Justice ALITO delivered the opinion of the Court.

Pratik A. Shah, for U.S. as amicus curiae, by special leave of the Court, supporting the Petitioner.

Thurbert E. Baker, Atty. Gen. of Ga, Dennis R. Dunn, Deputy Atty. Gen., Shalen S. Nelson, Sen. Asst., Atty. Gen., Elizabeth M. Williamson, Asst. Atty. Gen., Atlanta, GA, Mark H. Cohen, Troutman Sanders LLP, Atlanta, Georgia, for Petitioners.

Marcia Robinson Lowry, Ira P. Lustbader, New York, NY, Jeffrey O. Bramlett, Michael A. Caplan, Bondurant, Mixson & Elmore, LLP, Atlanta, GA, Paul D. Clement, Adam Conrad, King & Spalding, LLP, Washington, DC, for Respondents.

Opinion

Justice ALITO delivered the opinion of the Court. This case presents the question whether the calculation of an attorney's fee, under federal fee-shifting statutes, based on the “lodestar,” i.e., the number of hours worked multiplied by the prevailing hourly rates, may be increased due to superior performance and results. We have stated in previous cases that such an increase is permitted in extraordinary circumstances, and we reaffirm that rule. But as we have also said in prior cases, there is a strong presumption that the lodestar is sufficient; factors subsumed in the lodestar calculation cannot be used as a ground for increasing an award above the lodestar; and a party seeking fees has the burden of identifying a factor that the lodestar does not adequately take into account and proving with specificity that an enhanced fee is justified. Because the District Court did not apply these standards, we reverse the decision below and remand for further proceedings consistent with this opinion. I

A

Respondents (plaintiffs below) are children in the Georgia foster-care system and their next friends. They filed this class action on behalf of 3,000 children in foster care and named as defendants the Governor of Georgia and various state officials (petitioners in this case). Claiming that deficiencies in the foster-care system in two counties near Atlanta violated their federal and state constitutional and statutory rights, respondents sought injunctive and declaratory relief, as well as attorney's fees and expenses.

The United States District Court for the Northern District of Georgia eventually referred the case to mediation, where the parties entered into a consent decree, which the District Court approved. The consent decree resolved all pending issues other than the fees that respondents' attorneys were entitled to receive under 42 U.S.C. § 1988.

B

Respondents submitted a request for more than $14 million in attorney's fees. Half of that amount was based on their calculation of the lodestar—roughly 30,000 hours multiplied by hourly rates of $200 to $495 for attorneys and $75 to $150 for non attorneys. In support of their fee request, respondents submitted affidavits asserting that these rates were within the range of prevailing market rates for legal services in the relevant market. The other half of the amount that respondents sought represented a fee enhancement for superior work and results. Affidavits submitted in support of this request claimed that the lodestar amount “would be generally insufficient to induce lawyers of comparable skill, judgment, professional representation and experience” to litigate this case. See, e.g., App. 80. Petitioners objected to the fee request, contending that some of the proposed hourly rates were too high, that the hours claimed were excessive, and that the enhancement would duplicate factors that were reflected in the lodestar amount.

The District Court awarded fees of approximately $10.5 million. See 454 F.Supp.2d 1260, 1296 (N.D.Ga.2006). The District Court found that the hourly rates proposed by respondents were “fair and reasonable,” id., at 1285, but that some of the entries on counsel's billing records were vague and that the hours claimed for many of the billing categories were excessive. The court therefore cut the non-travel hours by 15% and halved the hourly rate for travel hours. This resulted in a lodestar calculation of approximately $6 million.

The court then enhanced this award by 75%, concluding that the lodestar calculation did not take into account “(1) the fact that class counsel were required to advance case expenses of $1.7 million over a three-year period with no on[-]going reimbursement, (2) the fact that class counsel were not paid on an on-going basis as the work was being performed, and (3) the fact that class counsel's ability to recover a fee and expense reimbursement were completely contingent on the outcome of the case.” Id., at 1288. The court stated that respondents' attorneys had exhibited “a higher degree of skill, commitment, dedication, and professionalism ... than the Court has seen displayed by the attorneys in any other case during its 27 years on the bench.” Id., at 1289. The court also commented that the results obtained were “ ‘extraordinary’ ” and added that “[a]fter 58 years as a practicing attorney and federal judge, the Court is unaware of any other case in which a plaintiff class has achieved such a favorable result on such a comprehensive scale.” Id., at 1290. The enhancement resulted in an additional $4.5 million fee award.

Relying on prior Circuit precedent, a panel of the Eleventh Circuit affirmed. 532 F.3d 1209 (2008). The panel held that the District Court had not abused its discretion by failing to make a larger reduction in the number of hours for which respondents' attorneys sought reimbursement, but the panel commented that it “would have cut the billable hours more if we were deciding the matter in the first instance” and added that the hourly rates approved by the District Court also “appear[ed] to be on the generous side.” Id., at 1220, and n. 2. On the question of the enhancement, however, the panel splintered, with each judge writing a separate opinion.

Judge Carnes concluded that binding Eleventh Circuit precedent required that the decision of the District Court be affirmed, but he opined that the reasoning in our opinions suggested that no enhancement should be allowed in this case. He concluded that the quality of the attorneys' performance was “adequately accounted for ‘either in determining the reasonable number of hours expended on the litigation or in setting the reasonable hourly rates.’ ” Id., at 1225 (quoting Pennsylvania v. Delaware Valley Citizens' Council for Clean Air, 478 U.S. 546, 565–566, 106 S.Ct. 3088, 92 L.Ed.2d 439 (1986) (Delaware Valley I) ). He found that an enhancement could not be justified based on delay in the recovery of attorney's fees and reimbursable expenses because such delay is a routine feature of cases brought under 42 U.S.C. § 1983. And he reasoned that the District Court had contravened our holding in Burlington v. Dague, 505 U.S. 557, 112 S.Ct. 2638, 120 L.Ed.2d 449 (1992), when it relied on “ ‘the fact that class counsel's compensation was totally contingent upon prevailing in this action.’ ” 532 F.3d, at 1226, 1228 (quoting affidavit in support of fee request). Judge Wilson concurred in the judgment but disagreed with Judge Carnes' view that Eleventh Circuit precedent is inconsistent with our decisions. Judge Hill also concurred in the judgment but expressed no view about the correctness of the prior Circuit precedent.

The Eleventh Circuit denied rehearing en banc over the dissent of three judges. See 547 F.3d 1319 (2008). Judge Wilson filed an opinion concurring in the denial of rehearing; Judge Carnes, joined by Judges Tjoflat and Dubina, filed an opinion dissenting from the denial of rehearing; and Judge Tjoflat filed a separate dissent, contending, among other things, that the District Court, by basing the enhancement in large part on a comparison of the performance of respondents' attorneys with all of the unnamed attorneys whose work he had observed during his professional career, had improperly rendered a decision that was effectively unreviewable on appeal and had essentially served as a witness in support of the enhancement. Id., at 1326–1327.

We granted certiorari. 556 U.S. 1165, 129 S.Ct. 1907, 173 L.Ed.2d 1056 (2009).

II

The general rule in our legal system is that each party must pay its own attorney's fees and expenses, see Hensley v. Eckerhart, 461 U.S. 424, 429, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983), but Congress enacted 42 U.S.C. § 1988 in order to ensure that federal rights are adequately enforced. Section 1988 provides that a prevailing party in certain civil rights actions may recover “a reasonable attorney's fee as part of the costs.” Unfortunately, the statute does not explain what Congress meant by a “reasonable” fee, and therefore the task of identifying an appropriate methodology for determining a “reasonable” fee was left for the courts.

One possible method was set out in Johnson v. Georgia Highway Express,

Inc., 488 F.2d 714, 717–719 (C.A.5 1974), which listed 12 factors that a court should consider in determining a reasonable fee. This method, however, “gave very little actual guidance to district courts. Setting attorney's fees by reference to a series of sometimes subjective factors placed unlimited discretion in trial judges and produced disparate results.” Delaware Valley I, supra, at 563, 106 S.Ct. 3088.

An alternative, the lodestar approach, was pioneered by the Third Circuit in Lindy Bros. Builders, Inc. of Philadelphia v. American Radiator & Standard Sanitary Corp., 487 F.2d 161 (1973), appeal after remand, 540 F.2d 102 (1976), and “achieved dominance in the federal courts” after our decision in Hensley. Gisbrecht v. Barnhart, 535 U.S. 789, 801, 122 S.Ct. 1817, 152 L.Ed.2d 996 (2002). “Since that time, ‘[t]he “lodestar” figure has, as its name suggests, become the guiding light of our fee-shifting jurisprudence.’ ” Ibid. (quoting Dague, supra, at 562, 112 S.Ct. 2638).

Although the lodestar method is not perfect, it has several important virtues. First, in accordance with our understanding of the aim of fee-shifting statutes, the lodestar looks to “the prevailing market rates in the relevant community.” Blum v. Stenson, 465 U.S. 886, 895, 104 S.Ct. 1541, 79 L.Ed.2d 891 (1984). Developed after the practice of hourly billing had become widespread, see Gisbrecht, supra, at 801, 122 S.Ct. 1817, the lodestar method produces an award that roughly approximates the fee that the prevailing attorney would have received if he or she had been representing a paying client who was billed by the hour in a comparable case. Second, the lodestar method is readily administrable, see Dague, 505 U.S., at 566, 112 S.Ct. 2638; see also Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health and Human Resources, 532 U.S. 598, 609, 121 S.Ct. 1835, 149 L.Ed.2d 855 (2001); and unlike the Johnson approach, the lodestar calculation is “objective,” Hensley, supra, at 433, 103 S.Ct. 1933, and thus cabins the discretion of trial judges, permits meaningful judicial review, and produces reasonably predictable results.

III

Our prior decisions concerning the federal fee-shifting statutes have established six important rules that lead to our decision in this case.

First, a “reasonable” fee is a fee that is sufficient to induce a capable attorney to undertake the representation of a meritorious civil rights case. See Delaware Valley I, 478 U.S., at 565, 106 S.Ct. 3088 (“[I]f plaintiffs ... find it possible to engage a lawyer based on the statutory assurance that he will be paid a ‘reasonable fee,’ the purpose behind the fee-shifting statute has been satisfied”); Blum, supra, at 897, 104 S.Ct. 1541 (“[A] reasonable attorney's fee is one that is adequate to attract competent counsel, but that does not produce windfalls to attorneys” (ellipsis, brackets, and internal quotation marks omitted)). Section 1988's aim is to enforce the covered civil rights statutes, not to provide “a form of economic relief to improve the financial lot of attorneys.” Delaware Valley I, supra, at 565, 106 S.Ct. 3088.

Second, the lodestar method yields a fee that is presumptively sufficient to achieve this objective. See Dague, supra, at 562, 112 S.Ct. 2638; Delaware Valley I, supra, at 565, 106 S.Ct. 3088; Blum, supra, at 897, 104 S.Ct. 1541; see also Gisbrecht, supra, at 801–802, 122 S.Ct. 1817. Indeed, we have said that the presumption is a “strong” one. Dague, supra, at 562, 112 S.Ct. 2638; Delaware Valley I, supra, at 565, 106 S.Ct. 3088.

Third, although we have never sustained an enhancement of a lodestar amount for performance, see Brief for United States as Amicus Curiae 12, 17, we have repeatedly said that enhancements may be awarded in “ ‘rare’ ” and “ ‘exceptional’ ” circumstances. Delaware Valley I, supra, at 565, 106 S.Ct. 3088; Blum, supra, at 897, 104 S.Ct. 1541; Hensley, 461 U.S., at 435, 103 S.Ct. 1933. Fourth, we have noted that “the lodestar figure includes most, if not all, of the relevant factors constituting a ‘reasonable’ attorney's fee,” Delaware Valley I, supra, at 566, 106 S.Ct. 3088, and have held that an enhancement may not be awarded based on a factor that is subsumed in the lodestar calculation, see Dague, supra, at 562–563, 112 S.Ct. 2638; Pennsylvania v. Delaware Valley Citizens' Council for Clean Air, 483 U.S. 711, 726–727, 107 S.Ct. 3078, 97 L.Ed.2d 585 (1987) (Delaware Valley II) (plurality opinion); Blum, 465 U.S., at 898, 104 S.Ct. 1541. We have thus held that the novelty and complexity of a case generally may not be used as a ground for an enhancement because these factors “presumably [are] fully reflected in the number of billable hours recorded by counsel.” Ibid. We have also held that the quality of an attorney's performance generally should not be used to adjust the lodestar “[b]ecause considerations concerning the quality of a prevailing party's counsel's representation normally are reflected in the reasonable hourly rate.” Delaware Valley I, supra, at 566, 106 S.Ct. 3088.

Fifth, the burden of proving that an enhancement is necessary must be borne by the fee applicant. Dague, supra, at 561, 112 S.Ct. 2638; Blum, 465 U.S., at 901–902, 104 S.Ct. 1541.

Finally, a fee applicant seeking an enhancement must produce “specific evidence” that supports the award. Id., at 899, 901, 104 S.Ct. 1541 (An enhancement must be based on “evidence that enhancement was necessary to provide fair and reasonable compensation”). This requirement is essential if the lodestar method is to realize one of its chief virtues, i.e., providing a calculation that is objective and capable of being reviewed on appeal.

IV

A

In light of what we have said in prior cases, we reject any contention that a fee determined by the lodestar method may not be enhanced in any situation. The lodestar method was never intended to be conclusive in all circumstances. Instead, there is a “strong presumption” that the lodestar figure is reasonable, but that presumption may be overcome in those rare circumstances in which the lodestar does not adequately take into account a factor that may properly be considered in determining a reasonable fee.

B

In this case, we are asked to decide whether either the quality of an attorney's performance or the results obtained are factors that may properly provide a basis for an enhancement. We treat these two factors as one. When a plaintiff's attorney achieves results that are more favorable than would have been predicted based on the governing law and the available evidence, the outcome may be attributable to superior performance and commitment of resources by plaintiff's counsel. Or the outcome may result from inferior performance by defense counsel, unanticipated defense concessions, unexpectedly favorable rulings by the court, an unexpectedly sympathetic jury, or simple luck. Since none of these latter causes can justify an enhanced award, superior results are relevant only to the extent it can be shown that they are the result of superior attorney performance. Thus, we need only consider whether superior attorney performance can justify an enhancement. And in light of the principles derived from our prior cases, we inquire whether there are circumstances in which superior attorney performance is not adequately taken into account in the lodestar calculation. We conclude that there are a few such circumstances but that these circumstances are indeed “rare” and “exceptional,” and require specific evidence that the lodestar fee would not have been “adequate to attract competent counsel,” Blum, supra, at 897, 104 S.Ct. 1541 (internal quotation marks omitted).

First, an enhancement may be appropriate where the method used in determining the hourly rate employed in the lodestar calculation does not adequately measure the attorney's true market value, as demonstrated in part during the litigation. This may occur if the hourly rate is determined by a formula that takes into account only a single factor (such as years since admission to the bar) or perhaps only a few similar factors. In such a case, an enhancement may be appropriate so that an attorney is compensated at the rate that the attorney would receive in cases not governed by the federal fee-shifting statutes. But in order to provide a calculation that is objective and reviewable, the trial judge should adjust the attorney's hourly rate in accordance with specific proof linking the attorney's ability to a prevailing market rate.

Second, an enhancement may be appropriate if the attorney's performance includes an extraordinary outlay of expenses and the litigation is exceptionally protracted. As Judge Carnes noted below, when an attorney agrees to represent a civil rights plaintiff who cannot afford to pay the attorney, the attorney presumably understands that no reimbursement is likely to be received until the successful resolution of the case, 532 F.3d, at 1227, and therefore enhancements to compensate for delay in reimbursement for expenses must be reserved for unusual cases. In such exceptional cases, however, an enhancement may be allowed, but the amount of the enhancement must be calculated using a method that is reasonable, objective, and capable of being reviewed on appeal, such as by applying a standard rate of interest to the qualifying outlays of expenses. Third, there may be extraordinary circumstances in which an attorney's performance involves exceptional delay in the payment of fees. An attorney who expects to be compensated under § 1988 presumably understands that payment of fees will generally not come until the end of the case, if at all. See 532 F.3d, at 1227 (opinion of Carnes, J.). Compensation for this delay is generally made “either by basing the award on current rates or by adjusting the fee based on historical rates to reflect its present value.” Missouri v. Jenkins, 491 U.S. 274, 282, 109 S.Ct. 2463, 105 L.Ed.2d 229 (1989) (internal quotation marks omitted). But we do not rule out the possibility that an enhancement may be appropriate where an attorney assumes these costs in the face of unanticipated delay, particularly where the delay is unjustifiably caused by the defense. In such a case, however, the enhancement should be calculated by applying a method similar to that described above in connection with exceptional delay in obtaining reimbursement for expenses.

We reject the suggestion that it is appropriate to grant performance enhancements on the ground that departures from hourly billing are becoming more common. As we have noted, the lodestar was adopted in part because it provides a rough approximation of general billing practices, and accordingly, if hourly billing becomes unusual, an alternative to the lodestar method may have to be found. However, neither respondents nor their amici contend that that day has arrived. Nor have they shown that permitting the award of enhancements on top of the lodestar figure corresponds to prevailing practice in the general run of cases.

We are told that, under an increasingly popular arrangement, attorneys are paid at a reduced hourly rate but receive a bonus if certain specified results are obtained, and this practice is analogized to the award of an enhancement such as the one in this case. Brief for Respondents 55–57. The analogy, however, is flawed. An attorney who agrees, at the outset of the representation, to a reduced hourly rate in exchange for the opportunity to earn a performance bonus is in a position far different from an attorney in a § 1988 case who is compensated at the full prevailing rate and then seeks a performance enhancement in addition to the lodestar amount after the litigation has concluded. Reliance on these comparisons for the purposes of administering enhancements, therefore, is not appropriate.

V

In the present case, the District Court did not provide proper justification for the large enhancement that it awarded. The court increased the lodestar award by 75% but, as far as the court's opinion reveals, this figure appears to have been essentially arbitrary. Why, for example, did the court grant a 75% enhancement instead of the 100% increase that respondents sought? And why 75% rather than 50% or 25% or 10%?

The District Court commented that the enhancement was the “minimum enhancement of the lodestar necessary to reasonably compensate [respondents'] counsel.” 454 F.Supp.2d, at 1290. But the effect of the enhancement was to increase the top rate for the attorneys to more than $866 per hour, and the District Court did not point to anything in the record that shows that this is an appropriate figure for the relevant market.

The District Court pointed to the fact that respondents' counsel had to make extraordinary outlays for expenses and had to wait for reimbursement, id., at 1288, but the court did not calculate the amount of the enhancement that is attributable to this factor. Similarly, the District Court noted that respondents' counsel did not receive fees on an ongoing basis while the case was pending, but the court did not sufficiently link this factor to proof in the record that the delay here was outside the normal range expected by attorneys who rely on § 1988 for the payment of their fees or quantify the disparity. Nor did the court provide a calculation of the cost to counsel of any extraordinary and unwarranted delay. And the court's reliance on the contingency of the outcome contravenes our holding in Dague. See 505 U.S., at 565, 112 S.Ct. 2638.

Finally, insofar as the District Court relied on a comparison of the performance of counsel in this case with the performance of counsel in unnamed prior cases, the District Court did not employ a methodology that permitted meaningful appellate review. Needless to say, we do not question the sincerity of the District Court's observations, and we are in no position to assess their accuracy. But when a trial judge awards an enhancement on an impressionistic basis, a major purpose of the lodestar method—providing an objective and reviewable basis for fees, see id., at 566, 112 S.Ct. 2638—is undermined.

Determining a “reasonable attorney's fee” is a matter that is committed to the sound discretion of a trial judge, see 42 U.S.C. § 1988 (permitting court, “in its discretion,” to award fees), but the judge's discretion is not unlimited. It is essential that the judge provide a reasonably specific explanation for all aspects of a fee determination, including any award of an enhancement. Unless such an explanation is given, adequate appellate review is not feasible, and without such review, widely disparate awards may be made, and awards may be influenced (or at least, may appear to be influenced) by a judge's subjective opinion regarding particular attorneys or the importance of the case. In addition, in future cases, defendants contemplating the possibility of settlement will have no way to estimate the likelihood of having to pay a potentially huge enhancement. See Marek v. Chesny, 473 U.S. 1, 7, 105 S.Ct. 3012, 87 L.Ed.2d 1 (1985) ( “ ‘[M]any a defendant would be unwilling to make a binding settlement offer on terms that left it exposed to liability for attorney's fees in whatever amount the court might fix on motion of the plaintiff’ ”).

Section 1988 serves an important public purpose by making it possible for persons without means to bring suit to vindicate their rights. But unjustified enhancements that serve only to enrich attorneys are not consistent with the statute's aim. In many cases, attorney's fees awarded under § 1988 are not paid by the individuals responsible for the constitutional or statutory violations on which the judgment is based. Instead, the fees are paid in effect by state and local taxpayers, and because state and local governments have limited budgets, money that is used to pay attorney's fees is money that cannot be used for programs that provide vital public services. Cf. Horne v. Flores, 557 U.S. 433, ––––, 129 S.Ct. 2579, 2593–2594, 174 L.Ed.2d 406 (2009) (payment of money pursuant to a federal-court order diverts funds from other state or local programs).

* * * For all these reasons, the judgment of the Court of Appeals is reversed, and the case is remanded for proceedings consistent with this opinion.

It is so ordered.

Justice KENNEDY, concurring.

If one were to ask an attorney or a judge to name the significant cases of his or her career, it would be unsurprising to find the list includes a case then being argued or just decided. When immersed in a case, lawyers and judges find within it a fascination, an intricacy, an importance that transcends what the detached observer sees. So the pending or just completed case will often seem extraordinary to its participants. That is the dynamic of the adversary system, the system that so well serves the law.

It is proper for the Court today to reject the proposition that all enhancements are barred; still, it must be understood that extraordinary cases are presented only in the rarest circumstances.

With these comments, I join in full the opinion of the Court.

Justice THOMAS, concurring.

Nearly 30 years ago, a group of attorneys sought a fee award under 42 U.S.C. § 1988 after “achiev[ing] only limited success” litigating their clients' constitutional claims. Hensley v. Eckerhart, 461 U.S. 424, 431, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983). This Court's opinion resolving their claim for fees observed that “in some cases of exceptional success an enhanced award” of attorney's fees under § 1988 “may be justified.” Id., at 435, 103 S.Ct. 1933 (emphasis added). That observation plainly was dicta, but one year later this Court relied on it to reject the “argument that an ‘upward adjustment’ ” to the lodestar calculation “is never permissible.” Blum v. Stenson, 465 U.S. 886, 897, 104 S.Ct. 1541, 79 L.Ed.2d 891 (1984). Yet “we have never sustained an enhancement of a lodestar amount for performance,” ante, at 1673, and our jurisprudence since Blum has charted “a decisional arc that bends decidedly against enhancements,” 532 F.3d 1209, 1221 (C.A.11 2008) (Carnes, J.). See also ante, at 1672 – 1674.

Today the Court holds, consistent with Hensley and Blum, that a lodestar fee award under § 1988 may be enhanced for attorney performance in a “few” circumstances that “are indeed ‘rare’ and ‘exceptional.’ ” Ante, at 1674. But careful readers will observe the precise limitations that the Court imposes on the availability of such enhancements. See ante, at 1674 – 1675; see also ante, at 1677 (KENNEDY, J., concurring) (“[I]t must be understood that extraordinary cases are presented only in the rarest circumstances”). These limitations preserve our prior cases and advance our attorney's fees jurisprudence further along the decisional arc that Judge Carnes described. I agree with the Court's approach and its conclusion because, as the Court emphasizes, see ante, at 1672 – 1673, the lodestar calculation will in virtually every case already reflect all indicia of attorney performance relevant to a fee award.

Justice BREYER, with whom Justice STEVENS, Justice GINSBURG, and Justice SOTOMAYOR join, concurring in part and dissenting in part.

We granted certiorari in this case to consider “whether the calculation of an attorney's fee” that is “based on the ‘lodestar,’ ” ante, at 1669 (opinion of the Court), can “ever be enhanced based solely on [the] quality of [the lawyers'] performance and [the] results obtained,” Pet. for Cert. i (emphasis added). The Court answers that question in the affirmative. See ante, at 1669 (“We have stated in previous cases that such an increase is permitted in extraordinary circumstances, and we reaffirm that rule”); see also ante, p. 1677 (KENNEDY, J., concurring). As our prior precedents make clear, the lodestar calculation “does not end the [fee] inquiry” because there “remain other considerations that may lead the district court to adjust the fee upward.” Hensley v. Eckerhart, 461 U.S. 424, 434, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983). For that reason, “[t]he lodestar method was never intended to be conclusive in all circumstances.” Ante, at 1674. Instead, as the Court today reaffirms, when “superior attorney performance,” ibid., leads to “exceptional success an enhanced award may be justified,” Hensley supra, at 435, 103 S.Ct. 1933 ; see also Pennsylvania v. Delaware Valley Citizens' Council for Clean Air, 478 U.S. 546, 565, 106 S.Ct. 3088, 92 L.Ed.2d 439 (1986); Blum v. Stenson, 465 U.S. 886, 896–900, 104 S.Ct. 1541, 79 L.Ed.2d 891 (1984). I agree with that conclusion.

Where the majority and I part ways is with respect to a question that is not presented, but that the Court obliquely, and in my view inappropriately, appears to consider nonetheless—namely, whether the lower courts correctly determined in this case that exceptional circumstances justify a lodestar enhancement. See Parts IV–V, ante; see also ante, p. 1677 (KENNEDY, J., concurring). I would not reach that issue, which lies beyond the narrow question that we agreed to consider. See 556 U.S. 1165, 129 S.Ct. 1907, 173 L.Ed.2d 1056 (2009) (limiting review to the first question presented); Pet. for Cert. i (stating question); see also Glover v. United States, 531 U.S. 198, 205, 121 S.Ct. 696, 148 L.Ed.2d 604 (2001) (“As a general rule ... we do not decide issues outside the questions presented ...”). Nor do I believe that this Court, which is twice removed from the litigation underlying the fee determination, is properly suited to resolve the fact-intensive inquiry that 42 U.S.C. § 1988 demands. But even were I to engage in that inquiry, I would hold that the District Court did not abuse its discretion in awarding an enhancement. And I would therefore affirm the judgment of the Court of Appeals.

As the Court explains, the basic question that must be resolved when considering an enhancement to the lodestar is whether the lodestar calculation “adequately measure[s]” an attorney's “value,” as “demonstrated” by his performance “during the litigation.” Ante, at 1674. While I understand the need for answering that question through the application of standards, I also believe that the answer inevitably involves an element of judgment. Moreover, when reviewing a district court's answer to that question, an appellate court must inevitably give weight to the fact that a district court is better situated to provide that answer. For it is the district judge, and only the district judge, who will have read all of the motions filed in the case, witnessed the proceedings, and been able to evaluate the attorneys' overall performance in light of the objectives, context, legal difficulty, and practical obstacles present in the case. In a word, the district judge will have observed the attorney's true “value, as demonstrated ... during the litigation.” Ibid. (emphasis added). By contrast, a court of appeals, faced with a cold and perhaps lengthy record, will inevitably have less time and opportunity to determine whether the lawyers have done an exceptionally fine job. And this Court is yet less suited to performing that inquiry. Accordingly, determining whether a fee enhancement is warranted in a given case “is a matter that is committed to the sound discretion of a trial judge,” ante, at 1676, and the function of appellate courts is to review that judge's determination for an abuse of such discretion. See Pierce v. Underwood, 487 U.S. 552, 571, 108 S.Ct. 2541, 101 L.Ed.2d 490 (1988); see also General Elec. Co. v. Joiner, 522 U.S. 136, 143, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997) (“[D]eference ... is the hallmark of abuse-of-discretion review”).

This case well illustrates why our tiered and functionally specialized judicial system places the task of determining an attorney's fee award primarily in the district court's hands. The plaintiffs' lawyers spent eight years investigating the underlying facts, developing the initial complaint, conducting court proceedings, and working out final relief. The District Court's docket, with over 600 entries, consists of more than 18,000 pages. Transcripts of hearings and depositions, along with other documents, have produced a record that fills 20 large boxes. Neither we, nor an appellate panel, can easily read that entire record. Nor should we attempt to second-guess a district judge who is aware of the many intangible matters that the written page cannot reflect.

My own review of this expansive record cannot possibly be exhaustive. But those portions of the record I have reviewed lead me to conclude, like the Court of Appeals, that the District Judge did not abuse his discretion when awarding an enhanced fee. I reach this conclusion based on four considerations.

First, the record indicates that the lawyers' objective in this case was unusually important and fully consistent with the central objectives of the basic federal civil-rights statute, 42 U.S.C. § 1983. Moreover, the problem the attorneys faced demanded an exceptionally high degree of skill and effort. Specifically, these lawyers and their clients sought to have the State of Georgia reform its entire foster-care system—a system that much in the record describes as well below the level of minimal constitutional acceptability. The record contains investigative reports, mostly prepared by Georgia's own Office of the Child Advocate, which show, for example, the following:

• The State's foster-care system was unable to provide essential medical and mental health services; children consequently and unnecessarily suffered illness and life long medical disabilities, such as permanent hearing loss, due to failures on the part of the State

to administer basic care and antibiotics. See, e.g., Doc. 3, Exh. 3C, pp. 11–13.

• Understaffing and improper staffing placed children in the care of individuals with dangerous criminal records; children were physically assaulted by the staff, locked outside of the shelters at night as punishment, and abused in other ways. See, e.g., Doc. 50, pp. 32–36, 55; Doc. 3, Exh. 3A, pp. 2–6; Doc. 3, Exh. 2, pp. 4–5; Doc. 52, Exh. 1, pp. 6, 12–15, 34.

•

The shelters themselves were “unsanitary and dilapidated,” “ unclean,” infested with rats, “overcrowded,” unsafe, and “ ‘out of control.’ ” See, e.g., Doc. 3, Exh. 3A, at 1–2; Doc. 3, Exh. 3B, p. 2; Doc. 50, at 29.

• Due to improper supervision and other deficiencies at the shelters, 20% of the children abused drugs; some also became victims of child prostitution. See id., at 39; Doc. 3, Exh. 3A, at 3.

• Systemic failures also caused vulnerable children to suffer regular beatings and sexual abuse, including rape, at the hands of more aggressive shelter residents. See, e.g., Doc. 50, at 18–22, 54–55; Doc. 52, Exh. 1, at 7–10, 26; Doc. 3, Exh. 3B, at 3 (“[A child] was beaten so badly by eight other [children] that he suffered severe internal bleeding”); id., at 4 (describing violent sexual assault and rape).

• Not surprisingly, many children—upwards of 5 per day and over 750 per year—tried to escape these conditions; others tried to commit suicide. See, e.g., Doc. 50, at 27–28, 54; Doc. 52, Exh. 18, p. 4 (under seal) (at least 25% of children run away from shelters); Doc. 52, Exh. 18E, pp. 1–11, 18–19 (under seal) (daily logs); see also Doc. 50, Exh. 1, at 37, 54 (describing suicide attempts) (all docket entries above and hereinafter refer to No. 1:02–cv–1686 (ND Ga.) (case below)).

The State's Office of the Child Advocate, whose reports provide much of the basis for the foregoing description, concluded that the system was “operating in crisis mode” and that any private operator who ran such a system “would never be licensed to care for children.” Office of the Child Advocate for the Protection of Children Ann. Rep. 10, 14 (2001), Record, Doc. 3, Exh. 3C (hereinafter OCA 2001 Report); accord, id., Exh. 3A, at 1. The advocate noted that neither her investigative reports nor national news publicity (including a television program that highlighted a 5–year–old foster child's death from beatings) had prompted corrective action by the State. OCA 2001 Report 1, 14.

The advocate further stated that litigation was necessary to force reform. Id., at 14–15. And she repeatedly asked the State to give her office the authority to conduct that litigation. See Office of the Child Advocate Advisory Committee,Ann.Effectiveness Rep.4(2002),onlineathttp://www.georgia. gov/vgn/images/portal/cit _ 1210/7/22/ 84622967effectiveness2003.pdf (all Internet materials as visited Apr. 16, 2010, and available in Clerk of Court's case file) (“[F]or the Office to be truly effective, it must possess the authority to compel change [and] ... to initiate litigation on behalf of children. Such authority is widely considered by other states' Child Advocates as crucial to effecting meaningful change for children”); Office of the Child Advocate Advisory Committee, Ann. Effectiveness Rep. 13 (2003–2004) (same), online at http://gachildadvocate.org/ vgn/images/portal/cit _ 1210/48/16/84624761OCA _ Effectivenes s_ Report2003 _ 2004.doc; Office of the Child Advocate Advisory Committee, Ann. Effectiveness Rep. 11 (2004–2005) (same), online at http://www.georgia.gov/vgn/images/ portal/ cit_ 1210/31/23/102387685OCA% 20Effectiveness% 20Report % 202004–2005.doc. But the State did not grant the Child Advocate's office the litigating authority she sought. See 2000 Ga. Laws p. 245, as codified, Ga.Code Ann. § 15–11–173 (2008).

The upshot is that the plaintiffs' attorneys did what the child advocate could not do: They initiated this lawsuit. They thereby assumed the role of “a ‘private attorney general’ ” by filling an enforcement void in the State's own legal system, a function “that Congress considered of the highest priority,” Newman v. Piggie Park Enterprises, Inc., 390 U.S. 400, 402, 88 S.Ct. 964, 19 L.Ed.2d 1263 (1968) (per curiam), and “meant to promote in enacting § 1988,” Texas State Teachers Assn. v. Garland Independent School Dist., 489 U.S. 782, 793, 109 S.Ct. 1486, 103 L.Ed.2d 866 (1989).

Second, the course of the lawsuit was lengthy and arduous. The plaintiffs and their lawyers began with factual investigations beyond those which the child advocate had already conducted. See, e.g., Record, Docs. 50–52 (partially under seal). They then filed suit. And the State met the plaintiffs' efforts with a host of complex procedural, as well as substantive, objections. The State, for example, argued that the law forbade the plaintiffs to investigate the shelters; on the eve of a state-court decision that might have approved the investigations, the State then removed the case to federal court; the State then sought protective orders preventing the attorneys from speaking to the shelters' staff; and, after losing its motions, the State delayed to the point where the District Court “was forced to admonish [the] State Defendants for ‘relying on technical legal objections to discovery requests in order to delay and hinder the discovery process.’ ” 454 F.Supp.2d. 1260, 1268 (N.D.Ga.2006) (quoting Record, Doc. 145, p. 4). See also Doc. 1; Doc. 3, pp. 9–10; Docs. 26, 28–29, 44, 60.

In the meantime, the State moved for dismissal, basing the motion on complex legal doctrines such as Younger abstention and the Rooker–Feldman doctrine, which the District Court found inapplicable. 218 F.R.D. 277, 284–290 (N.D.Ga.2003). See Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971); Rooker v. Fidelity Trust Co., 263 U.S. 413, 44 S.Ct. 149, 68 L.Ed. 362 (1923), and District of Columbia Court of Appeals v. Feldman, 460 U.S. 462, 103 S.Ct. 1303, 75 L.Ed.2d 206 (1983). The State also opposed the petitioners' request to certify a class of the 3,000 children in foster care, but the District Court again rejected the State's argument. 218 F.R.D., at 299–302. And, after that, the State filed a lengthy motion for summary judgment, Record, Docs. 243–245, which plaintiffs' attorneys opposed in thorough briefing supported by comprehensive exhibits, see Docs. 254–258, 260. After losing that motion and eventually agreeing to mediation, the State forced protracted litigation as to who should be the mediator. See Docs. 363–364, 366, 369–370, 373, 376, 380. All told, in opposing the plaintiffs' efforts to have the foster-care system reformed, the State spent $2.4 million on outside counsel (who, because they charge the State reduced rates, worked significantly more hours than that figure alone indicates) and tapped its own law department for an additional 5,200 hours of work. 454 F.Supp.2d, at 1287.

Third, in the face of this opposition, the results obtained by the plaintiffs' attorneys appear to have been exceptional. The 47–page consent decree negotiated over the course of the mediation sets forth 31 specific steps that the State will take in order to address the specific deficiencies of the sort that I described above. See id., at 1289; see also App. 92–207 (consent decree). And it establishes a reporting and oversight mechanism that is backed up by the District Court's enforcement authority. See 454 F.Supp.2d, at 1289. As a result of the decree, the State agreed to comprehensive reforms of its foster-care system, to the benefit of children in many different communities. And informed observers have described the decree as having brought about significant positive results. See, e.g., Record, Doc. 632, p. 4 (most recent court-appointed overseers' report) (“The State's overall performance ... continues the trend of steady improvement ...”); id., at 4–10 (detailing substantial health, safety, and welfare improvements); see also, Office of the Child Advocate Ann. Report (2008), Letter from Tom C. Rawlings, Child Advocate, to Sonny Perdue, Governor of Georgia (Jan. 16, 2009), online at http://oca.georgia.gov/vgn/images/portal/cit_ 1210/48/0/131408008OCA% 202008% 20Annual% 20Report.pdf (“[W]e are generally pleased with the direction of our state's child welfare system ...”); cf. Weinstein & Weinstein, Before It's Too Late: Neuropsychological Consequences of Child Neglect and their Implications for Law and Social Policy, 33 U. Mich. J. L. Reform 561, 590–591 (2000) (describing in general the broad social impact of dysfunctional child-welfare systems (quoting National Institute of Health, Research on Child Neglect (1999), online at http:// grants.nih.gov/grants/guide/rfa–files/RFA–OD–99–06.html)). But see Record, Doc. 632, at 10–13 (noting areas in which Georgia's system still needs improvement).

Fourth and finally, the District Judge, who supervised these proceedings, who saw the plaintiffs amass, process, compile, and convincingly present vast amounts of factual information, who witnessed their defeat of numerous state procedural and substantive motions, and who was in a position to evaluate the ultimate mediation effort, said:

1. the “mediation effort in this case went far beyond anything that this Court has seen in any previous case,” 454 F.Supp.2d, at 1282;

2. “based on its personal observation of plaintiffs' counsel's performance throughout this litigation, the Court finds that ... counsel brought a higher degree of skill, commitment, dedication, and professionalism to this litigation than the Court has seen displayed by the attorneys in any other case during its 27 years on the bench,” id., at 1288–1290;

3. the Consent Decree “provided extraordinary benefits to the plaintiff class ... .” Id., at 1282. “[T]he settlement achieved by plaintiffs' counsel is comprehensive in its scope and detailed in its coverage .... After 58 years as a practicing attorney and federal judge, the Court is unaware of any other case in which a plaintiff class has achieved such a favorable result on such a comprehensive scale,” id., at 1289–1290.

Based on these observations and on its assessment of the attorneys' performance during the course of the litigation, the District Court concluded that “the evidence establishes that the quality of service rendered by class counsel ... was far superior to what consumers of legal services in the legal marketplace ... could reasonably expect to receive for the rates used in the lodestar calculation.” Id., at 1288.

On the basis of what I have read, I believe that assessment was correct. I recognize that the ordinary lodestar calculation yields a large fee award. But by my assessment, the lodestar calculation in this case translates to an average hourly fee per attorney of $249. See id., at 1287 (lodestar calculation and attorney hours). (The majority's reference to an hourly fee of $866, ante, at 1675, refers to the rate associated with the single highest paid of the 17 attorneys under the enhanced fee, not the average hourly rate under the lodestar. The lay reader should also bear in mind that a lawyer's “fee” is substantially greater than his “profit,” given that attorneys must sometimes cover case-specific costs (which in this case exceeded $800,000, see 454 F.Supp.2d, at 1291) and also must cover routine overhead expenses, which typically consume 40% of their fees, see Altman Weil Publications, Inc., Survey of Law Firm Economics 30 (2007 ed.)).

At $249 per hour, the lodestar would compensate this group of attorneys—whom the District Court described as extraordinary—at a rate lower than the average rate charged by attorneys practicing law in the State of Georgia, where the average hourly rate is $268. See id., at 89. Accordingly, even the majority would seem to acknowledge that some form of an enhancement is appropriate in this case. See ante, at 1674 (“[A]n enhancement may be appropriate where the method used in determining the hourly rate employed in the lodestar calculation does not adequately measure the attorney's true market value, as demonstrated in part during the litigation”). Indeed, the fact that these exceptional results were achieved in a case where “much of the work,” ante, at 1675 – 1676, n. 7, was performed by relatively inexperienced attorneys (who, accordingly, would be compensated by the lodestar “below the market average,” ibid.) is all the more reason to think that their service rendered their outstanding performance worthy of an enhancement. By comparison, the District Court's enhanced award—a special one-time adjustment unique to this exceptional case—would compensate these attorneys, on this one occasion, at an average hourly rate of $435, which is comparable to the rates charged by the Nation's leading law firms on average on every occasion. See Firm–by–Firm Sampling of Billing Rates Nationwide, National Law Journal, Dec. 11, 2006, p. S2 (listing 13 firms at which average hourly rate is between $400 and $510); J. Barnett, Certification Drag: The Opinion Puzzle and Other Transactional Curiosities, 33 Iowa J. Corp. L. 95, 110, n. 58 (2007) (“These numbers are probably an underestimate given that many of the highest-billing national law firms decline to take part in the National Law Journal Survey”). Thus, it would appear that the enhanced award is wholly consistent with the purpose of § 1988, which was enacted to ensure that “ counsel for prevailing parties [are] paid as is traditional with attorneys compensated by a fee-paying client.” S.Rep. No. 94–1011, p. 6 (1976); see H.R.Rep. No. 94–1558, p. 9 (1976) (“[C]ivil rights plaintiffs should not be singled out for different and less favorable treatment”); see also Blum, 465 U.S., at 893, 897, 104 S.Ct. 1541.

In any event, the circumstances I have listed likely make this a “rare” or “exceptional” case warranting an enhanced fee award. And they certainly make clear that it was neither unreasonable nor an abuse of discretion for the District Court to reach that conclusion. Indeed, if the facts and circumstances that I have described are even roughly correct, then it is fair to ask: If this is not an exceptional case, what is?

* * *

My disagreement with the Court is limited. As I stated at the outset, we are in complete agreement with respect to the answer to the question presented: “[A]n increase” to the lodestar “due to superior performance and results” “is permitted in extraordinary circumstances.” Ante, at 1678. Unlike Justice THOMAS, I do not read the Court's opinion to “advance our attorney's fees jurisprudence further along the decisional arc” toward a point where enhancements are “virtually” barred in all cases. Ante, at 1678 (concurring opinion). Our prior cases make clear that enhancements are permitted in “ ‘exceptional’ cases,” Delaware Valley, 478 U.S., at 565, 106 S.Ct. 3088, where the attorney achieves “ exceptional success,” Hensley, 461 U.S., at 435, 103 S.Ct. 1933; see also Blum, 465 U.S., at 896–901, 104 S.Ct. 1541. By definition, such exceptional circumstances occur only rarely. See ante, p. 1677 (KENNEDY, J., concurring). I do not see how the Court could “advance” our fee enhancement jurisprudence so as to further discourage lodestar enhancements without overruling the precedents I have just cited, which the Court has not done. To the contrary, today the Court “reaffirm[s]” those precedents, which allow enhancements for exceptional performance. Ante, at 1669. And with respect to that central holding we are unanimous.

Nor is my disagreement with the Court absolute with respect to the proper resolution of the case before us, for the Court does not purport to prohibit the District Court from awarding an enhanced fee on remand if that court provides more detailed reasoning supporting its decision. Ante, at 1675; cf. Tr. of Oral Arg. 47. But the majority and I do disagree in this respect: I would not disturb the judgment below. “A request for attorney's fees should not result in a second major litigation.” Hensley, 461 U.S., at 437, 103 S.Ct. 1933. Nor should it lead to years of protracted appellate review. See id., at 455–456, 103 S.Ct. 1933 (Brennan, J., concurring in part and dissenting in part). We did not grant certiorari in this case to consider the fact-intensive dispute over whether this is, in fact, an exceptional case that merits a lodestar enhancement. The District Court has already resolved that question and the Court of Appeals affirmed its judgment, having found no abuse of discretion. I would have been content to resolve no more than the question presented. But, even were I to follow the Court's inclination to say more, I would hold that the principles upon which we agree—including the applicability of abuse-of-discretion review to a District Court's fee determination—require us to affirm the judgment below.