of today’s decision. See post, at 17–20. To hear the principal dissent tell it, the world will end not in fire, or ice, but in a bankruptcy court. The response to these ominous predictions is the same now as it was when Justice Brennan, dissenting in Schor, first made them nearly 30 years ago: “This is not to say, of course, that if Congress created a phalanx of non-Article III tribunals equipped to handle the entire business of the Article III courts without any Article III supervision or control and without evidence of valid and specific legislative necessities, the fact that the parties had the election to proceed in their forum of choice would necessarily save the scheme from constitutional attack. But this case obviously bears no resemblance to such a sce-nario . . . .” 478 U. S., at 855 (citations omitted). Adjudication based on litigant consent has been a consistent feature of the federal court system since its inception. Reaffirming that unremarkable fact, we are confident, poses no great threat to anyone’s birthrights, constitutional or otherwise. III Sharif contends that to the extent litigants may validly consent to adjudication by a bankruptcy court, such consent must be express. We disagree. Nothing in the Constitution requires that consent to adjudication by a bankruptcy court be express. Nor does the relevant statute, 28 U. S. C. §157, mandate express consent; it states only that a bankruptcy court must obtain “the consent”—consent simpliciter—“of all parties to the proceeding” before hearing and determining a non-core claim. §157(c)(2). And a requirement of express consent would be in great tension with our decision in Roell v. Withrow, 538 U. S. 580 (2003) . That case concerned the interpretation of §636(c), which authorizes magistrate judges to “conduct any or all proceedings in a jury or nonjury civil matter and order the entry of judgment in the case,” with “the consent of the parties.”[12] The specific question in Roell was whether, as a statutory matter, the “consent” required by §636(c) had to be express. The dissent argued that “[r]eading §636(c)(1) to require express consent not only is more consistent with the text of the statute, but also” avoids constitutional concerns by “ensur[ing] that the parties knowingly and voluntarily waive their right to an Article III judge.” 538 U. S., at 595 (opinion of Thomas, J.). But the majority—thus placed on notice of the constitutional concern—was untroubled by it, opining that “the Article III right is substantially honored” by permitting waiver based on “actions rather than words.” Id., at 589, 590. The implied consent standard articulated in Roell supplies the appropriate rule for adjudications by bankruptcy courts under §157. Applied in the bankruptcy context, that standard possesses the same pragmatic virtues—increasing judicial efficiency and checking gamesmanship—that motivated our adoption of it for consent-based adjudications by magistrate judges. See id., at 590. It bears emphasizing, however, that a litigant’s consent—whether express or implied—must still be knowing and voluntary. Roell makes clear that the key inquiry is whether “the litigant or counsel was made aware of the need for consent and the right to refuse it, and still voluntarily appeared to try the case” before the non-Article III adjudicator. Ibid.; see also id., at 588, n. 5 (“notification of the right to refuse” adjudication by a non-Article III court “is a prerequisite to any inference of consent”).[13] IV It would be possible to resolve this case by determining whether Sharif in fact consented to the Bankruptcy Court’s adjudication of count V of Wellness’ adversary complaint. But reaching that determination would require a deeply factbound analysis of the procedural history unique to this protracted litigation. Our resolution of the consent question—unlike the antecedent constitutional question—would provide little guidance to litigants or the lower courts. Thus, consistent with our role as “a court of review, not of first view,” Nautilus, Inc. v. Biosig Instruments, Inc., 572 U. S. ___, ___ (2014) (slip op., at 14) (internal quotation marks omitted), we leave it to the Seventh Circuit to decide on remand whether Sharif’s actions evinced the requisite knowing and voluntary consent, and also whether, as Wellness contends, Sharif forfeited his Stern argument below. * * * The Court holds that Article III permits bankruptcy courts to decide Stern claims submitted to them by consent. The judgment of the United States Court of Appeals for the Seventh Circuit is therefore reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Congress has authorized 179 circuit judgeships and 677 district judgeships, a total of 856. United States Courts, Status of Article III Judgeships, http://www.uscourts.gov/Statistics/JudicialBusiness/2014/status-article-iii-judgeships.aspx (all Internet materials as visitedMay 22, 2015, and available in Clerk of Court’s case file).The number of authorized magistrate and bankruptcy judgeships currently stands at 883: 534 full-time magistrate judgeships and349 bankruptcy judgeships. United States Courts, Appointments of Magistrate Judges, http://www.uscourts.gov/Statistics/JudicialBusiness/2014/appointments-magistrate-judges.aspx; United States Courts, Status of Bankruptcy Judgeships, http://www.uscourts.gov/Statistics/JudicialBusiness/2014/status-bankruptcy-judgeships.aspx. 2 Between October 1, 2013, and September 30, 2014, for example, litigants filed 963,739 cases in bankruptcy courts—more thandouble the total number filed in district and circuit courts. United States Courts, Judicial Caseload Indicators, http://www.uscourts.gov/Statistics/JudicialBusiness/2014/judicial-caseload-indicators.aspx. 3 Congress appears to have drawn the term “core” from Northern Pipeline’s description of “the restructuring of debtor-creditor relations” as “the core of the federal bankruptcy power.” Northern Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U. S., 50, 71 (1982). 4 Individual petitioners Ralph and Cathy Oats are Wellness’ founders. This opinion refers to all petitioners collectively as “Wellness.” 5 Although the Seventh Circuit referred to Sharif’s failure to raise his Stern argument in a timely manner as a waiver, that court has since clarified that its decision rested on forfeiture. See Peterson v. Somers Dublin Ltd., 729 F. 3d 741, 747 (2013) (“The issue in Wellness International Network was forfeiture rather than waiver”). 6 The Seventh Circuit concluded its opinion by considering the rem-edy for the Bankruptcy Court’s purportedly unconstitutional issuance of a final judgment. The court determined that if count V of Wellness’ complaint raised a core claim, the only statutorily authorized remedy would be for the District Court to withdraw the reference to the Bankruptcy Court and set a new discovery schedule. The Seventh Circuit’s reasoning on this point was rejected by our decision last Term in Executive Benefits, which held that district courts may treat Stern claims like non-core claims and thus are not required to restart proceedings entirely when a bankruptcy court improperly enters final judgment. 7 Because the Court concludes that the Bankruptcy Court could val-idly enter judgment on Wellness’ claim with the parties’ consent, this opinion does not address, and expresses no view on, Wellness’ alternative contention that the Seventh Circuit erred in concluding the claim in count V of its complaint was a Stern claim. 8 In relevant part, the Act provides that district courts may assign magistrate judges certain enumerated duties as well as “such additional duties as are not inconsistent with the Constitution and the laws ofthe United States.” 28 U. S. C. §636(b)(3). 9 Discounting the relevance of Gomez and Peretz, the principal dissent emphasizes that neither case concerned the entry of final judgment by a non-Article III actor. See post, at 16 (opinion of Roberts, C. J.). Here again, the principal dissent’s insistence on formalism leads it astray. As we explained in Peretz, the “responsibility and importance [of] presiding over voir dire at a felony trial” is equivalent to the “supervision of entire civil and misdemeanor trials,” 501 U. S., at 933, tasks in which magistrate judges may “order the entry of judgment” with the parties’ consent, §636(c)(1). 10 The principal dissent accuses us of making Sharif’s consent “ ‘dispositive’ in curing [a] structural separation of powers violation,” con-trary to the holding of Schor. Post, at 16. That argument misapprehends both Schor and the nature of our analysis. What Schor forbids is using consent to excuse an actual violation of Article III. See 478 U. S., at 850–851 (“To the extent that th[e] structural principle [protected by Article III] is implicated in a given case, the parties cannot by consent cure the constitutional difficulty . . .” (emphasis added)). But Schor confirms that consent remains highly relevant when determining, as we do here, whether a particular adjudication in fact raises constitutional concerns. See id., at 855 (“separation of powers concerns are diminished” when “the decision to invoke [a non-Article III] forum is left entirely to the parties”). Thus, we do not rely on Sharif’s consent to “cur[e]” a violation of Article III. His consent shows, in part, why no such violation has occurred. Cf. Meltzer, Legislative Courts, Legislative Power, and the Constitution, 65 Ind. L. J. 291, 303 (1990) (“[C]onsent provides, if not complete, at least very considerable reason to doubt that the tribunal poses a serious threat to the ideal of federal adjudicatory independence”); Fallon, Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 915, 992 (1988) (when the parties consent, “there is substantial assurance that the agency is not generally behaving arbitrarily or otherwise offending separation-of-powers values. Judicial integrity is not at risk”). 11 In advancing its restrictive view of Stern, the principal dissent ignores the sweeping jurisprudential implications of its position. If, as the principal dissent suggests, consent is irrelevant to the Article III analysis, it is difficult to see how Schor and Peretz were not wrongly decided. But those decisions obviously remain good law. It is the principal dissent’s position that breaks with our precedents. See Plaut v. Spendthrift Farm, Inc., 514 U. S. 211, 231 (1995) (“[T]he proposition that legal defenses based upon doctrines central to the courts’ struc-tural independence can never be waived simply does not accord with our cases”). 12 Consistent with our precedents, the Courts of Appeals have unanimously upheld the constitutionality of §636(c). See Sinclair v. Wainwright, 814 F. 2d 1516, 1519 (CA11 1987); Bell & Beckwith v. United States, 766 F. 2d 910, 912 (CA6 1985); Gairola v. Virginia Dept. of Gen. Servs., 753 F. 2d 1281, 1285 (CA4 1985); D. L. Auld Co. v. Chroma Graphics Corp., 753 F. 2d 1029, 1032 (CA Fed. 1985); United States v. Dobey, 751 F. 2d 1140, 1143 (CA10 1985); Fields v. Washington Metropolitan Area Transit Auth., 743 F. 2d 890, 893 (CADC 1984); Geras v. Lafayette Display Fixtures, Inc., 742 F. 2d 1037, 1045 (CA7 1984); Lehman Bros. Kuhn Loeb Inc. v. Clark Oil & Refining Corp., 739 F. 2d 1313, 1316 (CA8 1984) (en banc); Puryear v. Ede’s Ltd., 731 F. 2d 1153, 1154 (CA5 1984); Goldstein v. Kelleher, 728 F. 2d 32, 36 (CA1 1984); Collins v. Foreman, 729 F. 2d 108, 115–116 (CA2 1984); Pacemaker Diagnostic Clinic, Inc. v. Instromedix, Inc., 725 F. 2d 537, 540 (CA9 1984) (en banc) (Kennedy, J.); Wharton-Thomas v. United States, 721 F. 2d 922, 929–930 (CA3 1983). 13 Even though the Constitution does not require that consent be express, it is good practice for courts to seek express statements of consent or nonconsent, both to ensure irrefutably that any waiver of the right to Article III adjudication is knowing and voluntary and to limit subsequent litigation over the consent issue. Statutes or judicial rules may require express consent where the Constitution does not. Indeed, the Federal Rules of Bankruptcy Procedure already require that pleadings in adversary proceedings before a bankruptcy court “contain a statement that the proceeding is core or non-core and, if non-core, that the pleader does or does not consent to entry of final orders or judgment by the bankruptcy judge.” Fed. Rule Bkrtcy. Proc. 7008 (opening pleadings); see Fed. Rule Bkrtcy. Proc. 7012 (responsive pleadings). The Bankruptcy Court and the parties followed that procedure in this case. See App. 6, 24; supra, at 5–6. And this Court recently submitted to Congress, pursuant to the Rules Enabling Act, proposed amendments to Rules 7008 and 7012 that remove references to the core/non-core distinction and thus require parties in all bankruptcy proceedings to state expressly whether they consent to the bankruptcy court’s entry of judgment. Report of the Judicial Conference, Committee on Rules of Practice and Procedure, pp. 6, 7, 9. (Sept. 2013).

574.US.265

Petitioner Whitfield, fleeing a botched bank robbery, entered 79-year-old Mary Parnell’s home and guided a terrified Parnell from a hallway to a room a few feet away, where she suffered a fatal heart attack. He was convicted of, among other things, violating 18 U. S. C. §2113(e), which establishes enhanced penalties for anyone who “forces any person to accompany him without the consent of such person” in the course of committing or fleeing from a bank robbery. On appeal, the Fourth Circuit held that the movement Whitfield required Parnell to make satisfied the forced-accompaniment requirement, rejecting his argument that §2113(e) requires “substantial” movement. Held: A bank robber “forces [a] person to accompany him,” for purposes of §2113(e), when he forces that person to go somewhere with him, even if the movement occurs entirely within a single building or over a short distance, as was the case here. At the time the forced-accompaniment provision was enacted, just as today, to “accompany” someone meant to “go with” him. The word does not, as Whitfield contends, connote movement over a substantial distance. Accompaniment requires movement that would normally be described as from one place to another. Here, Whitfield forced Parnell to accompany him for at least several feet, from one room to another, and that surely sufficed. The severity of the penalties for a forced-accompaniment conviction—a mandatory minimum of 10 years, and a maximum of life imprisonment—does not militate against this interpretation, for the danger of a forced accompaniment does not vary depending on the distance traversed. This reading also does not make any other part of §2113’s graduated penalty scheme superfluous. Pp. 2–5. 548 Fed. Appx. 70, affirmed. Scalia, J., delivered the opinion for a unanimous Court.

Federal law establishes enhanced penalties for anyone who “forces any person to accompany him” in the course of committing or fleeing from a bank robbery.18 U. S. C. §2113(e). We consider whether this provision applies when a bank robber forces someone to move with him over a short distance.I Petitioner Larry Whitfield, fleeing police after a botched bank robbery, entered the home of 79-year-old Mary Parnell through an unlocked door. Once inside, he encountered a terrified Parnell and guided her from the hallway to a computer room (which Whitfield estimates was between four and nine feet away. Brief for Petitioner 5). There, Parnell suffered a fatal heart attack. Whitfield fled, and was found hiding nearby. A grand jury indicted Whitfield for, among other things, violating §2113(e) by forcing Parnell to accompany him in the course of avoiding apprehension for a bank robbery. That section provides:“Whoever, in committing any offense defined in this section, or in avoiding or attempting to avoid apprehension for the commission of such offense . . . forces any person to accompany him without the consent of such person, shall be imprisoned not less than ten years, or if death results shall be punished by death or life imprisonment.” (Emphasis added.)Whitfield pleaded not guilty to the offense but a jury convicted him. On appeal, Whitfield challenged the sufficiency of the evidence, arguing that §2113(e) requires “substantial” movement, and that his movement with Parnell did not qualify. Brief for Appellant in No. 10–5217 (CA4),pp. 50–52. The Fourth Circuit disagreed, holding that, “[a]lthough Whitfield required Mrs. Parnell to accompany him for only a short distance within her own home, and for a brief period, no more is required to prove that a forced accompaniment occurred.” 695 F. 3d 288, 311 (2012). After further proceedings in the District Court and Court of Appeals, 548 Fed. Appx. 70 (2013), we granted certiorari, 573 U. S. ___ (2014).II Congress enacted the forced-accompaniment provision in 1934 after “an outbreak of bank robberies committed by John Dillinger and others.” Carter v. United States,530 U. S. 255,280 (2000) (Ginsburg, J., dissenting). Section 2113 has been amended frequently, but the relevant phrase—“forces any person to accompany him without the consent of such person”—has remained unchanged, and so presumptively retains its original meaning. Vermont Agency of Natural Resources v. United States ex rel. Stevens,529 U. S. 765,783, n.12 (2000). In 1934, just as today, to “accompany” someone meant to “go with” him. See Oxford English Dictionary 60 (1st ed. 1933) (defining “accompany” as: “To go in company with, to go along with”). The word does not, as Whitfield contends, connote movement over a substantial distance. It was, and still is, perfectly natural to speak of accompanying someone over a relatively short distance, for example: from one area within a bank “to the vault”;[1] “to the altar” at a wedding;[2] “up the stairway”;[3] or into, out of, or across a room.[4] English literature is replete with examples. See, e.g., C. Dickens, David Copperfield 529 (Modern Library ed. 2000) (Uriah “accompanied me into Mr. Wickfield’s room”); J. Austen, Pride and Prejudice 182 (Greenwich ed. 1982) (Elizabeth “accompanied her out of the room”). It is true enough that accompaniment does not embrace minimal movement—for example, the movement of a bank teller’s feet when the robber grabs her arm. It must constitute movement that would normally be described as from one place to another, even if only from one spot withina room or outdoors to a different one. Here, Whitfield forced Parnell to accompany him for at least several feet, from one room to another. That surely sufficed. In an attempt to support his position that “accompany” should be read to mean “accompany over a substantial distance,” Whitfield observes that a forced-accompaniment conviction carries severe penalties: a mandatory minimum sentence of 10 years, and a maximum sentence of life imprisonment. In 1934, a forced-accompaniment conviction could even be punished with death. Act of May 18, 1934, ch. 304, §3,48Stat.783. The severity of these sentences, Whitfield says, militates against interpreting subsection (e) to capture forced accompaniment occurring over a small distance. But it does not seem to us that the danger of a forced accompaniment varies with the distance traversed. Consider, for example, a hostage-taker’s movement of one of his victims a short distance to a window, where she would be exposed to police fire; or his use of the victim as a human shield as he approaches the door. And even if we thought otherwise, we would have no authority to add a limitation the statute plainly does not contain. The Congress that wrote this provision may well have had most prominently in mind John Dillinger’s driving off with hostages, but it enacted a provision which goes wellbeyond that. It is simply not in accord with English usage to give “accompany” a meaning that covers only large distances. Whitfield also contends that “accompany” must be read narrowly in light of §2113’s graduated penalty scheme. That scheme prescribes: (1) a 20-year maximum sentence for bank robbers who use “force and violence” or “intimidation,” §2113(a), (2) a 25-year maximum sentence for those who “assaul[t]” or “pu[t] in jeopardy the life of” another “by the use of a dangerous weapon or device,” §2113(d), and (3) a minimum sentence of 10 years, and a maximum sentence of life, for forced accompaniment, §2113(e). According to Whitfield, bank robbers almost always “exert some control over the movement of the bank’s employ-ees.” Brief for Petitioner 22. Therefore, he says, unless “accompany” is limited to forced movement over sub-stantial distances, nearly all §2113 violations will be punishable under subsection (e), making subsections (a) and(d) pointless. We disagree. Even if Whitfield is right that bank robbers always “exert some control” over others, it does not follow that they always force others to accompany them somewhere—that is, to go somewhere with them. As we have no reason to think that to be the case, and because subsections (a), (d), and (e) all cover distinct conduct, our interpretation of “accompany” does not make any part of §2113 superfluous.* * * We hold that a bank robber “forces [a] person to accompany him,” for purposes of §2113(e), when he forces that person to go somewhere with him, even if the movement occurs entirely within a single building or over a short distance. Defined in this manner, Whitfield forced Parnell to “accompany him.” §2113(e). The judgment of the Fourth Circuit is affirmed.It is so ordered.Notes1 Addison State Bank Robbed, DuPage County Register, Apr. 6, 1928.2 Salmon-Peters Marriage Announcement, N. Y. Times, Dec. 7, 1930, p. N7.3 Woman’s Version of Norris Killing Being Challenged, Washington Post, Jan. 20, 1927, p. 3.4 See, e.g., Davis v. Potter, 51 Idaho 81, 84, 2 P. 2d 318, 319 (1931) (“[T]he patient, still under the influence of the anesthetic, was wheeled to her room, accompanied by Dr. Sturges”); Union & Planters’ Bank & Trust Co. v. Rylee, 130 Miss. 892, 906, 94 So. 796, 798 (1923) (“This witness further testified that when J. N. Rylee left the decedent’s room she accompanied him into the hall”); Vernonia School Dist. 47J v. Acton,515 U. S. 646,650 (1995) (“The student then enters an empty locker room accompanied by an adult monitor of the same sex”).

575.US.2014_13-1499

Florida is one of 39 States where voters elect judges at the polls. To promote public confidence in the integrity of the judiciary, the Florida Supreme Court adopted Canon 7C(1) of its Code of Judicial Conduct, which provides that judicial candidates “shall not personally solicit campaign funds . . . but may establish committees of responsible persons” to raise money for election campaigns. Petitioner Lanell Williams-Yulee (Yulee) mailed and posted online a letter soliciting financial contributions to her campaign for judicial office. The Florida Bar disciplined her for violating a Florida Bar Rule requiring candidates to comply with Canon 7C(1), but Yulee contended that the First Amendment protects a judicial candidate’s right to personally solicit campaign funds in an election. The Florida Supreme Court upheld the disciplinary sanctions, concluding that Canon 7C(1) is narrowly tailored to serve the State’s compelling interest. Held: The judgment is affirmed. 138 So. 3d 379, affirmed. Chief Justice Roberts delivered the opinion of the Court, except as to Part II, concluding that the First Amendment permits Canon 7C(1)’s ban on the personal solicitation of campaign funds by judicial candidates. Pp. 8–22. (a) Florida’s interest in preserving public confidence in the integrity of its judiciary is compelling. The State may conclude that judges, charged with exercising strict neutrality and independence, cannot supplicate campaign donors without diminishing public confidence in judicial integrity. Simply put, the public may lack confidence in a judge’s ability to administer justice without fear or favor if he comes to office by asking for favors. This Court’s precedents have recognized the “vital state interest” in safeguarding “ ‘public confidence in the fairness and integrity of the nation’s elected judges,’ ” Caperton v. A. T. Massey Coal Co., 556 U. S. 868 . Unlike the legislature or the executive, the judiciary “has no influence over either the sword or the purse,” Federalist No. 78, p. 465 (A. Hamilton), so its authority depends in large measure on the public’s willingness to respect and follow its decisions. Public perception of judicial integrity is accordingly “ ‘a state interest of the highest order.’ ” 556 U. S., at 889. A State’s interest in preserving public confidence in the integrity of its judiciary extends beyond its interest in preventing the appearance of corruption in legislative and executive elections, because a judge’s role differs from that of a politician. Republican Party of Minn. v. White, 536 U. S. 765 . Unlike a politician, who is expected to be appropriately responsive to the preferences of supporters, a judge in deciding cases may not follow the preferences of his supporters or provide any special consideration to his campaign donors. As in White, therefore, precedents applying the First Amendment to political elections have little bearing on the issues here. The vast majority of elected judges in States allowing personal solicitation serve with fairness and honor, but in the eyes of the public, a judicial candidate’s personal solicitation could result (even unknowingly) in “a possible temptation . . . which might lead him not to hold the balance nice, clear and true.” Tumey v. Ohio, 273 U. S. 510 . That risk is especially pronounced where most donors are lawyers and litigants who may appear before the judge they are supporting. In short, it is the regrettable but unavoidable appearance that judges who personally ask for money may diminish their integrity that prompted the Supreme Court of Florida and most other States to sever the direct link between judicial candidates and campaign contributors. Pp. 9–12. (b) Canon 7C(1) raises no fatal underinclusivity concerns. The solicitation ban aims squarely at the conduct most likely to undermine public confidence in the integrity of the judiciary: personal requests for money by judges and judicial candidates. The Canon applies evenhandedly to all judges and judicial candidates, regardless of viewpoint or means of solicitation. And unlike some laws that have been found impermissibly underinclusive, Canon 7C(1) is not riddled with exceptions. Yulee relies heavily on the provision of Canon 7C(1) that allows solicitation by a candidate’s campaign committee. But Florida, along with most other States, has reasonably concluded that solicitation by the candidate personally creates a categorically different and more severe risk of undermining public confidence than does solicitation by a campaign committee. When the judicial candidate himself asks for money, the stakes are higher for all involved. A judicial candidate asking for money places his name and reputation behind the request, and the solicited individual knows that the same person who signed the fundraising letter might one day sign the judgment. This dynamic inevitably creates pressure for the recipient to comply, in a way that solicitation by a third party does not. Just as inevitably, the personal involvement of the candidate in the solicitation creates the public appearance that the candidate will remember who says yes, and who says no. However similar the two solicitations may be in substance, a State may conclude that they present markedly different appearances to the public. Permitting a judicial candidate to write thank you notes to campaign donors likewise does not detract from the State’s interest in preserving public confidence in the integrity of the judiciary. The State’s compelling interest is implicated most directly by the candidate’s personal solicitation itself. A failure to ban thank you notes for contributions not solicited by the candidate does not undercut the Bar’s rationale. In addition, the State has a good reason for allowing candidates to write thank you notes and raise money through committees. These accommodations reflect Florida’s effort to respect the First Amendment interests of candidates and their contributors—to resolve the “fundamental tension between the ideal character of the judicial office and the real world of electoral politics.” Chisom v. Roemer, 501 U. S. 380 . The State should not be punished for leaving open more, rather than fewer, avenues of expression, especially when there is no indication of a pretextual motive for the selective restriction of speech. Pp. 12–16. (c) Canon 7C(1) is also not overinclusive. By any measure, it restricts a narrow slice of speech. It leaves judicial candidates free to discuss any issue with any person at any time; to write letters, give speeches, and put up billboards; to contact potential supporters in person, on the phone, or online; and to promote their campaigns through the media. Though they cannot ask for money, they can direct their campaign committees to do so. Yulee concedes that Canon 7C(1) is valid in numerous applications, but she contends that the Canon cannot constitutionally be applied to her chosen form of solicitation: a letter posted online and distributed via mass mailing. This argument misperceives the breadth of the compelling interest underlying Canon 7C(1). Florida has reasonably determined that personal appeals for money by a judicial candidate inherently create an appearance of impropriety that may cause the public to lose confidence in the integrity of the judiciary. That interest may be implicated to varying degrees in particular contexts, but the interest remains whenever the public perceives the judge person-ally asking for money. Canon 7C(1) must be narrowly tailored, not “perfectly tailored.” Burson v. Freeman, 504 U. S. 191 . The First Amendment does not confine a State to addressing evils in their most acute form. Here, Florida has concluded that all personal solicitations by judicial candidates create a public appearance that undermines confidence in the integrity of the judiciary; banning all personal solicitations by judicial candidates is narrowly tailored to address that concern. Yulee errs in contending that Florida can accomplish its compelling interest through recusal rules and campaign contribution limits. A rule requiring recusal in every case in which a lawyer or litigant made a campaign contribution would disable many jurisdictions, and a flood of postelection recusal motions could exacerbate the very appearance problem the State is trying to solve. As for contribution limits, Florida already applies them to judicial elections, and this Court has never held that adopting such limits precludes a State from pursuing its compelling interests through additional means. The desirability of judicial elections is a question that has sparked disagreement for more than 200 years, but it is not the Court’s place to resolve that enduring debate. The Court’s limited task is to apply the Constitution to the question presented in this case. Judicial candidates have a First Amendment right to speak in support of their campaigns. States have a compelling interest in preserving public confidence in their judiciaries. When the State adopts a narrowly tailored restriction like the one at issue here, those principles do not conflict. A State’s decision to elect judges does not compel it to compromise public confidence in their integrity. Pp. 16–22. Roberts, C. J., delivered the opinion of the Court, except as to Part II. Breyer, Sotomayor, and Kagan, JJ., joined that opinion in full, and Ginsburg, J., joined except as to Part II. Breyer, J., filed a concurring opinion. Ginsburg, J., filed an opinion concurring in part and concurring in the judgment, in which Breyer, J., joined as to Part II. Scalia, J., filed a dissenting opinion, in which Thomas, J., joined. Kennedy, J., and Alito, J., filed dissenting opinions.

. The “closely drawn” standard is a poor fit for this case. The Court adopted that test in Buckley to address a claim that campaign contribution limits violated a contributor’s “freedom of political association.” Id., at 24–25. Here, Yulee does not claim that Canon 7C(1) violates her right to free association; she argues that it violates her right to free speech. And the Florida Bar can hardly dispute that the Canon infringes Yulee’s freedom to discuss candidates and public issues—namely, herself and her qualifications to be a judge. The Bar’s call to import the “closely drawn” test from the contribution limit context into a case about solicitation therefore has little avail. As several of the Bar’s amici note, we applied the “closely drawn” test to solicitation restrictions in McConnell v. Federal Election Comm’n,540 U. S. 93,136 (2003), overruled in part by Citizens United v. Federal Election Comm’n,558 U. S. 310 (2010). But the Court in that case determined that the solicitation restrictions operated primarily to prevent circumvention of the contribution limits, which were the subject of the “closely drawn” test in the first place. 540 U. S., at 138–139. McConnell offers no help to the Bar here, because Florida did not adopt Canon 7C(1) as an anticircumvention measure. In sum, we hold today what we assumed in White: A State may restrict the speech of a judicial candidate only if the restriction is narrowly tailored to serve a compelling interest.III The Florida Bar faces a demanding task in defending Canon 7C(1) against Yulee’s First Amendment challenge. We have emphasized that “it is the rare case” in which a State demonstrates that a speech restriction is narrowly tailored to serve a compelling interest. Burson v. Freeman,504 U. S. 191,211 (1992) (plurality opinion). But those cases do arise. See ibid.; Holder v. Humanitarian Law Project,561 U. S. 1–39 (2010); McConnell, 540 U. S., at 314 (opinion of Kennedy, J.); cf. Adarand Constructors, Inc. v. Peña,515 U. S. 200,237 (1995) (“we wish to dispel the notion that strict scrutiny is ‘strict in theory, but fatal in fact’ ”). Here, Canon 7C(1) advances the State’s compelling interest in preserving public confidence in the integrity of the judiciary, and it does so through means narrowly tailored to avoid unnecessarily abridging speech. This is therefore one of the rare cases in which a speech restriction withstands strict scrutiny.A The Florida Supreme Court adopted Canon 7C(1) to promote the State’s interests in “protecting the integrity of the judiciary” and “maintaining the public’s confidence in an impartial judiciary.” 138 So. 3d, at 385. The way the Canon advances those interests is intuitive: Judges, charged with exercising strict neutrality and independence, cannot supplicate campaign donors without diminishing public confidence in judicial integrity. This principle dates back at least eight centuries to Magna Carta, which proclaimed, “To no one will we sell, to no one will we refuse or delay, right or justice.” Cl. 40 (1215), in W. McKechnie, Magna Carta, A Commentary on the Great Charter of King John 395 (2d ed. 1914). The same concept underlies the common law judicial oath, which binds a judge to “do right to all manner of people . . . without fear or favour, affection or ill-will,” 10 Encyclopaedia of the Laws of England 105 (2d ed. 1908), and the oath that each of us took to “administer justice without respect to persons, and do equal right to the poor and to the rich,”28 U. S. C. §453. Simply put, Florida and most other States have concluded that the public may lack confidence in a judge’s ability to administer justice without fear or favor if he comes to office by asking for favors. The interest served by Canon 7C(1) has firm support in our precedents. We have recognized the “vital state interest” in safeguarding “public confidence in the fairness and integrity of the nation’s elected judges.” Caperton v. A. T. Massey Coal Co.,556 U. S. 868,889 (2009) (internal quotation marks omitted). The importance of public confidence in the integrity of judges stems from the place ofthe judiciary in the government. Unlike the executive or the legislature, the judiciary “has no influence over either the sword or the purse; . . . neither force nor will but merely judgment.” The Federalist No. 78, p. 465 (C. Rossiter ed. 1961) (A. Hamilton) (capitalization altered). The judiciary’s authority therefore depends in large measure on the public’s willingness to respect and follow its decisions. As Justice Frankfurter once put it for the Court, “justice must satisfy the appearance of justice.” Offutt v. United States,348 U. S. 11,14 (1954). It follows that public perception of judicial integrity is “a state interest of the highest order.” Caperton, 556 U. S., at 889 (quoting White, 536 U. S., at 793 (Kennedy, J., concurring)). The principal dissent observes that bans on judicial candidate solicitation lack a lengthy historical pedigree. Post, at 1–2 (opinion of Scalia, J.). We do not dispute that fact, but it has no relevance here. As the precedent cited by the principal dissent demonstrates, a history and tradition of regulation are important factors in determining whether to recognize “new categories of unprotected speech.” Brown v. Entertainment Merchants Assn., 564 U. S. ___, ___ (2011) (slip op., at 3); see post, at 1. But nobody argues that solicitation of campaign funds by judicial candidates is a category of unprotected speech. As explained above, the First Amendment fully applies to Yulee’s speech. The question is instead whether that Amendment permits the particular regulation of speech at issue here. The parties devote considerable attention to our cases analyzing campaign finance restrictions in political elections. But a State’s interest in preserving public confidence in the integrity of its judiciary extends beyond its interest in preventing the appearance of corruption in legislative and executive elections. As we explained in White, States may regulate judicial elections differently than they regulate political elections, because the role of judges differs from the role of politicians. 536 U. S., at 783; id., at 805 (Ginsburg, J., dissenting). Politicians are expected to be appropriately responsive to the preferences of their supporters. Indeed, such “responsiveness is key to the very concept of self-governance through elected officials.” McCutcheon v. Federal Election Comm’n, 572 U. S. ___, ___ (2014) (plurality opinion) (slip op., at 39). The same is not true of judges. In deciding cases, a judge is not to follow the preferences of his supporters, or provide any special consideration to his campaign donors. A judge instead must “observe the utmost fairness,” striving to be “perfectly and completely independent, with nothing to influence or controul him but God and his conscience.” Address of John Marshall, in Proceedings and Debates of the Virginia State Convention of 1829–1830, p. 616 (1830). As in White, therefore, our precedents applying the First Amendment to political elections have little bearing on the issues here. The vast majority of elected judges in States that allow personal solicitation serve with fairness and honor. But “[e]ven if judges were able to refrain from favoring donors, the mere possibility that judges’ decisions may be moti-vated by the desire to repay campaign contributions is likely to undermine the public’s confidence in the judiciary.” White, 536 U. S., at 790 (O’Connor, J., concurring). In the eyes of the public, a judge’s personal solicitation could result (even unknowingly) in “a possible temptation . . . which might lead him not to hold the balance nice, clear and true.” Tumey v. Ohio,273 U. S. 510,532 (1927). That risk is especially pronounced because most donors are lawyers and litigants who may appear before the judge they are supporting. See A. Bannon, E. Velasco, L. Casey, & L. Reagan, The New Politics of Judicial Elections: 2011–12, p. 15 (2013). The concept of public confidence in judicial integrity does not easily reduce to precise definition, nor does it lend itself to proof by documentary record. But no one denies that it is genuine and compelling. In short, it is the regrettable but unavoidable appearance that judges who personally ask for money may diminish their integrity that prompted the Supreme Court of Florida and most other States to sever the direct link between judicial candidates and campaign contributors. As the Supreme Court of Oregon explained, “the spectacle of lawyers or potential litigants directly handing over money to judicial candidates should be avoided if the public is to have faith in the impartiality of its judiciary.” In re Fadeley, 310 Ore. 548, 565, 802 P. 2d 31, 41 (1990). Moreover, personal solicitation by a judicial candidate “inevitably places the solicited individuals in a position to fear retaliation if they fail to financially support that candidate.” Simes, 368 Ark., at 585, 247 S. W. 3d, at 882. Potential litigants then fear that “the integrity of the judicial system has been compromised, forcing them to search for an attorney in part based upon the criteria of which attorneys have made the obligatory contributions.” Ibid. A State’s decision to elect its judges does not require it to tolerate these risks. The Florida Bar’s interest is compelling.B Yulee acknowledges the State’s compelling interest in judicial integrity. She argues, however, that the Canon’s failure to restrict other speech equally damaging to judicial integrity and its appearance undercuts the Bar’s position. In particular, she notes that Canon 7C(1) allows a judge’s campaign committee to solicit money, which arguably reduces public confidence in the integrity of the judiciary just as much as a judge’s personal solicitation. Yulee also points out that Florida permits judicial candidates to write thank you notes to campaign donors, which ensures that candidates know who contributes and who does not. It is always somewhat counterintuitive to argue that a law violates the First Amendment by abridging too little speech. We have recognized, however, that underinclusiveness can raise “doubts about whether the government is in fact pursuing the interest it invokes, rather than disfavoring a particular speaker or viewpoint.” Brown, 564 U. S., at ___ (slip op., at 14). In a textbook illustration of that principle, we invalidated a city’s ban on ritual animal sacrifices because the city failed to regulate vast swaths of conduct that similarly diminished its asserted interests in public health and animal welfare. Church of Lukumi Babalu Aye, Inc. v. Hialeah,508 U. S. 520–547 (1993). Underinclusiveness can also reveal that a law does not actually advance a compelling interest. For example, a State’s decision to prohibit newspapers, but not electronic media, from releasing the names of juvenile defendants suggested that the law did not advance its stated purpose of protecting youth privacy. Smith v. Daily Mail Publishing Co.,443 U. S. 97–105 (1979). Although a law’s underinclusivity raises a red flag, the First Amendment imposes no freestanding “underinclusiveness limitation.” R. A. V. v. St. Paul,505 U. S. 377,387 (1992) (internal quotation marks omitted). A State need not address all aspects of a problem in one fell swoop; policymakers may focus on their most pressing concerns. We have accordingly upheld laws—even under strict scrutiny—that conceivably could have restricted even greater amounts of speech in service of their stated interests. Burson, 504 U. S., at 207; see McConnell, 540 U. S., at 207–208; Metromedia, Inc. v. San Diego,453 U. S. 490–512 (1981) (plurality opinion); Buckley, 424 U. S., at 105. Viewed in light of these principles, Canon 7C(1) raises no fatal underinclusivity concerns. The solicitation ban aims squarely at the conduct most likely to undermine public confidence in the integrity of the judiciary: personal requests for money by judges and judicial candidates. The Canon applies evenhandedly to all judges and judicial candidates, regardless of their viewpoint or chosen means of solicitation. And unlike some laws that we have found impermissibly underinclusive, Canon 7C(1) is not riddled with exceptions. See City of Ladue v. Gilleo,512 U. S. 43–53 (1994). Indeed, the Canon contains zero exceptions to its ban on personal solicitation. Yulee relies heavily on the provision of Canon 7C(1) that allows solicitation by a candidate’s campaign committee. But Florida, along with most other States, has reasonably concluded that solicitation by the candidate personally creates a categorically different and more severe risk of undermining public confidence than does solicitation by a campaign committee. The identity of the solicitor matters, as anyone who has encountered a Girl Scout selling cookies outside a grocery store can attest. When the judicial candidate himself asks for money, the stakes are higher for all involved. The candidate has personally invested his time and effort in the fundraising appeal; he has placed his name and reputation behind the request. The solicited individual knows that, and also knows that the solicitor might be in a position to singlehandedly make decisions of great weight: The same person who signed the fundraising letter might one day sign the judgment. This dynamic inevitably creates pressure for the recipient to comply, and it does so in a way that solicitation by a third party does not. Just as inevitably, the personal involvement of the candidate in the solicitation creates the public appearance that the candidate will remember who says yes, and who says no. In short, personal solicitation by judicial candidates implicates a different problem than solicitation by campaign committees. However similar the two solicitations may be in substance, a State may conclude that they present markedly different appearances to the public. Florida’s choice to allow solicitation by campaign commit-tees does not undermine its decision to ban solicitation by judges. Likewise, allowing judicial candidates to write thank you notes to campaign donors does not detract from the State’s interest in preserving public confidence in the integrity of the judiciary. Yulee argues that permitting thank you notes heightens the likelihood of actual bias by ensuring that judicial candidates know who supported their campaigns, and ensuring that the supporter knows that the candidate knows. Maybe so. But the State’s compelling interest is implicated most directly by the candidate’s personal solicitation itself. A failure to ban thank you notes for contributions not solicited by the candidate does not undercut the Bar’s rationale. In addition, the State has a good reason for allowing candidates to write thank you notes and raise money through committees. These accommodations reflect Flor-ida’s effort to respect the First Amendment interests of candidates and their contributors—to resolve the “fundamental tension between the ideal character of the judicial office and the real world of electoral politics.” Chisom v. Roemer,501 U. S. 380,400 (1991). They belie the principal dissent’s suggestion that Canon 7C(1) reflects general “hostility toward judicial campaigning” and has “nothing to do with the appearances created by judges’ asking for money.” Post, at 11. Nothing? The principal dissent also suggests that Canon 7C(1) is underinclusive because Florida does not ban judicial candidates from asking individuals for personal gifts or loans. Post, at 10. But Florida law treats a personal “gift” or “loan” as a campaign contribution if the donor makes it “for the purpose of influencing the results of an election,” Fla. Stat. §106.011(5)(a), and Florida’s Judicial Qualifications Commission has determined that a judicial candidate violates Canon 7C(1) by personally soliciting such a loan. See In re Turner, 76 So. 3d 898, 901–902 (Fla. 2011). In any event, Florida can ban personal solicitation of campaign funds by judicial candidates without making them obey a comprehensive code to leading an ethical life. Underinclusivity creates a First Amendment concern when the State regulates one aspect of a problem while declining to regulate a different aspect of the problem that affects its stated interest in a comparable way. See Flor-ida Star v. B. J. F.,491 U. S. 524,540 (1989). The principal dissent offers no basis to conclude that judicial candidates are in the habit of soliciting personal loans, football tickets, or anything of the sort. Post, at 10. Even under strict scrutiny, “[t]he First Amendment does not require States to regulate for problems that do not exist.” Burson, 504 U. S., at 207 (State’s regulation of political solicitation around a polling place, but not charitable or commercial solicitation, was not fatally underinclusive under strict scrutiny). Taken to its logical conclusion, the position advanced by Yulee and the principal dissent is that Florida may ban the solicitation of funds by judicial candidates only if the State bans all solicitation of funds in judicial elections. The First Amendment does not put a State to that all-or-nothing choice. We will not punish Florida for leaving open more, rather than fewer, avenues of expression, especially when there is no indication that the selective restriction of speech reflects a pretextual motive.C After arguing that Canon 7C(1) violates the First Amendment because it restricts too little speech, Yulee argues that the Canon violates the First Amendment because it restricts too much. In her view, the Canon is not narrowly tailored to advance the State’s compelling interest through the least restrictive means. See United States v. Playboy Entertainment Group, Inc.,529 U. S. 803,813 (2000). By any measure, Canon 7C(1) restricts a narrow slice of speech. A reader of Justice Kennedy’s dissent could be forgiven for concluding that the Court has just upheld a latter-day version of the Alien and Sedition Acts, approving “state censorship” that “locks the First Amendment out,” imposes a “gag” on candidates, and inflicts “dead weight” on a “silenced” public debate. Post, at 2–4. But in reality, Canon 7C(1) leaves judicial candidates free to discuss any issue with any person at any time. Candidates can write letters, give speeches, and put up billboards. They can contact potential supporters in person, on the phone, or online. They can promote their campaigns on radio, television, or other media. They cannot say, “Please give me money.” They can, however, direct their campaign committees to do so. Whatever else may be said of the Canon, it is surely not a “wildly disproportionate restriction upon speech.” Post, at 1 (Scalia, J., dissenting). Indeed, Yulee concedes—and the principal dissent seems to agree, post, at 8—that Canon 7C(1) is valid in numerous applications. Yulee acknowledges that Florida can prohibit judges from soliciting money from lawyers and litigants appearing before them. Reply Brief 18. In addition, she says the State “might” be able to ban “direct one-to-one solicitation of lawyers and individuals or businesses that could reasonably appear in the court for which the individual is a candidate.” Ibid. She also suggests that the Bar could forbid “in person” solicitation by judicial candidates. Tr. of Oral Arg. 7; cf. Ohralik v. Ohio State Bar Assn.,436 U. S. 447 (1978) (permitting State to ban in person solicitation of clients by lawyers). But Yulee argues that the Canon cannot constitutionally be applied to her chosen form of solicitation: a letter posted online and distributed via mass mailing. No one, she contends, will lose confidence in the integrity of the judiciary based on personal solicitation to such a broad audience. This argument misperceives the breadth of the compelling interest that underlies Canon 7C(1). Florida has reasonably determined that personal appeals for money by a judicial candidate inherently create an appearance of impropriety that may cause the public to lose confidence in the integrity of the judiciary. That interest may be implicated to varying degrees in particular contexts, but the interest remains whenever the public perceives the judge personally asking for money. Moreover, the lines Yulee asks us to draw are unwork-able. Even under her theory of the case, a mass mailing would create an appearance of impropriety if addressed to a list of all lawyers and litigants with pending cases. So would a speech soliciting contributions from the 100 most frequently appearing attorneys in the jurisdiction. Yulee says she might accept a ban on one-to-one solicitation, but is the public impression really any different if a judicial candidate tries to buttonhole not one prospective donor but two at a time? Ten? Yulee also agrees that in person solicitation creates a problem. But would the public’s concern recede if the request for money came in a phone call or a text message? We decline to wade into this swamp. The First Amendment requires that Canon 7C(1) be narrowly tailored, not that it be “perfectly tailored.” Burson, 504 U. S., at 209. The impossibility of perfect tailoring is especially apparent when the State’s compelling interest is as intangible as public confidence in the integrity of the judiciary. Yulee is of course correct that some personal solicitations raise greater concerns than others. A judge who passes the hat in the courthouse creates a more serious appearance of impropriety than does a judicial candidate who makes a tasteful plea for support on the radio. But most problems arise in greater and lesser gradations, and the First Amendment does not confine a State to addressing evils in their most acute form. See id., at 210. Here, Florida has concluded that all personal solicitations by judicial candidates create a public appearance that undermines confidence in the integrity of the judiciary; banning all personal solicitations by judicial candidates is narrowly tailored to address that concern. In considering Yulee’s tailoring arguments, we are mindful that most States with elected judges have determined that drawing a line between personal solicitation by candidates and solicitation by committees is necessary to preserve public confidence in the integrity of the judiciary. These considered judgments deserve our respect, especially because they reflect sensitive choices by States in an area central to their own governance—how to select those who “sit as their judges.” Gregory v. Ashcroft,501 U. S. 452,460 (1991). Finally, Yulee contends that Florida can accomplish its compelling interest through the less restrictive means of recusal rules and campaign contribution limits. We dis-agree. A rule requiring judges to recuse themselves from every case in which a lawyer or litigant made a campaign contribution would disable many jurisdictions. And a flood of postelection recusal motions could “erode public confidence in judicial impartiality” and thereby exacerbate the very appearance problem the State is trying to solve. Caperton, 556 U. S., at 891 (Roberts, C. J., dissenting). Moreover, the rule that Yulee envisions could create a perverse incentive for litigants to make campaign contributions to judges solely as a means to trigger their later recusal—a form of peremptory strike against a judge that would enable transparent forum shopping. As for campaign contribution limits, Florida already applies them to judicial elections. Fla. Stat. §106.08(1)(a). A State may decide that the threat to public confidence created by personal solicitation exists apart from the amount of money that a judge or judicial candidate seeks. Even if Florida decreased its contribution limit, the ap-pearance that judges who personally solicit funds might improperly favor their campaign donors would remain. Although the Court has held that contribution limits advance the interest in preventing quid pro quo corruption and its appearance in political elections, we have never held that adopting contribution limits precludes a State from pursuing its compelling interests through additional means. And in any event, a State has compelling interests in regulating judicial elections that extend beyond its interests in regulating political elections, because judges are not politicians. In sum, because Canon 7C(1) is narrowly tailored to serve a compelling government interest, the First Amendment poses no obstacle to its enforcement in this case. As a result of our decision, Florida may continue to prohibit judicial candidates from personally soliciting campaign funds, while allowing them to raise money through committees and to otherwise communicate their electoral messages in practically any way. The principal dissent faults us for not answering a slew of broader questions, such as whether Florida may cap a judicial candidate’s spending or ban independent expenditures by corporations. Post, at 8–9. Yulee has not asked these questions, and for good reason—they are far afield from the narrow regulation actually at issue in this case. We likewise have no cause to consider whether the citizens of States that elect their judges have decided anything about the “oracular sanctity of judges” or whether judges are due “a hearty helping of humble pie.” Post,at 12. The principal dissent could be right that the decision to adopt judicial elections “probably springs,” at least in part, from a desire to make judges more accountable to the public, ibid., although the history on this matter is more complicated. See J. Shugerman, The People’s Courts, at 5 (arguing that States adopted judicial elections to increase judicial independence). In any event, it is a long way from general notions of judicial accountability to the principal dissent’s view, which evokes nothing so much as Delacroix’s painting of Liberty leading a determined band of citoyens, this time against a robed aristocracy scurrying to shore up the ramparts of the judicial castle through disingenuous ethical rules. We claim no similar insight into the People’s passions, hazard no assertions about ulterior motives of those who promulgated Canon 7C(1), and firmly reject the charge of a deceptive “pose of neutrality” on the part of those who uphold it. Post, at 12.* * * The desirability of judicial elections is a question that has sparked disagreement for more than 200 years. Hamilton believed that appointing judges to positions with life tenure constituted “the best expedient which can be devised in any government to secure a steady, upright, and impartial administration of the laws.” The Federalist No. 78, at 465. Jefferson thought that making judges “dependent on none but themselves” ran counter to the principle of “a government founded on the public will.” 12 The Works of Thomas Jefferson 5 (P. Ford ed. 1905). The federal courts reflect the view of Hamilton; most States have sided with Jefferson. Both methods have given our Nation jurists of wisdom and rectitude who have devoted themselves to maintaining “the public’s respect . . . and a reserve of public goodwill, without becoming subservient to public opinion.” Rehnquist, Judicial Independence, 38 U. Rich. L. Rev. 579, 596 (2004). It is not our place to resolve this enduring debate. Our limited task is to apply the Constitution to the question presented in this case. Judicial candidates have a First Amendment right to speak in support of their campaigns. States have a compelling interest in preserving public confidence in their judiciaries. When the State adopts a narrowly tailored restriction like the one at issue here, those principles do not conflict. A State’s decision to elect judges does not compel it to compromise public confidence in their integrity. The judgment of the Florida Supreme Court isAffirmed. Notes1* Yulee also contended that she had not violated Canon 7C(1), which applies to “a judicial office that is filled by public election between competing candidates,” because the incumbent judge had not declared his campaign for reelection at the time she sent her solicitation letter. She has since abandoned that argument.

574.US.528

While conducting an offshore inspection of a commercial fishing vessel in the Gulf of Mexico, a federal agent found that the ship’s catch contained undersized red grouper, in violation of federal conservation regulations. The officer instructed the ship’s captain, petitioner Yates, to keep the undersized fish segregated from the rest of the catch until the ship returned to port. After the officer departed, Yates instead told a crew member to throw the undersized fish overboard. For this offense, Yates was charged with destroying, concealing, and covering up undersized fish to impede a federal investigation, in violation of 18 U. S. C. §1519. That section provides that a person may be fined or imprisoned for up to 20 years if he “knowingly alters, destroys, mutilates, conceals, covers up, falsifies, or makes a false entry in any record, document, or tangible object with the intent to impede, obstruct, or influence” a federal investigation. At trial, Yates moved for a judgment of acquittal on the §1519 charge. Pointing to §1519’s origin as a provision of the Sarbanes-Oxley Act of 2002, a law designed to protect investors and restore trust in financial markets following the collapse of Enron Corporation, Yates argued that §1519’s reference to “tangible object” subsumes objects used to store information, such as computer hard drives, not fish. The District Court denied Yates’s motion, and a jury found him guilty of violating §1519. The Eleventh Circuit affirmed the conviction, concluding that §1519 applies to the destruction or concealment of fish because, as objects having physical form, fish fall within the dictionary definition of “tangible object.” Held: The judgment is reversed, and the case is remanded. 733 F.3d 1059, reversed and remanded. Justice Ginsburg, joined by The Chief Justice, Justice Breyer, and Justice Sotomayor, concluded that a “tangible object” within §1519’s compass is one used to record or preserve information. Pp. 6–20. (a) Although dictionary definitions of the words “tangible” and “object” bear consideration in determining the meaning of “tangible object” in §1519, they are not dispositive. Whether a statutory term is unambiguous “is determined [not only] by reference to the language itself, [but also by] the specific context in which that language is used, and the broader context of the statute as a whole.” Robinson v. Shell Oil Co., 519 U.S. 337, 341. Identical language may convey varying content when used in different statutes, sometimes even in different provisions of the same statute. See, e.g., FAA v. Cooper, 566 U. S. ___, ___. Pp. 7–10. (b) Familiar interpretive guides aid the construction of “tangible object.” Though not commanding, §1519’s heading—“Destruction, alteration, or falsification of records in Federal investigations and bankruptcy”—conveys no suggestion that the section prohibits spoliation of any and all physical evidence, however remote from records. Section 1519’s position within Title 18, Chapter 73, further signals that §1519 was not intended to serve as a cross-the-board ban on the destruction of physical evidence. Congress placed §1519 at the end of Chapter 73 following immediately after pre-existing specialized provisions expressly aimed at corporate fraud and financial audits. The contemporaneous passage of §1512(c)(1), which prohibits a person from “alter[ing], destroy[ing], mutilat[ing], or conceal[ing] a record, document, or other object . . . with the intent to impair the object’s integrity or availability for use in an official proceeding,” is also instructive. The Government argues that §1512(c)(1)’s reference to “other object” includes any and every physical object. But if §1519’s reference to “tangible object” already included all physical objects, as the Government also contends, then Congress had no reason to enact §1512(c)(1). Section 1519 should not be read to render superfluous an entire provision passed in proximity as part of the same Act. See Marx v. General Revenue Corp., 568 U. S. ___, ___. The words immediately surrounding “tangible object” in §1519—“falsifies, or makes a false entry in any record [or] document”—also cabin the contextual meaning of that term. Applying the canons noscitur a sociis and ejusdem generis, “tangible object,” as the last in a list of terms that begins “any record [or] document,” is appropriately read to refer, not to any tangible object, but specifically to the subset of tangible objects used to record or preserve information. This moderate interpretation accords with the list of actions §1519 proscribes; the verbs “falsif[y]” and “mak[e] a false entry in” typically take as grammatical objects records, documents, or things used to record or preserve information, such as logbooks or hard drives. See Gustafson v. Alloyd Co., 513 U.S. 561, 575. Use of traditional tools of statutory interpretation to examine markers of congressional intent within the Sarbanes-Oxley Act and §1519 itself thus call for rejection of an aggressive interpretation of “tangible object.” Furthermore, the meaning of “record, document, or thing” in a provision of the 1962 Model Penal Code (MPC) that has been interpreted to prohibit tampering with any kind of physical evidence is not a reliable indicator of the meaning Congress assigned to “record, document, or tangible object” in §1519. There are significant differences between the offense described by the MPC provision and the offense created by §1519. Pp. 10–18. (c) Finally, if recourse to traditional tools of statutory construction leaves any doubt about the meaning of “tangible object” in §1519, it would be appropriate to invoke the rule of lenity. Pp. 18–19. Justice Alito concluded that traditional rules of statutory construction confirm that Yates has the better argument. Title 18 U. S. C. §1519’s list of nouns, list of verbs, and title, when combined, tip the case in favor of Yates. Applying the canons noscitur a sociis and ejusdem generis to the list of nouns—“any record, document, or tangible object”—the term “tangible object” should refer to something similar to records or documents. And while many of §1519’s verbs—“alters, destroys, mutilates, conceals, covers up, falsifies, or makes a false entry in”—could apply to far-flung nouns such as salamanders or sand dunes, the term “makes a false entry in” makes no sense outside of filekeeping. Finally, §1519’s title—“Destruction, alteration, or falsification of records in Federal investigations and bankruptcy”—also points toward filekeeping rather than fish. Pp. 1–4. Ginsburg, J., announced the judgment of the Court and delivered an opinion, in which Roberts, C. J., and Breyer and Sotomayor, JJ., joined. Alito, J., filed an opinion concurring in the judgment. Kagan, J., filed a dissenting opinion, in which Scalia, Kennedy, and Thomas, JJ., joined.

in which The Chief Justice, Justice Breyer, and Justice Sotomayor join. John Yates, a commercial fisherman, caught undersized red grouper in federal waters in the Gulf of Mexico. To prevent federal authorities from confirming that he had harvested undersized fish, Yates ordered a crew member to toss the suspect catch into the sea. For this offense, he was charged with, and convicted of, violating18 U. S. C. §1519, which provides: “Whoever knowingly alters, destroys, mutilates, conceals, covers up, falsifies, or makes a false entry in any record, document, or tangible object with the intent to impede, obstruct, or influence the investigation or proper administration of any matter within the jurisdiction of any department or agency of the United States or any case filed under title 11, or in relation to or contemplation of any such matter or case, shall be fined under this title, imprisoned not more than 20 years, or both.”Yates was also indicted and convicted under §2232(a), which provides: “Destruction or Removal of Property to Prevent Seizure.—Whoever, before, during, or after any search for or seizure of property by any person authorized to make such search or seizure, knowingly destroys, damages, wastes, disposes of, transfers, or otherwise takes any action, or knowingly attempts to destroy, damage, waste, dispose of, transfer, or otherwise take any action, for the purpose of preventing or impairing the Government’s lawful authority to take such property into its custody or control or to continue holding such property under its lawful custody and control, shall be fined under this title or imprisoned not more than 5 years, or both.”Yates does not contest his conviction for violating §2232(a), but he maintains that fish are not trappedwithin the term “tangible object,” as that term is used in §1519. Section 1519 was enacted as part of the Sarbanes-Oxley Act of 2002,116Stat.745, legislation designed to protect investors and restore trust in financial markets following the collapse of Enron Corporation. A fish is no doubt an object that is tangible; fish can be seen, caught, andhandled, and a catch, as this case illustrates, is vulnerable to destruction. But it would cut §1519 loose from itsfinancial-fraud mooring to hold that it encompasses any and all objects, whatever their size or significance, destroyed with obstructive intent. Mindful that in Sarbanes-Oxley, Congress trained its attention on corporate and accounting deception and cover-ups, we conclude that a matching construction of §1519 is in order: A tangible object captured by §1519, we hold, must be one used to record or preserve information.I On August 23, 2007, the Miss Katie, a commercial fishing boat, was six days into an expedition in the Gulf of Mexico. Her crew numbered three, including Yates, the captain. Engaged in a routine offshore patrol to inspect both recreational and commercial vessels, Officer John Jones of the Florida Fish and Wildlife Conservation Commission decided to board the Miss Katie to check on the vessel’s compliance with fishing rules. Although the Miss Katie was far enough from the Florida coast to be in exclusively federal waters, she was nevertheless within Officer Jones’s jurisdiction. Because he had been deputized as a federal agent by the National Marine Fisheries Service, Officer Jones had authority to enforce federal, as well as state, fishing laws. Upon boarding the Miss Katie, Officer Jones noticed three red grouper that appeared to be undersized hanging from a hook on the deck. At the time, federal conservation regulations required immediate release of red grouper less than 20 inches long. 50 CFR §622.37(d)(2)(ii) (effective April 2, 2007). Violation of those regulations is a civil offense punishable by a fine or fishing license suspension. See 16 U. S. C. §§1857(1)(A), (G), 1858(a), (g). Suspecting that other undersized fish might be on board, Officer Jones proceeded to inspect the ship’s catch, setting aside and measuring only fish that appeared to him to be shorter than 20 inches. Officer Jones ultimately determined that 72 fish fell short of the 20inch mark. A fellow officer recorded the length of each of the undersized fish on a catch measurement verification form. With few exceptions, the measured fish were between 19 and 20 inches; three were less than 19 inches; none were less than 18.75 inches. After separating the fish measuring below 20 inches from the rest of the catch by placing them in wooden crates, Officer Jones directed Yates to leave the fish, thus segregated, in the crates until the Miss Katie returned to port. Before departing, Officer Jones issued Yates a citation for possession of undersized fish. Four days later, after the Miss Katie had docked in Cortez, Florida, Officer Jones measured the fish contained in the wooden crates. This time, however, the measured fish, although still less than 20 inches, slightly exceeded the lengths recorded on board. Jones surmised that the fish brought to port were not the same as those he had detected during his initial inspection. Under questioning, one of the crew members admitted that, at Yates’s direction, he had thrown overboard the fish Officer Jones had measured at sea, and that he and Yates had replaced the tossed grouper with fish from the rest of the catch. For reasons not disclosed in the record before us, more than 32 months passed before criminal charges were lodged against Yates. On May 5, 2010, he was indicted for destroying property to prevent a federal seizure, in violation of §2232(a), and for destroying, concealing, and covering up undersized fish to impede a federal investigation, in violation of §1519.[1] By the time of the indictment, the minimum legal length for Gulf red grouper had been lowered from 20 inches to 18 inches. See 50 CFR §622.37(d)(2)(iv) (effective May 18, 2009). No measured fish in Yates’s catch fell below that limit. The record does not reveal what civil penalty, if any, Yates received for his possession of fish undersized under the 2007 regulation. See16 U. S. C. §1858(a). Yates was tried on the criminal charges in August 2011. At the end of the Government’s case in chief, he moved for a judgment of acquittal on the §1519 charge. Pointing to §1519’s title and its origin as a provision of the Sarbanes-Oxley Act, Yates argued that the section sets forth “a documents offense” and that its reference to “tangible object[s]” subsumes “computer hard drives, logbooks, [and] things of that nature,” not fish. App. 91–92. Yates acknowledged that the Criminal Code contains “sections that would have been appropriate for the [G]overnment to pursue” if it wished to prosecute him for tampering with evidence. App. 91. Section 2232(a), set out supra, at 1–2, fit that description. But §1519, Yates insisted, did not. The Government countered that a “tangible object” within §1519’s compass is “simply something other than a document or record.” App. 93. The trial judge expressed misgivings about reading “tangible object” as broadly as the Government urged: “Isn’t there a Latin phrase [about] construction of a statute . . . . The gist of it is . . . you take a look at [a] line of words, and you interpret the words consistently. So if you’re talking about documents, and records, tangible objects are tangible objects in the nature of a document or a record, as opposed to a fish.” Ibid. The first-instance judge nonetheless followed controlling Eleventh Circuit precedent. While recognizing that §1519 was passed as part of legislation targeting corporate fraud, the Court of Appeals had instructed that “the broad language of §1519 is not limited to corporate fraud cases, and ‘Congress is free to pass laws with language covering areas well beyond the particular crisis du jour that initially prompted legislative action.’ ” No. 2:10–cr–66–FtM–29SPC (MD Fla., Aug. 8, 2011), App. 116 (quoting United States v. Hunt, 526 F. 3d 739, 744 (CA11 2008)). Accordingly, the trial court read “tangible object” as a term “independent” of “record” or “document.” App. 116. For violating §1519 and §2232(a), the court sentenced Yates to imprisonment for 30 days, followed by supervised release for three years. App. 118–120. For life, he will bear the stigma of having a federal felony conviction. On appeal, the Eleventh Circuit found the text of §1519 “plain.” 733 F. 3d 1059, 1064 (2013). Because “tangible object” was “undefined” in the statute, the Court of Appeals gave the term its “ordinary or natural meaning,” i.e., its dictionary definition, “[h]aving or possessing physical form.” Ibid. (quoting Black’s Law Dictionary 1592 (9th ed. 2009)). We granted certiorari, 572 U. S. ___ (2014), and now reverse the Eleventh Circuit’s judgment.II The Sarbanes-Oxley Act, all agree, was prompted by the exposure of Enron’s massive accounting fraud and revelations that the company’s outside auditor, Arthur Andersen LLP, had systematically destroyed potentially incriminating documents. The Government acknowledges that §1519 was intended to prohibit, in particular, corporate document-shredding to hide evidence of financial wrong-doing. Brief for United States 46. Prior law made it an offense to “intimidat[e], threate[n], or corruptly persuad[e] another person” to shred documents. §1512(b) (emphasis added). Section 1519 cured a conspicuous omission by imposing liability on a person who destroys records himself. See S. Rep. No. 107–146, p. 14 (2002) (describing §1519 as “a new general anti shredding provision” and explaining that “certain current provisions make it a crime to persuade another person to destroy documents, but not a crime to actually destroy the same documents yourself”). The new section also expanded prior law by including within the provision’s reach “any matter within the jurisdiction of any department or agency of the United States.” Id., at 14–15. In the Government’s view, §1519 extends beyond the principal evil motivating its passage. The words of §1519, the Government argues, support reading the provision as a general ban on the spoliation of evidence, covering all physical items that might be relevant to any matter under federal investigation. Yates urges a contextual reading of §1519, tying “tangible object” to the surrounding words, the placement of the provision within the Sarbanes-Oxley Act, and related provisions enacted at the same time, in particular §1520 and §1512(c)(1), see infra, at 10, 12–13. Section 1519, he maintains, targets not all manner of evidence, but records, documents, and tangible objects used to preserve them, e.g., computers, servers, and other media on which information is stored. We agree with Yates and reject the Government’s unrestrained reading. “Tangible object” in §1519, we conclude, is better read to cover only objects one can use to record or preserve information, not all objects in the physical world.A The ordinary meaning of an “object” that is “tangible,” as stated in dictionary definitions, is “a discrete . . . thing,” Webster’s Third New International Dictionary 1555 (2002), that “possess[es] physical form,” Black’s Law Dictionary 1683 (10th ed. 2014). From this premise, the Government concludes that “tangible object,” as that term appears in §1519, covers the waterfront, including fish from the sea. Whether a statutory term is unambiguous, however, does not turn solely on dictionary definitions of its component words. Rather, “[t]he plainness or ambiguity of statutory language is determined [not only] by reference to the language itself, [but as well by] the specific context in which that language is used, and the broader context of the statute as a whole.” Robinson v. Shell Oil Co.,519 U. S. 337,341 (1997). See also Deal v. United States,508 U. S. 129,132 (1993) (it is a “fundamental principle of statutory construction (and, indeed, of language itself) that the meaning of a word cannot be determined in isolation, but must be drawn from the context in which it is used”). Ordinarily, a word’s usage accords with its dictionary definition. In law as in life, however, the same words, placed in different contexts, sometimes mean different things. We have several times affirmed that identical language may convey varying content when used in different statutes, sometimes even in different provisions of the same statute. See, e.g., FAA v. Cooper, 566 U. S. ___, ___–___ (2012), (slip op., at 6–7) (“actual damages” has different meanings in different statutes); Wachovia Bank, N. A. v. Schmidt,546 U. S. 303–314 (2006) (“located” has different meanings in different provisions of the National Bank Act); General Dynamics Land Systems, Inc. v. Cline,540 U. S. 581–597 (2004) (“age” has different meanings in different provisions of the Age Discrimination in Employment Act of 1967); United States v. Cleveland Indians Baseball Co.,532 U. S. 200,213 (2001) (“wages paid” has different meanings in different provisions of Title 26 U. S. C.); Robinson, 519 U. S., at 342–344 (“employee” has different meanings in different sections of Title VII of the Civil Rights Act of 1964); Merrell Dow Pharmaceuticals Inc. v. Thompson,478 U. S. 804–808 (1986) (“arising under” has different meanings in U. S. Const., Art. III, §2, and28 U. S. C. §1331); District of Columbia v. Carter,409 U. S. 418–421 (1973) (“State or Territory” has different meanings in42 U. S. C. §1982 and §1983); Atlantic Cleaners & Dyers, Inc. v. United States,286 U. S. 427–437 (1932) (“trade or commerce” has different meanings in different sections of the Sherman Act). As the Court observed in Atlantic Cleaners & Dyers, 286 U. S., at 433:“Most words have different shades of meaning and consequently may be variously construed . . . . Where the subject matter to which the words refer is not the same in the several places where [the words] are used, or the conditions are different, or the scope of the legislative power exercised in one case is broader than that exercised in another, the meaning well may vary to meet the purposes of the law, to be arrived at by a consideration of the language in which those purposes are expressed, and of the circumstances under which the language was employed.”[2]In short, although dictionary definitions of the words “tangible” and “object” bear consideration, they are not dispositive of the meaning of “tangible object” in §1519. Supporting a reading of “tangible object,” as used in §1519, in accord with dictionary definitions, the Government points to the appearance of that term in Federal Rule of Criminal Procedure 16. That Rule requires the prosecution to grant a defendant’s request to inspect “tangible objects” within the Government’s control that have utility for the defense. See Fed. Rule Crim. Proc. 16(a)(1)(E). Rule 16’s reference to “tangible objects” has been interpreted to include any physical evidence. See 5 W. LaFave, J. Israel, N. King, & O. Kerr, Criminal Procedure §20.3(g), pp. 405–406, and n. 120 (3d ed. 2007). Rule 16 is a discovery rule designed to protect defendants by compelling the prosecution to turn over to the defense evidence material to the charges at issue. In that context, a comprehensive construction of “tangible objects” is fitting. In contrast, §1519 is a penal provision that refers to “tangible object” not in relation to a request for information relevant to a specific court proceeding, but rather in relation to federal investigations or proceedings of every kind, including those not yet begun.[3] See Commissioner v. National Carbide Corp., 167 F. 2d 304, 306 (CA2 1948) (Hand, J.) (“words are chameleons, which reflect the color of their environment”). Just as the context of Rule 16 supports giving “tangible object” a meaning as broad as its dictionary definition, the context of §1519 tugs strongly in favor of a narrower reading.B Familiar interpretive guides aid our construction of the words “tangible object” as they appear in §1519. We note first §1519’s caption: “Destruction, alteration, or falsification of records in Federal investigations and bankruptcy.” That heading conveys no suggestion that the section prohibits spoliation of any and all physical evidence, however remote from records. Neither does the title of the section of the Sarbanes-Oxley Act in which §1519 was placed, §802: “Criminal penalties for altering documents.”116Stat.800. Furthermore, §1520, the only other provision passed as part of §802, is titled “Destruction of corporate audit records” and addresses only that specific subset of records and documents. While these headings are not commanding, they supply cues that Congress did not intend “tangible object” in §1519 to sweep within its reach physical objects of every kind, including things no one would describe as records, documents, or devices closely associated with them. SeeAlmendarez-Torres v. United States,523 U. S. 224,234 (1998) (“[T]he title of a statute and the heading of a section are tools available for the resolution of a doubt about the meaning of a statute.” (internal quotation marks omitted)). If Congress indeed meant to make §1519 an all-encompassing ban on the spoliation of evidence, as the dissent believes Congress did, one would have expected a clearer indication of that intent. Section 1519’s position within Chapter 73 of Title 18 further signals that §1519 was not intended to serve as a cross-the-board ban on the destruction of physical evidence of every kind. Congress placed §1519 (and its companion provision §1520) at the end of the chapter, following immediately after the pre-existing §1516, §1517, and §1518, each of them prohibiting obstructive acts in specific contexts. See §1516 (audits of recipients of federal funds); §1517 (federal examinations of financial institutions); §1518 (criminal investigations of federal health care offenses). See also S. Rep. No. 107–146, at 7 (observing that §1517 and §1518 “apply to obstruction in certain limited types of cases, such as bankruptcy fraud, examinations of financial institutions, and healthcare fraud”). But Congress did not direct codification of the Sarbanes-Oxley Act’s other additions to Chapter 73 adjacent to these specialized provisions. Instead, Congress directed placement of those additions within or alongside retained provisions that address obstructive acts relating broadly to official proceedings and criminal trials: Section 806, “Civil Action to protect against retaliation in fraud cases,” was codified as §1514A and inserted between the pre-existing §1514, which addresses civil actions to restrain harassment of victims and witnesses in criminal cases, and §1515, which defines terms used in §1512 and §1513. Section 1102, “Tampering with a record or otherwise impeding an official proceeding,” was codified as §1512(c) and inserted within the pre-existing §1512, which addresses tampering with a victim, witness, or informant to impede any official proceeding. Section 1107, “Retaliation against informants,” was codified as §1513(e) and inserted within the pre-existing §1513, which addresses retaliation against a victim, witness, or informant in any official proceeding. Congress thus ranked §1519, not among the broad proscriptions, but together with specialized provisions expressly aimed at corporate fraud and financial audits. This placement accords with the view that Congress’ conception of §1519’s coverage was considerably more limited than the Government’s.[4] The contemporaneous passage of §1512(c)(1), which was contained in a section of the Sarbanes-Oxley Act discrete from the section embracing §1519 and §1520, is also instructive. Section 1512(c)(1) provides: “(c) Whoever corruptly— “(1) alters, destroys, mutilates, or conceals a record, document, or other object, or attempts to do so, with the intent to impair the object’s integrity or availability for use in an official proceeding. . . . .“shall be fined under this title or imprisoned not more than 20 years, or both.”The legislative history reveals that §1512(c)(1) was drafted and proposed after §1519. See 148 Cong. Rec. 12518, 13088–13089 (2002). The Government argues, and Yates does not dispute, that §1512(c)(1)’s reference to “other object” includes any and every physical object. But if §1519’s reference to “tangible object” already included all physical objects, as the Government and the dissent contend, then Congress had no reason to enact §1512(c)(1): Virtually any act that would violate §1512(c)(1) no doubt would violate §1519 as well, for §1519 applies to “the investigation or proper administration of any matter within the jurisdiction of any department or agency of the United States . . . or in relation to or contemplation of any such matter,” not just to “an official proceeding.”[5] The Government acknowledges that, under its reading, §1519 and §1512(c)(1) “significantly overlap.” Brief for United States 49. Nowhere does the Government explain what independent function §1512(c)(1) would serve if the Government is right about the sweeping scope of §1519. We resist a reading of §1519 that would render superfluous an entire provision passed in proximity as part of the same Act.[6] See Marx v. General Revenue Corp., 568 U. S. ___, ___ (2013) (slip op., at 14) (“[T]he canon against surplusage is strongest when an interpretation would render superfluous another part of the same statutory scheme.”). The words immediately surrounding “tangible object” in §1519—“falsifies, or makes a false entry in any record [or] document”—also cabin the contextual meaning of that term. As explained in Gustafson v. Alloyd Co.,513 U. S. 561,575 (1995), we rely on the principle of noscitur a sociis—a word is known by the company it keeps—to “avoid ascribing to one word a meaning so broad that it is inconsistent with its accompanying words, thus giving unintended breadth to the Acts of Congress.” (internal quotation marks omitted). See also United States v. Williams,553 U. S. 285,294 (2008) (“a word is given more precise content by the neighboring words with which it is associated”). In Gustafson, we interpreted the word “communication” in §2(10) of the Securities Act of 1933 to refer to a public communication, rather than any communication, because the word appeared in a list with other words, notably “notice, circular, [and] advertisement,” making it “apparent that the list refer[red] to documents of wide dissemination.” 513 U. S., at 575–576. And we did so even though the list began with the word “any.” The noscitur a sociis canon operates in a similar manner here. “Tangible object” is the last in a list of terms that begins “any record [or] document.” The term is therefore appropriately read to refer, not to any tangible object, but specifically to the subset of tangible objects involving records and documents, i.e., objects used to record or preserve information. See United States Sentencing Commission, Guidelines Manual §2J1.2, comment., n. 1 (Nov. 2014) (“ ‘Records, documents, or tangible objects’ includes (A) records, documents, or tangible objects that are stored on, or that are, magnetic, optical, digital, other electronic, or other storage mediums or devices; and (B) wire or electronic communications.”). This moderate interpretation of “tangible object” accords with the list of actions §1519 proscribes. The section applies to anyone who “alters, destroys, mutilates, conceals, covers up, falsifies, or makes a false entry in any record, document, or tangible object” with the requisite obstructive intent. (Emphasis added.) The last two verbs, “falsif[y]” and “mak[e] a false entry in,” typically take as grammatical objects records, documents, or things used to record or preserve information, such as logbooks or hard drives. See, e.g., Black’s Law Dictionary 720 (10th ed. 2014) (defining “falsify” as “[t]o make deceptive; to counterfeit, forge, or misrepresent; esp., to tamper with (a document, record, etc.)”). It would be unnatural, for example, to describe a killer’s act of wiping his fingerprints from a gun as “falsifying” the murder weapon. But it would not be strange to refer to “falsifying” data stored on a hard drive as simply “falsifying” a hard drive. Furthermore, Congress did not include on §1512(c)(1)’s list of prohibited actions “falsifies” or “makes a false entry in.” See §1512(c)(1) (making it unlawful to “alte[r], destro[y], mutilat[e], or concea[l] a record, document, or other object” with the requisite obstructive intent). That contemporaneous omission also suggests that Congress intended “tangible object” in §1519 to have a narrower scope than “other object” in §1512(c)(1).[7] A canon related to noscitur a sociis, ejusdem generis, counsels: “Where general words follow specific words in a statutory enumeration, the general words are [usually] construed to embrace only objects similar in nature to those objects enumerated by the preceding specific words.” Washington State Dept. of Social and Health Servs. v. Guardianship Estate of Keffeler,537 U. S. 371,384 (2003) (internal quotation marks omitted). In Begay v. United States,553 U. S. 137–143 (2008), for example, we relied on this principle to determine what crimes were covered by the statutory phrase “any crime . . . that . . . is burglary, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another,”18 U. S. C. §924(e)(2)(B)(ii). The enumeration of specific crimes, we explained, indicates that the “otherwise involves” provision covers “only similar crimes, rather than every crime that ‘presents a serious potential risk of physical injury to another.’ ” 553 U. S., at 142. Had Congress intended the latter “all encompassing” meaning, we observed, “it is hard to see why it would have needed to include the examples at all.” Ibid. See also CSX Transp., Inc. v. Alabama Dept. of Revenue,562 U. S. 277, ___ (2011) (slip op., at 16) (“We typically use ejusdem generis to ensure that a general word will not render specific words meaningless.”). Just so here. Had Congress intended “tangible object” in §1519 to be interpreted so generically as to capture physical objects as dissimilar as documents and fish, Congress would have had no reason to refer specifically to “record” or “document.” The Government’s unbounded reading of “tangible object” would render those words misleading surplusage. Having used traditional tools of statutory interpretation to examine markers of congressional intent within the Sarbanes-Oxley Act and §1519 itself, we are persuaded that an aggressive interpretation of “tangible object” must be rejected. It is highly improbable that Congress would have buried a general spoliation statute covering objects of any and every kind in a provision targeting fraud in financial record-keeping. The Government argues, however, that our inquiry would be incomplete if we failed to consider the origins of the phrase “record, document, or tangible object.” Congress drew that phrase, the Government says, from a 1962 Model Penal Code (MPC) provision, and reform proposals based on that provision. The MPC provision and proposals prompted by it would have imposed liability on anyone who “alters, destroys, mutilates, conceals, or removes a record, document or thing.” See ALI, MPC §241.7(1), p. 175 (1962). Those proscriptions were understood to refer to all physical evidence. See MPC §241.7, Comment 3, at 179 (1980) (provision “applies to any physical object”). Accordingly, the Government reasons, and the dissent exuberantly agrees, post, at 4–5, Congress must have intended §1519 to apply to the universe of physical evidence. The inference is unwarranted. True, the 1962 MPC provision prohibited tampering with any kind of physical evidence. But unlike §1519, the MPC provision did not prohibit actions that specifically relate to records, documents, and objects used to record or preserve information. The MPC provision also ranked the offense as a misdemeanor and limited liability to instances in which the actor “believ[es] that an official proceeding or investigation is pending or about to be instituted.” MPC §241.7(1), at 175. Yates would have had scant reason to anticipate a felony prosecution, and certainly not one instituted at a time when even the smallest of the fish he caught came within the legal limit. See supra, at 4; cf. Bond v. United States, 572 U. S. ___, ___ (2014), (slip op., at 14) (rejecting “boundless reading” of a statutory term given “deeply serious consequences” that reading would entail). A proposed federal offense in line with the MPC provision, advanced by a federal commission in 1971, was similarly qualified. See Final Report of the National Commission on Reform of Federal Criminal Laws §1323, pp. 116–117 (1971). Section 1519 conspicuously lacks the limits built into the MPC provision and the federal proposal. It describes not a misdemeanor, but a felony punishable by up to 20 years in prison. And the section covers conduct intended to impede any federal investigation or proceeding, including one not even on the verge of commencement. Given these significant differences, the meaning of “record, document, or thing” in the MPC provision and a kindred proposal is not a reliable indicator of the meaning Congress assigned to “record, document, or tangible object” in §1519. The MPC provision, in short, tells us neither “what Congress wrote [nor] what Congress wanted,” cf. post, at 15, concerning Yates’s small fish as the subject of a federal felony prosecution.C Finally, if our recourse to traditional tools of statutory construction leaves any doubt about the meaning of “tangible object,” as that term is used in §1519, we would invoke the rule that “ambiguity concerning the ambit of criminal statutes should be resolved in favor of lenity.” Cleveland v. United States,531 U. S. 12,25 (2000) (quoting Rewis v. United States,401 U. S. 808,812 (1971)). That interpretative principle is relevant here, where the Government urges a reading of §1519 that exposes individuals to 20-year prison sentences for tampering with any physical object that might have evidentiary value in any federal investigation into any offense, no matter whether the investigation is pending or merely contemplated, or whether the offense subject to investigation is criminal or civil. See Liparota v. United States,471 U. S. 419,427 (1985) (“Application of the rule of lenity ensures that criminal statutes will provide fair warning concerning conduct rendered illegal and strikes the appropriate balance between the legislature, the prosecutor, and the court in defining criminal liability.”). In determining the meaning of “tangible object” in §1519, “it is appropriate, before we choose the harsher alternative, to require that Congress should have spoken in language that is clear and definite.” See Cleveland, 531 U. S., at 25 (quoting United States v. Universal C. I. T. Credit Corp.,344 U. S. 218,222 (1952)). See also Jones v. United States,529 U. S. 848–859 (2000) (rule of lenity “reinforces” the conclusion that arson of an owner-occupied residence is not subject to federal prosecution under18 U. S. C. §844(i) because such a residence does not qualify as property “used in” commerce or commerce-affecting activity).[8]* * * For the reasons stated, we resist reading §1519 expansively to create a coverall spoliation of evidence statute, advisable as such a measure might be. Leaving that important decision to Congress, we hold that a “tangible object” within §1519’s compass is one used to record or preserve information. The judgment of the U. S. Court of Appeals for the Eleventh Circuit is therefore reversed, and the case is remanded for further proceedings.It is so ordered.Notes1 Yates was also charged with making a false statement to federal law enforcement officers, in violation of18 U. S. C. §1001(a)(2). That charge, on which Yates was acquitted, is not relevant to our analysis.2 The dissent assiduously ignores all this, post, at 11–12, in insisting that Congress wrote §1519 to cover, along with shredded corporate documents, red grouper slightly smaller than the legal limit.3 For the same reason, we do not think the meaning of “tangible objects” (or “tangible things,” see Fed. Rule Civ. Proc. 26(b)) in other discovery prescriptions cited by the Government leads to the conclusion that “tangible object” in §1519 encompasses any and all physical evidence existing on land or in the sea.4 The dissent contends that nothing can be drawn from the placement of §1519 because, before and after Sarbanes-Oxley, “all of Chapter 73 was ordered chronologically.” Post, at 9. The argument might have some force if the factual premise were correct. In Sarbanes-Oxley, Congress directed insertion of §1514A before §1518, then the last section in Chapter 73. If, as the dissent argues, Congress adopted §1519 to fill out §1512, post, at 6–7, it would have made more sense for Congress to codify the substance of §1519 within §1512 or in a new §1512A, rather than placing §1519 among specialized provisions. Notably, in Sarbanes-Oxley, Congress added §1512(c)(1), “a broad ban on evidence-spoliation,” cf. post, at 9, n. 2, to §1512, even though §1512’s preexisting title and provisions all related to witness-tampering. 5 Despite this sweeping “in relation to” language, the dissent remarkably suggests that §1519 does not “ordinarily operate in th[e] context [of] federal court[s],” for those courts are not “department[s] or agenc[ies].” Post, at 10. That suggestion, which, as one would expect, lacks the Government’s endorsement, does not withstand examination. The Senate Committee Report on §1519, on which the dissent elsewhere relies, see post, at 6, explained that an obstructive act is within §1519’s scope if “done ‘in contemplation’ of or in relation to a matter or investigation.” S. Rep. 107–146, at 15. The Report further informed that §1519 “is . . . meant to do away with the distinctions, which some courts have read into obstruction statutes, between court proceedings, investigations, regulatory or administrative proceedings (whether formal or not), and less formal government inquiries, regardless of their title.” Ibid. If any doubt remained about the multiplicity of contexts in which §1519 was designed to apply, the Report added, “[t]he intent of the provision is simple; people should not be destroying, altering, or falsifying documents to obstruct any government function.” Ibid.6 Furthermore, if “tangible object” in §1519 is read to include any physical object, §1519 would prohibit all of the conduct proscribed by §2232(a), which imposes a maximum penalty of five years in prison for destroying or removing “property” to prevent its seizure by the Government. See supra, at 1–2. 7 The dissent contends that “record, document, or tangible object” in §1519 should be construed in conformity with “record, document, or other object” in §1512(c)(1) because both provisions address “the same basic problem.” Post, at 11–12. But why should that be so when Congress prohibited in §1519 additional actions, specific to paper and electronic documents and records, actions it did not prohibit in §1512(c)(1)? When Congress passed Sarbanes-Oxley in 2002, courts had already interpreted the phrase “alter, destroy, mutilate, or conceal an object” in §1512(b)(2)(B) to apply to all types of physical evidence. See, e.g., United States v. Applewhaite, 195 F. 3d 679, 688 (CA3 1999) (affirming conviction under §1512(b)(2)(B) for persuading another person to paint over blood spatter). Congress’ use of a formulation in §1519 that did not track the one used in §1512(b)(2)(B) (and repeated in §1512(c)(1)) suggests that Congress designed §1519 to be interpreted apart from §1512, not in lockstep with it.8 The dissent cites United States v. McRae, 702 F. 3d 806, 834–838 (CA5 2012), United States v. Maury, 695 F. 3d 227, 243–244 (CA3 2012), and United States v. Natal, 2014 U. S. Dist. LEXIS 108852, *24–*26 (Conn., Aug. 7, 2014), as cases that would not be covered by §1519 as we read it. Post, at 18–19. Those cases supply no cause for concern that persons who commit “major” obstructive acts, id. at 18, will go unpunished. The defendant in McRae, a police officer who seized a car containing a corpse and then set it on fire, was also convicted for that conduct under18 U. S. C. §844(h) and sentenced to a term of 120 months’ imprisonment for that offense. See 702 F. 3d, at 817–818, 839–840. The defendant in Natal, who repainted a van to cover up evidence of a fatal arson, was also convicted of three counts of violating18 U. S. C. §3 and sentenced to concurrent terms of 174 months’ imprisonment. See Judgment in United States v. Morales, No. 3:12–cr–164 (Conn., Jan. 12, 2015). And the defendant in Maury, a company convicted under §1519 of concealing evidence that a cement mixer’s safety lock was disabled when a worker’s fingers were amputated, was also convicted of numerous other violations, including three counts of violating18 U. S. C. §1505 for concealing evidence of other worker safety violations. See 695 F. 3d, at 244–245. See also United States v. Atlantic States Cast Iron Pipe Co., 2007 WL 2282514, *70 (NJ, Aug. 2, 2007) (setting forth charges against the company). For those violations, the company was fined millions of dollars and ordered to operate under the supervision of a court-appointed monitor. See 695 F. 3d, at 246.

576.US.2014_13-628

Petitioner Zivotofsky was born to United States citizens living in Jerusalem. Pursuant to §214(d) of the Foreign Relations Authorization Act, Fiscal Year 2003, his mother asked American Embassy officials to list his place of birth as “Israel” on, inter alia, his passport. Section 214(d) states for “purposes of the registration of birth, certification of nationality, or issuance of a passport of a United States citizen born in the city of Jerusalem, the Secretary shall, upon the request of the citizen or the citizen’s legal guardian, record the place of birth as Israel.” The Embassy officials refused to list Zivotofsky’s place of birth as “Israel” on his passport, citing the Executive Branch’s longstanding position that the United States does not recognize any country as having sovereignty over Jerusalem. Zivotofsky’s parents brought suit on his behalf in federal court, seeking to enforce §214(d). Ultimately, the D. C. Circuit held the statute unconstitutional, concluding that it contradicts the Executive Branch’s exclusive power to recognize foreign sovereigns. Held: 1. The President has the exclusive power to grant formal recognition to a foreign sovereign. Pp. 6–26. (a) Where, as here, the President’s action is “incompatible with the expressed or implied will of Congress,” the President “can rely [for his authority] only upon his own constitutional powers minus any constitutional powers of Congress over the matter,” Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579 (Jackson, J., concurring). His asserted power must be both “exclusive” and “conclusive” on the issue, id., at 637–638, and he may rely solely on powers the Constitution grants to him alone, id., at 638. To determine whether the President’s power of recognition is exclusive, this Court examines the Constitution’s text and structure and relevant precedent and history. Pp. 6–7. (b) The Constitution’s text and structure grant the President the power to recognize foreign nations and governments. The Reception Clause directs that the President “shall receive Ambassadors and other public Ministers,” Art. II, §3. And at the time of the founding, receiving an ambassador was considered tantamount to recognizing the sending state’s sovereignty. It is thus logical and proper to infer that the Reception Clause would be understood to acknowledge the President’s power to recognize other nations. This inference is further supported by the President’s additional Article II powers: to negotiate treaties and to nominate the Nation’s ambassadors and dispatch other diplomatic agents. Though ratifying a treaty and confirming an ambassador require congressional approval, Congress lacks authority to initiate the actions without the President’s involvement. The President, unlike Congress, also has the power to open diplomatic channels simply by engaging in direct diplomacy with foreign heads of state and their ministers. The Constitution thus assigns the President, not Congress, means to effect recognition on his own initiative. Functional considerations also suggest that the President’s recognition power is exclusive. The Nation must “speak . . . with one voice” regarding which governments are legitimate in the eyes of the United States and which are not, American Insurance Assn. v. Garamendi, 539 U. S. 396 , and only the Executive has the characteristic of unity at all times. Unlike Congress, the President is also capable of engaging in the delicate and often secret diplomatic contacts that may lead to a recognition decision, see, e.g., United States v. Pink, 315 U. S. 203 , and is better positioned to take the decisive, unequivocal action necessary to recognize other states at international law. The President has also exercised unilateral recognition power since the founding, a practice endorsed by this Court, see, e.g., Banco Nacional de Cuba v. Sabbatino, 376 U. S. 398 . Under basic separation-of-powers principles, Congress, which has the central role in making laws, see Art. I, §8, cl. 18, does have substantial authority regarding many policy determinations that precede and follow an act of recognition. The President’s recognition determination is thus only one part of a political process. Pp. 7–14. (b) A fair reading of relevant precedent illustrates that this Court has long considered recognition to be the exclusive prerogative of the Executive. See, e.g., Williams v. Suffolk Ins. Co., 13 Pet. 415, 420; United States v. Belmont, 301 U. S. 324 ; United States v. Pink, supra, at 229; Banco Nacional de Cuba v. Sabbatino, supra, at 410; National City Bank of N. Y. v. Republic of China, 348 U. S. 356 . United States v. Curtiss-Wright Export Corp., 299 U. S. 304 , does not support a broader definition of the Executive’s power over foreign relations that would permit the President alone to determine the whole content of the Nation’s foreign policy. The Executive is not free from the ordinary controls and checks of Congress merely because foreign affairs are at issue. See, e.g., Medellín v. Texas, 552 U. S. 491 –532. Nonetheless, it is for the President alone to make the specific decision of what foreign power he will recognize as legitimate, and his position must be clear. Pp. 14–20. (c) The weight of historical evidence also indicates Congress has accepted that the recognition power is exclusive to the Presidency. Cf. NLRB v. Noel Canning, 573 U. S. ___. Since the first Administration, the President has claimed unilateral authority to recognize foreign sovereigns. And Congress, for the most part, has acquiesced, generally respecting the Executive’s policies and positions on formal recognition and even defending the President’s constitutional prerogative. Pp. 20–26. 2. Because the power to recognize foreign states resides in the President alone, §214(d) infringes on the Executive’s consistent decision to withhold recognition with respect to Jerusalem. See Nixon v. Administrator of General Services, 433 U. S. 425 . The provision forces the President, through the Secretary of State, to identify, upon request, citizens born in Jerusalem as being born in Israel when, as a matter of United States policy, neither Israel nor any other country is acknowledged as having sovereignty over Jerusalem. If the recognition power is to mean anything, it must mean that the President not only makes the initial, formal recognition determination but also may maintain that determination in his and his agent’s statements. Under international law, recognition may be effected by written or oral declaration. In addition, an act of recognition must leave no doubt as to the intention to grant it. Thus, if Congress could alter the President’s statements on matters of recognition or force him to contradict them, Congress in effect would exercise the recognition power. An “exclusive” Presidential power “disabl[es] the Congress from acting upon the subject.” Youngstown, supra, at 638 (Jackson, J., concurring). If Congress may not pass a law, speaking in its own voice, effecting formal recognition, then it may not force the President, through §214(d), to contradict his prior recognition determination in an official document issued by the Secretary of State. See Urtetiqui v. D’Arcy, 9 Pet. 692, 698. Section 214(d)’s flaw is further underscored by the fact that the statute’s purpose was to infringe on the President’s exclusive recognition power. While Congress may have power to enact passport legislation of wide scope, it may not “aggrandiz[e] its power at the expense of another branch” by requiring the President to contradict an earlier recognition determination in an official Executive Branch document. Freytag v. Commissioner, 501 U. S. 868 . Pp. 26–29. 725 F. 3d 197, affirmed. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Breyer, J., filed a concurring opinion. Thomas, J., filed an opinion concurring in the judgment in part and dissenting in part. Roberts, C. J., filed a dissenting opinion, in which Alito, J., joined. Scalia, J., filed a dissenting opinion, in which Roberts, C. J., and Alito, J., joined.

A delicate subject lies in the background of this case. That subject is Jerusalem. Questions touching upon the history of the ancient city and its present legal and international status are among the most difficult and complex in international affairs. In our constitutional system these matters are committed to the Legislature and the Executive, not the Judiciary. As a result, in this opinion the Court does no more, and must do no more, than note the existence of international debate and tensions respecting Jerusalem. Those matters are for Congress and the President to discuss and consider as they seek to shape the Nation’s foreign policies. The Court addresses two questions to resolve the interbranch dispute now before it. First, it must determine whether the President has the exclusive power to grant formal recognition to a foreign sovereign. Second, if he has that power, the Court must determine whether Congress can command the President and his Secretary of State to issue a formal statement that contradicts the earlier recognition. The statement in question here is a congressional mandate that allows a United States citizen born in Jerusalem to direct the President and Secretary of State, when issuing his passport, to state that his place of birth is “Israel.” I A Jerusalem’s political standing has long been, and remains, one of the most sensitive issues in American foreign policy, and indeed it is one of the most delicate issues in current international affairs. In 1948, President Truman formally recognized Israel in a signed statement of “recognition.” See Statement by the President Announcing Recognition of the State of Israel, Public Papers of the Presidents, May 14, 1948, p. 258 (1964). That statement did not recognize Israeli sovereignty over Jerusalem. Over the last 60 years, various actors have sought to assert full or partial sovereignty over the city, including Israel, Jordan, and the Palestinians. Yet, in contrast to a consistent policy of formal recognition of Israel, neither President Truman nor any later United States President has issued an official statement or declaration acknowledging any country’s sovereignty over Jerusalem. Instead, the Executive Branch has maintained that “ ‘the status of Jerusalem . . . should be decided not unilaterally but in consultation with all concerned.’ ” United Nations Gen. Assembly Official Records, 5th Emergency Sess., 1554th Plenary Meetings, United Nations Doc. No. 1 A⁄PV.1554, p. 10 (July 14, 1967); see, e.g., Remarks by President Obama in Address to the United Nations Gen. Assembly (Sept. 21, 2011), 2011 Daily Comp. of Pres. Doc. No. 00661, p. 4 (“Ultimately, it is the Israelis and the Palestinians, not us, who must reach agreement on the issues that divide them,” including “Jerusalem”). In a letter to Congress then-Secretary of State Warren Christopher expressed the Executive’s concern that “[t]here is no issue related to the Arab-Israeli negotiations that is more sensitive than Jerusalem.” See 141 Cong. Rec. 28967 (1995) (letter to Robert Dole, Majority Leader, (June 20, 1995)). He further noted the Executive’s opinion that “any effort . . . to bring it to the forefront” could be “very damaging to the success of the peace process.” Ibid. The President’s position on Jerusalem is reflected in State Department policy regarding passports and consular reports of birth abroad. Understanding that passports will be construed as reflections of American policy, the State Department’s Foreign Affairs Manual instructs its employees, in general, to record the place of birth on a passport as the “country [having] present sovereignty over the actual area of birth.” Dept. of State, 7 Foreign Affairs Manual (FAM) §1383.4 (1987). If a citizen objects to the country listed as sovereign by the State Department, he or she may list the city or town of birth rather than the country. See id., §1383.6. The FAM, however, does not allow citizens to list a sovereign that conflicts with Executive Branch policy. See generally id., §1383. Because the United States does not recognize any country as having sovereignty over Jerusalem, the FAM instructs employees to record the place of birth for citizens born there as “Jerusalem.” Id., §1383.5–6 (emphasis deleted). In 2002, Congress passed the Act at issue here, the Foreign Relations Authorization Act, Fiscal Year 2003, 116Stat. 1350. Section 214 of the Act is titled “United States Policy with Respect to Jerusalem as the Capital of Israel.” Id., at 1365. The subsection that lies at the heart of this case, §214(d), addresses passports. That subsection seeks to override the FAM by allowing citizens born in Jerusalem to list their place of birth as “Israel.” Titled “Record of Place of Birth as Israel for Passport Purposes,” §214(d) states “[f ]or purposes of the registration of birth, certification of nationality, or issuance of a passport of a United States citizen born in the city of Jerusalem, the Secretary shall, upon the request of the citizen or the citizen’s legal guardian, record the place of birth as Israel.” Id., at 1366. When he signed the Act into law, President George W. Bush issued a statement declaring his position that §214 would, “if construed as mandatory rather than advisory, impermissibly interfere with the President’s constitutional authority to formulate the position of the United States, speak for the Nation in international affairs, and determine the terms on which recognition is given to foreign states.” Statement on Signing the Foreign Relations Authorization Act, Fiscal Year 2003, Public Papers of the Presidents, George W. Bush, Vol. 2, Sept. 30, 2002, p. 1698 (2005). The President concluded, “U. S. policy regarding Jerusalem has not changed.” Ibid. Some parties were not reassured by the President’s statement. A cable from the United States Consulate in Jerusalem noted that the Palestine Liberation Organization Executive Committee, Fatah Central Committee, and the Palestinian Authority Cabinet had all issued statements claiming that the Act “ ‘undermines the role of the U. S. as a sponsor of the peace process.’ ” App. 231. In the Gaza Strip and elsewhere residents marched in protest. See The Associated Press and Reuters, Palestinians Stone Police Guarding Western Wall, The Seattle Times, Oct. 5, 2002, p. A7. In response the Secretary of State advised diplomats to express their understanding of “Jerusalem’s importance to both sides and to many others around the world.” App. 228. He noted his belief that America’s “policy towards Jerusalem” had not changed. Ibid. B In 2002, petitioner Menachem Binyamin Zivotofsky was born to United States citizens living in Jerusalem. App. 24–25. In December 2002, Zivotofsky’s mother visited the American Embassy in Tel Aviv to request both a passport and a consular report of birth abroad for her son. Id., at 25. She asked that his place of birth be listed as “ ‘Jerusalem, Israel.’ ” Ibid. The Embassy clerks explained that, pursuant to State Department policy, the passport would list only “Jerusalem.” Ibid. Zivotofsky’s parents objected and, as his guardians, brought suit on his behalf in the United States District Court for the District of Columbia, seeking to enforce §214(d). Pursuant to §214(d), Zivotofsky claims the right to have “Israel” recorded as his place of birth in his passport. See Zivotofsky v. Clinton, 566 U. S. ___, ___ (2012) (slip op., at 4) (“[W]hile Zivotofsky had originally asked that ‘Jerusalem, Israel’ be recorded on his passport, ‘[b]oth sides agree that the question now is whether §214(d) entitles [him] to have just ‘Israel’ listed’ ”). The arguments in Zivotofsky’s brief center on his passport claim, as opposed to the consular report of birth abroad. Indeed, in the court below, Zivotofsky waived any argument that his consular report of birth abroad should be treated differently than his passport. Zivotofsky v. Secretary of State, 725 F. 3d 197, 203, n. 3 (CADC 2013). He has also waived the issue here by failing to differentiate between the two documents. As a result, the Court addresses Zivotofsky’s passport arguments and need not engage in a separate analysis of the validity of §214(d) as applied to consular reports of birth abroad. After Zivotofsky brought suit, the District Court dismissed his case, reasoning that it presented a nonjusticiable political question and that Zivotofsky lacked standing. App. 28–39. The Court of Appeals for the District of Columbia Circuit reversed on the standing issue, Zivotofsky v. Secretary of State, 444 F. 3d 614, 617–619 (2006), but later affirmed the District Court’s political question determination. See Zivotofsky v. Secretary of State, 571 F. 3d 1227, 1228 (2009). This Court granted certiorari, vacated the judgment, and remanded the case. Whether §214(d) is constitutional, the Court held, is not a question reserved for the politi-cal branches. In reference to Zivotofsky’s claim the Court observed “the Judiciary must decide if Zivotofsky’s interpretation of the statute is correct, and whether the statute is constitutional”—not whether Jerusalem is, in fact, part of Israel. Zivotofsky v. Clinton, supra, at___ (slip op., at 7). On remand the Court of Appeals held the statute unconstitutional. It determined that “the President exclusively holds the power to determine whether to recognize a foreign sovereign,” 725 F. 3d, at 214, and that “section 214(d) directly contradicts a carefully considered exercise of the Executive branch’s recognition power.” Id., at 217. This Court again granted certiorari. 572 U. S. ___ (2014). II In considering claims of Presidential power this Court refers to Justice Jackson’s familiar tripartite framework from Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579 –638 (1952) (concurring opinion). The framework divides exercises of Presidential power into three categories: First, when “the President acts pursuant to an express or implied authorization of Congress, his authority is at its maximum, for it includes all that he possesses in his own right plus all that Congress can delegate.” Id., at 635. Second, “in absence of either a congressional grant or denial of authority” there is a “zone of twilight in which he and Congress may have concurrent authority,” and where “congressional inertia, indifference or quiescence may” invite the exercise of executive power. Id., at 637. Finally, when “the President takes measures incompatible with the expressed or implied will of Congress . . . he can rely only upon his own constitutional powers minus any constitutional powers of Congress over the matter.” Ibid. To succeed in this third category, the President’s asserted power must be both “exclusive” and “conclusive” on the issue. Id., at 637–638. In this case the Secretary contends that §214(d) in-fringes on the President’s exclusive recognition power by “requiring the President to contradict his recognition posi-tion regarding Jerusalem in official communications with foreign sovereigns.” Brief for Respondent 48. In so doing the Secretary acknowledges the President’s power is “at its lowest ebb.” Youngstown, 343 U. S., at 637. Because the President’s refusal to implement §214(d) falls into Justice Jackson’s third category, his claim must be “scrutinized with caution,” and he may rely solely on powers the Constitution grants to him alone. Id., at 638. To determine whether the President possesses the exclusive power of recognition the Court examines the Constitution’s text and structure, as well as precedent and history bearing on the question. A Recognition is a “formal acknowledgement” that a particular “entity possesses the qualifications for statehood” or “that a particular regime is the effective government of a state.” Restatement (Third) of Foreign Relations Law of the United States §203, Comment a, p. 84 (1986). It may also involve the determination of a state’s territorial bounds. See 2 M. Whiteman, Digest of International Law §1, p. 1 (1963) (Whiteman) (“[S]tates may recognize or decline to recognize territory as belonging to, or under the sovereignty of, or having been acquired or lost by, other states”). Recognition is often effected by an express “written or oral declaration.” 1 J. Moore, Digest of International Law §27, p. 73 (1906) (Moore). It may also be implied—for example, by concluding a bilateral treaty or by sending or receiving diplomatic agents. Ibid.; I. Brownlie, Prin-ciples of Public International Law 93 (7th ed. 2008) (Brownlie). Legal consequences follow formal recognition. Recognized sovereigns may sue in United States courts, see Guaranty Trust Co. v. United States, 304 U. S. 126, 137 (1938) , and may benefit from sovereign immunity when they are sued, see National City Bank of N. Y. v. Republic of China, 348 U. S. 356 –359 (1955). The actions of a recognized sovereign committed within its own territory also receive deference in domestic courts under the act of state doctrine. See Oetjen v. Central Leather Co., 246 U. S. 297 –303 (1918). Recognition at international law, furthermore, is a precondition of regular diplomatic relations. 1 Moore §27, at 72. Recognition is thus “useful, even necessary,” to the existence of a state. Ibid. Despite the importance of the recognition power in foreign relations, the Constitution does not use the term “recognition,” either in Article II or elsewhere. The Secretary asserts that the President exercises the recognition power based on the Reception Clause, which directs that the President “shall receive Ambassadors and other public Ministers.” Art. II, §3. As Zivotofsky notes, the Reception Clause received little attention at the Constitutional Convention. See Reinstein, Recognition: A Case Study on the Original Understanding of Executive Power, 45 U. Rich. L. Rev. 801, 860–862 (2011). In fact, during the ratification debates, Alexander Hamilton claimed that the power to receive ambassadors was “more a matter of dignity than of authority,” a ministerial duty largely “without consequence.” The Federalist No. 69, p. 420 (C. Rossiter ed. 1961). At the time of the founding, however, prominent international scholars suggested that receiving an ambassador was tantamount to recognizing the sovereignty of the sending state. See E. de Vattel, The Law of Nations §78, p. 461 (1758) (J. Chitty ed. 1853) (“[E]very state, truly possessed of sovereignty, has a right to send ambassadors” and “to contest their right in this instance” is equivalent to “contesting their sovereign dignity”); see also 2 C. van Bynkershoek, On Questions of Public Law 156–157 (1737) (T. Frank ed. 1930) (“Among writers on public law it is usually agreed that only a sovereign power has a right to send ambassadors”); 2 H. Grotius, On the Law of War and Peace 440–441 (1625) (F. Kelsey ed. 1925) (discussing the duty to admit ambassadors of sovereign powers). It is a logical and proper inference, then, that a Clause directing the President alone to receive ambassadors would be understood to acknowledge his power to recognize other nations. This in fact occurred early in the Nation’s history when President Washington recognized the French Revolutionary Government by receiving its ambassador. See A. Hamilton, Pacificus No. 1, in The Letters of Pacificus and Helvidius 5, 13–14 (1845) (reprint 1976) (President “acknowledged the republic of France, by the reception of its minister”). After this incident the import of the Reception Clause became clear—causing Hamilton to change his earlier view. He wrote that the Reception Clause “includes th[e power] of judging, in the case of a revolution of government in a foreign country, whether the new rulers are competent organs of the national will, and ought to be recognised, or not.” See id., at 12; see also 3 J. Story, Commentaries on the Constitution of the United States §1560, p. 416 (1833) (“If the executive receives an ambassador, or other minister, as the representative of a new nation . . . it is an acknowledgment of the sovereign authority de facto of such new nation, or party”). As a result, the Reception Clause provides support, although not the sole authority, for the President’s power to recognize other nations. The inference that the President exercises the recognition power is further supported by his additional Article II powers. It is for the President, “by and with the Advice and Consent of the Senate,” to “make Treaties, provided two thirds of the Senators present concur.” Art. II, §2, cl. 2. In addition, “he shall nominate, and by and with the Advice and Consent of the Senate, shall appoint Ambassadors” as well as “other public Ministers and Consuls.” Ibid. As a matter of constitutional structure, these additional powers give the President control over recognition decisions. At international law, recognition may be effected by different means, but each means is dependent upon Presidential power. In addition to receiving an ambassador, recognition may occur on “the conclusion of a bilateral treaty,” or the “formal initiation of diplomatic relations,” including the dispatch of an ambassador. Brownlie 93; see also 1 Moore §27, at 73. The President has the sole power to negotiate treaties, see United States v. Curtiss-Wright Export Corp., 299 U. S. 304, 319 (1936) , and the Senate may not conclude or ratify a treaty without Presidential action. The President, too, nominates the Nation’s ambassadors and dispatches other diplomatic agents. Congress may not send an ambassador without his involvement. Beyond that, the President himself has the power to open diplomatic channels simply by engaging in direct diplo-macy with foreign heads of state and their ministers. The Constitution thus assigns the President means to effect recognition on his own initiative. Congress, by contrast, has no constitutional power that would enable it to initiate diplomatic relations with a foreign nation. Because these specific Clauses confer the recognition power on the President, the Court need not consider whether or to what extent the Vesting Clause, which provides that the “executive Power” shall be vested in the President, provides further support for the President’s action here. Art. II, §1, cl. 1. The text and structure of the Constitution grant the President the power to recognize foreign nations and governments. The question then becomes whether that power is exclusive. The various ways in which the President may unilaterally effect recognition—and the lack of any similar power vested in Congress—suggest that it is. So, too, do functional considerations. Put simply, the Nation must have a single policy regarding which governments are legitimate in the eyes of the United States and which are not. Foreign countries need to know, before entering into diplomatic relations or commerce with the United States, whether their ambassadors will be received; whether their officials will be immune from suit in federal court; and whether they may initiate lawsuits here to vindicate their rights. These assurances cannot be equivocal. Recognition is a topic on which the Nation must “ ‘speak . . . with one voice.’ ” American Ins. Assn. v. Garamendi, 539 U. S. 396, 424 (2003) (quoting Crosby v. National Foreign Trade Council, 530 U. S. 363, 381 (2000) ). That voice must be the President’s. Between the two political branches, only the Executive has the characteristic of unity at all times. And with unity comes the ability to exercise, to a greater degree, “[d]ecision, activity, secrecy, and dispatch.” The Federalist No. 70, p. 424 (A. Hamilton). The President is capable, in ways Congress is not, of engaging in the delicate and often secret diplomatic contacts that may lead to a decision on recognition. See, e.g., United States v. Pink, 315 U. S. 203, 229 (1942) . He is also better positioned to take the decisive, unequivocal action necessary to recognize other states at international law. 1 Oppenheim’s International Law §50, p. 169 (R. Jennings & A. Watts eds., 9th ed. 1992) (act of recognition must “leave no doubt as to the intention to grant it”). These qualities explain why the Framers listed the traditional avenues of recognition—receiving ambassadors, making treaties, and sending ambassadors—as among the President’s Article II powers. As described in more detail below, the President since the founding has exercised this unilateral power to recognize new states—and the Court has endorsed the practice. See Banco Nacional de Cuba v. Sabbatino, 376 U. S. 398, 410 (1964) ; Pink, supra, at 229; Williams v. Suffolk Ins. Co., 13 Pet. 415, 420 (1839). Texts and treatises on international law treat the President’s word as the final word on recognition. See, e.g., Restatement (Third) of Foreign Relations Law §204, at 89 (“Under the Constitution of the United States the President has exclusive authority to recognize or not to recognize a foreign state or government”); see also L. Henkin, Foreign Affairs and the U. S. Constitution 43 (2d ed. 1996) (“It is no longer questioned that the President does not merely perform the ceremony of receiving foreign ambassadors but also determines whether the United States should recognize or refuse to recognize a foreign government”). In light of this author-ity all six judges who considered this case in the Court of Appeals agreed that the President holds the exclusive recognition power. See 725 F. 3d, at 214 (“[W]e conclude that the President exclusively holds the power to determine whether to recognize a foreign sovereign”); Zivotofsky, 571 F. 3d, at 1231 (“That this power belongs solely to the President has been clear from the earliest days of the Republic”); id., at 1240 (Edwards, J., concurring) (“The Executive has exclusive and unreviewable authority to recognize foreign sovereigns”). It remains true, of course, that many decisions affecting foreign relations—including decisions that may determine the course of our relations with recognized countries—require congressional action. Congress may “regulate Commerce with foreign Nations,” “establish an uniform Rule of Naturalization,” “define and punish Piracies and Felonies committed on the high Seas, and Offences against the Law of Nations,” “declare War,” “grant Letters of Marque and Reprisal,” and “make Rules for the Government and Regulation of the land and naval Forces.” U. S. Const., Art. I, §8. In addition, the President cannot make a treaty or appoint an ambassador without the approval of the Senate. Art. II, §2, cl. 2. The President, furthermore, could not build an American Embassy abroad without congressional appropriation of the necessary funds. Art. I, §8, cl. 1. Under basic separation-of-powers principles, it is for the Congress to enact the laws, including “all Laws which shall be necessary and proper for carrying into Execution” the powers of the Federal Government. §8, cl. 18. In foreign affairs, as in the domestic realm, the Constitution “enjoins upon its branches separateness but interdependence, autonomy but reciprocity.” Youngstown, 343 U. S., at 635 (Jackson, J., concurring). Although the President alone effects the formal act of recognition, Congress’ powers, and its central role in making laws, give it substantial authority regarding many of the policy determinations that precede and follow the act of recognition itself. If Congress disagrees with the President’s recognition policy, there may be consequences. Formal recognition may seem a hollow act if it is not accompanied by the dispatch of an ambassador, the easing of trade restrictions, and the conclusion of treaties. And those decisions require action by the Senate or the whole Congress. In practice, then, the President’s recognition determination is just one part of a political process that may require Congress to make laws. The President’s exclusive recognition power encompasses the authority to acknowledge, in a formal sense, the legitimacy of other states and governments, including their territorial bounds. Albeit limited, the exclusive recognition power is essential to the conduct of Presidential duties. The formal act of recognition is an executive power that Congress may not qualify. If the President is to be effective in negotiations over a formal recognition determination, it must be evident to his counterparts abroad that he speaks for the Nation on that precise question. A clear rule that the formal power to recognize a foreign government subsists in the President therefore serves a necessary purpose in diplomatic relations. All this, of course, underscores that Congress has an important role in other aspects of foreign policy, and the President may be bound by any number of laws Congress enacts. In this way ambition counters ambition, ensuring that the democratic will of the people is observed and respected in foreign affairs as in the domestic realm. See The Federalist No. 51, p. 322 (J. Madison). B No single precedent resolves the question whether the President has exclusive recognition authority and, if so, how far that power extends. In part that is because, until today, the political branches have resolved their disputes over questions of recognition. The relevant cases, though providing important instruction, address the division of recognition power between the Federal Government and the States, see, e.g., Pink, 315 U. S. 203 , or between the courts and the political branches, see, e.g., Banco Nacional de Cuba, 376 U. S., at 410—not between the President and Congress. As the parties acknowledge, some isolated statements in those cases lend support to the position that Congress has a role in the recognition process. In the end, however, a fair reading of the cases shows that the President’s role in the recognition process is both central and exclusive. During the administration of President Van Buren, in a case involving a dispute over the status of the Falkland Islands, the Court noted that “when the executive branch of the government” assumes “a fact in regard to the sovereignty of any island or country, it is conclusive on the judicial department.” Williams, 13 Pet., at 420. Once the President has made his determination, it “is enough to know, that in the exercise of his constitutional functions, he has decided the question. Having done this under the responsibilities which belong to him, it is obligatory on the people and government of the Union.” Ibid. Later, during the 1930’s and 1940’s, the Court addressed issues surrounding President Roosevelt’s decision to recognize the Soviet Government of Russia. In United States v. Belmont, 301 U. S. 324 (1937) , and Pink, 315 U. S. 203 , New York state courts declined to give full effect to the terms of executive agreements the President had concluded in negotiations over recognition of the Soviet regime. In particular the state courts, based on New York public policy, did not treat assets that had been seized by the Soviet Government as property of Russia and declined to turn those assets over to the United States. The Court stated that it “may not be doubted” that “recognition, establishment of diplomatic relations, . . . and agreements with respect thereto” are “within the competence of the President.” Belmont, 301 U. S., at 330. In these matters, “the Executive ha[s] authority to speak as the sole organ of th[e] government.” Ibid. The Court added that the President’s authority “is not limited to a determination of the government to be recognized. It includes the power to determine the policy which is to govern the question of recognition.” Pink, supra, at 229; see also Guaranty Trust Co., 304 U. S., at 137–138 (The “political department[’s] . . . action in recognizing a foreign government and in receiving its diplomatic representatives is conclusive on all domestic courts”). Thus, New York state courts were required to respect the executive agreements. It is true, of course, that Belmont and Pink are not direct holdings that the recognition power is exclusive. Those cases considered the validity of executive agreements, not the initial act of recognition. The President’s determination in those cases did not contradict an Act of Congress. And the primary issue was whether the executive agreements could supersede state law. Still, the language in Pink and Belmont, which confirms the President’s competence to determine questions of recognition, is strong support for the conclusion that it is for the President alone to determine which foreign governments are legitimate. Banco Nacional de Cuba contains even stronger statements regarding the President’s authority over recognition. There, the status of Cuba’s Government and its acts as a sovereign were at issue. As the Court explained, “Political recognition is exclusively a function of the Executive.” 376 U. S., at 410. Because the Executive had recognized the Cuban Government, the Court held that it should be treated as sovereign and could benefit from the “act of state” doctrine. See also Baker v. Carr, 369 U. S. 186, 213 (1962) (“[I]t is the executive that determines a person’s status as representative of a foreign government”); National City Bank of N. Y., 348 U. S., at 358 (“The status of the Republic of China in our courts is a matter for determination by the Executive and is outside the competence of this Court”). As these cases illustrate, the Court has long considered recognition to be the exclusive prerogative of the Executive. The Secretary now urges the Court to define the executive power over foreign relations in even broader terms. He contends that under the Court’s precedent the President has “exclusive authority to conduct diplomatic relations,” along with “the bulk of foreign-affairs powers.” Brief for Respondent 18, 16. In support of his submission that the President has broad, undefined powers over foreign affairs, the Secretary quotes United States v. Curtiss-Wright Export Corp., which described the President as “the sole organ of the federal government in the field of international relations.” 299 U. S., at 320. This Court declines to acknowledge that unbounded power. A formulation broader than the rule that the President alone determines what nations to formally recognize as legitimate—and that he consequently controls his statements on matters of recognition—presents different issues and is unnecessary to the resolution of this case. The Curtiss-Wright case does not extend so far as the Secretary suggests. In Curtiss-Wright, the Court considered whether a congressional delegation of power to the President was constitutional. Congress had passed a joint resolution giving the President the discretion to prohibit arms sales to certain militant powers in South America. The resolution provided criminal penalties for violation of those orders. Id., at 311–312. The Court held that the delegation was constitutional, reasoning that Congress may grant the President substantial authority and discretion in the field of foreign affairs. Id., at 315–329. Describing why such broad delegation may be appropriate, the opinion stated: “In this vast external realm, with its important, complicated, delicate and manifold problems, the President alone has the power to speak or listen as a representative of the nation. He makes treaties with the advice and consent of the Senate; but he alone negotiates. Into the field of negotiation the Senate cannot intrude; and Congress itself is powerless to invade it. As Marshall said in his great argument of March 7, 1800, in the House of Representatives, ‘The President is the sole organ of the nation in its external relations, and its sole representative with foreign nations.’ [10 Annals of Cong.] 613. ” Id., at 319. This description of the President’s exclusive power was not necessary to the holding of Curtiss-Wright—which, after all, dealt with congressionally authorized action, not a unilateral Presidential determination. Indeed, Curtiss-Wright did not hold that the President is free from Congress’ lawmaking power in the field of internationalrelations. The President does have a unique role in communi-cating with foreign governments, as then-Congressman John Marshall acknowledged. See 10 Annals of Cong. 613 (1800) (cited in Curtiss-Wright, supra, at 319). But whether the realm is foreign or domestic, it is still the Legislative Branch, not the Executive Branch, that makes the law. In a world that is ever more compressed and interdependent, it is essential the congressional role in foreign affairs be understood and respected. For it is Congress that makes laws, and in countless ways its laws will and should shape the Nation’s course. The Executive is not free from the ordinary controls and checks of Congress merely because foreign affairs are at issue. See, e.g., Medellín v. Texas, 552 U. S. 491 –532 (2008); Youngstown, 343 U. S., at 589; Little v. Barreme, 2 Cranch 170, 177–179 (1804); Glennon, Two Views of Presidential Foreign Affairs Power: Little v. Barreme or Curtiss-Wright? 13 Yale J. Int’l L. 5, 19–20 (1988); cf. Dames & Moore v. Regan, 453 U. S. 654 –681 (1981). It is not for the President alone to determine the whole content of the Nation’s foreign policy. That said, judicial precedent and historical practice teach that it is for the President alone to make the specific decision of what foreign power he will recognize as legitimate, both for the Nation as a whole and for the purpose of making his own position clear within the context of recognition in discussions and negotiations with foreign nations. Recognition is an act with immediate and powerful significance for international relations, so the President’s position must be clear. Congress cannot require him to contradict his own statement regarding a determination of formal recognition. Zivotofsky’s contrary arguments are unconvincing. The decisions he relies upon are largely inapposite. This Court’s cases do not hold that the recognition power is shared. Jones v. United States, 137 U. S. 202 (1890) , and Boumediene v. Bush, 553 U. S. 723 (2008) , each addressed the status of territories controlled or acquired by the United States—not whether a province ought to be recognized as part of a foreign country. See also Vermilya-Brown Co. v. Connell, 335 U. S. 377, 380 (1948) (“[D]etermination of [American] sovereignty over an area is for the legislative and executive departments”). And no one disputes that Congress has a role in determining the status of United States territories. See U. S. Const., Art. IV, §3, cl. 2 (Congress may “dispose of and make all needful Rules and Regulations respecting the Territory or other Property belonging to the United States”). Other cases describing a shared power address the recognition of Indian tribes—which is, similarly, a distinct issue from the recognition of foreign countries. See Cherokee Nation v. Georgia, 5 Pet. 1 (1831). To be sure, the Court has mentioned both of the political branches in discussing international recognition, but it has done so primarily in affirming that the Judiciary is not responsible for recognizing foreign nations. See Oetjen, 246 U. S., at 302 (“ ‘Who is the sovereign, de jure or de facto, of a territory is not a judicial, but is a political question, the determination of which by the legislative and executive departments of any government conclusively binds the judges’ ” (quoting Jones, supra, at 212)); United States v. Palmer, 3 Wheat. 610, 643 (1818) (“[T]he courts of the union must view [a] newly constituted government as it is viewed by the legislative and executive departments of the government of the United States”). This is consistent with the fact that Congress, in the ordinary course, does support the President’s recognition policy, for instance by confirming an ambassador to the recognized foreign government. Those cases do not cast doubt on the view that the Executive Branch determines whether the United States will recognize foreign states and governments and their territorial bounds. C Having examined the Constitution’s text and this Court’s precedent, it is appropriate to turn to accepted understandings and practice. In separation-of-powers cases this Court has often “put significant weight upon historical practice.” NLRB v. Noel Canning, 573 U. S. ___, ___ (2014) (slip op., at 6) (emphasis deleted). Here, history is not all on one side, but on balance it provides strong support for the conclusion that the recognition power is the President’s alone. As Zivotofsky argues, certain historical incidents can be interpreted to support the position that recognition is a shared power. But the weight of historical evidence supports the opposite view, which is that the formal determination of recognition is a power to be exercised only by the President. The briefs of the parties and amici, which have been of considerable assistance to the Court, give a more complete account of the relevant history, as do the works of scholars in this field. See, e.g., Brief for Respondent 26–39; Brief for Petitioner 34–57; Brief for American Jewish Committee as Amicus Curiae 6–24; J. Goebel, The Recognition Policy of the United States 97–170 (1915) (Goebel); 1 Moore §§28–58, 74–164; Reinstein, Is the President’s Recognition Power Exclusive? 86 Temp. L. Rev. 1, 3–50 (2013). But even a brief survey of the major historical examples, with an emphasis on those said to favor Zivotofsky, establishes no more than that some Presidents have chosen to cooperate with Congress, not that Congress itself has exercised the recognition power. From the first Administration forward, the President has claimed unilateral authority to recognize foreign sovereigns. For the most part, Congress has acquiesced in the Executive’s exercise of the recognition power. On occasion, the President has chosen, as may often be prudent, to consult and coordinate with Congress. As Judge Tatel noted in this case, however, “the most striking thing” about the history of recognition “is what is absent from it: a situation like this one,” where Congress has enacted a statute contrary to the President’s formal and considered statement concerning recognition. 725 F. 3d, at 221 (concurring opinion). The first debate over the recognition power arose in 1793, after France had been torn by revolution. See Prakash & Ramsey, The Executive Power over Foreign Affairs, 111 Yale L. J. 231, 312 (2001). Once the Revolutionary Government was established, Secretary of State Jefferson and President Washington, without consulting Congress, authorized the American Ambassador to resume relations with the new regime. See Letter to Gouverneur Morris (Mar. 12, 1793), in 25 Papers of Thomas Jefferson 367, 367–368 (J. Catanzariti ed. 1992); Goebel 99–104. Soon thereafter, the new French Government proposed to send an ambassador, Citizen Genet, to the United States. See id., at 105. Members of the President’s Cabinet agreed that receiving Genet would be a binding and public act of recognition. See Opinion on the Treaties with France (Apr. 28, 1793), in 25 Papers of Thomas Jefferson, at 608, 612 (“The reception of the Minister at all . . . is an ackno[w]le[d]gement of the legitimacy of their government”); see also Letter from A. Hamilton to G. Washington (Cabinet Paper) (Apr. 1793), in 4 Works of Alexander Hamilton 369, 369–396 (H. Lodge ed. 1904). They de-cided, however, both that Genet should be received and that consultation with Congress was not necessary. See T. Jefferson, Anas (Apr. 18, 1793), in 1 Writings of Thomas Jefferson 226, 227 (P. Ford ed. 1892); Cabinet Opinion on Washington’s Questions on Neutrality and the Alliance with France (Apr. 19, 1793), in 25 Papers of Thomas Jefferson, at 570. Congress expressed no disagreement with this position, and Genet’s reception marked the Nation’s first act of recognition—one made by the President alone. See Prakash, supra, at 312–313. The recognition power again became relevant when yet another revolution took place—this time, in South America, as several colonies rose against Spain. In 1818, Speakerof the House Henry Clay announced he “intended mov-ing the recognition of Buenos Ayres and probably of Chile.” Goebel 121. Clay thus sought to appropriate money “ ‘[f ]or one year’s salary’ ” for “ ‘a Minister’ ” topresent-day Argentina. 32 Annals of Cong. 1500 (1818). President Monroe, however, did not share that view. Although Clay gave “one of the most remarkable speeches of his career,” his proposed bill was defeated. Goebel 123; 32 Annals of Cong. 1655. That action has been attributed, in part, to the fact that Congress agreed the recognition power rested solely with the President. Goebel 124; see, e.g., 32 Annals of Cong. 1570 (statement of Rep. Alexander Smyth) (“[T]he acknowledgment of the independence of a new Power is an exercise of Executive authority; consequently, for Congress to direct the Executive how he shall exercise this power, is an act of usurpation”). Four years later, after the President had decided to recognize the South American republics, Congress did pass a resolution, on his request, appropriating funds for “such missions to the independent nations on the American continent, as the President of the United States may deem proper.” Act of May 4, 1822, ch. 52, 3Stat. 678. A decade later, President Jackson faced a recognition crisis over Texas. In 1835, Texas rebelled against Mexico and formed its own government. See Goebel 144–147. But the President feared that recognizing the new government could ignite a war. See A. Jackson, To the Senate and House of Representatives of the United States (Dec. 21, 1836), in 3 Messages and Papers of the Presidents 265, 266–267 (J. Richardson ed. 1899). After Congress urged him to recognize Texas, see Cong. Globe, 24th Cong., 1st Sess., 453 (1836); H. R. Rep. No. 854, 24th Cong., 1st Sess. (1836), the President delivered a message to the Legislature. He concluded there had not been a “deliberate inquiry” into whether the President or Congress possessed the recognition power. See A. Jackson, in 3 Messages and Papers of the Presidents, at 267. He stated, however, “on the ground of expediency, I am disposed to concur” with Congress’ preference regarding Texas. Ibid. In response Congress appropriated funds for a “diplomatic agent to be sent to the Republic of Texas, whenever the President of the United States . . . shall deem it expedient to appoint such minister.” Act of Mar. 3, 1837, 5Stat. 170. Thus, although he cooperated with Congress, the President was left to execute the formal act of recognition. President Lincoln, too, sought to coordinate with Congress when he requested support for his recognition of Liberia and Haiti. In his first annual message to Congress he said he could see no reason “why we should persevere longer in withholding our recognition of the independence and sovereignty of Hayti and Liberia.” Lincoln’s First Annual Message to Congress (Dec. 3, 1861), in 6 Messages and Papers of the Presidents 44, 47. Nonetheless, he was “[u]nwilling” to “inaugurate a novel policy in regard to them without the approbation of Congress.” Ibid. In response Congress concurred in the President’s recognition determination and enacted a law appropriating funds to appoint diplomatic representatives to the two countries—leaving, as usual, the actual dispatch of ambassadors and formal statement of recognition to the President. Act of June 5, 1862, 12Stat. 421. Three decades later, the branches again were able to reach an accord, this time with regard to Cuba. In 1898, an insurgency against the Spanish colonial government was raging in Cuba. President McKinley determined to ask Congress for authorization to send armed forces to Cuba to help quell the violence. See 31 Cong. Rec. 3699–3702 (1898). Although McKinley thought Spain was to blame for the strife, he opposed recognizing either Cuba or its insurgent government. Id., at 3701. At first, the House proposed a resolution consistent with McKinley’s wishes. Id., at 3810. The Senate countered with a resolution that authorized the use of force but that did recognize both Cuban independence and the insurgent government. Id., at 3993. When the Senate’s version reached the House, the House again rejected the language recognizing Cuban independence. Id., at 4017. The resolution went to Conference, which, after debate, reached a compromise. See Reinstein, 86 Temp. L. Rev., at 40–41. The final resolution stated “the people of the Island of Cuba are, and of right ought to be, free and independent,” but made no mention of recognizing a new Cuban Government. Act of Apr. 20, 1898, 30Stat. 738. Accepting the compromise, the President signed the joint resolution. See Reinstein, 86 Temp. L. Rev., at 41. For the next 80 years, “[P]residents consistently recognized new states and governments without any serious opposition from, or activity in, Congress.” Ibid.; see 2 Whiteman §§6–60, at 133–242 (detailing over 50 recognition decisions made by the Executive). The next debate over recognition did not occur until the late 1970’s. It concerned China. President Carter recognized the People’s Republic of China (PRC) as the government of China, and derecognized the Republic of China, located on Taiwan. See S. Kan, Cong. Research Serv., China/Taiwan: Evolution of the “One China” Policy—Key Statements from Washington, Beijing, and Taipei 1, 10 (Oct. 10, 2014). As to the status of Taiwan, the President “acknowledge[d] the Chinese position” that “Taiwan is part of China,” id., at 39 (text of U. S.–PRC Joint Communique on the Establishment of Diplomatic Relations (Jan. 1, 1979)), but he did not accept that claim. The President proposed a new law defining how the United States would conduct business with Taiwan. See Hearings on Taiwan Legislation before the House Committee on Foreign Affairs, 96th Cong., 1st Sess., 2–6 (1979) (statement of Warren Christopher, Dep-uty Secretary of State). After extensive revisions, Congress passed, and the President signed, the Taiwan Relations Act, 93Stat. 14 (1979) (codified as amended at 22 U. S. C. §§3301–3316). The Act (in a simplified summary) treated Taiwan as if it were a legally distinct entity from China—an entity with which the United States intended to maintain strong ties. See, e.g., §§3301, 3303(a), (b)(1), (b)(7). Throughout the legislative process, however, no one raised a serious question regarding the President’s exclusive authority to recognize the PRC—or to decline to grant formal recognition to Taiwan. See, e.g., 125 Cong. Rec. 6709 (1979) (statement of Sen. Jacob Javits) (“Neither bill [proposed by either Chamber] sought to reestablish official relations between the United States and the Republic of China on Taiwan; Congress . . . does not have the author-ity to do that even if it wanted to do so”). Rather, Congress accepted the President’s recognition determination as a completed, lawful act; and it proceeded to outline the trade and policy provisions that, in its judgment, were appropriate in light of that decision. This history confirms the Court’s conclusion in the instant case that the power to recognize or decline to recognize a foreign state and its territorial bounds resides in the President alone. For the most part, Congress has respected the Executive’s policies and positions as to formal recognition. At times, Congress itself has defended the President’s constitutional prerogative. Over the last 100 years, there has been scarcely any debate over the President’s power to recognize foreign states. In this respect the Legislature, in the narrow context of recognition, on balance has acknowledged the importance of speaking “with one voice.” Crosby, 530 U. S., at 381. The weight of historical evidence indicates Congress has accepted that the power to recognize foreign states and governments and their territorial bounds is exclusive to the Presidency. III As the power to recognize foreign states resides in the President alone, the question becomes whether §214(d) infringes on the Executive’s consistent decision to withhold recognition with respect to Jerusalem. See Nixon v. Administrator of General Services, 433 U. S. 425, 443 (1977) (action unlawful when it “prevents the Executive Branch from accomplishing its constitutionally assigned functions”). Section 214(d) requires that, in a passport or consular report of birth abroad, “the Secretary shall, upon the request of the citizen or the citizen’s legal guardian, record the place of birth as Israel” for a “United States citizen born in the city of Jerusalem.” 116Stat. 1366. That is, §214(d) requires the President, through the Secretary, to identify citizens born in Jerusalem who so request as being born in Israel. But according to the President, those citizens were not born in Israel. As a matter of United States policy, neither Israel nor any other country is acknowledged as having sovereignty over Jerusalem. In this way, §214(d) “directly contradicts” the “carefully calibrated and longstanding Executive branch policy of neutrality toward Jerusalem.” 725 F. 3d, at 217, 216. If the power over recognition is to mean anything, it must mean that the President not only makes the initial, formal recognition determination but also that he may maintain that determination in his and his agent’s statements. This conclusion is a matter of both common sense and necessity. If Congress could command the President to state a recognition position inconsistent with his own, Congress could override the President’s recognition determination. Under international law, recognition may be effected by “written or oral declaration of the recognizing state.” 1 Moore §27, at 73. In addition an act of recognition must “leave no doubt as to the intention to grant it.” 1 Oppenheim’s International Law §50, at 169. Thus, if Congress could alter the President’s statements on matters of recognition or force him to contradict them, Congress in effect would exercise the recognition power. As Justice Jackson wrote in Youngstown, when a Presidential power is “exclusive,” it “disabl[es] the Congress from acting upon the subject.” 343 U. S., at 637–638 (concurring opinion). Here, the subject is quite narrow: The Executive’s exclusive power extends no further than his formal recognition determination. But as to that determination, Congress may not enact a law that directly contradicts it. This is not to say Congress may not express its disagreement with the President in myriad ways. For example, it may enact an embargo, decline to confirm an ambassador, or even declare war. But none of these acts would alter the President’s recognition decision. If Congress may not pass a law, speaking in its own voice, that effects formal recognition, then it follows that it may not force the President himself to contradict his earlier statement. That congressional command would not only prevent the Nation from speaking with one voice but also prevent the Executive itself from doing so in conducting foreign relations. Although the statement required by §214(d) would not itself constitute a formal act of recognition, it is a mandate that the Executive contradict his prior recognition determination in an official document issued by the Secretary of State. See Urtetiqui v. D’Arcy, 9 Pet. 692, 699 (1835) (a passport “from its nature and object, is addressed to foreign powers” and “is to be considered . . . in the character of a political document”). As a result, it is unconstitu-tional. This is all the more clear in light of the longstanding treatment of a passport’s place-of-birth section as an official executive statement implicating recognition. See 725 F. 3d, at 224 (Tatel, J., concurring). The Secretary’s position on this point has been consistent: He will not place information in the place-of-birth section of a passport that contradicts the President’s recognition policy. See 7 FAM §1383. If a citizen objects to the country listed as sovereign over his place of birth, then the Secretary will accommodate him by listing the city or town of birth rather than the country. See id., §1383.6. But the Secretary will not list a sovereign that contradicts the President’s recognition policy in a passport. Thus, the Secretary will not list “Israel” in a passport as the country containing Jerusalem. The flaw in §214(d) is further underscored by the undoubted fact that that the purpose of the statute was to infringe on the recognition power—a power the Court now holds is the sole prerogative of the President. The statute is titled “United States Policy with Respect to Jerusalem as the Capital of Israel.” §214, 116Stat. 1365. The House Conference Report proclaimed that §214 “contains four provisions related to the recognition of Jerusalem as Israel’s capital.” H. R. Conf. Rep. No. 107–671, p. 123 (2002). And, indeed, observers interpreted §214 as altering United States policy regarding Jerusalem—which led to protests across the region. See supra, at 4. From the face of §214, from the legislative history, and from its reception, it is clear that Congress wanted to express its displeasure with the President’s policy by, among other things, commanding the Executive to contradict his own, earlier stated position on Jerusalem. This Congress may not do. It is true, as Zivotofsky notes, that Congress has substantial authority over passports. See Haig v. Agee, 453 U. S. 280 (1981); Zemel v. Rusk, 381 U. S. 1 (1965) ; Kent v. Dulles, 357 U. S. 116 (1958) . The Court does not question the power of Congress to enact passport legislation of wide scope. In Kent v. Dulles, for example, the Court held that if a person’s “ ‘liberty’ ” to travel “is to be regulated” through a passport, “it must be pursuant to the law-making functions of the Congress.” See id., at 129. Later cases, such as Zemel v. Rusk and Haig v. Agee, also proceeded on the assumption that Congress must authorize the grounds on which passports may be approved or denied. See Zemel, supra, at 7–13; Haig, supra, at 289–306. This is consistent with the extensive lawmaking power the Constitution vests in Congress over the Nation’s foreign affairs. The problem with §214(d), however, lies in how Congress exercised its authority over passports. It was an improper act for Congress to “aggrandiz[e] its power at the expense of another branch” by requiring the President to contradict an earlier recognition determination in an official document issued by the Executive Branch. Freytag v. Commissioner, 501 U. S. 868, 878 (1991) . To allow Congress to control the President’s communication in the context of a formal recognition determination is to allow Congress to exercise that exclusive power itself. As a result, the statute is unconstitutional. * * * In holding §214(d) invalid the Court does not question the substantial powers of Congress over foreign affairs in general or passports in particular. This case is confined solely to the exclusive power of the President to control recognition determinations, including formal statements by the Executive Branch acknowledging the legitimacy of a state or government and its territorial bounds. Congress cannot command the President to contradict an earlier recognition determination in the issuance of passports. The judgment of the Court of Appeals for the District of Columbia Circuit is Affirmed.

577.US.2015_14-1382

Respondents, corporate citizens of Delaware, Nebraska, and Illinois, sued petitioner Americold Realty Trust, a “real estate investment trust” organized under Maryland law, in a Kansas court. Americold removed the suit to Federal District Court based on diversity-of-citizenship jurisdiction. See 28 U. S. C. §§1332(a)(1), 1441(b). The District Court accepted jurisdiction and ruled in Americold’s favor. On appeal, the Tenth Circuit held that the District Court lacked jurisdiction to hear the suit. Since Americold was not a corporation, the court reasoned, its citizenship for diversity jurisdiction purposes should be based on the citizenship of its members, which included its shareholders. Because no record of those shareholders’ citizenship existed, diversity was not proved. Held: For purposes of diversity jurisdiction, Americold’s citizenship is based on the citizenship of its members, which include its shareholders. Pp. 2–6. (a) Historically, the relevant citizens for jurisdictional purposes in a suit involving a “mere legal entity” were that entity’s “members,” or the “real persons who come into court” in the entity’s name. Bank of United States v. Deveaux, 5 Cranch 61, 86, 91. But for the limited exception of jurisdictional citizenship for corporations, see Louisville, C. & C. R. Co. v. Letson, 2 How. 497, 558, this Court continues to “adhere to [the] oft-repeated rule that diversity jurisdiction in a suit by or against the entity depends on the citizenship of ‘all [its] members,’ ” Carden v. Arkoma Associates, 494 U. S. 185 . Applying the rule here, Americold possesses the citizenship of all its members, who, under Maryland law, include its shareholders. See, e.g., Md. Corp. & Assns. Code Ann. §8–101(c). Pp. 2–4. (b) Americold argues that anything called a “trust” possesses the citizenship of its trustees alone. Traditionally, a trust was considered a “fiduciary relationship” between multiple people and could not be haled into court; hence, legal proceedings involving a trust were brought by or against the trustees in their own name, Deveaux, 5 Cranch, at 91. Americold confuses the traditional trust with the variety of unincorporated entities that many States have given the “trust” label. Under Maryland law, the real estate investment trust at issue is treated as a “separate legal entity” that can sue or be sued. §§8–102(2), 8–301(2). Despite what such an entity calls itself, so long as it is unincorporated, this Court will apply the “oft-repeated rule” that it possesses the citizenship of all its members. Pp. 4–6. 776 F. 3d 1175, affirmed. Sotomayor, J., delivered the opinion for a unanimous Court.

Federal law permits federal courts to resolve certain nonfederal controversies between “citizens” of different States. This rule is easy enough to apply to humans, but can become metaphysical when applied to legal entities. This case asks how to determine the citizenship of a “real estate investment trust,” an inanimate creature of Maryland law. We answer: While humans and corporations can assert their own citizenship, other entities take the citizenship of their members. I This action began as a typical state-law controversy, one involving a contract dispute and an underground food-storage warehouse fire. A group of corporations whose food perished in that 1991 fire continues to seek compensation from the warehouse’s owner, now known as Americold Realty Trust. After the corporations filed their latest suit in Kansas court, Americold removed the suit to the Federal District Court for the District of Kansas. The District Court accepted jurisdiction and resolved the dispute in favor of Americold. On appeal, however, the Tenth Circuit asked for supplemental briefing on whether the District Court’s exercise of jurisdiction was appropriate. The parties responded that the District Court possessed jurisdiction because the suit involved “citizens of different States.” 28 U. S. C. §§1332(a)(1), 1441(b). The Tenth Circuit disagreed. The court considered the corporate plaintiffs citizens of the States where they were chartered and had their principal places of business: Delaware, Nebraska, and Illinois. See ConAgra Foods, Inc. v. Americold Logistics, LLC, 776 F. 3d 1175, 1182 (2015); §1332(c)(1) (specifying the citizenship of corporations for jurisdictional purposes). The court applied a different test to determine Americold’s citizenship because Americold is a “real estate investment trust,” not a corporation. Distilling this Court’s precedent, the Tenth Circuit reasoned that the citizenship of any “non-corporate artificial entity” is determined by considering all of the entity’s “members,” which include, at minimum, its shareholders. Id., at 1180–1181 (citing Carden v. Arkoma Associates, 494 U. S. 185 (1990) ). As there was no record of the citizenship of Americold’s shareholders, the court concluded that the parties failed to demonstrate that the plaintiffs were “citizens of different States” than the defendants. See Strawbridge v. Curtiss, 3 Cranch 267 (1806). We granted certiorari to resolve confusion among the Courts of Appeals regarding the citizenship of unincorporated entities. 576 U. S. ___ (2015). We now affirm. II Exercising its powers under Article III, the First Congress granted federal courts jurisdiction over controversies between a “citizen” of one State and “a citizen of another State.” 1Stat. 78. For a long time, however, Congress failed to explain how to determine the citizenship of a nonbreathing entity like a business association. In the early 19th century, this Court took that silence literally, ruling that only a human could be a citizen for jurisdictional purposes. Bank of United States v. Deveaux, 5 Cranch 61, 86–91 (1809). If a “mere legal entity” like a corporation were sued, the relevant citizens were its “members,” or the “real persons who come into court” in the entity’s name. Id., at 86, 91. This Court later carved a limited exception for corporations, holding that a corporation itself could be considered a citizen of its State of incorporation. See Louisville, C. & C. R. Co. v. Letson, 2 How. 497, 558 (1844). Congress etched this exception into the U. S. Code, adding that a corporation should also be considered a citizen of the State where it has its principal place of business. 28 U. S. C. §1332(c) (1958 ed.). But Congress never expanded this grant of citizenship to include artificial entities other than corporations, such as joint-stock companies or limited partnerships. For these unincorporated entities, we too have “adhere[d] to our oft-repeated rule that diversity jurisdiction in a suit by or against the entity depends on the citizenship of ‘all [its] members.’ ” Carden, 494 U. S., at 195–196 (quoting Chapman v. Barney, 129 U. S. 677, 682 (1889) ). Despite our oft-repetition of the rule linking unincorporated entities with their “members,” we have never expressly defined the term. But we have equated an association’s members with its owners or “ ‘the several persons composing such association.’ ” Carden, 494 U. S., at 196 (quoting Great Southern Fire Proof Hotel Co. v. Jones, 177 U. S. 449, 456 (1900) ). Applying this principle with reference to specific States’ laws, we have identified the members of a joint-stock company as its shareholders, the members of a partnership as its partners, the members of a union as the workers affiliated with it, and so on. See Carden, 494 U. S., at 189–190 (citing Chapman, 129 U. S., at 682; Great Southern, 177 U. S., at 457; and Steelworkers v. R. H. Bouligny, Inc., 382 U. S. 145, 146 (1965) ). This case asks us to determine the citizenship of Americold Realty Trust, a “real estate investment trust” organized under Maryland law. App. 93. As Americold is not a corporation, it possesses its members’ citizenship. Nothing in the record designates who Americold’s members are. But Maryland law provides an answer. In Maryland, a real estate investment trust is an “unincorporated business trust or association” in which prop-erty is held and managed “for the benefit and profit of any person who may become a shareholder.” Md. Corp. & Assns. Code Ann. §§8–101(c), 8–102 (2014). As with joint-stock companies or partnerships, shareholders have “ownership interests” and votes in the trust by virtue of their “shares of beneficial interest.” §§8–704(b)(5), 8–101(d). These shareholders appear to be in the same position as the shareholders of a joint-stock company or the partners of a limited partnership—both of whom we viewed as members of their relevant entities. See Carden, 494 U. S., at 192–196; see also §8–705(a) (linking the term “beneficial interests” with “membership interests” and “partnership interests”). We therefore conclude that for purposes of diversity jurisdiction, Americold’s members include its shareholders. III Americold disputes this conclusion. It cites a case called Navarro Savings Assn. v. Lee, 446 U. S. 458 (1980) , to argue that anything called a “trust” possesses the citizenship of its trustees alone, not its shareholder beneficiaries as well. As we have reminded litigants before, however, “Navarro had nothing to do with the citizenship of [a] ‘trust.’ ” Carden, 494 U. S., at 192–193. Rather, Navarro reaffirmed a separate rule that when a trustee files a lawsuit in her name, her jurisdictional citizenship is the State to which she belongs—as is true of any natural person. 446 U. S., at 465. This rule coexists with our discussion above that when an artificial entity is sued in its name, it takes the citizenship of each of its members. That said, Americold’s confusion regarding the citizenship of a trust is understandable and widely shared. See Emerald Investors Trust v. Gaunt Parsippany Partners, 492 F. 3d 192, 201–206 (CA3 2007) (discussing various approaches among the Circuits). The confusion can be explained, perhaps, by tradition. Traditionally, a trust was not considered a distinct legal entity, but a “fiduciary relationship” between multiple people. Klein v. Bryer, 227 Md. 473, 476–477, 177 A. 2d 412, 413 (1962); Restatement (Second) of Trusts §2 (1957). Such a relationship was not a thing that could be haled into court; legal proceedings involving a trust were brought by or against the trustees in their own name. Glenn v. Allison, 58 Md. 527, 529 (1882); Deveaux, 5 Cranch, at 91. And when a trustee files a lawsuit or is sued in her own name, her citizenship is all that matters for diversity purposes. Navarro, 446 U. S., at 462–466. For a traditional trust, therefore, there is no need to determine its membership, as would be true if the trust, as an entity, were sued. Many States, however, have applied the “trust” label to a variety of unincorporated entities that have little in common with this traditional template. Maryland, for example, treats a real estate investment trust as a “separate legal entity” that itself can sue or be sued. Md. Corp. & Assns. Code Ann. §§8–102(2), 8–301(2). So long as such an entity is unincorporated, we apply our “oft-repeated rule” that it possesses the citizenship of all its members. Carden, 494 U. S., at 195. But neither this rule nor Navarro limits an entity’s membership to its trustees just because the entity happens to call itself a trust. We therefore decline to apply the same rule to an unincorporated entity sued in its organizational name that applies to a human trustee sued in her personal name. We also decline an amicus’ invitation to apply the same rule to an unincorporated entity that applies to a corporation—namely, to consider it a citizen only of its State of establishment and its principal place of business. See Brief for National Association of Real Estate Investment Trusts 11–21. When we last examined the “doctrinal wall” between corporate and unincorporated entities in 1990, we saw no reason to tear it down. Carden, 494 U. S., at 190. Then as now we reaffirm that it is up to Congress if it wishes to incorporate other entities into 28 U. S. C. §1332(c)’s special jurisdictional rule. * * * For these reasons, the judgment of the Court of Appeals is Affirmed.

578.US.2015_14-770

American nationals may seek money damages from state sponsors of terrorism in the courts of the United States. See 28 U. S. C. §1605A. Prevailing plaintiffs, however, often face practical and legal difficulties enforcing their judgments. To place beyond dispute the availability of certain assets for satisfaction of judgments rendered in terrorism cases against Iran, Congress enacted the Iran Threat Reduction and Syria Human Rights Act of 2012. As relevant here, the Act makes a designated set of assets available to satisfy the judgments underlying a consolidated enforcement proceeding which the statute identifies by the District Court’s docket number. 22 U. S. C. §8772. Section 8772(a)(2) requires a court, before allowing execution against these assets, to determine, inter alia, “whether Iran holds equitable title to, or the beneficial interest in, the assets.” Respondents—more than 1,000 victims of Iran-sponsored acts of terrorism, their estate representatives, and surviving family members—rank within 16 discrete groups, each of which brought suit against Iran. To enforce judgments they obtained by default, the 16 groups moved for turnover of about $1.75 billion in bond assets held in a New York bank account—assets that, respondents alleged, were owned by Bank Markazi, the Central Bank of Iran. The turnover proceeding began in 2008. In 2012, the judgment holders updated their motions to include execution claims under §8772. Bank Markazi maintained that §8772 could not withstand inspection under the separation-of-powers doctrine, contending that Congress had usurped the judicial role by directing a particular result in the pending enforcement proceeding. The District Court disagreed, concluding that §8772 permissibly changed the law applicable in a pending litigation. The Second Circuit affirmed. Held: Section 8772 does not violate the separation of powers. Pp. 12–24. (a) Article III of the Constitution establishes an independent Judiciary with the “province and duty . . . to say what the law is” in particular cases and controversies. Marbury v. Madison, 1 Cranch 137, 177. Necessarily, that endowment of authority blocks Congress from “requir[ing] federal courts to exercise the judicial power in a manner that Article III forbids.” Plaut v. Spendthrift Farm, Inc., 514 U. S. 211 . Although Article III bars Congress from telling a court how to apply pre-existing law to particular circumstances, Robertson v. Seattle Audubon Soc., 503 U. S. 429 –439, Congress may amend a law and make the amended prescription retroactively applicable in pending cases, Landgraf v. USI Film Products, 511 U. S. 244 –268; United States v. Schooner Peggy, 1 Cranch 103, 110. In United States v. Klein, 13 Wall. 128, 146, this Court enigmatically observed that Congress may not “prescribe rules of decision to the Judicial Department . . . in [pending] cases.” More recent decisions have clarified that Klein does not inhibit Congress from “amend[ing] applicable law.” Robertson, 503 U. S., at 441; Plaut, 514 U. S., at 218. Section 8772 does just that: It requires a court to apply a new legal standard in a pending postjudgment enforcement proceeding. No different result obtains because, as Bank Markazi argues, the outcome of applying §8772 to the facts in the proceeding below was a “foregone conclusio[n].” Brief for Petitioner 47. A statute does not impinge on judicial power when it directs courts to apply a new legal standard to undisputed facts. See Pope v. United States, 323 U. S. 1 . Pp. 12–19. (b) Nor is §8772 invalid because, as Bank Markazi further objects, it prescribes a rule for a single, pending case identified by caption and docket number. The amended law upheld in Robertson also applied to cases identified in the statute by caption and docket number. 503 U. S., at 440. Moreover, §8772 is not an instruction governing one case only: It facilitates execution of judgments in 16 suits. While consolidated for administrative purposes at the execution stage, the judgment-execution claims were not independent of the original actions for damages and each retained its separate character. In any event, the Bank’s argument rests on the flawed assumption that legislation must be generally applicable. See Plaut, 514 U. S., at 239, n. 9. This Court and lower courts have upheld as a valid exercise of Congress’ legislative power laws governing one or a very small number of specific subjects. Pp. 19–21. (c) Adding weight to this decision, §8772 is an exercise of congressional authority regarding foreign affairs, a domain in which the controlling role of the political branches is both necessary and proper. Measures taken by the political branches to control the disposition of foreign-state property, including blocking specific foreign-state assets or making them available for attachment, have never been rejected as invasions upon the Article III judicial power. Cf. Dames & Moore v. Regan, 453 U. S. 654 . Notably, before enactment of the Foreign Sovereign Immunities Act, the Executive regularly made case-specific determinations whether sovereign immunity should be recognized, and courts accepted those determinations as binding. See, e.g., Republic of Austria v. Altmann, 541 U. S. 677 –691. This practice, too, was never perceived as an encroachment on the federal courts’ jurisdiction. Dames & Moore, 453 U. S., at 684–685. Pp. 21–23. 758 F. 3d 185, affirmed. Ginsburg, J., delivered the opinion of the Court, in which Kennedy, Breyer, Alito, and Kagan, JJ., joined, and in all but Part II–C of which Thomas, J., joined. Roberts, C. J., filed a dissenting opinion, in which Sotomayor, J., joined.

” in the assets). Finally, prior to the statute’s enactment, Bank Markazi and Clearstream had argued that the assets in question were located in Luxembourg, not New York. Supra, at 10. Leaving the issue for court resolution, Congress, in §8772(a)(1), required the District Court to determine whether the assets were “held in the United States.” 22 Recall, again, that respondents are judgment creditors who prevailed on the merits of their respective cases. Section 8772 serves to facilitate their ability to collect amounts due to them from assets of the judgment debtor. 23 The dissent also analogizes §8772 to a law that makes “conclusive” one party’s flimsy evidence of a boundary line in a pending property dispute, notwithstanding that the governing law ordinarily provides that an official map establishes the boundary. Post, at 1. Section 8772, however, does not restrict the evidence on which a court may rely in making the required findings. A more fitting analogy for depicting §8772’s operation might be: In a pending property dispute, the parties contest whether an ambiguous statute makes a 1990 or 2000 county map the relevant document for establishing boundary lines. To clarify the matter, the legislature enacts a law specifying that the 2000 map supersedes the earlier map. 24 At oral argument, Bank Markazi clarified that its argument extended beyond a single pending case, encompassing as well “a limited category of cases.” Tr. of Oral Arg. 5. See also id., at 57–58. 25 Section 8772’s limitation to one consolidated proceeding operates unfairly, Bank Markazi suggests, because other judgment creditors “would be subject to a completely different rule” if they “sought to execute against the same assets” outside No. 10–CIV–4518. Brief for 26 District courts routinely consolidate multiple related matters for a single decision on common issues. See, e.g., Securities Investor Protection Corp. v. Bernard L. Madoff Inv. Securities LLC, 476 B. R. 715, 717 (SDNY 2012) (deciding several legal questions arising in over 80 cases concerning “the massive Ponzi scheme perpetrated by Bernard L. Madoff”). 27 Questioning this understanding of the proceedings below, The Chief Justice emphasizes that many of the judgment creditors were joined in the Peterson enforcement proceeding by interpleader. See post, at 8, n. 1. That is true, supra, at 8, n. 9, but irrelevant. As explained above, execution proceedings are continuations of merits proceedings, not new lawsuits. Thus, the fact that many creditors joined by interpleader motion did not transform execution claims in 16 separate suits into “a single case.” Post, at 8, n. 1. 28 Laws narrow in scope, including “class of one” legislation, may violate the Equal Protection Clause if arbitrary or inadequately justified. Village of Willowbrook v. Olech, 528 U. S. 562, 564 (2000) (per curiam) (internal quotation marks omitted); New Orleans v. Dukes, 427 U. S. 297 –306 (1976) (per curiam). 29 The Chief Justice correctly notes that the Court in Dames & Moore v. Regan, 453 U. S. 654, 661 (1981) , urged caution before extending its analysis to “other situations” not presented in that case. Post, at 15. Much of the Court’s cause for concern, however, was the risk that the ruling could be construed as license for the broad exercise of unilat-

578.US.2015_14-1457

Petitioner Brandon Betterman pleaded guilty to bail jumping after failing to appear in court on domestic assault charges. He was then jailed for over 14 months awaiting sentence, in large part due to institutional delay. He was eventually sentenced to seven years’ imprisonment, with four of the years suspended. Arguing that the 14-month gap between conviction and sentencing violated his speedy trial right, Betterman appealed, but the Montana Supreme Court affirmed the conviction and sentence, ruling that the Sixth Amendment’s Speedy Trial Clause does not apply to postconviction, presentencing delay. Held: The Sixth Amendment’s speedy trial guarantee does not apply once a defendant has been found guilty at trial or has pleaded guilty to criminal charges. Pp. 3–11. (a) Criminal proceedings generally unfold in three discrete phases. First, the State investigates to determine whether to arrest and charge a suspect. Once charged, the suspect is presumed innocent until conviction upon trial or guilty plea. After conviction, the court imposes sentence. There are checks against delay geared to each particular phase. P. 3. (b) Statutes of limitations provide the primary protection against delay in the first stage, when the suspect remains at liberty, with the Due Process Clause safeguarding against fundamentally unfair prosecutorial conduct. United States v. Lovasco, 431 U. S. 783 . P. 3. (c) The Speedy Trial Clause right attaches when the second phase begins, that is, upon a defendant’s arrest or formal accusation. United States v. Marion, 404 U. S. 307 –321. The right detaches upon conviction, when this second stage ends. Before conviction, the accused is shielded by the presumption of innocence, Reed v. Ross, 468 U. S. 1 , which the Speedy Trial Clause implements by minimizing the likelihood of lengthy incarceration before trial, lessening the anxiety and concern associated with a public accusation, and limiting the effects of long delay on the accused’s ability to mount a defense, Marion, 404 U. S., at 320. The Speedy Trial Clause thus loses force upon conviction. This reading comports with the historical understanding of the speedy trial right. It “has its roots at the very foundation of our English law heritage,” Klopfer v. North Carolina, 386 U. S. 213 , and it was the contemporaneous understanding of the Sixth Amendment’s language that “accused” described a status preceding “convicted” and “trial” meant a discrete episode after which judgment (i.e., sentencing) would follow. The Court’s precedent aligns with the text and history of the Speedy Trial Clause. See Barker v. Wingo, 407 U. S. 514 –533. Just as the right to speedy trial does not arise prearrest, Marion, 404 U. S., at 320–322, adverse consequences of postconviction delay are outside the purview of the Speedy Trial Clause. The sole remedy for a violation of the speedy trial right—dismissal of the charges—fits the preconviction focus of the Clause, for it would be an unjustified windfall to remedy sentencing delay by vacating validly obtained convictions. This reading also finds support in the federal Speedy Trial Act of 1974 and numerous state analogs, which impose time limits for charging and trial but say nothing about sentencing. The prevalence of guilty pleas and the resulting scarcity of trials in today’s justice system do not bear on the presumption-of-innocence protection at the heart of the Speedy Trial Clause. Moreover, a central feature of contemporary sentencing—the preparation and review of a presentence investigation report—requires some amount of wholly reasonable presentencing delay. Pp. 3–9. (d) Although the Constitution’s presumption-of-innocence-protective speedy trial right is not engaged in the sentencing phase, statutes and rules offer defendants recourse. Federal Rule of Criminal Procedure 32(b)(1), for example, directs courts to “impose sentence without unnecessary delay.” Further, as at the prearrest stage, due process serves as a backstop against exorbitant delay. Because Betterman advanced no due process claim here, however, the Court expresses no opinion on how he might fare under that more pliable standard. Pp. 9–11. 378 Mont. 182, 342 P. 3d 971, affirmed. Ginsburg, J., delivered the opinion for a unanimous Court. Thomas, J., filed a concurring opinion, in which Alito, J., joined. Sotomayor, J., filed a concurring opinion.

The Sixth Amendment to the U. S. Constitution provides that “[i]n all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury . . . .” Does the Sixth Amendment’s speedy trial guarantee apply to the sentencing phase of a criminal prosecution? That is the sole question this case presents. We hold that the guarantee protects the accused from arrest or indictment through trial, but does not apply once a defendant has been found guilty at trial or has pleaded guilty to criminal charges. For inordinate delay in sentencing, although the Speedy Trial Clause does not govern, a defendant may have other recourse, including, in appropriate circumstances, tailored relief under the Due Process Clauses of the Fifth and Fourteenth Amendments. Petitioner Brandon Betterman, however, advanced in this Court only a Sixth Amendment speedy trial claim. He did not preserve a due process challenge. See Tr. of Oral Arg. 19. We, therefore, confine this opinion to his Sixth Amendment challenge. I Ordered to appear in court on domestic assault charges, Brandon Betterman failed to show up and was therefore charged with bail jumping. 378 Mont. 182, 184, 342 P. 3d 971, 973 (2015). After pleading guilty to the bail-jumping charge, he was jailed for over 14 months awaiting sentence on that conviction. Id., at 184–185, 342 P. 3d, at 973–974. The holdup, in large part, was due to institutional delay: the presentence report took nearly five months to complete; the trial court took several months to deny two presentence motions (one seeking dismissal of the charge on the ground of delay); and the court was slow in setting a sentencing hearing. Id., at 185, 195, 342 P. 3d, at 973–974, 980. Betterman was eventually sentenced to seven years’ imprisonment, with four of those years suspended. Id., at 185, 342 P. 3d, at 974. Arguing that the 14-month gap between conviction and sentencing violated his speedy trial right, Betterman appealed. The Montana Supreme Court affirmed his conviction and sentence, ruling that the Sixth Amendment’s Speedy Trial Clause does not apply to postconviction, presentencing delay. Id., at 188–192, 342 P. 3d, at 975–978. We granted certiorari, 577 U. S. ___ (2015), to resolve a split among courts over whether the Speedy Trial Clause applies to such delay.[1] Holding that the Clause does not apply to delayed sentencing, we affirm the Montana Supreme Court’s judgment. II Criminal proceedings generally unfold in three discrete phases. First, the State investigates to determine whether to arrest and charge a suspect. Once charged, the suspect stands accused but is presumed innocent until conviction upon trial or guilty plea. After conviction, the court imposes sentence. There are checks against delay throughout this progression, each geared to its particular phase. In the first stage—before arrest or indictment, when the suspect remains at liberty—statutes of limitations provide the primary protection against delay, with the Due Process Clause as a safeguard against fundamentally unfair prosecutorial conduct. United States v. Lovasco, 431 U. S. 783, 789 (1977) ; see id., at 795, n. 17 (Due ProcessClause may be violated, for instance, by prosecutorial delay that is “tactical” or “reckless” (internal quotation marks omitted)). The Sixth Amendment’s Speedy Trial Clause homes in on the second period: from arrest or indictment through conviction. The constitutional right, our precedent holds, does not attach until this phase begins, that is, when a defendant is arrested or formally accused. United States v. Marion, 404 U. S. 307 –321 (1971). Today we hold that the right detaches upon conviction, when this second stage ends.[2] Prior to conviction, the accused is shielded by the presumption of innocence, the “bedrock[,] axiomatic and elementary principle whose enforcement lies at the foundation of the administration of our criminal law.” Reed v. Ross, 468 U. S. 1, 4 (1984) (internal quotation marks omitted). The Speedy Trial Clause implements that presumption by “prevent[ing] undue and oppressive incarceration prior to trial, . . . minimiz[ing] anxiety and concern accompanying public accusation[,] and . . . limit[ing] the possibilities that long delay will impair the ability of an accused to defend himself.” Marion, 404 U. S., at 320 (internal quotation marks omitted). See also Barker v. Wingo, 407 U. S. 514 –533 (1972). As a measure protecting the presumptively innocent, the speedy trial right—like other similarly aimed measures—loses force upon conviction. Compare In re Winship, 397 U. S. 358, 364 (1970) (requiring “proof beyond a reasonable doubt of every fact necessary to constitute the crime”), with United States v. O’Brien, 560 U. S. 218, 224 (2010) (“Sentencing factors . . . can be proved . . . by a preponderance of the evidence.”). Compare also 18 U. S. C. §3142(b) (bail presumptively available for accused awaiting trial) with §3143(a) (bail presumptively unavailable for those convicted awaiting sentence). Our reading comports with the historical understanding. The speedy trial right, we have observed, “has its roots at the very foundation of our English law heritage. Its first articulation in modern jurisprudence appears to have been made in Magna Carta (1215) . . . .” Klopfer v. North Carolina, 386 U. S. 213, 223 (1967) . Regarding the Framers’ comprehension of the right as it existed at the founding, we have cited Sir Edward Coke’s Institutes of the Laws of England. See id., at 223–225, and nn. 8, 12–14, 18. Coke wrote that “the innocent shall not be worn and wasted by long imprisonment, but . . . speedily come to his tria[l].” 1 E. Coke, Second Part of the Institutes of the Laws of England 315 (1797) (emphasis added). Reflecting the concern that a presumptively innocent person should not languish under an unresolved charge, the Speedy Trial Clause guarantees “the accused” “the right to a speedy . . . trial.” U. S. Const., Amdt. 6 (emphasis added). At the founding, “accused” described a status preceding “convicted.” See, e.g., 4 W. Blackstone, Commentaries on the Laws of England 322 (1769) (commenting on process in which “persons accused of felony . . . were tried . . . and convicted” (emphasis added)). And “trial” meant a discrete episode after which judgment (i.e., sentencing) would follow. See, e.g., id., at 368 (“We are now to consider the next stage of criminal prosecution, after trial and conviction are past . . . : which is that of judgment.”).[3] This understanding of the Sixth Amendment language—“accused” as distinct from “convicted,” and “trial” as separate from “sentencing”—endures today. See, e.g., Black’s Law Dictionary 26 (10th ed. 2014) (defining “accused” as “a person who has been arrested and brought before a magistrate or who has been formally charged” (emphasis added)); Fed. Rule Crim. Proc. 32 (governing “Sentencing and Judgment,” the rule appears in the chapter on “Post-Conviction Procedures,” which follows immediately after the separate chapter headed “Trial”).[4] This Court’s precedent aligns with the text and history of the Speedy Trial Clause. Detaining the accused pretrial, we have said, disadvantages him, and the imposition is “especially unfortunate” as to those “ultimately found to be innocent.” Barker, 407 U. S., at 532–533. And in Marion, 404 U. S., at 320, addressing “the major evils protected against by the speedy trial guarantee,” we observed: “Arrest is a public act that may seriously interfere with the defendant’s liberty, whether he is free on bail or not, and that may disrupt his employment, drain his financial resources, curtail his associations, subject him to public obloquy, and create anxiety in him, his family and his friends.” We acknowledged in Marion that even pre-arrest—a stage at which the right to a speedy trial does not arise—the passage of time “may impair memories, cause evidence to be lost, deprive the defendant of witnesses, and otherwise interfere with his ability to defend himself.” Id., at 321. Nevertheless, we determined, “this possibility of prejudice at trial is not itself sufficient reason to wrench the Sixth Amendment from its proper [arrest or charge triggered] context.” Id., at 321–322. Adverse consequences of postconviction delay, though subject to other checks, see infra, at 10–11, are similarly outside the purview of the Speedy Trial Clause.[5] The sole remedy for a violation of the speedy trial right—dismissal of the charges, see Strunk v. United States, 412 U. S. 434, 440 (1973) ; Barker, 407 U. S., at 522—fits the preconviction focus of the Clause. It would be an unjustified windfall, in most cases, to remedy sentencing delay by vacating validly obtained convictions. Betterman concedes that a dismissal remedy ordinarily would not be in order once a defendant has been convicted. See Tr. of Oral Arg. 5–6; cf. Bozza v. United States, 330 U. S. 160, 166 (1947) (“[A]n error in passing the sentence” does not permit a convicted defendant “to escape punishment altogether.”).[6] The manner in which legislatures have implemented the speedy trial guarantee matches our reading of the Clause. Congress passed the Speedy Trial Act of 1974, 18 U. S. C. §3161 et seq., “to give effect to the sixth amendment right.” United States v. MacDonald, 456 U. S. 1 , n. 7 (1982) (quoting S. Rep. No. 93–1021, p. 1 (1974)). “The more stringent provisions of the Speedy Trial Act have mooted much litigation about the requirements of the Speedy Trial Clause . . . .” United States v. Loud Hawk, 474 U. S. 302 , n. 1 (1986) (citation omitted). With certain exceptions, the Act directs—on pain of dismissal of the charges, §3162(a)—that no more than 30 days pass between arrest and indictment, §3161(b), and that no more than 70 days pass between indictment and trial, §3161(c)(1). The Act says nothing, however, about the period between conviction and sentencing, suggesting that Congress did not regard that period as falling within the Sixth Amendment’s compass. Numerous state analogs similarly impose precise time limits for charging and trial; they, too, say nothing about sentencing.[7] Betterman asks us to take account of the prevalence of guilty pleas and the resulting scarcity of trials in today’s justice system. See Lafler v. Cooper, 566 U. S. ___, ___ (2012) (slip op., at 11) (“[C]riminal justice today is for the most part a system of pleas, not a system of trials.”). The sentencing hearing has largely replaced the trial as the forum for dispute resolution, Betterman urges. Therefore, he maintains, the concerns supporting the right to a speedy trial now recommend a speedy sentencing hear-ing. The modern reality, however, does not bear on the presumption-of-innocence protection at the heart of the Speedy Trial Clause. And factual disputes, if any there be, at sentencing, do not go to the question of guilt;they are geared, instead, to ascertaining the proper sentence within boundaries set by statutory minimums and maximums. Moreover, a central feature of contemporary sentencing in both federal and state courts is preparation by the probation office, and review by the parties and the court, of a presentence investigation report. See 18 U. S. C. §3552; Fed. Rule Crim. Proc. 32(c)–(g); 6 W. LaFave, J. Israel, N. King, & O. Kerr, Criminal Procedure §26.5(b), pp. 1048–1049 (4th ed. 2015) (noting reliance on presentence reports in federal and state courts). This aspect of the system requires some amount of wholly reasonable presentencing delay.[8] Indeed, many—if not most—disputes are resolved, not at the hearing itself, but rather through the presentence-report process. See N. Demleitner, D. Berman, M. Miller, & R. Wright, Sentencing Law and Policy 443 (3d ed. 2013) (“Criminal justice is far more commonly negotiated than adjudicated; defendants and their attorneys often need to be more concerned about the charging and plea bargaining practices of prosecutors and the presentence investigations of probation offices than . . . about the sentencing procedures of judges or juries.”); cf. Bierschbach & Bibas, Notice-and-Comment Sentencing, 97 Minn. L. Rev. 1, 15 (2012) (“[T]oday’s sentencing hearings . . . rubber-stamp plea-bargained sentences.”). As we have explained, at the third phase of thecriminal-justice process, i.e., between conviction and sentencing, the Constitution’s presumption-of-innocence-protective speedy trial right is not engaged.[9] That does not mean, however, that defendants lack any protection against undue delay at this stage. The primary safeguard comes from statutes and rules. The federal rule on point directs the court to “impose sentence without unnecessary delay.” Fed. Rule Crim. Proc. 32(b)(1). Many States have provisions to the same effect,[10] and some States prescribe numerical time limits.[11] Further, as at the prearrest stage, due process serves as a backstop against exorbitant delay. See supra, at 3. After conviction, a defendant’s due process right to liberty, while diminished, is still present. He retains an interest in a sentencing proceeding that is fundamentally fair. But because Betterman advanced no due process claim here, see supra, at 1, we express no opinion on how he might fare under that more pliable standard. See, e.g., United States v. $8,850, 461 U. S. 555 –565 (1983).[12] * * * The course of a criminal prosecution is composed of discrete segments. During the segment between accusation and conviction, the Sixth Amendment’s Speedy Trial Clause protects the presumptively innocent from long enduring unresolved criminal charges. The Sixth Amendment speedy trial right, however, does not extend beyond conviction, which terminates the presumption of innocence. The judgment of the Supreme Court of Montana is therefore Affirmed.Notes 1 Compare Burkett v. Cunningham, 826 F. 2d 1208, 1220 (CA3 1987); Juarez-Casares v. United States, 496 F. 2d 190, 192 (CA5 1974); Ex parte Apicella, 809 So. 2d 865, 869 (Ala. 2001); Gonzales v. State, 582 P. 2d 630, 632 (Alaska 1978); Jolly v. State, 358 Ark. 180, 191, 189 S. W. 3d 40, 45 (2004); Trotter v. State, 554 So. 2d 313, 316 (Miss. 1989), superseded by statute on other grounds, Miss. Code Ann. §99–35–101 (2008); Commonwealth v. Glass, 526 Pa. 329, 334, 586 A. 2d 369, 371 (1991); State v. Leyva, 906 P. 2d 910, 912 (Utah 1995); and State v. Dean, 148 Vt. 510, 513, 536 A. 2d 909, 912 (1987) (Speedy Trial Clause applies to sentencing delay), with United States v. Ray, 578 F. 3d 184, 198–199 (CA2 2009); State v. Drake, 259 N. W. 2d 862, 866 (Iowa 1977), abrogated on other grounds by State v. Kaster, 469 N. W. 2d 671, 673 (Iowa 1991); State v. Pressley, 290 Kan. 24, 29, 223 P. 3d 299, 302 (2010); State v. Johnson, 363 So. 2d 458, 460 (La. 1978); 378 Mont. 182, 192, 342 P. 3d 971, 978 (2015) (case below); and Ball v. Whyte, 170 W. Va. 417, 418, 294 S. E. 2d 270, 271 (1982) (Speedy Trial Clause does not apply to sentencing delay). 2 We reserve the question whether the Speedy Trial Clause applies to bifurcated proceedings in which, at the sentencing stage, facts that could increase the prescribed sentencing range are determined (e.g., capital cases in which eligibility for the death penalty hinges on aggravating factor findings). Nor do we decide whether the right reattaches upon renewed prosecution following a defendant’s successful appeal, when he again enjoys the presumption of innocence. 3 As Betterman points out, at the founding, sentence was often imposed promptly after rendition of a verdict. Brief for Petitioner 24–26. But that was not invariably the case. For the court’s “own convenience, or on cause shown, [sentence could be] postpone[d] . . . to a future day or term.” 1 J. Bishop, Criminal Procedure §1291, p. 767 (3d ed. 1880) (footnote omitted). See also 1 J. Chitty, A Practical Treatise on the Criminal Law 481 (1819) (“The sentence . . . is usually given immediately after the conviction, but the court may adjourn to another day and then give judgment.”). 4 We do not mean to convey that provisions of the Sixth Amendment protecting interests other than the presumption of innocence are inapplicable to sentencing. In this regard, we have held that the right to defense counsel extends to some postconviction proceedings. See Mempa v. Rhay, 389 U. S. 128 –137 (1967). 5 Smith v. Hooey, 393 U. S. 374 (1969) , on which Betterman relies, is not to the contrary. There we concluded that a defendant, though already convicted and imprisoned on one charge, nevertheless has a right to be speedily brought to trial on an unrelated charge. Id., at 378. “[T]here is reason to believe,” we explained in Smith, “that an outstanding untried charge (of which even a convict may, of course, be innocent) can have fully as depressive an effect upon a prisoner as upon a person who is at large.” Id., at 379. Smith is thus consistent with comprehension of the Speedy Trial Clause as protective of the presumptively innocent. 6 Betterman suggests that an appropriate remedy for the delay in his case would be reduction of his sentence by 14 months—the time between his conviction and sentencing. See Tr. of Oral Arg. 6. We have not read the Speedy Trial Clause, however, to call for a flexible or tailored remedy. Instead, we have held that violation of the right demands termination of the prosecution. 7 See, e.g., Alaska Rule Crim. Proc. 45 (2016); Ark. Rules Crim. Proc. 28.1 to 28.3 (2015); Cal. Penal Code Ann. §1382 (West 2011); Colo. Rev. Stat. §18–1–405 (2015); Conn. Rules Crim. Proc. 43–39 to 43–42 (2016); Fla. Rule Crim. Proc. 3.191 (2016); Haw. Rule Crim. Proc. 48 (2016); Ill. Comp. Stat., ch. 725, §5/103–5 (West 2014); Ind. Rule Crim. Proc. 4 (2016); Iowa Rule Crim. Proc. 2.33 (2016); Kan. Stat. Ann. §22–3402 (2014 Cum. Supp.); La. Code Crim. Proc. Ann., Art. 701 (West Cum. Supp. 2016); Mass. Rule Crim. Proc. 36 (2016); Neb. Rev. Stat. §§29–1207, 29–1208 (2008); Nev. Rev. Stat. §178.556 (2013); N. Y. Crim. Proc. Law Ann. §30.30 (West Cum. Supp. 2016); Ohio Rev. Code Ann. §§2945.71 to 2945.73 (Lexis 2014); Ore. Rev. Stat. §§135.745, 135.746, 135.748, 135.750, 135.752 (2015); Pa. Rule Crim. Proc. 600 (2016); S. D. Codified Laws §23A–44–5.1 (Cum. Supp. 2015); Va. Code Ann. §19.2–243 (2015); Wash. Rule Crim. Proc. 3.3 (2016); Wis. Stat. §971.10 (2011–2012); Wyo. Rule Crim. Proc. 48 (2015). 8 “In federal prosecutions,” the Solicitor General informs us, “the median time between conviction and sentencing in 2014 was 99 days.” Brief for United States as Amicus Curiae 31, n. 5. A good part of this time no doubt was taken up by the drafting and review of a presentence report. See Fed. Rule Crim. Proc. 32(c)–(g) (detailing presentence-report process). 9 It is true that during this period the defendant is often incarcerated. See, e.g., §3143(a) (bail presumptively unavailable for convicted awaiting sentence). Because postconviction incarceration is considered punishment for the offense, however, a defendant will ordinarily earn time-served credit for any period of presentencing detention. See §3585(b); A. Campbell, Law of Sentencing §9:28, pp. 444–445, and n. 4 (3d ed. 2004) (“[State c]rediting statutes routinely provide that any period of time during which a person was incarcerated in relation to a given offense be counted toward satisfaction of any resulting sentence.”). That such detention may occur in a local jail rather than a prison is of no constitutional moment, for a convicted defendant has no right to serve his sentence in the penal institution he prefers. See Meachum v. Fano, 427 U. S. 215 –225 (1976). 10 See, e.g., Alaska Rule Crim. Proc. 32(a) (2016); Colo. Rule Crim. Proc. 32(b)(1) (2015); Del. Super. Ct. Crim. Rule 32(a)(1) (2003); Fla. Rule Crim. Proc. 3.720 (2016); Haw. Rule Penal Proc. 32(a) (2016); Kan. Stat. Ann. §22–3424(c) (2014 Cum. Supp.); Ky. Rule Crim. Proc. 11.02(1) (2016); La. Code Crim. Proc. Ann., Art. 874 (West 2016); Me. Rule Crim. Proc. 32(a)(1) (2015); Mass. Rule Crim. Proc. 28(b) (2016); Mich. Ct. Rule 6.425(E)(1) (2011); Mo. Sup. Ct. Rule 29.07(b)(1) (2011); Mont. Code Ann. §46–18–115 (2015); Nev. Rev. Stat. §176.015(1) (2013); N. H. Rule Crim. Proc. 29(a)(1) (2016); N. J. Ct. Rule 3:21–4(a) (2016); N. Y. Crim. Proc. Law Ann. §380.30(1) (West Cum. Supp. 2016); N. D. Rule Crim. Proc. 32(a)(1) (2011); Ohio Rule Crim. Proc. 32(A) (2013); R. I. Super. Ct. Rule 32(a)(1) (2015); S. D. Codified Laws §23A–27–1 (Cum. Supp. 2015); Vt. Rule Crim. Proc. 32(a)(1) (2010); Va. Sup. Ct. Rule 3A:17.1(b) (2012); W. Va. Rule Crim. Proc. 32(a) (2006); Wyo. Rule Crim. Proc. 32(c)(1) (2015). 11 See, e.g., Ariz. Rule Crim. Proc. 26.3(a)(1) (2011); Ark. Rule Crim. Proc. 33.2 (2015); Cal. Penal Code Ann. §1191 (West 2015); Ind. Rule Crim. Proc. 11 (2016); N. M. Rule Crim. Proc. 5–701(B) (2016); Ore. Rev. Stat. §137.020(3) (2015); Pa. Rule Crim. Proc. 704(A)(1) (2016); Tenn. Code Ann. §40–35–209(a) (2014); Utah Rule Crim. Proc. 22(a) (2015); Wash. Rev. Code §9.94A.500(1) (2016 Cum. Supp.). These sentencing provisions are separate from state analogues to the Speedy Trial Act. See supra, at 8, and n. 7. 12 Relevant considerations may include the length of and reasons for delay, the defendant’s diligence in requesting expeditious sentencing, and prejudice.

577.US.2015_14-844

The Prison Litigation Reform Act of 1995 provides that prisoners qualified to proceed in forma pauperis (IFP) must nonetheless pay an initial partial filing fee, set as “20 percent of the greater of” the average monthly deposits in the prisoner’s account or the average monthly balance of the account over the preceding six months. 28 U. S. C. §1915(b)(1). They must then pay the remainder of the fee in monthly installments of “20 percent of the preceding month’s income credited to the prisoner’s account.” §1915(b)(2). The initial partial fee is assessed on a per-case basis, i.e., each time the prisoner files a lawsuit. The initial payment may not be exacted if the prisoner has no means to pay it, §1915(b)(4), and no monthly installments are required unless the prisoner has more than $10 in his account, §1915(b)(2). In contest here is the calculation of subsequent monthly installment payments when more than one fee is owed. Petitioner Antoine Bruce, a federal inmate and a frequent litigant, argued that the monthly filing-fee payments do not become due until filing-fee obligations previously incurred in other cases are satisfied. The D. C. Circuit disagreed, holding that Bruce’s monthly payments were due simultaneously with monthly payments in the earlier cases. Held: Section 1915(b)(2) calls for simultaneous, not sequential, recoupment of multiple monthly installment payments. Pp. 5–8. (a) Bruce and the Government present competing interpretations of the IFP statute, which does not explicitly address how multiple filing fees should be paid. In urging a per-prisoner approach under which he would pay 20 percent of his monthly income regardless of the number of cases he has filed, Bruce relies principally on the contrast between the singular “clerk” and the plural “fees” as those nouns appear in §1915(b)(2), which requires payments to be forwarded “to the clerk of the court . . . until the filing fees are paid.” Even when more than one filing fee is owed, Bruce contends, §1915(b)(2) instructs that only one clerk will receive payment each month. In contrast, the Government urges a per-case approach. Emphasizing that §1915 as a whole has a single-case focus, providing instructions for each case, the Government contends that it would be anomalous to treat paragraph (b)(1)’s initial partial payment, admittedly directed at a single case, differently than paragraph (b)(2)’s subsequent monthly payments. Pp. 5–7. (b) Section 1915’s text and context support the per-case approach. Just as §1915(b)(1) calls for assessment of “an initial partial filing fee” each time a prisoner “brings a civil action or files an appeal” (emphasis added), so its allied provision, §1915(b)(2), calls for monthly 20 percent payments simultaneously for each action pursued. Section 1915(b)(3), which imposes a ceiling on fees permitted “for the commencement of a civil action or an appeal” (emphasis added), and §1915(b)(4), which protects the right to bring “a civil action or appea[l] a . . . judgment” (emphasis added), confirm that subsection (b) as a whole is written from the perspective of a single case. Pp. 7–8. 761 F. 3d 1, affirmed. Ginsburg, J., delivered the opinion for a unanimous Court.

This case concerns the payment of filing fees for civil actions commenced by prisoners in federal courts. Until 1996, indigent prisoners, like other indigent persons, could file a civil action without paying any filing fee. See 28 U. S. C. §1915(a)(1). In the Prison Litigation Reform Act of 1995 (PLRA), 110Stat. 1321–66, Congress placed several limitations on prisoner litigation in federal courts. Among those limitations, Congress required prisoners qualified to proceed in forma pauperis nevertheless to pay an initial partial filing fee. That fee is statutorily set as “20 percent of the greater of” the average monthly deposits in the prisoner’s account or the average monthly balance of the account over the preceding six months. §1915(b)(1). Thereafter, to complete payment of the filing fee, prisoners must pay, in monthly installments, “20 percent of the preceding month’s income credited to the prisoner’s account.” §1915(b)(2). The initial partial filing fee may not be exacted if the prisoner has no means to pay it, §1915(b)(4), and no monthly installments are required unless the prisoner has more than $10 in his account, §1915(b)(2). It is undisputed that the initial partial filing fee is to be assessed on a per-case basis, i.e., each time the prisoner files a lawsuit. In contest here is the calculation of subsequent monthly installment payments. Petitioner Antoine Bruce urges a per-prisoner approach under which he would pay 20 percent of his monthly income regardless of the number of cases he has filed. The Government urges, and the court below followed, a per-case approach under which a prisoner would pay 20 percent of his monthly income for each case he has filed. Courts of Appeals have divided on which of these two approaches §1915(b)(2) orders.[1] To resolve the conflict, we granted certiorari. 576 U. S. ___ (2015). We hold that monthly installment payments, like the initial partial payment, are to be assessed on a per-case basis. Nothing in §1915’s current design supports treating a prisoner’s second or third action unlike his first lawsuit. I A In 1892, Congress enacted the in forma pauperis (IFP) statute, now codified at 28 U. S. C. §1915, “to ensure that indigent litigants have meaningful access to the federal courts.” Neitzke v. Williams, 490 U. S. 319, 324 (1989) . Reacting to “a sharp rise in prisoner litigation,” Woodford v. Ngo, 548 U. S. 81, 84 (2006) , Congress in 1996 enacted the PLRA, which installed a variety of measures “designed to filter out the bad claims [filed by prisoners] and facilitate consideration of the good,” Coleman v. Tollefson, 575 U. S. ___, ___ (2015) (slip op., at 3) (quoting Jones v. Bock, 549 U. S. 199, 204 (2007) ; alteration in original). Among those measures, Congress required prisoners to pay filing fees for the suits or appeals they launch. The provisions on fee payment, set forth in §1915(b), read: “(1) . . . [I]f a prisoner brings a civil action or files an appeal in forma pauperis, the prisoner shall be required to pay the full amount of a filing fee. The court shall assess and, when funds exist, collect, as a partial payment of any court fees required by law, an initial partial filing fee of 20 percent of the greater of— “(A) the average monthly deposits to the prisoner’s account; or “(B) the average monthly balance in the prisoner’s account for the 6-month period immediately preceding the filing of the complaint or notice of appeal. “(2) After payment of the initial partial filing fee, the prisoner shall be required to make monthly payments of 20 percent of the preceding month’s income credited to the prisoner’s account. The agency having custody of the prisoner shall forward payments from the prisoner’s account to the clerk of the court each time the amount in the account exceeds $10 until the filing fees are paid.” The monthly installment scheme described in §1915(b)(2) also applies to costs awarded against prisoners when they are judgment losers. §1915(f)(2)(B). To further contain prisoner litigation, the PLRA introduced a three-strikes provision: Prisoners whose suits or appeals are dismissed three or more times as frivolous, malicious, or failing to state a claim on which relief may be granted are barred from proceeding IFP “unless the prisoner is under imminent danger of serious physical injury.” §1915(g). In other words, for most three strikers, all future filing fees become payable in full upfront. Congress included in its 1996 overhaul of §1915 a safety-valve provision to ensure that the fee requirementswould not bar access to the courts: “In no event shall a prisoner be prohibited from bringing a civil action or appealing a civil or criminal judgment for the reason that the prisoner has no assets and no means by which to pay the initial partial filing fee.” §1915(b)(4). B Petitioner Antoine Bruce, a federal inmate serving a 15-year sentence, is a frequent litigant.[2] In the instant case, Bruce challenges his placement in a special management unit at the Federal Correctional Institution in Talladega, Alabama. Pinson v. Samuels, 761 F. 3d 1, 3–4 (CADC 2014).[3] Bruce had previously incurred filing-fee obligations in other cases and maintained that the monthly filing-fee payments for this case would not become due until those prior obligations were satisfied. Id., at 4, 7. The Court of Appeals for the District of Columbia Circuit, whose decision is before us for review, rejected Bruce’s argument. Id., at 8–10. Bruce must make monthly filing-fee payments in this case, the court held, simultaneously with such payments in earlier commenced cases. Id., at 8. We agree with the appeals court that §1915(b)(2) calls for simultaneous, not sequential, recoupment of multiple filing fees. II The IFP statute does not explicitly address whether multiple filing fees (after the initial partial payment) should be paid simultaneously or sequentially. Bruce and the Government present competing interpretations. A In support of the per-prisoner approach, Bruce relies principally on what he sees as a significant contrast between the singular “clerk” and the plural “fees” as those nouns appear in 28 U. S. C. §1915(b)(2). That provision requires payments to be forwarded “to the clerk of the court . . . until the filing fees are paid.” Ibid. (empha-sis added). Even when more than one filing fee isowed, Bruce contends, the statute instructs that only one clerk will receive payment each month; in other words,fee payments are to be made sequentially rather than simultaneously. The initial partial payment, which is charged on a per-case basis, plus the three-strikes provision, Bruce urges, together suffice to satisfy the PLRA’s purpose, which is to “force prisoners to think twice about the case and not just file reflexively,” 141 Cong. Rec. 14572 (1995) (remarks of Sen. Kyl). The additional economic disincentive that the per-case approach would occasion, Bruce asserts, could excessively encumber access to federal courts. Furthermore, Bruce points out, the per-case approach breaks down when a prisoner incurs more than five obligations. Nothing will be left in the account to pay the sixth fee, Bruce observes. Necessarily, therefore, its payment will be entirely deferred. Why treat the second obligation unlike the sixth, Bruce asks. Isn’t the statute sensibly read to render all monthly payments sequential? Bruce notes in this regard that, under the per-case approach, his ability to use his account to purchase amenities will be progressively curtailed; indeed, the account might be reduced to zero upon his filing or joining a fifth case. Finally, Bruce argues, administrative difficulties counsel against the per-case approach. Costs could dwarf the monetary yield if prisons, under a per-case regime, were obliged to send as many as five checks to five different courts each month. And the problems faced by state-prison officials—who sometimes must choose which of several claims on a prisoner’s income (e.g., child-support, medical copayments) should take precedence—would be exacerbated under a system demanding simultaneous payment of multiple litigation charges. B The Government emphasizes that §1915 as a whole has a single-case focus, providing instructions for each case. It would be anomalous, the Government urges, to treat paragraph (b)(1)’s initial partial payment, which Bruce concedes is directed at a single case, differently than paragraph (b)(2)’s subsequent monthly payments. The two paragraphs, the Government observes, are linked by paragraph (b)(2)’s opening clause: “After payment of the initial partial filing fee.” The per-case approach, the Government adds, better comports with the purpose of the PLRA to deter frivolous suits. See Newlin v. Helman, 123 F. 3d 429, 436 (CA7 1997) (Easterbrook, J.) (“Otherwise a prisoner could file multiple suits for the price of one, postponing payment of the fees for later-filed suits until after the end of imprisonment (and likely avoiding them altogether [because fees are often uncollectable on a prisoner’s release]).”), overruled in part on other grounds by Lee v. Clinton, 209 F. 3d 1025 (CA7 2000), and Walker v. O’Brien, 216 F. 3d 626 (CA7 2000). The Government further observes that the generally small size of the initial partial fee—here, $0.64, App. to Pet. for Cert. 21a—provides scant disincentive, on its own, for multiple filings. Responding to Bruce’s observation that, for a prisoner with more than five charges, even the per-case approach resorts to sequential payments, the Government agrees, but tells us that this scenario arises infrequently. “[M]ost prisoners,” the Government states, “would accrue three strikes (and therefore be required to pay the full filing fees upfront) by the time they incurred the obligation for their sixth case.” Brief for Respondents 29. Finally, answering Bruce’s concern that the per-case approach could leave a prisoner without money for amenities, the Government points out that prisons “are constitutionally bound to provide inmates with adequate food, clothing, shelter, and medical care,” id., at 48 (citing Farmer v. Brennan, 511 U. S. 825, 832 (1994) ), and must furnish “ ‘paper and pen to draft legal documents’ and ‘stamps to mail them,’ ” Brief for Respondents 48 (quoting Bounds v. Smith, 430 U. S. 817, 824, 825 (1977) ). More-over, the Government notes, the Federal Bureau of Prisons (BOP) “goes beyond those requirements,” providing inmates “articles necessary for maintaining personal hygiene,” and free postage “not only for legal mailings but also to enable the inmate to maintain community ties.” Brief for Respondents 48, n. 21 (internal quotation marks omitted). III The Circuits following the per-case approach, we conclude, better comprehend the statute. Just as §1915(b)(1) calls for assessment of “an initial partial filing fee” each time a prisoner “brings a civil action or files an appeal” (emphasis added), so its allied provision, §1915(b)(2), triggered immediately after, calls for “monthly payments of 20 percent of the preceding month’s income” simultaneously for each action pursued. The other two paragraphs of §1915(b) confirm that the subsection as a whole is written from the perspective of a single case. See §1915(b)(3) (imposing a ceiling on fees permitted “for the commencement of a civil action or an appeal” (emphasis added)); §1915(b)(4) (protecting the right to “brin[g] a civil action or appea[l] a civil or criminal judgment” (emphasis added)). There is scant indication that the statute’s perspective shifts partway through paragraph (2).[4] Bruce’s extratextual points do not warrant a departure from the interpretation suggested by the text and context. The per-case approach more vigorously serves the statutory objective of containing prisoner litigation, while the safety-valve provision, see supra, at 4, ensures against denial of access to federal courts. Bruce’s administrability concerns carry little weight given reports from several States that the per-case approach is unproblematic. See Brief for State of Michigan et al. as Amici Curiae 18–20. * * * For the reasons stated, the judgment of the Court of Appeals for the District of Columbia Circuit is Affirmed.Notes 1 Compare Atchison v. Collins, 288 F. 3d 177, 181 (CA5 2002) (per curiam); Newlin v. Helman, 123 F. 3d 429, 436 (CA7 1997), overruled in part on other grounds by Lee v. Clinton, 209 F. 3d 1025 (CA7 2000), and Walker v. O’Brien, 216 F. 3d 626 (CA7 2000); Lefkowitz v. Citi-Equity Group, Inc., 146 F. 3d 609, 612 (CA8 1998); Christensen v. Big Horn Cty. Bd. of Cty. Comm’rs, 374 Fed. Appx. 821, 829–833 (CA10 2010); and Pinson v. Samuels, 761 F. 3d 1, 7–10 (CADC 2014) (case below) (adopting per-case approach), with Whitfield v. Scully, 241 F. 3d 264, 276–277 (CA2 2001); Siluk v. Merwin, 783 F. 3d 421, 427–436 (CA3 2015); and Torres v. O’Quinn, 612 F. 3d 237, 241–248 (CA4 2010) (adopting per-prisoner approach). 2 At oral argument, Bruce’s counsel informed the Court that Bruce had framed or joined 19 prison-litigation cases, although “the last seven or so have not been filed . . . because [Bruce] had had three strikes by the 12th.” Tr. of Oral Arg. 23. See Brief for Respondents 40 (stating that Bruce filed three new lawsuits during the pendency of his case before this Court). 3 The Court of Appeals construed the pleadings in this case as a petition for a writ of mandamus. 761 F. 3d, at 3. We assume without deciding that a mandamus petition qualifies as a “civil action” or “appeal” for purposes of 28 U. S. C. §1915(b). 4 Use of the plural “fees” in that paragraph does not persuade us otherwise. Congress has been less than meticulous in its employment of the singular “fee” and the plural “fees,” sometimes using those words interchangeably. See, e.g., 28 U. S. C. §1930(a) (“The parties commencing a case under title 11 shall pay to the clerk . . . the following filing fees: [enumerating several options]. In addition to the filing fee paid to the clerk, [an additional fee shall be paid].” (emphasis added)); 42 U. S. C. §1988(b) (“[T]he court . . . may allow the prevailing party . . . a reasonable attorney’s fee as part of the costs, except that in any action brought against a judicial officer . . . such officer shall not be held liable for any costs, including attorney’s fees . . . .” (emphasis added)). See also Dictionary Act, 1 U. S. C. §1 (“In determining the meaning of any Act of Congress, unless the context indicates otherwise—words importing the singular include and apply to several persons, parties, or things; words importing the plural include the singular . . . .”).

577.US.2015_14-857

The United States Navy contracted with petitioner Campbell-Ewald Company (Campbell) to develop a multimedia recruiting campaign that included the sending of text messages to young adults, but only if those individuals had “opted in” to receipt of marketing solicitations on topics that included Navy service. Campbell’s subcontractor Mindmatics LLC generated a list of cellular phone numbers for consenting 18- to 24-year-old users and then transmitted the Navy’s message to over 100,000 recipients, including respondent Jose Gomez, who alleges that he did not consent to receive text messages and, at age 40, was not in the Navy’s targeted age group. Gomez filed a nationwide class action, alleging that Campbell violated the Telephone Consumer Protection Act (TCPA), 47 U. S. C. §227(b)(1)(A)(iii), which prohibits “using any automatic dialing system” to send a text message to a cellular telephone, absent the recipient’s prior express consent. He sought treble statutory damages for a willful and knowing TCPA violation and an injunction against Campbell’s involvement in unsolicited messaging. Before the deadline for Gomez to file a motion for class certification, Campbell proposed to settle Gomez’s individual claim and filed an offer of judgment pursuant to Federal Rule of Civil Procedure 68. Gomez did not accept the offer and allowed the Rule 68 submission to lapse on expiration of the time (14 days) specified in the Rule. Campbell then moved to dismiss the case pursuant to Rule 12(b)(1) for lack of subject-matter jurisdiction. Campbell argued first that its offer mooted Gomez’s individual claim by providing him with complete relief. Next, Campbell urged that Gomez’s failure to move for class certification before his individual claim became moot caused the putative class claims to become moot as well. The District Court denied the motion. After limited discovery, the District Court granted Campbell’s motion for summary judgment. Relying on Yearsley v. W. A. Ross Constr. Co., 309 U. S. 18 , the court held that Campbell, as a contractor acting on the Navy’s behalf, acquired the Navy’s sovereign immunity from suit under the TCPA. The Ninth Circuit reversed. It agreed that Gomez’s case remained live but concluded that Campbell was not entitled to “derivative sovereign immunity” under Yearsley or on any other basis. Held: 1. An unaccepted settlement offer or offer of judgment does not moot a plaintiff’s case, so the District Court retained jurisdiction to adjudicate Gomez’s complaint. Article III’s “cases” and “controversies” limitation requires that “an actual controversy . . . be extant at all stages of review, not merely at the time the complaint is filed,” Arizonans for Official English v. Arizona, 520 U. S. 43 (internal quotation marks omitted), but a case does not become moot as “long as the parties have a concrete interest, however small,” in the litigation’s outcome, Chafin v. Chafin, 568 U. S. ___, ___ (internal quotation marks omitted). Gomez’s complaint was not effaced by Campbell’s unaccepted offer to satisfy his individual claim. Under basic principles of contract law, Campbell’s settlement bid and Rule 68 offer of judgment, once rejected, had no continuing efficacy. With no settlement offer operative, the parties remained adverse; both retained the same stake in the litigation they had at the outset. Neither Rule 68 nor the 19th-century railroad tax cases California v. San Pablo & Tulare R. Co., 149 U. S. 308 , Little v. Bowers, 134 U. S. 547 , and San Mateo County v. Southern Pacific R. Co., 116 U. S. 138 , support the argument that an unaccepted settlement offer can moot a complaint. Pp. 6–12. 2. Campbell’s status as a federal contractor does not entitle it to immunity from suit for its violation of the TCPA. Unlike the United States and its agencies, federal contractors do not enjoy absolute immunity. A federal contractor who simply performs as directed by the Government may be shielded from liability for injuries caused by its conduct. See Yearsley, 309 U. S., at 20–21. But no “derivative immunity” exists when the contractor has “exceeded [its] authority” or its authority “was not validly conferred.” Id., at 21. The summary judgment record includes evidence that the Navy authorized Campbell to send text messages only to individuals who had “opted in” to receive solicitations, as required by the TCPA. When a contractor violates both federal law and the Government’s explicit instructions, as alleged here, no immunity shields the contractor from suit. Pp. 12–14. 768 F. 3d 871, affirmed and remanded. Ginsburg, J., delivered the opinion of the Court, in which Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed an opinion concurring in the judgment. Roberts, C. J., filed a dissenting opinion, in which Scalia and Alito, JJ., joined. Alito, J., filed a dissenting opinion.

Is an unaccepted offer to satisfy the named plaintiff’s individual claim sufficient to render a case moot when the complaint seeks relief on behalf of the plaintiff and a class of persons similarly situated? This question, on which Courts of Appeals have divided, was reserved in Genesis HealthCare Corp. v. Symczyk, 569 U. S. ___, ___, ___, n. 4 (2013) (slip op., at 5, 6, n. 4). We hold today, in accord with Rule 68 of the Federal Rules of Civil Procedure, that an unaccepted settlement offer has no force. Like other unaccepted contract offers, it creates no lasting right or obligation. With the offer off the table, and the defendant’s continuing denial of liability, adversity between the parties persists. This case presents a second question. The claim in suit concerns performance of the petitioner’s contract with the Federal Government. Does the sovereign’s immunity from suit shield the petitioner, a private enterprise, as well? We hold that the petitioner’s status as a Government contractor does not entitle it to “derivative sovereign immunity,” i.e., the blanket immunity enjoyed by the sovereign. I The Telephone Consumer Protection Act (TCPA or Act) 48Stat. 1064, 47 U. S. C. §227(b)(1)(A)(iii), prohibits any person, absent the prior express consent of a telephone-call recipient, from “mak[ing] any call . . . using any automatic telephone dialing system . . . to any telephone number assigned to a paging service [or] cellular telephone service.” A text message to a cellular telephone, it is undisputed, qualifies as a “call” within the compass of §227(b)(1)(A)(iii). 768 F. 3d 871, 874 (CA9 2014). For damages occasioned by conduct violating the TCPA, §227(b)(3) authorizes a private right of action. A plaintiff successful in such an action may recover her “actual monetary loss” or $500 for each violation, “whichever is greater.” Damages may be trebled if “the defendant willfully or knowingly violated” the Act. Petitioner Campbell-Ewald Company (Campbell) is a nationwide advertising and marketing communications agency. Beginning in 2000, the United States Navy engaged Campbell to develop and execute a multimedia recruiting campaign. In 2005 and 2006, Campbell proposed to the Navy a campaign involving text messages sent to young adults, the Navy’s target audience, encouraging them to learn more about the Navy. The Navy approved Campbell’s proposal, conditioned on sending the messages only to individuals who had “opted in” to receipt of marketing solicitations on topics that included service in the Navy. App. 42. In final form, the message read: “Destined for something big? Do it in the Navy. Get a career. An education. And a chance to serve a greater cause. For a FREE Navy video call [ phone number].” 768 F. 3d, at 873. Campbell then contracted with Mindmatics LLC, which generated a list of cellular phone numbers geared to the Navy’s target audience—namely, cellular phone users between the ages of 18 and 24 who had consented to receiving solicitations by text message. In May 2006, Mindmatics transmitted the Navy’s message to over 100,000 recipients. Respondent Jose Gomez was a recipient of the Navy’s recruiting message. Alleging that he had never consented to receiving the message, that his age was nearly 40, and that Campbell had violated the TCPA by sending the message (and perhaps others like it), Gomez filed a class-action complaint in the District Court for the Central District of California in 2010. On behalf of a nationwide class of individuals who had received, but had not consented to receipt of, the text message, Gomez sought treble statutory damages, costs, and attorney’s fees, also an injunction against Campbell’s involvement in unsolicited messaging. App. 16–24. Prior to the agreed-upon deadline for Gomez to file a motion for class certification, Campbell proposed to settle Gomez’s individual claim and filed an offer of judgment pursuant to Federal Rule of Civil Procedure 68. App. to Pet. for Cert. 52a–61a.[1] Campbell offered to pay Gomez his costs, excluding attorney’s fees, and $1,503 per message for the May 2006 text message and any other text message Gomez could show he had received, thereby satisfying his personal treble-damages claim. Id., at 53a. Campbell also proposed a stipulated injunction in which it agreed to be barred from sending text messages in violation of the TCPA. The proposed injunction, however, denied liability and the allegations made in the complaint, and disclaimed the existence of grounds for the imposition of an injunction. Id., at 56a. The settlement offer did not include attorney’s fees, Campbell observed, because the TCPA does not provide for an attorney’s-fee award. Id., at 53a. Gomez did not accept the settlement offer and allowed Campbell’s Rule 68 submission to lapse after the time, 14 days, specified in the Rule. Campbell thereafter moved to dismiss the case pursuant to Federal Rule of Civil Procedure 12(b)(1) for lack of subject-matter jurisdiction. No Article III case or controversy remained, Campbell urged, because its offer mooted Gomez’s individual claim by providing him with complete relief. Gomez had not moved for class certification before his claim became moot, Campbell added, so the putative class claims also became moot. The District Court denied Campbell’s motion. 805 F. Supp. 2d 923 (CD Cal. 2011).[2] Gomez was not dilatory in filing his certification request, the District Court determined; consequently, the court noted, the class claims would “relat[e] back” to the date Gomez filed the complaint. Id., at 930–931. After limited discovery, Campbell moved for summary judgment on a discrete ground. The U. S. Navy enjoys the sovereign’s immunity from suit under the TCPA, Campbell argued. The District Court granted the motion. Relying on our decision in Yearsley v. W. A. Ross Constr. Co., 309 U. S. 18 (1940) , the court held that, as a contractor acting on the Navy’s behalf, Campbell acquired the Navy’s immunity. No. CV 10–02007DMG (CD Cal., Feb. 22, 2013), App. to Pet. for Cert. 22a–34a, 2013 WL 655237.The Court of Appeals for the Ninth Circuit reversed the summary judgment entered for Campbell. 768 F. 3d 871. The appeals court disagreed with the District Court’s ruling on the immunity issue, but agreed that Gomez’s case remained live. Concerning Gomez’s individual claim, the Court of Appeals relied on its then-recent decision in Diaz v. First American Home Buyers Protection Corp., 732 F. 3d 948 (2013). Diaz held that “an unaccepted Rule 68 offer that would fully satisfy a plaintiff’s [individual] claim is insufficient to render th[at] claim moot.” Id., at 950. As to the class relief Gomez sought, the Ninth Circuit held that “an unaccepted Rule 68 offer of judgment—for the full amount of the named plaintiff’s individual claim and made before the named plaintiff files a motion for class certification—does not moot a class action.” 768 F. 3d, at 875 (quoting Pitts v. Terrible Herbst, Inc., 653 F. 3d 1081, 1091–1092 (CA9 2011)). Next, the Court of Appeals held that Campbell was not entitled to “derivative sovereign immunity” under this Court’s decision in Yearsley or on any other basis. 768 F. 3d, at 879–881. Vacating the District Court’s judg-ment, the Ninth Circuit remanded the case for further proceedings.[3] We granted certiorari to resolve a disagreement among the Courts of Appeals over whether an unaccepted offer can moot a plaintiff’s claim, thereby depriving federal courts of Article III jurisdiction. Compare Bais Yaakov v. Act, Inc., 798 F. 3d 46, 52 (CA1 2015); Hooks v. Landmark Industries, Inc., 797 F. 3d 309, 315 (CA5 2015); Chapman v. First Index, Inc., 796 F. 3d 783, 787 (CA7 2015); Tanasi v. New Alliance Bank, 786 F. 3d 195, 200 (CA2 2015); Stein v. Buccaneers Limited Partnership, 772 F. 3d 698, 703 (CA11 2014); Diaz, 732 F. 3d, at 954–955 (holding that an unaccepted offer does not render a plaintiff’s claim moot), with Warren v. Sessoms & Rogers, P. A., 676 F. 3d 365, 371 (CA4 2012); O’Brien v. Ed Donnelly Enterprises, Inc., 575 F. 3d 567, 574–575 (CA6 2009); Weiss v. Regal Collections, 385 F. 3d 337, 340 (CA3 2004) (holding that an unaccepted offer can moot a plaintiff’s claim). We granted review as well to resolve the federal contractor immunity question Campbell’s petition raised. 575 U. S. ___ (2015). II Article III of the Constitution limits federal-court jurisdiction to “cases” and “controversies.” U. S. Const., Art. III, §2. We have interpreted this requirement to demand that “an actual controversy . . . be extant at all stages of review, not merely at the time the complaint is filed.” Arizonans for Official English v. Arizona, 520 U. S. 43, 67 (1997) (quoting Preiser v. Newkirk, 422 U. S. 395, 401 (1975) ). “If an intervening circumstance deprives the plaintiff of a ‘personal stake in the outcome of the lawsuit,’ at any point during litigation, the action can no longer proceed and must be dismissed as moot.” Genesis HealthCare Corp., 569 U. S., at ___ (slip op., at 4) (quoting Lewis v. Continental Bank Corp., 494 U. S. 472 –478 (1990)). A case becomes moot, however, “only when it is impossible for a court to grant any effectual relief what-ever to the prevailing party.” Knox v. Service Employees, 567 U. S. ___, ___ (2012) (slip op., at 7) (internal quotation marks omitted). “As long as the parties have a concrete interest, however small, in the outcome of the litigation, the case is not moot.” Chafin v. Chafin, 568 U. S. ___, ___ (2013) (slip op., at 6) (internal quotation marks omitted). In Genesis HealthCare, the Court considered a collective action brought by Laura Symczyk, a former employee of Genesis HealthCare Corp. Symczyk sued on behalf of herself and similarly situated employees for alleged violations of the Fair Labor Standards Act of 1938, 29 U. S. C. §201 et seq. In that case, as here, the defendant served the plaintiff with an offer of judgment pursuant to Rule 68 that would have satisfied the plaintiff’s individual dam-ages claim. 569 U. S., at ___ (slip op., at 2). Also as here, the plaintiff allowed the offer to lapse by failing to respond within the time specified in the Rule. Ibid. But unlike the case Gomez mounted, Symczyk did not dispute in the lower courts that Genesis HealthCare’s offer mooted her individual claim. Id., at ___ (slip op., at 5). Because of that failure, the Genesis HealthCare majority refused to rule on the issue. Instead, the majority simply assumed, without deciding, that an offer of complete relief pursuant to Rule 68, even if unaccepted, moots a plaintiff’s claim. Ibid. Having made that assumption, the Court proceeded to consider whether the action remained justiciable on the basis of the collective-action allegations alone. Absent a plaintiff with a live individual case, the Court concluded, the suit could not be maintained. Id., at ___ (slip op., at 6). Justice Kagan, writing in dissent, explained that she would have reached the threshold question and would have held that “an unaccepted offer of judgment cannot moot a case.” Id., at ___ (slip op., at 3). She reasoned: “When a plaintiff rejects such an offer—however good the terms—her interest in the lawsuit remains just what it was before. And so too does the court’s ability to grant her relief. An unaccepted settlement offer—like any unaccepted contract offer—is a legal nullity, with no operative effect. As every first-year law student learns, the recipient’s rejection of an offer ‘leaves the matter as if no offer had ever been made.’ Minneapolis & St. Louis R. Co. v. Columbus Rolling Mill, 119 U. S. 149, 151 (1886) . Nothing in Rule 68 alters that basic principle; to the contrary, that rule specifies that ‘[a]n unaccepted offer is considered withdrawn.’ Fed. Rule Civ. Proc. 68(b). So assuming the case was live before—because the plaintiff had a stake and the court could grant relief—the litigation carries on, unmooted.” Ibid. We now adopt Justice Kagan’s analysis, as has every Court of Appeals ruling on the issue post Genesis HealthCare.[4] Accordingly, we hold that Gomez’s complaint was not effaced by Campbell’s unaccepted offer to satisfy his individual claim. As earlier recounted, see supra, at 3–4, Gomez commenced an action against Campbell for violation of the TCPA, suing on behalf of himself and others similarly situated. Gomez sought treble statutory damages and an injunction on behalf of a nationwide class, but Campbell’s settlement offer proposed relief for Gomez alone, and it did not admit liability. App. to Pet. for Cert. 58a. Gomez rejected Campbell’s settlement terms and the offer of judgment. Under basic principles of contract law, Campbell’s settlement bid and Rule 68 offer of judgment, once rejected, had no continuing efficacy. See Genesis HealthCare, 569 U. S., at ___ (Kagan, J., dissenting) (slip op., at 3). Absent Gomez’s acceptance, Campbell’s settlement offer remained only a proposal, binding neither Campbell nor Gomez. See App. to Pet. for Cert. 59a (“Please advise whether Mr. Gomez will accept [Campbell’s] offer . . . .”). Having rejected Campbell’s settlement bid, and given Campbell’s continuing denial of liability, Gomez gained no entitlement to the relief Campbell previously offered. See Eliason v. Henshaw, 4 Wheat. 225, 228 (1819) (“It is an undeniable principle of the law of contracts, that an offer of a bargain by one person to another, imposes no obligation upon the former, until it is accepted by the latter . . . .”). In short, with no settlement offer still operative, the parties remained adverse; both retained the same stake in the litigation they had at the outset. The Federal Rule in point, Rule 68, hardly supports the argument that an unaccepted settlement offer can moot a complaint. An offer of judgment, the Rule provides, “is considered withdrawn” if not accepted within 14 daysof its service. Fed. Rule Civ. Proc. 68(a), (b). The sole built-in sanction: “If the [ultimate] judgment . . . is not more favorable than the unaccepted offer, the offeree must pay the costs incurred after the offer was made.” Rule 68(d). In urging that an offer of judgment can render a controversy moot, Campbell features a trio of 19th-century railroad tax cases: California v. San Pablo & Tulare R. Co., 149 U. S. 308 (1893) , Little v. Bowers, 134 U. S. 547 (1890) , and San Mateo County v. Southern Pacific R. Co., 116 U. S. 138 (1885) . None of those decisions suggests that an unaccepted settlement offer can put a plaintiff out of court. In San Pablo, California had sued to recover state and county taxes due from a railroad. In response, the railroad had not merely offered to pay the taxes in question. It had actually deposited the full amount demanded in a California bank in the State’s name, in accord with a California statute that “extinguished” the railroad’s tax obligations upon such payment. 149 U. S., at 313–314. San Pablo thus rested on California’s substantive law, which required the State to accept a taxpayer’s full payment of the amount in controversy. San Mateo and Little similarly involved actual payment of the taxes for which suit was brought. In all three cases, the railroad’s payments had fully satisfied the asserted tax claims, and so extinguished them. San Mateo, 116 U. S., at 141–142; Little, 134 U. S., at 556.[5] In contrast to the cases Campbell highlights, when the settlement offer Campbell extended to Gomez expired, Gomez remained emptyhanded; his TCPA complaint, which Campbell opposed on the merits, stood wholly unsatisfied. Because Gomez’s individual claim was not made moot by the expired settlement offer, that claim would retain vitality during the time involved in determining whether the case could proceed on behalf of a class. While a class lacks independent status until certified, see Sosna v. Iowa, 419 U. S. 393, 399 (1975) , a would-be class representative with a live claim of her own must be accorded a fair opportunity to show that certification is warranted. The Chief Justice’s dissent asserts that our decision transfers authority from the federal courts and “hands it to the plaintiff.” Post, at 10. Quite the contrary. The dissent’s approach would place the defendant in the driver’s seat. We encountered a kindred strategy in U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18 (1994) . The parties in Bancorp had reached a voluntary settlement while the case was pending before this Court. Id., at 20. The petitioner then sought vacatur of the Court of Appeals’ judgment, contending that it should be relieved from the adverse decision on the ground that the settlement made the dispute moot. The Court rejected this gambit. Id., at 25. Similarly here, Campbell sought to avoid a potential adverse decision, one that could expose it to damages a thousand-fold larger than the bid Gomez declined to accept. In sum, an unaccepted settlement offer or offer of judgment does not moot a plaintiff’s case, so the District Court retained jurisdiction to adjudicate Gomez’s complaint. That ruling suffices to decide this case. We need not, and do not, now decide whether the result would be different if a defendant deposits the full amount of the plaintiff’s individual claim in an account payable to the plaintiff, and the court then enters judgment for the plaintiff in that amount. That question is appropriately reserved for a case in which it is not hypothetical. III The second question before us is whether Campbell’s status as a federal contractor renders it immune from suit for violating the TCPA by sending text messages to unconsenting recipients. The United States and its agencies, it is undisputed, are not subject to the TCPA’s prohibitions because no statute lifts their immunity. Brief for Peti-tioner 2; Brief for Respondent 43. Do federal contractors share the Government’s unqualified immunity from liability and litigation? We hold they do not. “[G]overnment contractors obtain certain immunity in connection with work which they do pursuant to their contractual undertakings with the United States.” Brady v. Roosevelt S. S. Co., 317 U. S. 575, 583 (1943) . That immunity, however, unlike the sovereign’s, is not absolute. See id., at 580–581. Campbell asserts “derivative sovereign immunity,” Brief for Petitioner 35, but can offer no authority for the notion that private persons performing Government work acquire the Government’s embracive immunity. When a contractor violates both federal law and the Government’s explicit instructions, as here alleged, no “derivative immunity” shields the contractor from suit by persons adversely affected by the violation. Campbell urges that two of our decisions support its “derivative immunity” defense: Yearsley, 309 U. S. 18 , and Filarsky v. Delia, 566 U. S. ___ (2012). In Yearsley, a landowner asserted a claim for damages against a private company whose work building dikes on the Missouri River pursuant to its contract with the Federal Government had washed away part of the plaintiff’s land. We held that the contractor was not answerable to the landowner. “[T]he work which the contractor had done in the river bed,” we observed, “was all authorized and directed by the Government of the United States” and “performed pursuant to the Act of Congress.” 309 U. S., at 20 (internal quotation marks omitted). Where the Government’s “authority to carry out the project was validly conferred, that is, if what was done was within the constitutional power of Congress,” we explained, “there is no liability on the part of the contractor” who simply performed as the Government directed. Id., at 20–21.[6] The Court contrasted with Yearsley cases in which a Government agent had “exceeded his authority” or the authority “was not validly conferred”; in those circumstances, the Court said, the agent could be held liable for conduct causing injury to another. Id., at 21.[7] In Filarsky, we considered whether a private attorney temporarily retained by a municipal government as an investigator could claim qualified immunity in an action brought under 42 U. S. C. §1983. Finding no distinction in the common law “between public servants and private individuals engaged in public service,” we held that the investigator could assert “qualified immunity” in the lawsuit. 566 U. S., at ___, ___ (slip op., at 8, 5). Qualified immunity reduces the risk that contractors will shy away from government work. But the doctrine is bounded in a way that Campbell’s “derivative immunity” plea is not. “Qualified immunity may be overcome . . . if the defendant knew or should have known that his conduct violated a right ‘clearly established’ at the time of the episode in suit.” Id., at ___ (Ginsburg, J., concurring) (slip op., at 1) (citing Harlow v. Fitzgerald, 457 U. S. 800, 818 (1982) ). Campbell does not here contend that the TCPA’s requirements or the Navy’s instructions failed to qualify as “clearly established.” At the pretrial stage of litigation, we construe the record in a light favorable to the party seeking to avoid summary disposition, here, Gomez. Matsushita Elec. Industrial Co. v. Zenith Radio Corp., 475 U. S. 574, 587 (1986) . In opposition to summary judgment, Gomez presented evidence that the Navy authorized Campbell to send text messages only to individuals who had “opted in” to receive solicitations. App. 42–44; 768 F. 3d, at 874. A Navy representative noted the importance of ensuring that the message recipient list be “kosher” (i.e., that all recipients had consented to receiving messages like the recruiting text), and made clear that the Navy relied on Campbell’s representation that the list was in compliance. App. 43. See also ibid. (noting that Campbell itself encouraged the Navy to use only an opt-in list in order to meet national and local law requirements). In short, the current record reveals no basis for arguing that Gomez’s right to remain message-free was in doubt or that Campbell complied with the Navy’s instructions. We do not overlook that subcontractor Mindmatics, not Campbell, dispatched the Navy’s recruiting message to unconsenting recipients. But the Federal Communications Commission has ruled that, under federal common-law principles of agency, there is vicarious liability for TCPA violations. In re Joint Petition Filed by Dish Network, LLC, 28 FCC Rcd. 6574 (2013). The Ninth Circuit deferred to that ruling, 768 F. 3d, at 878, and we have no cause to question it. Campbell’s vicarious liability for Mindmatics’ conduct, however, in no way advances Campbell’s contention that it acquired the sovereign’s immunity from suit based on its contract with the Navy. * * * For the reasons stated, the judgment of the Court of Appeals for the Ninth Circuit is affirmed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Federal Rule of Civil Procedure 68 provides, in relevant part: 2 Because Campbell had already answered the complaint, the District Court construed Campbell’s motion as a request for summary judgment. 805 F. Supp. 2d, at 927, n. 2. 3 The Court of Appeals stayed its mandate pending proceedings in this Court. App. to Pet. for Cert. 62a–63a. 4 See Bais Yaakov v. Act, Inc., 798 F. 3d 46, 51–52 (CA1 2015); Hooks v. Landmark Industries, Inc., 797 F. 3d 309, 314–315 (CA5 2015); Chapman v. First Index, Inc., 796 F. 3d 783, 786–787 (CA7 2015); Tanasi v. New Alliance Bank, 786 F. 3d 195, 199–200 (CA2 2015); Stein v. Buccaneers Limited Partnership, 772 F. 3d 698, 702–703 (CA11 2014); Diaz v. First American Home Buyers Corp., 732 F. 3d 948, 953–955 (CA9 2013). 5 In addition to California v. San Pablo & Tulare R. Co., 149 U. S. 308 (1893) , The Chief Justice maintains, two recent decisions of the Court support its position: Alvarez v. Smith, 558 U. S. 87 (2009) , and Already, LLC v. Nike, Inc., 568 U. S. ___ (2013). See post, at 6–9 (dissenting opinion). The Court’s reasoning in those opinions, however, is consistent with our decision in this case. In Alvarez, the Court found moot claims for injunctive and declaratory relief in relation to cars and cash seized by the police. Through separate state-court proceedings, the State had “returned all the cars that it seized,” and the plaintiff-property owners had “either forfeited any relevant cash or ha[d] accepted as final the State’s return of some of it.” 558 U. S., at 89, 95–96. Alvarez thus resembles the railroad tax cases described above: The Alvarez plaintiffs had in fact received all the relief they could claim, all “underlying property disputes” had ended, id., at 89, and as the complaint sought “only declaratory and injunctive relief, not damages,” id., at 92, no continuing controversy remained. 6 If there had been a taking of the plaintiff’s property, the Courtnoted, “a plain and adequate remedy” would be at hand, i.e., recovery from the United States of “just compensation.” Yearsley, 309 U. S., at 21. 7 We disagree with the Court of Appeals to the extent that it described Yearsley as “establish[ing] a narrow rule regarding claims arising out of property damage caused by public works projects.” 768 F. 3d, at 879. Critical in Yearsley was not the involvement of public works, but the contractor’s performance in compliance with all federal directions.

578.US.2015_14-1375

Petitioner CRST, a trucking company using a system under which two employees share driving duties on a single truck, requires its drivers to graduate from the company’s training program before becoming a certified driver. In 2005, new driver Monika Starke filed a charge with the Equal Employment Opportunity Commission (Commission), alleging that she was sexually harassed by two male trainers during the road-trip portion of her training. Following the procedures set out in Title VII of the Civil Rights Act of 1964, see 42 U. S. C. §2000e–5(b), the Commission informed CRST about the charge and investigated the allegation, ultimately informing CRST that it had found reasonable cause to believe that CRST subjected Starke and “a class of employees and prospective employees to sexual harassment” and offering to conciliate. In 2007, having determined that conciliation had failed, the Commission, in its own name, filed suit against CRST under §706 of Title VII. During discovery, the Commission identified over 250 allegedly aggrieved women. The District Court, however, dismissed all of the claims, including those on behalf of 67 women, which, the court found, were barred on the ground that the Commission had not adequately investigated or attempted to conciliate its claims on their behalf before filing suit. The District Court then dismissed the suit, held that CRST is a prevailing party, and invited CRST to apply for attorney’s fees. CRST filed a motion for attorney’s fees. The District Court awarded the company over $4 million in fees. The Eighth Circuit reversed the dismissal of only two claims—on behalf of Starke and one other employee—but that led it to vacate, without prejudice, the attorney’s fees award. On remand, the Commission settled the claim on behalf of Starke and withdrew the other. CRST again sought attorney’s fees, and the District Court again awarded it more than $4 million, finding that CRST had prevailed on the claims for over 150 of the allegedly aggrieved women, including the 67 claims dismissed because of the Commission’s failure to satisfy its presuit requirements. The Eighth Circuit reversed and remanded once more. It held that a Title VII defendant can be a “prevailing party” only by obtaining a “ruling on the merits,” and that the District Court’s dismissal of the claims was not a ruling on the merits. Held: A favorable ruling on the merits is not a necessary predicate to find that a defendant is a prevailing party. Pp. 11–16. (a) Common sense undermines the notion that a defendant cannot “prevail” unless the relevant disposition is on the merits. A plaintiff seeks a material alteration in the legal relationship between the parties. But a defendant seeks to prevent an alteration in the plaintiff’s favor, and that objective is fulfilled whenever the plaintiff’s challenge is rebuffed, irrespective of the precise reason for the court’s decision, i.e., even if the court’s final judgment rejects the plaintiff’s claim for a nonmerits reason. There is no indication that Congress intended that defendants should be eligible to recover attorney’s fees only when courts dispose of claims on the merits. Title VII’s fee-shifting statute allows prevailing defendants to recover whenever the plaintiff’s “claim was frivolous, unreasonable, or groundless.” Christiansburg Garment Co. v. EEOC, 434 U. S. 412 . Congress thus must have intended that a defendant could recover fees expended in frivolous, unreasonable, or groundless litigation when the case is resolved in the defendant’s favor, whether on the merits or not. Christiansburg itself involved a defendant’s request for attorney’s fees in a case where the District Court had rejected the plaintiff’s claim for a non-merits reason. Various Courts of Appeals likewise have applied the Christiansburg standard when claims were dismissed for nonmerits reasons. Pp. 11–14. (b) The Court declines to decide the argument, raised by the Commission for the first time during the merits stage of this case, whether a defendant must obtain a preclusive judgment in order to prevail. The Commission’s failure to articulate its preclusion theory earlier has resulted in inadequate briefing on the issue, and the parties dispute whether the District Court’s judgment was in fact preclusive. The Commission also submits that the Court should affirm on the alternative ground that, even if CRST is a prevailing party, the Commission’s position that it had satisfied its presuit obligations was not frivolous, unreasonable, or groundless. These matters are left for the Eighth Circuit to consider in the first instance. It is not this Court’s usual practice to adjudicate either legal or predicate factual questions in the first instance, see Adarand Constructors, Inc. v. Mineta, 534 U. S. 103 , and that is the proper course here, given the extensive record in this case and the Commission’s change of position between the certiorari and merits stages. Pp. 14–16. 774 F. 3d 1169, vacated and remanded. Kennedy, J., delivered the opinion for a unanimous Court. Thomas, J., filed a concurring opinion.

This case involves the interpretation of a statutory provision allowing district courts to award attorney’s fees to defendants in employment discrimination actions. Under Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U. S. C. §2000e et seq., which prohibits discrimination in employment, a district court may award attorney’s fees to “the prevailing party.” §2000e–5(k). The Court of Appeals for the Eighth Circuit held that a Title VII defendant prevails only by obtaining a “ruling on the merits.” 774 F. 3d 1169, 1179 (2014); Marquart v. Lodge 837, Machinists and Aerospace Workers, 26 F. 3d 842, 851–852 (1994). This Court disagrees with that conclusion. The Court now holds that a favorable ruling on the merits is not a necessary predicate to find that a defendant has prevailed. I Title VII of the Civil Rights Act of 1964 authorizes an award of attorney’s fees in certain circumstances. The statute provides that “[i]n any action or proceeding under this subchapter the court, in its discretion, may allow the prevailing party, other than the [Equal Employment Opportunity] Commission or the United States, a reasonable attorney’s fee (including expert fees) as part of the costs, and the Commission and the United States shall be liable for costs the same as a private person.” §2000e–5(k). Before deciding whether an award of attorney’s fees is appropriate in a given case, then, a court must determine whether the party seeking fees has prevailed in the litigation. Texas State Teachers Assn. v. Garland Independent School Dist., 489 U. S. 782, 789 (1989) ; Hensley v. Eckerhart, 461 U. S. 424, 433 (1983) . Congress has included the term “prevailing party” in various fee-shifting statutes, and it has been the Court’s approach to interpret the term in a consistent manner. See Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health and Human Resources, 532 U. S. 598 , and n. 4 (2001). The Court has said that the “touchstone of the prevailing party inquiry must be the material alteration of the legal relationship of the parties.” Texas State Teachers Assn., supra, at 792–793. This change must be marked by “judicial imprimatur.” Buckhannon, 532 U. S., at 605. The Court has explained that, when a plaintiff secures an “enforceable judgmen[t] on the merits” or a “court-ordered consent decre[e],” that plaintiff is the prevailing party because he has received a “judicially sanctioned change in the legal relationship of the par-ties.” Id., at 604–605. The Court, however, has not set forth in detail how courts should determine whether a defendant has prevailed. Although the Court has not articulated a precise test for when a defendant is a prevailing party, in the Title VII context it has addressed how defendants should be treated under the second part of the inquiry—whether the district court should exercise its discretion to award fees to the prevailing party. When a defendant is the prevailing party on a civil rights claim, the Court has held, district courts may award attorney’s fees if the plaintiff’s “claim was frivolous, unreasonable, or groundless,” or if “the plaintiff continued to litigate after it clearly became so.” Christiansburg Garment Co. v. EEOC, 434 U. S. 412, 422 (1978) ; see also id., at 421. The Court of Appeals’ determination of the first part of the fee-shifting inquiry—whether petitioner is a prevailing party—presents the central issue in this case. Before addressing this question, however, a discussion of the facts and complex procedural history is warranted. II Petitioner CRST is a trucking company that employs a team driving system under which two employees share driving duties on a single truck. CRST requires its drivers to graduate from the company’s training program before becoming a certified driver. Part of that training is a 28-day over-the-road trip with a veteran driver. In 2005, a new driver named Monika Starke filed a charge of discrimination with the Equal Employment Opportunity Commission (Commission) alleging that two male trainers sexually harassed her during her over-the-road training trip. The Commission’s receipt of a charge of an unlawful workplace practice starts Title VII’s “detailed, multi-step procedure through which the Commission enforces the statute’s prohibition on employment discrimination.” Mach Mining, LLC v. EEOC, 575 U. S. ___, ___ (2015) (slip op., at 1). Under §706 of Title VII, the Commission first must inform the employer about the charge and the details of the allegations. 42 U. S. C. §2000e–5(b). The Commission next must investigate the allegation. Ibid. If the agency “determines after such investigation that there is not reasonable cause to believe that the charge is true,” it shall dismiss the charge and notify the parties. Ibid. At that point, the Commission is no longer involved, and the aggrieved individual may sue the employer in his or her own name. §2000e–5(f)(1). If, on the other hand, the Commission determines that there is reasonable cause to believe that a Title VII violation did occur, it “shall endeavor to eliminate any such alleged unlawful employment practice by informal methods of conference, conciliation, and persuasion.” §2000e–5(b). Only if the agency’s attempt at conciliation fails may it file a court action in its own name on behalf of the aggrieved person who brought the charge. §2000e–5(f)(1). Following these procedures, the Commission notified CRST of Starke’s charge and requested information regarding Starke’s allegations. In response CRST denied any wrongdoing. During the investigation, the Commission discovered that four other women had filed formal charges against the company with the Commission. The Commission then sent CRST several followup requests. It asked if CRST had received other allegations of harassment, demanded contact information for any women who were instructed by the trainers Starke accused of harassment, and sought “detailed contact information for” CRST’s dispatchers and female drivers. EEOC v. CRST Van Expedited, Inc., 679 F. 3d 657, 667 (CA8 2012). Over a year and a half after Starke filed her charge, the Commission sent CRST a letter of determination informing the company that the Commission had found reason-able cause to believe that CRST subjected Starke and “a class of employees and prospective employees to sexual harassment” and offering to conciliate. App. 811. Counsel for the Commission and for CRST discussed conciliation, but were unable to reach an agreement, and the Commission promptly notified the company that, in the agency’s view, the conciliation efforts had failed. In September 2007 the Commission, in its own name, filed suit against CRST under §706 of Title VII. It alleged that CRST subjected Starke and “[o]ther similarly situated . . . employees of CRST . . . to sexual harassment and a sexually hostile and offensive work environment” in violation of §§703(a) and 704(a) of Title VII, 42 U. S. C. §§2000e–2 and 2000e–3. App. 794–795. The Commission is allowed to “seek specific relief for a group of aggrieved individuals [under §706] without first obtaining class certification pursuant to” Federal Rule of Civil Procedure 23, because that rule “is not applicable to” a §706 enforcement action. General Telephone Co. of Northwest v. EEOC, 446 U. S. 318 –334 (1980). The Commission sought to enjoin CRST from engaging in discriminatory employment practices and to obtain an order requiring CRST to take proactive steps to remedy and prevent sex-based discrimination in the workplace. The Commission also sought damages and costs. During discovery, the Commission identified over 250 allegedly aggrieved women—far more than the Commission had forecast. CRST filed a motion for an order to show cause, alleging that the Commission “did not have a good-faith basis” for seeking relief on behalf of all the women. EEOC v. CRST Van Expedited, Inc., 2009 WL 2524402, *10 (ND Iowa, Aug. 13, 2009). The District Court did not strike any allegedly aggrieved persons at that time, although it did note its concern “that CRST still might unfairly face a ‘moving target’ of prospective plaintiffs as discovery winds down and trial approaches.” Ibid. (alteration and internal quotation marks omitted). The District Court proceeded to dispose of the Commission’s claims in a series of orders responsive to various motions filed by CRST. Section 707 of Title VII authorizes the Commission to bring a claim “that any person or group of persons is engaged in a pattern or practice” of illegal sex-based discrimination. See 42 U. S. C. §2000e–6. In the early stage of this litigation the Commission “made clear to the [district] court and CRST that it believe[d] CRST had engaged in ‘a pattern or practice’ of tolerating sexual harassment.” Order in No. 07–CV–95 (ND Iowa), Doc. 197, p. 25. CRST sought summary judgment on the Commission’s perceived pattern-or-practice claim. The District Court granted the motion. The court explained that, although courts have allowed the Commission to use a pattern-or-practice theory when litigating a §706 claim, the Commission did not plead a violation of §707 or use the phrase “pattern or practice” in its complaint. Id., at 24–25. Instead, the “[Commission’s] Complaint reads as if the [Commission] were asserting a prototypical §706 action.” Ibid. But, the court noted, CRST did not argue that the Commission failed to state a pattern-or-practice claim in the complaint; and the court presumed that CRST would not have sought summary judgment on a claim “it does not believe to exist.” Id., at 26. Because both parties accepted that the claim was live, “the court assume[d] without deciding that this is a sexual harassment pattern or practice case.” Ibid. After reviewing the parties’ arguments, the court held that the Commission had “not established a pattern or practice of tolerating sexual harassment” and dismissed with prejudice the assumed pattern-or-practice claim. Id., at 67. The court, as a final matter, advised that “[n]othing in this opinion . . . should be construed as a final ruling on the individual claims of sexual harassment that the [Commission] presses in this action.” Ibid. Next, the District Court ruled in several orders that the Commission’s claims on behalf of all but 67 of the women were barred on a variety of grounds. The court had previously dismissed claims on behalf of nearly 100 women as a discovery sanction due to the Commission’s failure to produce the women for deposition. In rejecting the Commission’s other claims, the court relied on (1) the expiration of the statute of limitations; (2) judicial estoppel; (3) the employee’s failure to report the alleged harassment in a timely fashion; (4) CRST’s prompt and effective response to reports of harassment; and (5) the lack of severity or pervasiveness of the alleged harassment. The District Court then barred the Commission from seeking relief for the remaining 67 women on the ground that the Commission had not satisfied its §706 presuit requirements before filing the lawsuit. The court concluded that the suit was “one of those exceptionally rare” cases where the Commission “wholly abandoned its statutory duties” to investigate and conciliate. CRST Van Expedited, Inc., 2009 WL 2524402, at *16. The court noted, how-ever, that it “expresse[d] no view as to whether the [Commission’s] investigation, determination and conciliation of Starke’s Charge would be sufficient to support a pattern[-]or-practice lawsuit.” Ibid., n. 21. The District Court then dismissed the suit, held that CRST is a prevailing party, and invited CRST to apply for attorney’s fees. CRST filed a motion for attorney’s fees. After describing how it disposed of the Commission’s claims piece by piece, the District Court held that the Commission’s failure to satisfy its presuit obligations for its claims on behalf of the final 67 women was “unreasonable,” and that an award of attorney’s fees was therefore appropriate. App. 140. The court awarded CRST over $4 million in attorney’s fees. Id., at 173–174. The Commission appealed the District Court’s order dismissing the claims on behalf of the 67 women that the District Court rejected for failure to satisfy Title VII’s presuit requirements as well as the District Court’s dismissal of some of the Commission’s other claims. As relevant here, the Court of Appeals held that the District Court’s dismissal of the 67 claims for a lack of investigation and conciliation was proper. The Commission, according to the Court of Appeals, “did not reasonably investigate the class allegations of sexual harassment during a reasonable investigation of the charge,” but rather used “discovery in the resulting lawsuit as a fishing expedition to uncover more violations.” CRST Van Expedited, Inc., 679 F. 3d, at 676 (internal quotation marks omitted). The Commission in fact “did not investigate the specific allegations of any of the 67 allegedly aggrieved persons . . . until after the Complaint was filed.” Ibid. (internal quotation marks omitted). The Court of Appeals affirmed the District Court’s dismissal of almost all of the other claims on which the Commission had appealed, reversing only the claims on behalf of Starke and one other employee—Tillie Jones— for reasons not material to the question at issue here. Like the District Court before it, the Court of Appeals declined to comment on whether the presuit investigation and attempted conciliation would have been sufficient to support a pattern-or-practice claim. The Court of Appeals also vacated, without prejudice, the attorney’s fees award. “In light of our reversals” of the District Court’s summary-judgment orders with respect to Starke and Jones, the court reasoned, “CRST is no longer a ‘prevailing’ defendant because the [Commission] still asserts live claims against it.” Id., at 694–695. Judge Murphy dissented from the court’s holding that the Commission had failed to satisfy its obligation to investigate and conciliate the final 67 claims, arguing that the Commission did not need to “complete its presuit duties for each individual alleged victim of discrimination when pursuing a class claim.” Id., at 695. After the case was remanded, the Commission withdrew its claim on behalf of Jones and settled its claim on behalf of Starke. The Commission thus had no claims left. The company again moved for attorney’s fees, and the District Court again awarded CRST more than $4 million in fees. The court first concluded “that this case contained multiple and distinct claims for relief,” thereby rejecting the Commission’s contention that it had brought a single claim on which it had prevailed. 2013 WL 3984478, *9 (ND Iowa, Aug. 1, 2013). Noting that the defendant does not have to prevail on every claim in a suit to obtain attorney’s fees, see Fox v. Vice, 563 U. S. 826 (2011) , the court then determined the claims on which CRST had prevailed. Applying Circuit precedent requiring a ruling on the merits of a claim before a defendant can be considered a prevailing party, the court found that CRST did not prevail on the claims that were dismissed because of the Commission’s failure to produce many of the allegedly aggrieved women for deposition. The court also found that CRST had not prevailed on the merits with respect to a handful of the Commission’s other claims. The court found that CRST did prevail, however, on the Commission’s pattern-or-practice claim and on the claims on behalf of over 150 of the allegedly aggrieved women, including the 67 claims dismissed because of the Commission’s failure to satisfy its presuit requirements. The court held that its dismissal of those 67 claims was a ruling on the merits because the Commission’s obligation to investigate and conciliate “is not a jurisdictional prerequisite; rather, it is an ingredient of the [Commission’s] claim.” 2013 WL 3984478, at *10. The court further concluded that an award of attorney’s fees was appropriate because the Commission’s failure to investigate and conciliate those 67 claims was unreasonable, as were the pattern-or-practice claim and the other claims on which it prevailed. The Commission appealed, and the Court of Appeals again reversed and remanded. The Court of Appeals first agreed with the District Court that the Commission brought many individual claims, not just a single claim. The Court of Appeals disagreed, however, with the District Court’s conclusion that CRST could recover attorney’s fees for the pattern-or-practice claim. The Commission did not allege a pattern-or-practice claim in its complaint, the Court of Appeals noted, and the District Court had “merely assumed without deciding that the [Commission]brought a pattern-or-practice claim.” 774 F. 3d, at 1179. The Court of Appeals concluded that the District Court erred by awarding fees “based on a purported” claim. Ibid. The Court of Appeals, bound by its own precedent in Marquart, then held that before a defendant can be deemed to have prevailed and to be eligible for fees there must have been a favorable “ ‘judicial determination . . . on the merits.’ ” 774 F. 3d, at 1179 (quoting Marquart, 26 F. 3d, at 852). A merits-based disposition is necessary, the court reasoned, because “ ‘[p]roof that a plaintiff’s case is frivolous, unreasonable, or groundless is not possible without a judicial determination of the plaintiff’s case on the merits.’ ” 774 F. 3d, at 1179 (quoting Marquart, supra, at 852). A case has not been decided on the merits, according to the Court of Appeals, if the defendant secured a “dismissal for lack of subject matter jurisdiction, on res judicata grounds, . . . on statute-of-limitations grounds,” or for something similar. 774 F. 3d, at 1179. The Court of Appeals distinguished “claim elements,” on the one hand, from “jurisdictional prerequisites or nonjurisdictional prerequisites to filing suit,” on the other. Id., at 1180. As relevant here, the court held that because Title VII’s presuit requirements are not elements of a Title VII claim, the dismissal of the claims regarding the 67 women on the ground that the Commission failed to investigate or conciliate was not a ruling on the merits, and CRST did not prevail on those claims. Id., at 1181. As a result, the court concluded, CRST was “not entitled to an award of attorneys’ fees on such claims.” Ibid. The Court of Appeals also criticized the District Court for “mak[ing] a universal finding that all of the [Commission’s] claims were without foundation,” instead of laying out “particularized findings . . . as to each individual claim upon which it granted summary judgment on the merits to CRST.” Id., at 1183. Such findings are necessary, the court reasoned, to avoid providing the defendant with “ ‘compensation for any fees that he would have paid in the absence of the frivolous claims.’ ” Ibid. (quoting Fox, supra, at 841). In particular, the court found it “problematic” that the District Court’s blanket finding included “(1) the purported pattern-or-practice claim and (2) the claims dismissed for the [Commission’s] failure to satisfy its presuit obligations.” 774 F. 3d, at 1183. The District Court was ordered to undertake a proper, particularized inquiry on remand. By precluding the defendant from recovering attorney’s fees when the claims in question have been dismissed because the Commission failed to satisfy its presuit obligations, the decision of the Court of Appeals conflicts with the decisions of three other Courts of Appeals. See EEOC v. Propak Logistics, Inc., 746 F. 3d 145, 152–154 (CA4 2014); EEOC v. Asplundh Tree Expert Co., 340 F. 3d 1256, 1261 (CA11 2003); EEOC v. Pierce Packing Co., 669 F. 2d 605, 608–609 (CA9 1982). This Court granted certiorari. 577 U. S. ___ (2015). III A The Court of Appeals held that CRST did not prevail on the claims brought on behalf of 67 women because the District Court’s disposition of these claims for failure to investigate and conciliate was not a ruling on the merits. In this Court the Commission now takes the position that the court erred by applying an on-the-merits requirement. Brief for Respondent 29 (“[A]sking whether a judgment is ‘on the merits’ in some abstract sense risks confusion”); Tr. of Oral Arg. 30 (“We have abandoned the Eighth Circuit’s view that you need a disposition on the merits”). This Court agrees and now holds that a defendant need not obtain a favorable judgment on the merits in order to be a “prevailing party.” Common sense undermines the notion that a defendant cannot “prevail” unless the relevant disposition is on the merits. Plaintiffs and defendants come to court with different objectives. A plaintiff seeks a material alteration in the legal relationship between the parties. A defendant seeks to prevent this alteration to the extent it is in the plaintiff’s favor. The defendant, of course, might prefer a judgment vindicating its position regarding the substantive merits of the plaintiff’s allegations. The defendant has, however, fulfilled its primary objective whenever the plaintiff’s challenge is rebuffed, irrespective of the precise reason for the court’s decision. The defendant may prevail even if the court’s final judgment rejects the plaintiff’s claim for a nonmerits reason. There is no indication that Congress intended that defendants should be eligible to recover attorney’s fees only when courts dispose of claims on the merits. The congressional policy regarding the exercise of district court discretion in the ultimate decision whether to award fees does not distinguish between merits-based and non-merits-based judgments. Rather, as the Court explained in Christiansburg Garment Co. v. EEOC, one purpose of the fee-shifting provision is “to deter the bringing of lawsuits without foundation.” 434 U. S., at 420 (internal quotation marks omitted); see also Fox, 563 U. S., at 836 (noting, in the context of 42 U. S. C. §1988’s closely related provision, that Congress wanted “to relieve defendants of the burdens associated with fending off frivolous litigation”). The Court, therefore, has interpreted the statute to allow prevailing defendants to recover whenever the plaintiff’s “claim was frivolous, unreasonable, or groundless.” Christiansburg, supra, at 422. It would make little sense if Congress’ policy of “sparing defendants from the costs of frivolous litigation,” Fox, supra, at 840, depended on the distinction between merits-based and non-merits-based frivolity. Congress must have intended that a defendant could recover fees expended in frivolous, unreasonable, or groundless litigation when the case is resolved in the defendant’s favor, whether on the merits or not. Imposing an on-the-merits requirement for a defendant to obtain prevailing party status would undermine that congressional policy by blocking a whole category of defendants for whom Congress wished to make fee awards available. Christiansburg itself involved a defendant’s request for attorney’s fees in a case where the District Court had rejected the plaintiff’s claim for a nonmerits reason. That case involved a claim under Title VII, as originally enacted, which did not give the Commission the authority to sue in its own name on behalf of an aggrieved person. Rosa Helm had filed a charge of discrimination against Christiansburg Garment Co. with the Commission in 1968. A few years later, the Commission determined that its conciliation efforts had failed and told Helm of her right to sue Christiansburg, which she did not exercise. Then in 1972, Congress amended Title VII to allow the Commission to sue in its own name on behalf of an aggrieved person, including where the employee’s charge was “pending with the Commission” when the amendments took effect. Equal Employment Opportunity Act of 1972, §14, 86Stat. 103. The Commission sued Christiansburg based on Helm’s charge, but the District Court granted summary judgment to the defendant on the ground that the charge was not pending on the amendments’ effective date. EEOC v. Christiansburg Garment Co., 376 F. Supp. 1067, 1073–1074 (WD Va. 1974). This Court was asked “what standard should inform a district court’s discretion in deciding whether to award attorney’s fees to a successful defendant in a Title VII action.” Christiansburg, 434 U. S., at 417 (emphasis deleted). If a ruling on the merits were necessary for the defendant to prevail and be eligible for attorney’s fees, the lack of a ruling on the merits would have been dispositive to this Court’s analysis. But the Court said nothing to suggest that the fact that the ruling was not on the merits ended the inquiry. Its reasoning was to the contrary. This Court noted with approval that the District Court had applied the correct standard and found that the “Commission’s statutory interpretation of §14 of the 1972 amendments was not frivolous.” Id., at 424 (internal quotation marks omitted). Various Courts of Appeals likewise have applied the Christiansburg standard when claims were dismissed for nonmerits reasons. A plaintiff’s claim may be frivolous, unreasonable, or groundless if the claim is barred by state sovereign immunity, C. W. v. Capistrano Unified School Dist., 784 F. 3d 1237, 1247–1248 (CA9 2015), or is moot, Propak Logistics, 746 F. 3d, at 152. See also Brief for Petitioner 33–34 (collecting Courts of Appeals cases in which the defendant received attorney’s fees and the District Court’s judgment was not on the merits). In cases like these, significant attorney time and expenditure may have gone into contesting the claim. Congress could not have intended to bar defendants from obtaining attorney’s fees in these cases on the basis that, although the litigation was resolved in their favor, they were nonetheless not prevailing parties. Neither the text of the fee-shifting statute nor the policy which underpins it counsels in favor of adopting the Court of Appeals’ on-the-merits requirement. B Having abandoned its defense of the Court of Appeals’ reasoning, the Commission now urges this Court to hold that a defendant must obtain a preclusive judgment in order to prevail. The Court declines to decide this issue, however. The Commission changed its argument between the certiorari and merits stages. As a result, the Commission may have forfeited the preclusion argument by not raising it earlier. The Commission’s failure to articulate its preclusion theory before the eleventh hour has resulted in inadequate briefing on the issue. The Commission and CRST dispute, moreover, whether the District Court’s judgment was in fact preclusive. Compare Brief for Respondent 38–45 with Reply Brief 8–13. The Court leaves these legal and factual issues for the Court of Appeals to consider in the first instance. The Commission submits the Court should affirm on the alternative ground that, even if CRST is a prevailing party, the Commission’s position that it had satisfied its presuit obligations was not frivolous, unreasonable, or groundless. The Commission acknowledges that the Court of Appeals has not decided this issue, but nevertheless invokes the Court’s authority to affirm “on any ground properly raised below.” Washington v. Confederated Bands and Tribes of Yakima Nation, 439 U. S. 463 , n. 20 (1979); see Brief for Respondent 4950. In light of this case’s intricate procedural history, see supra, at 3–11, this is not an appropriate case to reach and settle this fact-sensitive issue. It has been over 10 years since Starke first filed her charge and close to 9 years since the Commission filed its complaint. The dispute over the award of attorney’s fees has continued over much of that period and is still unresolved. When it appeared the litigation was coming to a close in the District Court, the trial judge considered this a case in which attorney’s fees should be assessed against the Commission. The Court of Appeals then made the rulings it considered proper in response, and there were further proceedings in the trial court and once again on appeal. Against this background of protracted and expensive litigation on the fee issue, the Court is aware of the need to resolve the outstanding issues without unnecessary delay. As the Court has noted in earlier cases, “the determination of fees ‘should not result in a second major litigation.’ ” Fox, 563 U. S., at 838 (quoting Hensley, 461 U. S., at 437). It is not prudent, however, for the Court to attempt to resolve all the pending issues under the circumstances here. It is not the Court’s usual practice to adjudicate either legal or predicate factual questions in the first instance. See Adarand Constructors, Inc. v. Mineta, 534 U. S. 103, 110 (2001) (per curiam) (noting “that this is a court of final review and not first view” (internal quotation marks omitted)). That precept is applicable here, especially in light of the extensive record in the case and the Commission’s change in its position. This Court is confident that the Court of Appeals, and, if necessary, the District Court, will resolve the case by taking any proper steps to expedite its resolution in a manner consistent with their own procedures and their responsibilities in other pending cases. The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

579.US.2015_15-446

The Leahy-Smith America Invents Act creates an agency procedure called “inter partes review” that allows a third party to ask the U. S. Patent and Trademark Office to reexamine the claims in an already-issued patent and to cancel any claim that the agency finds to be unpatentable in light of prior art. The Act, as relevant here, provides that the Patent Office’s decision “whether to institute an inter partes review . . . shall be final and non-appealable,” 35 U. S. C. §314(d), and grants the Patent Office authority to issue “regulations . . . establishing and governing inter partes review,” §316(a)(4). A Patent Office regulation issued pursuant to that authority provides that, during inter partes review, a patent claim “shall be given its broadest reasonable construction in light of the specification of the patent in which it appears.” 37 CFR §42.100(b). In 2012, Garmin International, Inc., and Garmin USA, Inc., sought inter partes review of all 20 claims of a patent held by petitioner Cuozzo Speed Technologies, LLC, asserting, among other things, that claim 17 was obvious in light of three prior patents. The Patent Office agreed to review claim 17. It also decided to reexamine claims 10 and 14 on that same ground because it determined those claims to be logically linked to the obviousness challenge to claim 17. The Patent Office, through its Patent Trial and Appeal Board, concluded that the claims were obvious in light of prior art, denied for reasons of futility Cuozzo’s motion to amend the claims, and canceled all three claims. Cuozzo appealed to the Federal Circuit. Cuozzo claimed that the Patent Office improperly instituted inter partes review with respect to claims 10 and 14, and it alleged that the Board improperly used the “broadest reasonable construction” standard to interpret the claims rather than the standard used by courts, which gives claims their “ordinary meaning . . . as understood by a person of skill in the art,” Phillips v. AWH Corp., 415 F. 3d 1303, 1314. The Federal Circuit rejected both arguments. It reasoned that §314(d) made the Patent Office’s decision to institute inter partes review “nonappealable,” and it concluded that the Patent Office’s regulation was a reasonable exercise of the agency’s rulemaking authority. Held: 1. Section 314(d) bars Cuozzo’s challenge to the Patent Office’s decision to institute inter partes review. Pp. 7–12. (a) The text of §314(d) expressly states that the Patent Office’s determinations whether to institute inter partes review “shall be final and nonappealable.” Moreover, construing §314(d) to permit judicial review of the Patent Office’s preliminary decision to institute inter partes review undercuts the important congressional objective of giving the agency significant power to revisit and revise earlier patent grants. Past practice in respect to related proceedings, including the predecessor to inter partes review, also supports the conclusion that Congress did not intend for courts to review these initial determinations. Finally, reading §314(d) as limited to interlocutory appeals would render the provision largely superfluous in light of the Administrative Procedure Act. Pp. 7–9. (b) The “strong presumption” favoring judicial review, Mach Mining, LLC v. EEOC, 575 U. S. ___, ___, is overcome here by these “ ‘clear and convincing’ ” indications that Congress intended to bar review, Block v. Community Nutrition Institute, 467 U. S. 340 . Given that presumption, however, the interpretation adopted here applies to cases in which the challenge is to the Patent Office’s determination “to initiate an inter partes review under this section,” or where the challenge consists of questions closely tied to the application and interpretation of statutes related to that determination. Cuozzo’s claim does not implicate a constitutional question, nor does it present other questions of interpretation that reach well beyond “this section” in terms of scope and impact. Rather, Cuozzo’s allegation that Garmin’s petition did not plead “with particularity” the challenge to claims 10 and 14 as required by §312 is little more than a challenge to the Patent Office’s conclusion under §314(a) that the “information presented in the petition” warranted review. Pp. 9–12. 2. The Patent Office regulation requiring the Board to apply the broadest reasonable construction standard to interpret patent claims is a reasonable exercise of the rulemaking authority granted to the Patent Office by statute. Pp. 12–20. (a) Where a statute leaves a gap or is ambiguous, this Court typically interprets a congressional grant of rulemaking authority as giving the agency leeway to enact rules that are reasonable in light of the text, nature, and purpose of the statute. United States v. Mead Corp., 533 U. S. 218 ; Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 –843. Here, the statute grants the Patent Office the authority to issue regulations “governing inter partes review,” and no statutory provision unambiguously mandates a particular claim construction standard. The Patent Office’s rulemaking authority is not limited to procedural regulations. Analogies to interpretations of other congressional grants of rulemaking authority in other statutes, which themselves do not unambiguously contain a limitation to procedural rules, cannot magically render unambiguous the different language in the different statutory grant of rulemaking authority at issue. The nature and purpose of inter partes review does not unambiguously require the Patent Office to apply one particular claim construction standard. Cuozzo’s contention that the purpose of inter partes review—to establish trial-like procedures for reviewing previously issued patents—supports the application of the ordinary meaning standard ignores the fact that in other significant respects, inter partes review is less like a judicial proceeding and more like a specialized agency proceeding. This indicates that Congress designed a hybrid proceeding. The purpose of inter partes review is not only to resolve patent-related disputes among parties, but also to protect the public’s “paramount interest in seeing that patent monopolies . . . are kept within their legitimate scope.” Precision Instrument Mfg. Co. v. Automotive Maintenance Machinery Co., 324 U. S. 806 . Neither the statute’s language, nor its purpose, nor its legislative history suggests that Congress decided what standard should apply in inter partes review. Pp. 12–17. (b) The regulation is a reasonable exercise of the Patent Office’s rulemaking authority. The broadest reasonable construction standard helps ensure precision in drafting claims and prevents a patent from tying up too much knowledge, which, in turn, helps members of the public draw useful information from the disclosed invention and understand the lawful limits of the claim. The Patent Office has used this standard for more than 100 years and has applied it in proceedings which, as here, resemble district court litigation. Cuozzo’s two arguments in response are unavailing. Applying the broadest reasonable construction standard in inter partes review is not, as Cuozzo suggests, unfair to a patent holder, who may move to amend at least once in the review process, and who has had several opportunities to amend in the original application process. And though the application of one standard in inter partes review and another in district court proceedings may produce inconsistent outcomes, that structure is inherent to Congress’ regulatory design, and it is also consistent with past practice, as the patent system has long provided different tracks for the review and adjudication of patent claims. The Patent Office’s regulation is reasonable, and this Court does not decide whether a better alternative exists as a matter of policy. Pp. 17–20. 793 F. 3d 1268, affirmed. Breyer, J., delivered the opinion for a unanimous Court with respect to Parts I and III, and the opinion of the Court with respect to Part II, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, and Kagan, JJ., joined. Thomas, J., filed a concurring opinion. Alito, J., filed an opinion concurring in part and dissenting in part, in which Sotomayor, J., joined.

The Leahy-Smith America Invents Act, 35 U. S. C. §100 et seq., creates a process called “inter partes review.” That review process allows a third party to ask the U. S. Patent and Trademark Office to reexamine the claims in analready-issued patent and to cancel any claim that the agency finds to be unpatentable in light of prior art. See §102 (requiring “novel[ty]”); §103 (disqualifying claims that are “obvious”). We consider two provisions of the Act. The first says: “No Appeal.—The determination by the Director [of the Patent Office] whether to institute an inter partes review under this section shall be final and non-appealable.” §314(d). Does this provision bar a court from considering whether the Patent Office wrongly “determin[ed] . . . to institute an inter partes review,” ibid., when it did so on grounds not specifically mentioned in a third party’s review request? The second provision grants the Patent Office the authority to issue “regulations . . . establishing and governing inter partes review under this chapter.” §316(a)(4). Does this provision authorize the Patent Office to issue a regulation stating that the agency, in inter partes review, “shall [construe a patent claim according to] its broadest reasonable construction in light of the specification of the patent in which it appears”? 37 CFR §42.100(b) (2015). We conclude that the first provision, though it may not bar consideration of a constitutional question, for example, does bar judicial review of the kind of mine-run claim at issue here, involving the Patent Office’s decision to institute inter partes review. We also conclude that the second provision authorizes the Patent Office to issue the regulation before us. See, e.g., United States v. Mead Corp., 533 U. S. 218, 229 (2001) ; Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837, 842 (1984) . I A An inventor obtains a patent by applying to the Patent Office. A patent examiner with expertise in the relevant field reviews an applicant’s patent claims, considers the prior art, and determines whether each claim meets the applicable patent law requirements. See, e.g., 35 U. S. C. §§101, 102, 103, 112. Then, the examiner accepts a claim, or rejects it and explains why. See §132(a). If the examiner rejects a claim, the applicant can resubmit a narrowed (or otherwise modified) claim, which the examiner will consider anew, measuring the new claim against the same patent law requirements. If the exam-iner rejects the new claim, the inventor typically has yet another chance to respond with yet another amended claim. Ultimately, the Patent Office makes a final decision allowing or rejecting the application. The applicant may seek judicial review of any final rejection. See §§141(a), 145. For several decades, the Patent Office has also possessed the authority to reexamine—and perhaps cancel—a patent claim that it had previously allowed. In 1980, for example, Congress enacted a statute providing for “ex parte reexamination.” Act to Amend the Patent and Trademark Laws, 35 U. S. C. §301 et seq. That statute (which remains in effect) gives “[a]ny person at any time” the right to “file a request for reexamination” on the basis of certain prior art “bearing on the patentability” of an already-issued patent. §§301(a)(1), 302. If the Patent Office concludes that the cited prior art raises “a substantial new question of patentability,” the agency can reexamine the patent. §303(a). And that reexamination can lead the Patent Office to cancel the patent (or some of its claims). Alternatively, the Director of the Patent Office can, on her “own initiative,” trigger such a proceeding. Ibid. And, as with examination, the patent holder can seek judicial review of an adverse final decision. §306. In 1999 and 2002, Congress enacted statutes that established another, similar procedure, known as “inter partes reexamination.” Those statutes granted third parties greater opportunities to participate in the Patent Office’s reexamination proceedings as well as in any appeal of a Patent Office decision. See, e.g., American Inventors Protection Act of 1999, §297 et seq. (2006 ed.) (superseded). In 2011, Congress enacted the statute before us. That statute modifies “inter partes reexamination,” which it now calls “inter partes review.” See H. R. Rep. No. 112–98, pt. 1, pp. 46–47 (2011) (H. R. Rep.). Like inter partes reexamination, any third party can ask the agency to initiate inter partes review of a patent claim. But the new statute has changed the standard that governs the Patent Office’s institution of the agency’s process. Instead of requiring that a request for reexamination raise a “substantial new question of patentability,” it now requires that a petition show “a reasonable likelihood that” the challenger “would prevail.” Compare §312(a) (2006 ed.) (repealed) with §314(a) (2012 ed.). The new statute provides a challenger with broader participation rights. It creates within the Patent Office a Patent Trial and Appeal Board (Board) composed of administrative patent judges, who are patent lawyers and former patent examiners, among others. §6. That Board conducts the proceedings, reaches a conclusion, and sets forth its reasons. See ibid. The statute sets forth time limits for completing this review. §316(a)(11). It grants the Patent Office the authority to issue rules. §316(a)(4). Like its predecessors, the statute authorizes judicial review of a “final written decision” canceling a patent claim. §319. And, the statute says that the agency’s initial decision “whether to institute an inter partes review” is “final and nonappealable.” §314(d); compare ibid. with §§312(a), (c) (2006 ed.)(repealed) (the “determination” that a petition forinter partes reexamination “raise[s]” “a substantial new question of patentability” is “final and non-appealable”), and §303(c) (2012 ed.) (similar in respect to ex partereexamination). B In 2002, Giuseppe A. Cuozzo applied for a patent covering a speedometer that will show a driver when he is driving above the speed limit. To understand the basic idea, think of the fact that a white speedometer needle will look red when it passes under a translucent piece of red glass or the equivalent (say, red cellophane). If you attach a piece of red glass or red cellophane to a speedometer beginning at 65 miles per hour, then, when the white needle passes that point, it will look red. If we attach the red glass to a plate that can itself rotate, if we attach the plate to the speedometer, if we connect the plate to a Global Positioning System (GPS) receiver, and if we enter onto a chip or a disk all the speed limits on all the Nation’s roads, then the GPS can signal where the car is, the chip or disk can signal the speed limit at that place, and the plate can rotate to the right number on the speedometer. Thus, if the speed limit is 35 miles per hour, then the white speedometer needle will pass under the red plate at 35, not 65, and the driver will know if he is driving too fast. In 2004, the Patent Office granted the patent. See U. S. Patent No. 6,778,074 (Cuozzo Patent). The Appendix contains excerpts from this patent, offering a less simplified (and more technical) description. C Petitioner Cuozzo Speed Technologies, LLC (Cuozzo), now holds the rights to the Cuozzo Patent. In 2012, Garmin International, Inc., and Garmin USA, Inc., filed a petition seeking inter partes review of the Cuozzo Patent’s 20 claims. Garmin backed up its request by stating, for example, that the invention described in claim 17 was obvious in light of three prior patents, the Aumayer, Evans, and Wendt patents. U. S. Patent No. 6,633,811; U. S. Patent No. 3,980,041; and U. S. Patent No. 2,711,153. Cf. Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U. S. 275, 280 (1944) (Black, J., dissenting) (“[S]omeone, somewhere, sometime, made th[is] discovery [but] I cannot agree that this patentee is that discoverer”). The Board agreed to reexamine claim 17, as well as claims 10 and 14. The Board recognized that Garmin had not expressly challenged claim 10 and claim 14 on the same obviousness ground. But, believing that “claim 17 depends on claim 14 which depends on claim 10,” the Board reasoned that Garmin had “implicitly” challenged claims 10 and 14 on the basis of the same prior inventions, and it consequently decided to review all three claims together. App. to Pet. for Cert. 188a. After proceedings before the Board, it concluded that claims 10, 14, and 17 of the Cuozzo Patent were obvious in light of the earlier patents to which Garmin had referred. The Board explained that the Aumayer patent “makes use of a GPS receiver to determine . . . the applicable speed limit at that location for display,” the Evans patent “describes a colored plate for indicating the speed limit,” and the Wendt patent “describes us[ing] a rotatable pointer for indicating the applicable speed limit.” Id., at 146a–147a. Anyone, the Board reasoned, who is “not an automaton”—anyone with “ordinary skill” and “ordinary creativity”—could have taken the automated approach suggested by the Aumayer patent and applied it to the manually adjustable signals described in the Evans and Wendt patents. Id., at 147a. The Board also concluded that Cuozzo’s proposed amendments would not cure this defect, id., at 164a–166a, and it consequently denied Cuozzo’s motion to amend its claims. Ultimately, it ordered claims 10, 14, and 17 of the Cuozzo Patent canceled, id., at 166a. Cuozzo appealed to the United States Court of Appeals for the Federal Circuit. Cuozzo argued that the Patent Office improperly instituted inter partes review, at least in respect to claims 10 and 14, because the agency found that Garmin had only implicitly challenged those two claims on the basis of the Aumayer, Evans, and Wendt patents, while the statute required petitions to set forth the grounds for challenge “with particularity.” §312(a)(3). Cuozzo also argued that the Board, when construing the claims, improperly used the interpretive standard set forth in the Patent Office’s regulation (i.e., it gave those claims their “broadest reasonable construction,” 37 CFR §42.100(b)), when it should have applied the standard that courts normally use when judging a patent’s validity (i.e., it should have given those claims their “ordinary meaning . . . as understood by a person of skill in the art,” Phillips v. AWH Corp., 415 F. 3d 1303, 1314 (CA Fed. 2005)(en banc)). A divided panel of the Court of Appeals rejected both arguments. First, the panel majority pointed out that 35 U. S. C. §314(d) made the decision to institute inter partes review “nonappealable.” In re Cuozzo Speed Technologies, LLC, 793 F. 3d 1268, 1273 (CA Fed. 2015) (internal quotation marks omitted). Second, the panel majority affirmed the application of the broadest reasonable construction standard on the ground (among others) that the regulation was a reasonable, and hence lawful, exercise of the Patent Office’s statutorily granted rulemaking authority. Id., at 1278–1279; see §314(a)(4). By a vote of 6 to 5, the Court of Appeals denied Cuozzo’s petition for rehearing en banc. In re Cuozzo Speed Technologies, LLC, 793 F. 3d 1297, 1298 (CA Fed. 2015). We granted Cuozzo’s petition for certiorari to review these two questions. II Like the Court of Appeals, we believe that Cuozzo’s contention that the Patent Office unlawfully initiated its agency review is not appealable. For one thing, that is what §314(d) says. It states that the “determination by the [Patent Office] whether to institute an inter partes review under this section shall be final and nonappealable.” (Emphasis added.) For another, the legal dispute at issue is an ordinary dispute about the application of certain relevant patent statutes concerning the Patent Office’s decision to institute inter partes review. Cuozzo points to a related statutory section, §312, which says that petitions must be pleaded “with particularity.” Those words, in its view, mean that the petition should have specifically said that claims 10 and 14 are also obvious in light of this same prior art. Garmin’s petition, the Government replies, need not have mentioned claims 10 and 14 separately, for claims 10, 14, and 17 are all logically linked; the claims “rise and fall together,” and a petition need not simply repeat the same argument expressly when it is so obviously implied. See 793 F. 3d, at 1281. In our view, the “No Appeal” provision’s language must, at the least, forbid an appeal that attacks a “determination . . . whether to institute” review by raising this kind of legal question and little more. §314(d). Moreover, a contrary holding would undercut one important congressional objective, namely, giving the Patent Office significant power to revisit and revise earlier patent grants. See H. R. Rep., at 45, 48 (explaining that the statute seeks to “improve patent quality and restore confidence in the presumption of validity that comes with issued patents”); 157 Cong. Rec. 9778 (2011) (remarks of Rep. Goodlatte) (noting that inter partes review “screen[s] out bad patents while bolstering valid ones”). We doubt that Congress would have granted the Patent Office this authority, including, for example, the ability to continue proceedings even after the original petitioner settles and drops out, §317(a), if it had thought that the agency’s final decision could be unwound under some minor statutory technicality related to its preliminary decision to institute inter partes review. Further, the existence of similar provisions in this, and related, patent statutes reinforces our conclusion. See §319 (limiting appellate review to the “final written decision”); §312(c) (2006 ed.) (repealed) (the “determination” that a petition for inter partes reexamination “raise[s]” a “substantial new question of patentability” is “final and non-appealable”); see also §303(c) (2012 ed.); In re Hiniker Co., 150 F. 3d 1362, 1367 (CA Fed. 1998) (“Section 303 . . . is directed toward the [Patent Office’s] authority to institute a reexamination, and there is no provision granting us direct review of that decision”). The dissent, like the panel dissent in the Court of Appeals, would limit the scope of the “No Appeal” provision to interlocutory appeals, leaving a court free to review the initial decision to institute review in the context of the agency’s final decision. Post, at 1, 5 (Alito, J., concurring in part and dissenting in part); 793 F. 3d, at 1291 (Newman, J., dissenting). We cannot accept this interpretation. It reads into the provision a limitation (to interlocutory decisions) that the language nowhere mentions and that is unnecessary. The Administrative Procedure Act already limits review to final agency decisions. 5 U. S. C. §704. The Patent Office’s decision to initiate inter partes review is “preliminary,” not “final.” Ibid. And the agency’s decision to deny a petition is a matter committed to the Patent Office’s discretion. See §701(a)(2); 35 U. S. C. §314(a) (no mandate to institute review); see also post, at 9, and n. 6. So, read as limited to such preliminary and discretionary decisions, the “No Appeal” provision would seem superfluous. The dissent also suggests that its approach is a “familiar practice,” consistent with other areas of law. Post, at 8. But the kind of initial determination at issue here—that there is a “reasonable likelihood” that the claims are unpatentable on the grounds asserted—is akin to decisions which, in other contexts, we have held to be unreviewable. See Kaley v. United States, 571 U. S. ___, ___ (2014) (slip op., at 8) (“The grand jury gets to say—without any review, oversight, or second-guessing—whether probable cause exists to think that a person committed a crime”). We recognize the “strong presumption” in favor of judicial review that we apply when we interpret statutes, including statutes that may limit or preclude review. Mach Mining, LLC v. EEOC, 575 U. S. ___, ___ (2015) (slip op., at 4) (internal quotation marks omitted). This presumption, however, may be overcome by “ ‘clear and convincing’ ” indications, drawn from “specific language,” “specific legislative history,” and “inferences of intent drawn from the statutory scheme as a whole,” that Congress intended to bar review. Block v. Community Nutrition Institute, 467 U. S. 340 –350 (1984). That standard is met here. The dissent disagrees, and it points to Lindahl v. Office of Personnel Management, 470 U. S. 768 (1985) , to support its view that, in light of this presumption, §314(d) should be read to permit judicial review of any issue bearing on the Patent Office’s preliminary decision to institute inter partes review. See post, at 4–5. Lindahl is a case about the judicial review of disability determinations for federal employees. We explained that a statute directing the Office of Personnel Management to “ ‘determine questions of disability,’ ” and making those decisions “ ‘final,’ ” “ ‘conclusive,’ ” and “ ‘not subject to review,’ ” barred a court from revisiting the “factual underpinnings of . . . disability determinations”—though it permitted courts to consider claims alleging, for example, that the Office of Personnel Management “ ‘substantial[ly] depart[ed] from important procedural rights.’ ” 470 U. S., at 771, 791. Thus, Lindahl’s interpretation of that statute preserved the agency’s primacy over its core statutory function in accord with Congress’ intent. Our interpretation of the “No Appeal” provision here has the same effect. Congress has told the Patent Office to determine whether inter partes review should proceed, and it has made the agency’s decision “final” and “nonappealable.” §314(d). Our conclusion that courts may not revisit this initial determination gives effect to this statutory command. Moreover, Lindahl’s conclusion was consistent with prior judicial practice in respect to those factual agency determinations, and legislative history “strongly suggest[ed]” that Congress intended to preserve this prior practice. Id., at 780. These features, as explained above, also support our interpretation: The text of the “No Appeal” provision, along with its place in the overall statutory scheme, its role alongside the Administrative Procedure Act, the prior interpretation of similar patent statutes, and Congress’ purpose in crafting inter partes review, all point in favor of precluding review of the Patent Office’s institution decisions. Nevertheless, in light of §314(d)’s own text and the presumption favoring review, we emphasize that our interpretation applies where the grounds for attacking the decision to institute inter partes review consist of questions that are closely tied to the application and interpretation of statutes related to the Patent Office’s decision to initiate inter partes review. See §314(d) (barring appeals of “determinations . . . to initiate an inter partes review under this section” (emphasis added)). This means that we need not, and do not, decide the precise effect of §314(d) on appeals that implicate constitutional questions, that depend on other less closely related statutes, or that present other questions of interpretation that reach, in terms of scope and impact, well beyond “this section.” Cf. Johnson v. Robison, 415 U. S. 361, 367 (1974) (statute precluding review of “any question of law or fact under any law administered by the Veterans’ Administration” does not bar review of constitutional challenges (emphasis deleted and internal quotation marks omitted)); Traynor v. Turnage, 485 U. S. 535 –545 (1988) (that same statute does not bar review of decisions made under different statutes enacted at other times). Thus, contrary to the dissent’s suggestion, we do not categorically preclude review of a final decision where a petition fails to give “sufficient notice” such that there is a due process problem with the entire proceeding, nor does our interpretation enable the agency to act outside its statutory limits by, for example, canceling a patent claim for “indefiniteness under §112” in inter partes review. Post, at 10–13. Such “shenanigans” may be properly reviewable in the context of §319 and under the Administrative Procedure Act, which enables reviewing courts to “set aside agency action” that is “contrary to constitutional right,” “in excess of statutory jurisdiction,” or “arbitrary [and] capricious.” Compare post, at 13, with 5 U. S. C. §§706(2)(A)–(D). By contrast, where a patent holder merely challenges the Patent Office’s “determin[ation] that the information presented in the petition . . . shows that there is a reasonable likelihood” of success “with respect to at least 1 of the claims challenged,” §314(a), or where a patent holder grounds its claim in a statute closely related to that decision to institute inter partes review, §314(d) bars judicial review. In this case, Cuozzo’s claim that Garmin’s petition was not pleaded “with particularity” under §312 is little more than a challenge to the Patent Office’s conclusion, under §314(a), that the “information presented in the petition” warranted review. Cf. United States v. Williams, 504 U. S. 36, 54 (1992) (“A complaint about the quality or adequacy of the evidence can always be recast as a complaint that the . . . presentation was ‘incomplete’ or ‘misleading’ ”). We therefore conclude that §314(d) bars Cuozzo’s efforts to attack the Patent Office’s determination to institute inter partes review in this case. III Cuozzo further argues that the Patent Office lacked the legal authority to issue its regulation requiring the agency, when conducting an inter partes review, to give apatent claim “its broadest reasonable construction in light of the specification of the patent in which it appears.” 37 CFR §42.100(b). Instead, Cuozzo contends that the Patent Office should, like the courts, give claims their “ordinary meaning . . . as understood by a person of skill in the art.” Phillips, 415 F. 3d, at 1314. The statute, however, contains a provision that grants the Patent Office authority to issue “regulations . . . establishing and governing inter partes review under this chapter.” 35 U. S. C. §316(a)(4). The Court of Appeals held that this statute gives the Patent Office the legal author-ity to issue its broadest reasonable construction regulation. We agree. A We interpret Congress’ grant of rulemaking authority in light of our decision in Chevron U. S. A. Inc., 467 U. S. 837 . Where a statute is clear, the agency must follow the statute. Id., at 842–843. But where a statute leaves a “gap” or is “ambigu[ous],” we typically interpret it as granting the agency leeway to enact rules that are reasonable in light of the text, nature, and purpose of the statute. Mead Corp., 533 U. S., at 229; Chevron U. S. A. Inc., supra, at 843. The statute contains such a gap: No statutory provision unambiguously directs the agency to use one standard or the other. And the statute “express[ly] . . . authoriz[es] [the Patent Office] to engage in the process of rulemaking” to address that gap. Mead Corp., supra, at 229. Indeed, the statute allows the Patent Office to issue rules “governing inter partes review,” §316(a)(4), and the broadest reasonable construction regulation is a rule that governs inter partes review. Both the dissenting judges in the Court of Appeals and Cuozzo believe that other ordinary tools of statutory interpretation, INS v. Cardoza-Fonseca, 480 U. S. 421 , and n. 12 (1987), lead to a different conclusion. The dissenters, for example, point to cases in which the Circuit interpreted a grant of rulemaking authority in a different statute, §2(b)(2)(A), as limited to procedural rules. See, e.g., Cooper Technologies Co. v. Dudas, 536 F. 3d 1330, 1335 (CA Fed. 2008). These cases, however, as we just said, interpret a different statute. That statute does not clearly contain the Circuit’s claimed limitation, nor is its language the same as that of §316(a)(4). Section 2(b)(2)(A) grants the Patent Office authority to issue “regulations” “which . . . shall govern . . . proceedings in the Office” (emphasis added), but the statute before us, §316(a)(4), does not refer to “proceedings”—it refers more broadly to regulations “establishing and governing inter partes review.” The Circuit’s prior interpretation of §2(b)(2)(A) cannot magically render unambiguous the different language in the different statute before us. Cuozzo and its supporting amici believe we will reach a different conclusion if we carefully examine the purpose of inter partes review. That purpose, in their view, is to modify the previous reexamination procedures and to replace them with a “ ‘trial, adjudicatory in nature.’ ” Brief for Petitioner 26 (quoting Google Inc. v. Jongerius Panoramic Techs., LLC, IPR 2013–00191, Paper No. 50, p. 4 (PTAB, Feb. 13, 2014))). They point out that, under the statute, an opposing party can trigger inter partes review. Parties can engage in “discovery of relevant evidence,” including “deposition[s], . . . affidavits or declarations” as well as anything “otherwise necessary in the interest of justice.” §316(a)(5). Parties may present “factual evidence and expert opinions” to support their arguments. §316(a)(8). The challenger bears the burden of proving unpatentability. §318(e). And, after oral argument before a panel of three of the Board’s administrative patent judges, it issues a final written decision. §§6, 316(a)(10), 318. Perhaps most importantly, a decision to cancel a patent normally has the same effect as a district court’s determination of a patent’s invalidity. In light of these adjudicatory characteristics, which make these agency proceedings similar to court proceedings, Congress, in Cuozzo’s view, must have designed inter partes review as a “surrogate for court proceedings.” Brief for Petitioner 28. Cuozzo points to various sources of legislative history in support of its argument. See H. R. Rep., at 48 (Inter partes review is a “quick and cost effective alternativ[e] to litigation”); id., at 46–47 (“The Act converts inter partes reexamination from an examinational to an adjudicative proceeding”); see also S. Rep. No.110–259, p. 20 (2008) (Inter partes review is “a quick, inexpensive, and reliable alternative to district court litigation”); 157 Cong. Rec. 3429–3430 (2011) (remarks of Sen. Kyl) (“Among the reforms that are expected to expedite these proceedings [is] the shift from an examinational to an adjudicative model”). And, if Congress intended to create a “surrogate” for court proceedings, why would Congress not also have intended the agency to use the claim construction standard that district courts apply (namely, the ordinary meaning standard), rather than the claim construction standard that patent examiners apply (namely, the broadest reasonable construction standard)? The problem with Cuozzo’s argument, however, is that, in other significant respects, inter partes review is less like a judicial proceeding and more like a specialized agency proceeding. Parties that initiate the proceeding need not have a concrete stake in the outcome; indeed, they may lack constitutional standing. See §311(a); cf. Consumer Watchdog v. Wisconsin Alumni Research Foundation, 753 F. 3d 1258, 1261–1262 (CA Fed. 2014). As explained above, challengers need not remain in the proceeding; rather, the Patent Office may continue to conduct an inter partes review even after the adverse party has settled. §317(a). Moreover, as is the case here, the Patent Office may intervene in a later judicial proceeding to defend its decision—even if the private challengers drop out. And the burden of proof in inter partes review is different than in the district courts: In inter partes review, the challenger (or the Patent Office) must establish unpatentability “by a preponderance of the evidence”; in district court, a challenger must prove invalidity by “clear and convincing evidence.” Compare §316(e) with Microsoft Corp. v. i4i Ltd. Partnership, 564 U. S. 91, 95 (2011) . Most importantly, these features, as well as inter partes review’s predecessors, indicate that the purpose of the proceeding is not quite the same as the purpose of district court litigation. The proceeding involves what used to be called a reexamination (and, as noted above, a cousin of inter partes review, ex parte reexamination, 35 U. S. C. §302 et seq., still bears that name). The name and accompanying procedures suggest that the proceeding offers a second look at an earlier administrative grant of a patent. Although Congress changed the name from “reexamination” to “review,” nothing convinces us that, in doing so, Congress wanted to change its basic purposes, namely, to reexamine an earlier agency decision. Thus, in addition to helping resolve concrete patent-related disputes among parties, inter partes review helps protect the public’s “paramount interest in seeing that patent monopolies . . . are kept within their legitimate scope.” Precision Instrument Mfg. Co. v. Automotive Maintenance Machinery Co., 324 U. S. 806, 816 (1945) ; see H. R. Rep., at 39–40 (Inter partes review is an “efficient system for challenging patents that should not have issued”). Finally, neither the statutory language, its purpose, or its history suggest that Congress considered what standard the agency should apply when reviewing a patent claim in inter partes review. Cuozzo contends that §301(d), explaining that the Patent Office should “determine the proper meaning of a patent claim,” reinforces its conclusion that the ordinary meaning standard should apply. But viewed against a background of language and practices indicating that Congress designed a hybrid proceeding, §301(d)’s reference to the “proper meaning” of a claim is ambiguous. It leaves open the question of which claim construction standard is “proper.” The upshot is, whether we look at statutory language alone, or that language in context of the statute’s purpose, we find an express delegation of rulemaking authority, a “gap” that rules might fill, and “ambiguity” in respect to the boundaries of that gap. Mead Corp., 533 U. S., at 229; see Chevron U. S. A. Inc., 467 U. S., at 843. We consequently turn to the question whether the Patent Office’s regulation is a reasonable exercise of its rulemakingauthority. B We conclude that the regulation represents a reasonable exercise of the rulemaking authority that Congress delegated to the Patent Office. For one thing, construing a patent claim according to its broadest reasonable construction helps to protect the public. A reasonable, yet unlawfully broad claim might discourage the use of the invention by a member of the public. Because an exam-iner’s (or reexaminer’s) use of the broadest reasonable construction standard increases the possibility that the examiner will find the claim too broad (and deny it), use of that standard encourages the applicant to draft narrowly. This helps ensure precision while avoiding overly broad claims, and thereby helps prevent a patent from tying up too much knowledge, while helping members of the public draw useful information from the disclosed invention and better understand the lawful limits of the claim. See §112(a); Nautilus, Inc. v. Biosig Instruments, Inc., 572 U. S. ___, ___ (2014) (slip op., at 10); see also In re Yamamoto, 740 F. 2d 1569, 1571 (CA Fed. 1984). For another, past practice supports the Patent Office’s regulation. See 77 Fed. Reg. 48697 (2012). The Patent Office has used this standard for more than 100 years. 793 F. 3d, at 1276. It has applied that standard in proceedings, which, as here, resemble district court litigation. See Bamberger v. Cheruvu, 55 USPQ 2d 1523, 1527 (BPAI 1998) (broadest reasonable construction standard applies in interference proceedings); Brief for Generic Pharmaceutical Association et al. as Amici Curiae 7–16 (describing similarities between interference proceedings and adjudicatory aspects of inter partes review); see also In re Yamamoto, supra, at 1571 (broadest reasonable construction standard applies in reexamination). It also applies that standard in proceedings that may be consolidated with a concurrent inter partes review. See 77 Fed. Reg. 48697–48698. Cuozzo makes two arguments in response. First, Cuozzo says that there is a critical difference between the Patent Office’s initial examination of an application to determine if a patent should issue, and this proceeding, in which the agency reviews an already-issued patent. In an initial examination of an application for a patent the examiner gives the claim its broadest reasonable construction. But if the patent examiner rejects the claim, then, as described above, Part I–A, supra, the applicant has a right to amend and resubmit the claim. And the examiner and applicant may repeat this process at least once more. This system—broad construction with a chance to amend—both protects the public from overly broad claims and gives the applicant a fair chance to draft a precise claim that will qualify for patent protection. In inter partes review, however, the broadest reasonable construction standard may help protect certain public interests, but there is no absolute right to amend any challenged patent claims. This, Cuozzo says, is unfair to the patent holder. The process however, is not as unfair as Cuozzo suggests. The patent holder may, at least once in the process, make a motion to do just what he would do in the examination process, namely, amend or narrow the claim. §316(d) (2012 ed.). This opportunity to amend, together with the fact that the original application process may have presented several additional opportunities to amend the patent, means that use of the broadest reasonable construction standard is, as a general matter, not unfair to the patent holder in any obvious way. Cuozzo adds that, as of June 30, 2015, only 5 out of 86 motions to amend have been granted. Brief for Petitioner 30; see Tr. of Oral Arg. 30 (noting that a sixth motion had been granted by the time of oral argument in this case). But these numbers may reflect the fact that no amendment could save the inventions at issue, i.e., that the patent should have never issued at all. To the extent Cuozzo’s statistical argument takes aim at the manner in which the Patent Office has exercised its authority, that question is not before us. Indeed, in this particular case, the agency determined that Cuozzo’s proposed amendment “enlarge[d],” rather than narrowed, the challenged claims. App. to Pet. for Cert. 165a–166a; see §316(d)(3). Cuozzo does not contend that the decision not to allow its amendment is “arbitrary” or “capricious,” or “otherwise [un]lawful.” 5 U. S. C. §706(2)(a). Second, Cuozzo says that the use of the broadest reasonable construction standard in inter partes review, together with use of an ordinary meaning standard in district court, may produce inconsistent results and cause added confusion. A district court may find a patent claim to be valid, and the agency may later cancel that claim in its own review. We recognize that that is so. This possibility, however, has long been present in our patent system, which provides different tracks—one in the Patent Office and one in the courts—for the review and adjudication of patent claims. As we have explained above, inter partes review imposes a different burden of proof on the challenger. These different evidentiary burdens mean that the possibility of inconsistent results is inherent to Congress’ regulatory design. Cf. One Lot Emerald Cut Stones v. United States, 409 U. S. 232 –238 (1972) ( per curiam). Moreover, the Patent Office uses the broadest reasonable construction standard in other proceedings, including interference proceedings (described above), which may implicate patents that are later reviewed in district court. The statute gives the Patent Office the power to consolidate these other proceedings with inter partes review. To try to create uniformity of standards would consequently prove difficult. And we cannot find unreasonable the Patent Office’s decision to prefer a degree of inconsistency in the standards used between the courts and the agency, rather than among agency proceedings. See 77 Fed. Reg. 48697–48698. Finally, Cuozzo and its supporting amici offer various policy arguments in favor of the ordinary meaning standard. The Patent Office is legally free to accept or reject such policy arguments on the basis of its own reasoned analysis. Having concluded that the Patent Office’s regulation, selecting the broadest reasonable construction standard, is reasonable in light of the rationales described above, we do not decide whether there is a better alternative as a policy matter. That is a question that Congress left to the particular expertise of the Patent Office. * * * For the reasons set forth above, we affirm the judgment of the Court of Appeals for the Federal Circuit. It is so ordered. APPENDIX SPEED LIMIT INDICATOR AND METHOD FOR DISPLAYING SPEED AND THE RELEVANT SPEED LIMIT Figure 1 * * * Figure 4 * * * DESCRIPTION OF THE CURRENT EMBODIMENT “In FIG. 1, a new and improved speed limit indicator and method for displaying speed and the relevant speed limit 10 . . . is illustrated . . . . More particularly, the speed limit indicator and method for displaying speed and the relevant speed limit 10 has a speedometer 12 mounted on a dashboard 26. [The] [s]peedometer 12 has a backplate 14 made of plastic, speed denoting markings 16 painted on [that] backplate 14, a colored display 18 made of a red plastic filter, and a plastic needle 20 rotably mounted in the center of [the] backplate 14. A [GPS] receiver 22 is positioned adjacent to the speedometer 12. Other gauges 24 typically present on a dashboard 26 are shown. . . . . . “[I]n FIG. 4, a new and improved speed limit indicator and method for displaying speed and the relevant speed limit 10 . . . is illustrated . . . . More particularly, the speed limit indicator and method for displaying speed and the relevant speed limit 10 has a backplate 14, colored display 18, housing 28, and axle 30. . . . . . “I claim: . . . . . “10. A speed limit indicator comprising: “a [GPS] receiver; “a display controller connected to said [GPS] receiver, wherein said display controller adjusts a colored display in response to signals from said [GPS] receiver to continu-ously update the delineation of which speed readings are in violation of the speed limit at a vehicle’s present location; and “a speedometer integrally attached to said colored display. . . . . . “14. The speed limit indicator as defined in claim 10, wherein said colored display is a colored filter. . . . . . “17. The speed limit indicator as defined in claim 14, wherein said display controller rotates said colored filter independently of said speedometer to continuously update the delineation of which speed readings are in violation of the speed limit at a vehicles present location.” Cuozzo Patent.

579.US.2015_15-458

Petitioner Rocky Dietz sued respondent Hillary Bouldin for negligence for injuries suffered in an automobile accident. Bouldin removed the case to Federal District Court. At trial, Bouldin admitted liability and stipulated to damages of $10,136 for Dietz’ medical expenses. The only disputed issue remaining was whether Dietz was entitled to more. During deliberations, the jury sent the judge a note asking whether Dietz’ medical expenses had been paid and, if so, by whom. Although the judge was concerned that the jury may not have understood that a verdict of less than the stipulated amount would require a mistrial, the judge, with the parties’ consent, responded only that the information being sought was not relevant to the verdict. The jury returned a verdict in Dietz’ favor but awarded him $0 in damages. After the verdict, the judge discharged the jury, and the jurors left the courtroom. Moments later, the judge realized the error in the $0 verdict and ordered the clerk to bring back the jurors, who were all in the building—including one who may have left for a short time and returned. Over the objection of Dietz’ counsel and in the interest of judicial economy and efficiency, the judge decided to recall the jury. After questioning the jurors as a group, the judge was satisfied that none had spoken about the case to anyone and ordered them to return the next morning. After receiving clarifying instructions, the reassembled jury returned a verdict awarding Dietz $15,000 in damages. On appeal, the Ninth Circuit affirmed. Held: A federal district court has a limited inherent power to rescind a jury discharge order and recall a jury in a civil case for further deliberations after identifying an error in the jury’s verdict. The District Court did not abuse that power here. Pp. 4–13. (a) The inherent powers that district courts possess “to manage their own affairs so as to achieve the orderly and expeditious disposition of cases,” Link v. Wabash R. Co., 370 U. S. 626 –631, have certain limits. The exercise of an inherent power must be a “reasonable response to the problems and needs” confronting the court’s fair administration of justice and cannot be contrary to any express grant of, or limitation on, the district court’s power contained in a rule or statute. Degen v. United States, 517 U. S. 820 –824. These two principles support the conclusion here. First, rescinding a discharge order and recalling the jury can be a reasonable response to correcting an error in the jury’s verdict in certain circumstances, and is similar in operation to a district court’s express power under Federal Rule of Civil Procedure 51(b)(3) to give the jury a curative instruction and order them to continue deliberating to correct an error in the verdict before discharge. Other inherent powers possessed by district courts, e.g., a district court’s inherent power to modify or rescind its orders before final judgment in a civil case, see Marconi Wireless Telegraph Co. of America v. United States, 320 U. S. 1 –48, or to manage its docket and courtroom with a view toward the efficient and expedient resolution of cases, see Landis v. North American Co., 299 U. S. 248 , also support this conclusion. Second, rescinding a discharge order to recall a jury does not violate any other rule or statute. No implicit limitation in Rule 51(b)(3) prohibits a court from rescinding its discharge order and reassembling the jury. Nor are such limits imposed by other rules dealing with postverdict remedies. See, e.g., Fed. Rules Civ. Proc. 50(b), 59(a)(1)(A). Pp. 4–7. (b) This inherent power must be carefully circumscribed, especially in light of the guarantee of an impartial jury. Because discharge releases a juror from the obligations to avoid discussing the case outside the jury room and to avoid external prejudicial information, the potential that a jury reassembled after being discharged might be tainted looms large. Thus, any suggestion of prejudice should counsel a district court not to exercise its inherent power. The court should determine whether any juror has been directly tainted and should also take into account additional factors that can indirectly create prejudice, which at a minimum, include the length of delay between discharge and recall, whether the jurors have spoken to anyone about the case after discharge, and any emotional reactions to the verdict witnessed by the jurors. Courts should also ask to what extent just-dismissed jurors accessed their smartphones or the internet. Applying those factors here, the District Court did not abuse its discretion. The jury was out for only a few minutes, and, with the exception of one juror, remained inside the courthouse. The jurors did not speak to any person about the case after discharge. And, there is no indication in the record that the verdict generated any kind of emotional reaction or electronic exchanges or searches that could have tainted the jury. Pp. 7–10. (c) Dietz’ call for a categorical bar on reempaneling a jury after discharge is rejected. Even assuming that at common law a discharged jury could never be brought back, the advent of modern federal trial practice limits the common law’s relevance as to the specific question raised here. There is no benefit to imposing a rule that says that as soon as a jury is free to go a judge categorically cannot rescind that order to correct an easily identified and fixable mistake. And Dietz’ “functional” discharge test, which turns on whether the jurors remain within the district court’s “presence and control,” i.e., within the courtroom, raises similar problems. Pp. 11–13. 794 F. 3d 1093; affirmed. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Alito, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Kennedy, J., joined.

In this case, a jury returned a legally impermissible verdict. The trial judge did not realize the error until shortly after he excused the jury. He brought the jury back and ordered them to deliberate again to correct the mistake. The question before us is whether a federal district court can recall a jury it has discharged, or whether the court can remedy the error only by ordering a new trial. This Court now holds that a federal district court has the inherent power to rescind a jury discharge order and recall a jury for further deliberations after identifying an error in the jury’s verdict. Because the potential of tainting jurors and the jury process after discharge is extraordinarily high, however, this power is limited in duration and scope, and must be exercised carefully to avoid any potential prejudice. I Petitioner Rocky Dietz was driving through an intersection in Bozeman, Montana, when Hillary Bouldin ran the red light and T-boned Dietz. As a result of the accident, Dietz suffered injuries to his lower back that caused him severe pain. He sought physical therapy, steroid injections, and other medications to treat his pain. Dietz sued Bouldin for negligence. Bouldin removed the case to Federal District Court. See 28 U. S. C. §§1332, 1441. At trial, Bouldin admitted that he was at fault for the accident and that Dietz was injured as a result. Bouldin also stipulated that Dietz’ medical expenses of $10,136 were reasonable and necessary as a result of the collision. The only disputed issue at trial for the jury to resolve was whether Dietz was entitled to damages above $10,136. During deliberations, the jury sent the judge a note asking: “ ‘Has the $10,136 medical expenses been paid; and if so, by whom?’ ” App. 36. The court discussed the note with the parties’ attorneys and told them he was unsure whether the jurors understood that their verdict could not be less than that stipulated amount, and that a mistrial would be required if the jury did not return a verdict of at least $10,136. The judge, however, with the consent of both parties, told the jury that the information they sought was not relevant to the verdict. The jury returned a verdict in Dietz’ favor but awarded him $0 in damages. The judge thanked the jury for its service and ordered them “discharged,” telling the jurors they were “free to go.” App. to Pet. for Cert. 25a. The jurors gathered their things and left the courtroom. A few minutes later, the court ordered the clerk to bring the jurors back. Speaking with counsel outside the jury’s presence, the court explained that it had “just stopped the jury from leaving the building,” after realizing that the $0 verdict was not “legally possible in view of stipulated damages exceeding $10,000.” Id., at 26a. The court suggested two alternatives: (1) order a new trial or (2) reempanel the jurors, instructing them to award at least the stipulated damages, and ordering them to deliberate anew. Dietz’ attorney objected to reempaneling the discharged jurors, arguing that the jury was no longer capable of returning a fair and impartial verdict. The court reiter-ated that none of the jurors had left the building, and asked the clerk whether any had even left the floor where the courtroom was located. The clerk explained that only one juror had left the building to get a hotel receipt and bring it back. Before the jurors returned, the judge told the parties that he planned to order the jury to deliberate again and reach a different verdict. The judge explained that he would “hate to just throw away the money and time that’s been expended in this trial.” Id., at 28a. When the jurors returned to the courtroom, the judge questioned them as a group and confirmed that they had not spoken to anyone about the case. The judge explained to the jurors the mistake in not awarding the stipulated damages. He informed the jurors that he was reempaneling them and would ask them to start over with clarifying instructions. He asked the jurors to confirm that they understood their duty and to return the next morning to deliberate anew. The next day, the reassembled jury returned a verdict awarding Dietz $15,000 in damages. On appeal, the Ninth Circuit affirmed. 794 F. 3d 1093 (2015). The court held that a district court could reempanel the jury shortly after dismissal as long as during the period of dismissal, the jurors were not exposed to any outside influences that would compromise their ability to reconsider the verdict fairly. This Court granted Dietz’ petition for a writ of certiorari to resolve confusion in the Courts of Appeals on whether and when a federal district court has the authority to recall a jury after discharging it. 577 U. S. ___ (2016). See Wagner v. Jones, 758 F. 3d 1030, 1034–1035 (CA8 2014), cert. denied, 575 U. S. ___ (2015); United States v. Figueroa, 683 F. 3d 69, 72–73 (CA3 2012); United States v. Rojas, 617 F. 3d 669, 677–678 (CA2 2010); United States v. Marinari, 32 F. 3d 1209, 1214 (CA7 1994); Summers v. United States, 11 F. 2d 583, 585–587 (CA4 1926). II A The Federal Rules of Civil Procedure set out many of the specific powers of a federal district court. But they are not all encompassing. They make no provision, for example, for the power of a judge to hear a motion in limine,[1] a motion to dismiss for forum non conveniens,[2] or many other standard procedural devices trial courts around the country use every day in service of Rule 1’s paramount command: the just, speedy, and inexpensive resolution of disputes. Accordingly, this Court has long recognized that a district court possesses inherent powers that are “governed not by rule or statute but by the control necessarily vested in courts to manage their own affairs so as to achieve the orderly and expeditious disposition of cases.” Link v. Wabash R. Co., 370 U. S. 626 –631 (1962); see also United States v. Hudson, 7 Cranch 32, 34 (1812). Al-though this Court has never precisely delineated the outer boundaries of a district court’s inherent powers, the Court has recognized certain limits on those powers. First, the exercise of an inherent power must be a “reasonable response to the problems and needs” confronting the court’s fair administration of justice. Degen v. United States, 517 U. S. 820 –824 (1996). Second, the exercise of an inherent power cannot be contrary to any express grant of or limitation on the district court’s power contained in a rule or statute. See id., at 823; Fed. Rule Civ. Proc. 83(b) (districts courts can “regulate [their] practice in any manner consistent with federal law”); see, e.g., Bank of Nova Scotia v. United States, 487 U. S. 250, 254 (1988) (holding that a district court cannot invoke its inherent power to circumvent the harmless-error inquiry prescribed by Federal Rule of Criminal Procedure 52(a)). These two principles—an inherent power must be a reasonable response to a specific problem and the power cannot contradict any express rule or statute—support the conclusion that a district judge has a limited inherent power to rescind a discharge order and recall a jury in a civil case where the court discovers an error in the jury’s verdict. First, rescinding a discharge order and recalling the jury can be a reasonable response to correcting an error in the jury’s verdict in certain circumstances. In the normal course, when a court recognizes an error in a verdict before it discharges the jury, it has the express power to give the jury a curative instruction and order them to continue deliberating. See Fed. Rule Civ. Proc. 51(b)(3) (“The court . . . may instruct the jury at any time before the jury is discharged”); 4 L. Sand et al., Modern Federal Jury Instructions–Civil ¶78.01, Instruction 78–10, p. 78–31 (2015) (Sand) (when a jury returns an inconsistent verdict, “[r]esubmitting the verdict . . . to resolve the inconsistencies is often the preferable course”). The decision to recall a jury to give them what would be an identical predischarge curative instruction could be, depending on the circumstances, similarly reasonable. This conclusion is buttressed by this Court’s prior cases affirming a district court’s inherent authority in analogous circumstances. For example, the Court has recognized that a district court ordinarily has the power to modify or rescind its orders at any point prior to final judgment in a civil case. Marconi Wireless Telegraph Co. of America v. United States, 320 U. S. 1 –48 (1943); see also Fed. Rule Civ. Proc. 54(b) (district court can revise partial final judgment order absent certification of finality); Fernandez v. United States, 81 S. Ct. 642, 644, 5 L. Ed. 2d 683 (1961) (Harlan, J., in chambers) (district court has inherent power to revoke order granting bail). Here, the District Court rescinded its order discharging the jury before it issued a final judgment. Rescinding the discharge order restores the legal status quo before the court dismissed the jury. The District Court is thus free to reinstruct the jury under Rule 51(b)(3). This Court has also held that district courts have the inherent authority to manage their dockets and courtrooms with a view toward the efficient and expedient resolution of cases. See, e.g., Landis v. North American Co., 299 U. S. 248, 254 (1936) (district court has inherent power to stay proceedings pending resolution of parallel actions in other courts); Link, 370 U. S., at 631–632 (district court has inherent power to dismiss case sua sponte for failure to prosecute); Chambers v. NASCO, Inc., 501 U. S. 32, 44 (1991) (district court has inherent power to vacate judgment procured by fraud); United States v. Morgan, 307 U. S. 183 –198 (1939) (district court has inherent power to stay disbursement of funds until revised payments are finally adjudicated). This Court’s recognition of these other inherent powers designed to resolve cases expeditiously is consistent with recognizing an inherent power to recall a discharged jury and reempanel the jurors with curative instructions. Compared to the alternative of conducting a new trial, recall can save the parties, the court, and society the costly time and litigation expense of conducting a new trial with a new set of jurors. Second, rescinding a discharge order to recall a jury does not violate any other rule or statute. Rule 51(b)(3) states that a court “may instruct the jury at any time before the jury is discharged.” A judge obviously cannot instruct a jury that is discharged—it is no longer there. But there is no implicit limitation in Rule 51(b)(3) that prohibits a court from rescinding its discharge order and reassembling the jury. See Link, 370 U. S., at 630 (holding that Rule 41(b)’s allowance for a party to move to dismiss for failure to prosecute did not implicitly abrogate the court’s power to dismiss sua sponte). Other rules dealing with postverdict remedies such as a motion for a new trial or a motion for judgment notwithstanding the verdict, see Fed. Rules Civ. Proc. 50(b), 59(a)(1)(A), similarly do not place limits on a court’s ability to rescind a prior order discharging a jury. Accordingly, a federal district court can rescind a discharge order and recall a jury in a civil case as an exercise of its inherent powers. B Just because a district court has the inherent power to rescind a discharge order does not mean that it is appropriate to use that power in every case. Because the exercise of an inherent power in the interest of promoting efficiency may risk undermining other vital interests related to the fair administration of justice, a district court’s inherent powers must be exercised with restraint. See Chambers, 501 U. S., at 44 (“Because of their very potency, inherent powers must be exercised with restraint and discretion”). The inherent power to rescind a discharge order and recall a dismissed jury, therefore, must be carefully circumscribed, especially in light of the guarantee of an impartial jury that is vital to the fair administration of justice. This Court’s precedents implementing this guarantee have noted various external influences that can taint a juror. E.g., Remmer v. United States, 347 U. S. 227, 229 (1954) (“In a criminal case, any private communication, contact, or tampering, directly or indirectly, with a juror during a trial about the matter pending before the jury is, for obvious reasons, deemed presumptively prejudicial”). Parties can accordingly ask that a juror be excused during trial for good cause, Fed. Rule Civ. Proc. 47(c), or challenge jury verdicts based on improper extraneous influences such as prejudicial information not admitted into evidence, comments from a court employee about the defendant, or bribes offered to a juror, Warger v. Shauers, 574 U. S. ___, ___ (2014) (slip op., at 10) (citing Tanner v. United States, 483 U. S. 107, 117 (1987) ); see also Mattox v. United States, 146 U. S. 140 –150 (1892) (external prejudicial information); Parker v. Gladden, 385 U. S. 363, 365 (1966) (per curiam) (bailiff comments on defendant); Remmer, 347 U. S., at 228–230 (bribe offered to juror). The potential for taint looms even larger when a jury is reassembled after being discharged. While discharged, jurors are freed from instructions from the court requiring them not to discuss the case with others outside the jury room and to avoid external prejudicial information. See, e.g., 4 Sand ¶71.02 (standard instruction to avoid extraneous influences); see also id., ¶71.01, Instructions 71–12 to 71–14 (avoid publicity). For example, it is not uncommon for attorneys or court staff to talk to jurors postdischarge for their feedback on the trial. See 1 K. O’Malley et al., Federal Jury Practice and Instructions §9:8 (6th ed. 2006) (debating appropriateness of practice). Any suggestion of prejudice in recalling a discharged jury should counsel a district court not to exercise its inherent power. A district court that is considering whether it should rescind a discharge order and recall a jury to correct an error or instead order a new trial should, of course, determine whether any juror has been directly tainted—for example, if a juror discusses the strength of the evidence with nonjurors or overhears others talking about the strength of the evidence. But the court should also take into account at least the following additional factors that can indirectly create prejudice in this context, any of which standing alone could be dispositive in a particular case. First, the length of delay between discharge and recall. The longer the jury has been discharged, the greater the likelihood of prejudice. Freed from the crucible of the jury’s group decisionmaking enterprise, discharged jurors may begin to forget key facts, arguments, or instructions from the court. In taking off their juror “hats” and returning to their lives, they may lose sight of the vital collective role they played in the impartial administration of justice. And they are more likely to be exposed to potentially prejudicial sources of information or discuss the case with others, even if they do not realize they have done so or forget when questioned after being recalled by the court. How long is too long is left to the discretion of the district court, but it could be as short as even a few minutes, depending on the case. Second, whether the jurors have spoken to anyone about the case after discharge. This could include court staff, attorneys and litigants, press and sketch artists, witnesses, spouses, friends, and so on. Even apparently innocu-ous comments about the case from someone like a courtroom deputy such as “job well done” may be sufficient to taint a discharged juror who might then resist reconsidering her decision. Third, the reaction to the verdict. Trials are society’s way of channeling disputes into fair and impartial resolutions. But these disputes can be bitter and emotional. And, depending on the case, those emotions may be broadcasted to the jury in response to their verdict. Shock, gasps, crying, cheers, and yelling are common reactions to a jury verdict—whether as a verdict is announced in the courtroom or seen in the corridors after discharge. In such a case, there is a high risk that emotional reactions will cause jurors to begin to reconsider their decision and ask themselves, “Did I make the right call?” Of course, this concern would be present even in a decision to reinstruct the jury to fix an error after the verdict is announced but before they are discharged. See Fed. Rule Civ. Proc. 51(b)(3). Even so, after discharging jurors from their obligations and the passage of time, a judge should be reluctant to reempanel a jury that has witnessed emotional reactions to its verdict. In considering these and any other relevant factors, courts should also ask to what extent just-dismissed jurors accessed their smartphones or the internet, which provide other avenues for potential prejudice. It is a now-ingrained instinct to check our phones whenever possible. Immediately after discharge, a juror could text something about the case to a spouse, research an aspect of the evidence on Google, or read reactions to a verdict on Twitter. Prejudice can come through a whisper or a byte. Finally, we caution that our recognition here of a court’s inherent power to recall a jury is limited to civil cases only. Given additional concerns in criminal cases, such as attachment of the double jeopardy bar, we do not address here whether it would be appropriate to recall a jury after discharge in a criminal case. See Smith v. Massachusetts, 543 U. S. 462 –474 (2005). Applying these factors, the District Court here did not abuse its discretion by rescinding its discharge order and recalling the jury to deliberate further. The jury was out for only a few minutes after discharge. Only one juror may have left the courthouse, apparently to retrieve a hotel receipt. The jurors did not speak to any person about the case after discharge. There is no indication in the record that this run-of-the-mill civil case—where the parties agreed that the defendant was liable and disputed damages only—generated any kind of emotional reaction or electronic exchanges or searches that could have tainted the jury. There was no apparent potential for prejudiceby recalling the jury here. III Dietz asks us to impose a categorical bar on reempaneling a jury after it has been discharged. He contends that, at common law, a jury once discharged could never be brought back together again. Accordingly, he argues, without a “ ‘long unquestioned’ power” of courts recalling juries, a federal district court lacks the inherent power to rescind a discharge order. See Carlisle v. United States, 517 U. S. 416 –427 (1996) (district court lacked inherent authority to grant untimely motion for judgment of acquittal). We disagree. Even assuming that the common-law tradition is as clear as Dietz contends, but see, e.g., Prussel v. Knowles, 5 Miss. 90, 95–97 (1839) (allowing postdischarge recall), the common law is less helpful to understanding modern civil trial practice. At common law, any error in the process of rendering a verdict, no matter how technical or inconsequential, could be remedied only by ordering a new trial. But modern trial practice did away with this system, replacing it with the harmless-error standard now embodied in Rule 61. See Kotteakos v. United States, 328 U. S. 750, 758, 760 (1946) (recognizing predecessor statute to Rule 61 codified the “salutary pol-icy” of “substitu[ing] judgment for automatic . . . rules”). Jury practice itself no longer follows the strictures of the common law. The common law required that juries be sequestered from the rest of society until they reached a verdict. Tellier, Separation or Dispersal of Jury in Civil Case After Submission, 77 A. L. R. 2d 1086 (1961). This generally meant no going home at night, no lunch breaks, no dispersing at all until they reached a verdict. Id., §2; see also Lester v. Stanley, 15 F. Cas. 396, 396–397 (No. 8,277) (Conn. 1808) (Livingston, Circuit Justice) (following common law). Courts are no longer required to impose these requirements on juries in order to prevent possible prejudice. See Nebraska Press Assn. v. Stuart, 427 U. S. 539, 554 (1976) (cases requiring sequestration to avoid trial publicity “are relatively rare”); Drake v. Clark, 14 F. 3d 351, 358 (CA7 1994) (“Sequestration is an extreme measure, one of the most burdensome tools of the many available to assure a fair trial”). Accordingly, while courts should not think they are generally free to discover new inherent powers that are contrary to civil practice as recognized in the common law, see Carlisle, 517 U. S., at 426–427, the advent of modern federal trial practice limits the common law’s relevance as to the specific question whether a judge can recall a just-discharged jury. Dietz also argues that the nature of a jury’s deliberative process means that something about the jury is irrevocably broken once the jurors are told they are free to go. According to Dietz, with their bond broken, the jurors cannot be brought back together again as a “jury.” In other words, once a jury is discharged, a court can never put the jury back together again by rescinding its discharge order—legally or metaphysically. We reject this “Humpty Dumpty” theory of the jury. Juries are of course an integral and special part of the American system of civil justice. Our system cannot function without the dedication of citizens coming together to perform their civic duty and resolve disputes. But there is nothing about the jury as an entity that ceases to exist simply because the judge tells the jury that they are excused from further service. A discharge order is not a magical invocation. It is an order, like any other order. And, like any order, it can be issued by mistake. All judges make mistakes. (Even us.) See Brown v. Allen, 344 U. S. 443, 540 (1953) (Jackson, J., concurring in judgment) (“We are not final because we are infallible, but we are infallible only because we are final”). There is no benefit to imposing a rule that says that as soon as a jury is free to go a judge categorically cannot rescind that order to correct an easily identified and fixable mistake, even as the jurors are still in the courtroom collecting their things. Dietz does not suggest the Court adopt a magic-words rule, but instead urges the adoption of a “functional” discharge test based on whether the jurors remain within the “presence and control” of the district court, where control is limited to the courtroom itself. Tr. of Oral Arg. 5–7. Similarly, the dissent suggests that it is the chance “to mingle with bystanders” that creates a discharge that cannot be undone. Post, at 1–2 (opinion of Thomas, J.) (internal quotation marks and brackets omitted). These tests do not avoid the problems that Dietz and the dissent identify with a prejudice inquiry. Under a courtroom test, what if a juror has one foot over the line? What if she just stepped out to use the restroom? Under a courthouse test, what if she is just outside the doors? Reached her car in the parking lot? Under a bystander test, is a courtroom deputy in the jury room a mingling bystander? There is no good reason to prefer a test based on geography or identity over an inquiry focused on potential prejudice. Finally, Dietz argues that the District Court in this case erred by questioning the discharged jurors as a group before reempaneling them instead of questioning each and every juror individually. While individual questioning could be the better practice in many circumstances, Dietz’ attorney raised no objection to this part of the court’s process. We decline to review this forfeited objection. See Fed. Rule Civ. Proc. 46. * * * Federal district courts have a limited inherent power to rescind a discharge order and recall a jury in a civil case. District courts should exercise this power cautiously and courts of appeals should review its invocation carefully. That was done here. The judgment of the Court of Appeals for the Ninth Circuit is therefore Affirmed.Notes 1 Luce v. United States, 469 U. S. 38 , n. 4 (1984). 2 Gulf Oil Corp. v. Gilbert, 330 U. S. 501 –508 (1947).

577.US.2015_14-462

Petitioner DIRECTV, Inc., and its customers entered into a service agreement that included a binding arbitration provision with a class-arbitration waiver. It specified that the entire arbitration provision was unenforceable if the “law of your state” made class-arbitration waivers unenforceable. The agreement also declared that the arbitration clause was governed by the Federal Arbitration Act. At the time that respondents, California residents, entered into that agreement with DIRECTV, California law made class-arbitration waivers unenforceable, see Discover Bank v. Superior Court, 36 Cal. 4th 148, 113 P. 3d 1100. This Court subsequently held in AT&T Mobility LLC v. Concepcion, 563 U. S. 333 , however, that California’s Discover Bank rule was pre-empted by the Federal Arbitration Act, 9 U. S. C. §2. When respondents sued petitioner, the trial court deniedDIRECTV’s request to order the matter to arbitration, and the California Court of Appeal affirmed. The court thought that California law would render class-arbitration waivers unenforceable, so it held the entire arbitration provision was unenforceable under the agreement. The fact that the Federal Arbitration Act pre-empted that California law did not change the result, the court said, because the parties were free to refer in the contract to California law as it would have been absent federal pre-emption. The court reasoned that the phrase “law of your state” was both a specific provision that should govern more general provisions and an ambiguous provision that should be construed against the drafter. Therefore, the court held, the parties had in fact included California law as it would have been without federal pre-emption. Held: Because the California Court of Appeal’s interpretation is pre-empted by the Federal Arbitration Act, that court must enforce the arbitration agreement. Pp. 5–11. (a) No one denies that lower courts must follow Concepcion, but that elementary point of law does not resolve the case because the parties are free to choose the law governing an arbitration provision, including California law as it would have been if not pre-empted. The state court interpreted the contract to mean that the parties did so, and the interpretation of a contract is ordinarily a matter of state law to which this Court defers, Volt Information Sciences, Inc. v. Board of Trustees of Leland Stanford Junior Univ., 489 U. S. 468 . The issue here is not whether the court’s decision is a correct statement of California law but whether it is consistent with the Federal Arbitration Act. Pp. 5–6. (b) The California court’s interpretation does not place arbitration contracts “on equal footing with all other contracts,” Buckeye Check Cashing, Inc. v. Cardegna, 546 U. S. 440 , because California courts would not interpret contracts other than arbitration contracts the same way. Several considerations lead to this conclusion. First, the phrase “law of your state” is not ambiguous and takes its ordinary meaning: valid state law. Second, California case law—that under “general contract principles,” references to California law incorporate the California Legislature’s power to change the law retroactively, Doe v. Harris, 57 Cal. 4th 64, 69–70, 302 P. 3d 598, 601–602—clarifies any doubt about how to interpret it. Third, because the court nowhere suggests that California courts would reach the same interpretation in any other context, its conclusion appears to reflect the subject matter, rather than a general principle that would include state statutes invalidated by other federal law. Fourth, the language the court uses to frame the issue focuses only on arbitration. Fifth, the view that state law retains independent force after being authoritatively invalidated is one courts are unlikely to apply in other contexts. Sixth, none of the principles of contract interpretation relied on by the California court suggests that other California courts would reach the same interpretation elsewhere. The court applied the canon that contracts are construed against the drafter, but the lack of any similar case interpreting similar language to include invalid laws indicates that the antidrafter canon would not lead California courts to reach a similar conclusion in cases not involving arbitration. Pp. 6–10. 225 Cal. App. 4th 338, 170 Cal. Rptr. 3d 190, reversed and remanded. Breyer, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Alito, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion. Ginsburg, J., filed a dissenting opinion, in which Sotomayor, J., joined.

The Federal Arbitration Act states that a “written provision” in a contract providing for “settle[ment] by arbitration” of “a controversy . . . arising out of” that “contract . . . shall be valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. §2. We here consider a California court’s refusal to enforce an arbitration provision in a contract. In our view, that decision does not rest “upon such grounds as exist . . . for the revocation of any contract,” and we consequently set that judgment aside. I DIRECTV, Inc., the petitioner, entered into a service agreement with its customers, including respondents Amy Imburgia and Kathy Greiner. Section 9 of that contract provides that “any Claim either of us asserts will be resolved only by binding arbitration.” App. 128. It then sets forth a waiver of class arbitration, stating that “[n]either you nor we shall be entitled to join or consolidate claims in arbitration.” Id., at 128–129. It adds that if the “law of your state” makes the waiver of class arbitration unenforceable, then the entire arbitration provision “is unenforceable.” Id., at 129. Section 10 of the contract states that §9, the arbitration provision, “shall be governed by the Federal Arbitration Act.” Ibid. In 2008, the two respondents brought this lawsuit against DIRECTV in a California state court. They seek damages for early termination fees that they believe violate California law. After various proceedings not here relevant, DIRECTV, pointing to the arbitration provision, asked the court to send the matter to arbitration. The state trial court denied that request, and DIRECTVappealed. The California Court of Appeal thought that the critical legal question concerned the meaning of the contractual phrase “law of your state,” in this case the law of California. Does the law of California make the contract’s class-arbitration waiver unenforceable? If so, as the contract provides, the entire arbitration provision is unenforceable. Or does California law permit the parties to agree to waive the right to proceed as a class in arbitration? If so, the arbitration provision is enforceable. At one point, the law of California would have made the contract’s class-arbitration waiver unenforceable. In 2005, the California Supreme Court held in Discover Bank v. Superior Court, 36 Cal. 4th 148, 162–163, 113 P. 3d 1100, 1110, that a “waiver” of class arbitration in a “consumer contract of adhesion” that “predictably involve[s] small amounts of damages” and meets certain other criteria not contested here is “unconscionable under California law and should not be enforced.” See Cohen v. DirecTV, Inc., 142 Cal. App. 4th 1442, 1446–1447, 48 Cal. Rptr. 3d 813, 815–816 (2006) (holding a class-action waiver similar to the one at issue here unenforceable pursuant to Discover Bank); see also Consumers Legal Remedies Act, Cal. Civ. Code Ann. §§1751, 1781(a) (West 2009) (invalidating class-action waivers for claims brought under that statute). But in 2011, this Court held that California’s Discover Bank rule “ ‘stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress’ ” embodied in the Federal Arbitration Act. AT&T Mobility LLC v. Concepcion, 563 U. S. 333, 352 (2011) (quoting Hines v. Davidowitz, 312 U. S. 52, 67 (1941) ); see Sanchez v. Valencia Holding Co., LLC, 61 Cal. 4th 899, 923–924, 353 P. 3d 741, 757 (2015) (holding that Concepcion applies to the Consumers Legal Remedies Act to the extent that it would have the same effect as Discover Bank). The Fed-eral Arbitration Act therefore pre-empts and invalidates that rule. 563 U. S., at 352; see U. S. Const., Art. VI, cl. 2. The California Court of Appeal subsequently held in this case that, despite this Court’s holding in Concepcion, “the law of California would find the class action waiver unenforceable.” 225 Cal. App. 4th 338, 342, 170 Cal. Rptr. 3d 190, 194 (2014). The court noted that Discover Bank had held agreements to dispense with class-arbitration procedures unenforceable under circumstances such as these. 225 Cal. App. 4th, at 341, 170 Cal. Rptr. 3d, at 194. It conceded that this Court in Concepcion had held that the Federal Arbitration Act invalidated California’s rule. 225 Cal. App. 4th, at 341, 170 Cal. Rptr. 3d, at 194. But it then concluded that this latter circumstance did not change the result—that the “class action waiver is unenforceable under California law.” Id., at 347, 170 Cal. Rptr. 3d, at 198. In reaching that conclusion, the Court of Appeal referred to two sections of California’s Consumers Legal Remedies Act, §§1751, 1781(a), rather than Discover Bank itself. See 225 Cal. App. 4th, at 344, 170 Cal. Rptr. 3d, at 195. Section 1751 renders invalid any waiver of the right under §1781(a) to bring a class action for violations of that Act. The Court of Appeal thought that applying “state law alone” (that is, those two sections) would render unenforceable the class-arbitration waiver in §9 of the contract. Id., at 344, 170 Cal. Rptr. 3d, at 195. But it nonetheless recognized that if it applied federal law “then the class action waiver is enforceable and any state law to the contrary is preempted.” Ibid. As far as those sections apply to class-arbitration waivers, they embody the Discover Bank rule. The California Supreme Court has recognized as much, see Sanchez, supra, at 923–924, 353 P. 3d, at 757, and no party argues to the contrary. See Supp. Brief for Respondents 2 (“The ruling in Sanchez tracks respondents’ position precisely”). We shall consequently refer to the here-relevant rule as the Discover Bank rule. The court reasoned that just as the parties were free in their contract to refer to the laws of different States or different nations, so too were they free to refer to California law as it would have been without this Court’s holding invalidating the Discover Bank rule. The court thought that the parties in their contract had done just that. And it set forth two reasons for believing so. First, §10 of the contract, stating that the Federal Arbitration Act governs §9 (the arbitration provision), is a general provision. But the provision voiding arbitration if the “law of your state” would find the class-arbitration waiver unenforceable is a specific provision. The court believed that the specific provision “ ‘is paramount to’ ” and must govern the general. 225 Cal. App. 4th, at 344, 170 Cal. Rptr. 3d, at 195 (quoting Prouty v. Gores Technology Group, 121 Cal. App. 4th 1225, 1235, 18 Cal. Rptr. 3d 178, 185–186 (2004); brackets omitted). Second, the court said that “ ‘a court should construe ambiguous language against the interest of the party that drafted it.’ ” 255 Cal. App. 4th, at 345, 170 Cal. Rptr. 3d, at 196 (quoting Mastrobuono v. Shearson Lehman Hutton, Inc., 514 U. S. 52, 62 (1995) ). DIRECTV had drafted the language; to void the arbitration provision was against its interest. Hence the arbitration provision was void. The Court of Appeal consequently affirmed the trial court’s denial of DIRECTV’s motion to enforce the arbitration provision. The California Supreme Court denied discretionary review. App. to Pet. for Cert. 1a. DIRECTV then filed a petition for a writ of certiorari, noting that the Ninth Circuit had reached the opposite conclusion on precisely the same interpretive question decided by the California Court of Appeal. Murphy v. DirecTV, Inc., 724 F. 3d 1218, 1226–1228 (2013). We granted the petition. II No one denies that lower courts must follow this Court’s holding in Concepcion. The fact that Concepcion was a closely divided case, resulting in a decision from which four Justices dissented, has no bearing on that undisputed obligation. Lower court judges are certainly free to note their disagreement with a decision of this Court. But the “Supremacy Clause forbids state courts to dissociate themselves from federal law because of disagreement with its content or a refusal to recognize the superior authority of its source.” Howlett v. Rose, 496 U. S. 356, 371 (1990) ; cf. Khan v. State Oil Co., 93 F. 3d 1358, 1363–1364 (CA7 1996), vacated, 522 U. S. 3 (1997) . The Federal Arbitration Act is a law of the United States, and Concepcion is an authoritative interpretation of that Act. Consequently, the judges of every State must follow it. U. S. Const., Art. VI, cl. 2 (“[T]he Judges in every State shall be bound” by “the Laws of the United States”). While all accept this elementary point of law, that point does not resolve the issue in this case. As the Court of Appeal noted, the Federal Arbitration Act allows parties to an arbitration contract considerable latitude to choose what law governs some or all of its provisions, including the law governing enforceability of a class-arbitration waiver. 225 Cal. App. 4th, at 342–343, 170 Cal. Rptr. 3d, at 194. In principle, they might choose to have portions of their contract governed by the law of Tibet, the law of pre-revolutionary Russia, or (as is relevant here) the law of California including the Discover Bank rule and irrespective of that rule’s invalidation in Concepcion. The Court of Appeal decided that, as a matter of contract law, the parties did mean the phrase “law of your state” to refer to this last possibility. Since the interpretation of a contract is ordinarily a matter of state law to which we defer, Volt Information Sciences, Inc. v. Board of Trustees of Leland Stanford Junior Univ., 489 U. S. 468, 474 (1989) , we must decide not whether its decision is a correct statement of California law but whether (assuming it is) that state law is consistent with the Federal Arbitration Act. III Although we may doubt that the Court of Appeal has correctly interpreted California law, we recognize that California courts are the ultimate authority on that law. While recognizing this, we must decide whether the decision of the California court places arbitration contracts “on equal footing with all other contracts.” Buckeye Check Cashing, Inc. v. Cardegna, 546 U. S. 440, 443 (2006) . And in doing so, we must examine whether the Court of Appeal’s decision in fact rests upon “grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. §2. That is to say, we look not to grounds that the California court might have offered but rather to those it did in fact offer. Neither this approach nor our result “steps beyond Concepcion” or any other aspect of federal arbitration law. See post, at 9 (Ginsburg, J., dissenting) (hereinafter the dissent). We recognize, as the dissent points out, post, at 4, that when DIRECTV drafted the contract, the parties likely believed that the words “law of your state” included California law that then made class-arbitration waivers unenforceable. But that does not answer the legal question before us. That is because this Court subsequently held in Concepcion that the Discover Bank rule was invalid. Thus the underlying question of contract law at the time the Court of Appeal made its decision was whether the “law of your state” included invalid California law. We must now decide whether answering that question in the affirmative is consistent with the Federal Arbitration Act. After examining the grounds upon which the Court of Appeal rested its decision, we conclude that California courts would not interpret contracts other than arbitration contracts the same way. Rather, several considerations lead us to conclude that the court’s interpretation of this arbitration contract is unique, restricted to that field. First, we do not believe that the relevant contract language is ambiguous. The contract says that “[i]f . . . the law of your state would find this agreement to dispense with class arbitration procedures unenforceable, then this entire Section 9 [the arbitration section] is unenforceable.” App. 129. Absent any indication in the contract that this language is meant to refer to invalid state law, it presumably takes its ordinary meaning: valid state law. Indeed, neither the parties nor the dissent refer us to any contract case from California or from any other State that in-terprets similar language to refer to state laws authoritatively held to be invalid. While we recognize that the dissent believes this phrase to be “ambiguous,” post, at 7, 9, or “anomalous,” post, at 10, we cannot agree with that characterization. Second, California case law itself clarifies any doubt about how to interpret the language. The California Supreme Court has held that under “general contract principles,” references to California law incorporate the California Legislature’s power to change the law retroactively. See Doe v. Harris, 57 Cal. 4th 64, 69–70, 302 P. 3d 598, 601–602 (2013) (holding that plea agreements, which are governed by general contract principles, are “ ‘ “deemed to incorporate and contemplate not only the existing law but the reserve power of the state to amend the law or enact additional laws” ’ ” (quoting People v. Gipson, 117 Cal. App. 4th 1065, 1070, 12 Cal. Rptr. 3d 478, 481 (2004))). And judicial construction of a statute ordinarily applies retroactively. Rivers v. Roadway Express, Inc., 511 U. S. 298 –313 (1994). As far as we are aware, the principle of California law announced in Harris, not the Court of Appeal’s decision here, would ordinarily govern the scope of phrases such as “law of your state.” Third, nothing in the Court of Appeal’s reasoning suggests that a California court would reach the same interpretation of “law of your state” in any context other than arbitration. The Court of Appeal did not explain why parties might generally intend the words “law of your state” to encompass “invalid law of your state.” To the contrary, the contract refers to “state law” that makes the waiver of class arbitration “unenforceable,” while an in-valid state law would not make a contractual provision unenforceable. Assuming—as we must—that the court’s reasoning is a correct statement as to the meaning of “law of your state” in this arbitration provision, we can find nothing in that opinion (nor in any other California case) suggesting that California would generally interpret words such as “law of your state” to include state laws held invalid because they conflict with, say, federal labor statutes, federal pension statutes, federal antidiscrimination laws, the Equal Protection Clause, or the like. Even given our assumption that the Court of Appeal’s conclusion is correct, its conclusion appears to reflect the subject matter at issue here (arbitration), rather than a general principle that would apply to contracts using similar language but involving state statutes invalidated by other federal law. Fourth, the language used by the Court of Appeal focused only on arbitration. The court asked whether “law of your state” “mean[s] ‘the law of your state to the extent it is not preempted by the [Federal Arbitration Act],’ or ‘the law of your state without considering the preemptive effect, if any of the [Federal Arbitration Act].’ ” 225 Cal. App. 4th, at 344, 170 Cal. Rptr. 3d, at 195. Framing the question in such terms, rather than in generally applicable terms, suggests that the Court of Appeal could well have meant that its holding was limited to the specific subject matter of this contract—arbitration. Fifth, the Court of Appeal reasoned that invalid state arbitration law, namely the Discover Bank rule, maintained legal force despite this Court’s holding in Concepcion. The court stated that “[i]f we apply state law alone . . . to the class action waiver, then the waiver is unenforceable.” 225 Cal. App. 4th, at 344, 170 Cal. Rptr. 3d, at 195. And at the end of its opinion it reiterated that “[t]he class action waiver is unenforceable under California law, so the entire arbitration agreement is unenforceable.” Id., at 347, 170 Cal. Rptr. 3d, at 198. But those statements do not describe California law. See Concepcion, 563 U. S., at 344, 352; Sanchez, 61 Cal. 4th, at 923–924, 353 P. 3d, at 757. The view that state law retains independent force even after it has been authoritatively invalidated by this Court is one courts are unlikely to accept as a general matter and to apply in other contexts. Sixth, there is no other principle invoked by the Court of Appeal that suggests that California courts would reach the same interpretation of the words “law of your state”in other contexts. The court said that the phrase “lawof your state” constitutes “ ‘a specific exception’ ” to the agreement’s “ ‘general adoption of the [Federal Arbitration Act].’ ” 225 Cal. App. 4th, at 344, 170 Cal. Rptr. 3d, at 195. But that tells us nothing about how to interpret the words “law of your state” elsewhere. It does not answer the relevant question: whether those words encompass laws that have been authoritatively held invalid. Cf. Prouty, 121 Cal. App. 4th, at 1235, 18 Cal. Rptr. 3d, at 185–186 (specific words govern only “when a general and a particular provision are inconsistent”). The court added that it would interpret “ ‘ambiguous language against the interest of the party that drafted it,’ ” namely DIRECTV. 225 Cal. App. 4th, at 345, 170 Cal. Rptr. 3d, at 196 (quoting Mastrobuono, 514 U. S., at 62). The dissent adopts a similar argument. See post, at 7–9. But, as we have pointed out, supra, at 8, were the phrase “law of your state” ambiguous, surely some court would have construed that term to incorporate state laws invalidated by, for example, federal labor law, federal pension law, or federal civil rights law. Yet, we have found no such case. Moreover, the reach of the canon construing contract language against the drafter must have limits, no matter who the drafter was. The fact that we can find no similar case interpreting the words “law of your state” to include invalid state laws indicates, at the least, that the antidrafter canon would not lead California courts to reach a similar conclusion in similar cases that do not involve arbitration. * * * Taking these considerations together, we reach a conclusion that, in our view, falls well within the confines of (and goes no further than) present well-established law. California’s interpretation of the phrase “law of your state” does not place arbitration contracts “on equal footing with all other contracts,” Buckeye Check Cashing, Inc., 546 U. S., at 443. For that reason, it does not give “due regard . . . to the federal policy favoring arbitration.” Volt Information Sciences, 489 U. S., at 476. Thus, the Court of Appeal’s interpretation is pre-empted by the Federal Arbitration Act. See Perry v. Thomas, 482 U. S. 483 , n. 9 (1987) (noting that the Federal Arbitration Act pre-empts decisions that take their “meaning precisely from the fact that a contract to arbitrate is at issue”). Hence, the California Court of Appeal must “enforc[e]” the arbitration agreement. 9 U. S. C. §2. The judgment of the California Court of Appeal is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered.

579.US.2015_15-415

The Fair Labor Standards Act (FLSA) requires employers to pay overtime compensation to covered employees who work more than 40 hours in a given week. In 1966, Congress enacted an exemption from the overtime compensation requirement for “any salesman, partsman, or mechanic primarily engaged in selling or servicing automobiles” at a covered dealership. Fair Labor Standards Amendments of 1966, §209, 80Stat. 836, codified as amended at 29 U. S. C. §213(b)(10)(A). Congress authorized the Department of Labor to promulgate necessary rules, regulations, or orders with respect to this new provision. The Department exercised that authority in 1970 and issued a regulation that defined “salesman” to mean “an employee who is employed for the purpose of and is primarily engaged in making sales or obtaining orders or contracts for sale of the vehicles . . . which the establishment is primarily engaged in selling.” 29 CFR §779.372(c)(1) (1971). The regulation excluded service advisors, who sell repair and maintenance services but not vehicles, from the exemption. Several courts, however, rejected the Department’s conclusion that service advisors are not covered by the statutory exemption. In 1978, the Department issued an opinion letter departing from its previous position and stating that service advisors could be exempt under 29 U. S. C. §213(b)(10)(A). In 1987, the Department confirmed its new interpretation by amending its Field Operations Handbook to clarify that service advisors should be treated as exempt under the statute. In 2011, however, the Department issued a final rule that followed the original 1970 regulation and interpreted the statutory term “salesman” to mean only an employee who sells vehicles. 76 Fed. Reg. 18859. The Department gave little explanation for its decision to abandon its decades-old practice of treating service advisors as exempt under §213(b)(10)(A). Petitioner is an automobile dealership. Respondents are or were employed by petitioner as service advisors. Respondents filed suit alleging that petitioner violated the FLSA by failing to pay them overtime compensation when they worked more than 40 hours in a week. Petitioner moved to dismiss, arguing that the FLSA overtime provisions do not apply to respondents because service advisors are covered by the §213(b)(10)(A) exemption. The District Court granted the motion, but the Ninth Circuit reversed in relevant part. Deferring under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 , to the interpretation set forth in the 2011 regulation, the court held that service advisors are not covered by the §213(b)(10)(A) exemption. Held: Section 213(b)(10)(A) must be construed without placing controlling weight on the Department’s 2011 regulation. Pp. 7–12. (a) When an agency is authorized by Congress to issue regulations and promulgates a regulation interpreting a statute it enforces, the interpretation receives deference if the statute is ambiguous and the agency’s interpretation is reasonable. See Chevron, supra, at 842–844. When Congress authorizes an agency to proceed through notice-and-comment rulemaking, that procedure is a “very good indicator” that Congress intended the regulation to carry the force of law, so Chevron should apply. United States v. Mead Corp., 533 U. S. 218 –230. But Chevron deference is not warranted where the regulation is “procedurally defective”—that is, where the agency errs by failing to follow the correct procedures in issuing the regulation. 533 U. S., at 227. One basic procedural requirement of administrative rulemaking is that an agency must give adequate reasons for its decisions. Where the agency has failed to provide even a minimal level of analysis, its action is arbitrary and capricious and so cannot carry the force of law. Agencies are free to change their existing policies, but in explaining its changed position, an agency must be cognizant that longstanding policies may have “engendered serious reliance interests that must be taken into account.” FCC v. Fox Television Stations, Inc., 556 U. S. 502 . An “[u]nexplained inconsistency” in agency policy is “a reason for holding an interpretation to be an arbitrary and capricious change from agency practice,” National Cable & Telecommunications Assn. v. Brand X Internet Services, 545 U. S. 967 , and an arbitrary and capricious regulation of this sort receives no Chevron deference. Pp. 7–10. (b) Applying those principles, the 2011 regulation was issued without the reasoned explanation that was required in light of the Department’s change in position and the significant reliance interests involved. The industry had relied since 1978 on the Department’s position that service advisors are exempt from the FLSA’s overtime pay requirements, and had negotiated and structured compensation plans against this background understanding. In light of this background, the Department needed a more reasoned explanation for its decision to depart from its existing enforcement policy. The Department instead said almost nothing. It did not analyze or explain why the statute should be interpreted to exempt dealership employees who sell vehicles but not dealership employees who sell services. This lack of reasoned explication for a regulation that is inconsistent with the Department’s longstanding earlier position results in a rule that cannot carry the force of law, and so the regulation does not receive Chevron deference. It is appropriate to remand for the Ninth Circuit to interpret §213(b)(10)(A) in the first instance. Pp. 10–12. 780 F. 3d 1267, vacated and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Ginsburg, J., filed a concurring opinion, in which Sotomayor, J., joined. Thomas, J., filed a dissenting opinion, in which Alito, J., joined.

This case addresses whether a federal statute requires payment of increased compensation to certain automobile dealership employees for overtime work. The federal statute in question is the Fair Labor Standards Act (FLSA), 29 U. S. C. §201 et seq., enacted in 1938 to “protect all covered workers from substandard wages and oppressive working hours.” Barrentine v. Arkansas-Best Freight System, Inc., 450 U. S. 728, 739 (1981) . Among its other provisions, the FLSA requires employers to pay overtime compensation to covered employees who work more than 40 hours in a given week. The rate of overtime pay must be “not less than one and one-half times the regular rate” of the employee’s pay. §207(a). Five current and former service advisors brought this suit alleging that the automobile dealership where they were employed was required by the FLSA to pay them overtime wages. The dealership contends that the position and duties of a service advisor bring these employees within §213(b)(10)(A), which establishes an exemption from the FLSA overtime provisions for certain employees engaged in selling or servicing automobiles. The case turns on the interpretation of this exemption. I A Automobile dealerships in many communities not only sell vehicles but also sell repair and maintenance services. Among the employees involved in providing repair and maintenance services are service advisors, partsmen, and mechanics. Service advisors interact with customers and sell them services for their vehicles. A service advisor’s duties may include meeting customers; listening to their concerns about their cars; suggesting repair and maintenance services; selling new accessories or replacement parts; recording service orders; following up with customers as the services are performed (for instance, if new problems are discovered); and explaining the repair and maintenance work when customers return for their vehicles. See App. 40–41; see also Brennan v. Deel Motors, Inc., 475 F. 2d 1095, 1096 (CA5 1973); 29 CFR §779.372(c)(4) (1971). Partsmen obtain the vehicle parts needed to perform repair and maintenance and provide those parts to the mechanics. See §779.372(c)(2). Mechanics perform the actual repair and maintenance work. See §779.372(c)(3). In 1961, Congress enacted a blanket exemption from the FLSA’s minimum wage and overtime provisions for all automobile dealership employees. Fair Labor Standards Amendments of 1961, §9, 75Stat. 73. In 1966, Congress repealed that broad exemption and replaced it with a narrower one. The revised statute did not exempt dealership employees from the minimum wage requirement. It also limited the exemption from the overtime compensation requirement to cover only certain employees—in particular, “any salesman, partsman, or mechanic primarily engaged in selling or servicing automobiles, trailers, trucks, farm implements, or aircraft” at a covered dealership. Fair Labor Standards Amendments of 1966, §209, 80Stat. 836. Congress authorized the Department of Labor to “promulgate necessary rules, regulations, or orders” with respect to this new provision. §602, id., at 844. The Department exercised that authority in 1970 and issued a regulation that defined the statutory terms “salesman,” “partsman,” and “mechanic.” 35 Fed. Reg. 5896 (1970) (codified at 29 CFR §779.372(c)). The Department intended its regulation as a mere interpretive rule explaining its own views, rather than a legislative rule with the force and effect of law; and so the Department did not issue the regulation through the notice-and-comment procedures of the Administrative Procedure Act. See 35 Fed. Reg. 5856; see also 5 U. S. C. §553(b)(A) (exempting interpretive rules from notice and comment). The 1970 interpretive regulation defined “salesman” to mean “an employee who is employed for the purpose of and is primarily engaged in making sales or obtaining orders or contracts for sale of the vehicles or farm implements which the establishment is primarily engaged in selling.” 29 CFR §779.372(c)(1) (1971). By limiting the statutory term to salesmen who sell vehicles or farm implements, the regulation excluded service advisors from the exemption, since a service advisor sells repair and maintenance services but not the vehicle itself. The regulation made that exclusion explicit in a later subsection: “Employees variously described as service manager, service writer, service advisor, or service salesman . . . are not exempt under [the statute]. This is true despite the fact that such an employee’s principal function may be disagnosing [sic] the mechanical condition of vehicles brought in for repair, writing up work orders for repairs authorized by the customer, assigning the work to various employees and directing and checking on the work of mechanics.” §779.372(c)(4). Three years later, the Court of Appeals for the Fifth Circuit rejected the Department’s conclusion that service advisors are not covered by the statutory exemption. Deel Motors, supra. Certain District Courts followed that precedent. See Yenney v. Cass County Motors, 81 CCH LC ¶33,506 (Neb. 1977); Brennan v. North Bros. Ford, Inc., 76 CCH LC ¶33,247 (ED Mich. 1975), aff’d sub nom. Dunlop v. North Bros. Ford, Inc., 529 F. 2d 524 (CA6 1976) (table); Brennan v. Import Volkswagen, Inc., 81 CCH LC ¶33,522 (Kan. 1975). In the meantime, Congress amended the statutory provision by enacting its present text, which now sets out the exemption in two subsections. Fair Labor Standards Amendments of 1974, §14, 88Stat. 65. The first subsection is at issue in this case. It exempts “any salesman, partsman, or mechanic primarily engaged in selling or servicing automobiles, trucks, or farm implements” at a covered dealership. 29 U. S. C. §213(b)(10)(A). The second subsection exempts “any salesman primarily engaged in selling trailers, boats, or aircraft” at a covered dealership. §213(b)(10)(B). The statute thus exempts certain employees engaged in servicing automobiles, trucks, or farm implements, but not similar employees engaged in servicing trailers, boats, or aircraft. In 1978, the Department issued an opinion letter departing from its previous position. Taking a position consistent with the cases decided by the courts, the opinion letter stated that service advisors could be exempt under §213(b)(10)(A). Dept. of Labor, Wage & Hour Div., Opinion Letter No. 1520 (WH–467) (1978), [1978–1981 Transfer Binder] CCH Wages–Hours Administrative Rulings ¶31,207. The letter acknowledged that the Department’s new policy “represent[ed] a change from the position set forth in section 779.372(c)(4)” of its 1970 regulation. In 1987, the Department confirmed its 1978 interpretation by amending its Field Operations Handbook to clarify that service advisors should be treated as exempt under §213(b)(10)(A). It observed that some courts had interpreted the statutory exemption to cover service advisors; and it stated that, as a result of those decisions, it would “no longer deny the [overtime] exemption for such employees.” Dept. of Labor, Wage & Hour Div., Field Operations Handbook, Insert No. 1757, 24L04–4(k)(Oct. 20, 1987), online at https://perma.cc/5GHD-KCJJ (all Internet materials as last visited June 16, 2016). The Department again acknowledged that its new position represented a change from its 1970 regulation and stated that the regulation would “be revised as soon as is practicable.” Ibid. Twenty-one years later, in 2008, the Department at last issued a notice of proposed rulemaking. 73 Fed. Reg. 43654. The notice observed that every court that had considered the question had held service advisors to be exempt under §213(b)(10)(A), and that the Department itself had treated service advisors as exempt since 1987. Id., at 43658–43659. The Department proposed to revise its regulations to accord with existing practice by interpreting the exemption in §213(b)(10)(A) to cover service advisors. In 2011, however, the Department changed course yet again. It announced that it was “not proceeding with the proposed rule.” 76 Fed. Reg. 18833. Instead, the Department completed its 2008 notice-and-comment rulemaking by issuing a final rule that took the opposite position from the proposed rule. The new final rule followed the original 1970 regulation and interpreted the statutory term “salesman” to mean only an employee who sells automobiles, trucks, or farm implements. Id., at 18859 (codified at 29 CFR §779.372(c)(1)). The Department gave little explanation for its decision to abandon its decades-old practice of treating service advisors as exempt under §213(b)(10)(A). It was also less than precise when it issued its final rule. As described above, the 1970 regulation included a separate subsection stating in express terms that service advisors “arenot exempt” under the relevant provision. 29 CFR §779.372(c)(4) (1971). In promulgating the 2011 regulation, however, the Department eliminated that separate subsection. According to the United States, this change appears to have been “an inadvertent mistake in drafting.” Tr. of Oral Arg. 50. B Petitioner is a Mercedes-Benz automobile dealership in the Los Angeles area. Respondents are or were employed by petitioner as service advisors. They assert that petitioner required them to be at work from 7 a.m. to 6 p.m. at least five days per week, and to be available for work matters during breaks and while on vacation. App. 39–40. Respondents were not paid a fixed salary or an hourly wage for their work; instead, they were paid commissions on the services they sold. Id., at 40–41. Respondents sued petitioner in the United States District Court for the Central District of California, alleging that petitioner violated the FLSA by failing to pay them overtime compensation when they worked more than 40 hours in a week. Id., at 42–44. Petitioner moved to dismiss, arguing that the FLSA overtime provisions do not apply to respondents because service advisors are covered by the statutory exemption in §213(b)(10)(A). The District Court agreed and granted the motion to dismiss. The Court of Appeals for the Ninth Circuit reversed in relevant part. It construed the statute by deferring under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) , to the interpretation set forth by the Department in its 2011 regulation. Applying that deference, the Court of Appeals held that service advisors are not covered by the §213(b)(10)(A) exemption. 780 F. 3d 1267 (2015). The Court of Appeals recognized, however, that its decision conflicted with cases from a number of other courts. Id., at 1274 (citing, inter alia, Walton v. Greenbrier Ford, Inc., 370 F. 3d 446 (CA4 2004); Deel Motors, 475 F. 2d 1095; Thompson v. J. C. Billion, Inc., 368 Mont. 299, 294 P. 3d 397 (2013)). This Court granted certiorari to resolve the question. 577 U. S. ___ (2016). II A The full text of the statutory subsection at issue states that the overtime provisions of the FLSA shall not apply to: “any salesman, partsman, or mechanic primarily engaged in selling or servicing automobiles, trucks, or farm implements, if he is employed by a nonmanufacturing establishment primarily engaged in the business of selling such vehicles or implements to ultimate purchasers.” §213(b)(10)(A). The question presented is whether this exemption should be interpreted to include service advisors. To resolve that question, it is necessary to determine what deference,if any, the courts must give to the Department’s 2011 interpretation. In the usual course, when an agency is authorized by Congress to issue regulations and promulgates a regulation interpreting a statute it enforces, the interpretation receives deference if the statute is ambiguous and if the agency’s interpretation is reasonable. This principle is implemented by the two-step analysis set forth in Chevron. At the first step, a court must determine whether Congress has “directly spoken to the precise question at issue.” 467 U. S., at 842. If so, “that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Id., at 842–843. If not, then at the second step the court must defer to the agency’s interpretation if it is “reasonable.” Id., at 844. A premise of Chevron is that when Congress grants an agency the authority to administer a statute by issuing regulations with the force of law, it presumes the agency will use that authority to resolve ambiguities in the statutory scheme. See id., at 843–844; United States v. Mead Corp., 533 U. S. 218 –230 (2001). When Congress authorizes an agency to proceed through notice-and-comment rulemaking, that “relatively formal administrative procedure” is a “very good indicator” that Congress intended the regulation to carry the force of law, so Chevron should apply. Mead Corp., supra, at 229–230. But Chevron deference is not warranted where the regulation is “procedurally defective”—that is, where the agency errs by failing to follow the correct procedures in issuing the regulation. 533 U. S., at 227; cf. Long Island Care at Home, Ltd. v. Coke, 551 U. S. 158 –176 (2007) (rejecting challenge to procedures by which regulation was issued and affording Chevron deference). Of course, a party might be foreclosed in some instances from challenging the procedures used to promulgate a given rule. Cf., e.g., JEM Broadcasting Co. v. FCC, 22 F. 3d 320, 324–326 (CADC 1994); cf. also Auer v. Robbins, 519 U. S. 452 –459 (1997) (party cannot challenge agency’s failure to amend its rule in light of changed circumstances without first seeking relief from the agency). But where a proper challenge is raised to the agency procedures, and those procedures are defective, a court should not accord Chevron deference to the agency interpretation. Respondents do not contest the manner in which petitioner has challenged the agency procedures here, and so this opinion assumes without deciding that the challenge was proper. One of the basic procedural requirements of administrative rulemaking is that an agency must give adequate reasons for its decisions. The agency “must examine the relevant data and articulate a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Assn. of United States, Inc. v. State Farm Mut. Automobile Ins. Co., 463 U. S. 29, 43 (1983) (internal quotation marks omitted). That requirement is satisfied when the agency’s explanation is clear enough that its “path may reasonably be discerned.” Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc., 419 U. S. 281, 286 (1974) . But where the agency has failed to provide even that minimal level of analysis, its action is arbitrary and capricious and so cannot carry the force of law. See 5 U. S. C. §706(2)(A); State Farm, supra, at 42–43. Agencies are free to change their existing policies as long as they provide a reasoned explanation for the change. See, e.g., National Cable & Telecommunications Assn. v. Brand X Internet Services, 545 U. S. 967 –982 (2005); Chevron, 467 U. S., at 863–864. When an agency changes its existing position, it “need not always provide a more detailed justification than what would suffice for a new policy created on a blank slate.” FCC v. Fox Television Stations, Inc., 556 U. S. 502, 515 (2009) . But the agency must at least “display awareness that it is changing position” and “show that there are good reasons for the new policy.” Ibid. (emphasis deleted). In explaining its changed position, an agency must also be cognizant that longstanding policies may have “engendered serious reliance interests that must be taken into account.” Ibid.; see also Smiley v. Citibank (South Dakota), N. A., 517 U. S. 735, 742 (1996) . “In such cases it is not that further justification is demanded by the mere fact of policy change; but that a reasoned explanation is needed for disregarding facts and circumstances that underlay or were engendered by the prior policy.” Fox Television Stations, supra, at 515–516. It follows that an “[u]nexplained inconsistency” in agency policy is “a reason for holding an interpretation to be an arbitrary and capricious change from agency practice.” Brand X, supra, at 981. An arbitrary and capricious regulation of this sort is itself unlawful and receives no Chevron deference. See Mead Corp., supra, at 227. B Applying those principles here, the unavoidable conclusion is that the 2011 regulation was issued without the reasoned explanation that was required in light of the Department’s change in position and the significant reliance interests involved. In promulgating the 2011 regulation, the Department offered barely any explanation. A summary discussion may suffice in other circumstances, but here—in particular because of decades of industry reliance on the Department’s prior policy—the explanation fell short of the agency’s duty to explain why it deemed it necessary to overrule its previous position. The retail automobile and truck dealership industry had relied since 1978 on the Department’s position that service advisors are exempt from the FLSA’s overtime pay requirements. See National Automobile Dealers Association, Comment Letter on Proposed Rule Updating Reg-ulations Issued Under the Fair Labor Standards Act(Sept. 26, 2008), online at https://www.regulations.gov/#!documentDetail;D=WHD-2008-0003-0038. Dealerships and service advisors negotiated and structured their compensation plans against this background understanding. Requiring dealerships to adapt to the Department’s new position could necessitate systemic, significant changes to the dealerships’ compensation arrangements. See Brief for National Automobile Dealers Association et al. as Amici Curiae 13–14. Dealerships whose service advisors are not compensated in accordance with the Department’s new views could also face substantial FLSA liability, see 29 U. S. C. §216(b), even if this risk of liability may be diminished in some cases by the existence of a separate FLSA exemption for certain employees paid on a commission basis, see §207(i), and even if a dealership could defend against retroactive liability by showing it relied in good faith on the prior agency position, see §259(a). In light of this background, the Department needed a more reasoned explanation for its decision to depart from its existing enforcement policy. The Department said that, in reaching its decision, it had “carefully considered all of the comments, analyses, and arguments made for and against the proposed changes.” 76 Fed. Reg. 18832. And it noted that, since 1978, ithad treated service advisors as exempt in certain circumstances. Id., at 18838. It also noted the comment from the National Automobile Dealers Association stating that the industry had relied on that interpretation. Ibid. But when it came to explaining the “good reasons for the new policy,” Fox Television Stations, supra, at 515, the Department said almost nothing. It stated only that it would not treat service advisors as exempt because “the statute does not include such positions and the Department recognizes that there are circumstances under which the requirements for the exemption would not be met.” 76 Fed. Reg. 18838. It continued that it “believes that this interpretation is reasonable” and “sets forth the appropriate approach.” Ibid. Although an agency may justify its policy choice by explaining why that policy “is more consistent with statutory language” than alternative policies, Long Island Care at Home, 551 U. S., at 175 (internal quotation marks omitted), the Department did not analyze or explain why the statute should be interpreted to exempt dealership employees who sell vehicles but not dealership employees who sell services (that is, service advisors). And though several public comments supported the Department’s reading of the statute, the Department did not explain what (if anything) it found persuasive in those comments beyond the few statements above. It is not the role of the courts to speculate on reasons that might have supported an agency’s decision. “[W]e may not supply a reasoned basis for the agency’s action that the agency itself has not given.” State Farm, 463 U. S., at 43 (citing SEC v. Chenery Corp., 332 U. S. 194, 196 (1947) ). Whatever potential reasons the Department might have given, the agency in fact gave almost no reasons at all. In light of the serious reliance interests at stake, the Department’s conclusory statements do not suffice to explain its decision. See Fox Television Stations, 556 U. S., at 515–516. This lack of reasoned explication for a regulation that is inconsistent with the Department’s longstanding earlier position results in a rule that cannot carry the force of law. See 5 U. S. C. §706(2)(A); State Farm, supra, at 42–43. It follows that this regulation does not receive Chevron deference in the interpretation of the relevant statute. * * * For the reasons above, §213(b)(10)(A) must be construed without placing controlling weight on the Department’s 2011 regulation. Because the decision below relied on Chevron deference to this regulation, it is appropriate to remand for the Court of Appeals to interpret the statute in the first instance. Cf. Mead, 533 U. S., at 238–239. The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

578.US.2015_14-940

Under the one-person, one-vote principle, jurisdictions must design legislative districts with equal populations. See Wesberry v. Sanders, 376 U. S. 1 –8, Reynolds v. Sims, 377 U. S. 533 . In the context of state and local legislative districting, States may deviate somewhat from perfect population equality to accommodate traditional districting objectives. Where the maximum population deviation between the largest and smallest district is less than 10%, a state or local legislative map presumptively complies with the one-person, one-vote rule. Texas, like all other States, uses total-population numbers from the decennial census when drawing legislative districts. After the 2010 census, Texas adopted a State Senate map that has a maximum total-population deviation of 8.04%, safely within the presumptively permissible 10% range. However, measured by a voter-population baseline—eligible voters or registered voters—the map’s maximum population deviation exceeds 40%. Appellants, who live in Texas Senate districts with particularly large eligible- and registered-voter populations, filed suit against the Texas Governor and Secretary of State. Basing apportionment on total population, appellants contended, dilutes their votes in relation to voters in other Senate districts, in violation of the one-person, one-vote principle of the Equal Protection Clause. Appellants sought an injunction barring use of the existing Senate map in favor of a map that would equalize the voter population in each district. A three-judge District Court dismissed the complaint for failure to state a claim on which relief could be granted. Held: As constitutional history, precedent, and practice demonstrate, a State or locality may draw its legislative districts based on total population. Pp. 7–19. (a) Constitutional history shows that, at the time of the founding, the Framers endorsed allocating House seats to States based on total population. Debating what would become the Fourteenth Amendment, Congress reconsidered the proper basis for apportioning House seats. Retaining the total-population rule, Congress rejected proposals to allocate House seats to States on the basis of voter population. See U. S. Const., Amdt. 14, §2. The Framers recognized that use of a total-population baseline served the principle of representational equality. Appellants’ voter-population rule is inconsistent with the “theory of the Constitution,” Cong. Globe, 39th Cong., 1st Sess., 2766–2767, this Court recognized in Wesberry as underlying not just the method of allocating House seats to States but also the method of apportioning legislative seats within States. Pp. 8–15. (b) This Court’s past decisions reinforce the conclusion that States and localities may comply with the one-person, one-vote principle by designing districts with equal total populations. Appellants assert that language in this Court’s precedent supports their view that States should equalize the voter-eligible population of districts. But for every sentence appellants quote, one could respond with a line casting the one-person, one-vote guarantee in terms of equality of representation. See, e.g., Reynolds, 377 U. S., at 560–561. Moreover, from Reynolds on, the Court has consistently looked to total-population figures when evaluating whether districting maps violate the Equal Protection Clause by deviating impermissibly from perfect population equality. Pp. 15–18. (c) Settled practice confirms what constitutional history and prior decisions strongly suggest. Adopting voter-eligible apportionment as constitutional command would upset a well-functioning approach to districting that all 50 States and countless local jurisdictions have long followed. As the Framers of the Constitution and the Fourteenth Amendment comprehended, representatives serve all residents, not just those eligible to vote. Nonvoters have an important stake in many policy debates and in receiving constituent services. By ensuring that each representative is subject to requests and suggestions from the same number of constituents, total-population apportionment promotes equitable and effective representation. Pp. 18–19. (d) Because constitutional history, precedent, and practice reveal the infirmity of appellants’ claim, this Court need not resolve whether, as Texas now argues, States may draw districts to equalize voter-eligible population rather than total population. P. 19. Affirmed. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed an opinion concurring in the judgment. Alito, J., filed an opinion concurring in the judgment, in which Thomas, J., joined except as to Part III–B.

Texas, like all other States, draws its legislative districts on the basis of total population. Plaintiffs-appellants are Texas voters; they challenge this uniform method of districting on the ground that it produces un-equal districts when measured by voter-eligible population.Voter-eligible population, not total population, they urge, must be used to ensure that their votes will not be devalued in relation to citizens’ votes in other districts. We hold, based on constitutional history, this Court’s decisions, and longstanding practice, that a State may draw its legislative districts based on total population. I A This Court long resisted any role in overseeing the process by which States draw legislative districts. “The remedy for unfairness in districting,” the Court once held, “is to secure State legislatures that will apportion prop-erly, or to invoke the ample powers of Congress.” Colegrovev. Green, 328 U. S. 549, 556 (1946) . “Courts ought not to enter this political thicket,” as Justice Frankfurter put it. Ibid. Judicial abstention left pervasive malapportionment unchecked. In the opening half of the 20th century, there was a massive population shift away from rural areas and toward suburban and urban communities. Nevertheless, many States ran elections into the early 1960’s based on maps drawn to equalize each district’s population as it was composed around 1900. Other States used maps allocating a certain number of legislators to each county regardless of its population. These schemes left many rural districts significantly underpopulated in comparison with urban and suburban districts. But rural legislators who benefited from malapportionment had scant incentive to adopt new maps that might put them out of office. The Court confronted this ingrained structural inequal-ity in Baker v. Carr, 369 U. S. 186 –192 (1962). Thatcase presented an equal protection challenge to a Tennessee state-legislative map that had not been redrawn since 1901. See also id., at 192 (observing that, in the meantime, there had been “substantial growth and redistribution” of the State’s population). Rather than steering clear of the political thicket yet again, the Court held for the first time that malapportionment claims are justiciable. Id., at 237 (“We conclude that the complaint’s allegations of a denial of equal protection present a justiciable constitutional cause of action upon which appellants are entitled to a trial and a decision.”). Although the Court in Baker did not reach the merits of the equal protection claim, Baker’s justiciability ruling set the stage for what came to be known as the one-person, one-vote principle. Just two years after Baker, in Wes-berry v. Sanders, 376 U. S. 1 –8 (1964), the Court invalidated Georgia’s malapportioned congressional map, underwhich the population of one congressional district was “two to three times” larger than the population of the others. Relying on Article I, §2, of the Constitution, the Court required that congressional districts be drawn with equal populations. Id., at 7, 18. Later that same Term, in Reynolds v. Sims, 377 U. S. 533, 568 (1964) , the Court upheld an equal protection challenge to Alabama’s malapportioned state-legislative maps. “[T]he Equal Protection Clause,” the Court concluded, “requires that the seats in both houses of a bicameral state legislature must be apportioned on a population basis.” Ibid. Wesberry and Reynolds together instructed that jurisdictions must design both congressional and state-legislative districts with equal populations, and must regularly reapportion districts to prevent malapportionment.[1] Over the ensuing decades, the Court has several times elaborated on the scope of the one-person, one-vote rule. States must draw congressional districts with populations as close to perfect equality as possible. See Kirkpatrick v. Preisler, 394 U. S. 526 –531 (1969). But, when drawing state and local legislative districts, jurisdictions are permitted to deviate somewhat from perfect population equality to accommodate traditional districting objectives, among them, preserving the integrity of political subdivisions, maintaining communities of interest, and creating geographic compactness. See Brown v. Thomson, 462 U. S. 835 –843 (1983). Where the maximum population deviation between the largest and smallest district is less than 10%, the Court has held, a state or local legislative map presumptively complies with the one-person, one-vote rule. Ibid.[2] Maximum deviations above 10% are presumptively impermissible. Ibid. See also Mahan v. Howell, 410 U. S. 315, 329 (1973) (approving a state-legislative map with maximum population deviation of 16% to accommodate the State’s interest in “maintaining the integrity of political subdivision lines,” but cautioning that this deviation “may well approach tolerable limits”). In contrast to repeated disputes over the permissibility of deviating from perfect population equality, little controversy has centered on the population base jurisdictions must equalize. On rare occasions, jurisdictions have relied on the registered-voter or voter-eligible populations of districts. See Burns v. Richardson, 384 U. S. 73 –94 (1966) (holding Hawaii could use a registered-voter population base because of “Hawaii’s special population problems”—in particular, its substantial temporary military population). But, in the overwhelming majority of cases, jurisdictions have equalized total population, as measured by the decennial census. Today, all States use total-population numbers from the census when designing congressional and state-legislative districts, and only seven States adjust those census numbers in any meaningful way.[3] B Appellants challenge that consensus. After the 2010 census, Texas redrew its State Senate districts using a total-population baseline. At the time, Texas was subject to the preclearance requirements of §5 of the Voting Rights Act of 1965. 52 U. S. C. §10304 (requiring jurisdictions to receive approval from the U. S. Department of Justice or the U. S. District Court for the District of Columbia before implementing certain voting changes). Once it became clear that the new Senate map, S148, would not receive preclearance in advance of the 2012 elections, the U. S. District Court for the Western District of Texas drew an interim Senate map, S164, which also equalized the total population of each district. See Davis v. Perry, No. SA–11–CV–788 (Nov. 23, 2011).[4] On direct appeal, this Court observed that the District Court had failed to “take guidance from the State’s recently enacted plan in drafting an interim plan,” and therefore vacated the District Court’s map. Perry v. Perez, 565 U. S. ___, ___, ___–___ (2012) (per curiam) (slip op., at 4, 8–10). The District Court, on remand, again used census data to draw districts so that each included roughly the same size total population. Texas used this new interim map, S172, in the 2012 elections, and, in 2013, the Texas Legislature adopted S172 as the permanent Senate map. See App. to Brief for Texas Senate Hispanic Caucus et al. as Amici Curiae 5 (reproducing the current Senate map). The permanent map’s maximum total-population deviation is 8.04%, safely within the presumptively permissible 10% range. But measured by a voter-population baseline—eligible voters or registered voters—the map’s maximum population deviation exceeds 40%. Appellants Sue Evenwel and Edward Pfenninger live in Texas Senate districts (one and four, respectively) with particularly large eligible- and registered-voter populations. Contending that basing apportionment on total population dilutes their votes in relation to voters in other Senate districts, in violation of the one-person, one-vote principle of the Equal Protection Clause,[5] appellants filed suit in the U. S. District Court for the Western District of Texas. They named as defendants the Governor and Secretary of State of Texas, and sought a permanent injunction barring use of the existing Senate map in favor of a map that would equalize the voter population in each district. The case was referred to a three-judge District Court for hearing and decision. See 28 U. S. C. §2284(a); Shapiro v. McManus, 577 U. S. ___, ___–___ (2015) (slip op., at 5–7). That court dismissed the complaint for failure to state a claim on which relief could be granted. Appellants, the District Court explained, “rel[y] upon a theory never before accepted by the Supreme Court or any circuit court: that the metric of apportionment employed by Texas (total population) results in an unconstitutional apportionment because it does not achieve equality as measured by Plaintiffs’ chosen metric—voter population.” App. to Juris. Statement 9a. Decisions of this Court, the District Court concluded, permit jurisdictions to use any neutral, nondiscriminatory population baseline, including total population, when drawing state and local legislative districts. Id., at 13a–14a.[6] We noted probable jurisdiction, 575 U. S. ___ (2015), and now affirm. II The parties and the United States advance different positions in this case. As they did before the District Court, appellants insist that the Equal Protection Clause requires jurisdictions to draw state and local legislative districts with equal voter-eligible populations, thus protecting “voter equality,” i.e., “the right of eligible voters to an equal vote.” Brief for Appellants 14.[7] To comply with their proposed rule, appellants suggest, jurisdictions should design districts based on citizen-voting-age-population (CVAP) data from the Census Bureau’s American Community Survey (ACS), an annual statistical sample of the U. S. population. Texas responds that jurisdic-tions may, consistent with the Equal Protection Clause, design districts using any population baseline—including total population and voter-eligible population—so long as the choice is rational and not invidiously discriminatory. Although its use of total-population data from the census was permissible, Texas therefore argues, it could have used ACS CVAP data instead. Sharing Texas’ position that the Equal Protection Clause does not mandate use of voter-eligible population, the United States urges us not to address Texas’ separate assertion that the Constitution allows States to use alternative population baselines, including voter-eligible population. Equalizing total population, the United States maintains, vindicates the principle of representational equality by “ensur[ing] that the voters in each district have the power to elect a representative who represents the same number of constituents as all other representatives.” Brief for United States as Amicus Curiae 5. In agreement with Texas and the United States, we reject appellants’ attempt to locate a voter-equality mandate in the Equal Protection Clause. As history, precedent, and practice demonstrate, it is plainly permissible for jurisdictions to measure equalization by the total population of state and local legislative districts. A We begin with constitutional history. At the time of the founding, the Framers confronted a question analogous to the one at issue here: On what basis should congressional districts be allocated to States? The Framers’ solution, now known as the Great Compromise, was to provide each State the same number of seats in the Senate, and to allocate House seats based on States’ total populations. “Representatives and direct Taxes,” they wrote, “shall be apportioned among the several States which may be included within this Union, according to their respective Numbers.” U. S. Const., Art. I, §2, cl. 3 (emphasis added). “It is a fundamental principle of the proposed constitution,” James Madison explained in the Federalist Papers, “that as the aggregate number of representatives allotted to the several states, is to be . . . founded on the aggregate number of inhabitants; so, the right of choosing this allotted number in each state, is to be exercised by such part of the inhabitants, as the state itself may designate.” The Federalist No. 54, p. 284 (G. Carey & J. McClellan eds. 2001). In other words, the basis of representation in the House was to include all inhabitants—although slaves were counted as only three-fifths of a person—even though States remained free to deny many of those inhabitants the right to participate in the selection of their representatives.[8] Endorsing apportionment based on total population, Alexander Hamilton declared: “There can be no truer principle than this—that every individual of the commu-nity at large has an equal right to the protection of govern-ment.” 1 Records of the Federal Convention of 1787, p. 473 (M. Farrand ed. 1911).[9] When debating what is now the Fourteenth Amendment, Congress reconsidered the proper basis for apportioning House seats. Concerned that Southern States would not willingly enfranchise freed slaves, and aware that “a slave’s freedom could swell his state’s population for purposes of representation in the House by one person, rather than only three-fifths,” the Framers of the Fourteenth Amendment considered at length the possibility of allocating House seats to States on the basis of voter population. J. Sneed, Footprints on the Rocks of the Mountain: An Account of the Enactment of the Fourteenth Amendment 28 (1997). See also id., at 35 (“[T]he apportionment issue consumed more time in the Fourteenth Amendment debates than did any other topic.”). In December 1865, Thaddeus Stevens, a leader of the Radical Republicans, introduced a constitutional amendment that would have allocated House seats to States “according to their respective legal voters”; in addition, the proposed amendment mandated that “[a] true census of the legal voters shall be taken at the same time with the regular census.” Cong. Globe, 39th Cong., 1st Sess., 10 (1866). Supporters of apportionment based on voter population employed the same voter-equality reasoning that appellants now echo. See, e.g., id., at 380 (remarks of Rep. Orth) (“[T]he true principle of representation in Congress is that voters alone should form the basis, and that each voter should have equal political weight in our Government. . . .”); id., at 404 (remarks of Rep. Lawrence) (use of total population “disregards the fundamental idea of all just representation, that every voter should be equal in political power all over the Union”). Voter-based apportionment proponents encountered fierce resistance from proponents of total-population apportionment. Much of the opposition was grounded in the principle of representational equality. “As an abstract proposition,” argued Representative James G. Blaine, a leading critic of allocating House seats based on voter population, “no one will deny that population is the true basis of representation; for women, children, and other non-voting classes may have as vital an interest in the legislation of the country as those who actually deposit the ballot.” Id., at 141. See also id., at 358 (remarks of Rep. Conkling) (arguing that use of a voter-population basis “would shut out four fifths of the citizens of the country—women and children, who are citizens, who are taxed, and who are, and always have been, represented”); id., at 434 (remarks of Rep. Ward) (“[W]hat becomes of that large class of non-voting tax-payers that are found in every section? Are they in no matter to be represented? They certainly should be enumerated in making up the whole number of those entitled to a representative.”). The product of these debates was §2 of the Fourteenth Amendment, which retained total population as the congressional apportionment base. See U. S. Const., Amdt. 14, §2 (“Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed.”). Introducing the final version of the Amendment on the Senate floor, Senator Jacob Howard explained: “[The] basis of representation is numbers . . . ; that is, the whole population except untaxed Indians and persons excluded by the State laws for rebellion or other crime. . . . The committee adopted numbers as the most just and satisfactory basis, and this is the principle upon which the Constitution itself was originally framed, that the basis of representation should depend upon numbers; and such, I think, after all, is the safest and most secure principle upon which the Government can rest. Numbers, not voters; numbers, not property; this is the theory of the Constitution.” Cong. Globe, 39th Cong., 1st Sess., 2766–2767 (1866). Appellants ask us to find in the Fourteenth Amendment’s Equal Protection Clause a rule inconsistent with this “theory of the Constitution.” But, as the Court recognized in Wesberry, this theory underlies not just the method of allocating House seats to States; it applies as well tothe method of apportioning legislative seats within States. “The debates at the [Constitutional] Convention,” the Court explained, “make at least one fact abundantly clear: that when the delegates agreed that the House should represent ‘people,’ they intended that in allocating Congressmen the number assigned to each state should be determined solely by the number of inhabitants.” 376 U. S., at 13. “While it may not be possible to draw congressional districts with mathematical precision,” the Court acknowledged, “that is no excuse for ignoring our Constitution’s plain objective of making equal representation for equal numbers of people the fundamental goalfor the House of Representatives.” Id., at 18 (emphasis added). It cannot be that the Fourteenth Amendment calls for the apportionment of congressional districts based on total population, but simultaneously prohibits States from apportioning their own legislative districts on the same basis. Cordoning off the constitutional history of congressional districting, appellants stress two points.[10] First, they draw a distinction between allocating seats to States, and apportioning seats within States. The Framers selected total population for the former, appellants and their amici argue, because of federalism concerns inapposite to intra-state districting. These concerns included the perceived risk that a voter-population base might encourage States to expand the franchise unwisely, and the hope that a total-population base might counter States’ incentive to undercount their populations, thereby reducing their share of direct taxes. Wesberry, however, rejected the distinction appellants now press. See supra, at 12. Even without the weight of Wesberry, we would find appellants’ distinction unconvincing. One can accept that federalism—or, as Justice Alito emphasizes, partisan and regional political advantage, see post, at 6–13—figured in the Framers’ selection of total population as the basis for allocating congressional seats. Even so, it remains beyond doubt that the principle of representational equality figured prominently in the decision to count people, whether or not they qualify as voters.[11] Second, appellants and Justice Alito urge, see post, at 5–6, the Court has typically refused to analogize to features of the federal electoral system—here, the constitutional scheme governing congressional apportionment—when considering challenges to state and local election laws. True, in Reynolds, the Court rejected Alabama’s argument that it had permissibly modeled its State Senate apportionment scheme—one Senator for each county—on the United States Senate. “[T]he federal analogy,” the Court explained, “[is] inapposite and irrelevant to state legislative districting schemes” because “[t]he system of representation in the two Houses of the Federal Congress” arose “from unique historical circumstances.” 377 U. S., at 573–574. Likewise, in Gray v. Sanders, 372 U. S. 368 –372, 378 (1963), Georgia unsuccessfully attempted to defend, by analogy to the electoral college, its scheme of assigning a certain number of “units” to the winner of each county in statewide elections. Reynolds and Gray, however, involved features of the federal electoral system that contravene the principles of both voter and representational equality to favor interests that have no relevance outside the federal context. Senate seats were allocated to States on an equal basis to respect state sovereignty and increase the odds that the smaller States would ratify the Constitution. See Wesberry, 376 U. S., at 9–13 (describing the history of the Great Compromise). See also Reynolds, 377 U. S., at 575 (“Political subdivisions of States—counties, cities, or whatever—never were and never have been considered as sovereign entities. . . . The relationship of the States to the Federal Government could hardly be less analogous.”). “The [Electoral] College was created to permit the most knowledge-able members of the community to choose the executive of anation whose continental dimensions were thought to preclude an informed choice by the citizenry at large.” Williams v. Rhodes, 393 U. S. 23 –44 (1968) (Harlan, J., concurring in result). See also Gray, 372 U. S., at 378 (“The inclusion of the electoral college in the Constitution, as the result of specific historical concerns, validated the collegiate principle despite its inherent numerical inequality.” (footnote omitted)). By contrast, as earlier developed, the constitutional scheme for congressional apportion-ment rests in part on the same representational concerns that exist regarding state and local legislative districting. The Framers’ answer to the apportionment question inthe congressional context therefore undermines appellants’ contention that districts must be based on voter population. B Consistent with constitutional history, this Court’s past decisions reinforce the conclusion that States and localities may comply with the one-person, one-vote principle by designing districts with equal total populations. Quoting language from those decisions that, in appellants’ view, supports the principle of equal voting power—and emphasizing the phrase “one-person, one-vote”—appellants contend that the Court had in mind, and constantly meant, that States should equalize the voter-eligible population of districts. See Reynolds, 377 U. S., at 568 (“[A]n individual’s right to vote for State legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living on other parts of the State.”); Gray, 372 U. S., at 379–380 (“The concept of ‘we the people’ under the Constitution visualizes no preferred class of voters but equality among those who meet the basic qualifications.”). See also Hadley v. Junior College Dist. of Metropolitan Kansas City, 397 U. S. 50, 56 (1970) (“[W]hen members of an elected body are chosen from separate districts, each district must be established on a basis that will insure, as far as is practicable, that equal numbers of voters can vote for proportionally equal numbers of officials.”). Appellants, however, extract far too much from selectively chosen language and the “one-person, one-vote” slogan. For every sentence appellants quote from the Court’s opinions, one could respond with a line casting the one-person, one-vote guarantee in terms of equality of representation, not voter equality. In Reynolds, for instance, the Court described “the fundamental principle of representative government in this country” as “one of equal representation for equal numbers of people.” 377 U. S., at 560–561. See also Davis v. Bandemer, 478 U. S. 109, 123 (1986) (“[I]n formulating the one person, one vote formula, the Court characterized the question posed by election districts of disparate size as an issue of fair representation.”); Reynolds, 377 U. S., at 563 (rejecting state districting schemes that “give the same number of representatives to unequal numbers of constituents”). And the Court has suggested, repeatedly, that districting based on total population serves both the State’s interest in preventing vote dilution and its interest in ensuring equality of representation. See Board of Estimate of City of New York v. Morris, 489 U. S. 688 –694 (1989) (“If districts of widely unequal population elect an equal number of representatives, the voting power of each citizen in the larger constituencies is debased and the citizens in those districts have a smaller share of representation than do those in the smaller districts.”). See also Kirkpatrick, 394 U. S., at 531 (recognizing in a congressional-districting case that “[e]qual representation for equal numbers of people is a principle designed to prevent debasement of voting power and diminution of access to elected representatives”).[12] Moreover, from Reynolds on, the Court has consistently looked to total-population figures when evaluating whether districting maps violate the Equal Protection Clause bydeviating impermissibly from perfect population equality. See Brief for Appellees 29–31 (collecting cases brought under the Equal Protection Clause). See also id., at 31, n. 9 (collecting congressional-districting cases). Appellants point to no instance in which the Court has determined the permissibility of deviation based on eligible- or registered-voter data. It would hardly make sense for the Court to have mandated voter equality sub silentio and then used a total-population baseline to evaluate compliance with that rule. More likely, we think, the Court has always assumed the permissibility of drawing districts to equalize total population. “In the 1960s,” appellants counter, “the distribution of the voting population generally did not deviate from the distribution of total population to the degree necessary to raise this issue.” Brief for Appellants 27. To support this assertion, appellants cite only a District Court decision, which found no significant deviation in the distribution of voter and total population in “densely populated areas of New York State.” WMCA, Inc. v. Lomenzo, 238 F. Supp. 916, 925 (SDNY), aff’d, 382 U. S. 4 (1965) ( per curiam). Had this Court assumed such equivalence on a national scale, it likely would have said as much.[13] Instead, in Gaffney v. Cummings, 412 U. S. 735 –747 (1973), the Court acknowledged that voters may be distributed un-evenly within jurisdictions. “[I]f it is the weight of a person’s vote that matters,” the Court observed, then “total population—even if stable and accurately taken—may not actually reflect that body of voters whose votes must be counted and weighed for the purposes of reapportionment, because ‘census persons’ are not voters.” Id., at 746. Nonetheless, the Court in Gaffney recognized that the one-person, one-vote rule is designed to facilitate “[f ]air and effective representation,” id., at 748, and evaluated compliance with the rule based on total population alone, id., at 750. C What constitutional history and our prior decisions strongly suggest, settled practice confirms. Adopting voter-eligible apportionment as constitutional command would upset a well-functioning approach to districting that all 50 States and countless local jurisdictions have followed for decades, even centuries. Appellants have shown no reason for the Court to disturb this longstanding use of total population. See Walz v. Tax Comm’n of City of New York, 397 U. S. 664, 678 (1970) (“unbroken practice” followed “openly and by affirmative state action, not covertly or by state inaction, is not something to be lightly cast aside”). See also Burson v. Freeman, 504 U. S. 191 –206 (1992) (plurality opinion) (upholding a law limiting campaigning in areas around polling places in part because all 50 States maintain such laws, so there is a “widespread and time-tested consensus” that legislation of this order serves important state interests). As the Framers of the Constitution and the Fourteenth Amendment comprehended, representatives serve all residents, not just those eligible or registered to vote. See supra, at 8–12. Nonvoters have an important stake in many policy debates—children, their parents, even their grandparents, for example, have a stake in a strong public-education system—and in receiving constituent services, such as help navigating public-benefits bureaucracies. By ensuring that each representative is subject to requests and suggestions from the same number of constituents, total-population apportionment promotes equitable and effective representation. See McCormick v. United States, 500 U. S. 257, 272 (1991) (“Serving constituents and supporting legislation that will benefit the district and individ-uals and groups therein is the everyday business of a legislator.”).[14] In sum, the rule appellants urge has no mooring in the Equal Protection Clause. The Texas Senate map, we therefore conclude, complies with the requirements of the one-person, one-vote principle.[15] Because history, precedent, and practice suffice to reveal the infirmity of appellants’ claims, we need not and do not resolve whether, as Texas now argues, States may draw districts to equalize voter-eligible population rather than total population. * * * For the reasons stated, the judgment of the United States District Court for the Western District of Texas is Affirmed.Notes 1 In Avery v. Midland County, 390 U. S. 474 –486 (1968), the Court applied the one-person, one-vote rule to legislative apportionment at the local level. 2 Maximum population deviation is the sum of the percentage deviations from perfect population equality of the most- and least-populated districts. See Chapman v. Meier, 420 U. S. 1, 22 (1975) . For example, if the largest district is 4.5% overpopulated, and the smallest district is 2.3% underpopulated, the map’s maximum population deviation is 6.8%. 3 The Constitutions and statutes of ten States—California, Delaware, Hawaii, Kansas, Maine, Maryland, Nebraska, New Hampshire, New York, and Washington—authorize the removal of certain groups from the total-population apportionment base. See App. to Brief for Appellees 1a–46a (listing relevant state constitutional and statutory provisions). Hawaii, Kansas, and Washington exclude certain non-permanent residents, including nonresident members of the military. Haw. Const., Art. IV, §4; Kan. Const., Art. 10, §1(a); Wash. Const., Art. II, §43(5). See also N. H. Const., pt. 2, Art. 9–a (authorizing the state legislature to make “suitable adjustments to the general census . . . on account of non-residents temporarily residing in this state”). California, Delaware, Maryland, and New York exclude inmates who were domiciled out-of-state prior to incarceration. Cal. Elec. Code Ann. §21003(5) (2016 West Cum. Supp.); Del. Code Ann., Tit. 29, §804A (Supp. 2014); Md. State Govt. Code Ann. §2–2A–01 (2014); N. Y. Legis. Law Ann. §83–m(b) (2015 West Cum. Supp.). The Constitutions of Maine and Nebraska authorize the exclusion of noncitizen immigrants, Me. Const., Art. IV, pt. 1, §2; Neb. Const., Art. III, §5, but neither provision is “operational as written,” Brief for United States as Amicus Curiae 12, n. 3. 4 Various plaintiffs had challenged Texas’ State House, State Senate, and congressional maps under, inter alia, §2 of the Voting Rights Act of 1965. They sought and received an injunction barring Texas’ use of the new maps until those maps received §5 preclearance. See Allen v. State Bd. of Elections, 393 U. S. 544, 561 (1969) (“[A]n individual may bring a suit for declaratory judgment and injunctive relief, claiming that a state requirement is covered by §5, but has not been subjected to the required federal scrutiny.”). 5 Apart from objecting to the baseline, appellants do not challenge the Senate map’s 8.04% total-population deviation. Nor do they challenge the use of a total-population baseline in congressional districting. 6 As the District Court noted, the Ninth Circuit has likewise rejected appellants’ theory, i.e., that voter population must be roughly equalized. See Garza v. County of L. A., 918 F. 2d 763, 773–776 (CA9 1990). Also declining to mandate voter-eligible apportionment, the Fourth and Fifth Circuits have suggested that the choice of apportionment base may present a nonjusticiable political question. See Chen v. Houston, 206 F. 3d 502, 528 (CA5 2000) (“[T]his eminently political question has been left to the political process.”); Daly v. Hunt, 93 F. 3d 1212, 1227 (CA4 1996) (“This is quintessentially a decision that should be made by the state, not the federal courts, in the inherently political and legislative process of apportionment.”). 7 In the District Court, appellants suggested that districting bodies could also comply with the one-person, one-vote rule by equalizing the registered-voter populations of districts, but appellants have not repeated that argument before this Court. See Tr. of Oral Arg. 22–23. 8 As the United States observes, the “choice of constitutional language reflects the historical fact that when the Constitution was drafted and later amended, the right to vote was not closely correlated with citizenship.” Brief for United States as Amicus Curiae 18. Restrictions on the franchise left large groups of citizens, including women and many males who did not own land, unable to cast ballots, yet the Framers understood that these citizens were nonetheless entitled to representation in government. 9 Justice Alito observes that Hamilton stated this principle while opposing allocation of an equal number of Senate seats to each State. Post, at 7–8 (opinion concurring in judgment). That context, however, does not diminish Hamilton’s principled argument for allocating seats to protect the representational rights of “every individual of the community at large.” 1 Records of the Federal Convention of 1787, p. 473 (M. Farrand ed. 1911). Justice Alito goes on to quote James Madison for the proposition that Hamilton was concerned, simply and only, with “the outcome of a contest over raw political power.” Post, at 8. Notably, in the statement Justice Alito quotes, Madison was not attributing that motive to Hamilton; instead, according to Madison, Hamilton was attributing that motive to the advocates of equal representation for States. Farrand, supra, at 466. One need not gainsay that Hamilton’s backdrop was the political controversies of his day. That reality, however, has not deterred this Court’s past reliance on his statements of principle. See, e.g., Printz v. United States, 521 U. S. 898 –924 (1997). 10 Justice Alito adds a third, claiming “the allocation of congres-sional representation sheds little light” on the meaning of the one-person,one-vote rule “because that allocation plainly violates one person, one vote.” Post, at 4. For this proposition, Justice Alito notes the constitutional guarantee of two Senate seats and at least one House seat to each State, regardless of its population. But these guarantees bear no kinship to the separate question that dominated the Fourteenth Amendment’s ratification debates: After each State has received its guaranteed House seat, on what basis should additional seats be allocated? 11 Justice Alito asserts that we have taken the statements of the Fourteenth Amendment’s Framers “out of context.” Post, at 9. See also post, at 12 (“[C]laims about representational equality were invoked, if at all, only in service of the real goal: preventing southern States from acquiring too much power in the national government.”). Like Alexander Hamilton, see supra, at 9, n. 9, the Fourteenth Amendment’s Framers doubtless made arguments rooted in practical political realities as well as in principle. That politics played a part, however, does not warrant rejecting principled argument. In any event, motivations aside, the Framers’ ultimate choice of total population rather than voter population is surely relevant to whether, as appellants now argue, the Equal Protection Clause mandates use of voter population rather than total population. 12 Appellants also observe that standing in one-person, one-vote cases has rested on plaintiffs’ status as voters whose votes were diluted. But the Court has not considered the standing of nonvoters to challenge a map malapportioned on a total-population basis. This issue, moreover, is unlikely ever to arise given the ease of finding voters willing to serve as plaintiffs in malapportionment cases. 13 In contrast to the insubstantial evidence marshaled by appellants, the United States cites several studies documenting the uneven distribution of immigrants throughout the country during the 1960’s. See Brief for United States as Amicus Curiae 16. 14 Appellants point out that constituents have no constitutional right to equal access to their elected representatives. But a State certainly has an interest in taking reasonable, nondiscriminatory steps to facilitate access for all its residents. 15 Insofar as appellants suggest that Texas could have roughly equalized both total population and eligible-voter population, this Court has never required jurisdictions to use multiple population baselines. In any event, appellants have never presented a map that manages to equalize both measures, perhaps because such a map does not exist, or because such a map would necessarily ignore other traditional redistricting principles, including maintaining communities of interest and respecting municipal boundaries.

577.US.2015_14-840

The Federal Power Act (FPA) authorizes the Federal Energy Regulatory Commission (FERC) to regulate “the sale of electric energy at wholesale in interstate commerce,” including both wholesale electricity rates and any rule or practice “affecting” such rates. 16 U. S. C. §§824(b), 824d(a), 824e(a). But it places beyond FERC’s power, leaving to the States alone, the regulation of “any other sale”—i.e., any retail sale—of electricity. §824(b). In an increasingly competitive interstate electricity market, FERC has undertaken to ensure “just and reasonable” wholesale rates, §824d(a), by encouraging the creation of nonprofit entities to manage regions of the nationwide electricity grid. These wholesale market operators administer their portions of the grid to ensure that the network conducts electricity reliably, and each holds competitive auctions to set wholesale prices. These auctions balance supply and demand continuously by matching bids to provide electricity from generators with orders from utilities and other “load-serving entities” (LSEs) that buy power at wholesale for resale to users. All bids to supply electricity are stacked from lowest to highest, and accepted in that order until all requests for power have been met. Every electricity supplier is paid the price of the highest-accepted bid, known as the locational marginal price (LMP). In periods of high electricity demand, prices can reach extremely high levels as the least efficient generators have their supply bids accepted in the wholesale market auctions. Not only do rates rise dramatically during these peak periods, but the increased flow of electricity threatens to overload the grid and cause substantial service problems. Faced with these challenges, wholesale market operators devised wholesale demand response programs, which pay consumers for commitments to reduce their use of power during these peak periods. Just like bids to supply electricity, offers from aggregators of multiple users of electricity or large individual consumers to reduce consumption can be bid into the wholesale market auctions. When it costs less to pay consumers to refrain from using power than it does to pay producers to supply more of it, demand response can lower these wholesale prices and increase grid reliability. Wholesale operators began integrating these programs into their markets some 15 years ago and FERC authorized their use. Congress subsequently encouraged further development of demand response. Spurred on by Congress, FERC issued Order No. 719, which, among other things, requires wholesale market operators to receive demand response bids from aggregators of electricity consumers, except when the state regulatory authority overseeing those users’ retail purchases bars demand response participation. 18 CFR §35.28(g)(1). Concerned that the order had not gone far enough, FERC then issued the rule under review here, Order No. 745. §35.28(g)(1)(v) (Rule). It requires market operators to pay the same price to demand response providers for conserving energy as to generators for producing it, so long as a “net benefits test,” which ensures that accepted bids actually save consumers money, is met. The Rule rejected an alternative compensation scheme that would have subtracted from LMP the savings consumers receive from not buying electricity in the retail market, a formula known as LMP-G. The Rule also rejected claims that FERC lacked statutory authority to regulate the compensation operators pay for demand response bids. The Court of Appeals for the District of Columbia Circuit vacated the Rule, holding that FERC lacked authority to issue the order because it directly regulates the retail electricity market, and holding in the alternative that the Rule’s compensation scheme is arbitrary and capricious under the Administrative Procedure Act. Held: 1. The FPA provides FERC with the authority to regulate wholesale market operators’ compensation of demand response bids. The Court’s analysis proceeds in three parts. First, the practices at issue directly affect wholesale rates. Second, FERC has not regulated retail sales. Taken together, these conclusions establish that the Rule complies with the FPA’s plain terms. Third, the contrary view would conflict with the FPA’s core purposes. Pp. 14–29. (a) The practices at issue directly affect wholesale rates. The FPA has delegated to FERC the authority—and, indeed, the duty—to ensure that rules or practices “affecting” wholesale rates are just and reasonable. §§824d(a), 824e(a). To prevent the statute from assuming near-infinite breadth, see e.g., New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U. S. 645 , this Court adopts the D. C. Circuit’s common-sense construction limiting FERC’s “affecting” jurisdiction to rules or practices that “directly affect the [wholesale] rate,” California Independent System Operator Corp. v. FERC, 372 F. 3d 395, 403 (emphasis added). That standard is easily met here. Wholesale demand response is all about reducing wholesale rates; so too the rules and practices that determine how those programs operate. That is particularly true here, as the formula for compensating demand response necessarily lowers wholesale electricity prices by displacing higher-priced generation bids. Pp. 14–17. (b) The Rule also does not regulate retail electricity sales in violation of §824(b). A FERC regulation does not run afoul of §824(b)’s proscription just because it affects the quantity or terms of retail sales. Transactions occurring on the wholesale market have natural consequences at the retail level, and so too, of necessity, will FERC’s regulation of those wholesale matters. That is of no legal consequence. See, e.g., Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U. S. 354 –373. When FERC regulates what takes place on the wholesale market, as part of carrying out its charge to improve how that market runs, then no matter the effect on retail rates, §824(b) imposes no bar. Here, every aspect of FERC’s regulatory plan happens exclusively on the wholesale market and governs exclusively that market’s rules. The Commission’s justifications for regulating demand response are likewise only about improving the wholesale market. Cf. Oneok, Inc. v. Learjet, Inc., 575 U. S. ___, ___. Pp. 17–25. (c) In addition, EPSA’s position would subvert the FPA. EPSA’s arguments suggest that the entire practice of wholesale demand response falls outside what FERC can regulate, and EPSA concedes that States also lack that authority. But under the FPA, wholesale demand response programs could not go forward if no entity had jurisdiction to regulate them. That outcome would flout the FPA’s core purposes of protecting “against excessive prices” and ensuring effective transmission of electric power. Pennsylvania Water & Power Co. v. FPC, 343 U. S. 414 ; see Gulf States Util. Co. v. FPC, 411 U. S. 747 . The FPA should not be read, against its clear terms, to halt a practice that so evidently enables FERC to fulfill its statutory duties of holding down prices and enhancing reliability in the wholesale energy market. Pp. 25–29. 2. FERC’s decision to compensate demand response providers at LMP—the same price paid to generators—instead of at LMP-G, is not arbitrary and capricious. Under the narrow scope of review in Motor Vehicle Mfrs. Assn. of United States, Inc. v. State Farm Mut. Automobile Ins. Co., 463 U. S. 29 , this Court’s important but limited role is to ensure that FERC engaged in reasoned decisionmaking—that it weighed competing views, selected a compensation formula with adequate support in the record, and intelligibly explained the reasons for making that decision. Here, FERC provided a detailed explanation of its choice of LMP and responded at length to contrary views. FERC’s serious and careful discussion of the issue satisfies the arbitrary and capricious standard. Pp. 29–33. 753 F. 3d 216, reversed and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined. Scalia, J. filed a dissenting opinion, in which Thomas, J., joined. Alito, J., took no part in the consideration or decision of the cases.Notes 1 Together with No. 14–841, EnerNOC, Inc., et al. v. Electric Power Supply Association et al., also on certiorari to the same court.

The Federal Power Act (FPA or Act), 41Stat. 1063, as amended, 16 U. S. C. §791a et seq., authorizes the Federal Energy Regulatory Commission (FERC or Commission) to regulate “the sale of electric energy at wholesale in interstate commerce,” including both wholesale electricity rates and any rule or practice “affecting” such rates. §§824(b), 824e(a). But the law places beyond FERC’s power, and leaves to the States alone, the regulation of “any other sale”—most notably, any retail sale—of electricity. §824(b). That statutory division generates a steady flow of jurisdictional disputes because—in point of fact if not of law—the wholesale and retail markets in electricity are inextricably linked. These cases concern a practice called “demand response,” in which operators of wholesale markets pay electricity consumers for commitments not to use power at certain times. That practice arose because wholesale market operators can sometimes—say, on a muggy August day—offer electricity both more cheaply and more reliably by paying users to dial down their consumption than by paying power plants to ramp up their production. In the regulation challenged here, FERC required those market operators, in specified circumstances, to compensate the two services equivalently—that is, to pay the same price to demand response providers for conserving energy as to generators for making more of it. Two issues are presented here. First, and fundamen-tally, does the FPA permit FERC to regulate these demand response transactions at all, or does any such rule impinge on the States’ authority? Second, even if FERC has the requisite statutory power, did the Commission fail to justify adequately why demand response providers and electricity producers should receive the same compensation? The court below ruled against FERC on both scores. We disagree. I A Federal regulation of electricity owes its beginnings to one of this Court’s decisions. In the early 20th century, state and local agencies oversaw nearly all generation, transmission, and distribution of electricity. But this Court held in Public Util. Comm’n of R. I. v. Attleboro Steam & Elec. Co., 273 U. S. 83 –90 (1927), that the Commerce Clause bars the States from regulating certain interstate electricity transactions, including wholesale sales (i.e., sales for resale) across state lines. That ruling created what became known as the “Attleboro gap”—a regulatory void which, the Court pointedly noted, only Congress could fill. See id., at 90. Congress responded to that invitation by passing the FPA in 1935. The Act charged FERC’s predecessor agency with undertaking “effective federal regulation of the expanding business of transmitting and selling electric power in interstate commerce.” New York v. FERC, 535 U. S. 1, 6 (2002) (quoting Gulf States Util. Co. v. FPC, 411 U. S. 747, 758 (1973) ). Under the statute, the Commission has authority to regulate “the transmission of electric energy in interstate commerce” and “the sale of electric energy at wholesale in interstate commerce.” 16 U. S. C. §824(b)(1). In particular, the FPA obligates FERC to oversee all prices for those interstate transactions and all rules and practices affecting such prices. The statute provides that “[a]ll rates and charges made, demanded, or received by any public utility for or in connection with” interstate transmissions or wholesale sales—as well as “all rulesand regulations affecting or pertaining to such rates or charges”—must be “just and reasonable.” §824d(a). And if “any rate [or] charge,” or “any rule, regulation, practice, or contract affecting such rate [or] charge[,]” falls short of that standard, the Commission must rectify the problem: It then shall determine what is “just and reasonable” and impose “the same by order.” §824e(a). Alongside those grants of power, however, the Act also limits FERC’s regulatory reach, and thereby maintains a zone of exclusive state jurisdiction. As pertinent here, §824(b)(1)—the same provision that gives FERC author-ity over wholesale sales—states that “this subchapter,” including its delegation to FERC, “shall not apply toany other sale of electric energy.” Accordingly, the Commission may not regulate either within-state wholesale sales or, more pertinent here, retail sales of electricity (i.e., sales directly to users). See New York, 535 U. S., at17, 23. State utility commissions continue to oversee those transactions. Since the FPA’s passage, electricity has increasingly become a competitive interstate business, and FERC’s role has evolved accordingly. Decades ago, state or local utilities controlled their own power plants, transmission lines, and delivery systems, operating as vertically integrated monopolies in confined geographic areas. That is no longer so. Independent power plants now abound, and almostall electricity flows not through “the local power networks of the past,” but instead through an interconnected “grid” of near-nationwide scope. See id., at 7 (“electricity that enters the grid immediately becomes a part of a vast pool of energy that is constantly moving in interstate commerce,” linking producers and users across the country). In this new world, FERC often forgoes the cost-based rate-setting traditionally used to prevent monopolistic pricing. The Commission instead undertakes to ensure “just and reasonable” wholesale rates by enhancing competition—attempting, as we recently explained, “to break down regulatory and economic barriers that hinder a free market in wholesale electricity.” Morgan Stanley Capital Group Inc. v. Public Util. Dist. No. 1 of Snohomish Cty., 554 U. S. 527, 536 (2008) . As part of that effort, FERC encouraged the creation of nonprofit entities to manage wholesale markets on a regional basis. Seven such wholesale market operators now serve areas with roughly two-thirds of the country’s electricity “load” (an industry term for the amount of electricity used). See FERC, Energy Primer: A Handbook of Energy Market Basics 58–59 (Nov. 2015) (EnergyPrimer). Each administers a portion of the grid, providing generators with access to transmission lines and ensuring that the network conducts electricity reliably. See ibid. And still more important for present purposes, each operator conducts a competitive auction to set wholesale prices for electricity. These wholesale auctions serve to balance supply and demand on a continuous basis, producing prices for electricity that reflect its value at given locations and times throughout each day. Such a real-time mechanism is needed because, unlike most products, electricity cannot be stored effectively. Suppliers must generate—every day, hour, and minute—the exact amount of power necessary to meet demand from the utilities and other “load-serving entities” (LSEs) that buy power at wholesale for resale to users. To ensure that happens, wholesale market operators obtain (1) orders from LSEs indicating how much electricity they need at various times and (2) bids from generators specifying how much electricity they can produce at those times and how much they will charge for it. Operators accept the generators’ bids in order of cost (least expensive first) until they satisfy the LSEs’ total demand. The price of the last unit of electricity purchased is then paid to every supplier whose bid was accepted, regardless of its actual offer; and the total cost is split among the LSEs in proportion to how much energy they have ordered. So, for example, suppose that at 9 a.m. on August 15 four plants serving Washington, D. C. can each produce some amount of electricity for, respectively, $10/unit, $20/unit, $30/unit, and $40/unit. And suppose that LSEs’ demand at that time and place is met after the operator accepts the three cheapest bids. The first three generators would then all receive $30/unit. That amount is (think back to Econ 101) the marginal cost—i.e., the added cost of meeting another unit of demand—which is the price an efficient market would produce. See 1 A. Kahn, The Economics of Regulation: Principles and Institutions 65–67 (1988). FERC calls that cost (in jargon that will soon become oddly familiar) the locational marginal price, or LMP.[1] As in any market, when wholesale buyers’ demand for electricity increases, the price they must pay rises correspondingly; and in those times of peak load, the grid’s reliability may also falter. Suppose that by 2 p.m. on August 15, it is 98 degrees in D. C. In every home, store, or office, people are turning the air conditioning up. To keep providing power to their customers, utilities and other LSEs must ask their market operator for more electricity. To meet that spike in demand, the operator will have to accept more expensive bids from suppliers. The operator, that is, will have to agree to the $40 bid that it spurned before—and maybe, beyond that, to bids of $50 or $60 or $70. In such periods, operators often must call on extremely inefficient generators whose high costs of production cause them to sit idle most of the time. See Energy Primer 41–42. As that happens, LMP—the price paid by all LSEs to all suppliers—climbs ever higher. And meanwhile, the increased flow of electricity through the grid threatens to overload transmission lines. See id., at 44. As every consumer knows, it is just when the weather is hottest and the need for air conditioning most acute that blackouts, brownouts, and other service problems tend to occur. Making matters worse, the wholesale electricity market lacks the self-correcting mechanism of other markets. Usually, when the price of a product rises, buyers natu-rally adjust by reducing how much they purchase. But con-sumers of electricity—and therefore the utilities and other LSEs buying power for them at wholesale—do not respond to price signals in that way. To use the economic term, demand for electricity is inelastic. That is in part because electricity is a necessity with few ready substitutes: When the temperature reaches 98 degrees, many people see no option but to switch on the AC. And still more: Many State regulators insulate consumers from short-term fluctuations in wholesale prices by insisting that LSEs set stable retail rates. See id., at 41, 43–44. That, one might say, short-circuits the normal rules of economic behavior. Even in peak periods, as costs surge in the wholesale market, consumers feel no pinch, and so keep running the AC as before. That means, in turn, that LSEs must keep buying power to send to those users—no matter that wholesale prices spiral out of control and increased usage risks overtaxing the grid. But what if there were an alternative to that scenario? Consider what would happen if wholesale market operators could induce consumers to refrain from using (and so LSEs from buying) electricity during peak periods. Whenever doing that costs less than adding more power, an operator could bring electricity supply and demand into balance at a lower price. And simultaneously, the operator could ease pressure on the grid, thus protecting against system failures. That is the idea behind the practice at issue here: Wholesale demand response, as it is called, pays consumers for commitments to curtail their use of power, so as to curb wholesale rates and prevent grid breakdowns. See id., at 44–46.[2] These demand response programs work through the operators’ regular auctions. Aggregators of multiple users of electricity, as well as large-scale individual users like factories or big-box stores, submit bids to decrease electricity consumption by a set amount at a set time for a set price. The wholesale market operators treat those offers just like bids from generators to increase supply. The operators, that is, rank order all the bids—both to produce and to refrain from consuming electricity—from least to most expensive, and then accept the lowest bids until supply and demand come into equipoise. And, once again, the LSEs pick up the cost of all those payments. So, to return to our prior example, if a store submitted an offer not to use a unit of electricity at 2 p.m. on August 15 for $35, the operator would accept that bid before calling on the generator that offered to produce a unit of power for $40. That would result in a lower LMP—again, wholesale market price—than if the market operator could not avail itself of demand response pledges. See ISO/RTO Council, Harnessing the Power of Demand: How ISOs and RTOs Are Integrating Demand Response Into Wholesale Electricity Markets 40–43 (2007) (estimating that, in one market, a demand response program reducing electricity usage by 3% in peak hours would lead to price declines of 6% to 12%). And it would decrease the risk of blackouts and other service problems. Wholesale market operators began using demand response some 15 years ago, soon after they assumed the role of overseeing wholesale electricity sales. Recognizing the value of demand response for both system reliability and efficient pricing, they urged FERC to allow them to implement such programs. See, e.g., PJM Interconnection, L. L. C., Order Accepting Tariff Sheets as Modified, 95 FERC ¶61,306 (2001); California Independent System Operator Corp., Order Conditionally Accepting for Filing Tariff Revisions, 91 FERC ¶61,256 (2000). And as demand response went into effect, market participants of many kinds came to view it—in the words of respondent Electric Power Supply Association (EPSA)—as an “important element[ ] of robust, competitive wholesale electricity markets.” App. 94, EPSA, Comments on Proposed Rule on Demand Response Compensation in Organized Wholesale Energy Markets (May 12, 2010). Congress added to the chorus of voices praising wholesale demand response. In the Energy Policy Act of 2005, 119Stat. 594 (EPAct), it declared as “the policy of the United States” that such demand response “shall be encouraged.” §1252(f), 119Stat. 966, 16 U. S. C. §2642 note. In particular, Congress directed, the deployment of “technology and devices that enable electricity customers to participate in . . . demand response systems shall be facilitated, and unnecessary barriers to demand response participation in energy . . . markets shall be eliminated.” Ibid.[3] B Spurred on by Congress, the Commission determined to take a more active role in promoting wholesale demand response programs. In 2008, FERC issued Order No. 719, which (among other things) requires wholesale market operators to receive demand response bids from aggregators of electricity consumers, except when the state regulatory authority overseeing those users’ retail purchases bars such demand response participation. See 73 Fed. Reg. 64119, ¶154 (codified 18 CFR §35.28(g)(1) (2015)). That original order allowed operators to compensate demand response providers differently from generators if they so chose. No party sought judicial review. Concerned that Order No. 719 had not gone far enough, FERC issued the rule under review here in 2011, with one commissioner dissenting. See Demand Response Competition in Organized Wholesale Energy Markets, Order No. 745, 76 Fed. Reg. 16658 (Rule) (codified 18 CFR §35.28(g)(1)(v)). The Rule attempts to ensure “just and reasonable” wholesale rates by requiring market operators to appropriately compensate demand response providers and thus bring about “meaningful demand-side participation” in the wholesale markets. 76 Fed. Reg. 16658, ¶1, 16660, ¶10; 16 U. S. C. §824d(a). The Rule’s most significant provision directs operators, under two specified conditions, to pay LMP for any accepted demand response bid, just as they do for successful supply bids. See 76 Fed. Reg. 16666–16669, ¶¶45–67. In other words, the Rule requires that demand response providers in those circumstances receive as much for conserving electricity as generators do for producing it. The two specified conditions ensure that a bid to use less electricity provides the same value to the wholesale market as a bid to make more. First, a demand response bidder must have “the capability to provide the service” offered; it must, that is, actually be able to reduce electricity use and thereby obviate the operator’s need to secure additional power. Id., at 16666, ¶¶48–49. Second, paying LMP for a demand response bid “must be cost-effective,” as measured by a standard called “the net benefits test.” Ibid., ¶48. That test makes certain that accepting a lower-priced demand response bid over a higher-priced supply bid will actually save LSEs (i.e., wholesale purchasers) money. In some situations it will not, even though accepting a lower-priced bid (by definition) reduces LMP. That is because (to oversimplify a bit) LSEs share the cost of paying successful bidders, and reduced electricity use makes some LSEs drop out of the market, placing a proportionally greater burden on those that are left. Each remaining LSE may thus wind up paying more even though the total bill is lower; or said otherwise, the costs associated with an LSE’s increased share of compensating bids may exceed the savings that the LSE obtains from a lower wholesale price.[4] The net benefits test screens out such counterproductive demand response bids, exempting them from the Rule’s compensation requirement. See id., at 16659, 16666–16667, ¶¶3, 50–53. What remains are only those offers whose acceptance will result in actual savings to wholesale purchasers (along with more reliable service to end users). See id., at 16671, ¶¶78–80. The Rule rejected an alternative scheme for compensating demand response bids. Several commenters had urged that, in paying a demand response provider, an operator should subtract from the ordinary wholesale price the savings that the provider nets by not buying electricity on the retail market. Otherwise, the commenters claimed, demand response providers would receive a kind of“double-payment” relative to generators. See id., at 16663, ¶24. That proposal, which the dissenting commissioner largely accepted, became known as LMP minus G, or more simply LMP-G, where “G” stands for the retail price of electricity. See id., at 16668, ¶60, 16680 (Moeller, dissenting). But FERC explained that, under the conditions it had specified, the value of an accepted demand response bid to the wholesale market is identical to that of an accepted supply bid because each succeeds in cost-effectively “balanc[ing] supply and demand.” Id., at 16667, ¶55. And, the Commission reasoned, that comparable value is what ought to matter given FERC’s goal of strengthening competition in the wholesale market: Rates should reflect not the costs that each market participant incurs, but instead the services it provides. See id., at 16668, ¶62. Moreover, the Rule stated, compensating demand response bids at their actual value—i.e., LMP—will help overcome various technological barriers, including a lack of needed infrastructure, that impede aggregators and large-scale users of electricity from fully participating in demand response programs. See id., at 16667–16668, ¶¶57–58. The Rule also responded to comments challenging FERC’s statutory authority to regulate the compensation operators pay for demand response bids. Pointing to the Commission’s analysis in Order No. 719, the Rule explained that the FPA gives FERC jurisdiction over such bids because they “directly affect[ ] wholesale rates.” Id., at 16676, ¶112 (citing 74 id., at 37783, ¶47, and 18 U. S. C. §824d). Nonetheless, the Rule noted, FERC would continue Order No. 719’s policy of allowing any state regulatory body to prohibit consumers in its retail market from taking part in wholesale demand response programs. See 76 Fed. Reg. 16676, ¶114; 73 id., at 64119, ¶154. Accordingly, the Rule does not require any “action[ ] that would violate State laws or regulations.” 76 id., at 16676, ¶114. C A divided panel of the Court of Appeals for the District of Columbia Circuit vacated the Rule as “ultra vires agency action.” 753 F. 3d 216, 225 (2014). The court held that FERC lacked authority to issue the Rule even though “demand response compensation affects the wholesale market.” Id., at 221. The Commission’s “jurisdiction to regulate practices ‘affecting’ rates,” the court stated, “does not erase the specific limit[ ]” that the FPA imposes on FERC’s regulation of retail sales. Id., at 222. And the Rule, the court concluded, exceeds that limit: In “luring . . . retail customers” into the wholesale market, and causing them to decrease “levels of retail electricity consumption,” the Rule engages in “direct regulation of the retail market.” Id., at 223–224. The Court of Appeals held, alternatively, that the Rule is arbitrary and capricious under the Administrative Procedure Act, 5 U. S. C. §706(2)(A), because FERC failed to “adequately explain[ ]” why paying LMP to demand response providers “results in just compensation.” 753 F. 3d, at 225. According to the court, FERC did not “properly consider” the view that such a payment would give those providers a windfall by leaving them with “the full LMP plus . . . the savings associated with” reduced consumption. Ibid. (quoting Demand Response Competition in Organized Wholesale Energy Markets: Order on Rehearing and Clarification, Order No. 745–A (Rehearing Order), 137 FERC ¶61,215, p. 62,316 (2011) (Moeller, dissenting)). The court dismissed out of hand the idea that “comparable contributions [could] be the reason for equal compensation.” 753 F. 3d, at 225. Judge Edwards dissented. He explained that the rules governing wholesale demand response have a “direct effect . . . on wholesale electricity rates squarely within FERC’s jurisdiction.” Id., at 227. And in setting those rules, he argued, FERC did not engage in “direct regulation of the retail market”; rather, “[a]uthority over retail rates . . . remains vested solely in the States.” Id., at 234 (internal quotation marks omitted). Finally, Judge Edwards rejected the majority’s view that the Rule is arbitrary and capricious. He noted the substantial deference due to the Commission in cases involving ratemaking, and concluded that FERC provided a “thorough” and “reasonable” explanation for choosing LMP as the appropriate compensation formula. Id., at 236–238. We granted certiorari, 575 U. S. ___ (2015), to decide whether the Commission has statutory authority to regulate wholesale market operators’ compensation of demand response bids and, if so, whether the Rule challenged here is arbitrary and capricious. We now hold that the Commission has such power and that the Rule is adequately reasoned. We accordingly reverse. II Our analysis of FERC’s regulatory authority proceeds in three parts. First, the practices at issue in the Rule—market operators’ payments for demand response commitments—directly affect wholesale rates. Second, in addressing those practices, the Commission has not regulated retail sales. Taken together, those conclusions establish that the Rule complies with the FPA’s plain terms. And third, the contrary view would conflict with the Act’s core purposes by preventing all use of a tool that no one (not even EPSA) disputes will curb prices and enhance reliability in the wholesale electricity market.[5] A The FPA delegates responsibility to FERC to regulate the interstate wholesale market for electricity—both wholesale rates and the panoply of rules and practices affecting them. As noted earlier, the Act establishesa scheme for federal regulation of “the sale of electric energy at wholesale in interstate commerce.” 16 U. S. C. §824(b)(1); see supra, at 3. Under the statute, “[a]ll rates and charges made, demanded, or received by any public utility for or in connection with” interstate wholesale sales “shall be just and reasonable”; so too shall “all rules and regulations affecting or pertaining to such rates orcharges.” §824d(a). And if FERC sees a violation of that standard, it must take remedial action. More specifically, whenever the Commission “shall find that any rate [or] charge”—or “any rule, regulation, practice, or contract affecting such rate [or] charge”—is “unjust [or] unreasonable,” then the Commission “shall determine the just and reasonable rate, charge[,] rule, regulation, practice or contract” and impose “the same by order.” §824e(a). That means FERC has the authority—and, indeed, the duty—to ensure that rules or practices “affecting” wholesale rates are just and reasonable. Taken for all it is worth, that statutory grant could extend FERC’s power to some surprising places. As the court below noted, markets in all electricity’s inputs—steel, fuel, and labor most prominent among them—might affect generators’ supply of power. See 753 F. 3d, at 221; id., at 235 (Edwards, J., dissenting). And for that matter, markets in just about everything—the whole economy, as it were—might influence LSEs’ demand. So if indirect or tangential impacts on wholesale electricity rates sufficed, FERC could regulate now in one industry, now in another, changing a vast array of rules and practices to implement its vision of reasonableness and justice. We cannot imagine that was what Congress had in mind. For that reason, an earlier D. C. Circuit decision adopted, and we now approve, a common-sense construction ofthe FPA’s language, limiting FERC’s “affecting” jurisdiction to rules or practices that “directly affect the [wholesale] rate.” California Independent System Operator Corp. v. FERC, 372 F. 3d 395, 403 (2004) (emphasis added); see 753 F. 3d, at 235 (Edwards, J., dissenting). As we have explained in addressing similar terms like “relating to” or “in connection with,” a non-hyperliteral reading is needed to prevent the statute from assuming near-infinite breadth. See New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U. S. 645, 655 (1995) (“If ‘relate to’ were taken to extend to the furthest stretch of its indeterminacy, then for all practical purposes [the statute] would never run its course”); Maracich v. Spears, 570 U. S. ___, ___ (2013) (slip op., at 9) (“The phrase ‘in connection with’ is essentially indeterminat[e] because connections, like relations, stop nowhere” (internal quotation marks omitted)). The Commission itself incorporated the D. C. Circuit’s standard in addressing its authority to issue the Rule. See 76 Fed. Reg. 16676, ¶112 (stating that FERC has jurisdiction because wholesale demand response “directly affects wholesale rates”). We think it right to do the same. Still, the rules governing wholesale demand response programs meet that standard with room to spare. In general (and as earlier described), wholesale market operators employ demand response bids in competitive auctions that balance wholesale supply and demand and thereby set wholesale prices. See supra, at 7–8. The operators accept such bids if and only if they bring down the wholesale rate by displacing higher-priced generation. And when that occurs (most often in peak periods), the easing of pressure on the grid, and the avoidance of service problems, further contributes to lower charges. See Brief for Grid Engineers et al. as Amici Curiae 26–27. Wholesale demand response, in short, is all about reducing wholesale rates; so too, then, the rules and practices that determine how those programs operate. And that is particularly true of the formula that operators use to compensate demand response providers. As in other areas of life, greater pay leads to greater participation. If rewarded at LMP, rather than at some lesser amount, more demand response providers will enter more bids capable of displacing generation, thus necessarily lowering wholesale electricity prices. Further, the Commission found, heightened demand response participation will put “downward pressure” on generators’ own bids, encouraging power plants to offer their product at reduced prices lest they come away empty-handed from the bidding process. 76 Fed. Reg. 16660, ¶10. That, too, ratchets down the rates wholesale purchasers pay. Compensation for demand response thus directly affects wholesale prices. Indeed, it is hard to think of a practice that does so more. B The above conclusion does not end our inquiry into the Commission’s statutory authority; to uphold the Rule, we also must determine that it does not regulate retail electricity sales. That is because, as earlier described, §824(b) “limit[s] FERC’s sale jurisdiction to that at wholesale,” reserving regulatory authority over retail sales (as well as intrastate wholesale sales) to the States. New York, 535 U. S., at 17 (emphasis deleted); see 16 U. S. C. §824(b); supra, at 3.[6] FERC cannot take an action transgressing that limit no matter how direct, or dramatic, its impact on wholesale rates. Suppose, to take a far-fetched example, that the Commission issued a regulation compelling every consumer to buy a certain amount of electricity on the retail market. Such a rule would necessarily determine the load purchased on the wholesale market too, and thus would alter wholesale prices. But even given that ineluctable consequence, the regulation would exceed FERC’s authority, as defined in §824(b), because it specifies terms of sale at retail—which is a job for the States alone.[7] Yet a FERC regulation does not run afoul of §824(b)’s proscription just because it affects—even substantially—the quantity or terms of retail sales. It is a fact of eco-nomic life that the wholesale and retail markets in electricity, as in every other known product, are not hermetically sealed from each other. To the contrary, transactions that occur on the wholesale market have natural consequences at the retail level. And so too, of necessity, will FERC’s regulation of those wholesale matters. Cf. Oneok, Inc. v. Learjet, Inc., 575 U. S. ___, ___ (2015) (slip op., at 13) (noting that in the similarly structured world of natural gas regulation, a “Platonic ideal” of strict separation between federal and state realms cannot exist). When FERC sets a wholesale rate, when it changes wholesale market rules, when it allocates electricity as between wholesale purchasers—in short, when it takes virtually any action respecting wholesale transactions—it has some effect, in either the short or the long term, on retail rates. That is of no legal consequence. See, e.g., Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U. S. 354 –373 (1988) (holding that an order regulating wholesale purchases fell within FERC’s jurisdiction, and preempted contrary state action, even though it clearly affected retail prices); Nantahala Power & Light Co. v. Thornburg, 476 U. S. 953 –961, 970 (1986) (same); FPC v. Louisiana Power & Light Co., 406 U. S. 621 –641 (1972) (holding similarly in the natural gas context). When FERC regulates what takes place on the wholesale market, as part of carrying out its charge to improve how that market runs, then no matter the effect on retail rates, §824(b) imposes no bar. And in setting rules for demand response, that is all FERC has done. The Commission’s Rule addresses—and addresses only—transactions occurring on the wholesale market. Recall once again how demand response works—and forgive the coming italics. See supra, at 7–8. Wholesale market operators administer the entire program, receiving every demand response bid made. Those operators accept such a bid at the mandated price when (and only when) the bid provides value to the wholesale market by balancing supply and demand more “cost-effective[ly]”—i.e., at a lower cost to wholesale purchasers—than a bid to generate power. 76 Fed. Reg. 16659, 16666, ¶2, 48. The compensation paid for a successful bid (LMP) is whatever the operator’s auction has determined is the marginal price of wholesale electricity at a particular location and time. See id., at 16659, ¶2. And those footing the bill are the same wholesale purchasers that have benefited from the lower wholesale price demand response participation has produced. See id., at 16674, ¶¶99–100. In sum, whatever the effects at the retail level, every aspect of the regulatory plan happens exclusively on the wholesale market and governs exclusively that market’s rules. What is more, the Commission’s justifications for regulating demand response are all about, and only about, improving the wholesale market. Cf. Oneok, 575 U. S., at ___ (slip op., at 11) (considering “the target at which [a] law aims” in determining whether a State is properly regulating retail or, instead, improperly regulating wholesale sales). In Order No. 719, FERC explained that demand response participation could help create a “well-functioning competitive wholesale electric energy market” with “reduce[d] wholesale power prices” and “enhance[d] reliability.” 73 Fed. Reg. 64103, ¶16. And in the Rule under review, FERC expanded on that theme. It listed the several ways in which “demand response in organized wholesale energy markets can help improve the functioning and competitiveness of those markets”: by replacing high-priced, inefficient generation; exerting “downward pressure” on “generator bidding strategies”; and “support[ing] system reliability.” 76 id., at 16660, ¶10; see Notice of Proposed Rulemaking for Order No. 745, 75 id., at 15363–15364, ¶4 (2010) (noting similar aims); supra, at 7–8. FERC, that is, focused wholly on the benefits that demand response participation (in the wholesale market) could bring to the wholesale market. The retail market figures no more in the Rule’s goals than in the mechanism through which the Rule operates. EPSA’s primary argument that FERC has usurped state power (echoed in the dissent) maintains that the Rule “effectively,” even though not “nominal[ly],” regulates retail prices. See, e.g., Brief for Respondents 1, 10, 23–27, 35–39; Tr. of Oral Arg. 26, 30; post, at 4–6. The argument begins on universally accepted ground: Under §824(b), only the States, not FERC, can set retail rates. See, e.g., FPC v. Conway Corp., 426 U. S. 271, 276 (1976) . But as EPSA concedes, that tenet alone cannot make its case, because FERC’s Rule does not set actual rates: States continue to make or approve all retail rates, and in doing so may insulate them from price fluctuations in the wholesale market. See Brief for Respondents 39. Still, EPSA contends, rudimentary economic analysis shows that the Rule does the “functional equivalen[t]” of setting—more particularly, of raising—retail rates. Id., at 36. That is because the opportunity to make demand response bids in the wholesale market changes consumers’ calculations. In deciding whether to buy electricity at retail, economically-minded consumers now consider both the cost of making such a purchase and the cost of forgoing a possible demand response payment. So, EPSA explains, if a factory can buy electricity for $10/unit, but can earn $5/unit for not buying power at peak times, then the effective retail rate at those times is $15/unit: the $10 the factory paid at retail plus the $5 it passed up. See id., at 10. And by thus increasing effective retail rates, EPSA concludes, FERC trespasses on the States’ ground. The modifier “effective” is doing quite a lot of work in that argument—more work than any conventional understanding of rate-setting allows. The standard dictionary definition of the term “rate” (as used with reference to prices) is “[a]n amount paid or charged for a good or service.” Black’s Law Dictionary 1452 (10th ed. 2014); see, e.g., 13 Oxford English Dictionary 208–209 (2d ed. 1989) (“rate” means “price,” “cost,” or “sum paid or asked for a . . . thing”). To set a retail electricity rate is thus to establish the amount of money a consumer will hand over in exchange for power. Nothing in §824(b) or any other part of the FPA suggests a more expansive notion, in which FERC sets a rate for electricity merely by altering consumers’ incentives to purchase that product.[8] And neither does anything in this Court’s caselaw. Our decisions uniformly speak about rates, for electricity and all else, in only their most prosaic, garden-variety sense. As the Solicitor General summarized that view, “the rate is what it is.” Tr. of Oral Arg. 7. It is the price paid, not the price paid plus the cost of a forgone economic opportunity. Consider a familiar scenario to see what is odd about EPSA’s theory. Imagine that a flight is overbooked. The airline offers passengers $300 to move to a later plane that has extra seats. On EPSA’s view, that offer adds $300—the cost of not accepting the airline’s proffered payment—to the price of every continuing passenger’s ticket. So a person who originally spent $400 for his ticket, and decides to reject the airline’s proposal, paid an “effective” price of $700. But would any passenger getting off the plane say he had paid $700 to fly? That is highly unlikely. And airline lawyers and regulators (including many, we are sure, with economics Ph. D.’s) appear to share that common-sensical view. It is in fact illegal to “increase the price” of “air transportation . . . after [such] air transportation has been purchased by the consumer.” 14 CFR §399.88(a) (2015). But it is a safe bet that no airline has ever gotten into trouble by offering a payment not to fly.[9] And EPSA’s “effective price increase” claim fares even worse when it comes to payments not to use electricity. In EPSA’s universe, a wholesale demand response program raises retail rates by compelling consumers to “pay” the price of forgoing demand response compensation. But such a consumer would be even more surprised than our air traveler to learn of that price hike, because the natural consequence of wholesale demand response programs is to bring down retail rates. Once again, wholesale market operators accept demand response bids only if those offers lower the wholesale price. See supra, at 7–8. And when wholesale prices go down, retail prices tend to follow, because state regulators can, and mostly do, insist that wholesale buyers eventually pass on their savings to consumers. EPSA’s theoretical construct thus runs headlong into the real world of electricity sales—where the Rule does anything but increase retail prices. EPSA’s second argument that FERC intruded into the States’ sphere is more historical and purposive in nature. According to EPSA, FERC deliberately “lured [retail customers] into the[ ] wholesale markets”—and, more, FERC did so “only because [it was] dissatisfied with the States’ exercise of their undoubted authority” under §824(b) to regulate retail sales. Brief for Respondents 23; see id., at 2–3, 31, 34. In particular, EPSA asserts, FERC disapproved of “many States’ continued preference” for stable pricing—that is, for insulating retail rates from short-term fluctuations in wholesale costs. Id., at 28. In promoting demand response programs—or, in EPSA’s somewhat less neutral language, in “forc[ing] retail customers to respond to wholesale price signals”—FERC acted “for the express purpose of overriding” that state policy. Id., at 29, 49. That claim initially founders on the true facts of how wholesale demand response came about. Contra EPSA, the Commission did not invent the practice. Rather, and as described earlier, the impetus came from wholesale market operators. See supra, at 8. In designing their newly organized markets, those operators recognized almost at once that demand response would lower wholesale electricity prices and improve the grid’s reliability. So they quickly sought, and obtained, FERC’s approval to institute such programs. Demand response, then, emerged not as a Commission power grab, but instead as a market-generated innovation for more optimally balancing wholesale electricity supply and demand. And when, years later (after Congress, too, endorsed the practice), FERC began to play a more proactive role, it did so for the identical reason: to enhance the wholesale, not retail, electricity market. Like the market operators, FERC saw that sky-high demand in peak periods threatened network breakdowns, compelled purchases from inefficient generators, and consequently drove up wholesale prices. See, e.g., 73 Fed. Reg. 64103, ¶16; 76 id., at 16660, ¶10; see supra, at 6–7. Addressing those problems—which demand response does—falls within the sweet spot of FERC’s statutory charge. So FERC took action promoting the practice. No doubt FERC recognized connections, running in both directions, between the States’ policies and its own. The Commission understood that by insulating consumers from price fluctuations, States contributed to the wholesale market’s difficulties in optimally balancing supply and demand. See 76 Fed. Reg. 16667–16668, ¶¶57, 59; supra, at 6–7. And FERC realized that increased use of demand response in that market would (by definition) inhibit retail sales otherwise subject to State control. See 73 Fed. Reg. 64167. But nothing supports EPSA’s more feverish idea that the Commission’s interest in wholesale demand response emerged from a yen to usurp State authority over, or impose its own regulatory agenda on, retail sales. In promoting demand response, FERC did no more than follow the dictates of its regulatory mission to improve the competitiveness, efficiency, and reliability of the wholesale market. Indeed, the finishing blow to both of EPSA’s arguments comes from FERC’s notable solicitude toward the States. As explained earlier, the Rule allows any State regulator to prohibit its consumers from making demand response bids in the wholesale market. See 76 id., at 16676, ¶114; 73 id., at 64119, ¶154; supra, at 12. Although claiming the ability to negate such state decisions, the Commission chose not to do so in recognition of the linkage between wholesale and retail markets and the States’ role in overseeing retail sales. See 76 Fed. Reg. 16676, ¶¶112–114. The veto power thus granted to the States belies EPSA’s view that FERC aimed to “obliterate[ ]” their regulatory authority or “override” their pricing policies. Brief for Respondents 29, 33. And that veto gives States the means to block whatever “effective” increases in retail rates demand response programs might be thought to produce. Wholesale demand response as implemented in the Rule is a program of cooperative federalism, in which the States retain the last word. That feature of the Rule removes any conceivable doubt as to its compliance with §824(b)’s allocation of federal and state authority. C One last point, about how EPSA’s position would subvert the FPA. EPSA’s jurisdictional claim, as may be clear by now, stretches very far. Its point is not that this single Rule, relating to compensation levels, exceeds FERC’s power. Instead, EPSA’s arguments—that rewarding energy conservation raises effective retail rates and that “luring” consumers onto wholesale markets aims to disrupt state policies—suggest that the entire practice of wholesale demand response falls outside what FERC can regulate. EPSA proudly embraces that point: FERC, it declares, “has no business regulating ‘demand response’ at all.” Id., at 24. Under EPSA’s theory, FERC’s earlier Order No. 719, although never challenged, would also be ultra vires because it requires operators to open their markets to demand response bids. And more: FERC could not even approve an operator’s voluntary plan to administer a demand response program. See Tr. of Oral Arg. 44. That too would improperly allow a retail customer to participate in a wholesale market. Yet state commissions could not regulate demand response bids either. EPSA essentially concedes this point. See Brief for Respondents 46 (“That may well be true”). And so it must. The FPA “leaves no room either for direct state regulation of the prices of interstate wholesales” or for regulation that “would indirectly achieve the same result.” Northern Natural Gas Co. v. State Corporation Comm’n of Kan., 372 U. S. 84, 91 (1963) . A State could not oversee offers, made in a wholesale market operator’s auction, that help to set wholesale prices. Any effort of that kind would be preempted. And all of that creates a problem. If neither FERC nor the States can regulate wholesale demand response, then by definition no one can. But under the Act, no electricity transaction can proceed unless it is regulable by someone. As earlier described, Congress passed the FPA precisely to eliminate vacuums of authority over the electricity markets. See supra, at 2–3. The Act makes federal and state powers “complementary” and “comprehensive,” so that “there [will] be no ‘gaps’ for private interests to subvert the public welfare.” Louisiana Power & Light Co., 406 U. S., at 631. Or said otherwise, the statute prevents the creation of any regulatory “no man’s land.” FPC v. Transcontinental Gas Pipe Line Corp., 365 U. S. 1, 19 (1961) ; see id., at 28. Some entity must have jurisdiction to regulate each and every practice that takes place in the electricity markets, demand response no less than any other.[10] For that reason, the upshot of EPSA’s view would be to extinguish the wholesale demand response program in its entirety. Under the FPA, each market operator must submit to FERC all its proposed rules and procedures. See 16 U. S. C. §§824d(c)–(d); 18 CFR §§35.28(c)(4), 35.3(a)(1). Assume that, as EPSA argues, FERC could not authorize any demand response program as part of that package. Nor could FERC simply allow such plans to go into effect without its consideration and approval. There are no “off the books” programs in the wholesale electricity markets—because, once again, there is no regulatory “no man’s land.” Transcontinental, 365 U. S., at 19. The FPA mandates that FERC review, and ensure the reasonableness of, every wholesale rule and practice. See 16 U. S. C. §§824d(a), 824e(a); supra, at 3, 14–15. If FERC could not carry out that duty for demand response, then those programs could not go forward. And that outcome would flout the FPA’s core objects. The statute aims to protect “against excessive prices” and ensure effective transmission of electric power. Pennsylvania Water & Power Co. v. FPC, 343 U. S. 414, 418 (1952) ; see Gulf States Util. Co. v. FPC, 411 U. S. 747, 758 (1973) . As shown above, FERC has amply explained how wholesale demand response helps to achieve those ends, by bringing down costs and preventing service interruptions in peak periods. See supra, at 20. No one taking part in the rulemaking process—not even EPSA—seriously challenged that account. Even as he objected to FERC’s compensation formula, Commissioner Moeller noted the unanimity of opinion as to demand response’s value: “[N]owhere did I review any comment or hear any testimony that questioned the benefit of having demand response resources participate in the organized wholesale energy markets. On this point, there is no debate.” 76 Fed. Reg. 16679; see also App. 82, EPSA, Comments on Proposed Rule (avowing “full[ ] support” for demand response participation in wholesale markets because of its “economic and operational” benefits).[11] Congress itself agreed, “encourag[ing]” greater use of demand response participation at the wholesale level. EPAct §1252(f ), 119Stat. 966. That undisputed judgment extinguishes any last flicker of life in EPSA’s argument. We will not read the FPA, against its clear terms, to halt a practice that so evidently enables the Commission to fulfill its statutory duties of holding down prices and enhancing reliability in the wholesale energy market. III These cases present a second, narrower question: Is FERC’s decision to compensate demand response providers at LMP—the same price paid to generators—arbitrary and capricious? Recall here the basic issue. See supra, at 9–12. Wholesale market operators pay a single price—LMP—for all successful bids to supply electricity at a given time and place. The Rule orders operators to pay the identical price for a successful bid to conserve electric-ity so long as that bid can satisfy a “net benefits test”—meaning that it is sure to bring down costs for wholesale purchasers. In mandating that payment, FERC rejected an alternative proposal under which demand response providers would receive LMP minus G (LMP-G), where G is the retail rate for electricity. According to EPSA and others favoring that approach, demand response providers get a windfall—a kind of “double-payment”—unless market operators subtract the savings associated with conserving electricity from the ordinary compensation level. 76 Fed. Reg. 16663, ¶24. EPSA now claims that FERC failed to adequately justify its choice of LMP rather than LMP-G. In reviewing that decision, we may not substitute our own judgment for that of the Commission. The “scope of review under the ‘arbitrary and capricious’ standard is narrow.” Motor Vehicle Mfrs. Assn. of United States, Inc. v. State Farm Mut. Automobile Ins. Co., 463 U. S. 29, 43 (1983) . A court is not to ask whether a regulatory decision is the best one possible or even whether it is better than the alternatives. Rather, the court must uphold a rule if the agency has “examine[d] the relevant [considerations] and articulate[d] a satisfactory explanation for its action[,] including a rational connection between the facts found and the choice made.” Ibid. (internal quotation marks omitted). And nowhere is that more true than in a technical area like electricity rate design: “[W]e afford great deference to the Commission in its rate decisions.” Morgan Stanley, 554 U. S., at 532. Here, the Commission gave a detailed explanation of its choice of LMP. See 76 Fed. Reg. 16661–16669, ¶¶18–67. Relying on an eminent regulatory economist’s views, FERC chiefly reasoned that demand response bids should get the same compensation as generators’ bids because both provide the same value to a wholesale market. See id., at 16662–16664, 16667–16668, ¶¶20, 31, 57, 61; see also App. 829–851, Reply Affidavit of Dr. Alfred E. Kahn (Aug. 30, 2010) (Kahn Affidavit). FERC noted that a market operator needs to constantly balance supply and demand, and that either kind of bid can perform that service cost-effectively—i.e., in a way that lowers costs for wholesale purchasers. See 76 Fed. Reg. 16667–16668, ¶¶56, 61. A compensation system, FERC concluded, therefore should place the two kinds of bids “on a competitive par.” Id., at 16668, ¶61 (quoting Kahn Affidavit); see also App. 830, Kahn Affidavit (stating that “economic efficiency requires” compensating the two equally given their equivalent function in a “competitive power market[ ]”). With both supply and demand response available on equal terms, the operator will select whichever bids, of whichever kind, provide the needed electricity at the lowest possible cost. See Rehearing Order, 137 FERC, at 62,301–62,302, ¶68 (“By ensuring that both . . . receive the same compensation for the same service, we expect the Final Rule to enhance the competitiveness” of wholesale markets and “result in just and reasonable rates”). That rationale received added support from FERC’s adoption of the net benefits test. The Commission realized during its rulemaking that in some circumstances a demand response bid—despite reducing the wholesale rate—does not provide the same value as generation. See 76 Fed. Reg. 16664–16665, ¶38. As described earlier, that happens when the distinctive costs associated with compensating a demand response bid exceed the savings from a lower wholesale rate: The purchaser then winds up paying more than if the operator had accepted the best (even though higher priced) supply bid available. See supra, at 10–11. And so FERC developed the net benefits test to filter out such cases. See 76 Fed. Reg. 16666–16667, ¶¶50–53. With that standard in place, LMP is paid only to demand response bids that benefit wholesale purchasers—in other words, to those that function as “cost-effective alternative[s] to the next highest-bid generation.” Id., at 16667, ¶54. Thus, under the Commission’s approach, a demand response provider will receive the same compensation as a generator only when it is in fact providing the same service to the wholesale market. See ibid., ¶53. The Commission responded at length to EPSA’s con-trary view that paying LMP, even in that situation, will overcompensate demand response providers because they are also “effectively receiv[ing] ‘G,’ the retail rate that they do not need to pay.” Id., at 16668, ¶60. FERC explained that compensation ordinarily reflects only the value of the service an entity provides—not the costs it incurs, or benefits it obtains, in the process. So when a generator presents a bid, “the Commission does not inquire into the costs or benefits of production.” Ibid., ¶62. Different power plants have different cost structures. And, indeed, some plants receive tax credits and similar incentive payments for their activities, while others do not. See Rehearing Order, 137 FERC, at 62,301, ¶65, and n. 122. But the Commission had long since decided that such matters are irrelevant: Paying LMP to all generators—although some then walk away with more profit and some with less—“encourages more efficient supply and demand decisions.” 76 Fed. Reg. 16668, ¶62 (internal quotation marks omitted). And the Commission could see no economic reason to treat demand response providers any differently. Like generators, they too experience a range of benefits and costs—both the benefits of not paying for electricity and the costs of not using it at a certain time. But, FERC again concluded, that is immaterial: To increase competition and optimally balance supply and demand, market operators should compensate demand response providers, like generators, based on their contribution to the wholesale system. See ibid.; 137 FERC, at 62,300, ¶60. Moreover, FERC found, paying LMP will help demand response providers overcome certain barriers to participation in the wholesale market. See 76 Fed. Reg. 16667–16668, ¶¶57–59. Commenters had detailed significant start-up expenses associated with demand response, including the cost of installing necessary metering technol-ogy and energy management systems. See id., at 16661, ¶18, 16667–16668, ¶57; see also, e.g., App. 356, Viridity Energy, Inc., Comments on Proposed Rule on Demand Response Compensation in Organized Wholesale Energy Markets (May 13, 2010) (noting the “capital investments and operational changes needed” for demand response participation). The Commission agreed that such factors inhibit potential demand responders from competing with generators in the wholesale markets. See 76 Fed. Reg. 16668, ¶59. It concluded that rewarding demand response at LMP (which is, in any event, the price reflecting its value to the market) will encourage that competition and, in turn, bring down wholesale prices. See ibid. Finally, the Commission noted that determining the “G” in the formula LMP-G is easier proposed than accomplished. See ibid., ¶63. Retail rates vary across and even within States, and change over time as well. Accordingly, FERC concluded, requiring market operators to incorporate G into their prices, “even though perhaps feasible,” would “create practical difficulties.” Ibid. Better, then, not to impose that administrative burden. All of that together is enough. The Commission, not this or any other court, regulates electricity rates. The dis-puted question here involves both technical understanding and policy judgment. The Commission addressed that issue seriously and carefully, providing reasons in support of its position and responding to the principal alternative advanced. In upholding that action, we do not discount the cogency of EPSA’s arguments in favor of LMP-G. Nor do we say that in opting for LMP instead, FERC made the better call. It is not our job to render that judgment, on which reasonable minds can differ. Our important but limited role is to ensure that the Commission engaged in reasoned decisionmaking—that it weighed competing views, selected a compensation formula with adequate support in the record, and intelligibly explained thereasons for making that choice. FERC satisfied that standard. IV FERC’s statutory authority extends to the Rule at issue here addressing wholesale demand response. The Rule governs a practice directly affecting wholesale electricity rates. And although (inevitably) influencing the retail market too, the Rule does not intrude on the States’ power to regulate retail sales. FERC set the terms of transactions occurring in the organized wholesale markets, so as to ensure the reasonableness of wholesale prices and the reliability of the interstate grid—just as the FPA contemplates. And in choosing a compensation formula, the Commission met its duty of reasoned judgment. FERC took full account of the alternative policies proposed, and adequately supported and explained its decision. Accordingly, we reverse the judgment of the Court of Appeals for the District of Columbia Circuit and remand the cases for further proceedings consistent with this opinion. It is so ordered. Justice Alito took no part in the consideration or decision of these cases.Notes 1 To be more precise, LMP generally includes, in addition to the price of the highest-accepted bid, certain costs of moving power through the grid. But those costs are not relevant here, and we therefore disregard them. 2 Differently designed demand response programs can operate in retail markets. Some States, for example, either encourage or require utilities to offer “critical-peak rebates” to customers for curtailing electricity use at times of high load. See Energy Primer 45. 3 The dissent misreads this subsection of the EPAct in suggesting that it encourages States’ use of retail demand response, rather than the wholesale programs at issue here. See post, at 8–9 (opinion of Scalia, J.); n. 2, supra. The prior subsection, §1252(e), as the dissent notes, promotes demand response in the States—but then the EPAct switches gears. Subsection (f) expressly addresses the programs of “regional electricity entit[ies]”—that is, wholesale market operators. Indeed, the provision lists all the markets those operators run: not just the electricity market involved here, but also the “capacity and ancillary service markets.” Those are established components of the wholesale system with no counterparts at the state level. See Energy Primer 59. 4 The explanation is a stylized version of the actual phenomenon. In reality, LSEs rarely drop out of the market entirely because of demand response; instead, they will merely order less electricity. But the effect is the same as in the text, because the total cost of accepted bids is spread among LSEs in proportion to the units of electricity they purchase; and as those units decline, each remaining one bears a greater share of the bill. 5 Because we think FERC’s authority clear, we need not address the Government’s alternative contention that FERC’s interpretation of the statute is entitled to deference under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) . 6 EPSA additionally cites §824(a) as constraining the Commission’s authority, see Brief for Respondent EPSA et al. 25, 31, 43 (Brief for Respondents), but that provision adds nothing to the analysis. Section 824(a), the FPA’s “declaration of policy,” states that federal regulation of electricity is to “extend only to those matters which are not subject to regulation by the States.” We have often explained that this declaration serves only to frame the Act’s basic structure and purpose. See, e.g., New York, 535 U. S., at 22 (Section 824(a) “broadly expresse[s] [the Act’s] purpose” (quoting FPC v. Southern Cal. Edison Co., 376 U. S. 205, 215 (1964) ); id., at 215 (Section 824(a) is “merely a ‘policy declaration . . . of great generality’ ” (quoting Connecticut Light & Power Co. v. FPC, 324 U. S. 515, 527 (1945) )). That means, as applied to the issue here, that §824(a) merely points toward the division of regulatory authority that §824(b) carries out. The operative provision is what counts. 7 The dissent disputes this framing of the issue, but its criticism (made by neither EPSA nor its amici) is irrelevant to deciding this case. According to the dissent, the FPA prohibits FERC from regulating not only retail sales of electricity (as we agree) but also any other sales of electricity aside from wholesale sales. See post, at 2–4. But the dissent turns out not to argue that the Rule regulates some kind of non-retail, non-wholesale sale of electric energy (whatever that might be). Rather, the dissent claims that the Rule regulates retail sales, see post, at 4–6—exactly the point that we address, and reject, in the following pages. And in any event, the dissent’s framing of the issue is wrong if and to the extent it posits some undefined category of other electricity sales falling within neither FERC’s nor the States’ regulatory authority. Sales of electric energy come in two varieties: wholesale and retail. The very case the dissent relies on recognizes that fact by referring to “other sales, that is, to direct sales for consumptive use.” Panhandle Eastern Pipe Line Co. v. Public Serv. Comm’n of Ind., 332 U. S. 507, 516 (1947) . FERC regulates interstate wholesale sales of electricity; the States regulate retail sales of electricity. And FERC may also regulate, as it did here, practices and rules affecting wholesale prices—that is, matters beyond wholesale sales themselves—so long as, in doing so, it does not trespass on the States’ authority to regulate retail sales of electric power. See supra, at 3. 8 The dissent offers, alternatively, a definition of “price,” but that only further proves our point. “Price,” says the dissent, is “[t]he amount of money or other consideration asked for or given in exchange for something else.” Post, at 6 (quoting Black’s Law Dictionary 1380). But the “effective” rates posited by EPSA and the dissent do not meet that test. If $10 is the actual rate for a unit of retail electricity, that is the only amount either “asked for” or “given” in exchange for power. A retail customer is asked to pay $10 by its LSE, and if it buys that electricity, it gives the LSE that same $10. By contrast, the $15 “effective” rate is neither asked for nor given by anyone. 9 The dissent replaces our simple, real-world example with a convoluted, fictitious one—but once again merely confirms our point. Suppose, the dissent says, that an airline cancels a passenger’s $400 ticket; gives him a refund plus an extra $300; and then tells him that if he wants to repurchase the ticket, he must pay $700. Aha!, says the dissent—isn’t the price now $700? See post, at 5–6. Well, yes it is, because that is now the actual amount the passenger will have to hand over to the airline to receive a ticket in exchange (or in the dissent’s definition of price, the amount “asked for” and “given,” see n. 8, supra). In other words, in search of an intuitive way to explain its “effective rate” theory, the dissent must rely on an “actual rate” hypothetical. But all that does is highlight the distance, captured in the law, between real prices (reflecting amounts paid) and effective ones (reflecting opportunity costs). 10 The dissent contests this point (complaining that our decades’ worth of precedents affirming it partly rely on legislative history), but the example the dissent offers in response misses the mark. See post, at 7–8. The dissent hypothesizes a rule enabling generators to sell directly to consumers and fixing all generation, transmission, and retail rates. But of course neither FERC nor the States could issue such a rule: If FERC did so, it would interfere with the States’ authority over retail sales and rates as well as (most) generation; if a State did so, it would interfere with FERC’s power over transmission. Thus, to implement such a scheme, the States would need to do some things and FERC to do others. And if the one or the other declined to cooperate, then the full scheme could not proceed. But that just goes to show that the FPA divides regulatory power over electricity matters between FERC and the States. The example does nothing to demonstrate that some electricity transactions can proceed outside any regulator’s authority. 11 EPSA now contends that wholesale demand response is unnecessary because state regulators can adopt programs to reduce demand at the retail level. See Brief for Respondents 46–47. For example, States can insist that utilities give rebates to customers for not using energy at certain times. See n. 2, supra. But according to both the Commission and market participants, state-level programs cannot offer nearly the same benefits as wholesale demand response because individual utilities lack the regional scope and real-time information needed to identify when demand response will lower prices and ensure reliability system-wide. See 73 Fed. Reg. 64103, ¶18; Energy Primer 45–46; Brief for NRG Energy, Inc., as Amicus Curiae 20–22. Similarly, FERC addressed and rejected the dissent’s suggestion that wholesale market operators could pay LSEs to reduce their electricity purchases: Because LSEs lose revenues whenever demand goes down, any demand response programs targeting those actors would be highly inefficient. See FERC, Assessment of Demand Response and Advanced Metering 72 (2006); Tr. of Oral Arg. 56 (Solicitor General noting that LSEs engaged in demand response would be “cannibaliz[ing] their own profits”).

579.US.2015_14-981

The University of Texas at Austin (University) uses an undergraduate admissions system containing two components. First, as required by the State’s Top Ten Percent Law, it offers admission to any students who graduate from a Texas high school in the top 10% of their class. It then fills the remainder of its incoming freshman class, some 25%, by combining an applicant’s “Academic Index”—the student’s SAT score and high school academic performance—with the applicant’s “Personal Achievement Index,” a holistic review containing numerous factors, including race. The University adopted its current admissions process in 2004, after a year-long-study of its admissions process—undertaken in the wake of Grutter v. Bollinger, 539 U. S. 306 , and Gratz v. Bollinger, 539 U. S. 244 —led it to conclude that its prior race-neutral system did not reach its goal of providing the educational benefits of diversity to its undergraduate students. Petitioner Abigail Fisher, who was not in the top 10% of her high school class, was denied admission to the University’s 2008 freshman class. She filed suit, alleging that the University’s consideration of race as part of its holistic-review process disadvantaged her and other Caucasian applicants, in violation of the Equal Protection Clause. The District Court entered summary judgment in the University’s favor, and the Fifth Circuit affirmed. This Court vacated the judgment, Fisher v. University of Tex. at Austin, 570 U. S. ___ (Fisher I), and remanded the case to the Court of Appeals, so the University’s program could be evaluated under the proper strict scrutiny standard. On remand, the Fifth Circuit again affirmed the entry of summary judgment for the University. Held: The race-conscious admissions program in use at the time of petitioner’s application is lawful under the Equal Protection Clause. Pp. 6–20. (a) Fisher I sets out three controlling principles relevant to assessing the constitutionality of a public university’s affirmative action program. First, a university may not consider race “unless the admissions process can withstand strict scrutiny,” i.e., it must show that its “purpose or interest is both constitutionally permissible and substantial, and that its use of the classification is necessary” to accomplish that purpose. 570 U. S., at ___. Second, “the decision to pursue the educational benefits that flow from student body diversity is, in substantial measure, an academic judgment to which some, but not complete, judicial deference is proper.” Id., at ___. Third, when determining whether the use of race is narrowly tailored to achieve the university’s permissible goals, the school bears the burden of demonstrating that “available” and “workable” “race-neutral alternatives” do not suffice. Id., at ___. Pp. 6–8. (b) The University’s approach to admissions gives rise to an unusual consequence here. The component with the largest impact on petitioner’s chances of admission was not the school’s consideration of race under its holistic-review process but the Top Ten Percent Plan. Because petitioner did not challenge the percentage part of the plan, the record is devoid of evidence of its impact on diversity. Remand for further factfinding would serve little purpose, however, because at the time of petitioner’s application, the current plan had been in effect only three years and, in any event, the University lacked authority to alter the percentage plan, which was mandated by the Texas Legislature. These circumstances refute any criticism that the University did not make good faith efforts to comply with the law. The University, however, does have a continuing obligation to satisfy the strict scrutiny burden: by periodically reassessing the admission program’s constitutionality, and efficacy, in light of the school’s experience and the data it has gathered since adopting its admissions plan, and by tailoring its approach to ensure that race plays no greater role than is necessary to meet its compelling interests. Pp. 8–11. (c) Drawing all reasonable inferences in her favor, petitioner has not shown by a preponderance of the evidence that she was denied equal treatment at the time her application was rejected. Pp. 11–19. (1) Petitioner claims that the University has not articulated its compelling interest with sufficient clarity because it has failed to state more precisely what level of minority enrollment would constitute a “critical mass.” However, the compelling interest that justifies consideration of race in college admissions is not an interest in enrolling a certain number of minority students, but an interest in obtaining “the educational benefits that flow from student body diversity.” Fisher I, 570 U. S., at ___. Since the University is prohibited from seeking a particular number or quota of minority students, it cannot be faulted for failing to specify the particular level of minority enrollment at which it believes the educational benefits of diversity will be obtained. On the other hand, asserting an interest in the educational benefits of diversity writ large is insufficient. A university’s goals cannot be elusory or amorphous—they must be sufficiently measurable to permit judicial scrutiny of the policies adopted to reach them. The record here reveals that the University articulated concrete and precise goals—e.g., ending stereotypes, promoting “cross-racial understanding,” preparing students for “an increasingly diverse workforce and society,” and cultivating leaders with “legitimacy in the eyes of the citizenry”—that mirror the compelling interest this Court has approved in prior cases. It also gave a “reasoned, principled explanation” for its decision, id., at ___, in a 39-page proposal written after a year-long study revealed that its race-neutral policies and programs did not meet its goals. Pp. 11–13. (2) Petitioner also claims that the University need not consider race because it had already “achieved critical mass” by 2003 under the Top Ten Percent Plan and race-neutral holistic review. The record, however, reveals that the University studied and deliberated for months, concluding that race-neutral programs had not achieved the University’s diversity goals, a conclusion supported by significant statistical and anecdotal evidence. Pp. 13–15. (3) Petitioner argues further that it was unnecessary to consider race because such consideration had only a minor impact on the number of minority students the school admitted. But the record shows that the consideration of race has had a meaningful, if still limited, effect on freshman class diversity. That race consciousness played a role in only a small portion of admissions decisions should be a hallmark of narrow tailoring, not evidence of unconstitutionality. P. 15. (4) Finally, petitioner argues that there were numerous other race-neutral means to achieve the University’s goals. However, as the record reveals, none of those alternatives was a workable means of attaining the University’s educational goals, as of the time of her application. Pp. 15–19. 758 F. 3d 633, affirmed. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, and Sotomayor, JJ., joined. Thomas, J., filed a dissenting opinion. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Thomas, J., joined. Kagan, J., took no part in the consideration or decision of the case.

The Court is asked once again to consider whether the race-conscious admissions program at the University of Texas is lawful under the Equal Protection Clause.IThe University of Texas at Austin (or University) relies upon a complex system of admissions that has undergone significant evolution over the past two decades. Until 1996, the University made its admissions decisions primarily based on a measure called “Academic Index” (or AI), which it calculated by combining an applicant’sSAT score and academic performance in high school. In assessing applicants, preference was given to racialminorities.In 1996, the Court of Appeals for the Fifth Circuit invalidated this admissions system, holding that any consideration of race in college admissions violates the Equal Protection Clause. See Hopwood v. Texas, 78 F. 3d 932, 934–935, 948.One year later the University adopted a new admissions policy. Instead of considering race, the University began making admissions decisions based on an applicant’s AI and his or her “Personal Achievement Index” (PAI). The PAI was a numerical score based on a holistic review of an application. Included in the number were the applicant’s essays, leadership and work experience, extracurricular activities, community service, and other “special characteristics” that might give the admissions committee insight into a student’s background. Consistent with Hopwood, race was not a consideration in calculating an applicant’s AI or PAI.The Texas Legislature responded to Hopwood as well. It enacted H. B. 588, commonly known as the Top Ten Percent Law. Tex. Educ. Code Ann. §51.803 (West Cum. Supp. 2015). As its name suggests, the Top Ten Percent Law guarantees college admission to students who graduate from a Texas high school in the top 10 percent of their class. Those students may choose to attend any of the public universities in the State.The University implemented the Top Ten Percent Law in 1998. After first admitting any student who qualified for admission under that law, the University filled the remainder of its incoming freshman class using a combination of an applicant’s AI and PAI scores—again, without considering race.The University used this admissions system until 2003, when this Court decided the companion cases of Grutter v. Bollinger,539 U. S. 306, and Gratz v. Bollinger,539 U. S. 244. In Gratz, this Court struck down the University of Michigan’s undergraduate system of admissions, which at the time allocated predetermined points to racial minority candidates. See 539 U. S., at 255, 275–276. In Grutter, however, the Court upheld the University of Michigan Law School’s system of holistic review—a system that did not mechanically assign points but rather treated race as a relevant feature within the broader context of a candidate’s application. See 539 U. S., at 337, 343–344. In upholding this nuanced use of race, Grutter implicitly overruled Hopwood’s categorical prohibition.In the wake of Grutter, the University embarked upon a year-long study seeking to ascertain whether its admissions policy was allowing it to provide “the educational benefits of a diverse student body . . . to all of the University’s undergraduate students.” App. 481a–482a (affidavit of N. Bruce Walker ¶11 (Walker Aff.)); see also id., at 445a–447a. The University concluded that its admissions policy was not providing these benefits. Supp. App. 24a–25a.To change its system, the University submitted a proposal to the Board of Regents that requested permission to begin taking race into consideration as one of “the many ways in which [an] academically qualified individual might contribute to, and benefit from, the rich, diverse, and challenging educational environment of the Univer-sity.” Id., at 23a. After the board approved the proposal, the University adopted a new admissions policy to implement it. The University has continued to use that admissions policy to this day.Although the University’s new admissions policy was a direct result of Grutter, it is not identical to the policy this Court approved in that case. Instead, consistent with the State’s legislative directive, the University continues to fill a significant majority of its class through the Top Ten Percent Plan (or Plan). Today, up to 75 percent of the places in the freshman class are filled through the Plan. As a practical matter, this 75 percent cap, which has now been fixed by statute, means that, while the Plan continues to be referenced as a “Top Ten Percent Plan,” a student actually needs to finish in the top seven or eight percent of his or her class in order to be admitted under this category.The University did adopt an approach similar to the one in Grutter for the remaining 25 percent or so of the incoming class. This portion of the class continues to be admitted based on a combination of their AI and PAI scores. Now, however, race is given weight as a subfactor within the PAI. The PAI is a number from 1 to 6 (6 is the best) that is based on two primary components. The first component is the average score a reader gives the applicant on two required essays. The second component is a full-file review that results in another 1-to-6 score, the “Personal Achievement Score” or PAS. The PAS is determined by a separate reader, who (1) rereads the applicant’s required essays, (2) reviews any supplemental information the applicant submits (letters of recommendation, resumes, an additional optional essay, writing samples, artwork, etc.), and (3) evaluates the applicant’s potential contributions to the University’s student body based on the applicant’s leadership experience, extracurricular activities, awards/honors, community service, and other “special circumstances.”“Special circumstances” include the socioeconomic status of the applicant’s family, the socioeconomic status of the applicant’s school, the applicant’s family responsibilities, whether the applicant lives in a single-parent home, the applicant’s SAT score in relation to the average SAT score at the applicant’s school, the language spoken at the applicant’s home, and, finally, the applicant’s race. See App. 218a–220a, 430a.Both the essay readers and the full-file readers who assign applicants their PAI undergo extensive training to ensure that they are scoring applicants consistently. Deposition of Brian Breman 9–14, Record in No. 1: 08–CV–00263, (WD Tex.), Doc. 96–3. The Admissions Office also undertakes regular “reliability analyses” to “measure the frequency of readers scoring within one point of each other.” App. 474a (affidavit of Gary M. Lavergne ¶8); see also id., at 253a (deposition of Kedra Ishop (Ishop Dep.)). Both the intensive training and the reliability analyses aim to ensure that similarly situated applicants are being treated identically regardless of which admissions officer reads the file.Once the essay and full-file readers have calculated each applicant’s AI and PAI scores, admissions officers from each school within the University set a cutoff PAI/AI score combination for admission, and then admit all of the applicants who are above that cutoff point. In setting the cutoff, those admissions officers only know how many applicants received a given PAI/AI score combination. They do not know what factors went into calculating those applicants’ scores. The admissions officers who make the final decision as to whether a particular applicant will be admitted make that decision without knowing the applicant’s race. Race enters the admissions process, then, at one stage and one stage only—the calculation of the PAS.Therefore, although admissions officers can consider race as a positive feature of a minority student’s application, there is no dispute that race is but a “factor of a factor of a factor” in the holistic-review calculus. 645 F. Supp. 2d 587, 608 (WD Tex. 2009). Furthermore, consideration of race is contextual and does not operate as a mechanical plus factor for underrepresented minorities. Id., at 606 (“Plaintiffs cite no evidence to show racial groups other than African-Americans and Hispanics are excluded from benefitting from UT’s consideration of race in admissions. As the Defendants point out, the consideration of race, within the full context of the entire application, may be beneficial to any UT Austin applicant—including whites and Asian-Americans”); see also Brief for Asian American Legal Defense and Education Fund et al. as Amici Curiae 12 (the contention that the University discriminates against Asian-Americans is “entirely unsupported by evidence in the record or empirical data”). There is also no dispute, however, that race, when considered in conjunction with other aspects of an applicant’s background, can alter an applicant’s PAS score. Thus, race, in this indirect fashion, considered with all of the other factors that make up an applicant’s AI and PAI scores, can make a difference to whether an application is accepted or rejected.Petitioner Abigail Fisher applied for admission to the University’s 2008 freshman class. She was not in the top 10 percent of her high school class, so she was evaluated for admission through holistic, full-file review. Petitioner’s application was rejected.Petitioner then filed suit alleging that the University’s consideration of race as part of its holistic-review process disadvantaged her and other Caucasian applicants, in violation of the Equal Protection Clause. See U. S. Const., Amdt. 14, §1 (no State shall “deny to any person within its jurisdiction the equal protection of the laws”). The District Court entered summary judgment in the University’s favor, and the Court of Appeals affirmed.This Court granted certiorari and vacated the judgment of the Court of Appeals, Fisher v. University of Tex. at Austin, 570 U. S. ___ (2013) (Fisher I ), because it had applied an overly deferential “good-faith” standard in assessing the constitutionality of the University’s program. The Court remanded the case for the Court of Appeals to assess the parties’ claims under the correct legal standard.Without further remanding to the District Court, the Court of Appeals again affirmed the entry of summary judgment in the University’s favor. 758 F. 3d 633 (CA5 2014). This Court granted certiorari for a second time, 576 U. S. ___ (2015), and now affirms.IIFisher I set forth three controlling principles relevant to assessing the constitutionality of a public university’s affirmative-action program. First, “because racial characteristics so seldom provide a relevant basis for disparate treatment,” Richmond v. J. A. Croson Co.,488 U. S. 469,505 (1989), “[r]ace may not be considered [by a university] unless the admissions process can withstand strict scru-tiny,” Fisher I, 570 U. S., at ___ (slip op., at 7). Strict scru-tiny requires the university to demonstrate with clarity that its “ ‘purpose or interest is both constitutionally permissible and substantial, and that its use of the classification is necessary . . . to the accomplishment of its purpose.’ ” Ibid.Second, Fisher I confirmed that “the decision to pursue ‘the educational benefits that flow from student body diversity’ . . . is, in substantial measure, an academic judgment to which some, but not complete, judicial deference is proper.” Id., at ___ (slip op, at 9). A university cannot impose a fixed quota or otherwise “define diversity as ‘some specified percentage of a particular group merely because of its race or ethnic origin.’ ” Ibid. Once, however, a university gives “a reasoned, principled explanation” for its decision, deference must be given “to the University’s conclusion, based on its experience and expertise, that a diverse student body would serve its educational goals.” Ibid. (internal quotation marks and citation omitted).Third, Fisher I clarified that no deference is owed when determining whether the use of race is narrowly tailored to achieve the university’s permissible goals. Id., at ___ (slip op., at 10). A university, Fisher I explained, bears the burden of proving a “nonracial approach” would not promote its interest in the educational benefits of diversity “about as well and at tolerable administrative expense.” Id., at ___ (slip op., at 11) (internal quotation marks omitted). Though “[n]arrow tailoring does not require exhaustion of every conceivable race-neutral alternative” or “require a university to choose between maintaining a reputation for excellence [and] fulfilling a commitment to provide educational opportunities to members of all racial groups,” Grutter, 539 U. S., at 339, it does impose “on the university the ultimate burden of demonstrating” that “race-neutral alternatives” that are both “available” and “workable” “do not suffice.” Fisher I, 570 U. S., at ___ (slip op., at 11).Fisher I set forth these controlling principles, while taking no position on the constitutionality of the admissions program at issue in this case. The Court held only that the District Court and the Court of Appeals had “confined the strict scrutiny inquiry in too narrow a way by deferring to the University’s good faith in its use of racial classifications.” Id., at ___ (slip op., at 12) The Court remanded the case, with instructions to evaluate the record under the correct standard and to determine whether the University had made “a showing that its plan is narrowly tailored to achieve” the educational benefits that flow from diversity. Id., at ___ (slip op., at 13). On remand, the Court of Appeals determined that the program conformed with the strict scrutiny mandated by Fisher I. See 758 F. 3d, at 659–660. Judge Garzadissented.IIIThe University’s program is sui generis. Unlike other approaches to college admissions considered by this Court, it combines holistic review with a percentage plan. This approach gave rise to an unusual consequence in this case: The component of the University’s admissions policy that had the largest impact on petitioner’s chances of admission was not the school’s consideration of race under its holistic-review process but rather the Top Ten Percent Plan. Because petitioner did not graduate in the top 10 percent of her high school class, she was categorically ineligible for more than three-fourths of the slots in the incoming freshman class. It seems quite plausible, then, to think that petitioner would have had a better chance of being admitted to the University if the school used race-conscious holistic review to select its entire incoming class, as was the case in Grutter.Despite the Top Ten Percent Plan’s outsized effect on petitioner’s chances of admission, she has not challenged it. For that reason, throughout this litigation, the Top Ten Percent Plan has been taken, somewhat artificially, as a given premise.Petitioner’s acceptance of the Top Ten Percent Plan complicates this Court’s review. In particular, it has led to a record that is almost devoid of information about the students who secured admission to the University through the Plan. The Court thus cannot know how students admitted solely based on their class rank differ in their contribution to diversity from students admitted through holistic review.In an ordinary case, this evidentiary gap perhaps could be filled by a remand to the district court for further factfinding. When petitioner’s application was rejected, however, the University’s combined percentage-plan/holistic-review approach to admission had been in effect for just three years. While studies undertaken over the eight years since then may be of significant value in determining the constitutionality of the University’s current admissions policy, that evidence has little bearing on whether petitioner received equal treatment when her application was rejected in 2008. If the Court were to remand, therefore, further factfinding would be limited to a narrow 3-year sample, review of which might yield little insight.Furthermore, as discussed above, the University lacks any authority to alter the role of the Top Ten Percent Plan in its admissions process. The Plan was mandated by the Texas Legislature in the wake of Hopwood, so the University, like petitioner in this litigation, has likely taken the Plan as a given since its implementation in 1998. If the University had no reason to think that it could deviate from the Top Ten Percent Plan, it similarly had no reason to keep extensive data on the Plan or the students admitted under it—particularly in the years before Fisher I clarified the stringency of the strict-scrutiny burden for a school that employs race-conscious review.Under the circumstances of this case, then, a remand would do nothing more than prolong a suit that has already persisted for eight years and cost the parties on both sides significant resources. Petitioner long since has graduated from another college, and the University’s policy—and the data on which it first was based—may have evolved or changed in material ways.The fact that this case has been litigated on a somewhat artificial basis, furthermore, may limit its value for prospective guidance. The Texas Legislature, in enacting the Top Ten Percent Plan, cannot much be criticized, for it was responding to Hopwood, which at the time was binding law in the State of Texas. That legislative response, in turn, circumscribed the University’s discretion in crafting its admissions policy. These circumstances refute any criticism that the University did not make good-faith efforts to comply with the law.That does not diminish, however, the University’s continuing obligation to satisfy the burden of strict scrutiny in light of changing circumstances. The University en-gages in periodic reassessment of the constitutionality, and efficacy, of its admissions program. See Supp. App. 32a; App. 448a. Going forward, that assessment must be undertaken in light of the experience the school has accumulated and the data it has gathered since the adoption of its admissions plan.As the University examines this data, it should remain mindful that diversity takes many forms. Formalistic racial classifications may sometimes fail to capture diversity in all of its dimensions and, when used in a divisive manner, could undermine the educational benefits the University values. Through regular evaluation of data and consideration of student experience, the University must tailor its approach in light of changing circumstances, ensuring that race plays no greater role than is neces-sary to meet its compelling interest. The University’s examination of the data it has acquired in the years since petitioner’s application, for these reasons, must proceed with full respect for the constraints imposed by the Equal Protection Clause. The type of data collected, and the manner in which it is considered, will have a significant bearing on how the University must shape its admissions policy to satisfy strict scrutiny in the years to come. Here, however, the Court is necessarily limited to the narrow question before it: whether, drawing all reasonable inferences in her favor, petitioner has shown by a preponderance of the evidence that she was denied equal treatment at the time her application was rejected.IVIn seeking to reverse the judgment of the Court of Appeals, petitioner makes four arguments. First, she argues that the University has not articulated its compelling interest with sufficient clarity. According to petitioner, the University must set forth more precisely the level of minority enrollment that would constitute a “critical mass.” Without a clearer sense of what the University’s ultimate goal is, petitioner argues, a reviewing court cannot assess whether the University’s admissions program is narrowly tailored to that goal.As this Court’s cases have made clear, however, the compelling interest that justifies consideration of race in college admissions is not an interest in enrolling a certain number of minority students. Rather, a university may institute a race-conscious admissions program as a means of obtaining “the educational benefits that flow from student body diversity.” Fisher I, 570 U. S., at ___ (slip op., at 9) (internal quotation marks omitted); see also Grutter, 539 U. S., at 328. As this Court has said, enrolling a diverse student body “promotes cross-racial understanding, helps to break down racial stereotypes, and enables students to better understand persons of different races.” Id., at 330 (internal quotation marks and alteration omitted). Equally important, “student body diversity promotes learning outcomes, and better prepares students for an increasingly diverse workforce and society.” Ibid. (internal quotation marks omitted).Increasing minority enrollment may be instrumental to these educational benefits, but it is not, as petitioner seems to suggest, a goal that can or should be reduced to pure numbers. Indeed, since the University is prohibited from seeking a particular number or quota of minority students, it cannot be faulted for failing to specify the particular level of minority enrollment at which it believes the educational benefits of diversity will be obtained.On the other hand, asserting an interest in the educational benefits of diversity writ large is insufficient. A university’s goals cannot be elusory or amorphous—they must be sufficiently measurable to permit judicial scrutiny of the policies adopted to reach them.The record reveals that in first setting forth its current admissions policy, the University articulated concrete and precise goals. On the first page of its 2004 “Proposal to Consider Race and Ethnicity in Admissions,” the Univer-sity identifies the educational values it seeks to realize through its admissions process: the destruction of stereotypes, the “ ‘promot[ion of] cross-racial understanding,’ ” the preparation of a student body “ ‘for an increasingly diverse workforce and society,’ ” and the “ ‘cultivat[ion of] a set of leaders with legitimacy in the eyes of the citizenry.’ ” Supp. App. 1a; see also id., at 69a; App. 314a–315a (deposition of N. Bruce Walker (Walker Dep.)), 478a–479a (Walker Aff. ¶4) (setting forth the same goals). Later in the proposal, the University explains that it strives to provide an “academic environment” that offers a “robust exchange of ideas, exposure to differing cultures, preparation for the challenges of an increasingly diverse workforce, and acquisition of competencies required of future leaders.” Supp. App. 23a. All of these objectives, as a general matter, mirror the “compelling interest” this Court has approved in its prior cases.The University has provided in addition a “reasoned, principled explanation” for its decision to pursue these goals. Fisher I, supra, at ___ (slip op., at 9). The Univer-sity’s 39-page proposal was written following a year-long study, which concluded that “[t]he use of race-neutral policies and programs ha[d] not been successful” in “provid[ing] an educational setting that fosters cross-racial understanding, provid[ing] enlightened discussion and learning, [or] prepar[ing] students to function in an increasingly diverse workforce and society.” Supp. App. 25a; see also App. 481a–482a (Walker Aff. ¶¶8–12) (describing the “thoughtful review” the University undertook when it faced the “important decision . . . whether or not to use race in its admissions process”). Further support for the University’s conclusion can be found in the depositions and affidavits from various admissions officers, all of whom articulate the same, consistent “reasoned, principled explanation.” See, e.g., id., at 253a (Ishop Dep.), 314a–318a, 359a (Walker Dep.), 415a–416a (Defendant’s Statement of Facts), 478a–479a, 481a–482a (Walker Aff. ¶¶4, 10–13). Petitioner’s contention that the University’s goal was insufficiently concrete is rebutted by the record.Second, petitioner argues that the University has no need to consider race because it had already “achieved critical mass” by 2003 using the Top Ten Percent Plan and race-neutral holistic review. Brief for Petitioner 46. Petitioner is correct that a university bears a heavy burden in showing that it had not obtained the educational benefits of diversity before it turned to a race-conscious plan. The record reveals, however, that, at the time of petitioner’s application, the University could not be faulted on this score. Before changing its policy the University conducted “months of study and deliberation, including retreats, interviews, [and] review of data,” App. 446a, and concluded that “[t]he use of race-neutral policies and programs ha[d] not been successful in achieving” sufficient racial diversity at the University, Supp. App. 25a. At no stage in this litigation has petitioner challenged the University’s good faith in conducting its studies, and the Court properly declines to consider the extrarecord materials the dissent relies upon, many of which are tangential to this case at best and none of which the University has had a full opportunity to respond to. See, e.g., post, at 45–46 (opinion of Alito, J.) (describing a 2015 report regarding the admission of applicants who are related to ‘‘politically connected individuals’’).The record itself contains significant evidence, both statistical and anecdotal, in support of the University’s position. To start, the demographic data the University has submitted show consistent stagnation in terms of the percentage of minority students enrolling at the Univer-sity from 1996 to 2002. In 1996, for example, 266 African-American freshmen enrolled, a total that constituted 4.1 percent of the incoming class. In 2003, the year Grutter was decided, 267 African-American students enrolled—again, 4.1 percent of the incoming class. The numbers for Hispanic and Asian-American students tell a similar story. See Supp. App. 43a. Although demographics alone are by no means dispositive, they do have some value as a gauge of the University’s ability to enroll students who can offer underrepresented perspectives.In addition to this broad demographic data, the University put forward evidence that minority students admitted under the Hopwood regime experienced feelings of loneliness and isolation. See, e.g., App. 317a–318a.This anecdotal evidence is, in turn, bolstered by further, more nuanced quantitative data. In 2002, 52 percent of undergraduate classes with at least five students had no African-American students enrolled in them, and 27 percent had only one African-American student. Supp. App. 140a. In other words, only 21 percent of undergraduate classes with five or more students in them had more than one African-American student enrolled. Twelve percent of these classes had no Hispanic students, as compared to 10 percent in 1996. Id., at 74a, 140a. Though a college must continually reassess its need for race-conscious review, here that assessment appears to have been done with care, and a reasonable determination was made that the University had not yet attained its goals.Third, petitioner argues that considering race was not necessary because such consideration has had only a “ ‘minimal impact’ in advancing the [University’s] compelling interest.” Brief for Petitioner 46; see also Tr. of Oral Arg. 23:10–12; 24:13–25:2, 25:24–26:3. Again, the record does not support this assertion. In 2003, 11 percent of the Texas residents enrolled through holistic review were Hispanic and 3.5 percent were African-American. Supp. App. 157a. In 2007, by contrast, 16.9 percent of the Texas holistic-review freshmen were Hispanic and 6.8 percent were African-American. Ibid. Those increases—of 54 percent and 94 percent, respectively—show that consideration of race has had a meaningful, if still limited, effect on the diversity of the University’s freshman class.In any event, it is not a failure of narrow tailoring for the impact of racial consideration to be minor. The fact that race consciousness played a role in only a small portion of admissions decisions should be a hallmark of narrow tailoring, not evidence of unconstitutionality.Petitioner’s final argument is that “there are numerous other available race-neutral means of achieving” the University’s compelling interest. Brief for Petitioner 47. A review of the record reveals, however, that, at the time of petitioner’s application, none of her proposed alternatives was a workable means for the University to attain the benefits of diversity it sought. For example, petitioner suggests that the University could intensify its outreach efforts to African-American and Hispanic applicants. But the University submitted extensive evidence of the many ways in which it already had intensified its outreach efforts to those students. The University has created three new scholarship programs, opened new regional admissions centers, increased its recruitment budget by half-a-million dollars, and organized over 1,000 recruitment events. Supp. App. 29a–32a; App. 450a–452a (citing affidavit of Michael Orr ¶¶4–20). Perhaps more significantly, in the wake of Hopwood, the University spent seven years attempting to achieve its compelling interest using race-neutral holistic review. None of these efforts succeeded, and petitioner fails to offer any meaningful way in which the University could have improved upon them at the time of her application.Petitioner also suggests altering the weight given to academic and socioeconomic factors in the University’s admissions calculus. This proposal ignores the fact that the University tried, and failed, to increase diversity through enhanced consideration of socioeconomic and other factors. And it further ignores this Court’s precedent making clear that the Equal Protection Clause does not force universities to choose between a diverse student body and a reputation for academic excellence. Grutter, 539 U. S., at 339.Petitioner’s final suggestion is to uncap the Top Ten Percent Plan, and admit more—if not all—the University’s students through a percentage plan. As an initial matter, petitioner overlooks the fact that the Top Ten Percent Plan, though facially neutral, cannot be understood apart from its basic purpose, which is to boost minority enrollment. Percentage plans are “adopted with racially segregated neighborhoods and schools front and center stage.” Fisher I, 570 U. S., at ___ (Ginsburg, J., dissenting) (slip op., at 2). “It is race consciousness, not blindness to race, that drives such plans.” Ibid. Consequently, petitioner cannot assert simply that increasing the University’s reliance on a percentage plan would make its admissions policy more race neutral.Even if, as a matter of raw numbers, minority enrollment would increase under such a regime, petitioner would be hard-pressed to find convincing support for the proposition that college admissions would be improved if they were a function of class rank alone. That approach would sacrifice all other aspects of diversity in pursuit of enrolling a higher number of minority students. A system that selected every student through class rank alone would exclude the star athlete or musician whose grades suffered because of daily practices and training. It would exclude a talented young biologist who struggled to maintain above-average grades in humanities classes. And it would exclude a student whose freshman-year grades were poor because of a family crisis but who got herself back on track in her last three years of school, only to find herself just outside of the top decile of her class.These are but examples of the general problem. Class rank is a single metric, and like any single metric, it will capture certain types of people and miss others. This does not imply that students admitted through holistic review are necessarily more capable or more desirable than those admitted through the Top Ten Percent Plan. It merely reflects the fact that privileging one characteristic above all others does not lead to a diverse student body. Indeed, to compel universities to admit students based on class rank alone is in deep tension with the goal of educational diversity as this Court’s cases have defined it. See Grutter, supra, at 340 (explaining that percentage plans “may preclude the university from conducting the individualized assessments necessary to assemble a student body that is not just racially diverse, but diverse along all the qualities valued by the university”); 758 F. 3d, at 653 (pointing out that the Top Ten Percent Law leaves out students “who fell outside their high school’s top ten percent but excelled in unique ways that would enrich the diversity of [the University’s] educational experience” and “leaves a gap in an admissions process seeking to create the multi-dimensional diversity that [Regents of Univ. of Cal. v. Bakke,438 U. S. 265 (1978),] envisions”). At its center, the Top Ten Percent Plan is a blunt instrument that may well compromise the University’s own definition of the diversity it seeks.In addition to these fundamental problems, an admissions policy that relies exclusively on class rank creates perverse incentives for applicants. Percentage plans “encourage parents to keep their children in low-performing segregated schools, and discourage students from taking challenging classes that might lower their grade point averages.” Gratz, 539 U. S., at 304, n. 10 (Ginsburg, J., dissenting).For all these reasons, although it may be true that the Top Ten Percent Plan in some instances may provide a path out of poverty for those who excel at schools lacking in resources, the Plan cannot serve as the admissions solution that petitioner suggests. Wherever the balance between percentage plans and holistic review should rest, an effective admissions policy cannot prescribe, realisti-cally, the exclusive use of a percentage plan.In short, none of petitioner’s suggested alternatives—nor other proposals considered or discussed in the course of this litigation—have been shown to be “available” and “workable” means through which the University could have met its educational goals, as it understood and defined them in 2008. Fisher I, supra, at ___ (slip op., at 11). The University has thus met its burden of showing that the admissions policy it used at the time it rejected petitioner’s application was narrowly tailored.* * *A university is in large part defined by those intangible “qualities which are incapable of objective measurement but which make for greatness.” Sweatt v. Painter,339 U. S. 629,634 (1950). Considerable deference is owed to a university in defining those intangible characteristics, like student body diversity, that are central to its identity and educational mission. But still, it remains an enduring challenge to our Nation’s education system to reconcile the pursuit of diversity with the constitutional promise of equal treatment and dignity.In striking this sensitive balance, public universities, like the States themselves, can serve as “laboratories for experimentation.” United States v. Lopez,514 U. S. 549,581 (1995) (Kennedy, J., concurring); see also New State Ice Co. v. Liebmann,285 U. S. 262,311 (1932) (Brandeis, J., dissenting). The University of Texas at Austin has a special opportunity to learn and to teach. The University now has at its disposal valuable data about the manner in which different approaches to admissions may foster diversity or instead dilute it. The University must con-tinue to use this data to scrutinize the fairness of its admis-sions program; to assess whether changing demographics have undermined the need for a race-conscious policy; and to identify the effects, both positive and negative, of the affirmative-action measures it deems necessary.The Court’s affirmance of the University’s admissions policy today does not necessarily mean the University may rely on that same policy without refinement. It is the University’s ongoing obligation to engage in constant deliberation and continued reflection regarding its admissions policies.The judgment of the Court of Appeals is affirmed.It is so ordered.Justice Kagan took no part in the consideration or decision of this case.