On June 20, 2011, the United States Supreme Court handed down the long-awaited decision in Wal-Mart v. Dukes, No. 10-277, 564 U.S. ___, 2011 US LEXIS 4567 (2011) (Scalia, J.), which reversed class certification in what is believed to be the largest class action certified in the United States. The decision is a tremendous victory for businesses defending against massive class actions, and employers in particular. After a series of employee-friendly decisions, the United States Supreme Court has finally rendered a decision in favor of large employers with widespread operations.
According to the class representatives, although the 1.5 million women in the class had different supervisors in different stores at different times, held different positions, and were subject to different decisions, they had one common complaint. Specifically, they contended that a company-wide "policy" of allowing local managers to use discretion in pay and promotion decisions resulted in across-the-board discriminatory decisions.
The Supreme Court's decision to reverse the class certification was unanimous, but the justices divided as to why. The majority and dissent vigorously disputed that the plaintiffs' proof of decentralized, subjective decision-making—allegedly coupled with a centralized yet pervasive culture and policies—was sufficient to support the aggregation of millions of claims by women from 3,400 Wal-Mart facilities nationwide.
Background of the Decision
The Dukes case was brought by three Wal-Mart employees seeking to represent a ten-year class of more than 1.5 million women, both salaried and hourly current and former employees, at over 3,400 Wal-Mart stores nationwide seeking billions of dollars of back pay and other relief based on alleged gender discrimination in promotion and pay.
In granting class certification, the district court based its commonality and cohesiveness analysis on: (1) a claimed company policy of having decentralized and highly discretionary pay and promotion decisions that provided a potential conduit for discrimination; (2) an expert's opinion that Wal-Mart supported a culture of personnel practices that made the company "vulnerable" to gender discrimination; (3) statistical evidence of class-wide gender disparities in pay and promotion; and (4) anecdotal evidence from a few of the class members about discriminatory attitudes held or tolerated by management.
The trial court certified the class under Federal Rule of Civil Procedure 23(b)(2), which is reserved for classes seeking an indivisible, collective remedy, such as an injunction or declaratory relief. A deeply divided Ninth Circuit affirmed virtually all of the district court's decision. Judge Ikuta's dissent concluded with the oft-cited observation that "[n]ever before has such a low bar been set for certifying such a gargantuan class." Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 652 (9th Cir. 2010) (Ikuta, J., dissenting).
In granting Wal-Mart's petition for certiorari, the Court chose to limit its consideration to "Question 1 of Wal-Mart's petition with a specific direction to brief: "Whether the class certification ordered under Rule 23(b)(2) was consistent with Rule 23(a)." Notwithstanding that initial inclination, the majority opinion addresses many of the points made by Wal-Mart in its certiorari papers.
What Is Common Proof of Discrimination?
The employees pursued both disparate impact and disparate treatment claims—disparate impact because the practice of granting discretion to local managers allegedly caused a disproportionately adverse effect on females, and disparate treatment because Wal-Mart allegedly knew about this adverse impact yet failed to correct it.
The Court held that the proposed class failed for lack of common questions under Rule 23(a). Specifically, the Court required proof of a specific employment practice that tied the class members' claims and common experiences together, which it found lacking in the case presented: "Merely showing that Wal-Mart's policy of discretion has produced an overall sex-based disparity does not suffice." Dukes, 2011 U.S. LEXIS 4567, at *34. For purposes of certifying this nationwide class, the "crux of this case [was] commonality." Id. at *18. Federal Rule of Civil Procedure 23(a)(2) requires the trial court to find common questions of law or fact as a threshold matter in order to certify a class. According to the Court, the commonality question asks not whether all of the 1.5 million female employees had managers with discretion over their pay or promotions, nor whether they believed such discretion led to unlawful acts under the same provision of Title VII. The Court found the key to class action treatment was not whether common questions could be posed, but rather whether a class proceeding could generate common answers that would resolve the litigation.
An example of a truly common question is when employees assert discriminatory bias on the part of the same supervisor, which makes sense when considering that disparate treatment authorities typically require an employee claiming discrimination to show someone outside of his/her protected class received preferential treatment in nearly identical circumstances. Employers have long defended such claims by demonstrating that the employee's proffered comparator was invalid or irrelevant because it involved a different supervisor or different circumstances. In Dukes, the Supreme Court reaffirms the principle that an employee's claim of intentional discrimination is a fact intensive inquiry into what actually happened for that particular employee—it is not an issue that can be decided by anecdotes from other employees in other unrelated circumstances.
The four-justice dissent charges the majority with "disqualif[ying] the class at the starting gate," by elevating the standard for common questions to an unrealistic and unprecedented level. Id. at *52 (Ginsburg, J., dissenting). According to the dissent, the majority's framing of the common question precluded the Court from giving sufficient deference to the trial court's findings of fact. Justice Ginsburg surveyed the evidentiary landscape, concluding that the questions identified were in fact common under Rule 23(a)(2).
Dukes does not hold that class actions as a whole are inappropriate for intentional discrimination claims. Rather, to establish sufficient commonality for a discrimination class action, employees may show that an employer used a biased testing procedure, or they may provide "significant proof that an employer operated under a general policy of discrimination[.]" Id. at 26. The Court rejected the argument that allowing local managers discretion in pay and promotion decisions was a "common" policy of discrimination. Instead, it reasoned that granting managers discretion was:
on its face . . . 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. It is also a very common and presumptively reasonable way of doing business -- one that [the Court has] said 'should itself raise no inference of discriminatory conduct . . . .
Id. at *29. Justice Scalia provided examples of how different managers could use discretion differently: by using sex-neutral, performance-based criteria; by using test scores or education criteria; or by using discriminatory motives. "In such a company, demonstrating the invalidity of one manager's use of discretion will do nothing to demonstrate the invalidity of another's. A party seeking to certify a nationwide class will be unable to show that all the employees' Title VII claims will in fact depend on the answers to common questions." Id. at *30-31.
The Court found no single common question existed, and the employees had no convincing proof of a company-wide discriminatory pay and promotion policy. Rather, the "class" involved a multitude of different employees with different jobs at different levels for different periods of times in different stores in different states with different supervisors subject to different polices. Quoting from Judge Kozinski's dissent in the Ninth Circuit, the Court agreed that: "[s]ome thrived while others did poorly. They have little in common but their sex and this lawsuit." Id. at *38.
Impact: While this decision is a major victory for employers, it also underscores the importance of having company policies forbidding discrimination. Indeed, the Court noted in the decision that Wal-Mart forbade discrimination and penalized employees who violated equal employment opportunity laws. Employers should make sure they have anti-discrimination policies in place, they should train employees on these policies, they should enforce these policies, and of course, they should investigate the allegations of any complaints they receive from employees. As Dukes teaches us, each employee's circumstances are unique, and when an employee raises a concern or complaint, employers should investigate and take any remedial action that may be appropriate for those particular circumstances.
Statistics, Extrapolation, and Social Science
The role of statistics was an important focus of the debate between the majority and dissent. The plaintiffs' expert, using regression analysis, considered data at a regional level and found statistically significant disparities between men and women in terms of compensation and promotion, concluding that the only explanation was gender discrimination. The Court found this data to be insufficient to support certification because it was too generalized to "establish the existence of disparities at individual stores, let alone raise the inference" of a company-wide policy "implemented by discretionary decisions at the store and district level." Dukes, 2011 LEXIS 4567, at *32. Further, the statistical proof did not account for local differences in the numerous stores and employment markets throughout the country.
The anecdotal proof was similarly rejected. The Dukes plaintiffs had submitted a number of declarations of current and former Wal-Mart employees about their personal experiences and observations of discriminatory attitudes held or tolerated by management. The Court found that the 120 declarations were not representative of a class of 1.5 million. It derided the lower courts' reliance on this evidence, which it considered "too weak to raise any inference that all individual discretionary personnel decisions are discriminatory," id. at *34, noting that the 120 anecdotes themselves showed disparities among the class members and were not remotely representative of the expansive class.
The plaintiffs attempted to weave together the anecdotal, statistical, and decentralized decision-making proof into a question common to all class members through the testimony of a social scientist, who opined that Wal-Mart supported a "strong corporate culture" of personnel practices that made the company "vulnerable" to gender discrimination. Id. at *27. However, he was not able to support this opinion with any "specificity how regularly stereotypes play a meaningful role in employment decisions at Wal-Mart." Id. The Court noted that, the expert "conceded that he could not calculate whether 0.5 percent or 95 percent of the employment decisions at Wal-Mart might be determined by stereotyped thinking." Id. Because he could not answer this "essential question on which [the plaintiffs'] theory of commonality depends," the Court felt it could "safely disregard what he has to say." Id. at *28.
The Supreme Court also struck down the Ninth Circuit's suggestion that extrapolation techniques could be used to decide classwide issues. The circuit court had relied upon Hilao v. Estate of Marcos, 103 F.3d 767 (9th Cir. 1996), involving a class of more than 10,000 Philippine nationals who had been subjected to torture and execution by Ferdinand Marcos' regime. Of those claims, a small percentage were randomly selected by an expert statistician and then evaluated by a special master, who made recommendations as to which of those claims were valid. Those findings were then extrapolated to the class—multiplying the percentage of valid claims by the average recommended award from the sample. This extrapolation theory has become an innovative tool to determine classwide issues to avoid trying a number of individual cases. The Ninth Circuit saw "no reason why a similar procedure . . . could not be employed in this case," and so concluded that "there exists at least one method of managing this large class action that, albeit somewhat imperfect, nonetheless protects the due process rights of all involved parties." Dukes v. Wal-Mart, 603 F.3d at 627. The Supreme Court rejected this "Trial by Formula" approach because "a class cannot be certified on the premise that Wal-Mart will not be entitled to litigate its statutory defenses to individual claims." Id. at *50.
Impact: A number of innovative class trial plans are adversely impacted by the Court's decision. A number of courts have allowed statistics and extrapolation techniques, fostered by dubious expert testimony, to establish the certification requirements, reasoning that the analysis was only preliminary. The effect of such decisions is to permit a class trial, which exponentially elevates the stakes. Though much of the Court's analysis about the statistical and expert testimony was not controlling to its ultimate decision, the Court provided a great deal of guidance—and foreshadowing—about these controversial practices that have the practical effect of depriving a defendant of its ability to present individualized defenses to class members' claims. These parts of the Dukes decision will have pervasive impact well beyond the employment context.
The Role of Daubert in Class Certifications
Another Dukes issue on which the circuit courts have split is the role of expert witnesses and challenges in the class certification context. The expert challenged in Dukes was a sociologist who testified that Wal-Mart's decentralized and subjective decision-making, combined with its strong corporate "culture" made its pay and promotion practices "vulnerable to gender discrimination." Dukes, 2011 LEXIS 4567, at *13. The Court did not expressly rule that Daubert challenges were proper in class certification challenges, although it expressed "doubt" about the district court's decision to the contrary. Id. at *27-28. Instead, as noted above, it held that the expert's views should be disregarded because he could not determine "whether 0.5 percent or 95 percent of the employment decisions at Wal-Mart might be determined by stereotyped thinking"—which was the "essential question on which [the plaintiffs'] theory of commonality depends." Id. at *29.
Impact: Although the Court did not expressly decide the Daubert issue, its guidance is abundantly clear. Many circuit courts consider dictum of the Supreme Court to be equally binding upon them, and a majority of the circuit courts of appeals had already decided that Daubert-typechallenges were appropriate when experts presented to establish the elements of Rule 23 provided scientifically or technically unreliable decisions. See American Honda Motor Co. v. Allen, 600 F.3d 813, 814 (7th Cir. 2010) (requiring district courts to "conclusively rule on the admissibility of an expert opinion prior to class certification . . . [where] that opinion is essential to the certification decision"); In re Hydrogen Peroxide Antitrust Litigation, 552 F.3d 305, 323 (3d Cir. 2008)(Expert testimony should not be "uncritically accepted as establishing a Rule 23 requirement.");In re Initial Public Offerings Securities Litigation, 471 F.3d 24, 42 (2d Cir. 2006) (trial court must resolve factual disputes, including conflicting expert testimony, before applying the Rule 23 legal standard); cf. Blades v. Monsanto Co., 400 F.3d 562, 569-70 (8th Cir. 2005) (rejecting Daubert challenge, but carefully weighing expert opinion testimony).
The Proper Roles of Rule 23(b)(2) and (b)(3) Classes
The Court also addressed whether and to what extent Rule 23(b)(2) classes can be certified when they seek monetary relief in addition to injunctive or declaratory remedies. Prior to Dukes, the circuit courts had developed a multitude of tests for answering this question.
Rule 23(b)(2) was designed for classes seeking injunctive and declaratory relief—an "indivisible" remedy—and so does not require protections for the individual rights of the absent class members, such as the right to notice and the ability to opt out. More individualized forms of relief, including damages, are typically certified under Rule 23(b)(3), which also requires a showing that common issues predominate and that a class is superior to other alternatives for resolving the claims. Although the district court certified the class under Rule 23(b)(2), the class members also sought back pay and punitive damages as part of their relief.
The Supreme Court held that individual claims for monetary relief "belong in Rule 23(b)(3)." Id. at *42. Further, Wal-Mart was entitled to an individualized determination of each employee's eligibility for back pay, better left to the pattern-or-practice approach to meeting the requirements for employment discrimination claims. As a result, all nine justices agreed that the class was improperly certified under Rule 23(b)(2).
Impact: This decision initially settles the circuit split on how monetary claims are to be treated and is likely to curb significantly the practice of gerrymandering class claims to fit within (b)(2), such as the plaintiffs in Dukes had done by sacrificing the class members' claims for compensatory damages. Id. at *46. As a practical matter, requiring monetary relief claims to be litigated under Rule 23(b)(3) significantly raises the stakes for class plaintiffs and their counsel. Although Justice Scalia was careful not to exclude all monetary claims from Rule 23(b)(2) classes, he also made it clear that they will be the exception, not the rule.
In the high-stakes world of complex litigation, courts, commentators, and practitioners alike have been wrestling with a number of vexing issues. In particular, consumer and labor class actions in particular have been on the rise—between 2001 and early 2007, they have increased 156% and employment class actions have increased over 200%. The trend has been growing against certification of class actions due to considerations of individual rights of both the plaintiffs and defendants, but the need for some sort of aggregate resolution of litigation stemming from mass disasters, wide-reaching business practices, and pervasive schemes has outpaced the historical judicial restraints. The Dukes decision is expected to reverse this trend and significantly limit the number of classes appropriate for certification. Its impact should also reach beyond the federal courts, as many states have patterned their class action practices on Federal Rule 23 and look to federal authorities as guidance for interpretation of their local rules.
This article was prepared by Marcy Greer (email@example.com or 512 536 4581), Dan McClure (firstname.lastname@example.org or 713 651 5159), Jonathan Franklin (email@example.com or 202 662 0466), Brian Greig (firstname.lastname@example.org or 512 536 4510) and Jaclyn Hermes (email@example.com or 512 536 2465) from Fulbright's Litigation & Appellate, Class Action, and Labor and Employment Practice Groups.