William Klopfenstein v. Fifth Third Bank

READLER, Circuit Judge. A class of borrowers sued Fifth Third Bank for failing to properly disclose the annual percentage rate (APR) for short term loans the bank offered through its “Early Access” program. The parties litigated their dispute for over a decade, with the class prevailing at summary judgment on its Truth in Lending Act (TILA) claim, 15 U.S.C. § 1601 et seq., but losing at trial on its breach of contract claim. In accordance with TILA’s command to award attorney’s fees to persons who successfully litigate a TILA suit, see id. § 1640(a)(3), the district court awarded the class about $3.3 million in attorney’s fees. Fifth Third challenges that award as unreasonable. Save for the issue of the rates applied to some counsel, we disagree and affirm the district court.

I.

Starting in 2008, Fifth Third Bank offered an “Early Access” cash advance loan program. In practice, Fifth Third charged $1 for every $10 borrowed from a customer’s account, collecting the fee either when a direct deposit of $100 or more posted or after 35 days, whichever came first. In re Fifth Third Early Access Cash Advance Litig., 925 F.3d 265, 269 (6th Cir. 2019). Fifth Third represented that this flat 10% fee translated to a 120% APR. APR, however, is a measure of the cost of credit over time. So the APR for a flat fee fluctuates based on how soon the customer repays the loan. Because customers often repaid the advances well in advance of 35 days—on average about 11 days—the annualized rate frequently exceeded 120%. Klopfenstein v. Fifth Third Bank, --- F.4th ---, 2026 WL 1506439, at *2 (6th Cir. May 29, 2026).

A. Recognizing that discrepancy and seeking to challenge the bank’s lending arrangement, William Klopfenstein, on behalf of a class of Fifth Third customers, filed this action in 2012 in the United States District Court for the Northern District of Ohio. (All plaintiffs and class members are hereinafter referred to collectively as “the class.”) The complaint alleged that Fifth Third misrepresented the cost of the Early Access product by disclosing an inaccurate APR. Based on that alleged misrepresentation, the class asserted claims for breach of contract, fraud, violations of Ohio’s usury law, conversion, unjust enrichment, and unconscionability. Early in the case’s long history, the class’s Ohio-based counsel coordinated with Tycko & Zavareei (Tycko), a Washington, D.C., firm specializing in these sorts of class actions. Tycko, alongside attorneys from two other, non-D.C. based firms, would later be appointed interim lead counsel for the class.

Klopfenstein’s case was transferred to the United States District Court for the Southern District of Ohio in Cincinnati, where Fifth Third is headquartered. And it was later consolidated with similar actions originally filed in Florida, Kentucky, Tennessee, and Illinois, all of which had likewise been transferred to the Southern District of Ohio. After consolidation, the class filed an amended complaint. Many of the claims in the original complaint were included in its amended complaint, including the class’s breach of contract claim, which alleged that Fifth Third breached the Early Access contract by charging APRs exceeding 120% and by failing to provide accurate APR summaries on monthly bank statements. In re Fifth Third, 925 F.3d at 280.

The class also added a new claim in the amended complaint, one premised on TILA, a statute that imposes strict liability disclosure requirements on those who extend credit to customers. 15 U.S.C. § 1601 et seq. TILA requires lenders to disclose the “annual percentage rate,” a term that represents the annual cost of borrowing money. Id. § 1638(a)(4). The amended complaint alleged that Fifth Third violated TILA by failing to provide accurate and meaningful APR disclosures. In re Fifth Third, 925 F.3d at 274–75. Because Fifth Third allegedly communicated this inaccurate APR to thousands of customers, the class sought certification of a nationwide class.

Following a few years of litigation, the parties reached a mediated settlement. The settlement amount—$8.5 million—exceeded the potential statutory recovery available under TILA, which caps damages at $2 million. The class, however, later withdrew from the settlement after an expert estimated that potential contract damages could be much higher.

With the litigation resumed, the district court dismissed all claims except the TILA claim. Klopfenstein v. Fifth Third Bank, No. 12-cv-00851, 2015 WL 1468382, at *9 (S.D. Ohio Mar. 30, 2015). Invoking Federal Rule of Civil Procedure 54(b), the class sought entry of final judgment on the dismissed contract claim to allow for an immediate appeal. The district court granted that request, and the class in turn timely appealed the dismissal of the contract claim. After affirming the Rule 54(b) certification ruling and turning to the appeal’s substantive component, we held that the contract language was facially ambiguous, which required remanding the contract claim for further proceedings. In re Fifth Third, 925 F.3d at 280.

On remand, the district court certified two classes under Rule 23(b)(3)—one for the breach of contract claim and one for the TILA claim. The parties then engaged in two years of additional discovery and trial preparation. Following discovery, the district court granted summary judgment to the TILA class on its claim and awarded the statutory maximum of $2 million in damages. But the court denied cross motions for summary judgment on the breach of contract claim, which proceeded to trial. There, a jury found that Fifth Third breached the contract but that the voluntary payment doctrine, an affirmative defense to breach of contract claims in Ohio, barred recovery.

B. Following trial, the class moved for attorney’s fees under TILA’s fee shifting provision, which entitles prevailing plaintiffs to reasonable attorney’s fees. 15 U.S.C. § 1640(a)(3). In making its request, the class excluded from its fee calculation time entries made after the summary judgment ruling on the TILA claim. As for fees incurred before that ruling, the class similarly excluded what it believed was work attributable solely to the contract claim. Factoring in those exclusions, the class sought $5,638,622.53 in fees, plus $315,572.13 in costs and $50,000 in service awards. The class derived the fees total using the adjusted Laffey rates, which are the hourly attorney rates for the Washington, D.C., market adjusted for inflation. They applied these rates for all attorney work, regardless of whether the work was performed by Tycko. Fifth Third countered that the court should apply the local forum’s rates—known as the Rubin rates, which fall below the adjusted Laffey rates.

Following its review of the billing entries, the district court first excluded time it found attributable solely to the contract claim, including time related to a damages expert retained only for that claim. Those reductions went beyond what the class had excluded and eliminated $1,736,518 from its fee request. The court then applied a 15% across the board reduction for excessive or duplicative work and costs. The court declined to further reduce the total hours based on the class’s lack of overall success, in particular, its loss on the breach of contract claim.

As to what hourly rates should apply to the fee request, the district court chose the Laffey rates. The court declined to apply the Rubin rates on the grounds that the case involved multiple jurisdictions and required specialized expertise associated with national practice. Critically, the court utilized the Laffey rates for each class attorney, regardless of where the counsel practiced or the extent of the counsel’s role in the litigation. In the end, the court awarded $3,317,128.60 in attorney’s fees—about 59% of the amount requested. On appeal, Fifth Third challenges that award as an abuse of discretion.

II.

A district court’s discretion in awarding fees is wide, in part because we do not wish to encourage further litigation that has “nothing to do with the underlying cause of action.” Waldo v. Consumers Energy Co., 726 F.3d 802, 828 (6th Cir. 2013) (Sutton, J., concurring in part and dissenting in part); see also Ne. Ohio Coal. for the Homeless v. Husted, 831 F.3d 686, 702–03 (6th Cir. 2016) (highlighting the “desirability of avoiding frequent appellate review of what essentially are factual matters” (citation modified)). We thus rely heavily on the district court’s familiarity with the litigation and its evaluation of billing records. See Imwalle v. Reliance Med. Prods., Inc., 515 F.3d 531, 551 (6th Cir. 2008); Gonter v. Hunt Valve Co., Inc., 510 F.3d 610, 620 (6th Cir. 2007). With these guideposts in mind, it follows that we will reverse the district court’s award of attorney’s fees only if it abused its discretion in resolving a fee request, Linneman v. Vita-Mix Corp., 970 F.3d 621, 623–24 (6th Cir. 2020), for instance, if the district court, “relie[d] upon clearly erroneous findings of fact, applie[d] the law improperly, or use[d] an erroneous legal standard.” Ne. Ohio Coal., 831 F.3d at 702 (quoting Imwalle, 515 F.3d at 551). In other words, we will reverse an attorney’s fee award only when we are convinced that the district court “committed a clear error of judgment.” Paschal v. Flagstar Bank, 297 F.3d 431, 434 (6th Cir. 2002). To facilitate our review, we ask that the district court provide us a “concise but clear explanation of its reasons for the fee award.” Hensley v. Eckerhart, 461 U.S. 424, 437 (1983).

Fifth Third argues that the district court abused its discretion in four respects: (1) concluding that the breach of contract claim and the TILA claim were “related” and awarding fees for time spent litigating both claims; (2) reducing the award for overstaffing by only 15%; (3) declining to reduce the award for lack of success; and (4) applying the Laffey rates to all counsel.

A. We begin with Fifth Third’s relatedness argument. But before turning to the specifics of that issue, it bears addressing the underlying legal framework governing attorney’s fees awards. The background rule in the United States is that each party bears its own litigation expenses, including attorney’s fees. Fox v. Vice, 563 U.S. 826, 832 (2011) (citing Alyeska Pipeline Serv. Co. v. Wilderness Soc’y, 421 U.S. 240, 247 (1975)). But in certain circumstances, Congress has authorized courts to deviate from this practice by enacting fee shifting provisions. Id. (citing City of Burlington v. Dague, 505 U.S. 557, 561–62 (1992) (listing examples)). TILA, the statute involved here, includes one such provision: A prevailing plaintiff under TILA is entitled to recover “the costs of the action, together with a reasonable attorney’s fee as determined by the court.” 15 U.S.C. § 1640(a)(3).

There is no dispute that the class, having prevailed on its TILA claim, is entitled to reasonable fees. But how does a district court go about determining a reasonable fee award? In a nutshell, the court begins by calculating what is known as the “lodestar” amount, that is, “the number of hours reasonably expended on the litigation multiplied by a reasonable hourly rate.” Hensley, 461 U.S. at 433 & n.7 (explaining that “the standards set forth in [Hensley] are generally applicable in all cases in which Congress has authorized an award of fees to a ‘prevailing party’”); Echols v. Express Auto, Inc., 857 F. App’x 224, 226 (6th Cir. 2021) (applying Hensley to the Equal Credit Opportunity Act); Iroanyah v. Bank of Am., 753 F.3d 686, 694 (7th Cir. 2014) (applying Hensley to TILA’s fee shifting provision). After calculating the lodestar (or baseline amount), the district court may make upward or downward adjustments to arrive at a final fee award. See Adcock-Ladd v. Sec’y of Treasury, 227 F.3d 343, 349 (6th Cir. 2000) (citing Johnson v. Ga. Highway Express, Inc., 488 F.2d 714, 717–19 (5th Cir. 1974)) (listing upward and downward adjustment factors a court may consider). In the end, the trial court’s goal is to ensure that the fee award is reasonable in relation to the overall success achieved. 10 Wright & Miller’s Federal Practice & Procedure § 2675.2 (4th ed. 2025) (explaining that the Supreme Court has cautioned against “excessive” fee awards relative to the success achieved).

Where a plaintiff prevails only on claims with fee shifting provisions, the lodestar calculation is relatively straightforward. A trial court first assures itself that the hours logged by the prevailing party are reasonable and accurate, and then multiplies those hours by the hourly rate to find the fee award. Hensley, 461 U.S. at 433. But where the plaintiff also litigates claims that do not independently authorize fees, the analysis becomes more complicated. As much of a litigator’s work often is devoted to the litigation as a whole rather than individual claims, litigation efforts across a multi-claim case are not always readily separable by claim. Recognizing the challenge in fairly calculating fees in these muddled situations, the Supreme Court in Hensley concluded that a plaintiff need not necessarily prevail on a claim presented to be eligible to recover some attorney’s fees for that claim. See 461 U.S. at 435–37. So long as an unsuccessful claim is “related” to a successful, fee-entitling claim, the plaintiff may receive fees for time spent on the unsuccessful claims, depending upon the circumstances of the litigation. See id. at 435. Claims are related when they “involve a common core of facts or [are] based on related legal theories.” Id. On the other hand, claims are not considered related when they are “distinct in all respects.” Id. at 440; see also Johnston v. Borders, 36 F.4th 1254, 1282–86 (11th Cir. 2022) (per curiam) (concluding that a defamation claim was unrelated to a § 1983 due process claim because the claims arose from the conduct of different defendants in different instances).

That leaves a district court to determine whether the unsuccessful claims relate to the successful, fee-entitling claims and, more broadly, whether the degree of success justifies the hours expended. Imwalle, 515 F.3d at 552 (citing Hensley, 461 U.S. at 434). That process, however, typically cannot be distilled into a precise mathematical “formula.” Hensley, 461 U.S. at 436. Instead, it is better described as an “equitable judgment,” one for which a district court enjoys “discretion in making.” Id. at 436–37.

1. With these understandings in mind, the district court did not abuse its discretion in concluding that the successful TILA claim (for which Congress authorized fee shifting) and the unsuccessful breach of contract claim were related for purposes of determining a fee award. The claims “involve[d] a common core of facts” in that they both arose from disclosure issues with Fifth Third’s Early Access program. Sakhawati v. Lynch, 839 F.3d 476, 480 (6th Cir. 2016) (quoting Hensley, 461 U.S. at 435). The breach of contract claim, recall, hinged on alleged misrepresentations about the APR, while the TILA claim concerned failure to make proper disclosures about the APR. In line with the claims’ shared origin, it is no surprise that much of the class counsel’s work on the TILA claim overlapped with the breach of contract claim. Take discovery, for example. Both claims required discovery into matters such as Fifth Third’s standardized loan agreements and monthly account statements containing the 120% APR disclosure. Likewise, internal documents and testimony regarding how the bank calculated the disclosed APR and understood the relationship between the flat 10% fee and annualized interest were relevant to both claims. The same goes for discovery of data showing when advances were typically paid back and what APR customers actually paid as well as Fifth Third’s policies and training materials governing how the product and its cost were explained to customers. Those related aspects were on display during motions practice as well, where many of the issues that the class addressed furthered both claims, such as in its motion for class certification.

We are not alone in viewing the issue this way. For purposes of calculating attorney’s fees, a number of courts have similarly deemed TILA and breach of contract claims or the like related. See, e.g., Greene v. Gibraltar Mortg. Inv. Corp., 529 F. Supp. 186, 188 (D.D.C. 1981) (concluding plaintiff’s TILA and common law claims were related because they “arose from the same operative facts” and “concerned the same issues, i.e., the lack of notice and failure to disclose significant terms”); Moore v. Bank of Am., N.A. (USA), No. 03-CV-0520, 2005 WL 8173189, at *5 (S.D. Cal. Oct. 6, 2005) (similar); cf. Williams v. First Gov’t Mortg. & Invs. Corp., 225 F.3d 738, 746 (D.C. Cir. 2000) (affirming a determination that a TILA and an unconscionability claim were related).

Of course, not all of the class’s work on the breach of contract claim related to its efforts to advance the TILA claim. One example is work concerning how the class understood the terms of its payments. Those efforts were relevant only to the breach of contract claim, as TILA imposes strict liability for a failure to disclose and does not depend on a consumer’s confusion or lack thereof. See 15 U.S.C. § 1640. That said, we note that the class did not request fees for work tied solely to its breach of contract claim. And the district court, for its part, undertook a line-by-line check and eliminated any fees it believed were associated with work unrelated to the prevailing TILA claim, resulting in a $1.7 million reduction to what the class requested. All things considered, we see no error in the district court’s assessment of the relationship between the class’s lead claims.

2. Fifth Third disagrees. To start, it highlights the fact that the district court, in ruling on the class’s Rule 54(b) motion, treated the two claims at issue as distinct for certain merits purposes, which the bank says undermines the court’s relatedness rationale at the fees stage.

The Rule 54(b) inquiry, however, is different from the one conducted at the fees stage. Rule 54(b) addresses whether claims are sufficiently distinct to support entry of final judgment; a claim may be “separate” for Rule 54(b) purposes even if it is not “entirely distinct from all the other claims in the action.” Wright & Miller’s Federal Practice & Procedure, supra, § 2657. This aligns with the Rule’s goal, which is to allow for a limited exception to the customary single appeal when the legal principles undergirding the resolution of one claim do not affect the other pending claims. Hensley, by contrast, asks a separate question: whether claims are so unrelated that work between them can be cleanly divided, describing claims as unrelated only when they are “distinct in all respects.” 461 U.S. at 440. Again, that standard reflects a practical concern, namely, that it is often “difficult [for a district court] to divide the hours expended on a claim-by-claim basis.” Id. at 435; see also Thorne v. City of El Segundo, 802 F.2d 1131, 1141 (9th Cir. 1986) (“The test for relatedness of claims is not precise.”).

In other words, claims may be distinct for Rule 54(b) purposes but still have overlap. And Hensley treats that overlap as a reason not to segregate fees. See Johnson v. Orr, 897 F.2d 128, 132 (3d Cir. 1990) (Becker, J., concurring) (“Claims can be sufficiently interrelated to require joint consideration for Hensley purposes but sufficiently discrete to be severable from one another on the merits under Rule 54(b) for purposes of appellate jurisdiction.”). That distinction resolves any apparent tension here. Even if the breach and TILA claims were sufficiently distinct to qualify as separate claims under Rule 54(b) and thus proceed on different procedural tracks, it was not unreasonable for the district court to find that they were not “distinct in all respects” for Hensley purposes. Hensley, 461 U.S. at 440. Both claims turned on the same alleged defect in the APR disclosure in the Early Access loan agreement and required overlapping factual and legal development. Because counsel’s work on the two claims substantially overlapped, the district court did not err in treating them as related and awarding fees for work performed on both.

Fifth Third adds that TILA and contract liability involve different legal elements. And that fact, says the bank, is yet one more reason why the work done on the respective claims was distinct, meaning that the class ought to have separated the fees associated with the two claims. We do not entirely discount Fifth Third’s point. As we noted in our first pass at this litigation, “[w]hile the breach-of-contract claim focuses on the amount of money that Fifth Third charged plaintiffs, the TILA claim is based on . . . [a] failure to make required disclosures.” In re Fifth Third, 925 F.3d at 274. Perhaps another district court might reasonably conclude that a TILA claim is distinct from a breach of contract claim given the differing conduct that gives rise to liability for each claim. But we do not deem the district court to have abused its discretion in holding otherwise.

Nor do we credit Fifth Third’s contention that the district court deemed the two claims “85%” in common leading up to its summary judgment ruling, suggesting the district court treated virtually all work as shared between the claims. Appellant Br. 21. That figure, it bears noting, came from class counsel’s fee petition, not the number of hours the district court credited. We likewise reject Fifth Third’s contention that the amount of time class counsel spent on the TILA claim was unreasonable because “[n]o reasonable attorney would spend thousands of hours to litigate a strict-liability TILA claim.” Id. at 22. The district court, for its part, found the claims related and then independently reviewed the billing records line by line, excluding time it concluded was not reasonably attributable to both claims. That methodology produced roughly a $1.7 million reduction in the fees requested. And that approach, coupled with the district court’s assessment of the relatedness of the TILA and breach of contract claims, is enough to satisfy us that the district court did not abuse its discretion in determining that the TILA and breach of contract claims were related.

B. Turn next to Fifth Third’s argument that the district court abused its discretion by declining to further reduce the fee award to reflect the class’s purported limited success. Fifth Third is correct that the “reasonableness of a fee award turns in part on the level of a plaintiff’s success—the extent to which he prevailed.” Waldo, 726 F.3d at 829 (Sutton, J., concurring in part and dissenting in part) (citation modified). In some cases, a plaintiff will be deemed to have secured so little success that no fee award is appropriate at all. See Farrar v. Hobby, 506 U.S. 103, 115 (1992) (holding that “the only reasonable fee” for a plaintiff who recovers only nominal damages may be “no fee at all,” despite his status as a prevailing party). Relatedly, in cases where a plaintiff achieves some (but not complete) success, the lodestar may be excessive to the extent it accounts for efforts that did not lead to victory—even where, as here, claims were interrelated and brought in good faith. Hensley, 461 U.S. at 436.

As these examples reveal, the “success” inquiry in this setting involves significant judgment. After all, as is true in many facets of life, how we measure success has no shortage of possibilities. In the fees setting, success is not measured mechanically, neither by counting claims won or lost nor by comparing damages sought to damages recovered. See Imwalle, 515 F.3d at 554–56. Instead, we employ a holistic approach, taking account of all corners of the litigation. Having done so, we then aim to arrive at a fee award that is “reasonable in relation to the results obtained.” Hensley, 461 U.S. at 440.

The district court reasonably concluded that the class achieved meaningful success over the course of the litigation. In particular, the class prevailed fully on one of two claims that made it to final judgment: The class obtained the maximum statutory recovery available under TILA— $2 million—for hundreds of thousands of consumers, an outcome that could fairly be deemed “excellent.” Waldo, 726 F.3d at 824 (quoting Hensley, 461 U.S. at 435). That the class’s recovery was capped is a product of the statutory scheme, not any failure of the claim itself. Cf. Hines v. City of Columbus, 676 F. App’x 546, 556 (6th Cir. 2017) (affirming a reduced fee award where the plaintiff recovered a small fraction of the available damages on the successful claim). And that successful claim, it bears adding, came with a right to attorney’s fees under a fee shifting statute. At the same time, the district court fairly accounted for the class’s mixed success overall, given its failure to prevail on other claims, including the breach of contract claim. The court excluded work attributable solely to the breach claim and reduced the overall fee request substantially. Having already incorporated the litigation’s mixed results into the fee calculation, the district court was not required to impose an additional reduction. For future purposes, it bears emphasizing then that the success determination will always turn on the facts of the case at hand. In the context of today’s case, we do not believe the district court abused its discretion in concluding that the class secured enough success to entitle it to a sizable fee award.

Fifth Third resists this conclusion on several grounds. First, the bank highlights the numerous theories raised by the class that were later dismissed. But the class does not seek fees for those claims; it seeks fees only for work related to its successful TILA claim. And, again, while some of the work on those claims advanced ultimately unsuccessful theories, work performed may still reasonably contribute to ultimate success even if the claims eventually narrow. See Hensley, 461 U.S. at 435. In this instance, the narrowing of theories over the course of a decade-long litigation alone does not require a further fee reduction.

Second, Fifth Third argues that the fee award is disproportionate to the damages recovered, noting that the class recovered “less than 1% of what they sought.” Appellant Br. 29 (emphasis omitted). True, the amount of damages a plaintiff recovered can inform our attorney’s fees analysis. See City of Riverside v. Rivera, 477 U.S. 561, 574 (1986) (plurality opinion); Isabel v. City of Memphis, 404 F.3d 404, 415 (6th Cir. 2005). At the same time, proportionality is not required. Waldo, 726 F.3d at 824 n.5. In the end, the district court reasonably accounted for proportionality concerns by excluding time devoted solely to the contract claim and reducing the requested lodestar.

Third, Fifth Third points to the class’s rejection of the bank’s 2016 settlement offer, which Fifth Third contends would have produced a larger and earlier recovery for the class, meaning that the class should not be entitled to any fees incurred thereafter. Here too, Fifth Third is correct to note that a district court may consider a rejected settlement offer when evaluating a fee award’s reasonableness, especially where a plaintiff’s later recovery is not appreciably larger. See McKelvey v. Sec’y of U.S. Army, 768 F.3d 491, 495 (6th Cir. 2014). But we do not require a court to do so. And our decision in McKelvey, contrary to Fifth Third’s suggestion, does not say otherwise. There, the plaintiff rejected a settlement offer that exceeded the relief legally available at trial, and much of the subsequent litigation pursued remedies the statute did not permit. Id. at 494. Against that backdrop, we unsurprisingly affirmed the district court’s decision to discount fees in light of the plaintiff’s limited success. Id. at 493–94. Here, on the other hand, the class had viable TILA and contract claims at the time of settlement. That the class ultimately failed to prevail on the contract claim following a jury trial does not automatically render the earlier decision to continue litigating unreasonable. See Dowling v. Litton Loan Servicing LP, 320 F. App’x 442, 449–50 (6th Cir. 2009). Nor is it improper for class counsel to continue litigating claims that they reasonably believe could produce substantially greater recovery for the class, especially when class counsel has a fiduciary obligation to vigorously prosecute viable claims for its clients. Cf. In re Dry Max Pampers Litig., 724 F.3d 713, 718, 721 (6th Cir. 2013) (reversing the district court’s approval of a settlement where class counsel accepted inadequate relief for absent class members).

All told, we reject Fifth Third’s invitation to reweigh the record and impose our own view of how success should be measured. Another district court could reasonably have reduced the fee award further to reflect the class’s mixed success. But that does not mean the district court here acted outside the range of permissible choices.

C. Turn next to Fifth Third’s assertion that the district court erred by declining to reduce the fee award in light of class counsel overstaffing the case, leading to unnecessary fees. A district court may reduce compensable hours where the record shows unnecessary duplication of effort. Hensley, 461 U.S. at 434. We review a district court’s determination as to whether claimed hours are excessive or duplicative for clear error, recognizing that the district court understandably is best positioned to assess staffing decisions and billing efficiency for litigation in its courtroom. Wayne v. Village of Sebring, 36 F.3d 517, 532 (6th Cir. 1994).

With that in mind, we see no basis to upset the district court’s assessment of Fifth Third’s overstaffing objections. After reviewing the class’s billing entries, the district court observed some duplication and inefficiency and applied a 15% across the board reduction. We defer to the district court’s findings in this respect. Orduno v. Pietrzak, 932 F.3d 710, 720 (8th Cir. 2019) (noting the need for “great deference” to a district court’s factual findings about redundant or excessive hours billed in awarding fees). Offering only generalities as to why the class billed excessively for legal research, as does Fifth Third, provides us no specific reason why the district court’s 15% reduction was a “clear error of judgment.” See Paschal, 297 F.3d at 434.

D. Lastly, Fifth Third challenges the hourly rates used by the district court to calculate the fee award. To the bank’s mind, the district court abused its discretion by applying billing rates untethered to the relevant legal market and failing to adequately justify that choice. Fifth Third is correct, but only in part.

As explained, attorney’s fees are calculated using the lodestar method: “the number of hours reasonably expended . . . multiplied by a reasonable hourly rate.” Hensley, 461 U.S. at 433. A “reasonable hourly rate” is the prevailing market rate in the “relevant community.” Linneman, 970 F.3d at 630 (quoting Hadix v. Johnson, 65 F.3d 532, 535–36 (6th Cir. 1995)); see Ne. Ohio Coal., 831 F.3d at 715–16. We utilize the geographic market where the district court sits as the default “relevant community,” Gonter, 510 F.3d at 618 (quoting Lamar Advert. Co. v. Charter Township of Van Buren, 178 F. App’x 498, 502 (6th Cir. 2006)), in recognition of the notion that “[l]ocal lawyers litigating a case in a local courthouse should receive local billing rates.” Linneman, 970 F.3d at 630. Departures from forum rates are permissible only in narrow circumstances, primarily when the party requesting the fees shows that competent local counsel was not a suitable option given the complexity and subject matter of the case, making it reasonable in the first instance to hire out-of-town-counsel. Hadix, 65 F.3d at 535.

Here, the dispute over rates is almost entirely about defining the relevant community. Is it the national market for class action attorneys? Or the geographic market in which the forum sits (here, the Southern District of Ohio)? If the former, the parties appear to agree that we should apply the Laffey rates, which reflect the market rate for complex federal litigation in Washington, D.C. If the latter, we typically apply the Rubin rates as a benchmark for prevailing local market rates. Linneman, 970 F.3d at 630.

1. Begin with the rates applied to co-lead counsel Tycko. The district court did not abuse its discretion in utilizing D.C. rates for Tycko’s work even though the matter was litigated in Cincinnati. To justify applying non-forum rates, the class had to show that nonlocal counsel’s participation was “reasonable in the first instance” and that “the rates sought by [Tycko] are reasonable for an attorney of [a comparable] degree of skill, experience, and reputation.” Hadix, 65 F.3d at 535. We see no basis to question the district court’s conclusion that the class satisfied both requirements.

To begin, it was reasonable for the class’s Ohio-based counsel to work with Tycko from the litigation’s early stages. A case’s complexity can justify the involvement of an out-of-town specialist where “competent counsel capable of handling [the] complex litigation would not have been available locally at a significantly lower rate.” Id. at 535 (citing Chrapliwy v. Uniroyal, Inc., 670 F.2d 760, 769 (7th Cir. 1982)). That is the case here. Although many TILA cases are simple disputes about disclosure requirements, today’s case is different. See, e.g., Purtle v. Eldridge Auto Sales, Inc., 91 F.3d 797, 800–02 (6th Cir. 1996). This litigation reflected a complicated dispute over how to calculate and disclose an APR for a product with a variable loan term and fixed fee, a convoluted issue that required expert testimony to unravel. And the case spanned more than a decade. Along the way, the litigation grew to encompass five separate class actions from multiple states and produced repeated appeals. In re Fifth Third, 925 F.3d at 269–272; Klopfenstein, 2026 WL 1506439, at *2. This complicated terrain was not unfamiliar to Tycko. The firm had previously litigated a materially identical case involving a similar product (a payday loan) and TILA disclosure, a case the district court described as indistinguishable from this one. See Small v. BOKF, N.A., No. 13-cv-01125, 2014 WL 3906257 (D. Colo. Aug. 7, 2014). That experience makes the firm’s involvement in this case especially rational. See Sigley v. Kuhn, 205 F.3d 1341, 2000 WL 145187, at *7 (6th Cir. 2000) (table) (explaining that a case’s complexity justifies hiring out-of-town counsel).

The rates applied to Tycko were likewise reasonable. In making that determination, district courts may “look to a national market . . . or any other market they believe appropriate to fairly compensate particular attorneys in individual cases.” Louisville Black Police Officers Org. v. City of Louisville, 700 F.2d 268, 278 (6th Cir. 1983) (citing Chrapliwy, 670 F.2d at 760). After reviewing declarations addressing Tycko’s expertise and historic billing rates, the district court applied the rates of the firm’s home forum, as indicated by the Laffey Matrix. That decision was reasonable, especially when one considers that the class made good use of Tycko’s abilities. Based on the district court’s fee award, Tycko undertook more work than the other co-lead counsel and the other attorneys for the class. To be sure, the district court could have done more to explain the need for Tycko in this litigation. But all things considered, we do not think the district court’s application of the firm’s home rates constitutes an abuse of discretion.

2. Just because it was reasonable to apply the relatively higher Laffey rates to Tycko, however, does not mean it was reasonable to do so across the board. In our eyes, the district court abused its discretion in failing to adequately explain why it awarded Laffey fees to counsel other than Tycko.

Recall how the other class counsel came to litigate in the Southern District of Ohio. Lawyers out of Florida, Illinois, Missouri, and Kentucky brought similar lawsuits against Fifth Third in a collection of district courts. See Laskaris v. Fifth Third Bank, No. 13-cv-20529 (S.D. Fla.); McQuillen v. Fifth Third Bank, No. 13-cv-00476 (W.D. Ky.); Harrison v. Fifth Third Bank, No. 13-cv-00046 (M.D. Tenn.); Fyock v. Fifth Third Bank, No. 3:13-cv-00438 (S.D. Ill.). Fifth Third succeeded in transferring venue for those cases to the Southern District of Ohio, where the cases were consolidated with the Klopfenstein case.

After transfer, the consolidated case was litigated in the Southern District of Ohio. Given the transfer, attorneys could have requested the rates that apply to the forum district (here, Rubin rates) or the rates of their home jurisdictions. See Wayne, 36 F.3d at 533 (affirming the district court’s application of local rates instead of the rates of attorneys’ home fora); see also A.R. ex rel. R.V. v. N.Y.C. Dep’t of Educ., 407 F.3d 65, 81 (2d Cir. 2005) (a district court has discretion “to determine the relevant community for calculating attorneys’ fees where the case was not commenced and litigated in a single federal district”). As no party asked for any rate other than the Laffey or Rubin rates, we resolve only whether the district court abused its discretion in applying Laffey rates rather than Rubin rates.

At bottom, the community market rule is rooted in geography. Attorneys receive prevailing market rates for comparable attorneys in the district where the court sits. See Gonter, 510 F.3d at 618 (citing Lamar Advert., 178 F. App’x at 502); Linneman, 970 F.3d at 630. In the Southern District of Ohio, those are the Rubin rates. That being the case, a district court must justify any departure from applying community market rates. See Linneman, 970 F.3d at 630. Doing so requires a “specific showing of necessity” as to why counsel is entitled to out-of-area rates. Id.; see also Adcock-Ladd, 227 F.3d at 350.

The district court came up short in this one modest respect. The court’s stated basis for applying the Laffey rates was its conclusion that Rubin rates would not provide “sufficient compensation” given counsel’s “status and ability.” R. 323, PageID 12735. For support, the court noted that the class “submitted declarations, resumes, and documentation, which confirm their nationwide status and practices;” that it “had the opportunity to read and observe all of the attorneys’ work in this case and is confident in their status and ability;” and emphasized that the litigation was “a national practice case, requiring specialized knowledge and expertise.” Id. That fairly high-level rationale, however, was not enough to justify departure from the community market rule. Although the district court recognized that the class counsel performed well, as one would hope to be the case, it did not explain why local billing rates would not have been “sufficient to encourage competent representation” from local counsel in the case. Linneman, 970 F.3d at 630. To the contrary, that local counsel typically initiated the cases near their home communities (only to have those cases later consolidated in this action) suggests that prevailing local rates were sufficient to incentivize competent representation. See id. The district court abused its discretion when it concluded otherwise without additional explanation. See Hadix, 65 F.3d at 535–37 (holding that failure to apply forum rates without required findings is an abuse of discretion).

Disagreeing with our assessment, the class insists that the “relevant community” in this setting should not be based on the court’s location. Instead, says the class, the appropriate community is “class action attorneys with nationwide practices.” Appellee Br. 34. But that is not what our cases dictate. Again, we view the relevant community as the location of the forum district. Linneman, 970 F.3d at 630. From there, a district court has the discretion to make upward and downward adjustments based on factors like the counsel’s performance or a case’s complexity. See Blum v. Stenson, 465 U.S. 886, 898 (1984). But the starting point is the prevailing market rate for attorneys in the relevant geographic community. See B & G Mining, Inc. v. Dir., Off. of Workers’ Comp. Programs, 522 F.3d 657, 663 (6th Cir. 2008).

Fifth Third’s own representation preferences provide no reason to depart from local rates. First, Fifth Third had local counsel. Second, it makes no difference, contrary to the class’s insistence, that Fifth Third was represented by “outside counsel from Williams & Connolly,” a well-known national litigation firm, which, like Tycko, is based in Washington, D.C. Appellee Br. 32. Defense counsel’s rates are of “questionable relevance” in assessing whether it was reasonable to award class counsel inflated rates. Brooks v. Ga. State Bd. of Elections, 997 F.2d 857, 869 (11th Cir. 1993) (citing Johnson v. Univ. Coll. of the Univ. of Ala. in Birmingham, 706 F.2d 1205, 1208 (11th Cir. 1983)). To our minds, a defendant’s choice to retain premium national counsel reflects litigation strategy and resource allocation, not the prevailing market rate applicable to fee shifting awards. See id.; see also McClain v. Lufkin Indus., Inc., 649 F.3d 374, 384 (5th Cir. 2011) (rejecting the idea that class counsel’s hourly rates “should approximate” those charged by the defense counsel). Said differently, a defendant does not acquiesce to paying a higher fee award should it not prevail simply by hiring the best counsel it can afford.

In sum, the district court by and large acted within its discretion in awarding attorney’s fees in this complex case. It abused its discretion only in determining the rates applied to class counsel other than Tycko.

* * *

We affirm the district court’s calculation of the fee award in all respects save for the rates applied to firms other than Tycko. We thus vacate the award and remand for the district court to apply the proper forum rates.