No. 24-5421

On Petition for Rehearing En Banc

United States District Court for the Western District of Tennessee at Memphis. No. 2:20-cv-02482 — Thomas L. Parker, District Judge.

Argued En Banc: March 18, 2026

Decided and Filed: April 24, 2026

Before: SUTTON, Chief Judge; MOORE, CLAY, GIBBONS, GRIFFIN, KETHLEDGE, THAPAR, BUSH, LARSEN, NALBANDIAN, READLER, MURPHY, DAVIS, MATHIS, BLOOMEKATZ, RITZ, and HERMANDORFER, Circuit Judges.

ARGUED EN BANC: Theodore J. Boutrous Jr., GIBSON, DUNN & CRUTCHER LLP, Los Angeles, California, for Appellant. Jacob L. Phillips, JACOBSON PHILLIPS PLLC, Winter Park, Florida, for Appellee. ON BRIEF: Theodore J. Boutrous Jr., Bradley J. Hamburger, Daniel R. Adler, Matt Aidan Getz, GIBSON, DUNN & CRUTCHER LLP, Los Angeles, California, Jeffrey B. Wall, Judson O. Littleton, SULLIVAN & CROMWELL LLP, Washington, D.C., Christopher L. Vescovo, LEWIS THOMASON, Memphis, Tennessee, Peter W. Herzog III, WHEELER TRIGG O'DONNELL LLP , St. Louis, Missouri, Eric L. Robertson, WHEELER TRIGG O’DONNELL LLP, Denver, Colorado, for Appellant. Jacob L. Phillips, JACOBSON PHILLIPS PLLC, Winter Park, Florida, Hank Bates, Lee Lowther, CARNEY BATES & PULLIAM, PLLC, Little Rock, Arkansas for Appellee. Adam G. Unikowsky, JENNER & BLOCK LLP, Washington, D.C., for Amici Curiae.

Page 2MURPHY, J., delivered the opinion of the court in which SUTTON, C.J., and GRIFFIN, KETHLEDGE, THAPAR, BUSH, LARSEN, NALBANDIAN, READLER, and HERMANDORFER, JJ., concurred. BUSH, J. (pp. 22 – 31), delivered a separate concurring opinion. GIBBONS, J. (pp. 31 – 46), delivered a separate dissenting opinion in which MOORE, CLAY, DAVIS, MATHIS, BLOOMEKATZ, and RITZ, JJ., concurred.

MURPHY, Circuit Judge. Class-action suits brought against automobile insurers in many courts from around the country have asked the same question. Suppose an insurer promises to pay the “actual cash value” of an insured’s vehicle if the vehicle gets dest royed in an accident. Suppose further that the insurer uses the same formula to calculate actual cash value. If car owners believe that this general formula includes an improper reduction, may they pursue a class action? Five circuit courts have now sai d “no” because individual issues about the unique value of each used car will dominate all other matters. See Ambrosio v. Progressive Preferred Ins. Co. , 154 F.4th 1107, 1110 – 13 (9th Cir. 2025); Freeman v. Progressive Direct Ins. Co. , 149 F.4th 461, 468 – 71 (4th Cir. 2025); Schroeder v. Progressive Paloverde Ins. Co. , 146 F.4th 567, 576 – 78 (7th Cir. 2025); Drummond v. Progressive Specialty Ins. Co. , 142 F.4th 149, 158 – 61 (3d Cir. 2025); Sampson v. United Servs. Auto. Ass’n , 83 F.4th 414, 417, 421 – 23 (5th Cir. 2023).

This case asks the same question. When calculating the “actual cash value” of destroyed vehicles, State Farm Mutual Automobile Insurance Company often relies on the advertised prices of comparable used vehicles. It then imposes a “typical negotiation” adjustment that reduces the estimated value of these comparators to account for negotiations that lower their final sales prices. Jessica Clippinger brought a class-action challenge to this typical-negotiation adjustment, claiming that it generally undervalues the comparator vehicles. Even if Clippinger were correct, though, we agree with the other circuit courts that she cannot pursue this theory on a class-wide basis. To determine whether State Farm paid “actual cash value” for the 90,000 used vehicles i n the class, a jury would have to consider unique evidence about each vehicle's value. And this individual valuation will 'predominate' over all other questions under Federal Rule of Civil Procedure 23(b)(3). To be sure, the district court tried to avoid the need for case-by-case evidence by holding Page 3that it could estimate every class member’s damages using a simple calculation that would refund the amount of the typical-negotiation adjustment. But this proposed solution would wrongly read Rule 23 to eliminate State Farm’s “substantive right” to prese nt unique evidence that it paid fair market value to a specific class member despite its use of the adjustment. Wal-Mart Stores, Inc. v. Dukes , 564 U.S. 338, 367 (2011) (quoting 28 U.S.C. § 2072(b)). We thus reverse the district court’s class -certification order and remand for proceedings consistent with this opinion.

State Farm uses a standard-form insurance policy to insure the vehicles of its Tennessee customers. The policy allows State Farm to pay for damaged cars in different ways. As relevant now, State Farm may “[p]ay the actual cash value of the covered vehicl e minus any applicable deductible.” Policy, R.146 -2, PageID 4018. Yet State Farm's policy does not define the key phrase: “actual cash value.” Instead, it identifies the process that State Farm and an insured should use to calculate this value. Id. , PageID 4018 – 19. The policy first proposes that the parties negotiate to an agreement: “The owner of the covered vehicle and we must agree upon the actual cash value of the covered vehicle.” Id. , PageID 4018. If they cannot agree, the policy allows either side to seek an appraisal: “If there is disagreement as to the actual cash value of the covered vehicle, then the disagreement will be resolved by appraisal upon written request of the owner or us[.]” Id. The policy goes on to identify the “procedures” f or this appraisal. Id. Each party gets to pick one appraiser, and the chosen appraisers pick a third (or a court selects one if the chosen appraisers cannot agree). Id. These appraisers have the power to do just one thing: “determine the actual cash value” of the totaled car. Id. , PageID 4019. If two of them agree on this number in a “written appraisal” that includes an “explanation,” that number binds both sides. Id. Car owners and State Farm must bear the expenses for their chosen appraiser and then split the cost of the third appraiser. Id. , PageID 4018.

Separately, a Tennessee regulation identifies the methods that insurers may use to calculate the “actual cash value” of totaled vehicles. Tenn. Comp. R. & Regs. § 0780-01-05-.09(1). If an insurer pays a “cash settlement,” it may base that settlement on the “actual cost” to buy a vehicle “comparable” to the totaled one. Id. § 0780-01-05-.09(1)(b). The regulation identifies optional ways to determine this cost. Id. The insurer may: (1) look to the “cost of two or more comparable Page 4automobiles in the local market'; (2) look to the “cost of two . . . or more comparable automobiles” in nearby markets; (3) ask two or more “licensed dealers” in the local market for “quotations” if the first two methods are unavailable; or (4) use a gener al “source for determining statistically valid fair market values[.]' Id. If an insurer takes the fourth path, the general source must (among other things) “give primary consideration to the values of vehicles in the local market” and contain a “database” that can estimate the value of 85% of all vehicles on the market for the last 15 years. Id. Alternatively, an insurer may “deviate[]” from these identified methods if the insurer both supports the “deviation” with “documentation giving particulars of the automobile condition” and “fully explain[s]” its approach to the policyholder. Id. § 0780-01-05-.09(1)(c).

During the time applicable to this suit, State Farm followed the fourth approach (use of a general source) to start negotiations with Tennessee customers over the actual cash value of their destroyed vehicles. It relied on a database of millions of advertised or recently sold vehicles compiled by a company named Audatex. When a customer called State Farm to report an accident, a State Farm estimator would inspect the damaged vehicle and collect information about its preaccident condition. After the inspection, the estimator would upload this information into Audatex’s database.

This database would generate an “Autosource Report” that proposed a fair market value for the damaged car. To create this report, Audatex generally used a 'comparable vehicle methodology” that compared the destroyed vehicle to the most similar vehicles fo r sale. Graff Decl., R.1461, PageID 3990. If the database identified the 'advertised prices' for these comparable vehicles, the report would include a “typical negotiation” adjustment to those prices. Id. , PageID 3990 – 91. As the theory for this adjust ment, Audatex assumed that “most people” negotiate with used-car dealers to get a reduced sales price as compared to the advertised one. Id. , PageID 3990. An expert suggested that Audatex would not use this adjustment if the database had the actual sales prices of the comparable vehicles or if a 'no -haggle' dealer advertised them. Lowell Decl., R.156, PageID 4950. That said, some evidence suggests that Audatex sometimes applied this adjustment even to no-haggle dealers. See Merritt Rep., R.184, PageID 5975. In all events, claims handlers would next review the Autosource Report for “accuracy,” make any other necessary adjustments, and approve the final report. Graff Decl., R.146-1, PageID 3991. The Page 5claims handlers would then talk with a customer about the estimated value and answer questions about this report. If the parties reached an agreement on the estimated value of the totaled vehicle, claims handlers would document that agreement in the claims file.

State Farm's interaction with Jessica Clippinger (the named plaintiff) shows how this process worked. (Clippinger has changed her name to Jessica Pyron, but we will refer to her by her prior name for consistency's sake.) In May 2019, Clippinger got into an accident that destroyed her 2017 Dodge minivan. So she made a claim with State Farm, her insurer. State Farm decided that her minivan was a total loss and obtained an Autosource Report for it. This report identified the advertised prices (ranging from $15,800 to $18,803) for four comparable minivans in Tennessee. The report used the typical-negotiation adjustment to reduce those four prices by between $790 and $940. It next reduced the four prices even further to account for differences in mileage a nd features between the comparable vans and Clippinger's van. The report lastly averaged these reduced prices together to identify the actual cash value for Clippinger's van: $14,490. State Farm and Clippinger agreed on this amount. State Farm thus paid that amount to Clippinger and her lender.

Within the next year, though, Clippinger had second thoughts about one part of State Farm’s valuation process: the typical -negotiation adjustment. Clippinger thus brought a putative class-action lawsuit against State Farm. She alleged that its use of this adjustment to reduce the advertised prices for comparable vehicles to fictional sales prices breached the insurance policy and the implied duty of good faith and fair dealing. According to Clippinger, the modern used-car market operates mostly on internet platforms where parties can comparison shop. She argues that this increased competition has forced dealers to advertise in a “transparent” way that leaves no room for negotiation. Appellee’s Br. 3. Clippinger adds that State Farm used biased data tha t overestimated the number of used cars selling below their advertised price. Apart from this adjustment, though, Clippinger agreed that State Farm conducted a “detailed, sound, and reliable appraisal” process. Id. at 13. In other words, Clippinger asserted that State Farm undervalued its customers’ cars by the exact amount of the typical -negotiation adjustment.

Once Clippinger sued, State Farm invoked the insurance policy’s appraisal process. The district court held that the policy required Clippinger to use this process. Each party obtained an Page 6appraiser, and the court appointed a third. The three appraisers did not initially agree on the valuation: State Farm's appraiser valued Clippinger's van at $14,432; Clippinger's appraiser valued it at $17,756.69, and the third appraiser valued it at $18, 476.13. But Clippinger’s appraiser eventually came around to the third appraiser's number, so the two issued a written decision valuing the van at $18,476. This number bound both sides under the policy’s terms. State Farm thus paid Clippinger over $4,000 more (the difference between this higher valuation and the original one along with the additional taxes).

State Farm sought to end the lawsuit after it paid Clippinger the higher amount. It argued that the appraisal had authoritatively decided the actual cash value of Clippinger’s van. And it claimed that its decision to pay the higher amount eliminated Clip pinger’s injury. But the district court disagreed with this mootness claim. It concluded that the typical-negotiation adjustment forced customers to either accept less than the actual cash value or bear the costs of the appraisal process. See , 630 F. Supp. 3d 947, 956 (W.D. Tenn. 2022). It added that a reasonable jury could find that the claimed breach injured Clippinger because the initial undervaluation required her to incur appraisal costs before obtaining her minivan’s actual cash value. Id. at 957.

Clippinger next moved to certify a class against State Farm. The district court granted her motion. See , 2023 WL 7213796, at *15 – 16 (W.D. Tenn. Aug. 25, 2023). The court first determined that Clippinger’s contract claims were “typical” of the class, which made her an 'adequate' class representative. Id. at *9 – 10. It also accepted Clippinger’s “common” question: Did the typical -negotiation adjustment “artificially reduce[]” State Farm’s actual -cash-value estimate because the adjustment did not “accurately reflect[] how cars are valued and sold in the” used -car market? Id. at *11 (citation omitted). And the court rejected State Farm’s claim that individual issues would predominate over this common question. State Farm had argued that it could show that it provided the “actual cash value” for every class member’s vehicle u sing vehicle-specific evidence. Id. at *12. But the court accepted Clippinger’s competing view that she could identify each class member’s inju ry using a simple mathematical calculation. See id. at *12, *14. According to Clippinger, if she convinced a jury that buyers rarely negotiate for reduced prices, the court could determine each class member’s damages by recalculating the Autosource Report without the typical-negoPage 7tiation adjustment. See id. The court agreed. See id. It thus certified a class of State Farm customers whose payment for a totaled vehicle had rested on an Autosource Report that “decreased” the estimated “actual cash value” “based upon typical negotiation adjustments” for comparable vehicles. Id. at *16.

A divided panel affirmed. See , 156 F.4th 724, 746 (6th Cir. 2025). The en banc court then vacated the panel decision and agreed to rehear this case. See , 165 F.4th 1028, 1029 (6th Cir. 2026) (en banc) (order).

Federal Rule of Civil Procedure 23 sets the ground rules to certify a class action when the plaintiffs seek damages for the class. The plaintiffs must first satisfy Rule 23(a)’s numerosity, commonality, typicality, and adequacy requirements. See Wal-Mart , 564 U.S. at 349; Speerly v. Gen. Motors, LLC , 143 F.4th 306, 316 (6th Cir. 2025) (en banc). They must then satisfy Rule 23(b)(3)’s predominance and superiority requirements. See Speerly , 143 F.4th at 316. Plaintiffs have a duty to “prove” with evidenc e that the proposed class meets these elements. Id. (quoting Wal-Mart , 564 U.S. at 350). And the district court has a duty to conduct a “rigorous analysis” to confirm that they have done so. Wal-Mart , 564 U.S. at 350 – 51 (quoting Gen. Tel. Co. of the Sw. v. Falcon , 457 U.S. 147, 161 (1982)). The court must “forecast how the parties will conduct the litigation”; it cannot certify a class first and ask questions about how the litigation will proceed later. Fox v. Saginaw County , 67 F.4th 284, 302 (6th Cir. 2023).

We review a district court’s order certifying a class for an abuse of discretion. See In re Nissan N. Am., Inc. Litig. , 122 F.4th 239, 245 – 46 (6th Cir. 2024). The court commits such an error when it makes a legal mistake. See Speerly , 143 F.4th at 315. State Farm argues that the district court committed several of these mistakes here. At the outset, the insurer asserts that Article III of the Constitution prohibited this class action because the class includes “untold” customers who suffered no injury from the typicalnegotiation adjustment. Appellant’s Supp. Br. 24; compare Speerly , 143 F.4th at 345 (Thapar, J., concurring), with id. at 347 (Nalbandian, J., concurring). The insurer also argues that Clippinger failed to prove Rule 23(a)'s typicality and adequacy elements because her injury (appraisal costs) does not match the class members’ injuries (alleged Page 8receipt of less than actual cash value). And State Farm argues that the class fails Rule 23(b)(3)’s superiority requirement because the appraisal process will resolve the class members’ many claims more efficiently than this large suit. Lastly, State Far m argues that the class fails Rule 23(b)(3)’s predominance requirement because individual issues will overwhelm any common ones.

We need not resolve State Farm’s first three arguments because its fourth argument (that individual issues predominate) alone shows that this suit cannot proceed as a class action. See Fox , 67 F.4th at 296. State Farm’s predominance argument requires us first to identify the potential common questions that this suit presents before we can weigh those questions against the individual ones that the court must decide on a class-member-by-class-member basis. See WalMart , 564 U.S. at 359; Fox , 67 F.4th at 300. We thus start with Rule 23(a)’s commonality element.

Plaintiffs must show that a class action contains at least one question “common to the class” to satisfy Rule 23(a). Fed. R. Civ. P. 23(a)(2); see Wal-Mart , 564 U.S. at 349, 359. But not just any question will do. Rather, the question must clear two hurdles. It first must yield a cohesive yes-or-no answer for all class members. See Wal-Mart , 564 U.S. at 350. If a jury could resolve the question differently for different class members —“‘yes’ for some” and “‘no’ for others”— it lacks a “common” answer. Nissan , 122 F.4th at 246 – 47. The question also must yield an answer “ central ” to the class’s common claims. Wal-Mart , 564 U.S. at 350 (emphasis added); see Speerly , 143 F.4th at 316 – 17, 319. In other words, the question must matter to at least one element of those claims. See Speerly , 143 F.4th at 316 – 17, 319. We have yet to decide whether the question must conclusively resolve (rather than just “affect”) an element. Nissan , 122 F.4th at 246 (quoting Doster v. Kendall , 54 F.4th 398, 430 (6th Cir. 2022), vacated as moot , 144 S. Ct. 481 (2023)). But there is no doubt that a question on a peripheral issue (such as whether all the class members got into an accident) will not suffice. See Wal-Mart , 564 U.S. at 349 – 50; Nissan , 122 F.4th at 248. Commonality thus requires courts to identify the elements of the class’s claims that the purported common question will affect. See Speerly , 143 F.4th at 316 – 17.

Here, Clippinger’s complaint raises two claims on behalf of the class under Tennessee law: breach of contract and breach of the implied duty of good faith and fair dealing. But Tennessee Page 9courts do not recognize breach of the implied duty of good faith and fair dealing as a standalone cause of action apart from breach of contract. See Cadence Bank, N.A. v. The Alpha Tr. , 473 S.W.3d 756, 773 (Tenn. Ct. App. 2015); see also Shah v. Racetrac Petroleum Co. , 338 F.3d 557, 572 (6th Cir. 2003). So the class must prove the three breach-of-contract elements: that the parties entered an enforceable contract; that State Farm’s conduct breached this contract; and that the breach caused damages. See Fed. Ins. Co. v. Winters , 354 S.W.3d 287, 291 (Tenn. 2011).

Do the class’s contract claims include any common questions? At the outset, State Farm has not disputed that the first element could generate a common answer for the class: Did State Farm enter an enforceable contract to pay the 'actual cash value' to cla ss members when an accident destroyed their vehicles? State Farm concedes that it uses a standard-form insurance policy for all its Tennessee customers. So a jury likely must answer this question for each class member in the same yes-or-no way. See Nissan , 122 F.4th at 247. And one could reasonably view the question as “central” to the breach -of-contract claim. Wal-Mart , 564 U.S. at 350. Without a contract, State Farm had nothing to breach. See Fed. Ins. Co. , 354 S.W.3d at 291. True, State Farm likely would not dispute that a uniform contract existed. Yet we have suggested that even undisputed questions can satisfy the commonality element if they meet the cohesion and centrality requirements. See Hicks v. State Farm Fire & Cas. Co. , 965 F.3d 452, 458 – 59 (6th Cir. 2020). And again, a “single” common question can suffice to satisfy Rule 23(a)(2). Wal-Mart , 564 U.S. at 359 (citation omitted); Speerly , 143 F.4th at 319. Without briefing on this issue, though, we will simply assume that the contract-formation question satisfied the commonality element.

That said, the district court did not even identify the contract-formation question as a common one. It instead identified the 'central question' as 'whether the [typical -negotiation adjustment] accurately reflects how cars are valued and sold in the mark et.” Clippinger , 2023 WL 7213796, at *11. It thus asked about the general state of the usedcar market: Does a used car’s advertised price typically match its sales price, such that State Farm's typical -negotiation adjustment to advertised prices systematically understates used-car values? See Schroeder , 146 F.4th at 576; Drummond , 142 F.4th at 153, 157. To evaluate whether this question satisfies the commonality requirement, we will consider the cohesion and centrality elements in turn.

Page 10Cohesion Element . Will this marketcondition question yield a “yes” or “no” answer for the entire class? See Wal-Mart , 564 U.S. at 350; Nissan , 122 F.4th at 247. The question likely has the required cohesion because it asks about market averages : Do used-car buyers usually negotiate discounts on advertised prices in the current market? And, if so, what is the average size of that discount? The answer thus seemingly will remain the same for the whole class. After all, that is how averages work — they represent the mean or median value of an entire data set, not all the data set’s individual inputs. Clippinger and State Farm thus can use “common evidence” like “expert testimony” or “empirical data” to answer this question. Schroeder , 146 F.4th at 576.

True, the parties disagree on the proper answer. State Farm says that the use of advertised prices of comparable vehicles alone would “artificially inflate” the fair market values of totaled vehicles because used cars usually sell for less than their sticker price. Lowell Decl., R.156, PageID 4949 – 50. Audatex thus reduces the advertised price by a regularly adjusted percentage to account for this negotiated discount. Id. , PageID 4950, 4955, 4957. Clippinger responds that internet comparison shopping has eliminated the tendency to negotiate prices. Felix Rep., R.13914, PageID 3073 –75. As she sees things, then, Audatex’s percentage reduction to advertised prices artificially deflates the fair market values of totaled vehicles. Id. ; Merritt Rep., R.140-14, PageID 3817. Either way, though, this disagreement about the state of the used-car market does not change the yes-or-no nature of the question. No matter who has the better answer, the average negotiation reduction will stay the same for all class members.

Centrality Element . Will the answer to this marketcondition question 'drive the resolution” of the class’s common breach -of-contract claims? Wal-Mart , 564 U.S. at 350 (citation omitted). The district court answered 'yes' by pointing to the breach element of these claims. See Clippinger , 2023 WL 7213796, at *8, *11. It reasoned that the answer was central to the claims because it would categorically resolve whether the typical-negotiation adjustment breached State Farm’s standard -form insurance policy for all class members. See id. at *8, *11 – 12.

The court was mistaken. To decide whether the market-condition question will automatically prove a “breach” of the policy for every class member, we must consider the policy’s terms. See Ambrosio , 154 F.4th at 1111; Schroeder , 146 F.4th at 576. While the commonality analysis thus requires us to delve into the “merits” of the class’s contract claims, that overlap with Page 11the merits “cannot be helped” in this case. Wal-Mart , 564 U.S. at 351. We may (indeed, must) evaluate the merits when necessary to decide 'whether the Rule 23 prerequisites for class certification are satisfied.' Amgen Inc. v. Conn. Ret. Plans & Tr. Funds , 568 U.S. 455, 466 (2013). To establish commonality, Clippinger must 'affirmatively' prove that the market -condition question “is central to” the class’s claims. Wal-Mart , 564 U.S. at 350. And one cannot evaluate the question’s importance without ana lyzing those claims. See id. at 351 – 52.

Here, the merits of a contract dispute must start with the contract. The insurance policy shows that State Farm promised just one thing: to pay each class member the “actual cash value” of the class member’s totaled vehicle. Policy, R.146 -2, PageID 4018; cf. Freeman , 149 F.4th at 468; Schroeder , 146 F.4th at 576. Tennessee law, in turn, equates actual cash value with fair market value. Tenn. Comp. R. & Regs. § 0780-01-05-.09(1)(b) ¶ 4. In other words, a car’s actual cash value represents 'the price a willing buyer would pay a willing seller' in a voluntary exchange. Lammert v. Auto-Owners (Mut.) Ins. Co. , 572 S.W.3d 170, 174 (Tenn. 2019) (citation omitted). Yet this price can be hard to calculate. Given the subjective nature of the fair-marketvalue inquiry, the policy offers two ways for State Farm and its customers to identify the fair market value. The parties can either negotiate to an agreement or submit to an appraisal. Policy, R.146-2, PageID 4018 – 19. Critically, though, no language in the policy prevented State Farm from using a typical-negotiation adjustment when calculating an amount to start the negotiations over the fair market value of the customer’s totaled car.

Clippinger's market -condition question thus has a centrality problem. Her complaint alleges that State Farm breached the policy by failing “to pay the actual cash value of total loss vehicles” as the policy required. Compl., R.1 -1, PageID 23 – 24. But a j ury’s answer to the question whether advertised prices match sales prices as a general matter will not “resolve” whether State Farm breached its promise to pay actual cash value to any specific class member. Schroeder , 146 F.4th at 576; cf. Amgen , 568 U.S. at 467 – 68. Consider the two answers that the jury might give to this question at trial. If the jury agreed with Clippinger that the adjustment did not accurately reflect the used-car market, it would not establish that State Farm failed to pay actual cash value to any specific class member. Rather, State Farm might still assert that it did pay actual cash value to individual class members despite the invalidity of the adjustment as a general matter. And if Page 12the jury agreed with State Farm that the adjustment did approximate the conditions of the used-car market, it would not “end the case for one and for all” (or render all claims “dead on arrival”). Amgen , 568 U.S. at 468, 474 (citation omitted). Rather, individual class members might still assert that they did not receive actual cash value despite the validity of the adjustment as a general matter.

The question whether a breach occurred will instead depend on a different inquiry: comparing what State Farm paid a class member to the fair market value of the class member’s car. If a jury concluded that State Farm paid an amount equal to (or more than) this fair-marketvalue number, the class member would lack any conceivable theory to assert that the adjustment breached the policy. See Freeman , 149 F.4th at 469; Lara v. First Nat’l Ins. Co. of Am. , 25 F.4th 1134, 1139 (9th Cir. 2022). If, by contrast, the jury concluded that State Farm paid an amount less than the fair market value, the class member might have a breach theory tied to the receipt of less than the actual cash value. See Freeman , 149 F.4th at 469; Lara , 25 F.4th at 1139. (Frankly, it is not obvious how this fact alone would prove breach. The policy tells the parties that they must determine actual cash value either through agreement or appraisal. If both parties voluntarily agreed to an amount, where is the breach? But this class-certification question does not require us to decide that merits issue.) Either way, State Farm’s use of a typical -negotiation adjustment would not show a breach in any specific case without evidence about a car’s fair market value. See Freeman , 149 F.4th at 469; Schroeder , 146 F.4th at 575 – 78; Lara , 25 F.4th at 1140.

A few hypotheticals illustrate the disconnect between Clippinger’s proposed question and a breach. Imagine that State Farm used other valuation evidence (like a valuation from the Kelley Blue Book or an expert appraiser) to convince a jury that its Autosource Report overvalued a specific class member’s car despite its use of the typical -negotiation adjustment. Or imagine that State Farm uncovered that the buyers of comparable vehicles identified in an Autosource Report had negotiated an actual sales price below State Farm’s estimated sales price using that adjustment. Or consider Clippinger's own case. Imagine that a jury believed the State Farm appraiser's valuation of her minivan ($14,432). This finding would show that her Autosource Report overvalued her van (at $14,490) even while using the typical-negotiation adjustment. Each of these class members would have received what State Farm promised (actual cash value) despite the typicalnegotiation adjustment’s claimed illegitimacy as a general matter. Like other district Page 13courts that have improperly certified similar classes, therefore, the district court here misread the policy when it found that the general market-conditions question would resolve the breach element for all class members. See Freeman , 149 F.4th at 468 – 69; Schroeder , 146 F.4th at 576 – 78; Drummond , 142 F.4th at 155 – 56; Sampson , 83 F.4th at 422.

In response, Clippinger argues that a jury could find that State Farm categorically breached its policy merely by using the typical-negotiation adjustment even if (as in these hypotheticals) many class members received an amount equal to or above the fair market value of their vehicles. Cf. Drummond , 142 F.4th at 157. Her reason? She points to nothing in the policy that categorically prohibited this adjustment. She instead relies on two outside-the-policy sources. To start, she says that the adjustment violated the Tennessee regulation, which Tennessee law likely treats as incorporated into her policy. See Martin v. Powers , 505 S.W.3d 512, 517 – 18 (Tenn. 2016). Recall that the regulation allows an insurer to pay the “actual cost” of a “comparable” vehicle and lists four non-exclusive ways to identify this cost. Tenn. Comp. R. & Regs. § 0780-01-05-.09(1)(b). According to C lippinger, the adjustment would violate the requirement to pay the “actual cost” of comparator vehicles if it systematically undervalued those vehicles. But this claim suffers from the same problem: State Farm would have violated this actual-cost mandate for any given class member only if it paid less than the fair market value for that class member’s car. No better is Clippinger’s reliance on a different part of this regulation that requires an insurer to apply only “specific and appropriate” “deductions” to the actual “cost” of a replac ement vehicle (such as deductions for the “salvage” value of the destroyed vehicle). Id. § 0780-01-05-.09(1)(c). The typical-negotiation adjustment does not qualify as a deduction to the actual cost of replacement vehicles; it represents part of the process for identifying that actual cost (whether rightly or wrongly). So this limitation on deductions is beside the point.

Next, Clippinger argues that Tennessee’s implied duty of good faith and fair dealing barred State Farm from using the typical-negotiation adjustment. She reasons that the policy granted State Farm 'discretion' over how to calculate actual cash value and t hat the good-faith duty prohibited State Farm from exercising its discretion in an unreasonable way that 'deprive[d]' insureds “the benefit of” their bargain. Appellee’s Supp. Br. 12 (quoting Glob. Mall P’ship v. Shelmar Retail Partners, LLC , 2017 WL 2839741, at *6 (Tenn. Ct. App. July 3, 2017)). But Page 14Tennessee courts do not recognize an independent good-faith cause of action, so parties cannot use this doctrine to add duties to a contract that do not fall within its four corners. See Cadence Bank , 473 S.W.3d at 769; see also Wallace v. Nat’l Bank of Com. , 938 S.W.2d 684, 686 – 87 (Tenn. 1996). Tennessee courts thus “have not recognized a duty to negotiate in good faith absent an express contractual agreement to do so.” Barnes & Robinson Co. v. OneSource Facility Servs., Inc. , 195 S.W.3d 637, 643 – 44 (Tenn. Ct. App. 2006). And here, Clippinger identifies no policy language that State Farm breached other than its promise to pay actual cash value. Yet the use of the adjustment would not have deprived a class member of the “benefit” of this promise if the class member received the promised actual cash value. Glob. Mall , 2017 WL 2839741, at *6. Like her theory tied to the Tennessee regulation, then, this theory also turns on a case-by-case inquiry into the actual cash value of each class member’s vehicle. A contrary holding would improperly “read an unwritten requireme nt” into the policy. Ambrosio , 154 F.4th at 1111.

Even if this marketcondition question does not resolve the breach element “in one stroke,” Wal-Mart , 564 U.S. at 350, might it still qualify as a common one under Rule 23(a)(2)? A group of amici asks us to hold that it cannot, seemingly on the ground that a common question must decide at least one element of a claim. Yet, even if the question does not resolve the breach element, it may still affect that element by giving class members “relevant evidence” to help prove individual breaches. Lara , 25 F.4th at 1140; cf. Speerly , 143 F.4th at 319 – 20. Common evidence about the typicalnegotiation adjustment’s inaccuracy might bolster (even if it did not prove) an individual class member’s claim that State Farm undervalued that member’s vehicle. Further, the amici’s conclusion th at the question does not represent a common one would have broad effect: it might foreclose not just a class action seeking damages (under Rule 23(b)(3)) but also a class action seeking an injunction against the adjustment (under Rule 23(b)(2)). Cf. Doster , 54 F.4th at 438 – 39. Ultimately, we see no need to reach these broader commonality issues. Rather, we will simply assume that this marketcondition question “resolves a central issue short of the breach issue” and thus satisfies Rule 23(a)(2). Schroeder , 146 F.4th at 577.

Even if Clippinger proved that “common” issues existed, she still must show that they 'predominate' over the individual issues. Fed. R. Civ. P. 23(a)(2), (b)(3). Under this 'more Page 15demanding” predominance element, courts must separate issues that a jury can decide for the class (common issues) from those that will require member-by-member decisions (individual issues). Amchem Prods., Inc. v. Windsor , 521 U.S. 591, 623 – 24 (1997); see Speerly , 143 F.4th at 317, 320. If, on balance, the latter outweigh the former, common issues do not predominate. See Fox , 67 F.4th at 300; see also Ambrosio , 154 F.4th at 1110. That is the case here.

1. 1. Individual valuation questions will predominate over any common issues.

The predominance factor forecloses this class action because an individual issue will matter more than any common ones. On the “common” side of things, we have assumed that Clippinger identified two questions capable of class-wide determination. First, did State Farm agree to the same enforceable duty to pay the “actual cash value” to all class members? And second, did State Farm’s typical -negotiation adjustment “accurately reflect[]” the way the modern used -car market values cars? Clippinger , 2023 WL 7213796, at *11.

Turning to the “individual” side of things, though, one question “likely will dominate” these common ones: What was the actual cash value (that is, the fair market value) of each class member’s vehicle? Tarrify Props., LLC v. Cuyahoga County , 37 F.4th 1101, 1106 – 07 (6th Cir. 2022). As we have explained, a jury would have to identify this value for each class member before it could resolve the breach and damages elements of that member's claim. See Drummond , 142 F.4th at 159 – 60; Lara , 25 F.4th at 1139 – 40. And this valuation analysis will require a factintensive review. Indeed, what we have already said about homes remains true for cars: “[d]etermining fair market value requires an independent and individualized assessment of each absent class member’s property.” Tarrify , 37 F.4th at 1106; see Fox , 67 F.4th at 301. Clippinger’s own appraiser explained that a used car’s fair market value depends on “several factors,” including “the year, make and model, mileage, options, and the overall condition of the vehicle” before an accident. Bent Appraisal, R.146-20, PageID 4288; see Freeman , 149 F.4th at 469.

Take Clippinger’s minivan. Identifying its fair market value required three appraisers, each of whom produced a different number based on a different approach. State Farm's appraiser estimated a value below what State Farm had paid Clippinger. This appr aiser used “prices in actual sale transactions” for four comparison vans, called local dealerships to confirm these sales prices, Page 16and then compared Clippinger’s van to the identified minivans. O’Rourke Appraisal, R.146 -19, PageID 4187, 4190 – 92. The other two appraisers, by contrast, valued the van above what State Farm had paid. Clippinger's appraiser considered 'nationally recognized valuation guides,' communicated with “local dealers and private sellers,” and compared the sales prices of similar vans. Bent Appraisal, R.146-20, PageID 4287 – 88. The third appraiser, for his part, methodically evaluated all aspects of Clippinger's van (including the jack inputs and rear cupholders) and compared it to similar vans. Chandler Appraisal, R.146-21, PageID 4296 –304. Clippinger’s own facts thus rebut her claim that State Farm’s individualized defenses are “entirely hypothetical.” Appell ee's Supp. Br. 16. To the contrary, the insurer would have the right to use its non -hypothetical expert appraiser’s valuation to show that it paid actual cash value in her own case.

What do these appraisals in Clippinger’s case show for the trial? A jury would have to study the “unique characteristics” of each used car. Freeman , 149 F.4th at 469. But unlike the appraisers, the jury (or perhaps multiple juries) would have to do so for all 90,000 vehicles in the class. It would first have to consider unique evidence to determine each car’s value. It would then have to compare this estimate to the amount that State Farm paid. So before the jury (or juries) could identify the class m embers that State Farm “undercompensated,” the parties would have to present “individual proof” on a “plaintiff -byplaintiff” basis. Drummond , 142 F.4th at 159 – 60.

Other factors could bog down the proceedings even further. For one thing, the Tennessee regulation gives State Farm the option to choose from among several “methods” for determining a vehicle's value. Tenn. Comp. R. & Regs. § 0780-01-05-.09(1). The regulation thus would permit State Farm to use evidence other than the Autosource Report to convince a jury that it paid fair market value to a class member. For another thing, the insurance policy gives State Farm the right to invoke the “appraisal” process. Policy, R.146-2, PageID 4018. So State Farm could seek appraisals for all 90,000 class members — just as it did for Clippinger. These factors confirm the individualized nature of the inquiry. Cf. Ambrosio , 154 F.4th at 1112; Speerly , 143 F.4th at 335.

Similar logic has led five other circuit courts to reject class certification in other actualcash-value cases on these predominance grounds. See Ambrosio , 154 F.4th at 1110 – 13; Freeman , 149 F.4th at 468 – 71; Schroeder , 146 F.4th at 576 – 78; Drummond , 142 F.4th at 158 – 61; Sampson , 83 F.4th at 417, 421 -23. The Ninth Circuit's Ambrosio decision exemplifies this uniform Page 17precedent. There, an insured challenged Progressive’s “projected sold adjustment,” which reduced the advertised prices for comparable vehicles to “reflect” that consumers “negotiat[e] a different price than the listed” one. Ambrosio , 154 F.4th at 1108. The district court refused to certify an Arizona class of Progressive customers subject to this adjustment because individual questions about vehicle valuations predominated over all other issues. Id. at 1109. The Ninth Circuit affirmed. Id. at 1113. I t reasoned that the question whether Progressive’s use of the adjustment breached its contract with an insured “would involve an inquiry specific to that person.” Id. at 1112 (quoting Lara , 25 F.4th at 1139). And that factspecific inquiry was the sort of “aggregation -defeating, individual issue[]” that precluded class certification. Id. (citation omitted).

Clippinger tries to distinguish Ambrosio by invoking our abuse-of-discretion standard of review. She reasons that this standard is deferential enough to authorize answers that go “either way” on the same discretionary question. See Cooley v. Carmike Cinemas, Inc. , 25 F.3d 1325, 1332 (6th Cir. 1994). And the court in Ambrosio affirmed the denial of a class certification, while we would have to reverse the district court’s grant of class certification here. But a district court’s legal mistakes always amount to an abuse of discretion. See Speerly , 143 F.4th at 315. As a result, many appellate courts have reversed district courts in actual-cash-value cases when those courts misinterpreted an insurance policy as categorically barring the challenged adjustment. See Freeman , 149 F.4th at 471; Schroeder , 146 F.4th at 573; Drummond , 142 F.4th at 155 – 57. The district court here committed the same type of legal mistake.

To its credit, the district court recognized the presumptive need for vehicle-specific evidence and proposed a possible solution. According to that court, Clippinger challenged only the typical-negotiation adjustment in the Autosource Reports. See Clippinger , 2023 WL 7213796, at *11. So the court accepted Clippinger’s proposal that it could determine each class member’s injury in one fell swoop: it would recalculate the actual cash value of each member’s vehicle by “running every step of State Farm’s methodology” except for the typical-negotiation adjustment. Id. at *12. The court reasoned that State Farm “cannot dispute” this model’s accuracy because the insurer itself relied on the Autosource Reports for its valuation. Id. (citation omitted). At first blush, then, the court’s “formulaic calculation” might solve the individual -evidence problem that Page 18should otherwise derail this class action. Fox , 67 F.4th at 301 (quoting 2 William B. Rubenstein, Newberg and Rubenstein on Class Actions § 4:54, at 284 (6th ed. 2022)).

In avoiding that problem, though, the district court created a much bigger one: its proposal would violate the Rules Enabling Act. Congress put important safeguards on the Supreme Court’s 'power' to set 'rules of . . . procedure,” including the rules for class actions. 28 U.S.C. § 2072(a). Among other things, those rules may not “abridge, enlarge or modify any substantive right” of a party. Id. § 2072(b); see Wal-Mart , 564 U.S. at 367. This provision requires a district court to implement the 'adventuresome innovation' that is a Rule 23(b)(3) class action in a way that respects the statutory and contractual rights of both sides. Wal-Mart , 564 U.S. at 362, 367 (citation omitted); see Ortiz v. Fibreboard Corp. , 527 U.S. 815, 845 (1999). Courts thus may not meet the predominance element by barring a defendant from raising unique statutory or contractual “defenses to individual claims.” Wal-Mart , 564 U.S. at 367. And a district court flouts the Rules Enabling Act if it seeks to make a class action more m anageable by “replac[ing]” the claim -byclaim proceedings required by “individualized” defenses with a “Trial by Formula” approach. Id.

The district court’s refusal to allow State Farm to introduce vehicle -specific evidence other than the Autosource Reports contradicts these rules. The insurance policy does not limit the evidence that State Farm may use to value destroyed vehicles when negotiating with customers over fair market value. Likewise, the Tennessee regulation allows State Farm to choose from many valuation methods. Tenn. Comp. R. & Regs. § 0780-01-05-.09(1); see Sampson , 83 F.4th at 420. In other words, State Farm would have the contractual and regulatory right to introduce myriad evidence apart from the Autosource Reports — other valuation sources, other expert appraisers, other comparable vehicles, or the like — to convince a jury that it paid a class member actual cash value despite its use of the typical-negotiation adjustment. And apart from the evidence that State Farm may use at trial, the policy also gave it the right to avoid a jury valuation altogether by requiring a dispute-resolution process tied to expert appraisers. Cf. Speerly , 143 F.4th at 335 – 36. But the district court’s approach for calculating damages— limiting the valuation evidence to the pre-litigation Autosource Reports — would wrongly bar State Farm from invoking these rights.

To make matters worse, this evidentiary limit would not just eliminate State Farm’s rights. It would also eliminate the rights of absent class members. These class members could ostensibly Page 19argue that the Autosource Report undervalued their vehicles by more than the typical-negotiation adjustment. Clippinger's case confirms this point. The Autosource Report used the typical -negotiation adjustment to reduce the value of the four comparable minivans by between $790 and $940. Under the district court’s proposed valuation model, then, Clippinger should have received an amount in that range. Yet Clippinger ultimately received much more because two appraisers ended up valuing her minivan at arou nd $4,000 above the Autosource Report’s estimation. To certify a class, Clippinger thus proposes a formula that would harm all other similarly situated class members (those whose Autosource Report undervalued their vehicles in other ways). Cf. Gates v. Rohm & Haas Co. , 655 F.3d 255, 274 (3d Cir. 2011).

Why did the district court conclude that it could restrict the parties’ rights in these ways? It overlooked the rights of absent class members. And it suggested that State Farm had forfeited its right to introduce unique evidence of a vehicle’s valuation because it relied on the Autosource Report to seek an agreement with members of the class. See Clippinger , 2023 WL 7213796, at *13 – 14. But the court failed to explain why this initial use of the Autosource Report should qualify as a forfeiture. Even if State Farm elected to use that report to reach informal agreements with customers, nothing in the insurance policy or the Tennessee regulation indicated that this election bound the insurer later in litigation. Cf. Sampson , 83 F.4th at 422 -23. Perhaps the policy's language (in particular, the provision saying that both sides may agree on the actual cash value) could have bound State Farm to the Autosource Report if it had agreed that the report represented the actual cash value in a specific case. But State Farm no more agreed that the reports accurately valued totaled vehicles without the typical-negotiation adjustment than the class members agreed that it provided an accurate assessment with that adjustment. We know of no law that would allow class members to adopt an à la carte approach to valuation by taking what they want from the reports and discarding the rest. Either they must accept the value in the reports or start from scratch in l itigation. In short, the district court’s decision to limit S tate Farm to one acceptable valuation method represented an “arbitrary choice” that violated State Farm’s contractual and regulatory right to rely on any acceptable method. Id. ; see also Bourque v. State Farm Mut. Auto. Ins. Co. , 89 F.4th 525, 527 – 29 (5th Cir. 2023).

Page 20Nor does this case resemble those on which the district court and Clippinger have relied. See Jama v. State Farm Mut. Auto. Ins. Co. , 113 F.4th 924, 927 (9th Cir. 2024); Hicks , 965 F.3d at 459 – 63. In Jama , the Ninth Circuit reversed the decertification of a class of State Farm customers in the State of Washington who had confronted a similar negotiation adjustment. See 113 F.4th at 927, 931 – 35. Yet the Jama court relied on a Washington regulation that prescribed the method for identifying actual cash value and categorically prohibited a negotiation adjustment. See id. at 932 –33. The district court thus could calculate the class members’ injuries in a simple way: their damages equaled 'the exact amount of the impermissible negotiation deduction.' Id. at 933.

The Tennessee regulation in this case, by contrast, does not categorically prohibit State Farm from using a typical-negotiation adjustment. See Tenn. Comp. R. & Regs. § 0780-01-05.09(1). State Farm’s payment to a class member would violate this regulation only if the payment did not equal the fair market value of the destroyed vehicle — something that a court would have to decide on a case-by-cas e basis. In fact, the Ninth Circuit's later decision in Ambrosio distinguished Jama on this precise ground. See Ambrosio , 154 F.4th at 1111. The Ambrosio court explained that the challenged adjustment was not “facially unlawful” under Arizona law (unlike the Washington regulation in Jama ). See id. Because the Tennessee law at issue here also does not render the adjustment “facially unlawful,” this case resembles Ambrosio , not Jama . See id.

The district court’s reliance on our Hicks decision fares no better. That case concerned State Farm’s insurance policy for Kentucky homeowners. Hicks , 965 F.3d at 455. This policy promised to pay customers the “actual cash value” when a home suffered damage. Id. It defined this phrase to mean replacement cost minus depreciation. Id. State Farm had included labor costs in the depreciation reduction even though Kentucky law prohibited that practice. Id. at 456. When a class of homeowners sued over this illegal labor deduction, we upheld a class certification. Id. at 466. The class had argued for a simple damages formula that would refund the amount of the illegal labor deduction plus prejudgment interest. See id. at 460. But State Farm responded that individual issues predominated because it would have the right to recalculate the entire actualcash-value number for every class member. See id. Under this theory, State Farm could show that it paid a class member the correct amount (despite the improper deduction) because, for example, Page 21it initially overestimated replacement costs. See id. We rejected State Farm’s argument because it conflicted with Kentucky law. See id. at 461. That law instructed us to treat any overestimate “as an error in the insured’s favor” that State Farm could not take back. Id. (citation omitted). Because State Farm lacked any statelaw “substantive right” to recalculate the actual cash value, 28 U.S.C. § 2072(b), an initial miscalculation did not pose any obstacle to identifying each class member’s injury u sing the simple formula, see Hicks , 965 F.3d at 461 – 62.

Here, however, Clippinger may not rely on Tennessee law to avoid the need for individual inquiries in the way that Hicks relied on Kentucky law. She has identified no Tennessee case, statute, or regulation that would prohibit State Farm from using a new valuation method to show that class members received the actual cash value of their vehicles despite its use of the typicalnegotiation adjustment. In other words, Tennessee law would not treat an initial overestimate of the actual cash value as “an error in the insured’s favor” in the same way that Kentucky law did for homeowner’s insurance. Id. at 461 (citation omitted). State Farm thus has the “substantive right” to present individual evidence concerning the value of each class member’s car that it lacked in Hicks . Wal-Mart , 564 U.S. at 367 (quoting 28 U.S.C. § 2072(b)). And that right means that individual issues will “dominate” all other issues in this litigation. Tarrify , 37 F.4th at 1107.

All told, the class’s common contract claims will require “mini trials as to each” class member. Freeman , 149 F.4th at 471. So individualized questions will predominate over common ones. That fact renders a class action an inappropriate vehicle to manage the litigation.

We reverse and remand for proceedings consistent with this opinion.

Page 22JOHN K. BUSH, Circuit Judge, concurring. I agree with the majority opinion that the district court abused its discretion in certifying this class, and I join that opinion in full. To certify a class, a plaintiff must satisfy all four prongs of Rule 23(a) and at least one prong of Rule 23(b) of the Federal Rules of Civil Procedure. See Amchem Prods., Inc. v. Windsor , 521 U.S. 591, 613 – 14 (1997). Because the majority correctly holds that Clippinger cannot satisfy the predominance requirement of Rule 23(b)(3), the court need not address whether the class would satisfy the other requirements of Rules 23(a) and (b)(3).¹ I write separately, however, to highlight some additional points under Rules 23(a) and (b)(3) that justify the denial of class certification here and warrant consideration in future cases.

Clippinger’s merits theory makes the case for class certification quite puzzling. Although the merits are not before us for definitive decision on class certification, the class certification analysis “[f]requently . . . will entail some overlap with the merits of the plaintiff’s claim.” WalMart Stores, Inc. v. Dukes , 564 U.S. 338, 351 (2011). So even though we may not conduct a minitrial on the merits at certification, see, e.g. , Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds , 568 U.S. 455, 459 – 60 (2013), we are required to acknowledge the merits and their potential outcome when that could impact how the case is litigated after certification, see Dukes , 564 U.S. at 351 – 52.

As I interpret it, Clippinger argues on the merits that the typical-negotiation adjustment (TNA) is a breach of the insurance policy and the implied duty of good faith and fair dealing Page 23because it violates (1) the plain text of the policy and (2) a Tennessee regulation. For many reasons discussed below, either theory augers against class certification.

To the extent that Clippinger relies on the policy language, certification would be improper under Rule 23(a)(2), which requires that there be “questions of law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). That is because, under the logic of her merits theory vis-à-vis the policy language, Clippinger’s proffered common question of law or fact— the validity of the TNA — falls out of the case.

Why is this so? Well, first note how the policy defines the actual cash value, or “ACV”: it is either (1) the number agreed to by State Farm and the policyholder or (2) the number agreed to by at least two of the three appraisers. Appraisal is binding if the parties cannot agree to a valuation. See R. 1462, Policy, PageID 4018 (“If there is disagreement as to the [ACV] of the covered vehicle, then the disagreement will be resolved by appraisal . . . .' (bolding and italics omitted)).

So either of two things will happen when a policyholder totals her car. The first option is that she will agree with State Farm on a valuation and accept the money. If that occurs, then the policyholder is barred by Tennessee estoppel doctrine from asserting that State Farm breached the policy. See Gold Kist, Inc. v. Pillow , 582 S.W.2d 77, 79 – 80 (Tenn. Ct. App. 1979) (accepting partial performance estopped plaintiff from asserting breach). The second option is that she will disagree with State Farm’s va luation, and one of the parties will demand appraisal. If that happens, the policyholder is bound by the number that two of three appraisers come up with. Either way, the policyholder receives the ACV as defined by the policy: the number to which either the parties themselves, or the appraisers selected by them, agree.²

Page 24Here, it just so happens that Clippinger was paid through both methods. She first accepted State Farm’s proffered settlement. Then, more than a year later, she sued and received payment under the policy’s appraisal process.

Clippinger's receipt of payments under both contracted -for scenarios thus dooms her motion for class certification based on the TNA issue. Whether viewed through the lens of estoppel or appraisal, Clippinger has received the ACV of her car. It also should end her individual lawsuit, which is ultimately based on State Farm’s failure to pay ACV, not whether the TNA is valid or not. See Thomas Sysco Food Servs. v. Martin , 983 F.2d 60, 62 (6th Cir. 1993) (“A case will become moot when the requested relief is granted or no live controversy remains.”).

She nonetheless rests both her suit and her motion for class certification (and the district court rested its grant of that motion) on one common question of fact or law: whether the TNA is valid. Yet, under the plain language of the policy and Tennessee contract law, the validity of the TNA is irrelevant to whether State Farm has committed a breach justifying damages over and above those provided by the appraisal process under the policy. That is because the absent class members are either estopped from asserting a breach of the policy because they accepted State Farm’s Autosource -generated offer, see Gold Kist , 582 S.W.2d at 79 – 80, or they have received exactly what the policy says they are entitled to if they object to the offer — the ACV as determined by at least two of three appraisers. In either situation, the validity of the TNA does not matter for the amount of recovery. Without the TNA question having any impact on the outcome of the dispute, Clippinger cannot show that there is a common question of law or fact as to the entire class.

What is more, the policy squarely forecloses Clippinger from receiving any further relief beyond the ACV as determined by the contract-mandated method of valuation. First, Clippinger accepted the amount State Farm offered. Then, despite this resolution, Clippinger contested the valuation, at which point State Farm went through the appraisal process and paid her what the appraisers said she was owed. Through State Farm’s belt -and-suspenders approach to payment, there can be no doubt that Clippinger received all of the money that she was owed under the policy. That means the only remaining question is whether Clippinger is entitled to her appraisal fees or litigation costs as consequential damages.

Page 25Clippinger is not entitled to either category of damages. Even if Clippinger were to show a breach under the first part of the total-loss-settlement process — that is, the phase where State Farm makes an offer and the insured can choose whether or not to accept it — the only monetary remedy provided by the policy is through the appraisal process itself. The appraisal provision provides that “[e]ach party will pay the cost of its own appraiser, attorneys, and expert witnesses, as well as any other expenses incurred by that party. Both parties will share equally the cost of the third appraiser.” R. 146 -2, Policy, PageID 4018. Thus, even if we assume Clippinger’s breach theory is correct, she has already received everything she would be entitled to under her insurance policy. And her claimed “damages” for appraiser’s or attorney’s fees are not recoverable damages at all, given the explicit policy language barring such recovery.

This logic renders Clippinger incapable of being an adequate class representative, as required by Rule 23(a)(4). When only one person represents the class, she is adequate only if, among other things, she could serve as a vehicle for resolving all of the absent class members’ claims. See Stout v. J.D. Byrider , 228 F.3d 709, 717 – 18 (6th Cir. 2000). Stated differently, if some or all of the absent class members' claims would remain unresolved after resolving the named plaintiff's claims, then she is inadeq uate. See id. at 717 – 18. That is the case with Clippinger. The district court could easily resolve Clippinger's remaining outstanding claim without deciding whether use of the TNA breaches the policy. That is because once the court holds that Clippinger is not entitled to any further damages as a matter of contract interpretation, her case is over. She has received everything she is entitled to under the policy, and there is no other relief left to grant her. But the absent class members’ claims would be left unresolved, thus demonstrating why Clippinger would be an inadequate class representative.

Additionally, Clippinger's merits theory vis -à-vis the Tennessee regulation at issue — § 0780-01-05.09(1)(b) of the Tennessee Compilation of Rules and Regulations — would render class certification inappropriate under Rules 23(a)(3) and (4). Under Rule 23(a)(3), “the claims or defenses of the representative parties” must be “typical of the claims or defenses of the class.” Fed. R. Civ. P. 23(a)(3). A class represen tative’s claims are typical when “her claims are based on the same legal theory” as the rest of the class and proving her “own claim” would “necessarily”Page 26 prove the absent class members’ claims. Beattie v. CenturyTel, Inc. , 511 F.3d 554, 561 (6th Cir. 2007) (first quoting In re Am. Med. Sys., Inc. , 75 F.3d 1069, 1082 (6th Cir. 1996); and then quoting Sprague v. Gen. Motors Corp. , 133 F.3d 388, 399 (6th Cir. 1998)). And under Rule 23(a)(4), the named plaintiff must be able to “fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). The class representative is adequate when she does not have any conflicts of interest with the absent class members and has a qualified lawyer to prosecute the case. Beattie , 511 F.3d at 562 – 63. Typicality and adequacy are often two sides of the same coin. See, e.g. , In re Whirlpool Corp. Front-Loading Washer Prods. Liab. Litig. , 722 F.3d 838, 852 – 53 (6th Cir. 2013).

Clippinger asserts that the TNA violates the regulation by preventing the policyholder from receiving the ACV of her car. However, the regulation itself (and also the policy language) forecloses this theory. Although the regulation gives State Farm discretion on how to calculate the ACV, the regulation ultimately requires that the “cash settlement” be “based upon the actual cost . . . to purchase a comparable automobile,” Tenn. Comp. R. & Regs. § 0780-0105.09(1)(b)(4), similar to what the policy requires, see R. 1462, Policy, PageID 4018 (“Pay the actual cash value of the covered vehicle minus any applicable deductible.” (bolding and italics omitted)). And as all agree in this case, some cars sell below sticker price, 3 while others sell at or above sticker price. That means there are at least some cars for which the TNA, or some other similar deduction, would be permissible because the sticker price would not perfectly reflect the actual cost of a comparable vehicle, justifying some sort of adjustment.

The majority rightly notes that this bifurcation of categories of vehicle owners raises serious predominance and commonality difficulties, see Majority 13 – 15, but it also causes problems under Rules 23(a)(3) and (4). When a class can be firmly divided into two groups with divergent interests, the named plaintiff is typical and adequate only for the group to which she belongs. See Amchem , 521 U.S. at 625 – 28; CE Design Ltd. v. King Architectural Metals, Inc. , 637 F.3d 721, 724 – 28 (7th Cir. 2011). According to Clippinger's evidence, this class can be divided into two groups: those whose ACVs are less than sticker price and those whose ACVs are Page 27greater than, or equal to, sticker price. These groups have divergent interests. The low ACV group is incentivized to take a settlement before appraisal because appraisal likely will not net them any money, and the cost of the appraisal might swallow up whatever money they receive. But the high ACV group is more incentivized to compel appraisal before settling because appraisal is likely to net them a damages award large enough to make it worth spending money on the appraisal. Clippinger can be in only one group or the other.

This conflict could impact settlement negotiations for the class. The group with the low ACV would presumably support a lower settlement value than the group with the high ACV. Clippinger cannot represent both groups. Therefore, even if Clippinger could somehow overcome her other deficiencies as a class representative, her claims cannot be typical of, and she cannot adequately represent, the entire class, as required by Rules 23(a)(3) and (4).

The economic realities of Clippinger’s claims further demonstrate that she cannot satisfy Rule 23 —particularly Rule 23(b)(3)’s requirement that litigating the case as a class action would be “superior to multiple individual suits” of the same kind as the c lass action. Bowles v. Sabree , 121 F.4th 539, 555 (6th Cir. 2024) (emphasis omitted). Because Clippinger's proposed class action is no different from individual suits as a practical matter, the motion for class certification should have been denied.

“The class action is an ingenious device for economizing on the expense of litigation and enabling small claims to be litigated.” Thorogood v. Sears, Roebuck and Co. , 547 F.3d 742, 744 (7th Cir. 2008) (Posner, J.). It is a particularly useful device in many consumer protection claims, where 'aggregation of small claims is likely the only realistic option for pursuing a claim.' Feeney v. Dell, Inc. , 908 N.E.2d 753, 763 (Mass. 2009). But some consumer class actions involve permember recoveries so small that large numbers of unnamed class members will not even seek distribution. See, e.g. , Briseño v. Henderson , 998 F.3d 1014, 1020 (9th Cir. 2021) (in case where class recovery was fifteen cents each, “barely more than one -half of one percent of [class members] submitted a claim”). After all, people value their time and energy, so they will do something only if they gain more reward from that effort than they value the time and energy lost doing it. And Page 28just as a litigant is unlikely to spend the time and energy filling out the required paperwork for a few cents, he is unlikely to litigate an entire damages trial from start to finish to recover a relatively small amount of money.

Clippinger’s proposed class runs into exactly that problem. Assuming that Clippinger is typical of the absent class members (in both the legal and colloquial sense), the average claim appears to be worth, at most, only a few thousand dollars. To recover that money, a class member would conceivably need to (1) receive notice of the class action, (2) track down his original claim paperwork, (3) figure out the value of his vehicle without the TNA, (4) develop admissible evidence to support that valuation, and (5) present that evidence to the district court. It is unlikely that a lawyer will take a case that small, so we would be asking each absent class member to do all of that and then present his case to a jury pro se. I tend to think this solution is unlikely to get class-wide buy-in.

Indeed, from a purely economic standpoint, Clippinger's theory of class certification defeats the purpose of certifying this class. Class actions are worthwhile precisely because carrying out thousands of small-dollar trials would not be worth the time and money required to litigate them. Thorogood , 547 F.3d at 745. And yet Clippinger is asking for the district court to do precisely that. True, the liability phase would be resolved under Clippinger’s view, but that only slightly eases the burden of trying these cases. To prove damages, the class member and State Farm would have the right to offer competing evidence of each vehicle’s value. That means the damages portion of each case could be far more complex and involved than the liability portion.

In fact, certifying this class action could harm consumers in the aggregate far more than it could help them. If State Farm were to exercise its right to an appraisal for all 90,000 absent class members, and if Clippinger’s claim is typical, one can imagi ne State Farm being forced to spend as much as tens of millions of dollars on appraisal fees and hundreds of millions in additional costs for thousands of vehicles for which no owners had ever contested State Farm’s valuation. If State Farm exercised its right to go to trial, it would also have to pay a substantial amount in legal fees, expert witness fees, and other costs. Or State Farm would be forced to settle the case, regardless of the strength or weakness of Clippinger’s liability theory, because pa ying a settlement if the class is certified may be cheaper than the anticipated litigation and appraisal costs.

Page 29Whether State Farm goes to trial or settles, the outcome threatens to cause economic harm to the absent class members who had no beef with State Farm's valuation prior to it being challenged by Clippinger, as well as any Tennessee drivers whose insurers settle claims in a fashion relevantly similar to State Farm’s process. State Farm ostensibly included the appraisal process in the policy to avoid the costs of potential litigation, pass at least some of those savings on to the policyholders in the form of lower premiums, and thereby improve its competitive position. See generally Fire Ass ’ n of Philadelphia v. Ballard , 112 S.W.2d 532, 534 (Tex. Civ. App. 1938) ('The purpose of an appraisal provision is apparently to afford a simple, speedy, inexpensive and fair method of determining the loss or damage resulting from the happening of a contingency insured against.”). B ut exposing State Farm to the risk (or reality) of this type of litigation may force State Farm — and any other insurance company that employs a similar totalloss provision —to raise premiums because such a “speedy, inexpensive” way to resolve coverage disputes would no longer be available. Id. Or the insurance companies may resort to additional contractual language, such as arbitration provisions and class action waivers in their policies — outcomes that some argue would be a net-negative result for consumer-protection enforcement. See Kristian v. Comcast Corp. , 446 F.3d 25, 61 (1st Cir. 2006). In short, granting the motion for class certification may be more of an economic detriment for the absent class members in the long run than the short-term legal problems it purports to solve.

Clippinger claims that “no one on this Court, knowing the data manipulation and deletion that occurred, could apply the [TNA] on Plaintiff and Class Members without feeling a pang of conscience were they the decisionmaker on totalloss claims.” Appellee’s En Banc Supp. Br. 13 n.4. But members of this court may also feel a pang of conscience were the class to be certified and the ultimate result of this litigation is to raise premiums for the very policyholders this class action is supposed to protect — thousands of policyholders, most of whom, again, never objected to State Farm’s calculation of ACV for their vehicles.

Of course, we decide this case based on law, not economics. But it is worth noting that denying class certification here may be a net economic benefit for Tennessee auto owners as a whole. At the very least, the full financial picture for Tennessee consumers cautions against glibly Page 30accepting Clippinger’s over -the-top rhetoric implying that class adjudication would be a superior method for resolving these disputes under Rule 23(b)(3).

State Farm raises another interesting argument against class certification — that too many absent class members lack standing to warrant certification. Though unavailing here, the insurer’s point deserves discussion for future class action cases, as it pertains to the numerosity requirement of Rule 23(a)(1).

Rule 23(a)(1) permits class certification only if “the putative class is so numerous that joinder of all members is impracticable.” Golden v. City of Columbus , 404 F.3d 950, 965 (6th Cir. 2005) (quoting Fed. R. Civ. P. 23(a)(1)) (quotation marks omitted). And our precedents require that a named plaintiff provide some non-speculative basis for believing that the putative class is sufficiently numerous. See id. at 965 – 66; Alkire v. Irving , 330 F.3d 802, 820 (6th Cir. 2003). Thus, whether unnamed class members have standing is relevant to whether the class is, in fact, so numerous that joinder of all members is impracticable. In other words, the district court was required to consider the unnamed class members’ standing.

Constitutional standing analysis rests on the premise that “[f]ederal courts are courts of limited jurisdiction.” Johnson v. Johnson , 157 F.4th 813, 817 (6th Cir. 2025) (quoting Ohio ex rel. Skaggs v. Brunner , 549 F.3d 468, 474 (6th Cir. 2008) (per curiam)). As such, the federal courts may hear a case only when both the Constitution and an Act of Congress permit them to do so. Id. One of those constitutional limitations is Article III standing, which ensures that only those litigants with a concrete interest in the outcome of a dispute may invoke the authority of the federal courts. FDA v. All. for Hippocratic Med. , 602 U.S. 367, 378 – 80 (2024). When a plaintiff lacks Article III standing, the district court lacks subject matter jurisdiction to decide the case, Tenn. Gen. Assembly v. U.S. Dep’t of State , 931 F.3d 499, 507 (6th Cir. 2019), and its decisions are void, In re Vista-Pro Auto., LLC , 109 F.4th 438, 443 – 44 (6th Cir. 2024), aff’d sub nom. Coney Island Auto Parts Unlimited, Inc. v. Burton Tr. for Vista-Pro Auto., LLC , 146 S. Ct. 579 (2026).

“The Supreme Court has not decided whether an unnamed class member’s lack of standing poses an Article III problem.” Speerly v. Gen. Motors, LLC , 143 F.4th 306, 315 (6th Cir. 2025)Page 31 (en banc). The Supreme Court has, however, held that each and every class member must have standing to recover damages. TransUnion, LLC v. Ramirez , 594 U.S. 413, 431 (2021). This has important implications for class certification.

Consider how it can play out in practice. In TransUnion , the class had 8,185 members, but only the named class member had standing to bring two of the claims. 594 U.S. at 418, 440. Such circumstances are far from unique. Many consumer class actions involve a small number of plaintiffs who have suffered a real Article III injury, and a large number of plaintiffs who have suffered a “bare procedural violation”— that is, a violation of a statute without any harm to the plaintiff. Spokeo, Inc. v. Robins , 578 U.S. 330, 341 (2016). In these cases, the district court would have subject matter jurisdiction over a class of just one person or a handful of people — a class that is not so numerous that joinder would be impractical. And since numerosity is a mandatory prerequisite to class certification, that failure would preclude class certification.

As I see it, then, a district court considering a class certification motion needs to consider Article III standing as part of its numerosity analysis. And when the vast majority of absent class members lack Article III standing, class certification will be inappropriate.

In a case like this, such a defect would not change the outcome. We assume that a plaintiff will win on the merits when we are considering her standing, Merck v. Walmart, Inc , 114 F.4th 762, 772 (6th Cir. 2024), and a Tennessee plaintiff cannot show a breach of contract unless she has suffered damages, Fed. Ins. Co. v. Winters , 354 S.W.3d 287, 291 (Tenn. 2011). That presumption means that we also assume every absent class member here has necessarily suffered a pocketbook injury that can support Article III standing. See, e.g. , Collins v. Yellen , 594 U.S. 220, 243 (2021).

That may not be true in a future case, though, requiring us to confront the standing issue head-on.

With these observations, I concur in the majority opinion.

Page 32JULIA SMITH GIBBONS, Circuit Judge, dissenting. The district court certified a class of Tennesseans insured by State Farm who received payouts from State Farm after their cars were totaled. Under its insurance policies and Tennessee state law, State Farm was obligated to pay its insured parties the actual cash value (ACV) of their totaled vehicles. Named plaintiff Jessica Clippinger accepts all aspects of State Farm’s methodology to calculate ACV, except for its use of a specific cost adjustment known as a Typical Negotiation Adjustment (TNA). Clippinger argues that the TNA is facially invalid as a breach of contract and Tennessee law because it does not accurately reflect market conditions and therefore uniformly and artificially reduced the class membe rs’ vehicles’ valuations below ACV. Indeed, Clippinger’s theory of liability, common to all putative class members, is that State Farm's application of the TNA always results in a depreciation of an insured’s valuation below ACV and is, therefore, a per se breach of contract. Because the district court did not abuse its discretion in concluding that these common questions of liability predominate over any individualized damages issues, we respectfully dissent.

The dispute in this case concerns the payouts that State Farm makes for totaled cars and how these payouts are calculated. During the time relevant to this case, State Farm worked with the company Audatex, which, upon State Farm’s request, found comparabl e cars — broadly, cars of the same model, year, and style as the totaled car — in its database. Audatex next determined the sold or advertised price of these cars in the 120-day period around the date the car was totaled. Audatex’s calculation then took an a dditional step challenged by the plaintiffs here: the application of the TNA to the prices of those comparable cars. After applying the TNA, Audatex adjusted the price of each comparable car for mileage, equipment, and options; averaged the adjusted prices; and adjusted the average to account for the “difference between the loss vehicle and the typical measured condition.” DE 140 -10, Sealed Lowell Dep. Tr., Page ID 3591 – 92. State Farm finally factored the adjusted prices of these comparable cars into the ACV figure.

Page 33According to State Farm, the TNA is intended to account for the difference between a vehicle’s advertised price and its fair market value, because buyers and sellers of used cars at dealerships typically negotiate a lower price than the one at which the car is listed. Clippinger claims, however, that the negotiation adjustment is unfair and categorically improper because it does not reflect typical practices or market reality. Based on her account, the modern used-car market involves online shopping and price comparison, and used cars much more often sell for their advertised price. Moreover, she asserts that the TNA is calculated using manipulated data. So, she claims, the TNA artificially decreases the ACV figure. And putative class members who received a payout based on an Audatex valuation with the TNA applied suffered a breach of contract because the application of an artificial adjustment reduced the valuation of their totaled cars, thus depriving them of the ACV that State Farm owed. Clippinger likewise alleges that State Farm’s application of the TNA violated Tennessee law, which was incorporated by statute into her insurance policy.

We start with our common ground with the majority opinion. We agree that the class contains at least one question “common to the class.” Fed. R. Civ. P. 23(a)(2); Maj. Op. at 11. One common question, as the majority assumes, is whether 'State Farm's typ ical-negotiation adjustment ‘accurately reflect[s]’ the way the modern used -car market values cars[.]” Maj. Op. at 18 (quoting DE 202, Order on Class Cert., Page ID 6940). But another, as the district court correctly noted, is whether State’s Farm ACV me thodology with the TNA applied breached its insurance contracts with its insureds by paying “an artificially reduced amount.” DE 202, Order on Class Cert., Page ID 6940 (quotation omitted). And we would go further than the majority and affirmatively find that these classwide questions qualify as common ones under Rule 23(a)(2).

To show commonality, the class must “identif[y] a ‘common question of law or fact’” that '(1) yield[s] a common answer with common evidence and (2) meaningfully progress[es] the lawsuit.' Speerly v. Gen. Motors, LLC , 143 F.4th 306, 316 (6th Cir. 2025) (en banc) (quoting Fed. R. Civ. P. 23(a)(2)). “A common question is one where the same evidence will suffice for each member to make a prima facie showing or the issue is susceptible to generalized, class-wide proof.” Tyson Foods, Inc. v. Bouaphakeo , 577 U.S. 442, 453 (2016) (citation modified). If a Page 34reasonable decisionmaker could answer the question differently for different members of the class based on the common evidence, the question is not common. Speerly , 143 F.4th at 316.

Here, the plaintiffs’ claims share a legal question that can yield a common answer with common evidence: whether State Farm’s methodology to calculate ACV with the TNA applied resulted in “an artificially reduced amount” rather than ACV “as contractually required.” DE 202, Order on Class Cert., Page ID 6940 (quotation omitted). The district court correctly noted that the answer to this legal question depended on classwide factual questions, such as whether the TNA was a fair reflection of market condition s. The “truth or falsity” of the plaintiffs’ legal theory can therefore be resolved “in one stroke.” Wal-Mart Stores, Inc. v. Dukes , 564 U.S. 338, 350 (2013). 'Plaintiffs' claims live and die on this common contention.' Speerly , 143 F.4th at 363 (Moore, J., dissenting).

Moreover, this common question is “central” to the litigation because “it affects at least one contested element” of Clippinger’s claims. Id. at 317 (quoting Wal-Mart , 564 U.S. at 350). The centrality requirement does not require 'rote explication of every element.' Id. at 367 (Moore, J., dissenting). Rather, “the plaintiffs must tie that debated question to the relevant elements of that claim.” Id. at 319 (citation modified). Clippinger did just that. The central legal question here is whether the a pplication of the TNA “means” that State Farm “[is] not calculating ACV as contractually required but [is], instead, calculating an artificially reduced amount.” DE 202, Order on Class Cert., Page ID 6940 (quoting DE 138, Sealed Mot. for Class Cert., Page ID 2517). And the plaintiffs claimed in their motion for class certification that the “breach element” of their breach of contract claim was subject to common proof about the statistical basis for the TNA and Audatex’s underlying valuation method. DE 13 8, Sealed Mot. for Class Cert., Page ID 2516 – 17.

More crucially, the majority opinion errs by failing to reject amici’s misguided contention that, for a question to be common, the question must decide at least one element of a claim. That argument runs counter to Sixth Circuit and Supreme Court precedent. Again, to satisfy commonality, the common question must merely “affect[] at least one contested element” central to the class’s claims. Speerly , 143 F.4th at 317. This means that the “class action must be able to ‘generate common answers apt to drive the resolution of the litigation.’” In re Ford Motor Co. , 86 F.4th 723, 727 (6th Cir. 2023) (per curiam) (quoting Wal-Mart , 564 U.S. at 350). The Supreme Page 35Court has explained that a common answer capable of “driv[ing] the resolution of the litigation” is one where the “truth or falsity” of a “common contention” “resolve[s] an issue that is central to the validity” of the class’s claims. Wal-Mart , 564 U.S. at 350 (emphasis added). That does not require that the common question “resolve” an “element” altogether. See id. ; Speerly , 143 F.4th at 317.

The issue here — whether the TNA is facially invalid and thus artificially reduced ACV for the insureds —undoubtedly “affects” whether State Farm uniformly breached its contracts with all class members. In our view, this issue would resolve the breach element altogether because Clippinger argues that the TNA is always negative and thus always results in class members receiving a valuation below their vehicles’ ACV. But even under the majority opinion’s theory, whether the TNA is facially invalid provides cla ss members “relevant evidence” helpful to proving the breach element. Lara v. First Nat’l Ins. Co. of Am. , 25 F.4th 1134, 1140 (9th Cir. 2022); Jama v. State Farm Mut. Auto. Ins. Co. , 113 F.4th 924, 936 – 37 (9th Cir. 2024); Maj. Op. at 17. To hold otherwise would improperly “transform Rule 23(a)(2) into a miniature Rule 23(b)(3) analysis.” Speerly , 143 F.4th at 319 (citation modified). Because the common answers to these common questions would facilitate 'driv[ing] the [litigation's] resolution,' Wal-Mart , 564 U.S. at 350 (quotation omitted), we would explicitly reject amici’s argument and conclude, rather than merely assume, that Clippinger satisfied Rule 23(a)(2)’s commonality requirement.

We depart from the majority opinion when it concludes that the district court abused its discretion in determining that individual damages issues predominate over common ones. The majority conflicts with circuit precedent and would prevent the certification of valuation-based classes in even the most egregious circumstances.

In a properly certified Rule 23(b)(3) class, “[c]ommon questions subject to classwide proof must predominate over individualized questions, prompting us to ask whether the proposed class action beats the conventional approach of resolving disputes on a case-by-case basis in terms of efficiency and administrability.” Tarrify Props., LLC v. Cuyahoga County , 37 F.4th 1101, 1106Page 36 (6th Cir. 2022). A class may be certified where a common issue predominates “even though other important matters will have to be tried separately, such as damages or some affirmative defenses peculiar to some individual class members.” Hicks v. State Farm Fire & Cas. Co. , 965 F.3d 452, 459 –60 (6th Cir. 2020) (citation modified). Indeed, “[w]hen adjudication of questions of liability common to the class will achieve economies of time and expense, the predominance standard is generally satisfied even if da mages are not provable in the aggregate.” In re Whirlpool Corp. Front-Loading Washer Prods. Liab. Litig. , 722 F.3d 838, 860 (6th Cir. 2013) (quoting Comcast Corp. v. Behrend , 569 U.S. 27, 41 (2013) (Ginsburg, J., dissenting)).

The district court found that the inquiry of “whether the TNA accurately reflects how cars are valued and sold in the market” and, relatedly, whether the valuation with the TNA applied resulted in 'an artificially reduced amount' rather than ACV, was the 'central question' susceptible to classwide proof. DE 202, Order on Class Cert., Page ID 6940 (quotation omitted). It therefore approved a class damages model that removed the TNA from Audatex’s valuations because it found that 'State Farm cannot dispute that applying every aspect of its valuation process other than the TNA leads to an accurate valuation, at least in situations where application of a TNA is inappropriate.” Id. at Page ID 6941 (citation modified). This liability question is common to a class that comprises Tennessee insureds who received an ACV payout from State Farm with the TNA applied.

The district court's conclusion accords with our circuit's precedent. In Hicks , we held that a similar challenge to an insurance payment adjustment presented a common legal issue that predominated over individual issues, even where damages could vary from person to person. 965 F.3d at 459 –60. There, as here, “State Farm argue[d] t hat individual issues predominate[d] over the [liability] issue because it may have miscalculated ACV payments based on individualized errors unrelated to [the challenged adjust ment].' Id. at 460. And there, as here, State Farm argued that it would “defend against the claims of individual class members by proving that some insureds were not damaged because it” may have “overestimated ACV payments” such “that the deduction of [the challenged adjustment] resulted in no damages or it mistakenly reimbursed . . . claimants for more than they were owed.” Id. We rejected this claim, finding unpersuasive “State Farm’s argument that its own potential overestimations show that individua lized inquiries predominate.”Page 37 Id. at 461. To be sure, in Hicks , we emphasized that “[u]nder Kentucky law, insurance contracts are 'construed strictly against the insurer and liberally in favor of the insured.'' Id. (quoting Hicks v. State Farm Fire & Cas. Co. , 751 F. App’x 703, 709 (6th Cir. 2018)). But contrary to the majority opinion's assertion, see Maj. Op. at 25, the same is true under Tennessee law, Tata v. Nichols , 848 S.W.2d 649, 650 (Tenn. 1993).

The majority opinion asserts that individual damages issues predominate because calculating ACV would be “fact -intensive” and would thus require a jury to consider factors like “the year, make and model, mileage, options, and the overall condition of the v ehicle before an accident.” Maj. Op. at 18 (citation modified) (quoting DE 146 -20, Bent Appraisal, Page ID 4288). But Audatex’s model already considered all these factors. State Farm’s witnesses agree that it did so correctly in the first instance. And Clippinger agrees that, except for the application of the TNA, State Farm’s methodology for calculating ACV is accurate.

These facts distinguish Clippinger’s case from the plaintiffs’ case in Tarrify . There, we found that individualized valuation questions predominated where neither party had previously valued the foreclosed homes at issue and the parties had to start anew in identifying the properties’ market value. Tarrify , 37 F.4th at 1104, 1107. Here, by contrast, Clippinger seeks to use the ACV calculations that State Farm already performed and relied on to pay the proposed class members for their totaled cars. At a minimum, the plaintiffs can rely on State Farm's own Audatex calculations as a 'starting place' to correctly measure injury. See Jama , 113 F.4th at 936. Indeed, nothing requires the “plaintiffs to undertake a separate valuation process or retain an expert to opine on the value of the loss vehicles.” Id. at 937 (citation modified). To be sure, State Farm could repudiate its own methodology at the classwide liability stage and assert its appraisal rights against other class members, as it did for Clippinger.¹ Or, of course, State Farm could choose instead to defend the TNA on the merits.

But the fact that Clippinger’s and the class members’ ultimate damages may differ because of a reference, or lack of reference, to appraisal costs does not mean that individual issues Page 38predominate. Even if damages issues must be tried separately, a common answer to whether the challenged adjustment is a breach of contract can be generated through evidence common to the class. See Pickett v. City of Cleveland , 140 F.4th 300, 310 – 11 (6th Cir. 2025). And just because “some members of [the proposed class] may not be eligible for damages, or may receive damages in varying amounts, does not invalidate the district court’s certification of the Class based on a comm on question.” Id. at 310. Clippinger and the class mount a direct, substantive challenge to the TNA by arguing that it is always negative and thus always results in an improperly reduced valuation for their cars. 'To decertify' here 'based on the existence of some questio ns on individual damages would undermine the very purpose of Rule 23(b)(3).” Id. at 311.

Indeed, for all of State Farm’s many assertions about the impossibility of class -wide proof, the very model that Clippinger proposes was successfully implemented in another district court. A certified class in Chadwick v. State Farm Mutal Automobile Insurance Co. challenged the same TNA at issue in this case. See generally No. 4:21-cv-1161-DPM, 2024 WL 1156944 (E.D. Ark. Mar. 18, 2024). The case was tried before a jury, and the jury returned a verdict for the class. Specifically, the jury found that “[b]y applying a ‘[TNA],’ . . . State Farm br[oke] its contract to pay Chadwick, and each class member, the ‘[ACV]’ (minus any applicable deductible) of their totaled vehicle,” and that Chadwick and the class members were not “estopped from disagreeing now wi th the ‘[ACV]’ calculated by State Farm.” Verdict at 1– 2, Chadwick , 2024 WL 1156944 (E.D. Ark. June 13, 2025), DE 248. The jury further found that 'the damages sustained by Chadwick[] and each class member” are “the difference between (a) the ‘[ACV]’ of the totaled vehicle as calculated by the Autosource Report, but removing any ‘[TNA]’ in that Report, and (b) the ‘[ACV]’ of the totaled vehicle as calculated by the Autosource Report, including any ‘[TNA]’ in that Report.” Id. at 3.

Following the liability verdict, the court noted that “in addition to hearing evidence about how Chadwick’s damages (and those of each member) could be calculated, the jury heard evidence about the numbers as to Chadwick. A similar calculation must now be done for each class member.” Order at 2, Chadwick , No. 4:21-cv-1161-DPM (E.D. Ark. July 17, 2025), DE 266. The district court helped structure document production for the damages phase and suggested that “[i]f motion practice and mini-trials are needed, ” it would “anticipate[]” such trials to be conducted Page 39over the course of two weeks, with many background stipulations, and with the help of some magistrate judges presiding. Id. at 3 – 4. The court recently granted preliminary approval of a proposed class settlement. Order at 1 – 16, Chadwick , No. 4:21-cv-1161-DPM, 2026 WL 852099 (E.D. Ark. Mar. 27, 2026), DE 288. Chadwick provides concrete proof of concept for Clippinger’s proposed class litigation.

B.

We acknowledge that in certain other circuits, insurance valuation class certifications have been defeated because of differing damage calculations that undermined predominance under Rule 23(b)(3). See Drummond v. Progressive Specialty Ins. Co. , 142 F.4th 149, 158 – 61 (3d Cir. 2025); Freeman v. Progressive Direct Ins. Co. , 149 F.4th 461, 468 – 71 (4th Cir. 2025); Sampson v. United Servs. Auto. Ass’n , 83 F.4th 414, 417, 421 – 23 (5th Cir. 2023); Schroeder v. Progressive Paloverde Ins. Co. , 146 F.4th 567, 576 – 78 (7th Cir. 2025); Ambrosio v. Progressive Preferred Ins. Co. , 154 F.4th 1107, 1110 – 13 (9th Cir. 2025). But several circuits have declined interlocutory review of the district court's class certification in contexts similar to this one. See Volino v. Progressive Cas. Ins. Co. , No. 21-6243, 2023 WL 2532836 (S.D.N.Y. Mar. 16, 2023), appeal denied , No. 23-472 (2d Cir. May 31, 2023); Chadwick , 2024 WL 1156944, appeal denied , No. 24-8004 (8th Cir. May 28, 2024); Curran v. Progressive Direct Ins. Co. , 345 F.R.D. 498 (D. Colo. 2023), appeal denied , No. 24-700 (10th Cir. Apr. 9, 2024); Gulick v. State Farm Mut. Auto. Ins. Co. , 780 F. Supp. 3d 1169 (D. Kan. 2025), appeal denied , No. 25-602 (10th Cir. July 22, 2025); Reynolds v. Progressive Direct Ins. Co. , 346 F.R.D. 120 (N.D. Ala. 2024), appeal denied , No. 24-90006 (11th Cir. June 14, 2024). And two of our sister circuits’ decisions were split, with dissents on those panels. See Ambrosio , 154 F.4th at 1113 – 23 (Wallach, J., dissenting); Freeman , 149 F.4th at 471 – 76 (Berner, J., dissenting).

The Ninth Circuit’s decisions in Jama and Lara are especially instructive. In Lara , the Ninth Circuit affirmed a district court’s refusal to certify a similar class on predominance grounds. 25 F.4th at 1140. Lara concerned a class of customers insured by Liberty Mutual in Washington state who argued that Liberty Mutual was breaching its insurance contracts with them, and breaching Washington trade practices law, when it applied a 'condition adjustment' to the calculated valuation of totaled cars and did not 'itemiz[e] or explain[]' the deduction. Id. at 1137.

Page 40Like State Farm’s contracts, the policies in Lara concerned the ACV of totaled cars and provided for a binding appraisal if a customer rejected an offer. Id. at 1136 – 37. But unlike here, the putative class could have included insured parties for whom an adjuster gave a valuation higher than the initial ACV calculation or people who underwent an appraisal. Id. at 1139.

Indeed, that adjustment merely represented the first step of the insurer’s ACV process. Id. at 1137. There, the insurer “also look[ed] at the actual pre -accident condition of the totaled car.” Id. If the car 'was in great condition, then [the Audatex -equivalent company] reverse[d] the negative adjustment and sometimes even applie[d] a positive adjustment.” Id. Thus, the Ninth Circuit concluded that the challenged “condition adjustment” did not uniformly depress value and that the class may have included individuals who received the ACV of their vehicle, notwithstanding the potential procedural violation of Washington law. Id. at 1138 – 40. It therefore held that the district court did not abuse its discretion in finding that classwide issues did not predominate. Id.

In Jama , confronting similar facts before Lara was decided, the district court had certified a class based on “condition” adjustments as well as “negotiation” adjustments. 113 F.4th at 926– 27. After the Ninth Circuit decided Lara , however, the district court decertified both classes, reasoning that Lara required the plaintiffs to 'demonstrate injury' in a way they had not done. Id. at 927. On appeal from the district court's order in Jama , the Ninth Circuit upheld the decertification of the “condition” class but held that the district court had abus ed its discretion in decertifying the “negotiation” class. Id.

The Ninth Circuit held that the negotiation class could prove and measure its injuries on a classwide basis because all members claimed they “received less than they were owed in the exact amount of the impermissible negotiation deduction.” Id. at 933.² While the condition adjustment could “have accurately reflected the condition of the car for some class members in Lara , there [was] no negotiation adjustment that could accurately price the negotiation discount [in Jama ] if Page 41Plaintiffs [were] correct that the adjustment is always unlawful, regardless of the amount.” Id. at 934.³ Indeed, the court emphasized that “ Lara rejected measuring injury based on a failure to 'follow the prescribed process' because a procedural violation does not necessarily lead to an incorrect result.” Id. at 935. And the condition adjustment, unlike the narrowed negotiation class in Jama , “included members who may not have been injured by the allegedly unlawful process.” Id. The negotiation class, by contrast, left only those plaintiffs “who (1) were paid based on the Autosource report, excluding those who negotiated or pursued an appraisal, and (2) were paid a negotiation adjustment that, according to the Plaintiffs, can never measure a lawful deduction.” Id. Moreover, Jama determined that the district court had been wrong to determine that 'the Autosource reports, and the amount of a challenged adjustment” could not be “evidence of value and injury'-even for the 'condition' adjustment. Id. at 936. The Ninth Circuit therefore concluded that the district court abused its discretion in decertifying the negotiation class. See id. at 935 – 36.

Here, as in Jama , Clippinger and the class argue that the TNA is always negative and thus always results in an improperly reduced valuation for their cars. Rather than asking State Farm merely to show its work, they claim that their valuations were too low because State Farm applied the TNA. Moreover, the class definition similarly includes only those whose “actual cash value was decreased based upon [TNAs] to the comparable vehicles used to determine actual cash value.” DE 202, Order on Class Cert., Pag e ID 6949. Like the virtually identical negotiation class in Jama , State Farm does not argue that any analogous counter-negotiation adjustment was made for class members. Instead, it argues much more broadly that the valuation process is simplyPage 42 “inherently . . . individualized,” which would seem to support a conclusion that the predominance inquiry could never be satisfied for any valuationbased class. CA6 R. 84, Supp. Appellant’s Br., at 1. Our decision in Hicks forecloses such a broad position. And defining the class to include only those insured parties for whom the TNA’s calculation and application reduced their ACV figures properly focuses the inquiry on those affected by the challenged adjustment.⁴

While we recognize that the Ninth Circuit has since narrowed Jama in Ambrosio , we disagree with Ambrosio 's attempt to reconcile Lara 's and Jama 's holdings. In Ambrosio , the Ninth Circuit affirmed a district court's finding that individual questions predominated where the plaintiffs challenged Progressive’s use of a “projected sold adjustment” (PSA), which, like the TNA here, involved a “reduction to the list prices of comparable vehicles to ‘reflect consumer purchasing behavior (negotiating a different pri ce than the listed price).’” 154 F.4th at 1108. The court held that “the key factual difference” between Lara and Jama was that “there [was] nothing facially unlawful about Progressive’s use of the PSA.” Id. at 1111. It concluded that “[w]ithout any evidence that the PSA is disallowed on its face, its mere existence is not common evidence of liability on its own.' Id. This differed from the negotiation adjustment in Jama , which the district court had found was prohibited by Washington state law. Id. The court therefore held that Progressive's use of the PSA could not serve as common evidence of liability because the insurance policy did not explicitly prohibit its application and “the existence of the PSA [did] not necessarily indicate measurable damages, which is required to prove a breach of contract in Arizona.' Id. As a result, the Ninth Circuit decided that the district court did not abuse its discretion in concluding that individual questions predominated. Id. at 1112 – 13.

Page 43We read Jama and Lara , in contrast, to permit class certification 'when the challenged adjustment categorically results in all class members receiving less than the [ACV].” Freeman , 149 F.4th at 475 (Berner, J., dissenting). Class certification is therefore appropriate, as Clippinger asserts here, where the insurer’s application of an artificial adjustment “ necessarily results in a class member receiving less than the [ACV].” Id. We do not read Lara and Jama like Ambrosio does, which reading would effectively permit class certification only when the disputed adjustment “is categorically barred by law.” Ambrosio , 154 F.4th at 1111.

The consequence of the majority opinion's contrary holding is that similarly situated classes may be foreclosed from certification on predominance grounds in even the most outrageous cases. Indeed, under the majority’s approach, there is “no articulable l imit (if any) to what methods are permitted beyond what is provided for in the Contract.' Ambrosio , 154 F.4th at 1120 (Wallach, J., dissenting). Under the majority's view, an insurer could avoid class certification where it calculated ACV by applying “an ‘Artificial Deduction – a deduction taken without factual basis to pay you less money’ in any state that does not foreclose such a deduction by regulation.” Id. Or, as State Farm concedes, the majority’s approach could prevent class certification even when, for example, an insurer calculated ACV by employing orangutans to pick valuations or by throwing darts at randomly placed valuations on a dartboard.

What is more, even accepting Ambrosio 's method of reconciling Jama and Lara , Clippinger has also asserted that State Farm’s application of the TNA was facially unlawful.⁵ Under Tennessee law, State Farm was required to calculate ACV based on the “actual cost . . . to purchase a comparable automobile.” Tenn. Comp. R. & Regs. § 0780 -01-05-.09(1)(b). Because State Farm chose the “general source” method for calculating ACV, i ts methodology needed to, among other things, actually “produce fair market values” based on current and local data in the area where the vehicle was primarily garaged. Id. § 0780-1-5.09(1)(b)(4)(iii). And here, Page 44Clippinger has presented evidence to suggest that the TNA is calculated using manipulated data that categorically decreases payouts below ACV and is therefore invalid in breach of state law.

Thus, like the plaintiffs in Jama , Clippinger 'contend[s] that [Tennessee] law flatly prohibits any negotiation adjustment; and if Plaintiffs are correct about that legal issue, then each Plaintiff suffered damages equal to the amount of the negotiation adjustment that State Farm made.” 113 F.4th at 932. Clippinger similarly asks us “to credit . . . t he whole [Audatex] report, minus one specific uniformly applied downward adjustment' that she says is categorically unlawful. Id. at 934 (footnote omitted). At bottom, Clippinger’s “argument is that State Farm accurately estimated the actual cash value of [the class’s] vehicles based on several permissible inputs and then applied one further subtraction that [Tennessee] law entirely forbids.” Id. at 935.

It is true, as the majority opinion says, that we had already determined that the relevant labor depreciation in Hicks was unlawful in one of our previous decisions. See Hicks , 751 F. App’x at 711; Maj. Op. at 25. But the plaintiffs are attempting to establish something very similar in this case: that the application of the TNA constitutes both a breach of contract and a violation of Tennessee regulations.⁶ We would still follow Hicks . Nor is there any reason that this liability question must have been answered in a separate proceeding to allow class certification. See Jama , 113 F.4th at 933 (establishing illegality of the negotiation adjustment under Washington law during district court proceeding in same case).

Moreover, we review a district court's decision to certify a class for abuse of discretion. Hicks , 965 F.3d at 457. A district court has 'substantial discretion in determining Page 45whether to certify a class, as it possesses the inherent power to manage and control its own pending litigation.” Rikos v. Procter & Gamble Co. , 799 F.3d 497, 504 (6th Cir. 2015) (quoting Beattie v. CenturyTel, Inc. , 511 F.3d 554, 559 –60 (6th Cir. 2007)). Thus, a district court’s certification of a “class is subject to a very limited review and will be reversed only upon a strong showing that the district court’s decision was a clear abuse of discretion.” Randleman v. Fid. Nat’l Title Ins. Co. , 646 F.3d 347, 351 (6th Cir. 2011) (quoting Beattie , 511 F.3d at 559 –60). And “when a matter is entrusted to the district court's discretion, 'it is possible for two judges, confronted with the identical record, to come to opposite conclusions and for the appellate court to affirm both.'' United States v. Hutchinson , 573 F.3d 1011, 1030 (10th Cir. 2009) (Gorsuch, J.) (citation modified) (quoting United States v. Williams , 81 F.3d 1434, 1437 (7th Cir. 1996)). A decision affirming a district court’s refusal to certi fy a class, by itself, does not necessarily indicate that the district court would have abused its discretion had it certified the class.

Nor does the Rules Enabling Act pose any barrier to class certification. To the majority opinion, class treatment abridges State Farm’s right to present statutory defenses. Maj. Op. at 21– 23 (citing Wal-Mart , 564 U.S. at 367); see also Sampson , 83 F.4th at 420. Wal-Mart , however, was referring to distinct statutory defenses under Title VII's 'detailed remedial scheme' for employment discrimination, as well as the Supreme Court's correspondingly intricate burden -shifting approach. 564 U.S. at 366 – 67. St ate Farm’s potential defenses, on the other hand, would go to the argument that it could have paid an individual class member ACV by calculating it differently. Because State Farm can still present individualized valuation evidence in later damages procee dings, certifying this class would not abridge State Farm’s substantive rights in defending its valuation practices. See Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. Co. , 559 U.S. 393, 408 (2010) (plurality op.) (class treatment under Rule 23 “leaves the parties’ legal rights and duties intact and the rules of decision unchanged'); Tyson Foods , 577 U.S. at 453 (predominance requirement can still be satisfied, “even though other important matters will have to be tried separately, such as damages or some affirmative defenses peculiar to some individual class members”) (quoting 7AA Wright and Miller’s Federal Practice and Procedure § 1778 (3d ed. 2005)); Hicks , 965 F.3d at 463.

Page 46Clippinger’s theory of liability is that State Farm’s application of the TNA categorically reduced her vehicle’s valuation below ACV in breach of contract and in violation of Tennessee law. These common questions unite the class. To be sure, a jury is fr ee to reject Clippinger’s common proof that Audatex’s calculations, without the TNA, were the correct measure of “true” ACV, or that State Farm’s application of the TNA violated Tennessee law. But this does not change the fact that, for predominance purpo ses, Clippinger has shown that the class members’ injuries are susceptible to common proof. See In re Whirlpool , 722 F.3d at 860. Indeed, Clippinger's 'inability . . . to prove [her categorical allegation] would not result in individual questions predominating.” Rikos , 799 F.3d at 521 (citation modified) (quoting In re Whirlpool , 722 F.3d at 858). “Instead, a failure of proof on these issues would end the case.” Id. (citation modified). For the foregoing reasons, we respectfully dissent from the majority opinion and would conclude that the district court did not abuse its discretion by granting Clippinger’s motion for class certification.