No. 24-4103

D.C. No. 3:16-cv-02120EMC

Appeal from the United States District Court for the Northern District of California Edward M. Chen, District Judge, Presiding

Argued and Submitted September 15, 2025 San Francisco, California

Filed April 6, 2026Page 2 Before: Milan D. Smith, Jr. and Patrick J. Bumatay, Circuit Judges, and J. Campbell Barker, District Judge. *

Opinion by Judge Bumatay; Partial Concurrence and Partial Dissent by Judge M. Smith

SUMMARY **

Reversing the district court’s order awarding attorneys’ fees and expenses under 31 U.S.C. § 3730(d)(2) to the plaintiff and remanding in a qui tam action under the False Claims Act, the panel held that the district court abused its discretion in awarding an enhancement above the lodestar calculation.

Under the lodestar method for determining what fees are reasonable, district courts calculate the prevailing market rate in the relevant community and then multiply it by the number of hours reasonably worked on the case. The panel held that only a “rare and exceptional case” will justify an enhancement above the lodestar calculation, and a multiplier may be awarded only upon specific evidence that the lodestar is unreasonably low. In awarding counsel a 1.75 multiplier, the district court cited (1) the exceptional result that counsel achieved in surviving a motion to dismiss andPage 3 (2) counsel's investigative work, but the panel concluded that these considerations were adequately subsumed in the lodestar calculation. The panel further concluded that even assuming a multiplier was appropriate, the district court abused its discretion by failing to provide a reasoned basis for its selection of a 1.75 multiplier.

Concurring in part and dissenting in part, Judge M. Smith wrote that he agreed with the majority’s recitation of the facts and procedural history but disagreed that the district court abused its discretion by applying an enhancement above the lodestar calculation. He would instead conclude that the district court acted within its discretion when it determined that this was the 'rare and exceptional' case justifying such an enhancement. Judge M. Smith agreed, however, that the district court abused its discretion by imposing a 1.75 multiplier without a reasonably specific explanation for that particular multiplier, and so he concurred in the majority’s decision to reverse the district court’s choice of multiplier and remand.

Sanford J. Rosen (argued) and Adrienne P. Harrold (argued), R. Van Swearingen, Rosen Bien Galvan & Grunfeld LLP, San Francisco, California; J. Nelson Thomas, Thomas & Solomon LLP, Rochester, New York; for Plaintiff-Appellee. Timothy P. Ofak (argued) and Joseph M. Katz (argued), Weiner Brodsky Kider PC, Washington, D.C.; Thomas M. McInerney, Ogletree Deakins Nash Smoak & Stewart PC, San Francisco, California; for Defendant-Appellant.

Page 4BUMATAY, Circuit Judge:

In our legal system, each party is generally responsible for paying its own attorneys' fees and expenses. See Hensley v. Eckerhart , 461 U.S. 424, 429 (1983). But Congress sometimes legislates to modify this default rule, providing that a losing party must pay the prevailing party’s attorneys’ fees. See , e.g. , 42 U.S.C. § 1988(b). Congress included such a fee-shifting provision in the False Claims Act ('FCA'). When a relator successfully litigates a claim (and the government has not taken over the claim), the relator is entitled to 'reasonable expenses which the court finds to have been necessarily incurred, plus reasonable attorneys' fees and costs.' 31 U.S.C. § 3730(d)(2); see Cochise Consultancy, Inc. v. United States ex rel. Hunt , 587 U.S. 262, 266 (2019).

To guide this open-ended inquiry into what fees are “reasonable,” the Supreme Court has endorsed a means of calculating these fees—the lodestar method. See Perdue v. Kenny A. ex rel. Winn , 559 U.S. 542, 551–52 (2010). Under the lodestar method, district courts calculate the “prevailing market rate[] in the relevant community” and then multiply it by the number of hours reasonably worked on the case. See id. at 551 (simplified); Edmo v. Corizon, Inc. , 97 F.4th 1165, 1168 (9th Cir. 2024). The benefit of the lodestar method is that it 'cabins the discretion of trial judges, permits meaningful judicial review, and produces reasonably predictable results.' Perdue , 559 U.S. at 552. This case asks when courts may depart from a straight application of the lodestar method and award enhancements multiple times greater than the lodestar calculation.

Page 5Because this is not the 'rare and exceptional' case justifying an enhancement above the lodestar calculation, id. (simplified), we reverse.

Under the FCA, the federal government may sue any person who knowingly submits, or causes to submit, false claims to the United States. 31 U.S.C. §§ 3729(a), 3730(a). The FCA also allows private citizens, called “relators,” to file claims alleging this fraud-known as “qui tam” actions—on behalf of the United States. Id. § 3730(b). If a relator brings a qui tam action, the government may choose to intervene and take over prosecution of the case. Id. § 3730(b)(4). Not only that, but the government may also intervene and seek to have the claim dismissed. Id. § 3730(c)(2)(A), see United States ex rel. Polansky v. Exec. Health Res., Inc. , 599 U.S. 419, 438 (2023). No matter whether the government intervenes, if the qui tam action succeeds, the relator is entitled to a portion of the recovery, and reasonable attorneys’ fees and costs “shall” be awarded. Id. § 3730(d)(1)–(2).

In 2016, Gwen Thrower sued her former employer, Academy Mortgage Corporation, under the FCA. Thrower alleged that Academy falsely certified compliance with requirements of the Federal Housing Administration’s Direct Endorsement Program. Under that program, mortgage lenders may underwrite and endorse residential mortgages for government insurance. If a borrower defaults, the Department of Housing and Urban Development is on the hook for the loan. Thrower was initially represented by Rochester-based law firm Thomas & Solomon LLP (“T&S”) on a contingency-fee basis.

Page 6Thrower’s claim quickly experienced setbacks. First, the government decided not to intervene in the qui tam action, telling Thrower’s attorneys that there was too little evidence of systemic fraud at Academy to justify the suit. Thrower proceeded anyway. T&S then undertook a multi-week investigation of Academy’s lending practices. T&S created a hundreds-of-pages-long call sheet of former Academy employees nationwide, cold-called the former employees, and conducted detailed interviews regarding Academy's underwriting practices. Armed with new evidence gained from this investigation, T&S filed an amended complaint.

Then came the second hurdle. After Academy moved to dismiss the suit, the government also entered the case and moved to dismiss based on its own cost-benefit analysis of the litigation. Thrower and her attorneys didn’t give up. In two rounds of supplemental briefing, T&S defended the suit against the motions to dismiss. The district court sided with Thrower and denied both the government’s and Academy’s motions. The government appealed. Without reviewing the merits, a panel of this court dismissed the appeal for lack of appellate jurisdiction. See United States v. United States ex rel. Thrower , 968 F.3d 996, 1000 (9th Cir. 2020).

After that, things went Thrower's way. In late 2022, Thrower and Academy settled. T&S and another law firm, Reese Marketos, helped negotiate the terms of the $38,500,000 settlement. That means that just under $27,000,000 would be returned to the government and Thrower would receive $11,511,500 for her trouble. Under an agreement with T&S, Thrower was to provide T&S 40% of her proceedings, or $4,604,600. But in a later agreement between T&S and Reese Marketos, T&S’s share of Thrower’s award was reduced to 16%, or $1,841,840.

Page 7The settlement did not resolve Thrower’s entitlement to attorneys’ fees under the FCA. In seeking attorneys’ fees, Thrower sought to double T&S’s fees with a 2.0 multiplier based on Thrower’s “extraordinary” victory. In 2024, the district court granted Thrower's fee request in part and denied it in part. The district court found the requested rates were “on the high end” for the local legal market, though 'not entirely out of range.' Instead, the district court adopted the rates requested by T&S attorneys in a previous Northern District of California case and added 20% to account for inflation in legal market rates. After arriving at the hourly rate and determining the number of hours reasonably worked, the district court arrived at a lodestar amount of $4,374,532.25 for T&S’s fees.

The district court then agreed with Thrower that an upward multiplier for T&S's work was warranted. It concluded that the firm achieved an “exceptional result” in keeping the case “alive” despite the government’s attempt to dismiss it. To the district court, Thrower not only had to 'battl[e]' Academy but also the government. To do this, T&S conducted “a multi-week investigation, . . . identifying, locating, and interviewing former employees across the country” to determine whether Academy’s alleged fraud was pervasive. That Thrower prevailed against the two was, “if not the first of its kind, very rare.' But the district court determined that a 2.0 multiplier was not appropriate and reduced it to 1.75 on account of T&S’s reduced involvement in later stages of the litigation. The district court did not, though, provide any calculations underpinning its determination of the multiplier figure. After applying the multiplier of 1.75 to T&S’s merits work, the district court awarded Thrower $8,585,530.20 in attorneys’ fees.

Page 8Academy appeals the award of the multiplier. We review a district court’s award of attorneys’ fees for abuse of discretion. See Kelly v. Wengler , 822 F.3d 1085, 1094 (9th Cir. 2016).

To determine 'reasonable attorneys' fees' under the FCA, we use the lodestar method. See Carter v. Caleb Brett LLC , 757 F.3d 866, 868 (9th Cir. 2014) (noting that the lodestar method is the correct framework for calculating reasonable attorneys’ fees under federal fee-shifting statutes). The lodestar method is useful because it “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.' Perdue , 559 U.S. at 551.

The lodestar method proceeds in two steps. See Edmo , 97 F.4th at 1168. First, the district court is to tally the number of hours an attorney “reasonably expended” on the case and then multiply it by the “reasonable hourly rate” in the market. Id. The result is the “presumptively reasonable award.” Id. Second, the district court considers whether a modification of the lodestar figure—either upward or downward—is appropriate. Id. Any modification must be 'based on factors not subsumed in the lodestar figure.' Kelly , 822 F.3d at 1099.

Given the comprehensive nature of the lodestar calculation, any enhancements are 'rare and exceptional.' Perdue , 559 U.S. at 554 (simplified); see also id. at 561 (Thomas, J., concurring) (“[T]he lodestar calculation will in virtually every case already reflect all indicia of attorney performance relevant to a fee award.' (simplified)). A multiplier may be awarded only upon 'specific evidence'Page 9 that the lodestar fee is “unreasonably low or unreasonably high.” Van Gerwen v. Guarantee Mut. Life Co ., 214 F.3d 1041, 1045 (9th Cir. 2000) (simplified). “Any reliance on factors . . . subsumed in the lodestar determination will be considered an abuse of the trial court's discretion.' Parsons v. Ryan , 949 F.3d 443, 467 (9th Cir. 2020) (simplified).

This is not the “rare and exceptional” case that overcomes the 'strong presumption that the lodestar is sufficient.' Chambers v. Whirlpool Corp. , 980 F.3d 645, 665 (9th Cir. 2020) (simplified). The district court provided two reasons for awarding T&S a 1.75 multiplier for its merits work. First, the district court cited the “exceptional result” that T&S achieved in surviving the motion to dismiss, “notwithstanding the Government’s attempts to extinguish [the case].” Second, the district court attributed the firm’s successful defense against dismissal to “T&S’s investigative work.' Because these considerations were adequately subsumed in the lodestar calculation, it was an abuse of discretion to grant the multiplier. And even if an enhancement were appropriate, the district court failed to provide a sufficient reason for its seemingly arbitrary 1.75 multiplier.

Consider first the “exceptional result” in surviving the government's motion to dismiss and obtaining a sizable settlement for Thrower and the government. The district court focused on the unprecedented nature of T&S's win. Recall that the government entered the case and pushed for dismissal because it believed that the narrow scope of the alleged fraud didn’t warrant the FCA suit. Thus, Thrower had to battle not only Academy, but also the government to Page 10defeat dismissal. The district court denied the government’s motion to dismiss because it discounted the government’s cost-benefit analysis—believing that the scope of the analysis was too geographically and temporally limited. As the district court observed in its order granting fees, the case was “unprecedented” and attracted media attention; no court had ever permitted a qui tam action to proceed against the government’s wishes. The district court thus credited T&S with obtaining a 'very rare' 'if not the first of its kind' settlement.

Without more, this doesn’t amount to “specific evidence” justifying a multiplier. Ordinarily, the “novelty or complexity of the issues should not be considered at the multiplier stage.” Van Gerwen , 214 F.3d at 1045 n.2 (citing Blum v. Stenson , 465 U.S. 886, 898-901 (1984)). That's because the “quality of representation and results obtained are ordinarily subsumed in the lodestar determination.” Id. Indeed, the complexity in battling two parties at once should be “fully reflected in the number of billable hours recorded by counsel.” Chambers , 980 F.3d at 666 (quoting Perdue , 559 U.S. at 553).

Of course, 'superior results' can sometimes justify a fee enhancement. See Perdue , 559 U.S. at 554. '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,' courts can reward 'superior attorney performance.' Id. But once again, an enhancement for “superior attorney performance” can be considered only when that “performance is not adequately taken into account in the lodestar calculation.' Id. Perdue offers three examples in which the lodestar figure might not capture superior performance: (1) when the lodestar hourly rate fails to measure “the attorney’s true market value,” id. at 554-55;Page 11 (2) when attorney performance includes 'an extraordinary outlay of expenses and litigation is exceptionally protracted,” id. at 555; and (3) when “extraordinary circumstances” involve an “exceptional delay in the payment of fees,” id. at 556. Though the Court didn’t say these examples were exhaustive, it stressed that these circumstances must be 'rare and exceptional.' See id. at 553–57 (simplified).

The district court made no finding that any circumstance left the lodestar calculation unreasonably low. The district court didn’t find that T&S risked an extraordinary outlay of expenses, endured an exceptionally protracted litigation, or suffered an exceptional delay in payments. Nor did the district court believe that the lodestar calculation undervalued T&S’s “true market value.” As we’ve said, an enhancement may be appropriate when the 'value of [an attorney’s] services exceeds what the court determines to be the prevailing market rate for otherwise comparable attorneys.” Kelly , 822 F.3d at 1102. Yet the district court made no finding that T&S's services exceeded the area's prevailing market rate. Instead, it found the opposite. It concluded that T&S’s requested rates were “on the high end of the scale for large firms in the area.' It then lowered T&S’s requested rates based on an analysis of the 2022 mean rates of litigation partners in San Francisco and the rates that T&S itself requested in other local cases from 2021. It found that the 2021 rates were the appropriate hourly rates but increased them by 20% to account for inflation in market rates since that time. Finally, the district court didn't indicate that it was intentionally lowering the hourly rate on the expectation that it would make up any shortfall on the back end through the multiplier, as Thrower suggests.

Page 12And, as the district court noted, no statutory cap artificially lowered T&S’s hourly fee, which is often used to justify an enhancement. Cf. Kelly , 822 F.3d at 1102 . In Kelly , we affirmed 1.3 and 2.0 multipliers because plaintiff’s counsel’s “extraordinary performance yield[ed] extraordinary results.” See id. at 1091, 1093-94, 1103. But importantly, Kelly was governed by the Prison Litigation Reform Act ('PLRA'), 42 U.S.C. § 1997e(d). In PLRA cases, courts must “calculate the lodestar figure using a base rate no greater than the PLRA rate specified in § 1997e(d)(3).' Id. at 1102-03. And 'the PLRA rate virtually always replaces the prevailing market rate as the presumptively reasonable rate in the court’s determination of the lodestar figure because actual prevailing rates are very unlikely to be as low as the PLRA rate.' Id. at 1103. Thus, the statutorily constrained lodestar analysis justified the enhancement based on superior performance. See id. at 1102. In other words, superior performance merited an enhancement only because the prevailing rate was artificially capped by statute. In that case, “the quality of the work that produced [extraordinary results] was underrepresented in the hourly fee” mandated by the PLRA. Id. at 1102 (simplified). But 'in a non-PLRA case,' the lodestar method “generally subsumes the factors relevant to the determination of a reasonable attorney’s fee, including the novelty and complexity of the case and the quality of the attorney’s performance.” Id. at 1103. That is the situation here.

The district court’s second reason for the multiplier—to account for T&S's investigative work-is even easier to discount. According to the district court, T&S engaged in a multi-week investigation to review Academy’s policies and Page 13practices nationwide. This investigation involved identifying, locating, and interviewing former Academy employees across the country to see if the alleged fraudulent practices were pervasive throughout the company. Through extensive interviews, T&S apparently uncovered other evidence of fraud relevant to Thrower’s FCA allegations.

While these Herculean efforts clearly led to the denial of the government's motion to dismiss and the eventual settlement of this case, the lodestar calculations adequately account for the investigative work. Indeed, the district court carefully looked through T&S’s time entries and granted the firm around 7,000 hours of billable work for its efforts—or around 875 workdays. And nothing shows that these massive billables don’t properly account for T&S’s investigative work. When 'time and labor spent by [counsel] were subsumed within th[e] lodestar figure,” they may not be used to 'justify an upward multiplier.' See Chambers , 980 F.3d at 665 (simplified). Likewise, the skill reflected by T&S’s attorneys during this investigation was subsumed in the determination of the reasonable market rate. See Perdue , 559 U.S. at 553 (“[C]onsiderations concerning the quality of a prevailing party’s counsel’s representation normally are reflected in the reasonable hourly rate.” (simplified)). As mentioned, the district court awarded hourly rates higher than the mean of litigation partners in San Francisco.

3

We agree with the dissent that district court judges may apply an enhancement for an “exceptional result.” But it’s wrong to say that we improperly disregard the district court’s evidence of an “exceptional result,” including T&S’s hardfought result and diligent investigative efforts. The flaw in Page 14the district court’s analysis was not a failure to adequately show an exceptional result or extraordinary investigative efforts. The critical issue is that it didn’t provide specific evidence for why the lodestar method could not reflect T&S's true market rate. Thus, unlike in Kelly , which specifically relied on statutory rate caps as justification for the enhancement, 822 F.3d at 1102-03, the district court identified no constraint on market rates here. Instead, the district court resorted too quickly and without adequate explanation to the assignment of a multiplier. This breaks the back of the lodestar method, allowing resort to a “judge’s subjective opinion regarding particular attorneys or the importance of the case” outside the proper constraints of the two-step lodestar process. Perdue , 559 U.S. at 558.

* * *

Departures from the lodestar rate are to be 'rare' and reserved only for when 'the lodestar does not adequately take into account a factor that may properly be considered in determining a reasonable fee.' Perdue , 559 U.S. at 554. Because the multiplier here was based on considerations already subsumed into the lodestar calculation, its award was an abuse of discretion.

Even assuming a multiplier was appropriate, the district court abused its discretion by failing to provide a reasoned basis for its selection of a 1.75 multiplier. “It is essential that the judge provide a reasonably specific explanation for all aspects of a fee determination, including any award of an enhancement.” Perdue , 559 U.S. at 558. Absent this, the multiplier leaves parties without any understanding of the risk of having to pay a sizable enhancement and defeats the Page 15lodestar’s benefit of “providing an objective and reviewable basis for fees.” See id. at 558–59.

To pass muster then, district courts must reasonably show their work in arriving at a multiplier amount. In other words, the district court should answer, 'why 75% rather than 50% or 25% or 10%?” Id. at 557. At a minimum, this includes identifying the factors justifying the enhancement and then showing how much of 'the enhancement . . . is attributable' to those factors. Id. at 558. For example, when an extraordinary delay justifies an enhancement, the district court should “provide a calculation of the cost to counsel” resulting from that delay. Id. Or when a district court relies on a comparison of counsel’s performance, the district court must offer “specific proof linking the attorney’s ability to a prevailing market rate.” Id. at 555. Perdue thus instructs that a district court must link the factors showing the lodestar's inadequacy with the amount of the multiplier needed to close the gap.

Consider again Kelly . Recall we affirmed the award of a 2.0 and a 1.3 multiplier. See Kelly , 822 F.3d at 1093, 1102. There, the district court noted that the PLRA-compliant lodestar rate was $213 but found that comparable market rate for the two attorneys in the case was roughly $435 for one and between $275 and $300 for the other. After finding an enhancement justified, the district court here applied a multiplier of 2.0 (raising one attorney's rates to $426 an hour) and 1.3 (raising the other's to $277). Given these calculations, the parties could understand and predict the fees involved and this court could provide meaningful appellate review.

The district court failed to explain how it selected a 1.75 multiplier. After concluding that an enhancement was Page 16warranted given the exceptional result and the investigative work involved, the district court didn't then link the 1.75 multiplier to a shortfall in the lodestar calculation based on those factors. Instead, as best we can tell, the district court just started with Thrower’s requested 2.0 multiplier and then reduced it to 1.75 because T&S had a reduced role in later stages of the litigation, didn’t negotiate the settlement, and was already guaranteed a share of Thrower’s portion of the settlement. This explanation isn't enough. Rather, the district court needed to show how T&S’s superior performance justified a rate 1.75 times the lodestar rate. As a result, the district court’s determination of a multiplier was not evidently derived from the factors necessary to bring the lodestar calculation to a more appropriate rate. Because the district court’s choice of a 1.75 multiplier lacked an adequately reviewable methodology, it is an independent reason to reverse the enhancement.

For these reasons, we reverse the award of a multiplier and remand.

Page 17M. SMITH, Circuit Judge, concurring in part and dissenting in part.

I agree with the majority's recitation of the facts and procedural history. However, I disagree that the district court abused its discretion by applying an enhancement above the lodestar calculation. I would instead conclude that the district court acted within its discretion when it determined that this is the 'rare and exceptional' case justifying such enhancement. Perdue v. Kenny A. ex rel. Winn , 559 U.S. 542, 552 (2010) (internal quotation marks omitted). That being said, I agree that the district court abused its discretion by imposing a 1.75 multiplier without a “reasonably specific explanation” for that particular multiplier, so I ultimately concur in the majority’s decision in Part II.4 to reverse the district court’s choice of multiplier and remand.

The majority concludes that the district court abused its discretion by applying a multiplier to its attorney’s fee award because its justifications for the enhancement “were adequately subsumed in the lodestar calculation.” Maj. Op. at 9. I agree that it is an abuse of discretion to rely on subsumed factors when awarding an enhancement. See Parsons v. Ryan , 949 F.3d 443, 467 (9th Cir. 2020). But I disagree that the district court relied on subsumed factors here. In other words, I disagree that it was an abuse of discretion for the district court to conclude that Thomas &Page 18 Sullivan LLP’s (T&S) “true market value” is measured by the lodestar hourly rate. Perdue , 559 U.S . at 554-55.¹

I will begin with the majority’s first conclusion: that it was an abuse of discretion for the district court to enhance the attorney’s fees award based on the “exceptional result” in this case. The Supreme Court has explained that courts can reward 'superior performance and commitment of resources by plaintiff’s counsel” when they determine that “a plaintiff’s attorney [has] achieve[d] results that are more favorable than would have been predicted based on the governing law and the available evidence.” Id. at 554. The district court’s analysis makes clear that it did so here.

While the district court noted that the recovery amount in this case was 'in line with a median range' of other settlements in cases brought pursuant to the False Claims Act (FCA), it concluded that “what is clearly distinct about this case is that [Thrower] prevailed in receiving a settlement, notwithstanding the Government seeking to prevent the case from proceeding.' It explained that Thrower “battl[ed]” not only Defendant-Appellant Academy Mortgage Corporation (Academy), but also the government, which not only declined to intervene in this qui tam action but actively sought to dismiss it while Academy did the same. Moreover, the district court highlighted the Page 19government’s own acknowledgement that “no court had ever permitted the continuation of a qui tam action in the United States' name that the United States sought to end.'² See Reply Brief for United States, United States v. United States ex. rel. Thrower , No. 18-16408 (9th Cir. Aug. 21, 2019), 2019 WL 4132218, at *2. “Accordingly,” the district court concluded, “an upward modifier [wa]s appropriate” because “T&S effected the exceptional result achieved in the case by keeping it alive against stacked odds at the outset.” According to the district court, the case was, “if not the first of its kind, very rare.”

The majority disagrees that the 'exceptional result' justifies an enhancement here. The majority concludes that an exceptional result is “ordinarily” subsumed as a reflection of the 'novelty or complexity of the issues.' See Van Gerwen v. Guarantee Mut. Life Co. , 214 F.3d 1041, 1045 n.2 (9th Cir. 2000). And it concludes that the “complexity in battling two parties at once” is reflected in the number of billable hours. Maj. Op. at 10. Even so, the majority misconstrues the district court’s analysis. The district court did not rely on the “complexity” of opposing two parties at once to justify the enhancement; it relied on Thrower's exceptional and unexpected result despite opposing two parties at once. We have repeatedly stated that “exceptional” success that is accompanied by “superior” attorney performance can justify an enhanced award. Van Gerwen , 214 F.3d at 1046. See also Edmo v. Corizon, Inc. , 97 F.4th 1165, 1170–71 (9th Cir. 2024); Kelly v. Wengler , 822 F.3dPage 20 1085, 1102–04 (9th Cir. 2016). The Supreme Court has said the same. See Perdue , 559 U.S. at 554; Blum v. Stenson , 465 U.S. 886, 901 (1984); Hensley v. Eckerhart , 461 U.S. 424, 435–36 (1983).

In this case, the district court found T&S’s work to be “substantial, successful and heroic,' 'allowing for the unlikely, positive settlement in [Thrower's] favor.' This conclusion is sufficient to justify an enhancement under our precedent. What was extraordinary in this case is not merely the effort and risk that T&S shouldered to litigate the case. What was extraordinary is that T&S successfully did so despite facing “stacked odds.” For example, consider that T&S is a firm of only seven attorneys, who worked on this litigation for over seven years (2016 to 2024). Yet the firm obtained a settlement of $38.5 million. Of that $38.5 million, just under $27 million was restored to the United States Treasury. The majority acknowledges that T&S's investigative efforts were “Herculean,” Maj. Op. at 13, but it minimizes that T&S—like Hercules—successfully fought Cerberus. I would hold that the district court did not abuse its discretion by concluding that this is the 'rare' and “extraordinary” circumstance that warrants an enhancement.

The majority disagrees because it concludes that 'the district court [did not] believe that the lodestar calculation undervalued T&S’s ‘true market value.’” Maj. Op. at 11. The fact that the district court enhanced the award suggests otherwise.³ The majority distinguishes this case from Kelly Page 21v. Wengler , 822 F.3d 1085, 1102-04 (9th Cir. 2016), in which we affirmed a district court's use of 1.3 and 2.0 multipliers for 'extraordinary performance' that led to “extraordinary results.” See id. at 1093-94. The majority distinguishes Kelly because it was governed by the Prison Litigation Reform Act (PLRA), 42 U.S.C. § 1997e(d), which imposes a statutory cap on attorney’s fees. See id. at 1093. Because there is no similar cap for FCA cases, the majority reads Kelly entirely out of the equation. Specifically, the majority concludes that the 'superior performance [in Kelly ] merited an enhancement only because the prevailing rate was artificially capped by statute,' and there is no similar justification in this case. Maj. Op. at 12 (emphasis added).

The majority's interpretation of Kelly is inconsistent with its text. In Kelly , we analyzed both whether an enhanced award is consistent with the PLRA ( i.e. , whether an enhancement could be applied in a PLRA case at all ), see 22 F.3d at 1100–02, and also “whether the district court in [that] case provided clear reasons, supported by specific evidence in the record, for enhancing the lodestar figure' ( i.e. , whether an enhancement was warranted in that case ), id. at 1102-05. We answered both questions in the affirmative. In doing so, we affirmed the district court's reliance on “non-subsumed factors to justify its attorney’s Page 22fees enhancement-superior performance and the need to attract competent counsel.' Id. at 1102. We explained that

Like the prevailing rate in a non-PLRA case, the PLRA rate generally subsumes the factors relevant to the determination of a reasonable attorney’s fee, including the novelty and complexity of the case and the quality of the attorney’s performance. However, as in non-PLRA cases, the district court may enhance the lodestar figure when plaintiff's counsel's 'superior performance and commitment of resources” is “rare” and “exceptional” as compared to the run-of-themill representation in such cases.

Id. at 1103 (emphasis added). In practical effect, it is difficult to see the majority doing anything other than reading “generally” as “always.” The PLRA may explain why the lodestar rate undervalues an attorney’s “true market value' in some cases. But Kelly does not suggest that a statutory cap is the only reason that an enhancement may be warranted. To reiterate: whether or not a statutory cap is implicated, district courts may apply an enhancement when the result is “more favorable than would have been predicted based on the governing law and the available evidence,” and that result is due to attorney performance. Perdue , 559 U.S . at 554. That is indeed the case here, where the government declined to intervene in the lawsuit specifically because it found insufficient evidence of systemic fraud, and it was T&S’s extensive investigative efforts that suggested otherwise.

Page 23The district court did not abuse its discretion in finding an extraordinary result, nor did it abuse its discretion in finding that T&S’s investigative work was the reason for that result. The district court’s descriptions of T&S’s performance (“substantial, successful and heroic,” successful in “battl[e]” against “stacked odds”) demonstrate its conclusion that T&S went above and beyond “otherwise comparable attorneys.” Kelly , 822 F.3d at 1102. In other words, the extraordinary result was the result of T&S's superior performance, representing a higher 'true market value” than the prevailing market rate that the district court used to determine the lodestar.⁴ Id. Certainly, the district court was in a better position to evaluate T&S’s true market value than we are sitting as a court of appeal.⁵

The majority acknowledges that it is not an abuse of discretion to award a multiplier where a party achieves an “unprecedented” result. Maj. Op. at 10, 13–14. It merely explains that '[ w ] ithout more , [an unprecedented result] doesn’t amount to ‘specific evidence’ justifying a multiplier,” and it reverses the enhancement because it finds nothing more here. Maj. Op. at 10 (emphasis added). IPage 24 disagree. I would conclude that an enhancement is justified by the extraordinary result, although I agree that the particular multiplier used to enhance the award here was insufficiently supported. In ruling otherwise, the majority “discount[s]” an important part of the district court’s reasoning: its consideration of T&S’s extensive investigative work. Maj. Op. at 12. I turn to this factor in the next section.

B.

The majority’s second conclusion is that T&S’s investigative work cannot be used to justify an enhancement because the lodestar calculation “adequately account[s] for the investigative work” at issue. Maj. Op. at 13. In my view, this asks the wrong question, resulting in the majority erroneously discounting the district court’s reasoning. The district court did not examine T&S’s extensive investigation as an independent justification to award an enhancement; it examined T&S’s investigative work to explain why it found that the “extraordinary” result in this case was due to T&S’s superior performance. It was necessary for the district court to do so because the Supreme Court has stated that “superior results are relevant only to the extent it can be shown that they are the result of superior attorney performance.' Perdue , 559 U.S. at 554 (emphasis added). Courts 'treat th[o]se two factors as one.” Id. “[I]nferior performance by defense counsel' or 'unexpected favorable rulings by the court' are not enough to warrant an enhancement. Id. Whether T&S’s time spent investigating is subsumed in the lodestar is not the relevant question. The relevant question is whether T&S’s investigative work is responsible for the extraordinary result in this case.

Page 25The district court found that it was, concluding that “the case surviving is directly attributable to T&S’s investigative work.” Accordingly, the district court awarded an enhancement because T&S’s superior attorney performance (demonstrated through its investigative work) led to the extraordinary result (surviving the government’s motion to dismiss). It explained that “the narrowness of the Government’s investigation, which was the basis for denying the motion to dismiss, was made apparent precisely because of the scope and breadth of the investigation that T&S engaged in in bringing its claims against Academy.'

Accordingly, I would conclude that T&S's superior attorney performance provides something “more” to justify the enhancement and would hold that the district court did not abuse its discretion by relying on the unprecedented, extraordinary result to award an enhancement here. When a district court reasonably finds that an attorney's superior performance resulted in an extraordinary result, such finding suggests that the attorney's 'true market value' is higher than the lodestar hourly rate.

That being said, I agree with the majority that the district court abused its discretion by failing to provide a “reasonably specific explanation” for its selection of a 1.75 multiplier. See Perdue , 559 U.S. at 558. The district court explained that it was reducing the requested 2.0 multiplier to 1.75 because 'the firm took a lesser role as the suit progressed, [it] was not primarily responsible for negotiating the settlement, and the risk it took was counterbalanced to some extent by the private contingency agreement entitling it to fully 40% of the settlement fee paid to [Thrower]Page 26 (roughly $4 million) in addition to the fee awarded.” 6 But when a case is the 'rare case where it is not possible to address fully the attorney's performance by setting the appropriate, reasonable hourly rate used in calculating the lodestar,” the district court must “determine, and [] explain how, counsel’s service—particularly given the results obtained-relates to the service and results reasonably expected of an attorney working the number of hours at the hourly rate the district court sets in determining the lodestar.” Van Gerwen , 214 F.3d at 1047. Put another way: After determining that an enhancement is justified, the district court must support its decision to enhance the award using a multiplier with 'specific evidence.' Edmo , 97 F.4th at 1169, 1170–71 (“If an enhancement is justified, then the court ‘may adjust the lodestar upward . . . using a multiplier,’” which “must be supported by ‘specific evidence.’” (quoting Van Gerwen , 214 F.3d at 1045)). That is, it must support “the precise amount of the enhancement” with specific evidence. Id. at 1171. I agree that the district court failed to support the “precise amount” of its enhancement here. Id.

Because I agree that the district court failed to adequately explain how T&S's services and the results obtained “relate[] to the service and results reasonably expected of an attorney working” at the lodestar rate, id. , I agree with the majority's decision to reverse the application of the 1.75 multiplier and remand. On remand, I would have the district court provide a 'reasonably specific explanation' for the enhancement that allows “adequate appellate review” with a “discernible methodology.” Perdue , 559 U.S. at 558–59 &Page 27 n.8. In contrast, the majority concludes that the district court abused its discretion by awarding an enhancement at all.

We do not hold all attorneys that practice before us to an extraordinary standard; doing so would turn the exception into the norm. The majority discounts as ordinary diligence what the district considered to be superior attorney performance that achieved an extraordinary result in the face of “stacked odds.” Where attorneys go above and beyond the ordinary levels of competence, diligence, and zealous advocacy, district courts have discretion to fashion an attorney’s fee award that reflects the “true market value” of their performance when they achieve “extraordinary results” that can be attributed to their performance. Here, I would conclude that it was not abuse of discretion for the district court to conclude that this is a 'rare and extraordinary circumstance” warranting an enhancement.

Because the majority would narrow the court’s ability to compensate attorneys beyond what our case law supports, I respectfully dissent from the majority’s conclusion that the district court abused its discretion by enhancing the attorney’s fee award here. However, because I agree that the district court abused its discretion by failing to provide a “reasonably specific explanation” for its selection of a 1.75 multiplier, I concur in the majority’s conclusion to reverse the enhancement and remand.