Since the U.S. Supreme Court’s decision in Wal-Mart v. Dukes, there has been a significant amount of educated speculation about the effect of that decision on class action litigation in general and more particularly on class actions involving claims of employment discrimination.  Dukes is seen as creating an impassable barrier for class actions claiming discrimination in multiple locations based on excess subjectivity arising from decentralized decision-making.  Dukes instead focuses the inquiry on the existence and discriminatory effect of enterprise-wide policies such as an employment test or standardized performance criterion.  The question remains: what constitutes an enterprise-wide policy or practice?  This is a question that has challenged practitioners since General Tel. Co. of the Southwest v. Falcon, 457 U.S. 147, 159 n. 15 (1982), and before.

Bringing a class action seeking to invalidate an enterprise-wide policy or practice has significant ramifications. Generally, a claim that an employment policy is discriminatory will be made using the disparate impact model; finding an enterprise-wide policy that is not facially neutral is a rare event, and showing that such a policy was the result of intentional discrimination may be exceedingly difficult.  Plaintiffs’ counsel, however, generally prefer the disparate-treatment model because relief in disparate impact cases is limited to equitable relief, including back pay.  Compensatory and punitive damages are not available in disparate impact litigation.  Moreover, disparate-treatment claims must be certified, if at all, under the rigorous standards of Rule 23(b)(3).  Thus, even before Dukes, plaintiffs’ counsel faced challenges when pursuing class claims with the primary objective of recovering class-wide compensatory and punitive damages.

In this context, the Seventh Circuit’s recent decision in McReynolds v. Merrill Lynch, No. 11-3639 (February 24, 2012), is significant in two ways.  First, it uses an expansive approach in deciding what constitutes an enterprise-wide policy or practice.  Second, it permits a challenge to the validity of an enterprise-wide policy or practice using the disparate impact model and Rule 23(b)(2), while preserving claims for compensatory and punitive damages (although not on a class-wide basis).

In McReynolds plaintiffs alleged class-wide discrimination against African-American brokers because of Merrill Lynch’s company-wide “teaming” and “account distribution” policies. “The teaming policy permits brokers in the same office to form teams. They are not required to form or join teams, and many prefer to work by themselves. But many others prefer to work as part of a team. Team members share clients, and the aim in forming or joining a team is to gain access to additional clients . . . . The teams are formed by brokers, and once formed a team decides whom to admit as a new member. Complex Directors and branch-office managers do not select the team’s members.”  The “account distribution policy” applies when a broker leaves the firm, and determines how her accounts are distributed by branch management to the remaining brokers.  The administration of the “account distribution policy” involved a combination of criteria established by the company and discretion of branch-office managers. Merrill Lynch argued that, in light of Dukes, this was precisely the kind of autonomous, decentralized decision-making that could not be challenged in a class action (and, with respect to the “teaming policy” involved decisions by employees, not managers).

Judge Posner rejected Merrill Lynch’s position, and in effect allowed plaintiffs to use Dukes offensively and to claim that class certification was appropriate because plaintiffs were challenging a company-wide policy.  In McReynolds the company-wide policy was “the framework established by the company” which permitted teaming and permitted discretion in account distribution. Thus, Judge Posner concluded that “within each branch office the brokers [and managers] exercise a good deal of autonomy, though only within a framework established by the company” (emphasis added).  Accordingly, McReynolds is significant because it authorizes an expansive use of the disparate impact model to challenge aspects of an employer’s “framework” in addition to express policies or established employment practices.

McReynolds is significant in another way because it permits plaintiffs to pursue a disparate impact claim and obtain class certification under Rule 23(b)(2), yet preserve claims for compensatory and punitive damages. Judge Posner directed the certification of an “issue class” pursuant to Rule 23(c)(4), permitting the issue of the validity of the “teaming” and “account distribution” policies to be decided in a class action, leaving individualized claims for damages (including compensatory and punitive damages) to be resolved in separate proceedings.  Thus, if the court deciding the “issue class” determined that either the “teaming” or “account distribution” policies created an unlawful disparate-impact, the court would enjoin the use of the discriminatory policy.  Individual African-American brokers – class members – could then seek individual relief in individual actions in which they would be entitled, when pursuing a disparate impact claim, to a rebuttable presumption that they suffered from the discriminatory policy, and in addition pursue a disparate-treatment claim in order to seek compensatory or punitive damages.

Although Rule 23(c)(4) authorizing issue classes has existed since the 1966 amendments to Rule 23, the use of issue classes in employment discrimination litigation has been exceedingly rare.  As McReynolds shows, however, the issue class permits plaintiffs’ counsel to pursue a disparate impact claim while preserving the right to seek compensatory and punitive damages in a separate proceeding, and this raises the distinct possibility that plaintiffs’ counsel will increase the use of this procedural tool in employment litigation.