EBEL, Circuit Judge:

Manuel Christiansen and Brian Ashton (Relators) brought a qui tam lawsuit against Everglades College, Inc., d.b.a. Keiser University (Keiser University) under the federal False Claims Act (FCA). They alleged that Keiser University, a participant in federal student financial-aid programs, falsely certified compliance with a federal law banning incentive payments to university admissions counselors. When the United States initially declined to intervene and take over the case, Relators pursued the action and won only a limited trial victory — no damages and only $11,000 in penalties — so they appealed to this Court. During the pendency of that appeal, however, the United States stepped in and settled the case with Keiser, securing a much larger monetary recovery than Relators procured at trial.

Confident that they could have prevailed on appeal, Relators believed the rug had been pulled out from under them. They argued the United States had no right to intervene so late in the proceedings, they challenged the underlying fairness of the settlement, and they asked for an evidentiary hearing and discovery into the government's settlement deliberations in search of a nefarious motive. The district court rejected these arguments and allowed the United States to intervene, approved the settlement, and denied Relators' requests for an evidentiary hearing and discovery pertaining to the government's decision to intervene and settle the case. Relators now appeal those rulings. Appeal No. 16-10849.

Relators also appeal from the district court's subsequent award of reduced attorneys' fees and costs. Appeal No. 16-11839. In light of the paltry outcome secured at trial and Relators' opposition to the eventual settlement, which had resulted in a significantly greater monetary recovery, the district court trimmed Relators' fees and costs to a small fraction of the requested award.

Recognizing that the settlement and the attorneys' fees are inextricably linked, we consolidated these appeals. Exercising jurisdiction under 28 U.S.C. § 1291, we AFFIRM in both Appeals Nos. 16-10849 and 16-11839.

I. BACKGROUND

A. The Alleged FCA Violations

Keiser University is a non-profit college offering undergraduate and graduate programs across more than a dozen campuses. Many of Keiser's students receive federally sponsored financial aid under Title IV of the Higher Education Act of 1965, 20 U.S.C. §§ 1070-1099d. In order to receive Title IV funds, schools must enter into "program participation agreements" with the Department of Education that condition eligibility for financial aid funds on the institution's compliance with various enumerated requirements. 20 U.S.C. § 1094(a). One of those requirements is known as the Incentive Compensation Ban (ICB), which prohibits a school from paying incentives to recruiters and admissions personnel based on the number of students they enroll. Id. § 1094(a)(20); see also 34 C.F.R. § 668.14(b)(22).

Relators, two former employees who worked in Keiser's admissions department, alleged that Keiser submitted more than 230,000 claims for a total of $1.2 billion in federal financial aid, all the while falsely certifying to the United States that Keiser was complying with the ICB. According to Relators, Keiser knew that its admissions personnel received incentive payments based on their success in securing enrollments. Even with that knowledge, Keiser expressly certified compliance with the ICB on multiple occasions and — more important to Relators' theory of the case — Keiser caused its students to submit claims for financial aid to the government, all the while knowing that it was violating the ICB. Relying on the "implied false certification" theory,1 Relators asserted that Keiser was liable not only for its own express certifications, but also for the enormous volume of student-submitted claims.

B. The District Court's Merits Decision

When Relators initially brought the FCA action, the United States declined to take over the case, which is the government's prerogative under 31 U.S.C. § 3730(b)(2). Relators thus exercised their statutory right to pursue the case on behalf of the United States. Id. § 3730(b)(4)(B). After a bench trial, the district court handed Relators a victory that fell far short of their expectations. At the outset, the court agreed that Keiser was in violation of the ICB during the relevant period. But the court made three conclusions that severely limited Relators' monetary victory.

First, the court rejected Relators' theory that each student-submitted application for financial aid was an actionable FCA claim, reasoning that Keiser could not control the content, number, and submission of student financial-aid requests. Second, turning to the certifications that Keiser itself actually sent to the government, the court found that Keiser's top policymakers did not become aware of the improper compensation scheme until November 20, 2009, after which it knowingly submitted only two certifications of compliance to the government.2 Keiser was thus liable for only those two false claims, for which the district court awarded the minimum statutory penalty of $5,500 each — a total of $11,000. Third, the court rejected Relators' argument that the government's damages were equal to the value of all educational assistance paid to Keiser during the period covered by the false certifications. It reasoned that the Department of Education would not have demanded reimbursement for the already-paid Title IV funds even if Keiser had disclosed its ICB violations. Thus the government, according to the district court, suffered no financial damages at all. Instead, the government was entitled only to the nominal statutory penalties for the two express certifications sent by Keiser.

C. The United States Intervenes to Settle the Case

Relators appealed the district court's decision to our Court. The United States, however, believed an appeal was risky because there was a chance the Eleventh Circuit would affirm the district court's narrow interpretation of FCA liability, thereby impairing FCA enforcement efforts throughout the circuit. Thus, after Relators' opening brief was filed in this Court, but before the United States moved to intervene in the qui tam action, the United States struck a tentative deal with Keiser. The tentative settlement agreement provided that Keiser would pay the United States $335,000 — more than thirty times the amount recovered by Relators at trial — and the United States would in turn release Keiser from any further administrative or civil claims, and even more importantly, would also refrain from suspending or terminating Keiser's eligibility for future Title IV funds based on Keiser's challenged conduct in this case.

Relators' merits appeal was still pending on this Court's docket at the time, so the district court did not have jurisdiction to enter any orders. Thus, after reaching a tentative settlement with Keiser, the United States moved the district court for an indicative ruling stating that, if the district court reacquired jurisdiction on remand from this Court, the district court would permit the United States' intervention and approve the proposed settlement as fair and reasonable. The district court agreed and issued an indicative order to that effect, reasoning that the United States could intervene as the real party in interest and that the settlement was fair and reasonable because it resulted in a recovery far exceeding the amount obtained by Relators at trial. In light of that indicative ruling, this Court then remanded the case back to the district court, whereupon the district court granted the government's intervention motion.

In that same order, the district court also affirmed a magistrate judge's determination that Relators were not entitled to discovery or a full-blown evidentiary hearing to probe the settlement deliberations and the government's rationale for ending the case. Finally, the district court scheduled a statutorily mandated hearing to confirm that the proposed settlement was fair and reasonable. After the fairness hearing, the district court approved the settlement, and dismissed the case with prejudice.

D. Attorneys' Fees and Costs

While these developments were unfolding, Relators also sought an award for attorneys' fees and costs. After the bench trial, but before the settlement, Relators asked for over $1 million in attorneys' fees and almost $76,000 in litigation costs. In light of Relators' limited success at trial, the district court reduced the fee award to $60,000 for their efforts at trial, and trimmed the award of costs to $27,000. After the eventual settlement, which Relators claimed was brought about by their appeal, they sought enhanced fees and costs based on the larger settlement figure that the United States had been able to procure with Keiser, as well as the additional fees and costs they incurred for their efforts on appeal. The district court denied that request on the ground that Relators objected to the settlement at every stage of the proceedings and should not reap the benefits of an outcome they so vigorously sought to prevent.

Relators now appeal all of these issues.

II. DISCUSSION

We resolve four principal issues on appeal: (1) whether the government was required to satisfy the FCA's good-cause intervention requirement as set forth in 31 U.S.C. § 3730(c)(3) when it intervened for the sole purpose of settling the dispute; (2) whether the proposed settlement was "fair, adequate, and reasonable" under § 3730(c)(2)(B); (3) whether Relators were entitled to an evidentiary hearing and discovery to probe the government's reasons for settling the case; and (4) whether the district court abused its discretion in awarding substantially reduced attorneys' fees based on Relators' limited success at trial and subsequent opposition to the settlement. On all issues, we affirm the district court.

A. The Good-Cause Intervention Requirement

The United States originally declined to proceed with the qui tam action, which authorized Relators to litigate the case on behalf of the government. 31 U.S.C. § 3730(c)(3) ("If the Government elects not to proceed with the action, the person who initiated the action shall have the right to conduct the action."). Relators were unsatisfied with the trial outcome, however, so they appealed. But the government was eager to avoid the risk that the Eleventh Circuit would affirm the ruling below, so it then stepped in and, while the case was pending on appeal, settled the case, thereby depriving Relators of their chance at reversal and a subsequent greater recovery upon remand. Relators argue that the district court erred in granting the government's motion to intervene for the purpose of settling the case.

Relators rely on 31 U.S.C. § 3730(c)(3), which provides: "When a [relator] proceeds with the action, the court, without limiting the status and rights of the [relator], may nevertheless permit the Government to intervene at a later date upon a showing of good cause." 31 U.S.C. § 3730(c)(3) (emphasis added). Relators insist that the United States lacked "good cause" to intervene here. We reject this contention. We hold that, in this case, the United States did not need to satisfy the good-cause intervention requirement for qui tam actions under 31 U.S.C. § 3730(c)(3) because that subsection applies only when the government intervenes for the purpose of actually proceeding with the litigation — not when it is stepping in only for the purpose of settling and ending the case. Because intervention was not required, we need not concern ourselves with whether the requirements of § 3730(c)(3), which addresses late intervention by the government for the purpose of continuing the litigation, have been met.3

A straightforward reading of the text supports this conclusion. First, subsection (b)(2) expressly links intervention to the government's decision to "proceed with the action." § 3730(b)(2) ("The Government may elect to intervene and proceed with the action...." (emphasis added)). Second, in subsections (c)(2)(A) and (B), the statute spells out the circumstances in which the government may settle or dismiss a qui tam case, and neither subsection conditions the government's rights on formally intervening in the case. Instead, they provide in unequivocal terms that "the Government may settle [or dismiss] the action with the defendant notwithstanding the objections of the person initiating the action." § 3730(c)(2)(A), (B) (emphasis added). In the context of dismissals, the court need only "provide[] the [relator] with an opportunity for a hearing," § 3730(c)(2)(A); and with settlements, the court must "determine[], after a hearing, that the proposed settlement is fair, adequate, and reasonable under all the circumstances," § 3730(c)(2)(B). We decline to import the good-cause intervention requirement from subsection (c)(3) into these provisions which specifically govern dismissals and settlements.

Our sibling circuits have reached the same conclusion. The D.C. Circuit held in United States ex rel. Schweizer v. Oce N.V. (Schweizer) that the government does not have to intervene before settling a qui tam action because "intervention is necessary only if the government wishes to `proceed with the action.'" 677 F.3d 1228, 1233 (D.C. Cir. 2012) (internal quotation marks omitted). The Fifth Circuit also reasoned that, even when the government does not intervene, it "may settle a case over a relator's objections if the relator receives notice and [a] hearing of the settlement." Riley v. St. Luke's Episcopal Hosp., 252 F.3d 749, 754 (5th Cir. 2001) (en banc).

And in the closely related context of dismissals, other circuits have also declined to require good cause or formal intervention. For instance, the Tenth Circuit in Ridenour v. Kaiser-Hill Co. focused on the plain statutory language, noting that it could "identify nothing in the language of [the subsection on dismissals] to suggest the authority of the Government to dismiss a qui tam action is dependent upon prior intervention in the case." 397 F.3d 925, 933 (10th Cir. 2005); see also Swift v. United States, 318 F.3d 250, 251-52 (D.C. Cir. 2003) (finding intervention necessary "only if the government wishes to `proceed with the action'" and "[e]nding the case by dismissing it is not proceeding with the action"); Riley, 252 F.3d at 753 (5th Cir.) (noting that "the government retains the unilateral power to dismiss an action `notwithstanding the objections of the person'" (quoting Searcy v. Philips Elecs. N. Am. Corp., 117 F.3d 154, 160 (5th Cir. 1997))).

Thus, consistent with the holdings of other circuits, we hold that the United States is not required to satisfy the good-cause intervention standard in § 3730(c)(3) when settling a qui tam action brought under the FCA.

B. Whether the Proposed Settlement Was "fair, adequate, and reasonable"

The United States "may settle the action with the defendant ... if the court determines, after a hearing, that the proposed settlement is fair, adequate, and reasonable under all the circumstances." 31 U.S.C. § 3730(c)(2)(B) (emphasis added). The United States and Keiser struck a deal providing that Keiser would pay the government $335,000 and the government would release Keiser from any further claims and preserve Keiser's eligibility for future Title IV funds. The district court, after a hearing, determined this proposed agreement was "fair, adequate, and reasonable" under the statute. Relators argue that the settlement was unreasonable because they were likely to win appellate reversal and thereafter secure a potentially enormous monetary recovery.4

In evaluating the fairness and reasonableness of the settlement, we first decide the proper standard for reviewing proposed settlements between the United States and an FCA defendant. We then turn to the substance of this particular agreement and conclude that it was "fair, adequate, and reasonable."

1. The proper standard for evaluating FCA settlements

Relators ask us to review the proposed FCA settlement under the same rubric used to evaluate proposed class settlement between private parties under Fed. R. Civ. P. 23. Some district courts have done just that. See United States ex rel. Schweizer v. Oce N. Am., Inc., 956 F.Supp.2d 1, 10-11 (D.D.C. 2013); United States ex rel. Resnick v. Weill Med. Coll. of Cornell Univ., No. 04 CIV. 3088 (WHP), 2009 WL 637137, at *2 (S.D.N.Y. Mar. 5, 2009) (unreported); United States ex rel. Nudelman v. Int'l Rehab. Assocs., Inc., No. Civ.A. 00-1837, 2004 WL 1091032, at *1 n.1 (E.D. Pa. May 14, 2004) (unreported). Rule 23(e)(2) uses almost the identical language found in 31 U.S.C. § 3730(c)(2)(B). It provides that a class settlement must be "fair, reasonable, and adequate" to protect the interests of absent class members. Fed. R. Civ. P. 23(e)(2). In reviewing the fairness of class settlements in the Eleventh Circuit, a court considers the likelihood of success at trial and the range of possible recovery, as well as the complexity and expense of the litigation and the stage of the proceedings at which the settlement is reached, among other factors. See, e.g., Day v. Persels & Assocs., LLC, 729 F.3d 1309, 1326 (11th Cir. 2013). In general, we analyze these factors without deference to the parties settling the dispute because the purpose of our judicial review is to protect absent class members from an unfair bargain. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997) (noting the review of class settlements "protects unnamed class members `from unjust or unfair settlements' ..." (quoting 7B Charles Alan Wright et al., Federal Practice and Procedure § 1797 (2d ed. 1986))).

But a qui tam FCA suit materially differs from class-action litigation between private parties for two reasons. First, unlike members of a plaintiff class, a qui tam relator has suffered no invasion of his own rights. See Vt. Agency of Nat. Res. v. United States ex rel. Stevens, 529 U.S. 765, 772-73, 120 S.Ct. 1858, 146 L.Ed.2d 836 (2000) (analyzing whether FCA relators have Article III standing when they have suffered no legally cognizable injury themselves). The relator instead brings the qui tam claim as the assignee of the United States' claim, with the government being the real party in interest to FCA actions. See id. at 773-74, 120 S.Ct. 1858; see also United States ex rel. Eisenstein v. City of New York, 556 U.S. 928, 934-35, 129 S.Ct. 2230, 173 L.Ed.2d 1255 (2009); Timson v. Sampson, 518 F.3d 870, 873 (11th Cir. 2008) (per curiam). So, unlike with class-action settlements, when the government settles a qui tam case, it is agreeing to compromise with respect to its own injuries only, not those of the relator or other absent parties.

Second, also unlike with private plaintiffs, the government's interests are not confined to maximizing recovery against the defendant. The United States may want to consider whether the maximum recovery is proportional to the seriousness of the misconduct. It may also wish to consider public policy consequences or political ramifications. Moreover, it could conclude that limited prosecutorial resources are not worth spending on continued monitoring of the case, which it often must do even when the relator is proceeding with the action. These considerations are entirely absent when evaluating a proposed class settlement between private parties. Thus, we conclude that the Rule 23 standard for class settlements between private parties is not the proper framework for evaluating FCA settlements between the government and the defendant.

Unlike reviewing class-action settlements, in the FCA context there must be considerable deference to the settlement rationale offered by the government. The need for deference arises out of the loose similarity between government-obtained FCA settlements and decisions by the government not to prosecute or enforce an administrative remedy, which are presumptively unreviewable. See, e.g., Heckler v. Chaney, 470 U.S. 821, 832, 105 S.Ct. 1649, 84 L.Ed.2d 714 (1985). For that reason, some circuits have said the government may dismiss a relator's claim so long as dismissal is rationally related to a valid government purpose. See Ridenour v. Kaiser-Hill Co., 397 F.3d 925, 936 (10th Cir. 2005); United States ex rel. Sequoia Orange Co. v. Baird-Neece Packing Corp., 151 F.3d 1139, 1145 (9th Cir. 1998).

Relators are correct, however, to point out that settlement under the FCA is treated differently than dismissal. When the government seeks to dismiss the FCA action, the statute does not prescribe a judicial determination of reasonableness, see 31 U.S.C. § 3730(c)(2)(A); whereas when the government wants to settle a qui tam claim, the court must find the settlement is "fair, adequate, and reasonable," § 3730(c)(2)(B). See United States ex rel. Schweizer v. Oce N.V., 677 F.3d 1228, 1233-34 (D.C. Cir. 2012) (offering a thorough analysis of why FCA settlements warrant more rigorous judicial review than dismissals). Because of this differential statutory treatment, government-obtained FCA settlements do not receive the same level of deference that applies to dismissals.

That being said, some limited deference is still warranted. The United States' decision to end a case through settlement is similar enough to a decision to dismiss the case — a choice committed to the discretion of the Executive Branch — that this Court must afford some degree of respect to the government's settlement rationale. At the same time, given the FCA's mandate that we review settlements for fairness, we cannot just rubber stamp the government's justifications. In light of these principles, our review of proposed settlements must account for the reasonableness of the rationale offered by the government as well as the potentially prejudicial effect on the relator. Thus, we ask whether the government has advanced a reasonable basis for concluding the settlement is in the best interests of the United States, and whether the settlement unfairly reduces the relator's potential qui tam recovery.

With this in mind, we proceed to evaluate the proposed settlement in this case.

2. The proposed settlement is "fair, adequate, and reasonable"

We conclude that the proposed settlement is "fair, adequate, and reasonable." It secures for the United States a recovery that is thirty times larger than the district court's award at trial — that means higher recovery for both the United States and Relators. See 31 U.S.C. § 3730(d) (establishing a qui tam relator's share as a variable percentage of the proceeds). From Relators' point of view, however, that is cold comfort. In their view, the potential recovery of billions of dollars is so large that the "expected value" — potential recovery multiplied by the odds of obtaining it — dwarfs the $335,000 settlement figure, rendering the government's settlement decision unreasonable.

But unlike with private actors, the government's decision to settle a case does not turn solely on whether settlement is most likely to maximize the monetary recovery. As discussed earlier, the government may conclude that settling the case imposes liability reasonably proportionate to the gravity of the misconduct. Or it might wish to avoid expending further resources in monitoring the case as it proceeds through the appellate and (if remanded) trial process. And most importantly for the United States here, the government must be wary of the precedential impact of a potentially adverse appellate decision. If the Eleventh Circuit affirmed the district court's restrictive reading of the FCA here, the government would be limited in its enforcement efforts all around the Circuit. That enforcement burden would ultimately translate into lower recoveries across the board.

The government was not unreasonable in hedging its bets in this regard. Much is made in the parties' briefs about whether Relators would have succeeded on appeal, and we offer no comment on the merits of that debate. But it is evident that, at the very least, Relators' case hinged on a proposition that is not settled in this Circuit: whether an educational institution that has falsely certified compliance with federal payment conditions can be liable under the FCA for financial-aid requests submitted by its students. Winning the bulk of Relators' claimed penalties and alleged damages relied crucially on being able to attribute the 230,000 student-submitted claims to Keiser, even though Keiser did not control the content or submission of those claims. But that is an open question in this Circuit, so the United States did not act unreasonably in preferring the certainty of a settlement to the uncertainty of an appeal.

The United States has thus advanced a reasonable basis for concluding this proposed settlement furthers its best interests without unfairly reducing Relators' qui tam recovery (in fact, it guarantees a higher recovery for them). It is thus "fair, adequate, and reasonable." We therefore find the proposed settlement satisfies § 3730(c)(2)(B).

C. Entitlement to an Evidentiary Hearing and Discovery

The FCA entitles a qui tam relator to a fairness hearing as of right when the United States proposes a settlement with an FCA defendant. 31 U.S.C. § 3730(c)(2)(B). That statutory entitlement, however, does not include the right to unearth new evidence at the hearing to attack the proposed settlement. Instead, the statutory hearing provision "serves a more limited purpose of forcing the government to provide some reasoning behind its decision to settle the case and giving the plaintiff-relators an opportunity to direct the court's attention to facts or allegations that would suggest the settlement was" unreasonable. United States ex rel. Schweizer v. Oce N. Am., Inc., 956 F.Supp.2d 1, 11 (D.D.C. 2013). In other words, the relator is afforded an opportunity to highlight existing evidence of impropriety or unreasonableness.

To conclude that the statute mandates the right to develop new evidence would risk ballooning the FCA settlement hearing into a mini-trial on the merits. The government would have to explain why it considered an appellate defeat possible in light of the facts and law governing the case at hand, the relator would then dispute that assessment, and a fight over the merits would ensue. But the United States often settles a case precisely to avoid such a fight. That "would put the cart before the horse, in essence making trial a precondition of settlement." Id. To avoid that result, we construe the statutory right to a hearing to include no more than a right to highlight existing evidence and make arguments, based on that evidence, that the proposed settlement is unreasonable or improper.5 That is all the statute compels.

A court still retains, however, the inherent equitable power to give more than the FCA minimally commands. A district court could give a qui tam relator the opportunity to present and develop new evidence when he shows a substantial and particularized need for such a hearing. A relator could, for example, potentially make that showing by presenting a colorable and non-speculative claim that the United States has neglected to investigate the allegations, or that its settlement decision was motivated by improper considerations, such as collusion or bribery.6 We stress that it is the relator's burden to demonstrate this substantial and particularized need — and absent such a showing the district court ordinarily will not afford an evidentiary hearing.

In this case, the district court denied Relators' request for an evidentiary hearing. We affirm that denial. Relators have not demonstrated a substantial or particularized need to develop new evidence because there is no colorable and non-speculative claim that the United States failed to investigate the allegations or acted with improper motives. Relators did offer several reasons why they have met that burden, but we reject those arguments as insufficient.

First, Relators repeat their contention that the government's settlement rationale was patently unreasonable. Therefore, in their view, the offered reason for the settlement must be a pretext for some ulterior motive. But that cannot be so. As we have already explained, the government's rationale was reasonable. So we decline to infer a nefarious motive from the facial reasonableness of the government's settlement rationale.

Second, Relators take issue with their exclusion from the settlement negotiations. Why would the United States have kept them in the dark about the ongoing settlement talks, Relators argue, if it were not hiding something? This allegation goes too far. The government explained that it excluded Relators from its direct negotiations with Keiser because Relators' confidence of appellate reversal and adamant demand for billions of dollars would have impeded productive settlement negotiations. In any event, the United States did invite Relators' counsel to meet and discuss the proposed settlement after a tentative deal was reached with Keiser. Relators' counsel accepted the invitation and explained his objection to the notion of settling the case while on appeal. Nevertheless, the United States decided to go forward with its settlement with Keiser. Relators were thus hardly kept in the dark such that we could plausibly infer improper motives by the government.

Third, Relators allege that the United States did not read their opening brief on appeal before settling the case. Thus, in their view, the government could not have been adequately informed about the likelihood of success on appeal. But the United States had been monitoring the case closely while it proceeded through trial — it knew the facts and made its own informed conclusion about the law. In any event, before the government submitted the proposed settlement to the district court, the government lawyer responsible for the final settlement decision did read Relators' opening appellate brief. It thus stretches credulity to believe that the United States made so uninformed a decision that Relators are entitled to probe further with an evidentiary hearing.7

In addition to the opportunity to develop evidence at a hearing, Relators requested discovery into the government's settlement rationale. They propounded interrogatories inquiring about each demand and offer made during negotiation of the proposed settlement, the government's evaluation of Relators' likelihood of success on appeal, the reasons the United States filed Statements of Interest after trial, and the identities of the government lawyers who participated in various aspects of the case. The district court shielded the United States from these requests, and we affirm that decision.

The FCA does not expressly entitle the objecting qui tam relator to discovery at all. See 31 U.S.C. § 3730(c)(2)(B). Further, much like an evidentiary hearing, discovery into the government's settlement rationale risks converting the court's evaluation of the settlement's reasonableness into a mini-trial on the merits. See Schweizer, 956 F.Supp.2d at 11. That is generally antithetical to the government's prerogative to end a qui tam case brought in its own name.

That being said, as with the evidentiary hearing, a district court has the equitable power to compel discovery when the relator demonstrates a substantial and particularized need for it. It is the relator's burden to come forward with a colorable and non-speculative claim that the government's settlement rationale is improper and that further disclosures are needed.8 But as earlier explained, Relators have come forward with no colorable and non-speculative claim of improper motive or inadequate investigation by the United States. Thus, the district court did not err in declining to compel discovery in this case.

D. Attorneys' Fees and Costs

After trial, Relators sought over $1 million in attorneys' fees and almost $76,000 in litigation expenses. The district court determined, however, that Relators' limited trial victory warranted an across-the-board reduction in their total fee award to $60,000. Relators now object to that reduction.9 Reviewing the award for abuse of discretion, see Bivins v. Wrap It Up, Inc., 548 F.3d 1348, 1351 (11th Cir. 2008), we reject most of Relators' arguments for reversal as meritless. But one particular contention warrants further discussion. They argue that the amount of recovery is not an appropriate consideration — or at least not a significant consideration — in awarding attorneys' fees for qui tam FCA cases. We disagree.

In Hensley v. Eckerhart, a civil-rights case involving attorneys' fees awarded under 42 U.S.C. § 1988, the Supreme Court held that the lodestar amount — which is "[t]he product of reasonable hours times a reasonable rate" — may be adjusted downward depending on the "results obtained." 461 U.S. 424, 434, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983) (internal quotation marks omitted). When the prevailing party achieved only a "partial or limited success," the lodestar figure "may be an excessive amount" because "the most critical factor is the degree of success obtained." Id. at 436, 103 S.Ct. 1933 (emphasis added). Moreover, when the district court "reduce[s] the award to account for the limited success[,] [t]he court necessarily has discretion in making this equitable judgment." Id. at 436-37, 103 S.Ct. 1933.

The Supreme Court then decided Farrar v. Hobby, which relied on Hensley in holding that a court may reduce the lodestar amount when the plaintiff won only nominal damages, i.e., when the plaintiff secured only a "technical" victory by proving a legal violation but suffered no actual damages. 506 U.S. 103, 114, 113 S.Ct. 566, 121 L.Ed.2d 494 (1992). "In a civil rights suit for damages," the court reasoned, "the awarding of nominal damages also highlights the plaintiff's failure to prove actual, compensable injury." Id. at 115, 113 S.Ct. 566. In light of this rationale, Farrar seems especially applicable to Relators' case here because Relators were not able to prove Keiser caused any actual damages — instead they proved that Keiser knowingly submitted two false claims, which entitled them to an award of only statutory penalties.

In spite of this direction from the Supreme Court, Relators ask us to limit Hensley and Farrar to claims for private damages — not FCA claims which involve a suit to recover from fraud on the public fisc. See Hensley, 461 U.S. at 440, 103 S.Ct. 1933 ("[P]laintiff's success is a crucial factor in determining the proper amount of an award of attorney's fees under 42 U.S.C. § 1988." (emphasis added)); Farrar, 506 U.S. at 114, 113 S.Ct. 566 ("Where recovery of private damages is the purpose of civil rights litigation, a district court, in fixing fees, is obligated to give primary consideration to the amount of damages awarded as compared to the amount sought." (emphasis added) (internal quotation marks omitted)). And Relators say there is a good public-policy reason not to extend the Hensley-Farrar principle to FCA cases because that would deter potential whistleblowers and their attorneys from bringing qui tam cases that might result in only nominal or a small amount of damages. See S. Rep. 99-345, at 29 (1986), reprinted in 1986 U.S.C.C.A.N. 5266, 5294 (explaining the importance of attorneys' fees in FCA qui tam actions).

But Relators offer no authority that declines to apply Hensley and Farrar to FCA claims. In fact, although we are not bound by it, we have already held that the amount of recovery is relevant to the fee award in FCA cases. See United States v. Patrol Servs., Inc., 202 Fed.Appx. 357, 359 (11th Cir. 2006) (unpublished) ("The district court may adjust the amount depending on a number of factors, including the quality of the result and representation of the litigation."). We reject Relators' plea to carve out an exception to the Hensley-Farrar principle for FCA cases. Relators contend that qui tam FCA claims are different because such claims serve a substantial public interest, but it is not clear why that makes them different from civil-rights cases which also serve an important public need — or even why that alleged difference should matter for the award of attorneys' fees. Accordingly, we affirm the district court's reliance on the degree of success in awarding attorneys' fees in this case.

That brings us to one final issue. Relators contend that the district court abused its discretion in reducing the lodestar amount by over ninety-five percent to reach $60,000. That reduction, Relators insist, was outside the district court's admittedly broad range of discretion under our recent case, Yellow Pages Photos, Inc. v. Ziplocal, LP (Yellow Pages), 846 F.3d 1159 (11th Cir. 2017) (per curiam). In that case, we reversed a fee award when the district court reduced the lodestar amount by more than ninety percent based on the plaintiff's limited success at trial. Id. at 1164. But we did so for a reason that does not apply in our case today.

In Yellow Pages, the district court adopted a "strictly mathematical approach" in arriving at its over ninety-percent reduction figure, which the Supreme Court has disfavored. Id. (citing Hensley, 461 U.S. at 435 n.11, 103 S.Ct. 1933; City of Riverside v. Rivera, 477 U.S. 561, 574, 106 S.Ct. 2686, 91 L.Ed.2d 466 (1986)). The district court there merely divided the damages awarded at trial by the maximum damages sought, and then mechanically applied that fraction to the lodestar amount. Id. at 1164-65. That did not happen here. The district court expressly considered the "policy considerations on all sides (including the encouragement of whistle-blowers, the costs incurred by [Keiser] of defending the lawsuit and considerations of judicial economy), and the fact that the suit may have had a positive impact on [Keiser's] conduct." This approach was consistent with this Court's opinion in Yellow Pages, which made clear that a reduction from a rote application of an arithmetic formula is not permissible. The district court considered all of the circumstances, including the public benefit Relators served. And under our precedent, a public benefit does not "foreclose a reduction in the lodestar amount." Popham v. City of Kennesaw, 820 F.2d 1570, 1580 (11th Cir. 1987). If it did, then "Hensley would be incomprehensible because a reduction is attorney's fees ... would never be warranted." Id. On this record, we hold the district court did not abuse its discretion in awarding fees and costs.10

III. CONCLUSION

We AFFIRM the district court on all issues.