Key Issue
Whether the district court properly denied class certification after excluding supporting expert testimony under Daubert v. Merrell Dow Pharmaceuticals, Inc.63
Background
Grodzitsky v. American Honda Motor Co. was brought in the Central District of California as a putative class action alleging that the window regulators installed in certain Honda vehicles were defectively designed.64 In support of their motion for class certification, the plaintiffs submitted an expert opinion regarding the alleged defect.65 The plaintiffs' expert opined that Honda had failed to test its window regulators with the kind of vibrational stresses they would incur during operation.66 Accordingly, the vehicles at issue allegedly suffered from a common defect: excessive "vibration[-] induced metal fatigue" that led to premature failure in the window regulators.67
Honda moved to exclude the expert's opinions as deficient under Daubert, and the district court granted Honda's motion.68 The district court found that the opinion was not reliable because its opinion about the typical lifespan of an automobile's window regulator was based on the expert's own conclusory assertions rather than on industry standards, peer-reviewed literature, or other external information.69 The district court was also skeptical of the methodology the expert employed to demonstrate the existence of a defect in the regulators. The expert examined "an extremely small sample size of window regulators"—twenty-six compared to over four hundred thousand installed in the class vehicles—and therefore, according to the district court, lacked a solid basis for opining that failed window regulators generally failed because of their alleged vulnerability to vibration.70
The district court excluded the expert's opinion and then subsequently denied the plaintiffs' motion for class certification, because without the expert's opinion as to the nature of a common defect in the Honda window regulators, the plaintiffs could not make the requisite showing of commonality under Rule 23.71 The plaintiffs sought a Rule 23(f) appeal of the order denying class certification, which the Ninth Circuit granted.72
Decision
On appeal, a split Ninth Circuit panel affirmed the district court's exclusion of the expert's testimony.73 The panel likewise agreed that the district court properly denied class certification because plaintiffs failed to offer any other evidence of a common design defect.74
Writing in dissent, Judge Murguia conceded that there were serious flaws with the expert's report, particularly with his ultimate conclusion that the window regulators likely failed because of their vibration-vulnerability defect.75 However, she reasoned that the district court went too far when it excluded the entirety of the expert's testimony.76 Those parts of the expert's testimony opining as to the existence of a design defect in the window regulators should not have been excluded, since they were adequately rooted in the expert's application of his forensic engineering expertise to his personal examination of sample regulators.77 Because those portions of the expert's opinion should not have been excluded, the district court should have considered them when evaluating class certification.78 Judge Murguia contended that plaintiffs "do not need to demonstrate that they will prevail on the merits to satisfy commonality."79 It was therefore enough that the expert's testimony offered proof that the window regulators all suffered from a common design defect, such that "a class-wide proceeding would generate common answers apt to drive the resolution of the litigation."80
The majority responded that the district court could not grant class certification on the basis of expert evidence "riddled with scientific and methodological flaws."81 Rather, the district court was required to conduct a "rigorous analysis" of plaintiffs' claim that the proposed class suffered from a common design defect.82 The majority concluded that the district court had conducted that analysis and determined that the expert's opinions going to commonality were afflicted by the same flaws that rendered the testimony unreliable.83
Thoughts & Takeaways
Grodzitsky is notable because both the majority and dissent appear to agree that the district court should not have considered the expert report in support of class certification to the extent it was inadmissible under Daubert. Their reasoning is in apparent tension with Sali v. Corona Regional Medical Center,84 where a Ninth Circuit panel held for the first time that the "evidentiary proof a plaintiff must submit in support of class certification . . . need not be admissible evidence."85
In Sali, named plaintiffs brought employment claims on behalf of several proposed classes of registered nurses, alleging that the class members were not fully compensated and were denied meal periods and rest breaks.86 In denying class certification, the district court held that the named plaintiffs failed to satisfy Rule 23(a)'s typicality requirement because plaintiffs did "not offer any admissible evidence of [their] injuries."87 Plaintiffs relied on a single declaration authored by a paralegal for plaintiffs' counsel (the "Ruiz Declaration").88 Ruiz prepared a spreadsheet of data he extracted from "a random sampling of" two of the named plaintiffs' timekeeper records and concluded from that data that the named plaintiffs did not receive full pay, meal periods, and rest breaks.89 The Court excluded the Ruiz Declaration because it suffered from "multiple evidentiary issues": (1) Ruiz could not "authenticate the manipulated Excel Spreadsheets and other data that he relied upon to conduct his analysis"; and (2) Ruiz had "not demonstrated that he [was] technically qualified to conduct this analysis" under Rule 702.90
A Ninth Circuit panel reversed the district court's denial of class certification and held that the district court erred by striking the Ruiz Declaration "solely on the basis that the evidence is inadmissible at trial."91 The panel's reasoning emphasized that class certification is a "preliminary stage," and the district court's "rigorous analysis" of the Rule 23 requirements should therefore not rise to the level of "a mini-trial" that rests on "formal strictures" of evidentiary requirements.92 The Ninth Circuit held that, in evaluating expert testimony at class certification, "a district court should evaluate admissibility under . . . Daubert."93 But this admissibility analysis "must not be dispositive"— rather, it should go only to the weight of the offered proof.94 As a practical matter, "[l]imiting class-certification- stage proof to admissible evidence risks terminating actions" prematurely because "the evidence needed to prove a class's case often lies in a defendant's possession and may be obtained only through discovery."95 Here, the district court erred by excluding "evidence that likely could have been presented in an admissible form at trial."96
While Sali signaled a lower hurdle for plaintiffs at class certification by allowing district courts to grant certification based on evidence that would be inadmissible at trial, Grodzitsky holds that the district court should not give any weight to an inadmissible expert report.
One way of reconciling these decisions might be to limit Sali's reach to situations where a district court rejects otherwise-persuasive evidence on the basis of strictly formal flaws that could be cured later, e.g. after conducting more discovery. Sali emphasized that the district court erred because it elevated form over substance by failing to consider unauthenticated data that supported the merits of plaintiffs' claim because the data "likely could have been presented in an admissible form at trial."97 On this reading of Sali, district courts may still refuse to consider evidence that is inadmissible for more fundamental reasons, e.g. where an expert's opinion rests on an unreliable methodology and thereby undermines the merits of plaintiffs' claim, as in Grodzitsky where the expert "failed to provide a reliable opinion demonstrating a common defect."98 Unlike issues of document authentication, some methodological flaws in expert opinions cannot be cured simply by more discovery as Sali contemplates, but rather suggest that plaintiffs lack the ability to prove their claims on a classwide basis. After all, the plaintiffs in Grodzitsky were already on their "third try at class certification" and despite substituting in a new expert, plaintiffs were unable to overcome the "same shortcomings" that plagued plaintiffs' previous expert testimony.99
The tension between the opinions in Sali and Grodzitsky suggests that the Ninth Circuit may be faced with appeals in the future challenging how district courts have applied these decisions.
In addition to these issues, Grodzitsky offers a useful counterbalance to Sali's framing of class certification as a preliminary stage. While discovery may be ongoing at the class certification stage, a district court's decision to certify a class is a significant milestone in any litigation and has powerful consequences for both parties.100 Grodzitsky reaffirms that courts should not make that decision freely, without regard to the court's gatekeeping role under Daubert and plaintiffs' evidentiary burden to satisfy the Rule 23 requirements. Despite reaching different conclusions, Grodzitsky and Sali are consistent in emphasizing that a district court should engage with the substance of plaintiffs' class certification evidence. Defendants and plaintiffs alike can benefit from that guidance by preparing to attack and defend class certification evidence on substantive grounds.
Read the opinion here.
Footnotes
63. 509 U.S. 579 (1993).
64. Grodzitsky v. Am. Honda Motor Co., —F.3d—, No. 18-55417, 2020 WL 2050659, at *1 (9th Cir. Apr. 29, 2020).
65. Id. at *2.
66. Id.
67. Id.
68. Id. at *3.
69. Id.
70. Id. at *3.
71. Id. at *4.
72. Id.
73. Id. at *4-5.
74. Id. at *7.
75. Grodzitsky, 2020 WL 2050659, at *7 (Murguia, J., dissenting).
76. Id. at *8.
77. Id.
78. Id. at *9.
79. Id. at *8.
80. Id. (internal quotation marks and citation omitted).
81. Id. at *6 (majority opinion).
82. Id.
83. Id. at *6-7.
84. 909 F.3d 996, 1004 (9th Cir. 2018).
85. Id. at 1004.
86. Sali v. Universal Health Services of Rancho Springs, Inc., No. 14-985 PSG (JPRx), 2015 WL 12656937, at *1 (C.D. Cal. June 3, 2015).
87. Id. at *10.
88. Id.
89. Sali, 909 F.3d at 1003.
90. Sali, 2015 WL 12656937, at *10.
91. Sali, 909 F.3d at 1004-07.
92. Id. at 1004.
93. Id.
94. Id.
95. Id.
96. Id. at 1006.
9.7 Sali, 909 F.3d at 1006.
98. Grodzitsky, 2020 WL 2050659, at *6.
99. Grodzitsky v. Am. Honda Motor Co., No. 2:12-cv-001142, 2017 WL 8943159, at *3 (C.D. Cal. Oct. 30, 2017).
100. See Sali v. Corona Regional Med. Center, 907 F.3d 1185, 1188 (9th Cir. 2019) (Bea, J.), dissent from denial of reh'g en banc ("[A] district court's determination on class certification often 'sounds the death knell of the litigation,' whether by dismissal, if class certification is denied, or by settlement, if class certification is granted.").
Please click here to read the Newsletter.
To view original article, please click here.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
