COOPER INDUSTRIES, INC. v. LEATHERMAN TOOL GROUP, INC.

Justice Stevens, delivered the opinion of the Court.

A jury found petitioner guilty of unfair competition and awarded respondent $50,000 in compensatory damages and $4.5 million in punitive damages. The District Court held that the punitive damages award did not violate the Federal Constitution. The Court of Appeals concluded that "the district court did not abuse its discretion in declining to reduce the amount of punitive damages." App. to Pet. for Cert. 4a. The issue in this case is whether the Court of Appeals applied the wrong standard of review in considering the constitutionality of the punitive damages award.

I

The parties are competing tool manufacturers. In the 1980's, Leatherman Tool Group, Inc. (Leatherman or respondent), introduced its Pocket Survival Tool (PST). The Court of Appeals described the PST as an

Cooper introduced the original ToolZall in August 1996 at the National Hardware Show in Chicago. At that show, it used photographs in its posters, packaging, and advertising materials that purported to be of a ToolZall but were actually of a modified PST. When those materials were prepared, the first of the ToolZalls had not yet been manufactured. A Cooper employee created a ToolZall "mock-up" by grinding the Leatherman trademark from handles and pliers of a PST and substituting the unique fastenings that were to be used on the ToolZall. At least one of the photographs was retouched to remove a curved indentation where the Leatherman trademark had been. The photographs were used, not only at the trade show, which normally draws an audience of over 70,000 people, but also in the marketing materials and catalogs used by Cooper's sales force throughout the United States. Cooper also distributed a touched-up line drawing of a PST to its international sales representatives.2

Shortly after the trade show, Leatherman filed this action asserting claims of trade-dress infringement, unfair competition, and false advertising under § 43(a) of the Trademark Act of 1946 (Lanham Act), 60 Stat. 441, as amended, 15 U. S. C. § 1125(a) (1994 ed. and Supp. V), and a common-law claim of unfair competition for advertising and selling an "imitation" of the PST. In December 1996, the District Court entered a preliminary injunction prohibiting Cooper from marketing the ToolZall and from using pictures of the modified PST in its advertising. Cooper withdrew the original ToolZall from the market and developed a new model with plastic coated handles that differed from the PST. In November 1996, it had anticipatorily sent a notice to its sales personnel ordering a recall of all promotional materials containing pictures of the PST, but it did not attempt to retrieve the materials it had sent to its customers until the following April. As a result, the offending promotional materials continued to appear in catalogs and advertisements well into 1997.

After a trial conducted in October 1997, the jury returned a verdict that answered several special interrogatories. With respect to the Lanham Act infringement claims, the jury found that Leatherman had trademark rights in the overall appearance of the PST and that the original ToolZall infringed those rights but that the infringement had not damaged Leatherman. It then found that the modified ToolZall did not infringe Leatherman's trademark rights in the PST. With respect to the advertising claims, it found Cooper guilty of passing off, false advertising, and unfair competition and assessed aggregate damages of $50,000 on those claims. It then answered "Yes" to the following interrogatory:

Because it answered this question in the affirmative, the jury was instructed to determine the "amount of punitive damages [that] should be awarded to Leatherman." Ibid. The jury awarded $4.5 million. Ibid.

After the jury returned its verdict, the District Court considered, and rejected, arguments that the punitive damages were "grossly excessive" under our decision in BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996). See App. to Pet. for Cert. 24a. It then entered its judgment, which provided that 60% of the punitive damages would be paid to the Criminal Injuries Compensation Account of the State of Oregon. The judgment also permanently enjoined Cooper from marketing its original ToolZall in the United States or in 22 designated foreign countries.

On appeal, Cooper challenged both the District Court's injunction against copying the PST and the punitive damages award. The Court of Appeals issued two opinions. In its published opinion it set aside the injunction. Leather- man Tool Group, Inc. v. Cooper Industries, Inc., supra. It held that the overall appearance of the PST was not protected under the trademark laws because its distinguishing features, and the combination of those features, were functional. Accordingly, even though Cooper had deliberately copied the PST, it acted lawfully in doing so.3

In its unpublished opinion, the Court of Appeals affirmed the punitive damages award. It first rejected Cooper's argument that the Oregon Constitution, which has been interpreted to prohibit awards of punitive damages for torts that impose liability for speech, precluded the jury's award of such damages in this case. It then reviewed the District Court's finding that the award "was proportional and fair, given the nature of the conduct, the evidence of intentional passing off, and the size of an award necessary to create deterrence to an entity of Cooper's size" and concluded "that the award did not violate Cooper's due process rights" under the Federal Constitution. App. to Pet. for Cert. 3a, judgt. order reported at 205 F.3d 1351 (CA9 1999). It noted that the "passing off" in this case was "very unusual" because "even assuming PST is a superior product," no superior features of the PST were perceivable in the photographs. App. to Pet. for Cert. 3a. "Any customer who bought based on what the photographs showed would have received essentially that for which he or she paid." Ibid. Thus, Cooper's use of the photographs of the PST did not involve "the same sort of potential harm to Leatherman or to customers as that which may arise from traditional passing off." Id., at 4a. The Court of Appeals made clear, however, that it did not condone the passing off. "[A]t a minimum," it observed, "[the passing off] gave Cooper an unfair advantage" by allowing it to use Leatherman's work product "to obtain a `mock-up' more cheaply, easily, and quickly" than if it had waited until its own product was ready. Ibid. Accordingly, the Court of Appeals concluded, "the district court did not abuse its discretion in declining to reduce the amount of punitive damages." Ibid.

Cooper's petition for a writ of certiorari asked us to decide whether the Court of Appeals reviewed the constitutionality of the punitive damages award under the correct standard and also whether the award violated the criteria we articulated in Gore. We granted the petition to resolve confusion among the Courts of Appeals on the first question.4 531 U.S. 923 (2000). We now conclude that the constitutional issue merits de novo review. Because the Court of Appeals applied an "abuse of discretion" standard, we remand the case for determination of the second question under the proper standard.

II

Although compensatory damages and punitive damages are typically awarded at the same time by the same decisionmaker, they serve distinct purposes. The former are intended to redress the concrete loss that the plaintiff has suffered by reason of the defendant's wrongful conduct. See Restatement (Second) of Torts § 903, pp. 453-454 (1979); Pacific Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 54 (1991) (O'Connor, J., dissenting). The latter, which have been described as "quasi-criminal," id., at 19, operate as "private fines" intended to punish the defendant and to deter future wrongdoing. A jury's assessment of the extent of a plaintiff's injury is essentially a factual determination, whereas its imposition of punitive damages is an expression of its moral condemnation. See Gertz v. Robert Welch, Inc., 418 U.S. 323, 350 (1974) ("[Punitive damages] are not compensation for injury. Instead, they are private fines levied by civil juries to punish reprehensible conduct and to deter its future occurrence"); Haslip, 499 U. S., at 54 (O'Connor, J., dissenting) ("[P]unitive damages are specifically designed to exact punishment in excess of actual harm to make clear that the defendant's misconduct was especially reprehensible").5

Legislatures have extremely broad discretion in defining criminal offenses, Schall v. Martin, 467 U.S. 253, 268-269, n. 18 (1984), and in setting the range of permissible punishments for each offense, ibid.; Solem v. Helm, 463 U.S. 277, 290 (1983). Judicial decisions that operate within these legislatively enacted guidelines are typically reviewed for abuse of discretion. See, e. g., Koon v. United States, 518 U.S. 81, 96, 99-100 (1996); cf. Apprendi v. New Jersey, 530 U.S. 466, 481 (2000) (it is permissible "for judges to exercise discretion . . . in imposing a judgment within the range prescribed by statute").

As in the criminal sentencing context, legislatures enjoy broad discretion in authorizing and limiting permissible punitive damages awards. Cf. Gore, 517 U. S., at 568 ("States necessarily have considerable flexibility in determining the level of punitive damages that they will allow in different classes of cases and in any particular case"). A good many States have enacted statutes that place limits on the permissible size of punitive damages awards.6 When juries make particular awards within those limits, the role of the trial judge is "to determine whether the jury's verdict is within the confines set by state law, and to determine, by reference to federal standards developed under Rule 59, whether a new trial or remittitur should be ordered." Browning-Ferris Industries of Vt., Inc. v. Kelco Disposal, Inc., 492 U.S. 257, 279 (1989). If no constitutional issue is raised, the role of the appellate court, at least in the federal system, is merely to review the trial court's "determination under an abuse-ofdiscretion standard." Ibid.7

Despite the broad discretion that States possess with respect to the imposition of criminal penalties and punitive damages, the Due Process Clause of the Fourteenth Amendment to the Federal Constitution imposes substantive limits on that discretion. That Clause makes the Eighth Amendment's prohibition against excessive fines and cruel and unusual punishments applicable to the States. Furman v. Georgia, 408 U.S. 238 (1972) (per curiam). The Due Process Clause of its own force also prohibits the States from imposing "grossly excessive" punishments on tortfeasors, Gore, 517 U. S., at 562; TXO Production Corp. v. Alliance Resources Corp., 509 U.S. 443, 453-455 (1993) (plurality opinion).

The Court has enforced those limits in cases involving deprivations of life, Enmund v. Florida, 458 U.S. 782, 787, 801 (1982) (death is not "a valid penalty under the Eighth and Fourteenth Amendments for one who neither took life, attempted to take life, nor intended to take life"); Coker v. Georgia, 433 U.S. 584, 592 (1977) (opinion of White, J.) (sentence of death is "grossly disproportionate" and excessive punishment for the crime of rape);8 deprivations of liberty, Solem v. Helm, 463 U. S., at 279, 303 (life imprisonment without the possibility of parole for nonviolent felonies is "significantly disproportionate"); and deprivations of property, United States v. Bajakajian, 524 U.S. 321, 324 (1998) (punitive forfeiture of $357,144 for violating reporting requirement was "grossly disporportional" to the gravity of the offense); Gore, 517 U. S., at 585-586 ($2 million punitive damages award for failing to advise customers of minor predelivery repairs to new automobiles was "grossly excessive" and therefore unconstitutional).

In these cases, the constitutional violations were predicated on judicial determinations that the punishments were "grossly disproportional to the gravity of . . . defendant[s'] offense[s]." Bajakajian, 524 U. S., at 334; see also Gore, 517 U. S., at 585-586; Solem, 463 U. S., at 303; Coker, 433 U. S., at 592 (opinion of White, J.). We have recognized that the relevant constitutional line is "inherently imprecise," Ba- jakajian, 524 U. S., at 336, rather than one "marked by a simple mathematical formula," Gore, 517 U. S., at 582. But in deciding whether that line has been crossed, we have focused on the same general criteria: the degree of the defendant's reprehensibility or culpability, see, e. g., Bajakajian, 524 U. S., at 337; Gore, 517 U. S., at 575-580; Solem, 463 U. S., at 290-291; Enmund, 458 U. S., at 798; Coker, 433 U. S., at 598 (opinion of White, J.); the relationship between the penalty and the harm to the victim caused by the defendant's actions, see, e. g., Bajakajian, 524 U. S., at 339; Gore, 517 U. S., at 580-583; Solem, 463 U. S., at 293; Enmund, 458 U. S., at 798; Coker, 433 U. S., at 598 (opinion of White, J.); and the sanctions imposed in other cases for comparable misconduct, see, e. g., Bajakajian, 524 U. S., at 340-343; Gore, 517 U. S., at 583-585; Solem, 463 U. S., at 291; Enmund, 458 U. S., at 789-796; Coker, 433 U. S., at 593-597 (opinion of White, J.). Moreover, and of greatest relevance for the issue we address today, in each of these cases we have engaged in an independent examination of the relevant criteria. See, e. g., Bajakajian, 524 U. S., at 337-344; Gore, 517 U. S., at 575-586; Solem, 463 U. S., at 295-300; Enmund, 458 U. S., at 788-801; Coker, 433 U. S., at 592-600 (opinion of White, J.).

In Bajakajian, we expressly noted that the courts of appeals must review the proportionality determination "de novo" and specifically rejected the suggestion of the respondent, who had prevailed in the District Court, that the trial judge's determination of excessiveness should be reviewed only for an abuse of discretion. "The factual findings made by the district courts in conducting the excessiveness inquiry, of course, must be accepted unless clearly erroneous. . . . But the question whether a fine is constitutionally excessive calls for the application of a constitutional standard to the facts of a particular case, and in this context de novo review of that question is appropriate." 524 U. S., at 336-337, n. 10 (citing Ornelas v. United States, 517 U.S. 690, 697 (1996)).

Likewise, in Ornelas, we held that trial judges' determinations of reasonable suspicion and probable cause should be reviewed de novo on appeal. The reasons we gave in support of that holding are equally applicable in this case. First, as we observed in Ornelas, the precise meaning of concepts like "reasonable suspicion" and "probable cause" cannot be articulated with precision; they are "fluid concepts that take their substantive content from the particular contexts in which the standards are being assessed." Id., at 696. That is, of course, also a characteristic of the concept of "gross excessiveness." Second, "the legal rules for probable cause and reasonable suspicion acquire content only through application. Independent review is therefore necessary if appellate courts are to maintain control of, and to clarify, the legal principles." Id., at 697. Again, this is also true of the general criteria set forth in Gore; they will acquire more meaningful content through case-by-case application at the appellate level. "Finally, de novo review tends to unify precedent" and "`stabilize the law.' " 517 U. S., at 697— 698. Justice Breyer made a similar point in his concurring opinion in Gore:

Our decisions in analogous cases, together with the reasoning that produced those decisions, thus convince us that courts of appeals should apply a de novo standard of review when passing on district courts' determinations of the constitutionality of punitive damages awards.9

III

"Unlike the measure of actual damages suffered, which presents a question of historical or predictive fact, see, e. g., [St. Louis, I. M. & S. R. Co. v. Craft, 237 U.S. 648 (1915)], the level of punitive damages is not really a `fact' `tried' by the jury." Gasperini v. Center for Humanities, Inc., 518 U.S. 415, 459 (1996) (Scalia, J., dissenting). Because the jury's award of punitive damages does not constitute a finding of "fact," appellate review of the district court's determination that an award is consistent with due process does not implicate the Seventh Amendment concerns raised by respondent and its amicus.10 See Brief for Respondent 18-24; Brief for Association of Trial Lawyers of America as Amicus Curiae 17-20. Our decisions in Gasperini and Hetzel v. Prince William County, 523 U.S. 208 (1998) (per curiam), both of which concerned compensatory damages, are not to the contrary.11

It might be argued that the deterrent function of punitive damages suggests that the amount of such damages awarded is indeed a "fact" found by the jury and that, as a result, the Seventh Amendment is implicated in appellate review of that award. Some scholars, for example, assert that punitive damages should be used to compensate for the underdeterrence of unlawful behavior that will result from a defendant's evasion of liability. See Polinsky & Shavell, Punitive Damages: An Economic Analysis, 111 Harv. L. Rev. 869, 890-891 (1998) (in order to obtain optimal deterrence, "punitive damages should equal the harm multiplied by . . . the ratio of the injurer's chance of escaping liability to his chance of being found liable"); see also Ciraolo v. New York, 216 F.3d 236, 244-245 (CA2 2000) (Calabresi, J., concurring). "The efficient deterrence theory thus regards punitive damages as merely an augmentation of compensatory damages designed to achieve economic efficiency." Galanter & Luban, Poetic Justice: Punitive Damages and Legal Pluralism, 42 Am. U. L. Rev. 1393, 1449 (1993).

However attractive such an approach to punitive damages might be as an abstract policy matter, it is clear that juries do not normally engage in such a finely tuned exercise of deterrence calibration when awarding punitive damages. See Sunstein, Schkade, & Kahneman, Do People Want Optimal Deterrence?, 29 J. Legal Studies 237, 240 (2000). After all, deterrence is not the only purpose served by punitive damages. See supra, at 432. And there is no dispute that, in this case, deterrence was but one of four concerns the jury was instructed to consider when setting the amount of punitive damages.12 Moreover, it is not at all obvious that even the deterrent function of punitive damages can be served only by economically "optimal deterrence." "[C]itizens and legislators may rightly insist that they are willing to tolerate some loss in economic efficiency in order to deter what they consider morally offensive conduct, albeit costbeneficial morally offensive conduct; efficiency is just one consideration among many." Galanter & Luban, 42 Am. U. L. Rev., at 1450.13

Differences in the institutional competence of trial judges and appellate judges are consistent with our conclusion. In Gore, we instructed courts evaluating a punitive damages award's consistency with due process to consider three criteria: (1) the degree or reprehensibility of the defendant's misconduct, (2) the disparity between the harm (or potential harm) suffered by the plaintiff and the punitive damages award, and (3) the difference between the punitive damages awarded by the jury and the civil penalties authorized or imposed in comparable cases. 517 U. S., at 574-575. Only with respect to the first Gore inquiry do the district courts have a somewhat superior vantage over courts of appeals, and even then the advantage exists primarily with respect to issues turning on witness credibility and demeanor.14 Trial courts and appellate courts seem equally capable of analyzing the second factor. And the third Gore criterion, which calls for a broad legal comparison, seems more suited to the expertise of appellate courts. Considerations of institutional competence therefore fail to tip the balance in favor of deferential appellate review.

IV

It is possible that the standard of review applied by the Court of Appeals will affect the result of the Gore analysis in only a relatively small number of cases. See Brief for Respondent 46-48; Brief for Association of American Railroads as Amicus Curiae 18; see also Gasperini, 518 U. S., at 448 (Stevens, J., dissenting). Nonetheless, it does seem likely that in this case a thorough, independent review of the District Court's rejection of petitioner's due process objections to the punitive damages award might well have led the Court of Appeals to reach a different result. Indeed, our own consideration of each of the three Gore factors reveals a series of questionable conclusions by the District Court that may not survive de novo review.

When the jury assessed the reprehensibility of Cooper's misconduct, it was guided by instructions that characterized the deliberate copying of the PST as wrongful. The jury's selection of a penalty to deter wrongful conduct may, therefore, have been influenced by an intent to deter Cooper from engaging in such copying in the future. Similarly, the District Court's belief that Cooper acted unlawfully in deliberately copying the PST design might have influenced its consideration of the first Gore factor. See App. to Pet. for Cert. 23a. But, as the Court of Appeals correctly held, such copying of the functional features of an unpatented product is lawful. See TrafFix Devices, Inc. v. Marketing Displays, Inc., ante, p. 23. The Court of Appeals recognized that the District Court's award of attorney's fees could not be supported if based on the premise that the copying was unlawful, but it did not consider whether that improper predicate might also have undermined the basis for the jury's large punitive damages award.

In evaluating the second Gore factor, the ratio between the size of the award of punitive damages and the harm caused by Cooper's tortious conduct, the District Court might have been influenced by respondent's submission that it was not the actual injury—which the jury assessed at $50,000—that was relevant, but rather "the potential harm Leatherman would have suffered had Cooper succeeded in its wrongful conduct." See Brief for Respondent 7; see also Record Doc. No. 323, p. 23. Respondent calculated that "potential harm" by referring to the fact that Cooper had anticipated "gross profits of approximately $3 million during the first five years of sales." Brief for Respondent 7; see also Record Doc. No. 323, at 23. Even if that estimate were correct, however, it would be unrealistic to assume that all of Cooper's sales of the ToolZall would have been attributable to its misconduct in using a photograph of a modified PST in its initial advertising materials. As the Court of Appeals pointed out, the picture of the PST did not misrepresent the features of the original ToolZall and could not have deceived potential customers in any significant way. Its use was wrongful because it enabled Cooper to expedite the promotion of its tool, but that wrongdoing surely could not be treated as the principal cause of Cooper's entire sales volume for a 5-year period.

With respect to the third Gore factor, respondent argues that Cooper would have been subject to a comparable sanction under Oregon's Unlawful Trade Practices Act. Brief for Respondent 49. In a suit brought by a State under that Act, a civil penalty of up to $25,000 per violation may be assessed. Ore. Rev. Stat. § 646.642(3) (1997). In respondent's view, each of the thousands of pieces of promotional material containing a picture of the PST that Cooper distributed warranted the maximum fine. Brief for Respondent 49. Petitioner, on the other hand, argues that its preparation of a single "mock-up" for use in a single distribution would have been viewed as a single violation under the state statute. Reply Brief for Petitioner 2-3. The Court of Appeals expressed no opinion on this dispute. It did, however, observe that the unfairness in Cooper's use of the picture apparently had nothing to do with misleading customers but was related to its inability to obtain a "mock-up" quickly and cheaply. App. to Pet. for Cert. 3a. This observation is more consistent with the single-violation theory than with the notion that the statutory violation would have been sanctioned with a multimillion dollar fine.

We have made these comments on issues raised by application of the three Gore guidelines to the facts of this case, not to prejudge the answer to the constitutional question, but rather to illustrate why we are persuaded that the Court of Appeals' answer to that question may depend upon the standard of review. The de novo standard should govern its decision. Because the Court of Appeals applied a less demanding standard in this case, we vacate the judgment and remand the case for further proceedings consistent with this opinion.

It is so ordered.

Justice Thomas, concurring.

I continue to believe that the Constitution does not constrain the size of punitive damages awards. See BMW of North America, Inc. v. Gore, 517 U.S. 559, 599 (1996) (Scalia, J., joined by Thomas, J., dissenting). For this reason, given the opportunity, I would vote to overrule BMW. This case, however, does not present such an opportunity. The only issue before us today is what standard should be used to review a trial court's ruling on a BMW challenge. Because I agree with the Court's resolution of that issue, I join the opinion of the Court.

Justice Scalia, concurring in the judgment.

I was (and remain) of the view that excessive punitive damages do not violate the Due Process Clause; but the Court held otherwise. See BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996); id., at 598 (Scalia, J., dissenting). And I was of the view that we should review for abuse of discretion (rather than de novo ) fact-bound constitutional issues which, in their resistance to meaningful generalization, resemble the question of excessiveness of punitive damages—namely, whether there exists reasonable suspicion for a stop and probable cause for a search; but the Court held otherwise. See Ornelas v. United States, 517 U.S. 690 (1996); id., at 700 (Scalia, J., dissenting). Finally, in a case in which I joined a dissent that made it unnecessary for me to reach the issue, the Court categorically stated that "the question whether a fine is constitutionally excessive calls for . . . de novo review." United States v. Bajakajian, 524 U.S. 321, 336-337, n. 10 (1998); see id. , at 344 (Kennedy, J., joined by Rehnquist, C. J.,and O'Connor and Scalia, JJ., dissenting). Given these precedents, I agree that de novo review of the question of excessive punitive damages best accords with our jurisprudence. Accordingly, I concur in the judgment of the Court.

Justice Ginsburg, dissenting.

In Gasperini v. Center for Humanities, Inc., 518 U.S. 415 (1996), we held that appellate review of a federal trial court's refusal to set aside a jury verdict as excessive is reconcilable with the Seventh Amendment if "appellate control [is] limited to review for `abuse of discretion.' " Id., at 419. Gasperini was a diversity action in which the defendant had challenged a compensatory damages award as excessive under New York law. The reasoning of that case applies as well to an action challenging a punitive damages award as excessive under the Constitution. I would hold, therefore, that the proper standard of appellate oversight is not de novo review, as the Court today concludes, but review for abuse of discretion.

"An essential characteristic of [the federal court] system is the manner in which, in civil common-law actions, it distributes trial functions between judge and jury and, under the influence—if not the command—of the Seventh Amendment, assigns the decisions of disputed questions of fact to the jury." Id., at 432 (citing Byrd v. Blue Ridge Rural Elec. Cooperative, Inc., 356 U.S. 525, 537 (1958)). The Seventh Amendment provides: "In Suits at common law . . . the right of trial by jury shall be preserved, and no fact tried by a jury, shall be otherwise re-examined in any Court of the United States, than according to the rules of the common law." In Gasperini, we observed that although trial courts traditionally had broad authority at common law to set aside jury verdicts and to grant new trials, 518 U. S., at 432-433, "appellate review of a federal trial court's denial of a motion to set aside a jury's verdict as excessive is a relatively late, and less secure, development," id., at 434. We ultimately concluded that the Seventh Amendment does not preclude such appellate review, id., at 436, but explained that "[w]ithin the federal system, practical reasons combine with Seventh Amendment constraints to lodge in the district court, not the court of appeals, primary responsibility for application of [an excessiveness standard]," id., at 438. "Trial judges have the unique opportunity to consider the evidence in the living courtroom context," we said, "while appellate judges see only the cold paper record." Ibid. (citations and internal quotation marks omitted). "If [courts of appeals] reverse, it must be because of an abuse of discretion. . . . The very nature of the problem counsels restraint. . . . [Appellate courts] must give the benefit of every doubt to the judgment of the trial judge." Id., at 438-439 (internal quotation marks omitted) (citing Dagnello v. Long Island R. Co., 289 F.2d 797, 806 (CA2 1961)).

Although Gasperini involved compensatory damages, I see no reason why its logic should be abandoned when punitive damages are alleged to be excessive. At common law, as our longstanding decisions reiterate, the task of determining the amount of punitive damages "has [always been] left to the discretion of the jury." Day v. Wood worth, 13 How. 363, 371 (1852); see Barry v. Edmunds, 116 U.S. 550, 565 (1886) ("[N]othing is better settled than that . . . it is the peculiar function of the jury to determine the amount [of punitive damages] by their verdict."). The commitment of this function to the jury, we have explained, reflects the historical understanding that "the degree of punishment to be thus inflicted must depend on the peculiar circumstances of each case." Day, 13 How., at 371. The relevant factors include "the conduct and motives of the defendant" and whether, "in committing the wrong complained of, he acted recklessly, or wilfully and maliciously, with a design to oppress and injure the plaintiff." 1 J. Sutherland, Law of Damages 720 (1882). Such inquiry, the Court acknowledges, "is a factsensitive undertaking." Ante, at 438, n. 11.

The Court nevertheless today asserts that a "jury's award of punitive damages does not constitute a finding of `fact' " within the meaning of the Seventh Amendment. Ante, at 437. An ultimate award of punitive damages, it is true, involves more than the resolution of matters of historical or predictive fact. See ibid. (citing Gasperini, 518 U. S., at 459 (Scalia, J., dissenting)). But there can be no question that a jury's verdict on punitive damages is fundamentally dependent on determinations we characterize as fact findings— e. g., the extent of harm or potential harm caused by the defendant's misconduct, whether the defendant acted in good faith, whether the misconduct was an individual instance or part of a broader pattern, whether the defendant behaved negligently, recklessly, or maliciously. Punitive damages are thus not "[u]nlike the measure of actual damages suffered," ante, at 437 (citation and internal quotation marks omitted), in cases of intangible, noneconomic injury. One million dollars' worth of pain and suffering does not exist as a "fact" in the world any more or less than one million dollars' worth of moral outrage. Both derive their meaning from a set of underlying facts as determined by a jury. If one exercise in quantification is properly regarded as fact finding, see St. Louis, I. M. & S. R. Co. v. Craft, 237 U.S. 648, 661 (1915) (compensation for pain and suffering "involves only a question of fact"), it seems to me the other should be so regarded as well.

In Browning-Ferris Industries of Vt., Inc. v. Kelco Disposal, Inc., 492 U.S. 257 (1989), we approved application of an abuse-of-discretion standard for appellate review of a district court's ruling on a motion to set aside a punitive damages award as excessive. See id., at 279. BrowningFerris reserved the question whether even such deferential appellate review might run afoul of the Seventh Amendment. At that time (i. e., pre-Gasperini ), the Court "ha[d] never held expressly that the Seventh Amendment allows appellate review of a district court's denial of a motion to set aside an award as excessive." 492 U. S., at 279, n. 25. We found it unnecessary to reach the Seventh Amendment question in Browning-Ferris because the jury verdict there survived lower court review intact. Id., at 279, n. 25, 280. Browning-Ferris, in short, signaled our recognition that appellate review of punitive damages, if permissible at all, would involve at most abuse-of-discretion review. "[P]articularly . . . because the federal courts operate under the strictures of the Seventh Amendment," we were "reluctant to stray too far from traditional common-law standards, or to take steps which ultimately might interfere with the proper role of the jury." Id., at 280, n. 26.

The Court finds no incompatibility between this case and Browning-Ferris, observing that Browning-Ferris presented for our review an excessiveness challenge resting solely on state law, not on the Constitution. See ante, at 433, and n. 7. It is unclear to me why this distinction should make a difference. Of the three guideposts BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996), established for assessing constitutional excessiveness, two were derived from common-law standards that typically inform state law. See id., at 575, n. 24 ("The principle that punishment should fit the crime is deeply rooted and frequently repeated in common-law jurisprudence." (citation and internal quotation marks omitted)); id., at 580 ("The principle that exemplary damages must bear a `reasonable relationship' to compensatory damages has a long pedigree."). The third guidepost—comparability of sanctions for comparable misconduct—is not similarly rooted in common law, nor is it similarly fact bound. As the Court states, "the third Gore criterion . . . calls for a broad legal comparison." Ante, at 440. But to the extent the inquiry is "legal" in character, there is little difference between review de novo and review for abuse of discretion. Cf. Gasperini, 518 U. S., at 448 (Stevens, J., dissenting) ("[I]t is a familiar . . . maxim that deems an error of law an abuse of discretion.").1

Apart from "Seventh Amendment constraints," an abuseof-discretion standard also makes sense for "practical reasons." Id., at 438. With respect to the first Gore inquiry (i. e., reprehensibility of the defendant's conduct), district courts have an undeniably superior vantage over courts of appeals. As earlier noted, supra, at 445, district courts view the evidence not on a "cold paper record," but "in the living courtroom context," Gasperini, 518 U. S., at 438. They can assess from the best seats the vital matter of witness credibility. And "it of course remains true that [a] Court of Appeals should defer to the District Court's findings of fact unless they are clearly erroneous." Ante, at 440, n. 14.2

The Court recognizes that district courts have the edge on the first Gore factor, ante, at 440, but goes on to say that "[t]rial courts and appellate courts seem equally capable of analyzing the second [Gore] factor" (i. e., whether punitive damages bear a reasonable relationship to the actual harm inflicted), ibid. Only "the third Gore criterion [i. e., intrajurisdictional and inter jurisdictional comparisons] . . . seems more suited to the expertise of appellate courts." Ibid.

To the extent the second factor requires a determination of "the actual harm inflicted on the plaintiff," Gore, 517 U. S., at 580, district courts may be better positioned to conduct the inquiry, especially in cases of intangible injury. I can demur to the Court's assessment of relative institutional strengths, however, for even accepting that assessment, I would disagree with the Court's conclusion that "[c]onsiderations of institutional competence . . . fail to tip the balance in favor of deferential appellate review," ante, at 440. Gore itself assigned particular importance to the first inquiry, characterizing "degree of reprehensibility" as "[p]erhaps the most important indicium of the reasonableness of a punitive damages award." 517 U. S., at 575. District courts, as just noted, supra, at 448 and this page, have a superior vantage over courts of appeals in conducting that fact-intensive inquiry. Therefore, in the typical case envisioned by Gore, where reasonableness is primarily tied to reprehensibility, an appellate court should have infrequent occasion to reverse.

This observation, I readily acknowledge, suggests that the practical difference between the Court's approach and my own is not large. An abuse-of-discretion standard, as I see it, hews more closely to "the strictures of the Seventh Amendment," Browning-Ferris, 492 U. S., at 280, n. 26. The Court's de novo standard is more complex. It requires lower courts to distinguish between ordinary common-law excessiveness and constitutional excessiveness, ante, at 433, and to separate out fact findings that qualify for "clearly erroneous" review, ante, at 440, n. 14. See also ante, at 440, n. 13 (suggesting abuse-of-discretion review might be in order "if a State were to adopt a scheme that tied the award of punitive damages more tightly to the jury's finding of compensatory damages"). The Court's approach will be challenging to administer. Complex as it is, I suspect that approach and mine will yield different outcomes in few cases.

The Ninth Circuit, I conclude, properly identified abuse of discretion as the appropriate standard in reviewing the District Court's determination that the punitive damages awarded against Cooper were not grossly excessive. For the Seventh Amendment and practical reasons stated, I would affirm the judgment of the Court of Appeals.