On September 16, 2011, the Ninth Circuit handed down one of the first decisions to interpret and apply the game-changing decision in Wal-Mart Stores, Inc. v. Dukes, et al., 564 U.S. ___ (2011) (“Dukes”). Although Ellis v. Costco Wholesale Corp. (“Ellis”) is not as ground-breaking as the Dukes decision, it does provide some insight as to how the courts will apply the arguably employer friendly precedent established by Dukes.

In facts similar to Dukes, the case alleged a nationwide pattern and practice gender discrimination by Costco whereby Costco allegedly discriminated against women by refusing to promote them to the highest level management positions. Costco, which operates over 350 warehouse-style retail establishments nationwide, employs a management structure consisting of a General Manager (“GM”), two to three Assistant General Managers (“AGM”) and three to four Senior Staff Managers. GMs are responsible for the entire operations of their respective stores with AGMs as their second in command. Costco promotes almost exclusively from within its organization, with only current AGMs eligible for promotion to GM. Costco has no written policies or procedures for the selection and promotion process and does not retain records for the same.

The named plaintiffs were three current and former female Costco employee, two of which worked as AGMs. The named plaintiffs filed gender discrimination charges with the Equal Employment Opportunity Commission (“EEOC”) in June 2004 and March 2005. The EEOC dismissed the charges and the women sued Costco under Title VII of the Civil Rights Act of 1964, alleging that Costco systematically denied women promotions to AGM and/or GM positions and seeking class certification.

Plaintiffs submitted the opinions of three experts to establish commonality under Rule 23(a): (1) Dr. Drogin, a statistician who concluded that female employees were promoted more slower rate than their male counterparts and that women were underrepresented at the higher management levels at issue; (2) Dr. Bendick, a labor economist who opined that females were underrepresented in the subject management jobs as compared to their peers at comparable companies; and (3) Dr. Reskin, a sociologist who diagnosed a pervasive culture of paternalism and gender stereotyping at Costco. In response, Costco submitted 200 declarations and the opinion of its own experts, including Dr. Saad, a statistician and labor economist who concluded that women were not underrepresented at Costco and that any gender disparity, to the extent that they exist, were limited to two geographical regions.

On January 11, 2011, the district court filed a 37-page order granting plaintiffs’ motion to certify the class. Judge Patel found that plaintiffs had established commonality through the experts, but in doing so she declined to resolve the dispute between the experts as to whether the statistics really showed a pattern nationwide, finding that such an inquiry would be an impermissible consideration of the merits. She found the named plaintiffs to be adequate class representatives and certified the class under Rule 23(b)(2), which Judge Patel found requires only that the plaintiffs’ primary motivation be to obtain injunctive relief, even if they also seek back pay and punitive damages. Judge Patel did not address certification under the stricter Rule 23(b)(3) standard which is normally reserved for cases seeking damages. Costco appealed.

The Ninth Circuit issued several rulings, but remanded many of the issues to the district court, raising more questions than answers. One of the initial decisions involved the named plaintiffs’ standing. The Ninth Circuit held that because one of the three plaintiffs remained a current employee who was denied a promotion to the positions at issue, that she had Article III standing to pursue the action as a class representative. On the other hand, the other two named plaintiffs, as former employees, lacked standing to seek injunctive relief as necessary to support certification under Rule 23(b)(2). The court further held that the fact that the one employee with standing was eventually promoted to AGM did not render her an inadequate class representative because her claim was that the company culture was discriminatory and such culture prevented her from being promoted to GM. According to the Ninth Circuit, this was sufficient to establish adequacy.

On the commonality issue that resulted in a reversal in Dukes, the court noted that Judge Patel had failed to undertake a rigorous enough analysis of the expert opinions to adequately decide commonality. In applying the case law established by Dukes, the Ninth Circuit stated that “it is insufficient to merely allege any common question . . . [i]nstead, [the plaintiffs] must pose a question that ‘will produce a common answer to the crucial question why was I disfavored.’” The court interpreted this mandate in the case at hand as requiring that “Plaintiffs must have a common question that will connect many individual promotional decisions to their claim for class relief.”

Judge Patel conducted a general admissibility analysis for all the experts and found them to be all admissible, but she failed to conduct an in depth analysis of the expert opinions to determine whether such opinions could provide the “common answer” to the liability questions at issue. The court stated that instead of judging the persuasiveness of the evidence presented, the district court seemed to, erroneously, end its analysis of the plaintiffs’ evidence after determining that such evidence was admissible. The Ninth Circuit held that it was error to do so and emphasized Dukes requirement that “a district court must consider the merits if they overlap with the Rule 23(a) requirements.”

The Ninth Circuit went on to caution, however, that this rule applied only to the merits as necessary to decide commonality (e.g., whether gender disparities are confined to only two of Costco’s eight regions) and not a determination on whether such disparity, in fact, arose from discrimination. As the court noted, “[i]f no such nationwide discrimination exists, Plaintiffs would face an exceedingly difficult challenge in proving that there are questions of fact and law common to the nationwide class.” (emphasis in original). Unlike Dukes, where the record clearly established that the plaintiffs could not establish commonality because their key expert admitted that he could not say whether the culture of stereotyping impacted 0.5% or 95% of promotion decisions, there was no similar severe statistical deviation here. Accordingly, the Ninth Circuit held that “the district court applied an impermissible legal criteria . . . and failed to resolve the critical factual disputes centering around the national versus regional nature of the alleged discrimination” and remanded to the district court for what essentially amounts to a “redo” on the analysis.

The Ninth Circuit held that Judge Patel had also used the wrong standard in deciding whether certification was proper under Rule 23(b)(2) because she used the Ninth Circuit standard rejected under Dukes. Specifically, the court noted that the long-standing general rule that “in Rule 23(b)(2) cases, monetary damage requests are generally allowable only if they are merely incidental to the litigation . . . has been called into doubt by the Supreme Court.” The court remanded, stating that Judge Patel must now “apply the legal standard established in [Dukes].” Judge Patel must now determine whether there was any way, consistent with Dukes, to certify a class for punitive damages assuming the other elements of Rule 23(a) are established. The court expressly recognized that Dukes foreclosed certification of back pay under Rule 23(b)(2).

Finally, the Ninth Circuit made no determination as to the manageability of the proposed class due to the district court’s failure to approve any particular trial plan, saying that it would be premature to do so at this time. It would seem, based upon this holding then, that as part and parcel of a class certification decision, the district court should require a trial plan to assess whether manageability is feasible.

Based upon the foregoing, it would seem that there are many useful insights into how the courts are likely to apply Dukes. Specifically, litigants, and particularly plaintiffs, can expect that, in addition to the Daubert analysis undertaken when experts are involved, that the district court must also consider the merits of the experts’ opinion(s) in order to determine of the experts can properly answer the “common” question that is the basis of the Rule 23(a) inquiry. This application of Dukes would seem to set up a fairly difficult hurdle for putative class-action plaintiffs to overcome and which would require, at the least, a detailed and thorough expert opinion and report from the plaintiffs at the early class certification stage. For example, if an expert can establish total damages to the class, but fails in his methodology to establish class-wide liability (e.g., he cannot say whether there are subsections of the class that did not suffer injury or even identify who could fall into these subsections), such failure could arguably provide a basis to defeat class certification even if the expert's methodology survives Daubert gate-keeping.