No. 24-2708

D.C. No. 2:22-cv-00342SMB

Appeal from the United States District Court for the District of Arizona Susan M. Brnovich, District Judge, Presiding

Argued and Submitted April 4, 2025 Phoenix, Arizona

Filed September 12, 2025

Before: Michael Daly Hawkins, Evan J. Wallach, and Ryan D. Nelson, Circuit Judges. *

Page 2Opinion by Judge Hawkins; Dissent by Judge Wallach

SUMMARY **

The panel affirmed the district court’s order declining to certify a proposed class in an action brought by former customers of Progressive Preferred Insurance Company who made property-damage claims on their vehicles, which Progressive determined to be covered as total losses.

Under policy terms, Progressive was required to pay a totaled vehicle’s actual cash value (“ACV”), defined as “the market value, age, and condition of the vehicle at the time the loss occurs.” As part of its calculation of market value, Progressive used a “projected sold adjustment” (“PSA”)— a reduction to the list prices of comparable vehicles to “reflect consumer purchasing behavior (negotiating a different price than the listed price).”

Appellants alleged that Progressive's use of the PSA always resulted in an inherently flawed negative line-item adjustment, amounting to a breach of the express terms of Progressive’s uniform automobile insurance policy because it prevents an ACV from being determined by the true “market value” of the vehicle.

Page 3The district court found that Appellants satisfied their burden as to numerosity, commonality, typicality, and adequacy under Fed. R. Civ. P. 23(a), but that individual questions surrounding the calculation of each Appellant's ACV predominated for the purposes of Rule 23(b)(3).

Noting that the PSA is not facially unlawful, the panel addressed whether its implementation by Progressive could serve as common evidence of liability, and concluded that it cannot. As a result, Appellants faced the same problem that the prospective class members in Lara v. First National Insurance Company of America , 25 F.4th 1134 (9th Cir. 2022), faced: figuring out whether each individual putative class member was harmed would involve an inquiry specific to that person. Each individual Appellant would need to compare their flawed “market value” with a correct one to win on the merits. Therefore, common issues do not predominate over any questions affecting only individual members under Rule 23(b)(3), and class certification is inappropriate.

Federal Circuit Judge Wallach dissented. He wrote that the class should be certified because questions of law or fact common to class members predominate over questions regarding the requirement of the contract, the breach, and the resulting injury; and that the majority failed to reconcile in an articulable, principled fashion intra- and inter-circuit conflicts.

Page 4Lee Lowther (argued) and Hank Bates, Carney Bates & Pulliam PLLC, Little Rock, Arkansas; Jacob L. Phillips, Jacobson Phillips PLLC, Winter Park, Florida; for PlaintiffsAppellants.

Jeffrey S. Cashdan (argued), James M. Brigman, Zachary A. McEntyre, Erin Munger, and Allison H. White, King & Spalding LLP, Atlanta, Georgia; Nicole Bronnimann, King & Spalding LLP, Houston, Texas; Paul A. Mezzina and Amy R. Upshaw, King & Spalding LLP, Washington, D.C.; Julia B. Bates, King & Spalding LLP, Austin, Texas; Lawrence Kasten, Papetti Samuels Weiss McKirgan LLP, Scottsdale, Arizona; for Defendants-Appellees.

Adam G. Unikowsky, Jenner & Block LLP, Washington, D.C.; Jonathan D. Urick and Audrey Beck, U.S. Chamber Litigation Group, Washington, D.C.; Stephen Skardon, American Property Casualty Insurance Association, Chicago, Illinois; for Amici Curiae the Chamber of Commerce of the United States of America and the American Property Casualty Insurance Association.

Page 5HAWKINS, Circuit Judge:

At issue here is whether class certification is appropriate for challenging certain deductions in auto insurance claims, or whether individual questions predominate. Because the district court did not abuse its discretion in finding that individual questions in this case predominated, we affirm.

Elliott Ambrosio and the other named Plaintiffs (“Appellants”) are former customers of Progressive Preferred Insurance Company (“Progressive”). Appellants made property-damage claims on their vehicles, which Progressive determined to be covered as total losses (“totaled”). According to the policy terms, Progressive was required to pay the actual cash value ('ACV') of their totaled vehicles. The policies defined ACV as “the market value, age, and condition of the vehicle at the time the loss occurs.”

Progressive calculated the 'market value' of each vehicle by comparing it to similar vehicles that were for sale, or recently sold, in the insureds’ market area. Progressive conducted this comparison by using the WorkCenter TotalLoss system licensed from Mitchell International, Inc. (the “Mitchell Report”). As part of its calculation, the Mitchell Report factored in a “projected sold adjustment” (“PSA”). The PSA is a reduction to the list prices of comparable vehicles to “reflect consumer purchasing be havior (negotiating a different price than the listed price).” Progressive’s use of the PSA is the sole source of controversy in this suit.

Page 6Appellants allege that Progressive's use of the PSA always resulted in an inherently flawed negative line-item adjustment. For example, the PSA allegedly does not consider why a vehicle may have sold below a list price, such as for inapplicable military or promotional discounts. Yet, these types of sales were factored into the PSA calculation. Appellants further allege that when aggregating this data on vehicle sales, the Mitchell Report's sources intentionally omitted vehicles sold above the list price as “outliers.” This resulted in the PSA only relying on data which inaccurately skewed vehicle prices downward. Thus, according to Appellants, the use of the PSA amounts to a breach of the express terms of Progressive’s uniform automobile insurance policy because it prevents an ACV from being determined by the true “market value” of the vehicle.

As a result, Appellants sued Progressive for breach of contract, breach of the covenant of good faith and fair dealing, unjust enrichment, and declaratory relief, alleging that Progressive failed to pay the ACV of their totaled vehicles by utilizing the PSA in their valuation reports. Appellants sought class certification, defining the proposed class as:

All persons who made a first-party claim on a policy of insurance issued by Progressive Preferred Insurance Company or Progressive Advanced Insurance Company to an Arizona resident where the claim was submitted from March 4, 2016, through the date an order granting class certification is entered, and Progressive determined that the vehicle was a total loss and based its claim payment on an Instant Report from Mitchell where a

Page 7Projected Sold Adjustment was applied to at least one comparable vehicle.

The district court found Appellants had satisfied their burden as to numerosity, commonality, typicality, and adequacy under Rule 23(a). Specifically, the district court found that “the legitimacy of PSAs as a means of calculating ACV' was a question common to the class sufficient for Rule 23(a)(2). Nevertheless, the district court found that individual questions surrounding the calculation of each Appellant ’ s ACV predominated for the purposes of Rule 23(b)(3) and declined to certify the proposed class. Appellants timely appealed.

A denial of class certification is reviewed for abuse of discretion. Parra v. Bashas ’, Inc. , 536 F.3d 975, 977 (9th Cir. 2008). 'A district court applying the correct legal standard abuses its discretion only if it (1) relies on an improper factor, (2) omits a substantial factor, or (3) commits a clear error of judgment in weighing the correct mix of factors.” Sali v. Corona Reg’l Med. Ctr. , 909 F.3d 996, 1002 (9th Cir. 2018) (internal quotation marks omitted). Fact findings are reviewed 'under the clea rly erroneous standard, meaning we will reverse them only if they are (1) illogical, (2) implausible, or (3) without 'support in inferences that may be drawn from the record.’” Abdullah v. U.S. Sec. Assocs., Inc. , 731 F.3d 952, 956 (9th Cir. 2013) (quoting United States v. Hinkson , 585 F.3d 1247, 1262 (9th Cir. 2009)).

Rule 23(b)(3) of the Federal Rules of Civil Procedure allows for class certification if 'the court finds that thePage 8 questions of law or fact common to class members predominate over any questions affecting only individual members.' 'The predominance inquiry tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.' Tyson Foods, Inc. v. Bouaphakeo , 577 U.S. 442, 453 (2016) (internal quotation marks omitted) (quoting Amchem Prods., Inc. v. Windsor , 521 U.S. 591, 623 (1997)). Typically, individual “damage[s] calculations alone cannot defeat certification.” Pulaski & Middleman, LLC v. Google, Inc. , 802 F.3d 979, 986 (9th Cir. 2015) (quoting Yokoyama v. Midland Nat’l Life Ins. Co. , 594 F.3d 1087, 1094 (9th Cir. 2010)). Instead, the analysis is 'whether the common, aggregation -enabling, issues in the case are more prevalent or important than the non-common, aggregationdefeating, individual issues.” Tyson Foods, Inc. , 577 U.S. at 453 (quotation omitted ). “An individual question is one where 'members of a proposed class will need to present evidence that varies from member to membe r,' while 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, classwide proof.’” Id. (quoting 2 W. Rubenstein, Newberg on Class Actions § 4:50, 196 – 97 (5th ed. 2012)).

We have wrestled with this predominance inquiry as it relates to auto insurance policies, coming to different results. First, in Lara v. First National Insurance Company of America , 25 F.4th 1134 (9th Cir. 2022), an auto insurer applied a downward “condition adjustment” to account for the difference in price between the average condition of privately sold cars and those sold at dealerships. Id. at 1137. The plaintiffs disputed this adjustment as a breach of contract and argued that the insurer’s failure to itemize the Page 9deduction violated Washington insurance regulations. Id. We found class certification to be inappropriate because showing “injury will require an individualized determination for each plaintiff' which in turn predominated over any common questions. Id. at 1138. This was because the insurer “only owed each putative class member the actual cash value of his or her car” and “if a putative class member was given that amount or more, then he or she cannot win on the merits.” Id. at 1139. Regarding the lack of itemization as it related to state regulations, we determined that '[a] violation of the regulation isn't a breach. Breach of contract requires not just a violation of the terms of the contract but also an injury.” Id. (footnote omitted).

Two years later in Jama v. State Farm Mutual Automobile Insurance Company , 113 F.4th 924 (9th Cir. 2024), we were presented with a variation of the predominance question. There, a different auto insurer applied not only a “condition adjustment” like in Lara , but also a 'negotiation adjustment.' Jama , 113 F.4th at 927. This “negotiation adjustment” was designed to capture the price difference between the list price at a dealership and the actual negotiated price a customer might typically pay. Id. Jama determined class certification was appropriate there because liability for this claim was not predicated on individual assessments. Id. at 933. Unlike 'condition adjustments,” Washington state law explicitly prohibited the use of “negotiation adjustments.” Id. Thus, the dispositive issue was not that the insurer “failed to follow the correct procedure for making permissible adjustments' but that “Washington law [did] not permit [the insurer] to apply a discount for typical negotiation at all. ” Id. Therefore, the plaintiffs in Jama could measure their injuries on a classPage 10wide basis by adding back the amount of the facially unlawful deduction. Id.

Considering the above guidance from Lara and Jama , we must assess how the PSA functions here with an eye on whether individual questions predominate. At first glance the purpose of the PSA seems to resemble the “negotiation adjustment' in Jama . See 113 F.4th at 927. Both were designed to reflect the fact that cars usually sell at dealerships for less than their listed price. See id.

However, there is nothing facially unlawful about Progressive’s use of the PSA here. That was the key factual difference which Jama relied upon in differing from Lara . See id. at 933. Appellants have alleged that the use of the PSA always results in an unlawfully reduced 'market value,” but that only addresses the outcomes produced by the PSA, not whether it is categorically barred by law. Cf. Sali v. Corona Reg'l Med. Ctr. , 909 F.3d 996, 1011 (9th Cir. 2018) (“Because the Code specifies that a violation of § 226 is a per se injury, there is no individualized issue of damages.'). Without any evidence that the PSA is disallowed on its face, its mere existence is not common evidence of liability on its own. That makes this case more like Lara than Jama .

If the PSA is not facially unlawful, and thus controlled by Jama , then we must determine whether its implementation by Progressive could serve as common evidence of liability. Several facts suggest it cannot. First, the remainder of the insurance policy does not explicitly prohibit the use of the PSA. Appellants counter that elsewhere in the policy Progressive is required to calculate the ACV using the “market value” of the vehicle, which they Page 11claim cannot be done accurately while using the PSA. Again though, no further explanation of how “market value” must be calculated exists in the policy. See Schroeder v. Progressive Paloverde Ins. Co. , No. 24-1559, 2025 WL 2083855, at *5 (7th Cir. July 24, 2025) (analyzing identical policy language and concluding that use of the PSA was not facially barred by the contractual terms or state law). We cannot now read an unwritten requirement into the contract of how to calculate 'market value.' See Shattuck v. Precision-Toyota, Inc. , 566 P.2d 1332, 1334 (Ariz. 1977).

Second, the existence of the PSA does not necessarily indicate measurable damages, which is required to prove a breach of contract in Arizona. See Chartone, Inc. v. Bernini , 83 P.3d 1103, 1111 (Ariz. Ct. App. 2004) (“[I]n an action based on breach of contract, the plaintiff has the burden of proving the existence of a contract, breach of the contract, and resulting damages.”). If the appraisal from the Mitchell Report resul ted in a fair 'market value' assessment, even while using the PSA, then the ACV would be accurate, and there would be no injury. This would be fatal to a potential class member's claim. See Thomas v. Montelucia Villas, LLC , 302 P.3d 617, 621 (Ariz. 2013) ('the non -breaching party [must have] actually suffered injury from the anticipatory [breach]' (citation omitted)); see also Griffey v. Magellan Health Inc. , 562 F. Supp. 3d 34, 51 (D. Ariz. 2021) (“In a breach of contract claim, the plaintiff must demonstrate that the breach caused an injury.' (citing Thomas , 302 P.3d at 621)).

This is not a dispute over the amount of any individual’s damages, which would be insufficient to prevent class certification, see Pulaski , 802 F.3d at 988, but over an essential element of each individual Appellant’s claim, see Lara , 25 F.4th at 1139. See id. (noting that because the Page 12insurer “only owed each putative class member the actual cash value of his or her car, if a putative class member was given that amount or more, then he or she cannot win on the merits”) ; see also Drummond v. Progressive Specialty Ins. Co. , 142 F.4th 149, 159 (3d Cir. 2025) (“[B]ecause proving the underpayment issue is essential to Progressive’s alleged liability for breach of contract, we believe that ‘individualized evidence’ of whether each class member’ s vehicle’s ACV was greater than the final settleme nt value would ‘overwhelm the case.’” (citation omitted)).

Third, Progressive is entitled to “invoke individualized issues and provide sufficient evidence that the individualized issues bar recovery on at least some claims, thus raising the spectre of class-member-by-class-member adjudication of the issue.' Van v. LLR, Inc. , 61 F.4th 1053, 1067 (9th Cir. 2023) (citation omitted). What constitutes “market value” here is inherently a question of fact. See Stockholders & Spouses of Carioca Co. v. Superior Court , 687 P.2d 1261, 1263 (Ariz. 1984) (defining “market value” as “the price at which a willing seller will sell to a willing buyer”). Progressive has provided evidence that at least two members of the proposed class received a higher 'market value' valuation from the Mitchell Report than they would have from other sources. Allowing Progressive to mount this defense towards the remaining proposed class members would render class certification inappropriate. See Zinser v. Accufix Rsch. Inst., Inc. , 253 F.3d 1180, 1189 (9th Cir. 2001) (“[I]f the main issues in a case require the separate adjudication of each class member's individual claim or defense, a Rule 23(b)(3) action would be inappropriate. . . . ” (quotation omitted)). Yet, denying Progressive this defense altogether would seem to violate due process. See Sampson v. United Servs. Auto. Ass'n , 83 F.4th 414, 420 (5th Cir.

Page 132023) (“This creates an explosion of predominance issues because [the insurer] has the due process right to argue, for each individual plaintiff, that damages should be determined by a different legally permissible method that would produce lower damages than NA DA (or no damages at all).”).

Appellants argue that these other sources, such as the National Automobile Dealers Association ('NADA') or Kelley Blue Book ('KBB'), cannot be used to calculate ACV because they do not comport with Progressive's policy. This again reads unwritten requirements into the contract. Appellants ’ own expert states that 'there is no single methodology to arrive at a proper opinion as to a vehicle’s [ACV] . . . .” Moreover, the record indicates that the first step in a market value appraisal is to examine a nationally recognized valuation guide like NADA or KBB. Accordingly, Progressive's attempted use of NADA and KBB is not “without support . . . .” Van , 61 F.4th at 1062 (internal citation and quotation marks omitted). Given that this evidence could 'bar recovery on some claims , ' we “must determine, based on the particular facts of the case, ‘whether individualized questions . . . will overwhelm common ones and render class certification inappropriate under Rule 23(b)(3).'' Id. (quoting Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC , 31 F.4th 651, 669 (9th Cir. 2022) (en banc) (internal citation and quotation marks omitted)).

As a result, Appellants face the same problem that the prospective class members in Lara faced: 'figuring out whether each individual putative class member was harmed would involve an inquiry specific to that person.” Lara , 25 F.4th at 1139. Given that the PSA cannot serve as common evidence of liability, each individual Appellant would need to compare their flawed “market value” with a correct one toPage 14 win on the merits. This is the exact type of “aggregation -defeating, individual issue[]' on which 'members of a proposed class will need to present evidence that varies from member to member” that Rule 23(b)(3) intends to prevent. Tyson Foods, Inc. , 577 U.S. at 453 (quotation omitted). Therefore, it cannot be said that common issues “predominate over any questions affecting only individual members.” Fed. R. Civ. P. 23(b)(3). Class certification is inappropriate here.

There was no abuse of discretion in concluding that individual questions predominate over common ones.

AFFIRMED.

WALLACH, Circuit Judge, dissenting.

Predictability is the essence of the rule of law. If a standard insurance policy clause may wildly gyre, then policy holders cannot know, or even guess their rights. Without predictability, life is reduced to a state in which might makes right and force equals success. A state of nature is, of course, one in which Thomas Hobbes famously said life is nasty, brutish, and short.

Here, not only is there a conflict amongst the circuits, there is one within this circuit: one which the majority has failed to reconcile in an articulable, principled fashion and which leaves Progressive’s policyholders with no loadstone for a guide.

I respectfully dissent.

Page 15Progressive’s Arizona Auto Policy (“the Contract”) is at issue in this case. The district court held questions of law or fact regarding liability affecting only individual members “predominate” over any questions common to class members, Fed. R. Civ. P. 23(b)(3); Ambrosio v. Progressive Preferred Ins. Co. , No. 22-cv-342, 2024 WL 915184, at *7 – 8 (D. Ariz. Mar. 4, 2024), ER-13 – 15, and declined to certify the class.

“[W]e first review a class certification determination for legal error under a de novo standard.” Sali v. Corona Reg’l Med. Ctr. , 909 F.3d 996, 1002, 1011 (9th Cir. 2018) (reversing district court decision to decertify class and explaining the “district court’s determination that individual questions predominate in the claims of” the proposed class 'was premised on legal error').² '[I]f no legal error occurred, we will proceed to review the' district court's “decision for abuse of discretion” and “findings of fact underPage 16 the clearly erroneous standard.' Id. (first quoting Yokoyama , 594 F.3d at 1091; then quoting Abdullah v. U.S. Sec. Assocs., Inc. , 731 F.3d 952, 956 (9th Cir. 2013)).

III. Analysis

First, I discuss the requirements of the Contract under its plain text. Second, I discuss that the alleged breach by Progressive turns on a question common to the class because it turns on Progressive's methodology being unacceptable under the Contract common to all Plaintiffs. Third, I discuss the injury, which, as alleged, is common to all Plaintiffs: the onesided deduction from their cars’ actual -cash-value Mitchell values. I take issue with this. The majority believes there can be no predominately common injury because Progressive contends there may be alternative methods to value cars on an individual basis that would have left some Plaintiffs even worse off and therefore, uninjured by what Progressive actually paid them. Progressive never followed these alleged methods, and the alleged alternatives were not specifically provided for in the Contract; they were legal chimeras. The class should be certified for a trial.

Questions of law or fact common to class members predominate over questions regarding the requirement of the Contract, the breach, and the resulting injury. “Rule 23(b)(3) requires a showing that questions common to the class predominate, not that those questions will be answered, on the merits, in favor of the class.' Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds , 568 U.S. 455, 459 (2013). Unlike the district court and majority, I would not conclusively resolve the merits of this case at the class certification stage. Questions of law dispositive of this case may be resolved on a class-wide basis, at appropriate stages of this case.

The Contract provides in pertinent part:

Payments for loss to a covered auto, nonowned auto, or custom parts or equipment are subject to the following provisions: . . . The actual cash value is determined by the market value, age, and condition of the vehicle at the time the loss occurs.

ER-97 –98. The Contract defines actual cash value (“ACV”) owed to Plaintiff by articulating how the ACV 'is determined”—“The actual cash value is determined by the market value, age, and condition of the vehicle at the time the loss occurs.' (emphasis added).³ Under its terms, no other inputs can be used to “determine[]” ACV except for those identified in the Contract: 'market value, age, and condition of the vehicle at the time the loss occurs.' Progressive 'has no discretion to choose a system that underpay s ACV.” Plaintiffs -Appellants’ Supp. Br. at 9 n.1. “Its discretion is fettered by the language of the contract and 'constrained by [the insurer's] obligation to exercise its discretion in good faith.’” Plaintiffs -Appellants’ Supp. Br.

Page 18at 9 n.1 (alteration in original) (quoting Beaudry v. Ins. Co. of the W. , 50 P.3d 836, 841 (Ariz. Ct. App. 2002), as amended (July 26, 2002)).

Progressive’s alleged breach concerned the method by which it determined “actual cash value” under the Contract. The district court found that “Progressive used Mitchell, and its [] software system to make estimates on what a totaled vehicle's ACV should be.' Additionally, 'Progressive, through Mitchell's ACV methodology applies a Projected Sold Adjustment.' 'Until July 2021 the data used to calculate the Projected Sold Adjustment calculation did not include transactions where the list price was equal to the sold price, or where the sold price was greater than the listed price.” After July 2021, Projected Sold Adjustment (“PSA”) continued to be calculated using transactions where the sold price was less than the listed price, but it added into the calculation transactions where the sold price was equal to the listed price. However, the PSA deduction is only applied as a downward adjustment to the Mitchell value — it is never an upward adjustment.

In other words, Progressive uses the Mitchell value for ACV, and then Progressive puts a thumb on the scale against the insured by applying an artificial deduction. Data that is intentionally taken only from the negative side of the scale does not fit under what either the Contract or I recognize as “the market value, age, and condition of the vehicle.” The Contract’s requirement to “determine[]” ACV using certain inputs creates a relevant duty that turns on a method. The Seventh Circuit in Schroeder sug gested this would be 'a methodological duty,” in which “each class member must show that Progressive used an invalid method to calculate Page 19actual cash value — independent from the result of the calculation —to prove a breach.” Schroeder v. Progressive Paloverde Ins. Co. , No. 24-1559, --- F.4th ----, ----, 2025 WL 2083855, at *5 (7th Cir. July 24, 2025).⁴

Progressive acknowledges that PSA is solely a downward adjustment and excludes transactions where cars sold for above list price. DefendantsAppellees’ Answering Br. at 9 & n.1 ('Red Brief') (stating PSA 'excludes transactions where vehicles sold for more than their advertised prices,' and until 2021, 'excluded transactions where vehicles sold for list price”). Progressive’s theory is that the Contract allows them to apply a PSA deduction because it “reflect[s]” a “common -sense reality” that “many used ca rs sell for less than list price.' Red Br. at 7. Moreover, the district court concluded that Plaintiffs had presented evidence and argument that was sufficient 'to allege common injury — that class wide PSA use led to devalued ACVs” for the purposes of Rule 23(a)’sPage 20 “commonality” inquiry. ER -9. The majority, by contrast, erroneously implies that it is only “alleged” by the Plaintiffs (and perhaps disputed) whether PSA is a downward adjustment. Maj. Op. at 6 (“Appellants further allege that when aggregating this data on vehicle sales, the Mitchell Report’s sources intentionally omitted vehicles sold above the list price as ‘outliers.’”).

The majority presents the issue of breach as turning on “whether [the PSA] is categorically barred by law.” Maj. Op. at 10. The majority errs in its ostensible fruitless search for a “bar[] by law” in two respects.

First, the majority asserts, without a reasoned basis, that Progressive’s use of PSA is not “barred” because “this case [is] more like Lara than Jama .' Maj. Op. at 10; see also Maj. Op. at 13 – 14 (suggesting that if “the PSA cannot serve as common evidence of liability,' it causes 'the same problem that the prospective class members in Lara faced”). Even the Seventh Circuit in Schroeder — with an overall holding amenable to the majority — would not stretch Ninth Circuit precedent as far as the majority here. Schroeder recognized that in both Lara and Jama , the Ninth Circuit “assumed or held that an insurance policy contains a promise by an insurer to calculate covered losses in accordance with a prescribed method or formula — independent of a promise by the insurer to pay for covered losses.” Schroeder , --- F.4th at ----, 2025 WL 2083855, at *8 (first citing Jama v. State Farm Mut. Auto. Ins. Co. , 113 F.4th 924, 933 (9th Cir. 2024), cert. denied , No. 24-933, 2025 WL 1678989 (U.S. June 16, 2025); then citing La ra v. First Nat’l Ins. Co. of Am. , 25 F.4th 1134, 1139 & n.4 (9th Cir. 2022)). The Seventh Circuit noted that in Lara specifically, the Ninth Circuit had 'assum[ed] that proof of an insurer's failure to itemize a condition adjustment, in violation of Page 21Washington insurance regulations, sufficed to establish the breach element of the plaintiffs’ breach -ofcontract claims.” Id. (citing Lara , 25 F.4th at 1139 & n.4).⁵ The majority’s quotation to Lara deletes footnote 4 although Schroeder cited and discussed that footnote. Compare id. (citing Lara , 25 F.4th at 1139 & n.4), with Maj. Op. at 9 (quoting Lara , 25 F.4th at 1139).

Second, the majority ignores that Arizona law recognizes contracts create enforceable obligations. “[I]mplicit in the contract and the relationship is the insurer's obligation to play fairly with its insured.' Rawlings v. Apodaca , 726 P.2d 565, 570 –71 (Ariz. 1986). “Common evidence Progressive exercised its discretion to always reduce ACV payments with baseless negotiation deductions also supports Plaintiffs’ claim for breach of the implied covenant of good faith and fair dealing.” Plaintiffs -Appellan ts’ Supp. Br. at 9 n.1 (citing Garza v. Gama , 379 P.3d 1004, 1008 (Ariz. Ct. App. 2016) (reversing order decertifying a class because the “claim for breach of the duty of good faith and fair dealing raises an issue common to all the members of the class”)). If Progressive’s adhesion contractual lan guage offers no certainty to its policy holders, then the role of courts is to protect those consumers against it.⁶

Questions of law or fact common to class members predominate about whether Progressive's breach of the Contract caused an injury to the Plaintiffs. The PSA deductions are all downward adjustments to the Mitchell Value, even though the PSA deductions may vary across individuals.⁷ “[E]ven if there is variation in the amount of each class members’ damages, this is an insufficient basis by itself to deny certification.” Sali , 909 F.3d at 1011.

The majority holds that actual breach by underpayment is not an injury if hypothetical performance through another method could have resulted in even less payment to Plaintiffs. Under the majority's view, it is acceptable for Defendants-Appellees to breach the insurance Contract by using Mitchell-minus-PSA, so long as Mitchell-minus-PSA is a higher claim payment than would have resulted from the values in certain guidebooks — National Automobile Dealers Association (“NADA”) and Kelly Blue Book (“KBB”). It i s error to conclude that the Contract definitively allowed NADA and KBB and that this eliminated any common Page 23injury. See Maj. Op. at 13 – 14 ('PSA cannot serve as common evidence of liability . . . . ”).⁸

The district court erred at the class certification stage by construing the Contract to conclusively permit using NADA and KBB to calculate ACV. There was insufficient evidence to do so. The district court conclusorily stated, “there are at least two other alternative methodologies for providing an ACV estimate, including the National Automobile Dealers Association (‘NADA’) and Kelly Blue Book (‘KBB’),” ER -3, and “[t]here are at least two other ways of estimating ACV —NADA and KBB,” ER -14.⁹ Progressive has not offered any of its own evidence, and Plaintiff has disputed, that NADA and KBB can be used to determine 'actual cash value” under the Contract. The majority supports the district court by relying on its shred of 'evidence,' which the m ajority refers to as “the record indicat[ing] that the first step Page 25in a market value appraisal is to examine a nationally recognized valuation guide like NADA or KBB.” Maj. Op. at 13; SER-41. This is a cherrypicked citation to Plaintiffs’ expert’s deposition. Plaintiffs’ expert also stated in the same deposition that NADA and KBB are “only a guide,” that do not take into account individual characteristics of the vehicles. SER-44. The same expert also reiterated in one of his reports that NADA and KBB do not provide valuations for specific vehicles or consider geographical location of the vehicle. ER-126 – 27; see also PlaintiffsAppellants’ Opening Br. at 42 ('Blue Brief'); Plaintiffs -Appellants’ Reply Br. at 23. The majority’s conclusion on this critical point, therefore, relies solely on a statement taken out of context and contradicted by other evidence.¹⁰

Under Arizona law, “a court will attempt to enforce a contract according to the parties' intent,' Taylor v. State Farm Mut. Auto. Ins. , 854 P.2d 1134, 1138 (Ariz. 1993), which generally happens in one of three ways. First, if a contract’s language is plain and unambiguous on its face, its meaning is determined from the four corners of the instrument. In re Mortgs. Ltd. , 559 B.R. 508, 517 (Bankr. D. Ariz. 2016) (citing Taylor , 854 P.2d at 1138 – 39). Second, to determine the meaning intended by the parties, a court may consider extrinsic evidence and surrounding Page 26circumstances, including negotiation, prior understandings, and subsequent conduct. Pure Wafer Inc. v. City of Prescott , 845 F.3d 943, 954 – 55 (9th Cir. 2017) (citing Taylor , 854 P.2d at 1139 – 40). Third, if contract language is reasonably susceptible to more than one interpretation, the intent of the parties is a question of fact left to the fact finder. Chopin v. Chopin , 232 P.3d 99, 102 (Ariz. Ct. App. 2010) (citing Taylor , 854 P.2d at 1144 – 45). It is unclear which method the district court used, or why it determined this was not a question for the factfinder, because it provided no analysis — that in itself is an abuse of discretion.

The district court should have held that questions of law or fact common to class members predominate interpretation of the Contract on the question of whether NADA or KBB are permissible to calculate ACV. It should have certified the class and performed the Contract interpretation at the correct stage — that is, either through interpretation by the court of the Contract in summary judgment, or in a trial before a factfinder to consider a factual controversy on the parties’ intent. The class should have bee n certified and the district court should have answered this question through standard contract interpretation methods under Arizona law.¹¹

Page 27If extrinsic evidence were considered, it would pointedly weigh against NADA or KBB falling under the parties' intent in the Contract. Progressive has never used either NADA or KBB under the Arizona Policy. Progressive states in its Red Brief that if the class was certified: '[t]he jury would need to consider any number of alternative methodologies for calculating ACV.' (emphasis added). However, Progressive does not state how many alternative methodologies the jury would need to consider; indeed, Progressive does not identify any . There is no proof that Progressive ever used NADA or KBB, and the only evidence offered by the majority is a vague, out-of-context, and contradicted statement by PlaintiffsAppellants’ own expert, which does not directly address whether NADA or KBB are usable to calculate ACV.

Unless and until proper analysis demonstrates the Contract allowed them, NADA and KBB cannot be considered alternatives to compare against the MitchellminusPSA for purposes of defeating Plaintiffs' common injuries. The district court did not consider contractual interpretation or reasonable reliance but rather, like the majority here, assumed that Progressive’s interpretation of the Contract as inclusive of NADA or KBB was correct. Unless and until proper contract interpretation demonstrates otherwise, the triable, injury issues are predominantly common across the class. “[E]ach class member was injured Page 28in precisely the same manner” by the PSA deduction applied to their vehicles' Mitchell valuation. See Sali , 909 F.3d at 1011. “[E]ven if there is variation in the amount of each class members’ damages, this is an insufficient basis by itself to deny certification.' Id. The district court abused its discretion by determining that the class was insufficiently cohesive. Id. at 1011 – 12 (reversing denial of class certification, determining the district court abused its discretion).

The core error in the district court's analysis is that it believes NADA and KBB are conclusively permitted by the Contract, and that accordingly they are also alternatives that must be compared for each individual Plaintiff against the Mitchell-minus-PSA values, for purposes of determining if individual Plaintiffs were injured. Even accepting hypothetically that NADA and KBB are permitted by the Contract, there is still predominance of common questions in Plaintiffs’ claims.

I disagree with the majority that, in effect, because Progressive theoretically could have traveled back in time to when it determined the ACV of a Plaintiff’s vehicle and used a NADA or KBB guidebook estimate instead of a Mitchellminus-PSA market valuati on, “relative privation” ceases to be privation. I share the view of other courts, that an injury does not cease to be an injury just because another injury could be worse. Reynolds v. Progressive Direct Ins. Co. , No. 22-cv-503, 346 F.R.D. 120, 132 (N.D. Ala. 2024)Page 29 (rejecting this defense because '[y]es, Progressive could have harmed members more had it used some other method of determining ACV. But the possibility of greater harm under the NADA standard does not imply a reciprocal benefit from using the PSA; relative privation is still privation.”). Arizona law guarantees that “a fair trial could be given on the issue of damages alone” in front of a jury, and so Plaintiffs at least should have been certified to proceed to trial on the common injury question. Chartone, Inc. v. Bernini , 83 P.3d 1103, 1114 (Ariz. Ct. App. 2004) (quoting Martinez v. Schneider Enters., Inc. , 873 P.2d 684, 687 (Ariz. Ct. App. 1994)).

Under Arizona law, an insurer has 'some duties of a fiduciary nature,” including “[e]qual consideration, fairness and honesty.' Zilisch v. State Farm Mut. Auto. Ins. Co. , 995 P.2d 276, 279 (Ariz. 2000) (alteration in original) (quoting Rawlings , 726 P.2d at 571). I agree with PlaintiffsAppellant's articulation in their Blue Brief, that even if NADA and KBB were contractually permissible alternatives, Plaintiffs have an issue to contest as a class at trial: that Progressive may, consistent with its fiduciary-like duties, defend itself for making a “rigged deduction” (PSA) by simply “scroung[ing] around for an even lower non -ACV estimate.” Bad faith, which could have been worse is still bad faith.

A problem with the majority's view is that it has no articulable limit (if any) to what methods are permitted beyond what is provided for in the Contract. Under the majority’s view, any method that results in any number of Plaintiffs receiving less ACV than the Mitchell-minus-PSA warrants a denial of class certification because Progressive hypothetically could have used that method and injured those Plaintiffs even more. Plaintiffs-Appellants echo the Page 30same problem — that if Progressive could have used a monkey throwing darts at a dartboard, and the dartboard’s values sometimes were lower than the Mitchell Values for PSA, that would defeat class certification in the majority’s view. Similarly, as Plaintiffs-Appellants argue, if Progressive’s sweeping interpretation prevailed, “then tomorrow it could simply replace the PSA deduction with an 'Artificial Deduction a deduction taken without factual basis to pay you less money' in any state that does not fore close such a deduction by regulation.”

Several Circuit Courts of Appeals have declined to block trial on the merits of Progressive’s use of PSA by denying permissive interlocutory review of the district court’s class certification. E.g. , Reynolds , 346 F.R.D. 120, appeal denied , No. 24-90006 (11th Cir. June 14, 2024); Curran v. Progressive Direct Ins. Co. , No. 22-cv-878, 345 F.R.D. 498 (D. Colo. 2023), appeal denied , No. 24-700 (10th Cir. Apr. 9, 2024).

There are numerous additional cases from U.S. district courts around the country that have certified classes challenging Progressive’s PSA deductions. E.g. , Costello v. Mountain Laurel Assurance Co. , No. 22-cv-35, 2024 WL 239849, at *14, *21 (E.D. Tenn. Jan. 22, 2024) (R&R); 12 Volino v. Progressive Cas. Ins. Co. , No. 21-cv-6243, 2023Page 31 WL 2532836, at *9, *12 – 13 (S.D.N.Y. Mar. 16, 2023); Brown v. Progressive Mountain Ins. Co. , No. 21-cv-175, 2023 WL 7219499, at *11 (N.D. Ga. Aug. 3, 2023). An Ohio Court of Appeals affirmed a state trial court’s granting class certification in a challenge to Progressive's PSA. Davenport v. Progressive Direct Ins. , No. 114306, 2025 WL 1902437 (Ohio Ct. App. July 10, 2025); see also ECF No. 66 (PlaintiffsAppellants’ 28(j) Letter discussing Davenport ); ECF No. 67 (Progressive’s 28(j) Letter asserting Davenport was “wrongly decided”).

The majority has chosen to resolve this appeal without even acknowledging that courts around the country have supported plaintiffs challenging Progressive’s use of PSA. I recognize that the Third, Seventh, and Fourth Circuits have granted permissive interlocutory review and decertified classes on this issue. Drummond v. Progressive Specialty Ins. Co. , No. 21-cv-4479, 2023 WL 5181596, at *5 – 6, *10 – 12 (E.D. Pa. Aug. 11, 2023), reversed , 142 F.4th at 152; Schroeder v. Progressive Paloverde Ins. Co. , 713 F. Supp. 3d 523, 541 (S.D. Ind. 2024), reversed , --- F.4th at ----, 2025 WL 2083855, at *1, *3; Freeman v. Progressive Direct Ins. Co. , 733 F. Supp. 3d 463, 487 (D.S.C. 2024), reversed , --- F.4th at ----, 2025 WL 2435366, at *1. There was a dissent on one of those panels. Freeman , --- F.4th at ----, 2025 WL 2435366, at *8 – 12 (Berner, J., dissenting) (dissenting from reversing class certification). As I noted, they starkly contrast with the Tenth and Eleventh Circuits’ denial of permissive interlocutory review of the district courts' class certification . Curran , No. 24-700 (10th Cir. Apr. 9, 2024); Reynolds , No. 24-90006 (11th Cir. June 14, 2024). The majority declines to address this incipient Circuit split. The majority has, thus, all but ensured future splits between the Ninth Circuit and other Courts of Appeals. For Page 32that reason, among others, this case would be a good choice for en banc review.

The majority misapplies this Circuit's precedent in Lara , Jama , and Olean . In doing so, the majority adds to an intracircuit split that can only be resolved by the Ninth Circuit taking this case en banc .

The Lara case is distinguishable. The majority's discussion of Lara elides critical details. See, e.g. , Maj. Op. at 8 – 9. Plaintiffs in that case were merely challenging an insurer’s failure to itemize deductions such that they could be verified as required by Washington state law. See Lara , 25 F.4th at 1137. The alleged injury was not that the condition adjustment was incorrect, but that the lack of disclosure made the condition adjustment 'impossible to verify.' Plaintiffs -Appellants' Supp. Br. at 13 (quoting Lara , 25 F.4th at 1137 (“Plaintiffs’ theory of the case is that Liberty violates Washington’s insurance regulations by not itemizing or explaining this downward ‘condition adjustment,’ which makes it impossible to verify.”)).

In Lara , those plaintiffs challenged an unverifiable and non-itemized vehicle condition adjustment that could affect a valuation in either direction, arguing it violated technical state insurance regulations. 25 F.4th at 1137. In Lara , the statute required that “the insurer only has to pay the ‘actual cash value' of the car,' which the applicable Washington statute defined as 'the 'fair market value.'' Id. at 1136Page 33 (quoting Wash. Admin. Code § 284-30-320(1)). Unlike in Lara , the Contract in this case provides specifically for how “actual cash value” is to be “determined.” Unlike here, “[n]o one in Lara disputed that Liberty could lawfully have applied a properly itemized and verifiable condition adjustment to calculate putative class members’ actual cash value.” See Jama , 113 F.4th at 933 (citing Lara , 25 F.4th 1137, 1139).

In Jama , Washington state law “permit[ted] insurers to apply only those deductions explicitly laid out in [a state law] and no others.” Id. The challenged conduct in Jama was a 'negotiation' discount meant to capture the typical amount buyers may negotiate down the price of a replacement car. Jama determined a socalled “negotiation class” was amenable to be certified in that case.

Although this case is not governed by a specific Arizona statute, it is governed by a contract, which requires that “[t]he actual cash value is determined by the market value, age, and condition of the vehicle at the time the loss occurs.” Just as the statute in Jama only allowed for certain deductions, so too, are the deductions available under the Contract limited to only include deductions that are “market value, age, and condition of the vehicle at the time the loss occurs.' The negotiation class in Jama included class members who were paid the value determined in an Autosource report with the negotiation discount applied, which took a deduction/adjustment for “typical negotiation.” Jama , 113 F.4th at 928, 931. Defining the class to include only individuals harmed by the challenged conduct, here: the onesided, PSA deduction from Mitchell, 'suffice[s] by itself to prevent many of the situations we discussed at length in Lara where class members might not have been injured by the putatively unlawful ad justment.” Id. at 932.

Page 34Even the Seventh Circuit has recognized an existing intra-circuit split between Lara and Jama in the Ninth Circuit regarding 'the evidence required for an insured to prove the actual cash value of her property.” Schroeder , --- F.4th at ----, 2025 WL 2083855, at *7. Schroeder recognized Jama as reasoning that whether each putative class member received the actual cash value of her car could be determined “by adding back to the value of their vehicles as calculated in the [insurer’s valuation] reports the a mount of the [challenged] discount,' while according to Lara it would involve a personspecific inquiry of “looking into the actual pre-accident value of the car and then comparing that with what each person was offered.' Id. (alteration in original) (first citing Jama , 113 F.4th at 933; then citing Lara , 25 F.4th at 1139). For that reason, Schroeder declined to follow the Ninth Circuit’s “different view of the evidence” to prove ACV. Id.

The Freeman case also shows that other courts may draw conflicting views from the Ninth Circuit precedent. The Freeman majority cited Lara for support and said nothing of Jama . --- F.4th at ----, 2025 WL 2435366, at *7. The Freeman dissent argued that if the Projected Sold Adjustment is an improper deduction, then “it categorically causes Progressive to pay less than actual cash value.” --- F.4th at ----, 2025 WL 2435366, at *11 (Berner, J., dissenting). The Freeman dissent said the Ninth Circuit has “endorsed this view,” and the “takeaway from Jama and Lara is that class certification is appropriate when the challenged adjustment categorically results in all class members receiving less than the actual cash value.' Id. at *11 – 12 (Berner, J., dissenting).

Page 35Although this panel could not resolve the Ninth Circuit split, the majority fails to advance the resolution of this issue to what is ultimately required: an en banc proceeding.

Additionally, the district court and the majority fail to reconcile the Ninth Circuit's en banc decision in Olean , which held that Rule 23 permits certification of a class even when the class “potentially includes more than a de minimis number of uninjured class members.' Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC , 31 F.4th 651, 669 (9th Cir. 2022). Rule 23’s predominance requirement is not to be used as an end-run around Olean because Olean dealt with that very issue. Id. Olean ’s holding, and the dissent it overcame, dealt directly with whether 'predominance' required no more than a de minimis number of uninjured members. Id. ; id. at 691 – 92 (Lee, J., dissenting). The majority’s focus on defeating predominance by positing that a certain number of Plaintiffs were potentially not injured shortcircuits the Ninth Circuit’s en banc Olean holding.

The parties and the law would have been better served if this panel had addressed the intra-circuit split, and the incipient inter-circuit split, both for the benefits of their analyses, and for the certainty of the results. I respectfully dissent.