gavel.jpgBy Gerald L. Maatman, Jr. and David Ross

Virtually every class action pending in federal court has undergone a re-examination based on the U.S. Supreme Court’s holding in Wal-Mart Stores Inc. v. Dukes, 131 S. Ct. 2541 (2011). The issues arises either based on a defense motion to strike or decertify a class based on Dukes or in the parties’ briefing of a motion for class certification. Defendants typically argue that the class theories are no longer viable under Dukes, and Plaintiffs often defend their class claims as being consistent with the SCOTUS ruling. Federal district court judges are left to analyze whether Dukes completely blocks, impedes in part, or green lights the Plaintiffs’ class claims.

The latest ruling framing the “Dukes effect” is Easterling, et al. v. Connecticut Department of Corrections, Case No. 08-CV-826 (D. Conn. Nov. 22, 2011). Plaintiffs in Easterling – who sued on behalf of a class of rejected female corrections officer (“CO”) applicants – alleged that a component of a Connecticut Department of Corrections (“DOC”) physical fitness test was gender biased. The facts of the case are straight forward. Easterling applied for work with the DOC in 2004. At the time, CO candidates had to first pass a written examination and then a physical fitness test that included a 1.5 mile run that needed to be completed within the time allotted for the candidate’s age/sex cohort. If a candidate passed both tests, they proceeded to the interview stage of the application process, followed by a background investigation and medical examination. The physical fitness test consisted of four parts, and failure on any one part constituted failure of the entire physical fitness test. Although Easterling passed the written exam and three of the four components of the physical fitness test, she failed to complete the 1.5 mile run within the allotted time and was not advanced to the interview stage. Id. at 2-3.She sued in 2008 under Title VII of the Civil Rights Act of 1964, asserting that the physical fitness test had a disparate impact on female CO applicants.

Earlier in the litigation, Judge Janet Hall granted class certification for plaintiffs’ claims under Rule 23(b)(2). The Court found that Easterling had standing to pursue disparate impact claims against the DOC. Among other things, she found that Easterling could redress her alleged injuries resulting from the physical fitness test through remedies such as an injunction barring the DOC from hiring from applicant lists that were the product of unlawful policies, practices, or customs. Id. at 3.

In the wake of the SCOTUS ruling in Dukes, Defendants moved for decertification. On November 22, 2011, the Court held that even though the initial certification order could not stand in light of Dukes, decertification was not warranted. In effect, Judge Hall rejected the defense motion in part, but ruled that Dukes required a conversion of the class into a “hybrid class” per Rule 23(b)(3). Specifically, Judge Hall reasoned that on account of Dukes, the class claims needed to be “re-booted” — i.e., certified under Rule 23(b)(2) for class-wide declaratory and injunctive relief, and under Rule 23(b)(3) for monetary and individualized injunctive relief.

This past May, the Court had granted summary judgment on the issue of liability to the class on the claim that the 1.5-mile run component of the physical fitness exam the DOC used for corrections officer applicants between 2004 and 2006 had a disparate impact on female candidates

The members of the class include “[a]ll female applicants for the position of Correction Officer at the State of Connecticut Department of Correction who participated in the selection process and failed only the 1.5 mile run portion of the physical fitness test at any time from June 28, 2004 and continuing to the date of final judgment in this matter.” Id. at 17.

Judge Hall determined that her original certification order under Rule 23(b)(2) could not stand in light of Dukes, which rejected the U.S. Court of Appeals for the Second Circuit’s broad reading of Rule 23(b)(2) in Robinson v. Metro-North Commuter Railroad Co., 267 F. 3d 147 (2d Cir. 2001). The Court found that while Dukes overturned an en banc opinion of the U.S. Court of Appeals for the Ninth Circuit, the Ninth Circuit’s predominance test was identical to that of the Second Circuit. Id. at 4. Specifically, prior to Dukes, Second Circuit precedent provided that Rule 23(b)(2) certification was proper of a class seeking both injunctive and monetary relief as long as the positive weight or value to the Plaintiffs of the injunctive or declaratory relief sought was predominant over the value or the monetary relief and class treatment would be efficient and manageable. Judge Hall opined that she had relied on Robinson‘s now-defunct predominance test when it initially granted Plaintiff motion for class certification. Given the ruling in Dukes, the Court reasoned that the Supreme Court’s wholesale disavowal of that test qualified as a compelling reason to re-examine the earlier certification order. Id. at 4-5.

Judge Hall framed the issues as how the Court should respond to a shift in controlling law. The choices were to: (i) revoke the certification order entirely and dismiss Easterling’s case; (ii) keep the order in place for Rule 23(b)(2) certification on the issues of liability and class-wide injunctive relief per Dukes, but require that all claims for monetary and individualized injunctive relief be pursued in separate, individual lawsuits by the class members; or (iii) adopt the hybrid approach suggested by Plaintiff, maintaining the Rule 23(b)(2) certification with regard to liability and class-wide injunctive relief and certifying a separate Rule 23(b)(3) class on the issues of monetary and individualized injunctive relief.

Finding that Dukes overturned Robinson on the propriety of a predominance test under Rule 23(b)(2), Judge Hall determined that Dukes did not overturn Robinson completely. She ruled that decertification was not appropriate because “Dukes did not reject … [the Second Circuit’s] interpretation of Rule 23(c)(4).”  Id. at 7. Under Rule 23(c)(4), Judge Hall reasoned, federal judges are exhorted to use the class action device liberally to reduce the range of disputed issues in complex litigation and achieve judicial efficiencies.  Although Robinson was a pattern or practice disparate treatment case, the Court found that “its logic is equally applicable to disparate impact suits” because liability in both of these types of cases is established via class-wide, statistical proof. Id. at 6. Further, in both types of actions, as soon as systemic adverse treatment or impact is established, the Court may enter class-wide injunctive relief without investigating the validity or value of individual class members’ claims.
For these reasons, Judge Hall determined that the prior Rule 23(b)(2) certification ruling could be kept in place as to the issues of liability and class-wide injunctive relief. Further, the hybrid approach suggested by Plaintiff could be adopted because certification of the claims for monetary and individualized injunctive relief was proper under Rule 23(b)(3).

To warrant certification under Rule 23(b)(3), a class must meet two requirements beyond the Rule 23(a) prerequisites of commonality, numerosity, typicality, and adequacy. Specifically, Plaintiff must establish that common questions predominate over any questions affecting only individual class members and class resolution is superior to all other available forms of adjudication. The predominance requirement is met if resolution of some of the factual or legal issues that qualify each class member’s case as a genuine controversy can be achieved through generalized evidence and are more substantial than questions subject only to individualized evidence. As is usually evident in most class action contexts, certification is far more difficult to establish under Rule 23(b)(3) than Rule 23(b)(2).

Nonetheless, while the remedial stage of disparate impact litigation generally is dominated by individualized proof, Judge Hall reasoned that this was not the case here. The Court opined that the goal at the remedies stage “is to recreate ‘as nearly as possible … the conditions and relationships that would have been had there been no unlawful discrimination.’ ”  Id. at 12. Although, “[i]n the present case, the Court finds that it would be impossible to determine exactly which class members would have received job offers from the DOC if a 1.5 mile run had not been part of the physical fitness test,”  the Court found that the aggregate amount of individual relief ascertainable in a 23(b)(3) proceeding (even though the precise methodology for proving the aggregate amount had not yet been determined). Id. at 14.

Based on the record, it was unsettled which class members would have passed the subsequent medical screening and criminal background check, or gotten through the interview that would have taken place had they not failed the physical fitness test. In addition, passing the medical screening, criminal background check, and interview stages still only would have resulted in a class member being certified as an applicant and placed on an eligibility list, with no assurance of being that they would have been selected from that list. Accordingly, Judge Hall indicated that rather “than resort to ‘mere guesswork,’ … the Court will make an aggregate calculation of the back pay to which the class is entitled. This sum can then be distributed to eligible class members on a pro rata basis.” Id. at 13.

The Court opined that “this method of assessing monetary relief ‘is a consequence of substantive Title VII law, and not a creative method of proof intended to accommodate the logistical demands of class proceedings,’ ” and also is appropriate for calculating the particularized individual relief the class requests, such a “priority hiring” or front pay. Id. Although individual issues still will exist after such calculations are made, “the Court finds that these individual questions are less substantial than the issues that will be subject to generalized proof.” Id. at 15.

As to superiority of class treatment under Rule 23(b)(3), Judge Hall found that “[i]t would be absurd to have more than a hundred class members separately litigate the issue of aggregate back pay, just as it would be absurd to have them separately litigate the question of class-wide liability.” Id. at 15-16. Plaintiffs asserted that there may be as many as 177 class members and the DOC acknowledged that there could be up to 125. Finally, the Court found that any difficulties likely to arise in managing this class action, “are far less daunting than the difficulties involved in litigating over a hundred separately captioned actions.” Id. at 16.

Implications For Employers
This decision differs from other post-Dukes cases where defendants more typically challenge the threshold issues of commonality under Rule 23(a) because plaintiffs failed to challenge the impact of a specific employment practice and relied instead on a theory of adverse impact arising from discretionary, subjective decision-making – a theory that Dukes largely rejects in class cases.  Here, plaintiffs specifically challenged the employment practice of a uniform pre-employment test, which Dukes allowed may support a finding of commonality under Rule 23(a) and, in principle, may support a 23(b)(2) class for injunctive and declaratory relief for the class as a whole, apart from monetary damages.  Dukes , however, did not address the use of “hybrid” certifications, where damage claims might be certified under the stricter provisions of 23(b)(3), in addition to a 23(b)(2) class for injunctive relief.  Notably, the Court’s analysis in Easterling finds the “hybrid” certification approach consistent with Dukes and applies a liberal interpretation to the 23(b)(3) requirement that common questions must predominate over individualized issues (and perhaps influenced by the Court’s reluctance to revoke class certification at a late stage in the litigation).  Because the Court found the aggregate liability ascertainable on a class basis, it also found this common issue to be predominate in determining damages, following the Robinson scheme, even though damages in any individual case would be subject to the defense that the candidate was not otherwise qualified, would not have been hired in any event, or failed to mitigate damages.

Easterling is yet another example of how “Dukes questions” continue to spawn new Rule 23 puzzles. It is also a window into how the plaintiffs’ class action bar is utilizing “re-booting” theories to find their way around some of the impediments to class certification created by Dukes.