Monday, June 20, 2011

Win for Wal-Mart in Dukes Case

Scotus The Supreme Court issued it's opinion today in Wal-Mart v. Dukes, the big class action gender discrimination case, and it's not entirely surprising that the Court reversed the Ninth Circuit's certification of the class. All nine of the justices disagreed that the class could be certified under 23(b)(3), which allows a class to be certified if the class is seeking injunctive relief unless monetary claims of the class members would predominate. On the question of certification under 23(a)(2), the split was 5-4 on the lines we've come to expect. Here's the syllabus:

1. The certification of the plaintiff class  was not consistent with Rule 23(a).  Pp. 8–20.

(a) Rule 23(a)(2) requires a party seeking class certification to prove that the class has common “questions of law or  fact.” Their claims must depend upon a common contention of such a nature that it is capable of classwide resolution—which means that determination of its truth or falsity will resolve an  issue that is central to the validity of each one of the claims in one stroke. Here, proof of commonality necessarily overlaps with respondents’ merits contention that Wal-Mart engages in a pattern or practice of discrimination. The crux of a Title VII  inquiry is “the reason for a particular employment decision,” Cooper v. Federal Reserve Bank of Richmond, 467 U. S. 867, 876, and respondents wish to sue for millions of employment decisions at once.  Without some glue holding together the alleged reasons for those decisions, it will be impossible to say that examination of  all  the class  members’ claims will produce a common answer to the crucial discrimination question.  Pp. 8–12.

(b) General Telephone Co. of Southwest v. Falcon, 457 U. S. 147, describes  the  proper approach to commonality.  On the facts of this case, the conceptual gap between an individual’s discrimination claimand “the existence of a class of persons who have suffered the sameinjury,”  id., at 157–158, must be  bridged  by “[s]ignificant proof that an employer operated under a general policy of discrimination,”  id., at 159, n. 15.  Such proof is absent here.  Wal-Mart’s announced policy forbids sex discrimination, and the company has penalties for denials of equal opportunity. Respondents’ only evidence of a general discrimination policy was a sociologist’s analysis asserting that WalMart’s corporate culture made it vulnerable to gender bias.  But because he could not estimate what percent of Wal-Mart employment decisions might be determined by stereotypical thinking,  his testimony was worlds away from “significant proof” that Wal-Mart “operated under a general policy of discrimination.” Pp. 12–14.

(c) The only corporate policy that the plaintiffs’ evidence convincingly establishes is Wal-Mart’s “policy” of giving local supervisors discretion over employment matters. While such a policy could be the basis of a Title VII disparate-impact claim, recognizing that a claim “can” exist does not mean that every employee in a company with that policy has a common claim. In a company of Wal-Mart’s size and geographical scope, it is unlikely that all managers would exercise their discretion in a common way without some common direction. Respondents’ attempt to show such direction by means of statistical and anecdotal evidence falls well short.  Pp. 14–20.

2. Respondents’ backpay claims were improperly  certified under Rule 23(b)(2).  Pp. 20–27. 

(a) Claims for monetary relief may not be certified under Rule23(b)(2), at least where the monetary relief is not incidental to the requested injunctive or declaratory relief.  It is unnecessary to decide whether monetary claims can ever be certified under  the Rule because, at a minimum, claims for individualized relief, like backpay, are excluded.  Rule 23(b)(2) applies only  when a single, indivisibleremedy would provide  relief to each  class member.   The Rule’s history and structure indicate  that individualized monetary claims belong instead in Rule 23(b)(3), with its procedural protections of predominance, superiority, mandatory notice, and  the right to opt out.Pp. 20–23.  

(b) Respondents nonetheless argue that their backpay claims were appropriately certified under Rule 23(b)(2) because those claimsdo not “predominate” over their injunctive and declaratory  relief requests.  That  interpretation has no basis  in the Rule’s text and does obvious violence to the Rule’s structural  features. The mere “predominance” of a proper (b)(2) injunctive claim does nothing to justify eliminating Rule 23(b)(3)’s procedural protections, and creates incentives for class representatives to place at risk potentially valid monetary relief claims. Moreover, a district court would have to reevaluate the roster of class members continuously to excise those who leave their employment and become ineligible for classwide injunctive or declaratory relief.  By contrast, in a properly certified (b)(3)class action for backpay, it would be irrelevant whether the plaintiffsare still employed at Wal-Mart. It follows that backpay claimsshould not be certified under Rule 23(b)(2). Pp. 23–26.

(c) It is unnecessary to decide whether there are any forms of “incidental” monetary relief that are consistent with the above interpretation of  Rule 23(b)(2) and  the Due Process Clause  because respondents’ backpay claims are  not  incidental  to  their requested injunction.  Wal-Mart is entitled to individualized determinations of each employee’s eligibility for backpay.  Once a plaintiff establishes apattern or practice of discrimination, a district court must usuallyconduct “additional proceedings . . . to determine the scope of individual relief.”  Teamsters v. United States, 431 U. S. 324, 361.  The company can  then raise individual affirmative defenses and demonstrate that its action was lawful.  Id., at 362.  The Ninth Circuit erred in trying to replace such proceedings with Trial by Formula. Because Rule 23 cannot be interpreted to “abridge, enlarge or modify any substantive right,” 28 U. S. C. §2072(b), a class cannot be certified on thepremise that Wal-Mart will not be entitled to litigate its statutory defenses to individual claims.  Pp. 26–27. 603 F. 3d 571, reversed. 

My initial iimpressions are probably pretty obvious. First, this case came down to the size and complexity of both Wal-Mart's operation and the potential class. A majority of the Court simply did not believe that all of the class members could possibly be injured in the same way given the multitude of decisionmakers at issue--even for the disparate impact claim, which doesn't require intent. Second, a majority of members of the Court do not seem to believe that causation can be proven by statistical analysis, something that is evident not just in employment discrimination cases, but in other areas as well. A majority of the Court seems very suspicious of the idea that implicit bias could support a claim for disparate treatment--that doesn't seem the right kind of bias that those Justices seem to think must form the basis for a disparate treatment case. Particularly damaging to the use of that evidence was an article by Walker, Monahan, and Miller that criticized Bielby's use of a method they've written about.

I'll continue to drill down, but feel free to add your own impressions in the comments. The decision does not bode well for any kind of systemic case, large class action, or implicit bias case, in my opinion.


Employment Discrimination, Labor and Employment News | Permalink

TrackBack URL for this entry:

Listed below are links to weblogs that reference Win for Wal-Mart in Dukes Case:


At 1st glance: Disappointingly weak and technical dissent from Ginsburg, given the stakes of this case. No attempt from either side to conceptualize systemic disparate treatment as anything more than the accumulation of individual episodes of disparate treatment. Terrible!

Posted by: Noah Zatz | Jun 20, 2011 11:34:50 AM

While all of this is in the context of a class action certification issue, the decision is quite unsettling as to the future of systemic cases substantively. J. Scalia's opinion entirely misses what the nature of the claim is, as I understand it. The essence is that, while controlling from Bentonville every aspect of every Wal-Mart location's operation, Wal-Mart grants unstructured discretion to store managers to set pay and to promote workers. While that is not discriminatory on its face, the question is whether in its administration it is a discriminatory practice amounting to systemic disparate treatment. The first step to prove that would be to show the statistics of how that policy workers. Plaintiffs had strong statistical evidence based on multiple regressions that rejected the hypothesis that sex was not related to pay and promotion decision. The second step is to determine whether the correlation is because of sex. That it cannot be said that pay and promotion are unrelated starts to get to a yes. Also, the data about all of what was happening was collected in Bentonville, so HQs knew this was happening. But, unlike other areas, they did nothing about the severe gender imbalance. Further, Professor Bielby’s report bolsters that, as does the anecdotal evidence. Finally, the juxtaposition of tight central control of almost everything other than these decisions bolsters, I think, aupports drawing the inference of causation without some explanation about why a general principle of Wal-Mart management does not apply to these decisions.

J. Scalia mixes up the whole analysis thereby potentially overruling Teamsters, Hazelwood, Basemore and even Yick Wo v. Hopkins. Dukes can be distinguished from these since it is an interpretation of Rule 23 and not of Title VII. But how likely is that?

J. Scalia overlooks how this policy injured all the class members. Given the grant of unstructured and unreviewed discretion of the store managers, all the class members had a sex-based risk that decisions made about their pay or their promotions would be discriminatory. The risk of discrimination is an injury since Northeastern Florida Contractors, an affirmative action case.

Posted by: Mike Zimmer | Jun 20, 2011 3:15:30 PM

I agree that the issue is really statistical proof of discrimination, not class certification, and that it is hard to see how one could bring a case like Hazelwood after this decision. Perhaps if the "anecdotes" were more similar or numerous. OTOH, I am no fan of the corporate-wide class action, for reasons well-explored by Michael Selmi, 81 Tex.L.Rev. 1249 (2003). Alexis Herman will now sit on fewer useless implementation committees, but few employees of color ever saw any benefit from cases like Texaco, Coca-Cola, or Home Depot.

Posted by: Alan Hyde | Jun 21, 2011 7:49:53 AM

Post a comment