LAW-Class Actions After 'Dukes'

The following information is used for educational purposes only.



Class Actions After 'Dukes'

Elise M. Bloom and Amanda D. Haverstick

New York Law Journal

10-24-2011


Nearly 80 percent of civil trials involve some form of expert testimony, and such testimony plays a particularly prominent role in large-scale class actions.1 The standard of review courts should apply to plaintiffs' expert evidence at the class certification stage, however, remains an unsettled question in need of clarification. The need is particularly pressing in the employment discrimination context, where putative class plaintiffs generally must present expert proof to satisfy the Rule 23 class certification requirements. Indeed, whether plaintiffs prevail at certification often will turn largely on the extent of scrutiny the court applies to their expert evidence. Without definitive guidance from the U.S. Supreme Court or Congress, however, courts have taken a wide range of approaches on this issue, leading to inconsistent outcomes.

Courts' choice on the level of scrutiny to apply to plaintiffs' expert evidence involves two related, but analytically distinct, questions:

1. Must the expert's opinions satisfy the admissibility standards of Daubert and Fed. R. Evid. (FRE) 702?2 and

2. What standard of review should courts apply to the expert's substantive opinions when weighing their probative value under Rule 23?3

The U.S. Supreme Court touched on both these questions in Wal-mart Stores Inc. v. Dukes,4 but it did not rule definitively on either of them. Further guidance is needed.

Although Dukes did resolve the long-unsettled question of whether, and to what extent, courts should consider the merits of plaintiffs' underlying claims at the class certification stage—holding that courts must conduct a rigorous analysis of the merits if they bear on the Rule 23 determination5—Dukes did not resolve what standard of review governs plaintiff's expert evidence that bears on that same determination.

Role of Expert Proof

It is beyond dispute that the fates of the parties in an employment discrimination class action rise and fall at the class certification stage. If a court certifies the employee-plaintiffs' class, the defendant-employer typically deems the benefit of potential victory at trial to be outweighed by the risks and costs of continued litigation; if the court denies certification, the employee-plaintiffs generally do not deem pursuit of their individual claims to be worthwhile or economically feasible. Either result usually leads to settlement and conclusion of the litigation.

Plaintiffs' expert evidence plays a critical role in this often outcome-determinative class certification decision, especially in employment discrimination cases where statistical evidence is generally a prerequisite to plaintiffs' satisfying the commonality, typicality, and other elements of Rule 23.6

For example, in a Reduction-in-Force-based disparate impact/pattern-or-practice case, where plaintiffs' proposed class includes employees from different job positions, geographic locations, and/or company departments, statistical regression studies can show that discharge rates of protected-group class members are consistently adverse when compared to that of others—irrespective of position, location and department; and that the class representatives' claims are typical of the full class.

Without such studies, employee-plaintiffs generally would be unable to satisfy Rule 23.7

Given the critical import of courts' certification rulings and the key role plaintiffs' expert evidence plays in determining those rulings, it is imperative that courts begin to apply consistent standards when reviewing plaintiffs' expert evidence at this stage of the litigation.

Inconclusive Guidance

The Dukes plaintiffs, who alleged gender-based pay and promotion discrimination, relied on two types of expert evidence to support their motion for certification of a nation-wide class of current and former female Wal-mart employees: (1) sociological evidence that Wal-mart had a corporate culture making it vulnerable to stereotyped thinking and gender-biased employment decisions; and (2) statistical evidence showing pay and promotion disparities between male and female employees.

As to the sociological evidence, the Court noted both the parties' disagreement as to whether it satisfied Daubert and the lower courts' conclusion that Daubert did not apply at the class certification stage. The Court stated, "we doubt that is so," but it did not rule on the issue, instead holding that "even if properly considered, [the evidence] does nothing to advance [plaintiffs'] case," since their expert's conclusions were "worlds away" from the "significant proof" required to show a common policy of discrimination for purposes of satisfying Rule 23.8

As to plaintiffs' statistical studies, the Court held—without referencing Daubert or articulating the standard of review that lower courts should apply: "Even if [the studies] are taken at face value, [they] are insufficient to establish that [plaintiffs' claims]…can be proven on a class-wide basis," since they show only regional and national-level disparities.9 The Court also held that even if the studies had shown store-level disparities, they would be insufficient standing alone to establish commonality.10

In short, the Dukes Court provided dicta suggesting that Daubert did apply at the class certification stage, and the Court rejected the statistical expert analysis even if "taken at face value," and rejected the sociological expert evidence "even if properly considered." Clearly, more conclusive guidance is required to ensure that lower courts adopt a consistent approach when reviewing expert evidence at the class certification stage.

Courts' Approaches

Courts have applied a wide variety of approaches when reviewing plaintiffs' expert evidence at the class certification stage.

The Uncertain Applicability of 'Daubert.' Whether courts should subject plaintiffs' expert proof to Daubert review before considering it for class certification purposes has been described as "uncertain."11

In one case decided after the Supreme Court's Dukes decision, the U.S. Court of Appeals for the Eighth Circuit took a middle ground on the question. The court upheld a district court's certification order despite its failure to conduct a full Daubert review, reasoning that the lower court had not erred by applying a "tailored" Daubert analysis that "scrutinized the reliability of the expert testimony in light of the criteria for class certification and the current state of the evidence," without making a ruling on the testimony's admissibility.12

In another post-Dukes circuit court decision, the U.S. Court of Appeals for the Eleventh Circuit adopted the Seventh Circuit's pre-Dukes conclusion that a full Daubert review is proper when an expert's opinions overlap with the court's Rule 23 determination.13 The Eleventh Circuit held that where the parties' experts present conflicting testimony at the class certification stage, courts must "declare a proverbial, yet tentative winner" by conducting a full Daubert review.14 The Seventh Circuit in 2010 had vacated and remanded a district court's certification order that had acknowledged reservations about the reliability of plaintiffs' expert report (in connection with the defendant's Daubert challenge), but nevertheless had granted class certification based on the report.15

The Seventh Circuit rejected the district court's approach, holding 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," by performing a full Daubert analysis "if the situation warrants," including reviewing both the expert's qualifications and "any challenge to the reliability of information provided by an expert if that information is relevant to establishing any of the Rule 23 requirements."16

By contrast, before Dukes many courts historically would refuse to apply Daubert at all at the certification stage. The Ninth Circuit's (now overturned) 2010 Dukes decision, for example, held that a district court properly refused to subject plaintiffs' expert testimony to Daubert admissibility standards when granting certification based in part on such testimony.17

Courts' Review on the Weight to Be Afforded to Expert Testimony. The inconsistency in courts' approaches regarding applicability of Daubert at the class certification stage parallels their inconsistent approaches to scrutinizing experts' substantive theories when evaluating whether they suffice to satisfy the Rule 23 requirements.

Some courts have accepted plaintiffs' expert proffers without rigorous scrutiny and refused to consider defendant-employers' rebuttal-expert evidence. The U.S. Court of Appeals for the Fourth Circuit in 2009, for example, vacated a district court's class certification denial, reasoning that its discrediting of plaintiffs' expert's proof as "statistically insignificant" constituted an improper foray into the merits, and when combined with plaintiffs' affidavit evidence, their statistical proof established commonality, even though it did not meet the two standard deviation thresholds.18

More recently, courts increasingly have subjected plaintiffs' expert proof to more rigorous review and also considered defendants' counter-evidence when evaluating the propriety of class certification. For instance, the Eighth Circuit in 2011 (after Dukes), affirmed denial of class certification in a disparate-impact case where plaintiffs' expert's analysis was based on assumptions that all defendant's departments followed common hiring and promotion policies, which defendant's counter-evidence belied.19 Other circuit courts of appeals have reached similar results.20

Conclusion

Given the continued inconsistency in courts' approaches to the review of plaintiffs' expert evidence at the class certification stage, as well as the impact that certification or denial of certification has on the overall resolution of class actions, concrete direction is needed to ensure a uniformity in approach going forward. The Supreme Court should rule, and/or Congress should amend Rule 23 to require, that courts must apply both a full Daubert review and a rigorous analysis of plaintiffs' expert evidence—to the same degree Dukes requires courts to review other merits issues that bear on class certification.

Such a requirement would strike an appropriate balance between protecting employee-plaintiffs' entitlement to class certification where they have reliable and well-founded expert evidence to support the commonality of their claims and the other Rule 23 elements, while still protecting the rights of defendant-employers who should not be forced into costly class litigation or settlements, where the plaintiffs' class is certified based on inadmissible, methodologically flawed, or otherwise unsound expert "proof."

Elise M. Bloom is a New York-based partner and co-chair of the labor and employment law department at Proskauer Rose. Amanda D. Haverstick is special employment law counsel at the firm, resident in the firm's Newark office. Nausheen Rokerya, an associate at the firm, assisted in the preparation of this article.

Endnotes:

1. Joe S. Cecil, Federal Judicial Center/ALI-ABA, Manual for Complex Litigation: Expert Scientific Evidence, Opinion and Expert Testimony in Federal and State Courts 469-71 (2008); Carol Krafka et al., "Judge and Attorney Experiences, Practices and Concerns Regarding Expert Testimony in Federal Civil Trials," 8 PSYCH. PUB. POL. AND L. 309, 318-19 (2002).

2. Daubert v. Merrell Dow Pharms. Inc., 509 U.S. 579 (1993), directs that trial courts act as "gatekeepers" to exclude expert evidence that is not both "reliable" and "relevant." Factors for determining the admissibility of an expert's theory/technique include whether it has been subjected to peer review and publication and generally is accepted in the relevant expert community; whether it can be/has been tested; its known or potential error rate; and the existence and maintenance of standards controlling its operation. FRE 702, which governs admissibility of expert testimony at trial, codifies Daubert's reliability and relevance requirements. Neither Daubert nor FRE 702, however, indicates whether these requirements apply at the class certification stage.

3. These questions are not necessarily coterminous or mutually exclusive. See, e.g., Ellis v. Costco Wholesale Corp., 2011 WL 4336668 (9th Cir. Sept. 16, 2011) (holding that district court erroneously conflated the two questions and impermissibly stopped short its analysis of plaintiffs' expert evidence after determining that it passed Daubert, without applying a "second tier" rigorous review as to whether the evidence was sufficient to establish Rule 23 commonality); Remien v. EMC Corp., 2008 WL 597439 (N.D. Ill. March 3, 3008) (holding that plaintiffs' statistical studies satisfied Daubert, but would not be considered because they were based on arbitrary factors).

4. 131 S. Ct. 2541 (2011).

5. Dukes, 131 S. Ct. at 2553.

6. See, e.g., Brief of Amicus Curiae, Labor Economists and Statisticians in Support of Respondents, Wal-mart Stores Inc. v. Dukes, 2011 WL 719643, at *10-11 (2011) (arguing that adoption of Wal-Mart's position on the required methodology for statistical analysis at class certification "would doom class actions in many, if not most, pattern-or-practice cases," for empirical reasons).

7. See id.

8. Id.

9. Id. (Emphasis added.)

10. Id.

11. Am. Honda Motor Co. Inc. v. Allen, 600 F.3d 813, 815 (7th Cir. 2010) (granting Rule 23(f) review of district court's certification order due to "the uncertainty surrounding the propriety of conducting a Daubert analysis at the class certification stage, and the frequency with which this issue arises").

12. In re Zurn Pex Plumbing Products Liability Litig., 644 F.3d 604, 614 (8th Cir. July 6, 2011).

13. Sher v. Raytheon Co., 2011 U.S. App. LEXIS 4902, at *8 (11th Cir. March 9, 2011) (adopting Seventh Circuit's approach).

14. Id.

15. Allen, 600 F.3d at 814.

16. Id. at 815-16.

17. 603 F.3d 571, 601-02, 608-09 (9th Cir. 2010) (citing other cases), reversed and remanded, 131 S. Ct. 2551 (2011). See also, e.g., Walsh v. Principal Life Ins. Co., 266 F.R.D. 232, 237-38 (S.D. Iowa 2010) (holding that Daubert did not apply at the class certification stage—rather, expert "testimony should be judged on the basis of whether it supports class certification—not whether it meets the standard of admission at trial," but court ultimately struck the expert report as improperly opining on legal issues).

18. Brown v. Nucor, 576 F.3d 149, 153-56 (4th Cir. 2009) ("[E]vidence need not be conclusive to be probative…and an in-depth assessment of the merits…at this stage would be improper."); In re Delta Air Lines, 301 F.3d 953 (6th Cir. 2002) (finding nothing improper with the district court's acceptance of plaintiffs' expert proffers and its refusal to weigh defendants' rebuttal-expert proof at the class certification stage); Velez v. Novartis Pharm. Corp., 244 F.R.D. 243, 258-60 (S.D.N.Y. 2007) (granting class certification based in part on expert proof and rejecting defendant-employer's attacks on such proof and its rebuttal-expert proof, reasoning that "'statistical dueling'" was improper at the certification stage).

19. Bennett v. Nucor Corp., 2011 WL 4389194 (8th Cir. Sept. 22, 2011).

20. See, e.g., Grant v. Metropolitan Gov't of Nashville, 2011 WL 3796329, at *4 (6th Cir. Aug. 26, 2011) (upholding certification denial where plaintiffs proffered evidence that black employees were disproportionately concentrated in lower paying positions with fewer advancement opportunities, since scrutiny of that evidence showed that plaintiffs' expert had compared the wrong groups of people and made unfounded assumptions); Gintis v. Bouchard Transp. Co., 596 F.3d 64 (1st Cir. 2010) (vacating denial of class certification because district court failed to assess the relative strength of the parties' contending expert proffers); Hydrogen Peroxide, 552 F.3d 305, 307 (3d Cir. 2008) ("[T]he court's obligation to consider all relevant evidence and arguments extends to expert testimony, whether offered by [plaintiffs or defendants]").



Copyright 2011. ALM Media Properties, LLC. All rights reserved.