The Dukes Decision
At issue in Dukes was whether, in attempting to certify their class, the plaintiffs could establish commonality as required by Rule 23(a). The Court ultimately concluded that the plaintiffs’ expert, who testified to a general pattern of discrimination, could not determine whether the pattern was pervasive enough to constitute a general operating policy of discrimination. The plaintiffs and defendant disagreed as to whether the expert’s testimony was even admissible under the Daubert standard. The district court had concluded that Daubert-level scrutiny of expert testimony was not necessary at that stage—a determination upheld by the Ninth Circuit Court of Appeals. Dukes v. Wal-Mart Stores, Inc., 222 F.R.D. 189 (N.D. Cal. 2004), aff’d, 603 F.3d 571 (9th Cir. 2010). In its opinion, the Supreme Court addressed the issue by noting that the Court “doubt[ed]” the district court’s conclusion that Daubert did not apply to expert testimony at the certification stage, but the Court did not conclusively resolve the issue. Dukes, 131 S. Ct. at 2554.
Daubert in Full Force
A number of circuit courts conduct a full Daubert analysis at the class certification stage. Even before Dukes, the Seventh Circuit determined that Daubert applied at class certification. In American Honda Motor Co. v. Allen, the court declared unambiguously that “when an expert’s report or testimony is critical to class certification . . . a district court must conclusively rule on any challenge to the expert’s qualifications or submissions prior to ruling on a class certification motion.” 600 F.3d 813, 815–16 (7th Cir. 2010) (citation omitted). The Seventh Circuit found that the plaintiffs’ expert’s testimony as to the cause of a potential motorcycle defect did not meet Daubert’s standards because it was not peer-reviewed or developed independently of the litigation process. The appellate court held that the district court erred “by failing to clearly resolve the issue of its admissibility before certifying the class. . . .” Id. at 817; see also Messner v. Northshore Univ. HealthSystem, 669 F.3d 802 (7th Cir. 2012) (vacating district court opinion on class certification for failure to rule on a Daubert motion).
The Eleventh Circuit reached a similar conclusion in Sher v. Raytheon Co., 419 F. App’x 887, 890 (11th Cir. 2011) (unpublished). In a groundwater contamination class action, the court, citing the Seventh Circuit’s language in Allen, reversed a district court’s decision to certify a class of homeowners. It explained that “[h]ere the district court refused to conduct a Daubert-like critique of the proffered experts’ qualifications. This was error. As we have noted, a district court must make the necessary factual and legal inquiries and decide all relevant contested issues prior to certification.” Id. at 890–91; see also Young v. Nationwide Mut. Ins. Co., No. 11-5015/11-5016/11-5018/11-5019/11-5020, 2012 U.S. Dist. LEXIS 18625, at *20 (6th Cir. Sept. 5, 2012) (approving the “full Daubert analysis” performed by the district court). The court then remanded the case for evaluation of the expert’s qualification and proposed damages calculation.
Following Dukes, the Ninth Circuit also has employed a full Daubert analysis at the class certification stage. In Ellis v. Costco Wholesale Corp., the court noted that it had to follow “new precedent altering existing case law . . .” and required a full Daubert analysis of any expert opinion offered in support of class certification. 657 F.3d 970, 974, 981 (9th Cir. 2011) (vacated and remanded on other grounds). Accordingly, it ruled that the district court correctly scrutinized both the plaintiffs’ and defendant’s experts’ testimony on methods of aggregating data when ruling on the defendant’s motion to strike the expert’s testimony. In denying the defendant’s motion, the district court noted that the defendant had not discredited the expert’s results or methods. The Ninth Circuit found that the district court “correctly applied the evidentiary standard set forth in Daubert. . . .” Id. at 982.
California district courts have followed suit. Even before the Ninth Circuit clarified its position in Ellis, the district court in In re Aftermarket Automotive Lighting Products Antitrust Litigation interpreted the dicta in Dukes to mean that Daubert applied during class certification:
The Supreme Court has recently strongly indicated that Daubert should be applied to expert testimony at the certification stage of class action proceedings. Accordingly, in order to grant class certification, this Court must first determine whether it may rely on the methodology used by Plaintiffs’ expert to decide whether the claims in this case are amenable to common proof.
276 F.R.D. 364, 370 (C.D. Cal. 2011) (citation omitted).
The court then carried out a full Daubert analysis of the expert’s regression analysis techniques before certifying the class.
The Central District of California has continued to apply the Daubert test after Ellis and Dukes. In Keegan v. American Honda Motor Co., a class action regarding an alleged automobile defect, the court explained that it understood Ellis and Dukes to require a Daubert-style analysis of an expert’s testimony:
Before addressing the merits of the certification motion, the court must consider the parties’ challenges under Daubert . . . to the opposing parties’ experts. While courts in this circuit have previously concluded that expert testimony is admissible in evaluating class certification without a rigorous Daubert inquiry, the Supreme Court in Dukes “doubt[ed] that this is so.” After Dukes, the Ninth Circuit approved the application of Daubert to expert testimony presented in support of or opposition to a motion for class certification.
No. CV10-9508, 2012 U.S. Dist. LEXIS 91394, at *3 (C.D. Cal. June 12, 2012) (citations omitted) (second alteration in original).
Accordingly, when the court found that a portion of the expert’s methodology violated established scientific practices, it excluded that portion of the expert’s testimony from the class certification analysis.
Similarly, in Ralston v. Mortgage Investors Group, Inc., the Northern District of California cited Ellis and Dukes in requiring a Daubert analysis of an expert’s testimony on class certification. No. 08-536, 2011 U.S. Dist. LEXIS 138149, at *11 (N.D. Cal. Nov. 30, 2011). To defeat the defendant’s motion to exclude an expert’s damages testimony, the plaintiff had to show
that the theory and model of damages is relevant and sufficiently reliable in order to satisfy common elements of class-wide proof, as required by Fed. R. Civ. P. 23. The touchstone of this analysis is reliability of the methodology, and not a “rigorous” assessment of the theory or calculation of damages on the merits.
2011 U.S. Dist. LEXIS 138149, at *24–25.
The court elaborated that this Daubert-style analysis was a “preliminary and predicate step to the trial court’s rigorous analysis of the evidence supporting class certification.” Id. at *25.
Courts espousing a full Daubert analysis at the class certification stage have not discussed whether the outcome of this analysis would carry through to trial, thus leaving open the possibility that courts would need to engage in a further Daubert inquiry at trial. See, e.g., Behrend v. Comcast Corp., 655 F.3d 182, 199 (3d Cir. 2011) (“We allow preliminary merits inquiries when necessary for Rule 23 because of the potentially ‘decisive effect on litigation’ of a certification decision, but those inquiries remain limited and non-binding on the merits at trial. . . .”) (citation omitted), cert. granted, 80 U.S.L.W. 3707 (2012) (No. 11-864).
A Limited Daubert Analysis
In other circuits, courts have declined to read Dukes as requiring a full Daubert analysis at the class certification stage. In In re Zurn Pex Plumbing Products Liability Litigation, 644 F.3d 604, 613 (8th Cir. 2011), the Eighth Circuit rejected the idea of a full Daubert-style examination at class certification. It instead promoted a “tailored Daubert analysis,” which “examined the reliability of the expert opinions in light of the available evidence and the purpose for which they were offered” but which stopped short of determining whether expert opinions would be admitted at trial. Id. at 612–13. A full Daubert analysis before merits discovery would have been “impractical,” the court explained, because the disputed expert testimony regarding the average time for product failure would have evolved after merits discovery provided access to warranty claims data. Id. at 612. For that reason, Daubertwas applicable only insofar as it helped the court determine the existence of the requirements of Rule 23(a):
[Defendant’s] desire for an exhaustive and conclusive Daubert inquiry before the completion of merits discovery cannot be reconciled with the inherently preliminary nature of pretrial evidentiary and class certification rulings.
The main purpose of Daubert exclusion is to protect juries from being swayed by dubious scientific testimony. That interest is not implicated at the class certification stage where the judge is the decision maker.
Id. at 613.
Finally, the court highlighted that any Daubert-style analysis done at the class certification stage should not be determinative of what evidence ultimately would be admissible at trial.
Courts in the Third Circuit have taken a similar approach. In Behrend v. Comcast Corp., an antitrust class action, the majority answered the dissent’s argument that the Dukes dicta commanded a full Daubert analysis by stating that “it need not turn class certification into a mini-trial.” 655 F.3d at 204, n.13. The majority emphasized that the task of the trial court, as it understood Dukes, was
to evaluate whether an expert is presenting a model which could evolve to become admissible evidence, and not requiring a district court to determine if a model is perfect at the certification stage . . . [A] district court does not have to determine which model should be used at the time of class certification.
Thus, the Third Circuit upheld the district court’s decision, in which the district court had examined whether the models presented “could be refined between the time when class certification was granted and trial so as to comply with Daubert.” Id.
In Bruce v. Harley-Davidson Motor Co., No. CV 09-6588 CAS (RZx), 2012 U.S. Dist. LEXIS 36723 (C.D. Cal. Jan. 26, 2012), one court broke from the ranks of California district courts that have required a full Daubert analysis at class certification. After considering Dukes, Zurn, and Behrend, the court concluded that Zurn’s “focused” approach was “the appropriate application of Daubert at the class certification stage.” Id. at *13. Applying that approach, the court declined to admit the plaintiffs’ expert’s testimony regarding the commonality of a defect in the defendant’s product. The court reasoned that the model the expert presented was not sufficiently scientifically supported: “[The expert] has not adequately explained the scientific basis for his proposed standard, which has not been accepted in the field. . . . [E]vidence . . . does not establish that [the expert’]s rule . . . is scientifically valid.” Id. at *15–16; but see Dewey v. Volkswagen AG, 681 F.3d 170 (3d Cir. 2012) (approving magistrate judge’s admission of expert evidence on a class settlement).
The court further noted that the expert’s testimony failed to eliminate other potential causes of the product’s defect and that the expert should have attempted to have his methods peer-reviewed. Ultimately, the court seemed to conclude that although a full Daubert analysis was inappropriate in general, it should be used where merits discovery would not fill in the gaps in an expert’s testimony.
The standard for admissibility of expert testimony at the class certification stage remains uncertain, with courts applying the full Daubert standard or a variation thereof. The U.S. Supreme Court recently indicated its willingness to clarify this ambiguity when it granted certiorari in Behrend. In that case, the Court intends to examine “whether a district court may certify a class action without resolving whether the plaintiff class has introduced admissible evidence, including expert testimony, to show that the case is susceptible to awarding damages on a class-wide basis.” 80 U.S.L.W. 3707 (2012) (No. 11-864).
It is expected that the Supreme Court’s ruling in Behrend will provide lower courts with more guidance on the proper standard to be applied when assessing the admissibility of expert opinion at the class certification stage. However, that decision could usher in a wave of new questions, such as whether a court’s evidentiary ruling at the class certification stage would apply at trial. It is unclear whether a trial court would be restricted to the findings it makes at the class certification phase of the case or whether it could revisit admissibility issues before trial. Several courts have suggested, as the Behrend court did, that its “preliminary merits inquiries . . . remain limited and non-binding on the merits at trial,” 655 F.3d at 199; however, should the Supreme Court require a full Daubert analysis at the class certification stage, lower courts may be inclined to allow broader discovery before class certification, and they may be less likely to revisit expert admissibility issues at subsequent stages of the case.
Given the uncertainty in this area, counsel should be mindful of the Supreme Court’s comments in Dukes and be prepared to respond to a full Daubert challenge at the class certification stage, as courts engaging in a “rigorous analysis” of Rule 23 may apply similar rigor in their review of expert testimony. To avoid having all or a portion of their expert’s opinion stricken, counsel should ensure that any expert opinion offered at the class certification stage satisfies Daubert in its strictest sense.
Keywords: litigation, pretrial practice and discovery, Daubert analysis, evidentiary standard, Federal Rule of Civil Procedure 23, merits discovery
Dennis S. Ellis, Katherine F. Murray, and Nicholas J. Begakis are with Paul Hastings LLP, in Los Angeles, California. Summer associate Margaret Buckles assisted in the preparation of this article.