Rule 23(b)(3) of the Federal Rules of Civil Procedure1×1. Fed. R. Civ. P. 23(b)(3). The rule requires the court to find that “questions of law or fact common to class members predominate” and that a class action be “superior to other available methods for . . . adjudicating the controversy.” Id. The rule lists several factors that are “pertinent” to these findings, the most relevant of which is “the likely difficulties in managing a class action.” Id. is a formidable barrier for parties seeking to certify a nationwide class for litigation of claims grounded in state law.2×2. See Alexandra D. Lahav, Symmetry and Class Action Litigation, 60 UCLA L. Rev. 1494, 1504 (2013) (identifying “the increasingly narrow reading of the requirements of Rule 23, especially the commonality requirement . . . and the predominance requirement” as one of the “biggest barriers to class actions”); Linda Silberman, The Role of Choice of Law in National Class Actions, 156 U. Pa. L. Rev. 2001, 2002–03 (2008). Courts regularly conclude that differences in state law preclude the findings of predominance and superiority necessary to certify a class action.3×3. See Howard M. Erichson, The Problem of Settlement Class Actions, 82 Geo. Wash. L. Rev. 951, 979 (2014); Linda S. Mullenix, Gridlaw: The Enduring Legacy of Phillips Petroleum Co. v. Shutts, 74 UMKC L. Rev. 651, 653 (2006); Silberman, supra note 2, at 2003. At the same time, courts continue to certify nationwide classes for settlement of state law claims.4×4. See Erichson, supra note 3, at 978–81; Lahav, supra note 2, at 1504–07. This divergence is hard to square with Amchem Products, Inc. v. Windsor,5×5. 521 U.S. 591 (1997). in which the Supreme Court held that courts do not have to consider trial manageability when certifying a settlement-only class, but that the other requirements of Rule 23 “demand undiluted, even heightened, attention.”6×6. Id. at 620. Recently, in In re Hyundai & Kia Fuel Economy Litigation,7×7. 926 F.3d 539 (9th Cir. 2019) (en banc). the Ninth Circuit, sitting en banc, affirmed the district court’s certification of a nationwide class for settlement in a lawsuit alleging violations of state consumer protection laws.8×8. Id. at 572. Many saw the decision as a victory for class counsel and corporate defendants because it vacated a panel opinion that implied that variations in state law would usually prevent nationwide settlement of state law claims.9×9. See, e.g., Linda Chiem, Full 9th Circ. Hyundai Reset Gives Clarity on Class Deals, Law360 (June 7, 2019, 7:09 PM), https://www.law360.com/articles/1166975/full-9th-circ-hyundai-reset-gives-clarity-on-class-deals [https://perma.cc/GER9-UHYT]; Pravin R. Patel, Ninth Circuit Approves Nationwide Class Action Settlement, A.B.A. (July 3, 2019), https://www.americanbar.org/groups/litigation/committees/mass-torts/practice/2019/ninth-circuit-approves-nationwide-class-action-settlement [https://perma.cc/BU4Y-YRUY]. But proponents of nationwide settlements should not read too much into the decision: the Ninth Circuit justified certification by concluding that under California’s choice-of-law rules, California law applied to the entire class.10×10. Fuel Econ. Litig., 926 F.3d at 561–62. In doing so, it correctly recognized that variations in state law are relevant to certification for settlement, even if they do not preclude certification of a nationwide settlement class.
In the early 2010s, Hyundai and Kia used the results of EPA-mandated fuel economy tests to tout the efficiency of their vehicles.11×11. Bill Vlasic, Hyundai and Kia Acknowledge Overstating the Gas Mileage of Vehicles, N.Y. Times (Nov. 2, 2012), https://nyti.ms/RBJMq6 [https://perma.cc/27UV-AXZ5]. But consumers quickly became frustrated by their vehicles’ worse-than-expected performance under normal driving conditions.12×12. Id. The EPA began investigating these discrepancies in late 2011.13×13. In re Hyundai & Kia Fuel Econ. Litig., 881 F.3d 679, 694 (9th Cir. 2018). Within a couple of months, plaintiffs filed a putative nationwide class action, Espinosa v. Hyundai Motor America, Inc.,14×14. No. 12-cv-800 (C.D. Cal. Sept. 14, 2015). alleging that Hyundai’s inaccurate advertising violated California’s consumer protection laws.15×15. Fuel Econ. Litig., 881 F.3d at 695. The defendants removed the case to the Central District of California, where Judge Wu scheduled a class certification hearing for November of 2012.16×16. Id. at 688, 695. Four weeks before the hearing, in response to the EPA investigation, Hyundai and Kia announced downward adjustments in the fuel economy ratings of several 2011–2013 models.17×17. Fuel Econ. Litig., 926 F.3d at 553. Plaintiffs’ lawyers filed a wave of new lawsuits against the automakers.18×18. Id.
In a tentative ruling that November, Judge Wu indicated he would not certify a nationwide litigation class in Espinosa.19×19. Fuel Econ. Litig., 881 F.3d at 695–96. Hyundai’s appendix on variations in state law — which catalogued “differences in the burden of proof, liability, damages, statutes of limitations, and attorneys’ fee awards” among state consumer protection laws — demonstrated that under California’s choice-of-law rules, the law of each state would govern the claims of its class members.20×20. Id. at 695. Under California’s three-part choice-of-law test, the proponent of foreign law must prove that: (1) there are material differences between California law and the law of a foreign state; (2) “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[s] would be ‘more impaired’ than California’s interest if California law were applied.” Id. at 709 (Nguyen, J., dissenting) (quoting Kearney v. Salomon Smith Barney, Inc., 137 P.3d 914, 922 (Cal. 2006)). Citing the Ninth Circuit’s decision in Mazza v. American Honda Motor Co.,21×21. 666 F.3d 581 (9th Cir. 2012). he concluded that this variation precluded the finding of predominance necessary to certify a nationwide class.22×22. Fuel Econ. Litig., 881 F.3d at 696. While a final decision on class certification was pending, the Judicial Panel on Multidistrict Litigation centralized twelve cases, including Espinosa, for pretrial proceedings before Judge Wu.23×23. Id. at 697. One week later, the parties in Espinosa informed Judge Wu that the plaintiffs in Espinosa and two other cases had reached a nationwide settlement with Hyundai.24×24. Id. Kia later agreed to the same terms.25×25. Id.
Ten months later, the settling plaintiffs moved for certification of a nationwide settlement class and preliminary approval of the settlement.26×26. Id. at 699. The plaintiffs in Gentry v. Hyundai Motor America, Inc.,27×27. No. 13-cv-00030 (W.D. Va. Feb. 12, 2019). a tag-along action,28×28. A “tag-along action” is one that shares common facts with cases that have already been coordinated or consolidated for pretrial proceedings. See Rules of Procedure of the U.S. Judicial Panel on Multidistrict Litigation 1.1(h). objected.29×29. Fuel Econ. Litig., 881 F.3d at 699. They argued California’s choice-of-law rule required the application of Virginia law to Virginia purchasers.30×30. Id. at 699–700, 699 n.19. The Gentry plaintiffs and two other groups of plaintiffs also objected to the terms of the settlement. Id. at 700.
In June 2014, Judge Wu circulated a tentative opinion on class certification and preliminary settlement approval.31×31. See Civil Minutes –– General, In re Hyundai & Kia Fuel Econ. Litig., No. MDL 13-2424, at 1, 2014 WL 12594158, at *1 (C.D. Cal. June 26, 2014). After concluding that the class satisfied the requirements of Rule 23(a), he considered whether the class met the predominance and superiority requirements of Rule 23(b)(3).32×32. Id. at 7–12, 2014 WL 12594158, at *4–7. If the case were being certified for litigation, the predominance inquiry would require “extensive choice of law analysis.”33×33. Id. at 10, 2014 WL 12594158, at *6. It was “unclear,” however, “to what extent a choice of law analysis is necessary” for certification of a nationwide settlement class.34×34. Id. He ultimately determined that “the [c]ourt need not conduct a further conflict of laws analysis” because the Gentry objectors “offer[ed] little direct argument . . . that [a choice-of-law] analysis [was] necessary in the settlement context.”35×35. Civil Minutes –– General, Fuel Econ. Litig., No. MDL 13-2424, at 3 (C.D. Cal. July 24, 2014), ECF No. 290 [hereinafter July Tentative Ruling]. In August, Judge Wu certified a nationwide settlement class and preliminarily approved the proposed settlement.36×36. Fuel Econ. Litig., 926 F.3d at 554. The following summer, the court granted final approval to the settlement and awarded attorneys’ fees. Id. at 555. Various objectors appealed the orders on class certification, settlement approval, and attorneys’ fees. Id.
In 2018, a divided panel of the Ninth Circuit vacated the district court’s order certifying the settlement class.37×37. Id. Writing for the majority,38×38. Judge Ikuta was joined by Judge Kleinfeld. Judge Ikuta held that the district court abused its discretion when it failed to consider differences in state law prior to certifying the class.39×39. In re Hyundai & Kia Fuel Econ. Litig., 881 F.3d 679, 702 & n.23 (9th Cir. 2018). The panel concluded that Mazza required the district court to conduct a choice-of-law analysis as part of its predominance inquiry in both the litigation and settlement context.40×40. See id. at 702. Without deciding the issue, they implied that a nationwide class could not be certified if the laws of multiple states applied to the claims.41×41. See id. at 703. The panel also found that the district court improperly extended a presumption of reliance to used-car purchasers, who were not necessarily exposed to the inaccurate fuel economy estimates, id. at 703–05, and that by relying on the settling parties’ estimate of the settlement value, the court failed to provide “an adequate explanation of whether the award” of attorneys’ fees was proportionate to the settlement’s benefits, id. at 706. Judge Nguyen dissented. She argued that the objectors failed to meet their burden under California’s choice-of-law rule to prove that non-California law should govern their claims. Id. at 709–11 (Nguyen, J., dissenting). She also disagreed with the majority’s conclusions that used-car owners needed to demonstrate individualized reliance, id. at 715–16, and that the district court had abused its discretion by awarding attorneys’ fees without independently calculating the value of the settlement, id. at 716–19.
On rehearing en banc, the Ninth Circuit affirmed the district court’s orders.42×42. Fuel Econ. Litig., 926 F.3d at 572. Writing for the majority,43×43. Chief Judge Thomas and Judges Fletcher, Berzon, Bybee, Christen, and Hurwitz joined in the opinion. Judge Rawlinson joined in all of the opinion except the portion concerning the attorneys’ fee award, from which she dissented. Judge Nguyen began by emphasizing that “[t]he criteria for class certification are applied differently in litigation classes and settlement classes.”44×44. Fuel Econ. Litig., 926 F.3d at 556. While “[s]ettlement benefits cannot form part of a Rule 23(b)(3) analysis,”45×45. Id. at 558 (alteration in original) (quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1022 (9th Cir. 1998) (emphasis added)). Such benefits include “prompt and fair compensation without the risk and cost of litigation.” Id. settlement is “a factor in the calculus,”46×46. Id. at 559 (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 622 (1997)). particularly because the court need not consider “whether the case, if tried, would present intractable management problems.”47×47. Id. at 558 (quoting Amchem, 521 U.S. at 620). The Ninth Circuit held that the district court had not abused its discretion by finding that common issues predominated, despite variations in state law.48×48. Id. at 566. In addition to its discussion of the choice-of-law issue, the court noted that the common issues identified by the district court were the “types of common issues . . . [that] can establish predominance in nationwide class actions.” Id. at 559. Because the objectors had failed to establish the applicability of any foreign law under California’s choice-of-law rules, California law could be applied to the entire class.49×49. Id. at 561–64. The majority also held that even if used-car purchasers were not entitled to a presumption of reliance under California law, individualized issues of reliance “primarily implicate trial management issues, which we do not consider when conducting a predominance analysis for a settlement class.” Id. at 560. The district court was not “obligated to address choice-of-law issues beyond those raised by the objectors.”50×50. Id. at 562. The court also rejected the objectors’ other arguments against certification, finding that the application of California law satisfied due process, id. at 564–66, and that the class counsel provided adequate representation, id. at 566–67. Finally, the court found that the settlement was fair51×51. The parties provided sufficient notice of the settlement to class members, id. at 567–68, the claim forms were not unduly burdensome, id. at 568, and there was no evidence of collusion between the plaintiffs’ counsel and the defendants, id. at 568–70. and that the attorneys’ fees were reasonable.52×52. The district court “properly exercised its discretion in . . . using the lodestar method” to calculate the fee award without independently estimating the value of the settlement. Id. at 570.
Judge Ikuta dissented.53×53. She was joined by Judges Kleinfeld and M. Smith. Judge Rawlinson joined in part. She argued that the majority’s decision was inconsistent with the Supreme Court’s ruling in Amchem and with Rule 23.54×54. Id. at 573 (Ikuta, J., dissenting). Variations in state law raise “problems beyond those of just manageability” that may prevent certification of a nationwide class.55×55. Id. at 575 (quoting In re Warfarin Sodium Antitrust Litig., 391 F.3d 516, 529–30 (3d Cir. 2004)). Because of these potential problems, the dissent argued that the district court erred when it concluded “that it was not necessary to consider what state law applied.”56×56. Id. at 575. She also argued that the objectors had effectively demonstrated that Virginia law governed some of the plaintiffs’ claims, id. at 577–79, and that the district court improperly awarded attorneys’ fees without calculating the value of the settlement, id. at 579–82.
Despite Judge Ikuta’s protestations, the Ninth Circuit’s decision is not a giveaway to defendants and class counsel hoping to secure nationwide settlements of state law claims: the decision correctly recognized that variations in state law are relevant to the certification of nationwide settlement classes. The Ninth Circuit declined to embrace the reasoning of the district court and the settling parties, who argued that variations in state law can be ignored when certifying a class for settlement only. By assuming instead that the district court had applied California law to the entire class, the court implicitly acknowledged that approach would have stretched Amchem too far. While settlement eliminates the need to consider trial manageability issues, judges must give undiluted attention to the effects of variations in state law on the other requirements of Rule 23. The court’s careful choice-of-law analysis allowed it to conclude that common issues predominated without relying on the settlement context to justify a more lenient certification standard. Ultimately, the decision rules out two extreme conclusions — that variations in state law are never relevant or always fatal to certification of nationwide classes for settlement — but leaves such settlements open to attack when objectors can demonstrate that multiple states’ laws must apply.
The majority quietly rejected the settling parties’ argument — and the tentative conclusion of the district court — that settlement rendered any choice-of-law analysis unnecessary. Before the district court, the settling parties argued that “because settlement classes do not present [trial] manageability concerns,” variations in state law were irrelevant to the court’s predominance analysis in the settlement context.57×57. Settling Parties’ Supplemental Brief in Support of Preliminary Approval of Class Settlement and of Certification of Settlement Class at 2, In re Hyundai & Kia Fuel Econ. Litig., MDL 13-02424 (C.D. Cal. July 24, 2014) [hereinafter Settling Parties’ Brief]. They drew support from Sullivan v. DB Investments, Inc.,58×58. 667 F.3d 273 (3d Cir. 2011) (en banc). in which the Third Circuit concluded that “state law variations are largely ‘irrelevant to certification of a settlement class.’”59×59. Id. at 304 (quoting Warfarin, 391 F.3d at 529); see also id. at 304–07 (affirming certification despite the fact that many class members could not state a claim under their states’ antitrust laws); Settling Parties’ Brief, supra note 57, at 6 (citing Sullivan, 667 F.3d at 303). The district court seemed to adopt this reasoning, concluding it “need not conduct a further conflict of laws analysis,” not because the objectors had failed to prove that foreign law should apply, but because they offered “little direct argument” that choice-of-law analysis “is necessary in the settlement context.”60×60. July Tentative Ruling, supra note 35, at 3. But when the settling parties advanced this argument again on appeal, asserting that “settlement eliminated the need” for the district court to determine what state law applied,61×61. Fuel Econ. Litig., 926 F.3d at 564 n.9. the en banc panel relegated this argument to a footnote and assumed instead that the district court had applied California law to the class.62×62. See id. This is especially significant because the majority explicitly relied on the settlement context to find that common issues predominated despite the fact that California law may have required some issues to be tried individually.63×63. See id. at 560. By ignoring the district court’s tentative rulings, the Ninth Circuit indicated that it was unwilling to declare variations of state law irrelevant to the certification of settlement classes.
The Ninth Circuit’s decision did not expressly reject the district court’s conclusion (or the Third Circuit’s holding in Sullivan), but its choice to affirm on other grounds is a tacit admission that their reasoning runs afoul of Amchem. In Amchem, the Supreme Court held that courts certifying a class for settlement need not consider “whether the case, if tried, would present intractable management problems.”64×64. Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620 (1997). The Court stressed, however, that the “other specifications” of Rule 23 “demand undiluted, even heightened, attention in the settlement context.”65×65. Id. In cases where nationwide classes were certified for litigation, appellate courts have emphasized that variations in state law implicate the “other specifications” of Rule 23, including the core question of whether common issues predominate.66×66. See Mullenix, supra note 3, at 652 (noting that variations in state law “implicate . . . predominance of common questions, superiority, and manageability”). Courts often consider the effect of variations in state law on predominance and trial manageability separately. See, e.g., Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1189–90, 1192 (9th Cir. 2001); Castano v. Am. Tobacco Co., 84 F.3d 734, 741–44 (5th Cir. 1996). But see Fuel Econ. Litig., 926 F.3d at 563 (arguing that Zinser treated “state law variations as a . . . trial manageability concern”). The concern that variations in state law may “swamp” common issues is conceptually distinct from the effect such variations may have on trial manageability.67×67. Castano, 84 F.3d at 741; see also Georgine v. Amchem Prods., Inc., 83 F.3d 610, 618 (3d Cir. 1996), aff’d sub nom. Amchem Prods., Inc. v. Windsor, 521 U.S. 591 (1997). While courts have relied on Amchem to certify for settlement class actions that might not be suitable for class-wide litigation,68×68. See Erichson, supra note 3, at 976; Lahav, supra note 2, at 1498. most have argued that common issues predominated despite variations in state law, not that settlement eliminated the need to consider those variations at all.69×69. See, e.g., In re Prudential Ins. Co. Litig., 148 F.3d 283, 315 (3d Cir. 1998) (affirming certification of a nationwide settlement class where the objector “failed to demonstrate that the differences in applicable state law” prevented “grouping similar state laws together and applying them as a unit”); Hanlon v. Chrysler Corp., 150 F.3d 1011, 1022–23 (9th Cir. 1998) (affirming certification of a nationwide settlement class because “idiosyncratic differences between state consumer protection laws [were] not sufficiently substantive to predominate over the shared claims”). The Ninth Circuit was right not to rely on the settlement context alone to justify certification.
Instead, the Ninth Circuit correctly applied California’s choice-of-law rule to affirm the district court’s certification order without violating Amchem. Federal courts sitting in diversity use the forum state’s choice-of-law rule to determine which law or laws apply.70×70. See Fuel Econ. Litig., 926 F.3d at 561; Patrick Woolley, Erie and Choice of Law After the Class Action Fairness Act, 80 Tul. L. Rev. 1723, 1735 (2006). When one state’s law can apply to the entire class, variations in state law should present no obstacle to certification, even for litigation classes.71×71. See 2 William B. Rubenstein, Newberg on Class Actions § 4:61 (5th ed. 2012); Woolley, supra note 70, at 1739 (criticizing courts for “conflat[ing] two separate inquiries: (1) whether the party who bears the choice-of-law burden has met its burden . . . and (2) whether the party seeking certification has demonstrated . . . that certification of a class suit would be appropriate”). California’s choice-of-law rule puts the burden on the proponent of foreign law to prove it should apply to the case at hand.72×72. See Wash. Mut. Bank, FA v. Superior Court, 15 P.3d 1071, 1080–81 (Cal. 2001). Before the district court, only the Gentry plaintiffs argued that non-California law applied to their claims,73×73. See July Tentative Ruling, supra note 35, at 2. but they failed to point out any actual differences between California and Virginia consumer protection law.74×74. Settling Parties’ Brief, supra note 57, at 9–10. The Gentry plaintiffs abandoned the argument before the June 2014 class certification hearing. See July Tentative Ruling, supra note 35, at 2–3. By finding that the objectors had not met their burden to prove that foreign law should apply to the claims of any class members, the Ninth Circuit was able to distinguish Mazza, apply one law to the entire class, and conclude that common issues predominated without declaring that variations in state law are irrelevant to certification for settlement.75×75. See Fuel Econ. Litig., 926 F.3d at 561–63. By contrast, the majority’s assertion that “predominance is ‘readily met’ in cases alleging consumer fraud,” see id. at 559 (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 625 (1997)), would have been a highly questionable basis for affirming certification had multiple states’ laws applied. In the twenty years since Amchem, variations in choice of law have become a much larger obstacle to class certification for litigation. See Samuel Issacharoff, Settled Expectations in a World of Unsettled Law: Choice of Law After the Class Action Fairness Act, 106 Colum. L. Rev. 1839, 1859–60 (2006). As Judge Ikuta pointed out, the Ninth Circuit “rejected that very conclusion” in Mazza. Fuel Econ. Litig., 926 F.3d at 576 (Ikuta, J., dissenting); see also Mazza v. Am. Honda Motor Co., 666 F.3d 581, 594 (9th Cir. 2012) (implying that variations in state law precluded certification of a single nationwide class).
While the Ninth Circuit’s en banc decision is more favorable to class certification than the vacated panel decision, it does not give parties seeking nationwide class certification for settlement carte blanche to ignore variations in state law. Many members of the class action bar saw the decision as an unqualified victory for proponents of nationwide settlements.76×76. See sources cited supra note 9. But the court’s careful application of California’s choice-of-law rules suggests that variations in state law are relevant to class certification in the settlement context. In fact, after its decision in Fuel Economy Litigation, the Ninth Circuit decertified a different nationwide settlement class because the district court failed to consider how significant differences in state antitrust laws affected representation, class certification, and the terms of the settlement.77×77. In re Lithium Ion Batteries Antitrust Litig., 777 F. App’x 221, 223 (9th Cir. 2019). While favorable choice-of-law rules and low rates of objection may shield nationwide settlement classes from some of the scrutiny those classes would face in the litigation context,78×78. See Theodore Eisenberg & Geoffrey Miller, The Role of Opt-Outs and Objectors in Class Action Litigation: Theoretical and Empirical Issues, 57 Vand. L. Rev. 1529, 1558 (2004) (finding that only about one percent of class members object to settlements); cf. Silberman, supra note 2, at 2017 (arguing that “so long as one state is prepared to apply a single law . . . that state will attract litigation on behalf of the nationwide class”).
parties hoping to settle class actions on a nationwide basis would be wise to negotiate settlements — and define settlement classes — that account for differences in state law.