Monthly Archives: April 2013

« February 2013 (1)
May 2013 (1) »

The Uncertain Impact of Wal-Mart v. Dukes

Elizabeth Tippett, Robbing a Barren Vault: the Implications of Dukes v. Wal-Mart for Cases Challenging Subjective Employment Practices, 29 Hofstra Lab. & Emp. L. J. 433 (2012).

Melissa Hart, Civil Rights and Systemic Wrongs, 32 Berk. J. of Emp. & Lab. Law, 455 (2011).

It has been less than two years since the Supreme Court’s controversial decision in Wal-Mart v. Dukes, 131 S.Ct. 2541 (2011). During this short period the Court’s opinion has been interpreted by numerous lower courts. It also, not surprisingly, has been the subject of a substantial amount of commentary in law reviews and numerous proposals for legislative reform to restore a promise of class action challenges to employment discrimination that the Dukes decision allegedly shattered. Drawing from this commentary, I would choose these two very different articles as useful guides for tracking the impact of Dukes on employment discrimination class action litigation. The articles, in my view, together make the case that at least in the absence of legislative or judicial qualification, the Dukes decision’s 5-4 split holding on the commonality requirement in FRCP 23(a)(2) may have less of an impact than the Court’s unanimous dicta on the limited remedies allowed for Rule 23(b)(2) classes and the unavailability of statistical modeling to facilitate the certification of Rule 23(b)(3) classes.

Professor Tippett’s article provides support for questioning the impact on employment discrimination litigation of the Court’s holding that the plaintiffs in Dukes failed to provide support for there being a question of law or fact that would have a common answer for all members of the class of female nation-wide Wal-Mart employees for whom class certification was sought. Tippett does not attempt to narrow the thrust of Justice Scalia’s opinion for the Dukes majority. She recognizes language in the opinion that holds “the bare” delegation of discretion does not qualify as a “specific employment practice” open to disparate impact challenge as a subjective employment practice under the Court’s prior decision in Watson v. Fort Worth Bank & Trust, 487 U.S. 977 (1988). Tippett also understands Scalia’s opinion broadly to preclude systemic disparate treatment pattern or practice claims being based on “a decentralized policy of subjective decision-making alone.”

Professor Tippett’s questioning of the impact of the Dukes holding instead is based on her comprehensive review of all cases filed from 2005 to mid-2011 making either disparate impact or pattern or practice claims against subjective employment practices.  She concludes that such cases were “very uncommon” pre-Dukes and that “about half” of the cases that achieved certification or survived summary judgment could do so post-Dukes. She also isolates characteristics of those cases that achieved certification that would distinguish Dukes, including the limitation of the class to a single city or facility or to those subject to a single or small group of decision-makers, and the decision-makers’ disregard of possible objective criteria. Tippett’s methodology had limitations, at least some of which she fully recognizes. Yet, I think her study provides a template or basis for comparison with post-Dukes class certification decisions in the lower courts. Are those decisions certifying classes that challenge the unguided discretion of a single or small group of decision-makers at a particular facility in a particular district? Are the decisions certifying classes that challenge other subjective practices than “bare” delegation, including the subjective application of objective criteria?

The impact of Dukes on employment discrimination litigation, however, may not be limited to its effects on the types of cases challenging subjective employment practices that can meet the standards it sets for achieving “commonality” under Rule 23(a)(2). It is of course not sufficient for class certification to meet the prerequisites, including commonality, set forth in Rule 23(a). Class certification also requires fitting in one of the three categories defined in Rule 23(b). Because Rule 23(b)(1) presumably does not offer a fit for employment discrimination classes, the Dukes Court’s unanimous conclusion that classes cannot be certified under Rule 23(b)(2) where any form of individualized monetary relief is sought makes the standards for certification under (b)(3) critical. Professor Hart’s succinct article, in my view, appropriately highlights the threat that the Dukes Court’s accompanying dicta against the use of statistical modeling in Title VII cases will make it “nearly impossible” in many employment discrimination cases to meet the (b)(3) standards of predominance and superiority, regardless of whether the cases challenge subjective or objective policies or practices. As Hart explains, Justice Scalia interpreted the Court’s decision in Teamsters v. United States, 431 U.S. 324 (1977) and section 706(g) of Title VII to require a determination through additional individualized proceedings of which particular members of the class are due monetary relief because of being actually injured by the discrimination. In most employment discrimination cases, the need for such individualized proceedings could make certification of a class of any significant size seem impractical. Hart thus views Justice Scalia’s rejection of the avoidance of individualized proceedings by any form of sampling to determine the aggregate level of class-wide damages to be distributed throughout the class to be as “troubling” as his treatment of commonality.

Professor Hart persuasively argues that Justice Scalia’s interpretation of Teamsters and of section 706(g) were incorrect. She explains further why statistical modeling may be both more efficient, and also fairer for a plaintiff class, while employers care only about their aggregate liability, rather than the identities of individual victims. Hart concludes not only with a proposal for legislative change, but also with the “hope” that courts will use their discretion under Rule 23(c)(4) to certify a class “with respect to particular issues” to allow certification of a (b)(2) class for purposes of obtaining injunctive relief against discriminatory practices without preclusion of subsequent individual actions for monetary relief.

Regardless of whether Professor Hart’s criticism of Justice Scalia’s analysis is fully correct, her article raises critical questions that must be answered by the lower courts before the impact of the Dukes decision can be fully assessed. For instance, to what extent will lower courts reject (b)(3) certification because of the prospect of burdensome individual proceedings, even though prior settlement is likely? How will the critical relationship between Rules 23(b)(3) and 23(c)(4) be resolved? The answers to questions like these posed by the court’s dicta against statistical modeling ultimately may be more important than those posed by the Court’s holding on commonality.