NOT RECOMMENDED FOR PUBLICATION File Name: 25a0453n.06 Case Nos. 24-1598/1676 UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT FILED Oct 06, 2025 ) KELLY L. STEPHENS, Clerk WAYSIDE CHURCH, individually and on ) behalf of itself and all others similarly situated, ) et al., ) ON APPEAL FROM THE UNITED Plaintiffs-Appellees, ) STATES DISTRICT COURT FOR ) THE WESTERN DISTRICT OF ANN MEDEMA, et al. (24-1598), ) MICHIGAN Objector Plaintiffs - Appellants, ) ) OPINION v. ) ) VAN BUREN COUNTY, individually and on ) behalf of itself and all others similarly situated, ) ) Defendants-Appellees. ) ) TAYLOR BEIGHTOL (24-1676), ) Proposed Intervenor-Appellant, ) ) Before: KETHLEDGE, READLER, and BLOOMEKATZ, Circuit Judges. READLER, J., delivered the opinion of the court in which KETHLEDGE and BLOOMEKATZ, JJ., concurred. KETHLEDGE, J. (pp. 23–26), delivered a separate concurring opinion. READLER, Circuit Judge. For years, counties in Michigan seized tax-delinquent homes, sold them, and then pocketed or gifted to others the windfall beyond the unpaid taxes. After considerable litigation, that practice met an unceremonious fate: It was declared unlawful three times over, first by the Michigan Supreme Court, then by this Court, and most recently by the U.S. Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. Supreme Court. Rafaeli, LLC v. Oakland County, 952 N.W.2d 434, 441 (Mich. 2020); Hall v. Meisner, 51 F.4th 185, 188 (6th Cir. 2022); Tyler v. Hennepin County, 143 S. Ct. 1369, 1376 (2023). The current chapter in this legal chronicle is a class action pursued against 43 Michigan counties, all located in the Western District of Michigan. The named plaintiffs along with the putative class sought to recover surplus proceeds from the sale of their properties by those counties. Following a decade of litigation, including 18 months of mediation, the parties reached a proposed resolution. But not everyone was satisfied. Over three dozen objectors challenged the settlement, asserting that the agreement was the product of collusion, and that continued settlement negotiations or a trial on the merits would have produced more favorable terms. The district court was unpersuaded. It certified the class, overruled the objections, and approved the settlement. We now affirm. I. Wayside Church once owned a property in Van Buren County, located near Michigan’s southwestern corner. But the church fell behind on its property taxes owed to the County, leaving $16,750 unpaid. To collect, the County foreclosed on the property. Following the foreclosure, the County sold the property for $206,000, far more than the amount of Wayside’s property tax delinquency. Yet Michigan’s General Property Tax Act authorized the County to keep the balance, which it did. In 2014, Wayside sued the County, claiming that the County’s collection practices amounted to an uncompensated taking, in violation of the Fifth Amendment. The church brought the case on its own behalf and on behalf of a class of others similarly situated, naming the County as a representative of a class of similarly situated Michigan counties. 2 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. The case, it is fair to say, did not resolve quickly. The parties have endured an over decade- long path of litigation, largely because of the intervening case law that has informed the legal backdrop. Op. and Order Approving Settlement and Appointing Special Master, R. 544, PageID 12826 (district court acknowledging that the “case has been before [it] for nearly ten years”). The district court first dismissed Wayside’s complaint for failure to state a claim, reasoning that a former owner has no property interest in surplus proceeds under Michigan law. On appeal, a divided panel agreed the case should be dismissed, but on jurisdictional grounds rather than the merits, relying on the Williamson County rule that takings plaintiffs must first pursue relief in state court. Wayside Church v. Van Buren County, 847 F.3d 812, 817 (6th Cir. 2017); see Williamson Cnty. Reg’l Planning Comm’n v. Hamilton Bank of Johnson City, 473 U.S. 172 (1985). But the County’s victory was short lived, as the Supreme Court not long thereafter overruled Williamson County in Knick v. Township of Scott. 139 S. Ct. 2162, 2177 (2019) (holding that takings claims may be brought in federal court without first seeking relief in state court). Meanwhile, the district court reopened the case but stayed it in light of the Michigan Supreme Court’s grant of leave to appeal in Rafaeli, LLC v. Oakland County, 919 N.W.2d 401 (Mich. 2018) (mem.), a then-pending case before the Michigan Supreme Court that was expected to resolve a state law issue that could also inform Wayside’s federal takings claim. That prediction was prescient: Rafaeli held that former Michigan property owners who had their property seized due to a tax delinquency retain a property interest in surplus proceeds from the subsequent sale of those properties, and, further, that counties violate the Michigan constitution by retaining those proceeds. 952 N.W.2d at 441–42. 3 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. Shortly thereafter, the district court lifted the stay in this case. Continued litigation resulted in an adverse ruling against the County on their claimed sovereign immunity. That decision, in turn, led to an interlocutory appeal by the County as well as another stay in the district court. While on appeal, the case turned toward settlement talks. Beginning in April 2021, the parties participated in more than 30 mediation sessions with our Circuit’s mediation office, spanning over a year and a half. (Resolution efforts undertaken by the mediation office are confidential, we note, and thus are not disclosed to members of our Court. See 6th Cir. R. 33(b)(4)(D). Our understanding of the mediation process here thus comes from the parties and the record.) During those sessions, Wayside’s counsel and the County’s counsel invited other counties in the Western District of Michigan (most of whom were putative defendants in the case) to join the negotiations. Over time, 42 additional counties agreed to participate in the mediation. While that process was unfolding, we issued our decision in Hall v. Meisner, 51 F.4th 185 (6th Cir. 2022). There, former homeowners in Oakland County, Michigan sued the county and its officials after the county foreclosed upon the plaintiffs’ homes without refunding them the surplus above their tax debts. Hall held that a county effects a taking of private property, as that concept is understood for Fifth Amendment purposes, when it forecloses on a home due to a tax delinquency and then gifts the property to a third party who sells it for an amount that exceeds the delinquency. Id. at 188–89. In so doing, we explained that the Takings Clause secures for the former owners a recognized property interest in surplus proceeds following a tax foreclosure sale. And that interest, we added, could not be erased by Michigan’s granting an interest of absolute title in the property to counties through foreclosure. Id. at 189, 194. That decision departed from the logic of an Eighth Circuit decision issued months earlier, Tyler v. Hennepin County, 26 F.4th 789 (8th Cir. 2022), which held that former property owners have no viable Fifth Amendment 4 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. takings claim to proceeds resulting from a tax foreclosure sale when state law does not recognize a property interest in those proceeds. 26 F.4th at 793. In December 2022, two months after Hall was issued, the parties here reached a settlement, which they memorialized in an executed agreement. As part of the settlement, 42 counties— having joined the mediation process and secured formal approval from their governing boards— entered into the agreement alongside Van Buren County, with the settlement agreement creating subclasses for each of the 43 counties. That same month, petitioners in Tyler filed their reply brief with the Supreme Court in support of their petition for a writ of certiorari, urging the Supreme Court to “resolve a direct conflict between the Eighth Circuit and Sixth Circuit” in view of our recent decision in Hall. Petitioner’s Reply in Support of Petition for Writ of Certiorari at 4, Tyler v. Hennepin County, 143 S. Ct. 1369 (2023) (No. 22-166). The Supreme Court granted certiorari in Tyler on January 13, 2023. Tyler v. Hennepin County, 143 S. Ct. 644 (2023) (mem.). Two months later, in March 2023, the district court preliminarily approved the settlement agreement. In conjunction with that resolution, the district court conditionally certified a class of “[a]ll persons, their heirs and successors, who held a non-contingent interest in an Eligible Property at the time that property was foreclosed by a County and which was sold during the Class Period by that County.” Class members had 135 days to submit claims, a period that was later extended to 165. In their claim forms, former property owners were required to identify their parcels and verify their eligibility to recover under the settlement, at which point they would be entitled to receive their share of the settlement fund. Notices explaining this process were mailed to potential class members. Claims were submitted for 3,728 parcels, accounting for roughly 74% of the total surplus dollars available for distribution, with 416 putative class members (either parcel owners or other interested parties) opting out. In the end, 39 opt-outs objected. 5 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. As the claims process evolved, so too did the governing takings law. In late May 2023, the Supreme Court decided Tyler v. Hennepin County, 139 S. Ct. 1369 (2023). Resolving the circuit split in line with Hall, a unanimous Supreme Court held that a former property owner states a valid claim under the Takings Clause when the state keeps surplus proceeds from a tax-foreclosure sale. Id. at 1376. In September 2023, one day after the claims deadline closed, we allowed affected property owners to recover not just the surplus over the tax debt, but also interest on that amount. Freed v. Thomas, 81 F.4th 655, 658 (6th Cir. 2023). On February 13, 2024, the district court held a fairness hearing to discuss the proposed settlement agreement. See Fed. R. Civ. P. 23(e)(2). In connection with the hearing, the court ordered supplemental briefing and heard argument on several issues, including the retroactivity of Rafaeli, the statute of limitations, the inclusion of lienholders in the class, the adequacy of notice to the class, and attorney’s fees. It also approved the substitution of new class representatives. On June 27, 2024, the district court gave final approval to the settlement and certified the class, overruling challenges asserted by objectors. Failing on that front, objectors timely appealed. II. Objectors challenge the district court’s decision approving the class settlement, which resolved the class claims against the 43 participating counties. Before approving the settlement, the district court was required to determine both that the proposed class met the requirements for class certification, see Fed. R. Civ. P. 23(a), and that the settlement was “fair, reasonable, and adequate,” see Fed. R. Civ. P. 23(e)(2). Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620–22 (1997). Objectors challenge both aspects of the district court’s analysis, which we, in turn, review for an abuse of discretion. In re Dry Max Pampers Litig., 724 F.3d 713, 717 (6th Cir. 2013) (citing Int’l Union, United Auto., Aerospace, & Agr. Implement Workers of Am. v. Gen. Motors Corp., 6 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. 497 F.3d 615, 625 (6th Cir. 2007)). “A district court abuses its discretion when it relies on a clearly erroneous factual determination, applies the wrong legal standard, misapplies the correct one, or makes a clear error of judgment.” In re Ford Motor Co., 86 F.4th 723, 727 (6th Cir. 2023) (per curiam) (citing Lyngaas v. Curaden AG, 992 F.3d 412, 427–28 (6th Cir. 2021)). Before turning to those issues, it bears addressing how this legal framework maps on to the specific claims asserted by objectors. Many of those claims are tied to the performance of class counsel, which objectors characterize as inadequate. Some of objectors’ criticisms are presented as reasons why the class does not satisfy Rule 23(a)’s requirement that adequate counsel be retained to represent the class. Others are raised as reasons why the settlement does not satisfy Rule 23(e)(2)’s requirement that class counsel reach an adequate settlement for the class. And sometimes, objectors use both vehicles to pursue in essence the same critique. See, e.g., Medema Obj. Br. 42–44 (arguing counsel was conflicted and inadequate); id. at 49–51 (reframing the same conflict as proof the settlement was not fair). Although Rule 23(a) is typically forward-looking and Rule 23(e) retrospective, the adequacy inquiry in this settlement context is substantively the same: Both provisions in Rule 23 ask whether the class’s interests are aligned with its representatives and vigorously pursued by qualified, conflict-free counsel. See 4 William B. Rubenstein, Newberg and Rubenstein on Class Actions § 13:48 (6th ed. 2025); Int’l Union, 497 F.3d at 626; see also Fed. R. Civ. P. 23(g)(1)(A) (directing courts to weigh counsel’s work, experience, and resources in appointing class counsel). For our part, we will address the objections in the manner objectors have raised them, recognizing the overlap at times in our analysis. A. We begin with Rule 23(a)’s prerequisites for class certification. Before a district court may certify a class, the class must meet the four familiar Rule 23(a) requirements: (1) numerosity; (2) commonality; (3) typicality; and (4) adequacy of representation. Fed. R. Civ. P. 23(a). 7 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. Objectors do not contest the first three, leaving our focus on Rule 23(a)(4)’s command that the representative parties “fairly and adequately protect the interests of the class.” Whether representative parties meet this fair and adequate benchmark has two dimensions—the role of the class representatives and the role of their counsel. But in many ways, those are two sides of the same coin. Included in the fair and adequate inquiry is whether it “appears that [the class representatives] will vigorously prosecute the interests of the class through qualified counsel,” Int’l Union, 497 F.3d at 626 (citation modified), and whether those representatives are “part of the class and possess the same interest and suffer the same injury as the class members.” Id. (citation modified). Further, as the terms “fairly” and “adequately” are not separate commands but instead complementary ones, they together direct courts to ensure that the class’s interests are both aligned with its representatives and pursued by qualified, conflict-free counsel. See Vassalle v. Midland Funding LLC, 708 F.3d 747, 757 (6th Cir. 2013) (quoting In re Am. Med. Sys., Inc., 75 F.3d 1069, 1083 (6th Cir. 1996); Stout v. J.D. Byrider, 228 F.3d 709, 717 (6th Cir. 2000)). That inquiry carries particular importance where, as here, the class was certified solely for settlement. Without the crucible of litigation to test the class definition, we apply heightened scrutiny to the district court’s Rule 23(a) analysis. Amchem, 521 U.S. at 620; Dry Max, 724 F.3d at 721–22 (“These [Rule 23(a)] requirements are scrutinized more closely, not less, in cases involving a settlement class,” id. at 721.). The bulk of objectors’ arguments concern the adequacy of class counsel, so we begin there. We see no abuse of discretion in the district court’s assessment of counsel’s work. As the district court observed, counsel are experienced, recognized class action litigators. Lead counsel David Fink, for instance, has repeatedly been appointed by federal judges to leadership roles in complex cases, and has received numerous accolades from the Michigan legal community over his career. 8 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. See In re Nat’l Football League Players’ Concussion Inj. Litig., 821 F.3d 410, 429 (3d Cir. 2016) (emphasizing class counsel’s decades of experience in complex litigation as evidence of adequacy). Class counsel was also among the first to bring federal takings challenges to Michigan’s tax foreclosure scheme, efforts that helped spur a wave of parallel litigation. See Garcia v. Title Check, LLC, No. 21-1449, 2022 U.S. App. LEXIS 981, at *1 (6th Cir. Jan. 12, 2022) (mem.) (describing “deluge” of surplus-proceeds suits). These pioneering efforts matter; they show that counsel played a central role in developing the legal theory underpinning the case. And, most importantly, counsel brought this experience to bear by vigorously litigating the complex takings claims at issue for more than a decade, conducting discovery, opposing dispositive motions, and participating in over 30 mediation sessions across 18 months. Such sustained advocacy is strong evidence of adequacy. Objectors respond by pointing to episodes where, in their view, counsel faltered—namely, in communicating with represented parties after preliminary approval of the settlement, sending inadequate postcard notice, and continuing negotiations after the deaths of two representatives. None renders counsel inadequate. Regarding communications, the record reflects that counsel believed in good faith that the rules permitted them to do outreach to potential class members after preliminary approval. Once the issue was raised, counsel sought guidance and ceased the practice. Unlike the deliberate misconduct we condemned in Wayside Church v. Van Buren County, 103 F.4th 1215, 1217, 1224 (6th Cir. 2024), the conduct here was inadvertent and caused no prejudice to the class. Much of the same is true as to notice. Although the first round of postcards counsel sent to the putative class members omitted some information regarding the pending settlement, long- 9 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. form notices (which included the previously omitted details) were promptly mailed, curing any defect. Cf. Tennille v. W. Union Co., 785 F.3d 422, 440 (10th Cir. 2015) (holding that “deficiencies in sending notice via email” are not fatal in class proceedings when they were supplemented with fully sufficient mail notices). That remedial step, together with reminder mailings and notice via publication, ensured robust coverage and produced a claims rate exceeding 50%. Generally speaking, technical imperfections in notice do not render a settlement inadequate where the notice program as a whole reasonably apprises the class of the relevant developments. See Elliot v. Humana Inc., No. 22-CV-00329, 2025 WL 1065755, at *8 (W.D. Ky. Apr. 9, 2025) (explaining that missteps in counsel’s communications, even those amounting to a court-order violation, do not necessarily prejudice the class or establish inadequacy). That leaves counsel’s actions following the deaths of two of the class representatives. Counsel’s failure to promptly inform the district court and substitute new representatives was less than ideal, especially with objectors being the first to raise the issue. But we see nothing so disqualifying here as to render their representation inadequate. The case objectors rely on, Hubbard v. Plaza Bonita, LP, 630 F. App’x 681 (9th Cir. 2015) (mem.), involved far more troubling facts: concealment of a client’s death and forgery of her signature. Id. at 683. Nothing of that sort occurred here. Admittedly after some delay, counsel eventually disclosed the deaths, sought substitution, and proceeded with adequate representatives in place. Rule 23(a)(4) demands adequacy, not flawlessness. Reliable Money Order, Inc. v. McKnight Sales Co., 704 F.3d 489, 498 (7th Cir. 2013) (“Not [just] any ethical breach justifies the grave option of denying class certification.”). Relatedly, objectors say, the passing of two representatives during litigation also undermines the adjacent requirement that the class representatives be adequate. In objectors’ 10 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. assessment, the remaining class representatives were “inherently inadequate” in light of these developments. Medema Obj. Br. 32, 50. But when those events eventually came to light, the deceased representatives were promptly replaced, in advance of certification or settlement approval. Substitution of representatives is “common practice,” one that typically does not defeat adequacy. Little Caesar Enters., Inc. v. Smith, 172 F.R.D. 236, 244 n.3 (E.D. Mich. 1997) (citing Davis v. Thornburgh, 903 F.2d 212, 233 (3d Cir. 1990) (Becker, J., concurring in part and dissenting in part)). And because the substituted representatives suffered the same injury and shared the same goal of recovering surplus proceeds, the district court correctly deemed the class representatives to be adequate. B. With certification appropriate, we turn to whether the district court erred in approving the settlement agreement. Because absent class members lack a voice at the bargaining table, district courts must “carefully scrutinize” proposed class settlements. Shane Group, Inc. v. Blue Cross Blue Shield of Mich., 825 F.3d 299, 309 (6th Cir. 2016) (quoting Dry Max, 724 F.3d at 718). Under Rule 23, the class settlement must be “fair, reasonable, and adequate.” Fed. R. Civ. P. 23(e)(2). Courts typically treat those terms as a single standard, not separate inquiries. See, e.g., Dry Max, 724 F.3d at 718. To guide that inquiry, Rule 23(e)(2) directs courts to consider four “core concerns”: whether class counsel and representatives adequately represented the class; whether the proposal was negotiated at arm’s length; whether the relief is adequate; and whether the settlement treats class members equitably relative to one another. Fed. R. Civ. P. 23(e)(2). The first factor overlaps with the adequacy analysis under Rule 23(a)(4) and (g), which we have already addressed and need not repeat. See supra. So our focus here is on the agreement itself and whether the district court abused its discretion in concluding that the negotiations were fair, the relief adequate, and the distribution equitable. Dry Max, 724 F.3d at 717. 11 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. 1. The district court did not abuse its discretion in approving the settlement. Start with the fairness of the negotiations. As already explained, the district court found, and the record confirms, that the settlement resulted from an extensive adversarial process, one that encompassed more than 30 sessions with the Sixth Circuit’s mediation office. See In re Wendy’s Co. S’holder Derivative Action, 44 F.4th 527, 536–37 (6th Cir. 2022) (recognizing that a protracted, court-supervised mediation process is strong evidence of a fairly negotiated settlement). Nor did the district court commit a clear error of judgment in concluding the relief provided by the settlement was adequate. The negotiated resolution guaranteed class members 80% of their surplus proceeds and 64 cents on the dollar in actual recovery. If one were assessing the settlement with the benefit of hindsight, taking in all the background legal developments, nearly all of which aided the class, that amount might seem lower than expected. But doing so would overlook the timing of the settlement, a critical feature. When the parties executed the settlement agreement in December 2022, before preliminary approval in March 2023, neither liability nor damages was assured. On liability, it is true that, by the time of the settlement, we had held in Hall that a county’s practice of retaining or gifting surplus proceeds after a tax foreclosure sale in Michigan effected a federal taking. 51 F.4th at 196. But the Eighth Circuit had reached the opposite conclusion with respect to a similar Minnesota scheme. Tyler, 26 F.4th at 793–94. And in so doing, it disagreed with our core conclusion in Hall that the Takings Clause itself secures traditionally recognized property interests beyond those defined by state law. See id. at 792–94. Put another way, at the time of settlement, there was a circuit split on the legal question at the core of this dispute, one that was the subject of a pending certiorari petition. Circuit splits, of course, are the traditional source of cases accepted onto the Supreme Court’s certiorari docket. See Sup. Ct. R. 10(a) (explaining that one of the few “compelling” grounds upon which a certiorari grant is 12 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. appropriate is when “a United States court of appeals has entered a decision in conflict with the decision of another United States court of appeals on the same important matter”). And the certiorari petition in Tyler, it bears adding, would later be granted, putting at risk the victory secured in Hall. The settlement, on the other hand, guaranteed a sizeable and immediate recovery for members of the class, many of whom had gone years without compensation. What is more, the Supreme Court did not resolve the circuit split until May 25, 2023, two months after the settlement’s preliminary approval in March. Again, while Tyler was ultimately resolved in a way that benefitted the class here, we do not review the settlement with the wisdom gained from hindsight. See Whitlock v. FSL Mgmt., LLC, 843 F.3d 1084, 1094–95 (6th Cir. 2016). The same goes for other decisions that favored the class but were issued after the settlement was executed. For instance, we did not resolve whether former owners in this setting could recover interest until our decision in Freed, 81 F.4th at 658–59, which issued one day after the expiration of the claims deadline. The class faced even more risks—legal and intangible alike—in the absence of a settlement. Beyond the central issue of liability resolved in Tyler, some class members confronted individualized legal barriers due to the specific circumstances underlying their claims, such as res judicata, statute of limitations, or other equitable defenses the counties might have asserted. See R. 544, PageID 12866 (noting debate about the appropriate statute of limitations). And, as a practical matter, with many class members having reached their golden years, further delay raised the specter that some would never see a penny of their asserted damages during their lifetimes. See UAW v. Gen. Motors Corp., No. 05-cv-73991, 2006 WL 891151, at *17 (E.D. Mich. Mar. 31, 13 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. 2006) (“The obvious costs and uncertainty of such lengthy and complex litigation weigh in favor of settlement.”). Against this backdrop, class members faced a tradeoff: accept a guaranteed recovery of the settlement amount now or continue to litigate in the face of unresolved questions about both liability and damages. We see no abuse of discretion by the district court in honoring the class’s desire to achieve a guaranteed recovery and bring this long-running litigation to an end. Does 1– 2 v. Deja Vu Servs., Inc., 925 F.3d 886, 899 (6th Cir. 2019) (affirming district court’s approval of settlement where “the settlement put[] an end to potentially long and protracted litigation” (citation modified)); In re Motor Fuel Temperature Sales Pracs. Litig., 872 F.3d 1094, 1117 (10th Cir. 2017) (recognizing that a valid concern in reviewing a settlement is “whether the value of an immediate recovery outweighs the mere possibility of future relief after protracted and expensive litigation” (citation modified)); 1988 Tr. for Allen Child. Dated 8/8/88 v. Banner Life Ins. Co., 28 F.4th 513, 526 (4th Cir. 2022) (affirming district court’s approval of settlement where the court found that doing so now “avoids protracted litigation costs and risks to the settlement class and provides them with immediate recovery”). We do not doubt that some parties facing similar circumstances, perhaps like those who opted out of the class here, might decide to accept these risks and hold out for the chance at a larger payment. That is the nature of risk; people tolerate it differently. In re Prudential Ins. Co. Am. Sales Prac. Litig. Agent Actions, 148 F.3d 283, 317 (3d Cir. 1998) (explaining that a “settlement is a compromise, a yielding of the highest hopes in exchange for certainty and resolution” (citation modified)). Yet the district court’s mission is not to discern whether the theoretical “median” claimant would have accepted these settlement terms. Nor must a district court employ an economic model to predict the most desirable decision for the class, plugging risk facts into an 14 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. algebraic formula. In re Baby Prods. Antitrust Litig., 708 F.3d 163, 173–74 (3d Cir. 2013) (“The role of a district court is not to determine whether the settlement is the fairest possible resolution” but rather “whether the compromises reflected in the settlement—including those terms relating to the allocation of settlement funds—are fair, reasonable, and adequate when considered from the perspective of the class as a whole.”). Rather, the district court merely seeks fair assurance that a settlement falls within the “ballpark of reasonableness.” See Rubenstein, supra § 13:51; Banner Life Ins., 28 F.4th at 527 (citing “ballpark of reasonableness” test to affirm district court’s approval of settlement); In re Pharm. Indus. Average Wholesale Price Litig., 588 F.3d 24, 33 (1st Cir. 2009) (“The district court enjoys considerable range in approving or disapproving a class settlement ” (citation modified)). If so, as here, the district court does not abuse its discretion in approving the class settlement. We similarly see no abuse of discretion in the district court’s conclusion that the settlement agreement treats class members “equitably relative to each other,” a required consideration under Rule 23(e)(2)(D). Parties often raise this concern where some claimants—typically the named plaintiffs—receive preferential treatment relative to the rest of the class. See Vassalle, 708 F.3d at 755, 757. But here, the named plaintiffs and the class members were all compensated on a pro rata basis, with each receiving 80% of the surplus proceeds from the sale of their foreclosed property. So although individual recoveries differed depending on property value, every claimant obtained the same proportionate share. That structure reinforced, rather than undermined, equitable treatment. Put another way, the equal pro rata recovery ensured that all claims were treated similarly, aligning the interests of all class members. See Marshall v. Nat’l Football League, 787 F.3d 502, 519 (8th Cir. 2015) (recognizing fairness requires accounting for interests of the entire class). 15 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. 2.a. Objectors have three responses. They begin by characterizing the parties’ negotiations as, in effect, a “reverse auction.” By that, they mean the familiar class-action dynamic in which defendants facing overlapping suits pit competing counsel against each other and settle with the lawyer most willing to accept the lowest settlement offer. In objectors’ telling, the counties picked “the most ineffectual class lawyers to negotiate a settlement with, in the hope that the district court [would] approve a weak settlement that [would] preclude other claims against” the counties. County Appellees Br. 29 n.4 (quoting Grainger v. Ottawa Cnty., 90 F.4th 507, 512 n.5 (6th Cir. 2024) (quotation omitted)). According to objectors, the record here reflects telltale signs of a reverse auction, from the existence of a parallel case with stronger prospects, to Wayside’s procedural disadvantages after being denied leave to expand its claims, to secretive negotiations, all of which, they say, culminated in a cut-rate deal for the class. Objectors’ collusion theory is largely a factual issue to which we owe the district court “substantial deference” as we review the decision to approve the settlement for an abuse of discretion. See Gascho v. Glob. Fitness Holdings, LLC, 822 F.3d 269, 279 (6th Cir. 2016). And measured by that forgiving standard, no error occurred. In the district court’s words, objectors presented “more speculation than concrete evidence of collusion.” At best, the court noted, objectors could point only to “[c]ompetition among lawyers for the same class.” But competition alone is not tantamount to collusion. If it was, every overlapping class action would be ripe for the same accusation, as multiple groups of lawyers invariably vie to represent the class, a routine practice in intersecting class actions. Absent some showing that the parties manipulated that rivalry, runner up counsel’s understandable remorse is not a basis upon which to undo a class settlement. See Negrete v. Allianz Life Ins. Co. of N. Am., 523 F.3d 1091, 1094, 1099–100 (9th Cir. 2008); Rutter & Wilbanks Corp. v. Shell Oil Co., 314 F.3d 1180, 1189 (10th Cir. 2002) (noting 16 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. that “concrete evidence” is required to prove collusion). And here, objectors offered neither direct evidence (such as communications suggesting a coordinated “reverse auction”) nor indirect evidence (for instance, an unreasonably one-sided settlement or disproportionate attorney’s fee award) that might hint at collusive behavior. All in all, we see no basis to dispute the district court’s finding of “a lack of collusion.” Nor are we persuaded that the existence of a parallel and allegedly stronger case— Grainger, Jr. v. County of Ottawa, No. 1:19-cv-501, 2021 WL 790771 (W.D. Mich. Mar. 2, 2021)—provides evidence that the settling parties colluded here to avoid litigating the separate suit brought by Grainger’s counsel. Nothing about Grainger’s procedural posture suggested the counties could escape liability more easily here than there. Even before settlement talks began in this litigation, Grainger faced a major setback: The district court denied Grainger’s motion for class certification, effectively reducing the case to an individual’s claim against one county. Id. at *15–16. And when counsel tried to reboot the case by having new plaintiffs intervene, that too failed. See Motion to Intervene, Grainger, Jr. v. County of Ottawa, No. 1:19-cv-501, (W.D. Mich. Mar. 5, 2021); Order Denying Motions to Intervene, Grainger, Jr. v. County of Ottawa, No. 1:19- cv-501, (W.D. Mich. Feb. 7, 2023). Thus, contrary to objectors’ contention, Grainger was not in a comparatively better position than Wayside. That matters. Again, the premise of objectors’ collusion theory is that the counties would strategically settle the weaker action to sidestep a stronger one. But as Grainger offered no such advantage, this notion loses any force. No more availing is objectors’ emphasis on the fact that Wayside could not sue beyond Van Buren County, where its former property was located. True, the district court denied Wayside’s motion to amend its complaint to name other defendant counties. But that development did not make the case a weaker or more collusion-prone vehicle for settlement. As Wayside’s 17 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. complaint was against a defendant class, with Van Buren County as its representative, see Compl., R. 1, PageID 23–24, that posture kept all relevant counties in play in the litigation. And before settlement talks began, the district court approved Wayside’s motion to narrow that class to the counties in western Michigan. When the court later denied Wayside the opportunity to name more defendant counties, it explained that doing so was “unnecessary” because Wayside had sued a defendant class. Thus, contrary to objectors’ assertions, at no point was Wayside “barred” from suing the 42 other western Michigan counties; the lawsuit remained a vehicle for recovering against those entities. Nor do we see any sign of collusion based on the confidential mediation proceedings that resulted in the settlement. Again, the involvement of a neutral mediator (here our Circuit’s mediation office) is a strong indicator of procedural fairness, one that runs counter to any allegation of collusion. In re Wendy’s, 44 F.4th at 536 (noting that the presence of a neutral mediator supports the fairness of a settlement); see also Rubenstein, supra § 13:48 (“[T]here appears to be no better evidence of [a truly adversarial bargaining process] than the presence of a neutral third party mediator . . . .”). That those proceedings took place in private does not alter this observation. To encourage candor between the negotiating parties, our Circuit rules mandate that mediation sessions proceed in confidence. See 6th Cir. R. 33(b)(4)(D); cf. Swinton v. SquareTrade, Inc., 960 F.3d 1001, 1006 (8th Cir. 2020) (observing that the exclusion of an objector from settlement discussions did not “suggest[] collusion, mendacity, or underhanded activity”). And, it bears noting, there is no indication that the settlement talks were merely a cover for collusive activity. If the mediation process was a ruse, it was an elaborate one; between April 2021 and December 2022, the parties engaged in over 30 mediation sessions, including two in-person meetings. As we have asked in a similar setting, “why,” if the parties “were all in cahoots, did it take months of 18 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. mediation with an independent mediator to resolve the case?” In re Wendy’s, 44 F.4th at 536. At bottom, we see no evidence of collusion sufficient to disturb the district court’s approval of the settlement. b. Objectors next maintain that the settlement was inequitable because named plaintiffs with strong claims accepted the same payout as class members who held weaker or time-barred claims. Generally speaking, even when class members’ claims vary in strength and potential recovery, that fact, standing alone, does not make a settlement unfair. Int’l Union, 497 F.3d at 631 (recognizing that compromise necessarily reflects variations in claim strength); Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020–21 (9th Cir. 1998) (emphasizing that class settlements involve tradeoffs across claims of differing merit), overruled on other grounds by Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011); Reynolds v. Beneficial Nat’l Bank, 288 F.3d 277, 282 (7th Cir. 2002) (noting that the touchstone is fundamental fairness, not identical recovery). Indeed, in considering the fairness of the settlement under Rule 23, the district court was required to “take into account the interests of the entire class,” not merely some claimants. Marshall, 787 F.3d at 519. And on that score, the district court fairly considered the fact that even if the merits were “relatively strong” for some class members, others would recover far less, perhaps even nothing at all, given their relatively weaker claims, thus requiring some compromise on the final award for some class members. To be sure, Rule 23(e)(2)(D) requires that the settlement “treat[] class members equitably relative to each other.” It does so to guard against a class counsel who sells “out some of the class members at the expense of others, or for their own benefit.” Rubenstein, supra § 13:56. That worry is echoed in our case law, decisions that laid the groundwork for Rule 23(e)(2)(D). Id.; see Fed. R. Civ. P. 23(e)(2) advisory committee’s note to 2018 amendment (explaining that the 19 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. amendment codified in Rule 23(e)(2) consists of a list of four “core concerns” drawn from pre- existing circuit precedent). For instance, in Vassalle, a decision issued before Rule 23(e)(2)(D)’s enactment, we held that a settlement is suspect when the named plaintiffs receive a special benefit not available to absent class members, creating a misalignment of interests and the risk of collusion with defense counsel. 708 F.3d at 755, 757. But the opposite is true here: The named plaintiffs accepted the same pro rata distribution as everyone else, even though their claims were stronger than some. By doing so, they ensured that absent class members were included in the recovery, aligning their interests with the class. That structure reinforced, rather than undermined, the settlement’s fairness. In cases where there exists a disparity in recovery between class members due to the specifics of the claims at issue, that concern is perhaps best addressed under Rule 23(a), which requires commonality across the class, and Rule 23(b)(3), which requires that common questions of law or fact for the class “predominate over any other questions affecting only individual members.” As to Rule 23(b)(3) in particular, wide variation in the damages recoverable by class members can justify denying certification because individualized damages inquiries may overwhelm common issues. See Comcast Corp. v. Behrend, 569 U.S. 27, 34–38 (2013); Speerly v. Gen. Motors, LLC, 143 F.4th 306, 321–22 (6th Cir. 2025) (en banc) (explaining that differences across states in the damages elements and remedies for warranty and consumer protection claims meant individualized issues predominated over common ones). But the district court here certified the settlement class without any objections under Rule 23(b), and objectors did not challenge commonality or predominance on appeal. c. Finally, objectors complain that lienholders were included in the settlement class, making the settlement unfair. Objectors hinge their argument on the fact that the settlement 20 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. agreement defined the class to include those with a “non-contingent” interest in property subject to foreclosure. But the agreement itself listed “any type of lien” as an example of such an interest. Proposed Long Form Class Notice, R. 220-3, PageID 3642. That makes sense. Mortgagees, for instance, hold rights that are impaired when the county forecloses. Once a debtor signs a mortgage, “the creditors immediately obtain[] a host of present rights in the [property],” including foreclosure. In re Glance, 487 F.3d 317, 322–23 (6th Cir. 2007). All told, we see nothing unfair about settlement funds going to lienholders. d. One issue remains: the award of attorney’s fees. Although objectors have not challenged the award as too high (in fact, they contend that it is too low, making it evidence of collusion), we still must “carefully scrutinize” whether class counsel satisfied its fiduciary obligations to class members. Dry Max, 724 F.3d at 718. Here too, we review the approval of the fee award for an abuse of discretion. Imwalle v. Reliance Med. Prods., Inc., 515 F.3d 531, 551 (6th Cir. 2008). That exercise of discretion, however, is entitled to “substantial deference because the rationale for the award is predominately fact-driven.” Id. We do not think class counsel’s fees undermine the settlement’s fairness. As today’s case is a § 1983 action, one in which fees would ordinarily be shifted to defendants under 42 U.S.C. § 1988(b), the settlement’s structure, which requires fees to come from the class’s recovery, merits scrutiny. See 42 U.S.C. § 1988(b). The district court evaluated the fees request under both the percentage-of-fund and lodestar methods, finding counsel’s $6.85 million lodestar (i.e., hours reasonably worked multiplied by reasonable rates) fairly supported a $7.8 million award. That $7.8 million award reflects a modest enhancement—a 1.14 multiplier on counsel’s lodestar—to account for the risk of nonpayment and the quality of results achieved. Multipliers of this size are within the range we have previously upheld. See Van Horn v. Nationwide Prop. & Cas. Ins. Co., 21 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. 436 F. App’x 496, 499–500 (6th Cir. 2011) (upholding approval of a 1.2 multiplier). Fees of this magnitude, reached after extensive adversarial mediation, are not disproportionate to the class relief achieved and did not divert value from absent members. Cf. Dry Max, 724 F.3d at 721–22. The district court thus acted well within its discretion in concluding the fee award was reasonable. * * * * * We affirm. 22 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. KETHLEDGE, Circuit Judge, concurring. More than eight years ago, in this same case, I wrote that “the defendant Van Buren County took property worth $206,000 to satisfy a $16,750 debt, and then refused to refund any of the difference. In some legal precincts that sort of behavior is called theft.” Wayside Church v. Van Buren County, 847 F.3d 812, 823 (6th Cir. 2017) (dissenting opinion). Since then our court and the Supreme Court have both held, unanimously, that this same conduct—taking property worth more than a tax debt, and refusing to refund the difference—amounts to an unlawful taking of property in violation of the federal Constitution. And that is what the defendant counties did to every one of the class members here. On the constitutional merits, therefore, these class members are entitled to get back every penny the counties unlawfully took from them. Yet one would hardly surmise that entitlement from the terms of the settlement agreement before us here. The counties get to keep 20 cents of every dollar they unconstitutionally took from these plaintiffs—meaning they get to keep at least $13 million, and likely more. Meanwhile, in § 1983 cases where (as here) a state entity violates a plaintiff’s constitutional rights, the plaintiff can usually recover her attorneys’ fees. See 42 U.S.C. § 1988. But under this agreement the class members themselves must pay their class counsel’s fees—which could total $10 million before the case is done. Moreover, when the government takes property, its “obligation is to put the owners in as good position pecuniarily as if the use of their property had not been taken.” Phelps v. U.S., 274 U.S. 341, 344 (1927). And that means—where compensation comes after the taking itself—the government must pay prejudgment interest to a successful plaintiff in a takings case. Id.; see also, e.g., Kirby Forest Industries, Inc. v. U.S., 467 U.S. 1, 10 (1984) (prejudgment interest is part of “just compensation” for a taking); Albrecht v. U.S., 329 U.S. 599, 602 (1947) (same); Schneider v. County of San Diego, 285 F.3d 784, 789 (9th Cir. 2002) (“An award of prejudgment 23 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. interest as compensation for a taking, however, takes on a constitutional dimension.”); Freed v. Thomas, 81 F.4th 655, 658 (6th Cir. 2023). But under this settlement the class members get zero prejudgment interest—even though the counties have held some $65 million of the class members’ money for years now (in the case of Wayside Church, for more than 11 years). The upshot is that these class members get only 64 cents of every dollar the counties unlawfully took from them years ago—with no compensation for inflation in the meantime. To make that gap concrete, compare Wayside Church’s recovery under this settlement with the judgment it likely would have received as an individual plaintiff in a § 1983 suit. Again, to satisfy a tax debt of $16,750, Van Buren sold Wayside’s property in August 2014 for $206,000. Wayside is constitutionally entitled to the full surplus of $189,250. Wayside would also receive about $105,000 in prejudgment interest, assuming a rate of 5%—representing the value that the county obtained (and Wayside lost) by holding that $189,250 for some 11 years now. And the county, rather than Wayside itself, would pay Wayside’s attorneys’ fees. (Surely no district court would exercise its discretion to the contrary.) So, in an individual suit, Wayside would receive almost $300,000, and the county would pay its attorneys’ fees. But in this settlement Wayside gets only $121,120—with the county retaining most of the difference. In this appeal, however, our standard of review is deferential; and circumstances have painted us into a corner here. The reality is that—by an almost tragic fortuity—federal and state law alike conspired to prevent these class members (and everyone similarly situated to them in Michigan) from asserting their federal constitutional rights in response to these takings. On the federal side, the Supreme Court’s decision in Williamson County Reg’l Planning Comm’n v. Hamilton Bank of Johnson City denied these class members a federal forum for their (federal) takings claims. 473 U.S. 172, 194 (1985); see also Wayside Church, 847 F.3d at 822-23. And the 24 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. ability of these class members to obtain relief in Michigan courts was far from clear. See id. at 823-24 (dissenting opinion). Indeed, so far as I can tell, nobody in these cases prevailed on a federal takings claim in Michigan state court before 2020. See generally Rafaeli LLC v. Oakland County, 952 N.W.2d 434 (Mich. 2020). But in Knick v. Township of Scott, 588 U.S. 180, 194 (2019), the Supreme Court overruled Williamson County—which finally cleared the way for the claims before us now. Fairness is measured at a point in time. By the time the Supreme Court overruled Williamson County, the claims of many Michigan residents whose property had been unlawfully taken were time-barred or subject to an adverse judgment in state court. See, e.g., Hall v. Meisner, 2022 WL 7478163 at 1-2 (6th Cir. 2022) (affirming the dismissal of such claims). Meanwhile, in litigating these claims, the counties have exploited all these difficulties to the utmost. And these class members—many of whom, surely, are lower-income or elderly—can wait only so long for relief. For many of them the economic harm inflicted by these takings was catastrophic. Blowing up this settlement now, and making these class members wait many months or even years longer for relief—with the counties holding the money they unlawfully took from these people all the while—would inflict significant harm of its own. Personally I would have welcomed a rejection of this settlement, ideally some time ago. But as this case comes to us now, I conclude with the greatest reluctance that the district court’s approval of the settlement falls within the latitude afforded it by our standard of review. That said, the precedential import of this settlement should be close to zero. The baleful effects of Williamson County make the circumstances of this settlement sui generis. For years the counties acted with near impunity because an almost offhand, alternative holding of the Supreme Court (in Williamson County) denied the victims of these takings a federal forum for their federal 25 Nos. 24-1598/1676, Wayside Church et al. v. Van Buren County et al. constitutional claims. Absent a similar bar to asserting federal claims in federal courts, this case should be easy to distinguish from future ones in which class members have meritorious claims. In closing, what is most remarkable, in the litigation of these claims over the past eight years, is the counties’ complete lack of remorse. Their victims for the most part were their own residents; and what the counties forcibly and unlawfully took from them sometimes amounted to their life savings. The facts of Wayside’s case and many others truly shock the conscience; and yet, apparently, none of these county officials ever asked whether they should keep, for example, $206,000 in property to pay off a resident’s $16,750 tax debt. Instead, the counties kept all this property “simply because the Michigan General Property Tax Act said [they] could.” Hall v. Meisner, 51 F.4th 185, 194 (6th Cir. 2022). After the Supreme Court (and our court) told these counties that their actions violated the federal Constitution, every one of them could have simply returned the property they took; the counties can bear the financial burdens of their own transgressions more easily than these unfortunate individuals can bear them. But instead the counties have employed every available legal artifice to keep as much of that money as they possibly can. As a technical, legal matter, that was their right; and yet one can be disappointed. Local governments should serve their people, not prey upon them. 26