On June 20, 2011, the United States Supreme Court granted employers some long-awaited relief by substantially raising the bar for plaintiffs (and their lawyers) seeking to certify large employment discrimination class actions. In Wal-Mart v. Dukes (No. 10-277), the Court reversed the Ninth Circuit Court of Appeals’ en banc decision upholding the certification of a class action filed on behalf of approximately 1.5 million hourly and salaried female employees alleging sex discrimination in pay and promotions. The potential damages were estimated to be more than a billion dollars.

As we have detailed in prior newsletters and bulletins, because of potentially large damage awards and fee-shifting provisions, employment class actions have been a boon for the Plaintiff’s bar while exposing employers to significant liability and litigation costs.  Although the Dukes decision will not put an end to class actions, it, at the very least, temporarily halts the large nationwide employment discrimination class actions. In its ruling, the Supreme Court significantly increased the plaintiffs’ burden of proof at the class certification phase and mandates that district courts look more carefully at whether class certification is appropriate, including a critical assessment of plaintiffs’ proof of class-wide discrimination. 

The Supreme Court’s Decision in Dukes

Following an increasing trend, the Dukes plaintiffs alleged that Wal-Mart discriminated against its female employees by delegating subjective decision making authority with respect to pay and promotion decisions to its local store managers and by building a corporate culture that fostered sex bias in these managerial decisions. Both the district court and the Ninth Circuit held that plaintiffs demonstrated that their class claims were appropriate for certification by relying on:

(i) statistical evidence purportedly demonstrating disparities in the pay and promotions of males and females; (ii) anecdotal reports of discrimination by 120 female employees; and (iii) the “expert” testimony of a sociologist who concluded that Wal-Mart’s culture was susceptible to gender discrimination.

Plaintiffs Did Not Satisfy Their “Commonality” Burden under Rule 23(a)

In a strongly worded opinion, Justice Scalia, writing for the 5–4 majority, disagreed that the Dukes plaintiffs’ evidence was sufficient to support class certification because it did not meet plaintiffs’ burden of satisfying Rule 23 of the Federal Rules of Civil Procedure requirements for certification. As recast by Justice Scalia, to meet these requirements, plaintiffs must provide “significant proof” that their class claims involve a common issue the resolution of which is “central to the validity of each one of the [class members’] claims in one stroke”; for example, discriminatory bias on the part of the same manager or the use of a discriminatory test.  

The majority’s decision removes any doubt that a trial court must conduct a “rigorous analysis” to ensure that plaintiffs have satisfied the Rule 23 elements, including a searching review of evidence that goes to the merits of the case. Exploration of the merits was appropriate in Dukes, the majority found, because it necessarily overlapped with the plaintiffs’ class-wide allegations that Wal-Mart engaged in a pattern and practice of discrimination.

In Dukes, the majority found plaintiffs’ evidence fell far short of the required “significant proof.” The Court did not reverse a prior decision that delegation of subjective decision making to individual managers could constitute a common discriminatory practice, but the majority found plaintiffs’ evidence lacking where Wal-Mart had a “general policy of non-discrimination” and where thousands of managers were making literally millions of pay and promotion decisions in some 3,400 stores. Specifically, the Court rejected plaintiffs’ sociological expert’s conclusion that Wal-Mart’s corporate culture made it more susceptible to gender bias in managerial decision making because the expert could not even opine, let alone show, that gender bias infected .5 percent or 95 percent of managerial decisions. The Court concluded that this was “the opposite of uniform policy that could provide commonality needed for a class action.”

The majority also found plaintiffs’ statistical and anecdotal evidence to be equally unpersuasive. Plaintiffs’ statistical expert conducted a region-by-region analysis and found that female representation in management positions was substantially less than in lower hourly positions and that females earned less than men. The Court discounted this proof, stating that any disparity at the regional level could not by itself establish that there were pay or promotion disparities at the individual stores, and even less so across all class members, which the majority stated was necessary to support plaintiffs’ theory of commonality. Furthermore, even if the statistics supported disparity at all the individual stores, the analysis did not consider potential assertions by Wal-Mart’s managers that women are not as readily available in certain store areas or the differences in the criteria used by the individual stores to make the decisions. The Court further found that the plaintiffs’ anecdotal evidence comprised of 120 affidavits representing the reporting experiences of only 1 out of every 12,500 class members and only 235 of Wal-Mart’s 3,400 stores could not show the whole company operated under a general policy of discrimination.

Plaintiffs Could Not Pursue Individualized Monetary Claims under Rule 23(b)(2)

The Court also unanimously resolved a split in the Courts of Appeal and held that the claims for backpay should not have been certified as a class action under Rule 23(b)(2) because such backpay damages were not “incidental” to the injunctive or declaratory relief sought. The Court concluded that certification under Rule 23(b)(2) is inappropriate when “each member would be entitled to an individualized award of monetary damages.” 

Instead, the Court held that the monetary claims involving individualized proof must proceed under Rule 23(b)(3), which permits class certification only upon a showing that common questions of law and fact predominate over questions affecting individuals and after providing notice of the class action to potential class members and an opportunity to opt out. The Court reasoned that these procedural safeguards were necessary to protect class members’ individual interests in monetary relief. 

The Court also rejected the position adopted by the Ninth Circuit that a statistical sample of class members could be used to determine the damages for the whole class without individualized proceedings. The Court reasoned that this sampling method was inconsistent with the procedures established by the Supreme Court for determining the scope or lack of individual damages in Title VII claims. The majority further suggested, without deciding, that this approach might also violate an employer’s right to individualized determinations of each class member’s eligibility for backpay.

Implications of the Court’s Decision in Dukes

The most immediate effect of the Dukes decision is that district courts will need to reconsider the appropriateness of employment discrimination class actions on their docket that were certified under Rule 23(b)(2). In the longer run, Dukes may not have sounded the death knell for all large discrimination class actions but it has made it very difficult for plaintiffs to mount class actions that seek to cover multiple types of claims, e.g., pay and promotions, and many different job classes, facilities and/or managers. As a consequence, future class actions are more likely to focus on more discrete claims of discrimination covering fewer locales and limited to common decision makers and covering a more homogenous class. In particular, Dukes is likely to curtail the bringing of class actions under the “delegation of subjective decision making” theory. Although the Court did not articulate clear evidentiary standards for establishing “commonality,” the Court emphasized the need to demonstrate a common allegedly operative discriminatory practice and injury across all putative class members. It is difficult to see how plaintiffs will mount class actions based on “subjective decision making” given the Court’s emphasis that “demonstrating the invalidity of one manager’s use of discretion will do nothing to demonstrate the invalidity of another’s.”

The Dukes decision also, as a practical matter, will require district courts to probe more deeply into the merits at the class certification stage, and the Supreme Court endorsed the consideration of Daubert motions to exclude expert testimony before class certification to assess such testimony’s adequacy. Moreover, although Dukes is restricted to class certification requirements, its emphasis on proving that the alleged discriminatory practice applied to and may have injured all class members may also lead to higher standards of proof in establishing class-wide discrimination on the merits.

Dukes also will lessen the incentive of plaintiffs’ attorneys to bring class actions by making it more difficult to seek monetary damages for large, diffuse classes.

How plaintiffs’ attorneys will respond is open to speculation. The attorneys representing Dukes profess their intent to bring individual and more discreet, localized class actions. This may become an overall trend. Employers should keep in mind that the Dukes decision has no immediate impact on the ability of the EEOC to bring company-wide pattern and practice suits because the EEOC generally is not required to satisfy the “commonality” principles espoused by the Supreme Court. Nevertheless, the Dukes decision, which comes on the heels of the Court’s May 2011 pro-employer decision in AT&T Mobility v. Concepcion et ux. seemingly validating the use of mandatory arbitration agreements to bar employees’ ability to litigate claims on a class basis, is a welcome change for employers.