Lessons learned from the Supreme Court’s Decision in Wal-Mart v. Dukes
June 22 , 2011
The June 20, 2011 decision by the United States Supreme Court in Wal-Mart Stores, Inc. v. Dukes, et al., 564 U. S. ____ (2011), has received significant public attention.The “crux” of the case – in which plaintiffs attempted to certify a nationwide class of 1.5 million women who were purportedly discriminated against in promotion and pay by giant retailer Wal-Mart – was the failure to satisfy Rule 23(a)(2)’s requirement of “commonality.” Of course, class actions were designed to apply where a large class of persons has suffered the same alleged harm.But alleged common harm to a large number of people turned out not to be enough.The Court, by a 5-4 majority,found that there was insufficient evidence of a company-wide policy of discrimination. In the end, it was the lack of commonality in the alleged circumstances of the purported discriminatory conduct that doomed class certification.
More broadly than its case-specific impact, the Court’s decision provides three important practical lessons for our employer clients.
Lesson One: Employers Are Protected by a Strong, Explicit Anti-Discrimination Policy Even When Employment Decisions Are Decentralized.
As the Court found, Wal-Mart explicitly forbade discrimination and also imposed penalties for denial of equal employment opportunity. Plaintiffs argued, nonetheless, that by giving local managers discretion over pay and promotions, the company permitted de facto bias against women. As support, plaintiff used evidence from a statistician and a labor economist who performed regression analyses and found a statistically significant disparity between promotion rates of women at Wal-Mart compared to those rates at Wal-Mart’s competitors. Plaintiffs buttressed this testimony with opinions of a sociologist specializing in “social framework analysis.” He testified that, despite the lack of explicit discriminatory policies at Wal-Mart, the company’s “strong corporate culture” made it “vulnerable” to “gender bias.”
The Supreme Court flatly rejected such opinions. The fact that Wal-Mart had explicit general policies against discrimination, and the fact that it left decisions about raises and promotions to local managers, was fatal to the claim of commonality in a nationwide class action. What was required, instead, was a specific employment practice that tied all 1.5 million claims together. As the Court emphasized, “[M]erely proving that the discretionary system has produced a racial or sexual disparity is not enough. . . . Merely showing that Wal-Mart’s policy of discretion has produced an overall sex-disparity does not suffice.” In short, the expert’s testimony – critical for establishing a class – failed to establish “’significant proof’ that Wal-Mart operated under a general policy of discrimination.’”
The lesson for employers is that properly drafted – and enforced – policies can be a powerful weapon against class discrimination claims. There are many good business reasons for decentralized decisions about hiring, promotion, compensation and the like. A company may safeguard its decentralized practices under the umbrella of strong central anti-discrimination policies, even when evaluations and decisions are local. Now would be a good time to review company policies to ensure they properly convey the company’s fealty to anti-discrimination laws and the extent to which decisions are central or local in nature. Of course, the ruling should not be seen as an “anything goes” invitation to decentralization. Rather, this is a time to ensure that compliance programs say – and do – the right things.
Lesson Two: Statistics Matter But Will Be Scrutinized with Care.
The Wal-Mart decision is not a complete rejection of statistics in employment cases but it does suggest some limits to their utility as well as potential future attacks in litigation. Notably, although the plaintiffs’ experts found a statistically significant disparity between the position of men and women at Wal-Mart, which the experts opined could only be explained by gender discrimination, the sheer size and geographic diversity of the cohort being sampled permitted the Court to reject the analysis.
First, the plaintiff’s expert studies were insufficient to establish that a culture of discrimination existed at Wal-Mart on a nationwide basis. The Court rejected the plaintiff sociologist’s reliance on the methodology of “social framework analysis” – a relatively new and sometimes controversial technique when applied in the context of employment cases. Although the Court did not explicitly invoke Daubert in this portion of its decision, the analysis is Daubert-like insofar as it concludes that the there is too great an inferential gap between the proffered studies and the claims of nationwide discrimination. Indeed, in what likely will be an oft-quoted line of dictum going forward, the Court noted that while the district court had concluded that Daubert did not apply in a class certification hearing, “We doubt that is so.” Daubert challenges to expert testimony at the certification stage almost certainly will rise as a result of this brief sentence.
Second, the relative paucity of anecdotal evidence, compared to the size of the class, was another factor defeating class certification. The Court distinguished the present case – where there were 120 affidavits reporting instances of discrimination, or about 1 for every 12,500 employees – with Teamsters v. United States, 431 U.S. 324 (1977), in which there was one account of anecdotal evidence for every eight members of the class. The sheer size of the class robbed the 120 affidavits of their force and permitted the Court to reject them as insufficient proof for certification.
There are several practical consequences of the Court’s ruling. It is likely that future class claims against large companies will be broken down by divisions or states, and the like, rather than being pursued as single, nation-wide class. Indeed, one of the top plaintiffs’ lawyers in the Wal-Mart case said that as a result of the ruling, there would be more class actions at the store or regional level, where it might not be as hard to show that local managers had engaged in sex or age discrimination.
Decentralization, or a focus on subclasses as a principal class, is both a good thing and a potentially bad thing for employers. While it removes the risk of a potentially enormous single verdict – and the attendant settlement pressures that any large company would feel – it also means that companies will face more (and potentially more factually powerful) cases with all of the costs of piecemeal litigation. The same sorts of methodological questions about statistical proof and claims of discriminatory practices should be applied in these smaller class cases. But the analysis applied by the Wal-Mart court may have less force in cases with a more refined class, where the power of the evidence will be less diluted and potentially less subject to attack.
Lesson Three: Courts Will Not Award Backpay in Cases Seeking Injunctive Relief.
The second holding in Wal-Mart was unanimous and is not open to significant uncertainty going forward. Put simply, the Court found that claims for monetary relief – specifically back pay claims – cannot be certified under FRCP 23(b)(2). The standards for Rule 23(b)(2) class certification are lower than those for back-pay claims. However, the Second and Ninth Circuits have in the past permitted plaintiffs to bootstrap back-pay claims on 23(b)(2) injunctive relief claims. The Court found that such monetary relief was not “merely incidental” to the injunctive relief sought, but instead would require individualized inquiries into each employee’s eligibility for backpay. The impact of this part of the decision will surely be to chill a variety of class actions.
Related Attorney: Sarah R. Marmor
