On June 20, in a headline-making decision, the United States Supreme Court reversed class certification in Dukes v. Wal-Mart, which had threatened to become the largest-ever employment class action against a private employer. Dukes proposed to certify a nationwide class of approximately 1.5 million current and former Wal-Mart employees, a breathtaking gambit by plaintiffs’ counsel to make new law under Rule 23 of the Federal Rules of Civil Procedure. If the plaintiffs’ lawyers had succeeded, employers would face a real liability threat from class actions of any size or shape, regardless of evidence of the employer’s centralization or legal violations. With the Dukes decision, however, the Supreme Court brought Rule 23’s interpretation back down to earth and provided instructive guidance for employers seeking to avoid and successfully defend against class actions.
The Issues and the Dukes Decision
Plaintiffs alleged that Wal-Mart discriminated against female employees by denying them equal pay or promotions in violation of Title VII. Although Wal-Mart’s written policies expressly condemned and forbade sex discrimination, Plaintiffs claimed that Wal-Mart’s “corporate culture,” coupled with its policy of giving local managers discretion to make subjective employment decisions, disproportionately favored men.
A 5-4 majority of the Court held that Plaintiffs did not satisfy the Rule 23’s basic “commonality” requirement, set forth in Rule 23(a)(2), because, even if statistical evidence revealed that female employees were paid less, and were promoted less often, than men, Plaintiffs failed to establish that this disparity resulted from a discriminatory company-wide policy, practice or procedure. In reaching this decision, the Court set forth important boundaries for Rule 23’s burden of proof. The Court clarified that movants for class certification must “affirmatively demonstrate” compliance with Rule 23 by offering “significant” and “convincing proof” to support their contentions. This proof is subject to “rigorous analysis” by the trial court, which notably must consider the merits of plaintiffs’ claims to the extent they overlap with certification issues.
The Court also held that merely reciting general questions of law or fact, such as whether the employer violated the substantive law at issue (Title VII in Dukes), would not support class certification. Instead, Plaintiffs are required to demonstrate that class members “suffered the same injury.” It is not enough under Rule 23 to show that they all suffered a violation of the same provision of law. Plaintiffs’ claims must also depend on a “common contention,” or reason for the violation, such that “examination of all the class members’ claims for relief will produce a common answer to the crucial question why was I disfavored.”
The majority held that Plaintiffs failed to provide a “common answer” because they sought to litigate more than one million employment decisions without “some glue holding the alleged reasons for all of those decisions together.” In order to establish commonality in this case, Plaintiffs had to identify the specific employment practice that ties all their 1.5 million claims together. This could be accomplished with “significant proof” that Wal-Mart used biased testing or evaluation procedures, or “operated under a general policy of discrimination.” There was no evidence of the former and the only company-wide practice established by Plaintiffs’ evidence was Wal-Mart’s “policy” of allowing discretion by local supervisors over employment matters. The Court held this was “just the opposite of a uniform employment practice that would provide the commonality needed for a class action; it is a policy against having uniform employment practices.”
The Court rejected Plaintiffs’ remaining evidence as falling short of “significant proof” of a policy of discrimination. First, it rejected as irrelevant the testimony of Plaintiffs’ sociological expert, who opined that Wal-Mart’s strong corporate culture made it vulnerable to gender bias, because the expert could not say the degree to which this “culture” may have influenced employment decisions. The Court also rejected Plaintiffs’ efforts to infer company-wide discrimination from regional statistical data, finding that such data could not establish disparities at different stores, much less uniform, store-by-store disparities across the nation. Finally, the Court rejected Plaintiffs’ anecdotal evidence – comprised of 120 affidavits, representing one account for every 12,500 employees, and relating to just 235 of Wal-Mart’s 3,400 stores – as insufficient as significant proof of a company-wide practice of discrimination.
The Court next ruled, in a unanimous portion of the opinion, that plaintiffs’ claims for back pay could not be certified under Rule 23(b)(2), a subsection of Rule 23 that allows certification of a class without requiring movants to show that either common issues predominate or that a class action is superior to other methods of certification. Though characterized as “equitable relief” under the particular statute in question, Title VII, the Court held that back pay is an individualized claim for monetary relief that would not be incidental to Plaintiffs’ claims for injunctive or declaratory relief. Such claims must be certified, if at all, under Rule 23(b)(3), which protects the rights of class members by ensuring they receive notice and an opportunity to opt out and that imposes the higher standards mentioned above. In addition, the Court held that employers have a due process right to assert affirmative defenses to plaintiffs’ legal theory of discrimination, and to obtain individualized determinations regarding each employee’s eligibility for back pay, which would not be available in a Rule 23(b)(2) action.
Analysis and Guidance
Dukes has wide-ranging implications, but this discussion will concentrate on the major points of the opinion and related “takeaway guidance” for employers facing current or threatened employment law class actions.
The level of decision-making authority of local management is now itself an important decision. The Court came to a common-sense conclusion that local managerial discretion cannot be a common, uniform “policy or practice” supporting class certification. This conclusion is not, however, a “silver bullet” defense to Title VII and other motive-based employment law class actions, because the Court identified alternate ways for plaintiffs to surmount their burden of proof under Rule 23. Notably, the Court also focused on Wal-Mart’s express anti-discrimination policy in highlighting the weakness of Plaintiffs’ theory of a common practice of discrimination that would be 180 degrees opposite of what Wal-Mart actually had written down. Thus, employers would be well advised to continue such centralized antidiscrimination and anti-harassment policies, actively enforce them, and maintain the evidence that they actually have actively enforced them. Second, how far employers want to go in enabling local managerial discretion in personnel decisions is not a simple question: most employers will want to strike a balance to ensure that there are controls preventing the odd individual supervisor from taking ill-advised or potentially discriminatory actions. Dukes should not become an inadvertent catalyst for an employer’s future spike in individual discrimination cases. And finally, where employers want to enable discretion, they should emphasize it in documents (other than just job descriptions) that can serve as an economical means of defense here. Besides discrimination class actions, the existence of discretion (and thus variance in management style) will be important in nearly any kind of employment class action, especially wage-hour class and collective actions. For example, the existence of supervisory discretion is directly at issue in executive-exemption misclassification-based class and collective actions.
Rule 23(a) now has real teeth. Rule 23(a) is now a realistic possibility for fact-based defenses to stop a class action in its tracks before a trial court even reaches the more discretionary “balancing act” type of tests posed by Rule 23(b). In other words, employers now have a better chance at a dispositive threshold defense to class certification by showing lack of “significant proof” of a single common issue. Although Plaintiffs by law must fulfill every Rule 23(a) requirement (numerosity, commonality, typicality, adequacy), courts typically in the past have disqualified few class actions under Rule 23(a) requirements, and often relied simply on the pleadings or the existence of the kind of generalized “common questions” that Dukes now disposes of as irrelevant to a court’s analysis. Dukes shows, at least with respect to Rule 23(a) commonality, that a meaningful quantum of evidence is needed; in the Court’s words, “Rule 23 does not set forth a mere pleading standard.” The fact that the broad ranging Dukes class action could not pass this initial test to pose a single common question raises the bar for plaintiffs in future class actions, and highlights that successful defense will focus just as much on deep understanding and argument about the facts for the Rule 23(a) test as it does for a Rule 23(b)(3) “predominance” test. This point from Dukes may also be reflected in collective actions under the Fair Labor Standards Act and state law class actions in those jurisdictions where courts look to federal Rule 23 cases for guidance.
No excuse exists any more for trial courts to shy away from considering merits issues in determining class certification. One aspect of class litigation that can sometimes hamstring defense arguments is where a trial court might avoid deciding class issues that are intertwined with merits determinations, focusing on language from past Supreme Court cases that the class and merits inquiries are separate. The clear language in Dukes indicates that this should no longer be a frustration in making defense arguments.
Expert evidence will be subject to successful attack if it is not (1) reasonably predictive and (2) reasonably granular to the workplace level and tied into a concrete practice. Much more should and will be written on this in the future, but Dukes gives employers two major guideposts for defenses based on attacking expert testimony. First, the Court discarded entirely the testimony of plaintiffs’ sociological expert based on a single deposition admission that his analysis could not calculate whether “0.5 percent or 95 percent” of Wal-Mart’s decisions suffered from “stereotyped thinking.” In other words, the expert’s admission of a gigantic margin of error in his methodology as applied to the employer’s state of mind made his testimony worthless. Defense lawyers will take note of this important holding. Second, the Court emphasized, in rejecting the regional studies of Wal-Mart carried out by plaintiffs’ statistical expert, that such disparities are meaningful for a nationwide class action only if they are shown to carry down to a large number of individual stores in the purported class. The Court opined that this would eliminate the possibility of concentrations of disparities among a few stores that then inaccurately portray broader-based disparity simply because they manifest themselves in a regional average or aggregate. Even more strikingly, the Court’s opinion can be read to require proof of a specific employment practice (e.g., a biased testing procedure) even before considering statistical evidence of disparity as “significant proof” to sustain class certification. In other words, the Court leaves the door open for employers to argue that statistics are totally meaningless without a specific employment practice identified by the plaintiffs and then tied into the disparity with “convincing evidence.”
Plaintiffs cannot rely on proportionally minute numbers of affidavits compared to the size of the class to carry the day on class certification. As noted, the Court rejected Plaintiffs’ anecdotal evidence as being simply too infinitesimally small to show certification, standing alone. Quantitative arguments over declarations are nothing new in class certification litigation, and Dukes will provide defenses against certain more extreme claims by plaintiffs based on very low numbers of affidavits. Notably, Dukes commented on anecdotal evidence being sufficient in the context of a prior case where there was one affidavit for every eight members of the purported class.
Where plaintiffs seek individualized backpay or monetary relief, it is usually going to force them to meet higher Rule 23(b)(3) class certification standards instead of the more lenient Rule 23(b)(2) standards. Although there will be great debate over whether the Court’s ruling here is specific to simply Title VII claims, due to the particular nature of Title VII, it is clear that the Court put its thumb on the scale of requiring Rule 23(b)(3) certification for any statute that involves individual monetary claims. This holding is extremely important, because in the past (as in the lower courts’ rulings in Dukes), employment plaintiffs have sought to avoid the higher class certification bar of Rule 23(b)(3), by pleading or presenting “proof” that the monetary relief sought is secondary to an injunction or other equitable relief. The Court’s holding here is also important, because it leaves room for the broader argument based on Dukes that the presence of individualized affirmative defenses, just like individual monetary claims, is a fatal impediment to the easier (for plaintiffs) class certification route of Rule 23(b)(2).
The employment class action is not going away. Contrary to some reports, although nationwide class actions are going to face major hurdles after Dukes, we can “look forward” to continued employment law class actions in the future, with litigation requiring a focus on more granular facts, and fought out on a more detail-specific level. State law class actions will become more prevalent, with more Class Action Fairness Act removal litigation in their wake, and there is still plenty of room for tightly proven class actions covering a smaller geography. The employment plaintiff’s bar of the United States possesses great talent, and will adapt to this ruling by bringing more cohesive and better sharpened sets of claims in the future, so Dukes is not a signal for employers to become less vigilant.