Full opinion text
III. Discussion A. Certification Before certifying a class, the district court must assure itself that the proposed class action satisfies four prerequisites: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class. Fed. R. Civ. P. 23(a). In addition to meeting the numerosity, commonality, typicality, and adequacy prerequisites, the class action must fall within one of the three types specified in Rule 23(b). Here, the district court certified the class under Rule 23(b)(3), which requires that "questions of law or fact common to class members" must "predominate over any questions affecting only individual members," and the class action must be "superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3). The district court's Rule 23(a) and (b) analysis must be "rigorous." Comcast Corp. v. Behrend , 569 U.S. 27, 33, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013) (quoting Wal-Mart Stores, Inc. v. Dukes , 564 U.S. 338, 351, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011) ). The criteria for class certification are applied differently in litigation classes and settlement classes. In deciding whether to certify a litigation class, a district court must be concerned with manageability at trial. However, such manageability is not a concern in certifying a settlement class where, by definition, there will be no trial. On the other hand, in deciding whether to certify a settlement class, a district court must give heightened attention to the definition of the class or subclasses. Amchem Prods., Inc. v. Windsor , 521 U.S. 591, 620, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997). The Supreme Court specifically addressed the difference between litigation and settlement classes in Amchem . The Court wrote: Confronted with a request for settlement-only class certification, a district court need not inquire whether the case, if tried, would present intractable management problems, see Fed. Rule Civ. Proc. 23(b)(3)(D), for the proposal is that there be no trial. But other specifications of the Rule-those designed to protect absentees by blocking unwarranted or overbroad class definitions-demand undiluted, even heightened, attention in the settlement context. Such attention is of vital importance, for a court asked to certify a settlement class will lack the opportunity, present when a case is litigated, to adjust the class, informed by the proceedings as they unfold. We addressed concerns about definitions of settlement classes and fairness of proposed settlements in Hanlon v. Chrysler Corp. , 150 F.3d 1011, 1021 (9th Cir. 1998) : District courts must be skeptical of some settlement agreements put before them because they are presented with a "bargain proffered for ... approval without benefit of an adversarial investigation." [ Amchem , 521 U.S. at 621, 117 S.Ct. 2231 ]. These concerns warrant special attention when the record suggests that settlement is driven by fees; that is, when counsel receive a disproportionate distribution of the settlement, or when the class receives no monetary distribution but class counsel are amply rewarded. In the case before us, however, we need not analyze all of those criteria, for objectors challenge only the district court's findings regarding the predominance of common factual or legal issues under Rule 23(b)(3) and adequacy of representation under Rule 23(a)(4). We address those findings in turn. 1. Predominance The predominance inquiry under Rule 23(b)(3) "tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation." Amchem , 521 U.S. at 623, 117 S.Ct. 2231. It "presumes that the existence of common issues of fact or law have been established pursuant to Rule 23(a)(2)," and focuses on whether the "common questions present a significant aspect of the case and they can be resolved for all members of the class in a single adjudication"; if so, "there is clear justification for handling the dispute on a representative rather than on an individual basis." Hanlon , 150 F.3d at 1022 (quoting 7A Charles Alan Wright et al., Federal Practice & Procedure § 1777 (2d ed. 1986) ). "Predominance is not, however, a matter of nose-counting. Rather, more important questions apt to drive the resolution of the litigation are given more weight in the predominance analysis over individualized questions which are of considerably less significance to the claims of the class." Torres , 835 F.3d at 1134 (internal citation omitted). Therefore, even if just one common question predominates, "the action may be considered proper under Rule 23(b)(3) even though other important matters will have to be tried separately." Tyson Foods, Inc. v. Bouaphakeo , --- U.S. ----, 136 S. Ct. 1036, 1045, 194 L.Ed.2d 124 (2016) (quoting 7AA Charles Alan Wright et al., Federal Practice and Procedure § 1778 (3d ed. 2005) ). Rule 23(b)(3) lists four non-exclusive factors "pertinent" to a predominance finding: (A) the class members' interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action. These factors must be considered in light of the reason for which certification is sought-litigation or settlement-which "is relevant to a class certification." Amchem , 521 U.S. at 619, 117 S.Ct. 2231. As noted above, in deciding whether to certify a settlement-only class, "a district court need not inquire whether the case, if tried, would present intractable management problems." Id. at 620, 117 S.Ct. 2231. At the same time, a proposal to certify a settlement class presents other concerns-the risk of collusion chief among them-that "demand undiluted, even heightened, attention" by the district court. Id. The adversarial nature of a trial ensures that class definitions will be tested and allows the district court "to adjust the class, informed by the proceedings as they unfold." Id. A settlement lacks these safeguards. Therefore, the aspects of Rule 23(a) and (b) that are important to certifying a settlement class are "those designed to protect absentees by blocking unwarranted or overbroad class definitions." Id. The focus is "on whether a proposed class has sufficient unity so that absent members can fairly be bound by decisions of class representatives." Id. at 621, 117 S.Ct. 2231. Objectors Peri Fetsch and Dana Roland dispute that settlement plays any role in the predominance inquiry, arguing that the test is "precisely the same for a settlement class as it [is] for a litigation class." However, they misunderstand both Amchem and our statement in Hanlon that "[s]ettlement benefits cannot form part of a Rule 23(b)(3) analysis." 150 F.3d at 1022 (emphasis added). Our point, and the Supreme Court's holding in Amchem , was that the recovery secured through a settlement cannot be the basis for finding that common issues predominate. Id. ; Amchem , 521 U.S. at 622-23, 117 S.Ct. 2231. In Amchem , the district court found that predominance was satisfied based in part on class members' common interest in the settlement benefits-prompt and fair compensation without the risk and cost of litigation. 521 U.S. at 622, 117 S.Ct. 2231. The Supreme Court held that this was error because predominance looks at the cohesiveness of "the legal or factual questions that qualify each class member's case as a genuine controversy, questions that preexist any settlement." Id. at 623, 117 S.Ct. 2231. But whether a proposed class is sufficiently cohesive to satisfy Rule 23(b)(3) is informed by whether certification is for litigation or settlement. A class that is certifiable for settlement may not be certifiable for litigation if the settlement obviates the need to litigate individualized issues that would make a trial unmanageable. See 2 William B. Rubenstein, Newberg on Class Actions § 4:63 (5th ed. 2018) ("Courts ... regularly certify settlement classes that might not have been certifiable for trial purposes because of manageability concerns."). The Supreme Court said as much in Amchem . There, the Third Circuit, which reversed the district court's certification, held that Rule 23(a) and (b)(3) 's requirements "must be satisfied without taking into account the settlement." Id. at 619, 117 S.Ct. 2231 (quoting Georgine v. Amchem Prods., Inc. , 83 F.3d 610, 626 (3d Cir. 1996) ). Disagreeing, the Supreme Court pointed out that the Third Circuit "should have acknowledged that settlement is a factor in the calculus." Id. at 622, 117 S.Ct. 2231. The Court concluded that "a remand [was] not warranted," however, because the class did not satisfy Rule 23 's requirements "with or without a settlement on the table." Id. The Court also recognized that predominance is "readily met" in cases alleging consumer fraud, id. at 625, 117 S.Ct. 2231, and the present case is no exception. In many consumer fraud cases, the crux of each consumer's claim is that a company's mass marketing efforts, common to all consumers, misrepresented the company's product-here, a vehicle's fuel efficiency. The class was defined as "[a]ll current and former owners and lessees of [specified vehicles] who were the owner or lessee, on or before November 2, 2012, of such [v]ehicle that was registered [domestically]." This cohesive group of individuals suffered the same harm in the same way because of the automakers' alleged conduct. This case is a far cry from Amchem , which involved a "sprawling" asbestos settlement class with members who had wide-ranging injuries, some exposure-only and others imminently fatal. 521 U.S. at 623-26, 117 S.Ct. 2231. As Hanlon explained in distinguishing Amchem , the "heart" of the problem there was the class members' conflicting interests: current claimants, who were sick, wanted to maximize the immediate payout, whereas healthy claimants had a strong interest in preserving funds in case they became ill in the future. Hanlon , 150 F.3d at 1020-21. These vast differences in Amchem required "caution [because] individual stakes are high and disparities among class members great." 521 U.S. at 625, 117 S.Ct. 2231. In contrast, here, class members were exposed to uniform fuel-economy misrepresentations and suffered identical injuries within only a small range of damages. Further, as in Hanlon , no material conflicts existed among class members. Id . at 1021. The district court found that the following undisputed common questions predominated over individualized issues: (1) "[w]hether the fuel economy statements were in fact inaccurate"; and (2) "whether [the automakers] knew that their fuel economy statements were false or misleading." The court also found that the alleged misrepresentations were "uniformly" made via "Monroney stickers and nationwide advertising." We have held that these types of common issues, which turn on a common course of conduct by the defendant, can establish predominance in nationwide class actions. See Hanlon , 150 F.3d at 1022-23 (holding that "[a] common nucleus of facts and potential legal remedies dominate[d]" over "idiosyncratic differences between state consumer protection laws" where a nationwide class of minivan buyers' claims turned on "questions of [the manufacturer's] prior knowledge of the [vehicle's] deficiency, the design defect, and a damages remedy"); Edwards v. First Am. Corp. , 798 F.3d 1172, 1182-83 (9th Cir. 2015) (reversing denial of class certification for nationwide class of homebuyers because the alleged "common scheme, if true, present[ed] a significant aspect of [the defendant's] transactions that warrant[ed] class adjudication"). a. The Inclusion of Used Car Purchasers in the Class Does Not Defeat Predominance Fetsch and Roland argue that used car purchasers may not have seen the automakers' fuel efficiency representations, because only new cars are required to display the Monroney stickers, and that including used car purchasers in the class creates a factual issue precluding predominance. Their argument ignores the district court's finding that the alleged misrepresentations were made "uniformly"-not only on the Monroney stickers, but also in "nationwide advertising." When misrepresentations are made as part of a nationwide, concerted marketing effort, it makes no difference to the predominance analysis whether consumers encounter them in different guises. See In re Tobacco II Cases , 46 Cal.4th 298, 93 Cal.Rptr.3d 559, 207 P.3d 20, 40-41 (2009) ; see also In re First All. Mortg. Co. , 471 F.3d 977, 991 (9th Cir. 2006) ("The exact wording of the ... misrepresentations ... is not the predominant issue. It is the underlying scheme which demands attention." (quoting In re Am. Cont'l Corp./Lincoln Sav. & Loan Sec. Litig. , 140 F.R.D. 425, 431 (D. Ariz. 1992) )). Whether or not Hyundai's and Kia's advertising was substantial enough to support an inference of reliance under In re Tobacco II , the potential individual questions of reliance for used-car purchasers do not predominate in the context of this proposed settlement class. That some individualized issues might need to be addressed does not in and of itself defeat predominance. The predominance inquiry is mainly concerned with "the balance between individual and common issues." Sali v. Corona Reg'l Med. Ctr. , 889 F.3d 623, 635 (9th Cir. 2018) (emphasis added) (quoting Wang v. Chinese Daily News, Inc. , 737 F.3d 538, 545-46 (9th Cir. 2013) ). Indeed, this sort of individual question would only apply to a subset of the class (used-car purchasers) and would primarily implicate trial management issues, which we do not consider when conducting a predominance analysis for a settlement class. Amchem , 521 U.S. at 620, 117 S.Ct. 2231. Similarly, even if, as the automakers' expert opined, used car buyers are in a "somewhat different market" than new car buyers and would require a different damages analysis, the district court did not abuse its discretion by finding that common issues of fact predominated. "[T]he mere fact that there might be differences in damage calculations is not sufficient to defeat class certification." Pulaski & Middleman, LLC v. Google, Inc. , 802 F.3d 979, 987 (9th Cir. 2015) (quoting Stearns v. Ticketmaster Corp. , 655 F.3d 1013, 1026 (9th Cir. 2011) ); see Fed. R. Civ. P. 23(b)(3) advisory committee's note to 1966 amendment ("[A] fraud perpetrated on numerous persons by the use of similar misrepresentations may be an appealing situation for a class action, and it may remain so despite the need, if liability is found, for separate determination of the damages suffered by individuals within the class."). Nor is it clear why the damages here would need to be calculated based on each consumer's willingness to pay for higher fuel efficiency. Fraud damages do not normally correlate with the degree of reliance. Cf. Tobacco II Cases , 93 Cal.Rptr.3d 559, 207 P.3d at 39 ("It is not ... necessary that [the plaintiff's] reliance upon the truth of the fraudulent misrepresentation be the sole or even the predominant or decisive factor influencing his conduct .... It is enough that the representation has played a substantial part, and so had been a substantial factor, in influencing his decision." (alteration in original) (quoting Engalla v. Permanente Med. Grp., Inc. , 15 Cal.4th 951, 64 Cal.Rptr.2d 843, 938 P.2d 903, 919 (1997) )). If a consumer were to establish the threshold level of reliance, the automaker would be liable for the consumer's entire loss from higher-than-expected fuel costs-an amount that can easily be calculated on an individual basis, as it was in the Reimbursement Program. b. Variations in State Law Do Not Defeat Predominance Fetsch and Roland also argue that the district court failed to address variations in state law affecting claims by used car purchasers and that it was required to do so under Mazza v. American Honda Motor Co. , 666 F.3d 581 (9th Cir. 2012). They are incorrect. Subject to constitutional limitations and the forum state's choice-of-law rules, a court adjudicating a multistate class action is free to apply the substantive law of a single state to the entire class. Phillips Petroleum Co. v. Shutts , 472 U.S. 797, 823, 105 S.Ct. 2965, 86 L.Ed.2d 628 (1985) ; see also Harmsen v. Smith , 693 F.2d 932, 946-47 (9th Cir. 1982) (explaining that a district court sitting in diversity must "apply the substantive law of the state in which it sits, including choice-of-law rules"); Klaxon Co. v. Stentor Elec. Mfg. Co. , 313 U.S. 487, 496, 61 S.Ct. 1020, 85 L.Ed. 1477 (1941) (noting that "the accident of diversity of citizenship would constantly disturb equal administration of justice in coordinate state and federal courts sitting side by side" if Erie Railroad Co. v. Tompkins , 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188 (1938) did not apply to conflict-of-laws rules). Here, no party argued that California's choice-of-law rules should not apply to this class settlement arising from an MDL in a California court. By default, California courts apply California law "unless a party litigant timely invokes the law of a foreign state," in which case it is "the foreign law proponent" who must "shoulder the burden of demonstrating that foreign law, rather than California law, should apply to class claims." Wash. Mut. Bank, FA v. Superior Court , 24 Cal.4th 906, 103 Cal.Rptr.2d 320, 15 P.3d 1071, 1080-81 (2001) (quoting Bernhard v. Harrah's Club , 16 Cal.3d 313, 128 Cal.Rptr. 215, 546 P.2d 719, 721 (1976) ); accord Pokorny v. Quixtar, Inc. , 601 F.3d 987, 995 (9th Cir. 2010). To meet their burden, the objectors must satisfy the three-step governmental interest test. Wash. Mut. , 103 Cal.Rptr.2d 320, 15 P.3d at 1080-81 ; Pokorny , 601 F.3d at 994-95. Under that test, the objectors must prove that (1) the law of the foreign state "materially differs from the law of California," Wash. Mut. , 103 Cal.Rptr.2d 320, 15 P.3d at 1080, meaning that the law differs "with regard to the particular issue in question"; (2) a "true conflict exists," meaning that each state has an interest in the application of its own law to "the circumstances of the particular case"; and (3) the foreign state's interest would be "more impaired" than California's interest if California law were applied. Kearney v. Salomon Smith Barney, Inc. , 39 Cal.4th 95, 45 Cal.Rptr.3d 730, 137 P.3d 914, 922 (2006) ; accord Pokorny , 601 F.3d at 994-95. If the objectors fail to meet their burden at any step in the analysis, the district court "may properly find California law applicable without proceeding" to the rest of the analysis. Pokorny , 601 F.3d at 995 (quoting Wash. Mut. , 103 Cal.Rptr.2d 320, 15 P.3d at 1081 ). i. Objectors Failed to Meet Their Burden of Showing That California Law Does Not Apply Fetsch and Roland do not suggest that application of California law gives rise to constitutional problems. And before the district court, no objector presented an adequate choice-of-law analysis or explained how, under the facts of this case, the governmental interest test's three elements were met. Further, no objector argued that differences between the consumer protection laws of all fifty states precluded certification of a settlement class. Consequently, neither the district court nor class counsel were obligated to address choice-of-law issues beyond those raised by the objectors, and we will not decertify a class action for lack of such analysis. See Harmsen , 693 F.2d at 947 (affirming district court's application of California law to multistate class where the proponent of foreign law "failed to show, as required by California law, that the law of other states relating to the [state law] claims is significantly different from California's and, more importantly, that the interests of other states would be impaired by application of California law to these nonresident plaintiffs"); Pokorny , 601 F.3d at 994-96 (affirming application of California law because the foreign law proponent failed to meet its burden under California's governmental interest test). Mazza is readily distinguishable. There, the foreign law proponent (the defendant) "exhaustively detailed the ways in which California law differs from the laws of the 43 other jurisdictions" and showed how applying the facts to those disparate state laws made "a difference in this litigation." Mazza , 666 F.3d at 590-91. Unlike class counsel here, the plaintiffs in Mazza did "not contest these differences." Id. at 591 n.3. Weighing these arguments and concessions, a divided panel concluded that the defendant had "met its burden" to show that foreign law applied "[u]nder the facts and circumstances of this case." Id. at 591, 594. Importantly, the Mazza class was certified for litigation purposes. The prospect of having to apply the separate laws of dozens of jurisdictions presented a significant issue for trial manageability, weighing against a predominance finding. See also Zinser v. Accufix Research Inst., Inc. , 253 F.3d 1180, 1190-92 (9th Cir. 2001) (treating state law variations as a subspecies of trial manageability concerns). In settlement cases, such as the one at hand, the district court need not consider trial manageability issues. Amchem , 521 U.S. at 620, 117 S.Ct. 2231. In Hanlon , we affirmed certification under Rule 23(b)(3) of a nationwide settlement class of car owners alleging violations of various state consumer laws. 150 F.3d at 1017, 1022. We held that common questions as to the defendant's knowledge and the existence of the problem (the same questions at issue here) predominated, notwithstanding "variations in state law." Id. at 1020, 1022-23. In rejecting the objectors' argument that "the idiosyncratic differences between state consumer protection laws" defeated predominance, we reasoned that the claims revolved around a "common nucleus of facts" and applied the longstanding rule that "differing remedies" do not preclude class certification. Id. at 1022-23. That same reasoning applies with even greater force here, where the class claims turn on the automakers' common course of conduct-their fuel economy statements-and no objector established that the law of any other states applied. ii. Application of California Law Satisfies Due Process Objector Linda Ruth Scott, a lead plaintiff in a Virginia class action that was transferred to California as part of the MDL, argues that application of California law violates her due process rights. She asserts that Virginia, unlike most jurisdictions, does not provide cross-jurisdictional tolling of the statutes of limitations on her claims notwithstanding that her case was stayed upon transfer and remained pending in Virginia at the time of certification. Thus, she argues, certification of a nationwide class left Virginia class members with no real opportunity to opt out because their claims would otherwise be dismissed as time-barred. But Scott does not dispute that she, like all class members, had the right to opt out. See Epstein v. MCA, Inc. , 179 F.3d 641, 648 (9th Cir. 1999) ; see also Ortiz v. Fibreboard Corp. , 527 U.S. 815, 848, 119 S.Ct. 2295, 144 L.Ed.2d 715 (1999) ("[B]efore an absent class member's right of action [is] extinguishable due process require[s] that the member 'receive notice plus an opportunity to be heard and participate in the litigation,' and ... 'at a minimum ... an absent plaintiff [must] be provided with an opportunity to remove himself from the class.' " (quoting Shutts , 472 U.S. at 812, 105 S.Ct. 2965 ) (last omission and last alteration in original)). Rather, she contends that as a practical matter, she and other Virginia class members "[could not] opt out of this nationwide class action settlement because the District Court refused to certify a Virginia subclass with recognized class representatives asserting Virginia causes of action." Attorney Feinman, however, acknowledged that he filed a class action (along with two other mass actions) in Virginia after the creation of the MDL to toll the statute of limitations there, preserving claims for Virginia plaintiffs who decided to opt out of the MDL settlement. Indeed, a handful of Virginia plaintiffs did opt out of the settlement and continued their litigation in Virginia, along with plaintiffs who purchased subject vehicles after November 2, 2012, without any statute of limitations issues. See Gentry v. Hyundai Motor Am., Inc. , No. 3:13-CV-00030, 2017 WL 354251 (W.D. Va. Jan. 23, 2017), aff'd in part, dismissed in part sub nom. Adbul-Mumit v. Alexandria Hyundai, LLC , 896 F.3d 278 (4th Cir. 2018), cert. denied , --- U.S. ----, 139 S.Ct. 607, 202 L.Ed.2d 431 (2018). Ultimately, the Virginia courts dismissed two out of the three actions for pleading deficiencies. The sole remaining action survived the pleading stage with only one lemon law claim regarding the onboard mileage calculator, not the fuel economy misrepresentations at issue in the MDL. The statute of limitations issue raised here did not feature in the district court's or Fourth Circuit's decisions. Even assuming that the statute of limitations would have barred the claims of Virginia plaintiffs who opted out of the settlement, Hanlon forecloses Scott's requested relief. In Hanlon , as here, multiple class actions were filed and then consolidated in California following a federal agency's investigation, with the defendant announcing a remedial plan and entering into a settlement only after the class moved for certification. See 150 F.3d at 1018. Like Scott, an objector in Hanlon filed a late class action in another state and sought to litigate it in contravention of the district court's orders. See id. at 1019. We explained that while the objector was free to opt out of the class by filing the out-of-state action, he had no right to do so on behalf of anyone else: The procedural due process rights of [class] members include an opportunity to be excluded from the action. The right to participate, or to opt-out, is an individual one and should not be made by the class representative or the class counsel. There is no class action rule, statute, or case that allows a putative class plaintiff or counsel to exercise class rights en masse, either by making a class-wide objection or by attempting to effect a group-wide exclusion from an existing class. Indeed, to do so would infringe on the due process rights of the individual class members, who have the right to intelligently and individually choose whether to continue in a suit as class members. Additionally, to allow representatives in variously asserted class actions to opt a class out without the permission of individual class members "would lead to chaos in the management of class actions." Id. at 1024 (internal citations omitted) (quoting Berry Petroleum Co. v. Adams & Peck , 518 F.2d 402, 412 (2d Cir. 1975) ). Scott claims that the statute of limitations issue and due process "require[ ] that the [Virginia] [sub]class as a whole be remanded" (emphasis added). But what she seeks to do here is exactly what Hanlon held was forbidden-opt out a state subclass. Finally, Scott argues that by "not creating a Virginia subclass," the district court was "using the [MDL] process and [ Rule] 23 to deny the Virginians their day in court." But she has it backwards. Scott seeks to displace the operation of federal law-the MDL statute and class certification rules-to accommodate a single state's tolling rule. The Supremacy Clause forecloses such an argument. See Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. , 559 U.S. 393, 398-99, 130 S.Ct. 1431, 176 L.Ed.2d 311 (2010) (explaining that if a proposed class meets Rule 23 's criteria, state law cannot prohibit certification). Scott relies heavily on Shutts to support her due process claim, but she misunderstands the due process rights it addressed. Shutts distinguished the "minimum contacts requirement" that can be asserted by "out-of-state defendants or parties in the procedural posture of a defendant" in multistate cases from the process due "to absent class-action plaintiffs" based on their "constitutionally recognized property interest" in "a chose in action." 472 U.S. at 807, 105 S.Ct. 2965. Because absent class plaintiffs face fewer litigation-related burdens than out-of-state defendants in nonclass suits, "the Due Process Clause need not and does not afford the former as much protection from ... jurisdiction as it does the latter." Id. at 811, 105 S.Ct. 2965. "[T]o bind an absent plaintiff concerning a claim for money damages or similar relief at law," the district court is obligated to provide only "minimal procedural due process protection." Id. at 811-12, 105 S.Ct. 2965. Shutts "identified various procedural safeguards that are necessary to bind absent class members, including notice, the opportunity to be heard, the opportunity to opt out, and adequate representation," Epstein , 179 F.3d at 648, all of which were present here. As for the minimum contacts requirement, which out-of-state defendants could raise, that the application of a state's law must not be "arbitrary [or] fundamentally unfair," California has extensive contacts that satisfy this due process requirement here. Id. at 818, 105 S.Ct. 2965. For example, Hyundai Motor America is incorporated and has its principal place of business in California and roughly 10.7% of the class vehicles were sold in California. * * * In sum, the district court did not abuse its discretion in finding that common issues predominated. 2. Adequacy Separate from the predominance analysis, due process "requires that the named plaintiff at all times adequately represent the interests of the absent class members." Shutts , 472 U.S. at 812, 105 S.Ct. 2965. This adequacy requirement, formalized in Rule 23(a)(4), "serves to uncover conflicts of interest between named parties and the class they seek to represent" as well as the "competency and conflicts of class counsel." Amchem , 521 U.S. at 625, 626 n.20, 117 S.Ct. 2231. To determine legal adequacy, we resolve two questions: "(1) do the named plaintiffs and their counsel have any conflicts of interest with other class members and (2) will the named plaintiffs and their counsel prosecute the action vigorously on behalf of the class?" Hanlon , 150 F.3d at 1020. Scott contends that class counsel were inadequate because they failed to protect the rights of absent Virginia class members to opt out of the settlement. Since class counsel did in fact protect Virginians' right to opt out, this argument is meritless. Scott also argues that class counsel Hagens Berman, the firm representing the Brady and Hunter plaintiffs, now has a potential conflict with the class. She asserts that more than two years after the settlement was signed, Hagens Berman and the firm representing Hyundai jointly represented consumers in a putative class action suit against Volkswagen for alleged fraud regarding vehicle emissions. Scott identifies no authority establishing that a co-counsel relationship between class counsel and defense counsel in a future, unrelated case presents a conflict. B. Settlement Approval A binding settlement must provide notice to the class in a "reasonable manner" and otherwise be "fair, reasonable, and adequate." Fed. R. Civ. P. 23(e)(1), (2). Various objectors allege inadequacies in the notice and claim forms and purported collusion between class counsel and the automakers. None of their claims have merit. 1. The Notice to Class Members Provided Sufficient Information Before the district court approves a class settlement under Rule 23(e), it is "critical" that class members receive adequate notice. Hanlon , 150 F.3d at 1025. To satisfy Rule 23(e)(1), settlement notices must "present information about a proposed settlement neutrally, simply, and understandably." Rodriguez v. W. Publ'g Corp. , 563 F.3d 948, 962 (9th Cir. 2009). "Notice is satisfactory if it 'generally describes the terms of the settlement in sufficient detail to alert those with adverse viewpoints to investigate and to come forward and be heard.' " Id. (quoting Churchill Vill., LLC v. Gen. Elec. , 361 F.3d 566, 575 (9th Cir. 2004) ). Fetsch and Roland argue that class members who were already participating in the automakers' voluntary Reimbursement Program "likely were unaware that additional compensation ... could be received" by remaining in the program because this information was "buried" on page 11 of the long form notice and was omitted from the short form and email notices. In fact, the very first page of the long form notice informed class members: "If you previously received money under the [Reimbursement Program], you may still be able to receive a payment from the Settlement." As for the short form notice, it was designed to be, as the name suggests, short. Its primary purpose was to alert class members to the settlement, provide a high-level overview of the process, including critical dates, and explain where class members could obtain additional information, such as eligibility information and claim forms. In addition, after outlining some of the potential compensation, it informed class members that "[o]ther settlement benefits exist" and invited them to use an online calculator to estimate their individual benefit under each of the various compensation options based on a host of personalized factors. The short form notice "highly recommended" that class members "use this reimbursement calculator to evaluate [their] options based on [their] own circumstances ... before submitting a claim." This notice was more than adequate. Nor was it misleading for the various notices to inform class members that "[h]igh mileage drivers may receive greater amounts by participating in the ... Reimbursement Program." Since compensation under the Reimbursement Program was proportional to the number of miles driven and thus theoretically unlimited, that statement was true. Moreover, it served the valuable purpose of warning high-mileage drivers that choosing a lump sum payment might not have been in their best interests. Finally, in arguing that the notices did not explain in a "step-by-step" formula how each class member's benefit is calculated, Fetsch and Roland seek to impose a higher standard than is required. A settlement notice need not "provide an exact forecast" of the award each class member would receive, let alone a detailed mathematical breakdown; it merely must give class members "enough information so that those with 'adverse viewpoints' could investigate and 'come forward and be heard.' " Online DVD-Rental , 779 F.3d at 946-47 (9th Cir. 2015) (quoting Lane v. Facebook, Inc. , 696 F.3d 811, 826 (9th Cir. 2012) ). 2. The Claim Forms Were Not Overly Burdensome Objectors Ahearn and York argue that no claim forms were necessary at all and that the automakers should have automatically made lump sum payments to class members who did not request another form of compensation. They cite no evidence that this was possible. The district court found that the automakers did "not have complete records of resales of the class vehicles," and Ahearn and York fail to explain how the automakers could have identified subsequent purchasers who were also part of the class. They do not dispute that it was reasonable for the settlement to provide class members with different monetary recovery options based on miles driven and ownership status-information also not in the automakers' possession. Given that the automakers lacked complete information to determine the identities of all class members and the amounts of their claims, the district court properly exercised its discretion in finding that "some sort of claims process is necessary in order to verify ... that the claimant is a current owner, former owner, or current or former lessee of a qualifying vehicle." Ahearn, York, Fetsch, and Roland contend that the claim forms required too much documentation, such as proof of a class member's current address and proof of sale or ownership, and that this documentation burden is reflected in low claim participation rates. However, class members could easily avoid most documentation requirements by submitting an online claim form, which pre-populated information after class members entered their vehicle identification number and the unique class member identification number provided by their notices. Fetsch and Roland cite no evidence that any claims submitted on the paper claim form were, as they speculate, "denied because one box was not checked or one piece of documentation was not turned in." Ahearn and York contend that the 21% of class members who had filed claims for lump sum payments as of March 31, 2015 was an unreasonably low participation rate and that the "daunting claim form" was to blame. As of May 31, 2015-more than a month before the July 6, 2015 claims deadline-the participation rate of lump sum claimants had increased to 23%. We have approved class action settlements "where less than five percent of class members file claims." Online DVD-Rental , 779 F.3d at 945 ; see also Rodriguez , 563 F.3d at 967 (holding that the district court did not abuse its discretion in approving settlement class where 14% of 376,301 putative class members returned claim forms). And the 23% participation rate here must be viewed in light of the 59% of class members who took advantage of the Reimbursement Program prior to notice of the settlement. As the district court recognized, many of these class members "would decide not to submit a claims form at all" if they were satisfied with the automakers' voluntary compensation. 3. There Is No Evidence of Collusion Between Class Counsel and the Automakers Rule 23(e) ensures that unnamed class members are protected "from unjust or unfair settlements affecting their rights." Amchem , 521 U.S. at 623, 117 S.Ct. 2231 (1997). When the district court determines that a proposed settlement is fundamentally fair, adequate, and reasonable, our review "is extremely limited." Hanlon , 150 F.3d at 1026. We consider the overall fairness of "the settlement taken as a whole, rather than the individual component parts," because "[n]either the district court nor this court ha[s] the ability to 'delete, modify or substitute certain provisions.' " Id. (quoting Officers for Justice v. Civil Serv. Comm'n , 688 F.2d 615, 630 (9th Cir. 1982) ). The objectors argue that the settlement reached here was a "sweetheart deal." To the contrary, the settlement bears none of the typical signs of collusion between class counsel and defendants, such as when class counsel "receive a disproportionate distribution of the settlement," Bluetooth Headset , 654 F.3d at 947 (quoting Hanlon , 150 F.3d at 1021 ), the agreement contains a "clear sailing" provision for attorney's fees "separate and apart from class funds," id. , or unawarded fees revert to the defendants rather than to the class, id. This case stands in contrast to Bluetooth Headset , in which the settlement paid the class "zero dollars" and contained a "clear sailing" provision in which "defendants agreed not to object" to an award of attorney's fees totaling eight times the cy pres award, and a "kicker" clause whereby "all fees not awarded would revert to defendants." Id. at 938, 947. The district court there made no findings under either the lodestar or the percentage method and instead awarded what "defendants agreed to pay." Id. at 943. The settlement also bears no resemblance to the one in Amchem , which allowed defendants to withdraw in ten years while the class remained bound in perpetuity, limited the number of plaintiffs who could reject it and pursue individual claims each year, set annual caps on claims for each disease, set numerical and dollar limits on extraordinary claims above the fixed compensation ranges, offered no adjustments for inflation, and provided no compensation for certain claims and injuries. Amchem , 521 U.S. at 604-05, 627, 117 S.Ct. 2231. Here, the settlement has no clear sailing or kicker clauses, the automakers successfully litigated a reduction in fees, the court made findings, and the class received tens of millions of dollars. Moreover, the settlement here "was negotiated over multiple mediation sessions with a respected and experienced mediator," class counsel were "experienced," and class members had plenty of opportunities to raise their concerns at seven hearings over seventeen months. Fetsch and Roland assert that the automakers "looked for a settling group of plaintiffs that would provide them the lowest settlement cost," a phenomenon known as a "reverse auction." See Negrete v. Allianz Life Ins. Co. of N. Am. , 523 F.3d 1091, 1099 (9th Cir. 2008). As in Negrete , however, they have "floated out the specter of a reverse auction, but brought forth no facts to give that eidolon more substance." Id. It is true, as Fetsch and Roland point out, that class action defendants are generally indifferent to the allocation of settlement funds between class and counsel, which can encourage a settlement that is overly generous to counsel at the expense of the class. But here such concerns are out of place. No objector disputes the district court's finding that the settlement "provides substantial relief," including a "substantial cash payout, ranging from $ 240 to $ 1,420" per class member. The settling parties agreed on the amount of class compensation more than six months before negotiating, "over multiple mediation sessions with a respected and experienced mediator," the "reasonable" attorney's fees provided in the settlement agreement. We have previously approved such an approach, see Hanlon , 150 F.3d at 1029, and "[w]e put a good deal of stock in the product of an arms-length, non-collusive, negotiated resolution," Rodriguez , 563 F.3d at 965. Providing further assurance that the agreement was not the product of collusion, class counsel McCuneWright did not reach an agreement with the automakers regarding the amount of attorney's fees to which they were entitled. After a contested fee motion, the district court awarded McCuneWright approximately half of the fees that they had requested. Finally, the objectors contend that the unreasonably high attorney's fees award evidences collusion in the settlement. Ahearn and York argue that the award "bears all the hallmarks of collusion," and Fetsch and Roland claim that class counsel was motivated to give the automakers a "great deal" in exchange for not opposing their requested fees. Courts in this circuit determine attorney's fees in class actions using either the lodestar method or the percentage-of-recovery method. Hanlon , 150 F.3d at 1029. "The lodestar calculation begins with the multiplication of the number of hours reasonably expended by a reasonable hourly rate." Id. The district court may then adjust the resulting figure upward or downward to account for various factors, see Kerr v. Screen Extras Guild, Inc. , 526 F.2d 67, 70 (9th Cir. 1975), including the quality of the representation, the benefit obtained for the class, the complexity and novelty of the issues presented, and the risk of nonpayment, Hanlon , 150 F.3d at 1029. In class action cases where the defendants provide monetary compensation to the plaintiffs, "courts have discretion to employ either the lodestar method or the percentage-of-recovery method." Bluetooth Headset , 654 F.3d at 942. In the percentage method, "the court simply awards the attorneys a percentage of the fund sufficient to provide class counsel with a reasonable fee," using 25% as a benchmark. Hanlon , 150 F.3d at 1029. Similar to the lodestar, the 25% benchmark can be adjusted upward or downward, depending on the circumstances. See Six (6) Mexican Workers v. Ariz. Citrus Growers , 904 F.2d 1301, 1311 (9th Cir. 1990). When valuing the settlement is difficult or impossible, the lodestar method may prove more convenient, see Hanlon , 150 F.3d at 1029, but "no presumption in favor of either the percentage or the lodestar method encumbers the district court's discretion to choose one or the other," In re Wash. Pub. Power Supply Sys. Sec. Litig. , 19 F.3d 1291, 1296 (9th Cir. 1994). Here, the district court properly exercised its discretion in calculating the fee award using the lodestar method. As the district court found, the automakers "will pay attorneys' fees separately from the amount allocated to those covered by the class." Moreover, it is difficult to estimate the settlement value's upper bound. The settlement extended the Reimbursement Program's enrollment deadline by a year and a half, allowing additional class members to participate. These class members will continue to receive compensation from the program for many years into the future, the present value of which will depend on how many miles they drive and their cost of fuel. Ahearn and York argue that the district court erred by not confirming that attorney's fees were 25% or less of the settlement's value. However, the district court in fact cross-checked the lodestar amount and specifically found that the "total amount of attorney's fees awarded in this case is far lower than the 25% of the settlement figure used as a 'benchmark' in many class action cases in the Ninth Circuit." We have affirmed fee awards totaling a far greater percentage of the class recovery than the fees here. See, e.g. , Vizcaino v. Microsoft Corp. , 290 F.3d 1043, 1047-48 (9th Cir. 2002) (no abuse of discretion to award fees constituting 28% of the class's recovery given "risk" assumed in litigating); In re Pac. Enters. Sec. Litig. , 47 F.3d 373, 379 (9th Cir. 1995) (no abuse of discretion where the "$ 4 million award (thirty-three percent [of the class's recovery] ) for attorneys' fees is justified because of the complexity of the issues and the risks"). In any event, we do not require courts employing the lodestar method to perform a "crosscheck" using the percentage method. This would make "little logical sense," 5 Rubenstein, supra , § 15:92, because "the lodestar method yields a fee that is presumptively [reasonable]." Perdue v. Kenny A. ex rel. Winn , 559 U.S. 542, 552, 130 S.Ct. 1662, 176 L.Ed.2d 494 (2010). The percentage method is merely a shortcut to be used "in lieu of the often more time-consuming task of calculating the lodestar," but only if "the benefit to the class is easily quantified." Bluetooth Headset , 654 F.3d at 942. Even then, it is at best a rough approximation of a reasonable fee. Ahearn and York also object to the district court's use of a lodestar multiplier. The district court, which had ably managed this complex litigation for several years and observed various counsel's performance during numerous hearings and through extensive briefing, was in the best position to evaluate each firm's contributions. The record shows that the district court carefully made this assessment in determining the appropriate amount of attorney's fees and explained the basis of its ruling. The district court applied downward multipliers of 27 to 80 percent to the lodestars for the non-settling parties' counsel because they "had a more minor role in the [MDL] and did not participate [in] negotiating the primary settlement." The court applied a multiplier of 1.22 to the fee award for class counsel Hagens Berman due to "the complexity and volume of work that counsel engaged in in order to diligently pursue this case and develop its primary theory of liability," finding the multiplier in line with others in comparable complex and multi-year multidistrict litigations. And the court applied a multiplier of 1.5521 to the fees for class counsel McCuneWright because they "assumed more risk than other firms" by being one of the first firms to take up this cause, having filed Espinosa nearly 10 months before the automakers announced the fuel efficiency revisions. These multipliers are modest or in-line with others we have affirmed. See, e.g. , Vizcaino v. Microsoft Corp. , 290 F.3d 1043, 1051 (9th Cir. 2002) (upholding a lodestar multiplier cross-check showing a multiplier of 3.65); Kelly v. Wengler , 822 F.3d 1085, 1093, 1105 (9th Cir. 2016) (affirming lodestar multipliers of 2.0 and 1.3). The district court's limited use of the multipliers was well within its broad discretion to determine the amount of reasonable fees, see Fox v. Vice , 563 U.S. 826, 838, 131 S.Ct. 2205, 180 L.Ed.2d 45 (2011) (emphasizing and instructing appellate courts to give "substantial deference" to attorney's fees calculations because "trial courts need not, and indeed should not, become green-eyeshade accountants" and because of "the district court's superior understanding of the litigation" (quoting Hensley v. Eckerhart , 461 U.S. 424, 437, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983) )), and does not support a finding of collusion. C. Denial of Attorney's Fees to Feinman Last, Scott and her counsel Feinman challenge the district court's ruling that Feinman was not entitled to attorney's fees because he conferred no benefit on the class. "[W]here objectors do not add any new legal argument or expertise, and do not participate constructively in the litigation or confer a benefit on the class, they are not entitled to an award premised on equitable principles." Rodriguez v. Disner , 688 F.3d 645, 659 (9th Cir. 2012). The district court denied Feinman's $ 800,172.79 fee request because he "did not meaningfully contribute to the class settlement." Feinman sought fees on the theory that his due process arguments, which were rejected below and we reject here, were somehow beneficial to the class. However, these arguments are not only baseless, but also detrimental to the class; if adopted, they would permit Scott to hold hostage any class recovery under the settlement until she received the unique benefit of being certified to represent a Virginia subclass. Furthermore, Feinman does not dispute that he engaged in obstructive conduct throughout the litigation, including moving for discovery despite a stay and moving to remand despite an ongoing MDL. The district court did not abuse its discretion in denying fees. IV. Conclusion Over the course of several years, the district court performed an admirable job of managing this complex litigation. After the settlement was announced, the district court held multiple status conferences and requested several rounds of briefing to ensure that all of the litigants' concerns were heard and addressed. It made careful findings, which the objectors here largely do not challenge, and which more than support the judgment. AFFIRMED. IKUTA, Circuit Judge, with whom KLEINFELD and M. SMITH, Circuit Judges, join, and with whom RAWLINSON, Circuit Judge, joins in part, dissenting: The district court in this case certified a multistate class action under Rule 23 of the Federal Rules of Civil Procedure without determining what law applied to the plaintiffs' claims. It then awarded attorneys' fees without determining the value of the benefit the class derived from the settlement. This is contrary to Rule 23 and Supreme Court precedent, see Amchem Prods., Inc. v. Windsor , 521 U.S. 591, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997). I dissent. I Defendants Hyundai and Kia overstated the fuel efficiency of certain vehicles that they manufactured and sold. After an EPA investigation confirmed this overstatement, Hyundai and Kia announced that they would lower the fuel efficiency estimates for the affected cars and simultaneously announced that they were each instituting a voluntary reimbursement program to compensate affected vehicle owners and lessees for the additional fuel costs that they had incurred and would incur in the future as a result of the overstated fuel efficiency statements. This announcement set off a flurry of litigation across the country. In Espinosa v. Hyundai Motor America , an action pending in district court in California, Hyundai filed an extensive "Appendix of Variations in State Laws," which detailed the differences in the applicable state consumer protection laws and common law fraud actions. The district court initially found these state law differences so material that it tentatively ruled the class could not be certified. Meanwhile, the Multidistrict Litigation (MDL) judicial panel consolidated over fifty other actions before the district court in which Espinosa was pending. After MDL consolidation, Hyundai and Kia moved for certification of a nationwide class and preliminary approval of a class settlement they had negotiated with counsel for three of the MDL cases, Espinosa , Hunter et al. v Hyundai Motor et al. , and Brady et al. v. Hyundai Motor et al. Objecting to class certification, plaintiffs in Gentry et al. v. Hyundai Motor America , an action that had been filed in a Virginia district court before being consolidated in the MDL, argued that variations in state law defeated the predominance of common questions. The Virginia plaintiffs argued that they had purchased their vehicles under sales contracts that contained valid choice of law provisions requiring Virginia law to be applied to any claims. The Virginia plaintiffs further claimed that Virginia law provided a materially different remedy to Virginia consumers for certain claims and such material differences between Virginia and California consumer protection law precluded certification of a nationwide class. Despite those objections, the district court declined to decide what law was applicable to the plaintiffs' claims and certified the class without ruling on this threshold legal issue or conducting a choice of law analysis. The district court acknowledged that the court "would need to engage in an extensive choice of law analysis" if the case were going to trial, but the district court erroneously thought that such an analysis was not required to certify a settlement class. In response to the Virginia plaintiffs' objections, the district court concluded that any substantial differences in state law could be addressed as part of the Rule 23(e) fairness hearing. See Fed. R. Civ. P. 23(e). The court certified the class and later approved the settlement. In its tentative ruling granting final settlement approval, the court estimated that the settlement value was some $ 210 million, relying on a rough estimate that the settling parties had provided a year earlier. According to the objectors, however, claims attributable to the settlement added up to about $ 21 million at the time of final approval. Although another $ 23 million in class claims were filed by class members, the objectors contend that those class members were already participating in Hyundai and Kia's voluntary reimbursement program before the settlement, and therefore the value of their claims could not be attributable to the settlement. Relying on this estimate that some $ 210 million was provided by the settlement, the district court awarded nearly $ 9 million in total attorneys' fees to class counsel. It used a lodestar multiplier of 1.22 for the Hunter and Brady plaintiffs' counsel on the ground that they undertook a large volume of complex work. It used a lodestar multiplier of 1.5521 for the Espinosa plaintiffs' counsel on the ground that they had assumed greater risk by filing a lawsuit before the EPA had announced the results of its investigation. A number of class members objected, arguing that the attorneys' fees award was excessive in proportion to the actual benefit obtained on behalf of the class. The district court summarily rejected these arguments. II A class action "may only be certified if the trial court is satisfied, after a rigorous analysis, that the prerequisites of [ Rule 23 of the Federal Rules of Civil Procedure ] have been satisfied." Gen. Tel. Co. of Sw. v. Falcon , 457 U.S. 147, 161, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982). For a class certified under Rule 23(b)(3), a court must find that "questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3). In order to determine whether the Rule 23 prerequisites are met, a district court must determine what state law (or laws) apply to the plaintiffs' claims. "Because the Rules Enabling Act forbids interpreting Rule 23 to 'abridge, enlarge or modify any substantive right,' " Wal-Mart Stores, Inc. v. Dukes , 564 U.S. 338, 367, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011) (quoting 28 U.S.C. § 2072(b) ), a court cannot certify a class if doing so would deprive litigants of the benefit of the appropriate substantive law applicable to their claims, even if a class action "would provide the most secure, fair, and efficient means" of compensating plaintiffs, Amchem , 521 U.S. at 628, 117 S.Ct. 2231. Identifying the applicable law is particularly crucial in a multi-state class action, because a district court cannot reasonably make a finding regarding predominance and superiority without doing so. See Castano v. Am. Tobacco Co. , 84 F.3d 734, 740-41 (5th Cir. 1996) (holding that "a district court must consider how variations in state law affect predominance and superiority"); see also Lozano v. AT&T Wireless Servs., Inc ., 504 F.3d 718, 728 (9th Cir. 2007) (holding that "the law on predominance requires the district court to consider variations in state law when a class action involves multiple jurisdictions"). The district court must identify the law that applies to plaintiffs' claims regardless whether the court is certifying a litigation class or a settlement class. While "a district court need not inquire whether the case, if tried, would present intractable management problems" when considering a request to certify a settlement class, "other specifications of the Rule-those designed to protect absentees by blocking unwarranted or overbroad class definitions-demand undiluted, even heightened, attention in the settlement context." Amchem , 521 U.S. at 620, 117 S.Ct. 2231. "Such attention is of vital importance, for a court asked to certify a settlement class will lack the opportunity, present when a case is litigated, to adjust the class, informed by the proceedings as they unfold." Id. It is well established "that problems beyond those of just manageability may exist when a district court is asked to certify a single nationwide class action suit, even for settlement purposes, when claims arise under the substantive laws of the fifty states." In re Warfarin Sodium Antitrust Litig. , 391 F.3d 516, 529-30 (3d Cir. 2004). If the plaintiffs are not governed by the same legal rules, e.g., if the law of consumer protection or the requisite mens rea differs from jurisdiction to jurisdiction, the court may not be able to find that "common questions of law or fact" predominate or that "a class action is superior to other available methods" to resolve a claim. See Lozano , 504 F.3d at 728 (holding that the district court reasonably concluded that predominance was defeated when the standard for upholding a class action waiver differed from state to state). We have scrutinized state law variations even when a class is proposed only for settlement in order to determine whether "the idiosyncratic differences between state consumer protection laws" were "sufficiently substantive to predominate over the shared claims." Hanlon v. Chrysler Corp. , 150 F.3d 1011, 1022-23 (9th Cir. 1998). Moreover, courts may