Welcome to the Washington Legal Foundation
HOME  |  CONTACT US  |  SUPPORT WLF
Welcome to the Washington Legal Foundation

Publication Detail


Wal-Mart v. Dukes: Rough Justice Without Due Process
Topic: Civil Justice Reform
By Andrew J. Trask, Counsel in the Complex Commercial Litigation department at McGuireWoods LLP.
Legal Opinion Letter, March 11, 2011, 2 pages
Download a PDF of the Publication
Full Publication Content:

 

WLF Legal Opinion Letter

Wal-Mart v. Dukes: Rough Justice Without Due Process

By Andrew J. Trask
March 11, 2011 (Vol. 20 No. 4)

It has been called the class too large to certify, and it is certainly the opinion that launched a dozen law-review articles.  In Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571 (9th Cir. 2010), an en banc panel of the U.S. Court of Appeals for the Ninth Circuit affirmed the certification of a class of "all women employed by Wal-Mart at any time after December 26, 1998."  Id. at 577.  The decision has sparked heated debate on several fronts, from advocates who hail it as an important step forward in protecting women's rights to critics who worry that the Ninth Circuit may have sought rough justice at the expense of due process. 

In this case, the critics have it right.  The Ninth Circuit has an established reputation for pushing legal boundaries to achieve results it deems just.  See Kevin M. Scott, Supreme Court Reversals of the Ninth Circuit, 48 ARIZ. L. REV. 341, 352-53 (2006).  Pushing boundaries is not necessarily a bad thing, if doing so produces better results consistent with due process.  But that is not what happened in Dukes

Dukes is a Title VII sex-discrimination case.  The plaintiffs alleged that, as women, they received less money for comparable work and were passed over for promotions.  Their complaint sought injunctive relief (not specified in the en banc opinion), "back pay" (technically monetary, but considered equitable by many courts), and punitive damages.  The trial court certified a class under Rule 23(b)(2), a provision generally used for injunctive relief as opposed to money damages.  Wal-Mart appealed, challenging (among other issues) the trial court's use of Rule 23(b)(2) and its reliance on plaintiffs' statistical expert to establish common issues. A three-judge panel of the Ninth Circuit upheld the certification, as did a subsequent en banc panel.

So Wal-Mart took the question to the Supreme Court, which granted certiorari to review two questions: (1) when plaintiffs can seek Rule 23(b)(2) certification for a class that seeks money damages, and (2) sua sponte, "[w]hether the class certification ordered under Rule 23(b)(2) was consistent with Rule 23(a)."

There are several questionable rulings in the en banc opinion.  (Among others, Wal-Mart challenged the reliance on anecdotal evidence, the decision to relieve the plaintiffs of their burden to prove store-by-store discrimination or discriminatory animus, and the denial of its rights to assert legitimate statutory defenses.)  But two holdings stand out as particularly misguided.

First, the en banc opinion blurred the lines between Rule 23(b)(2) and Rule 23(b)(3).  It upheld certification of class claims--including for "back pay"--under Rule 23(b)(2), by calling the back pay "restitution" (equitable relief) instead of monetary damages.  Why should this matter to anyone?  Because Rule 23(b)(2) does not have the same notice requirements as Rule 23(b)(3), and does not require a showing either that common issues predominate or that a class action is superior to other methods of resolving the controversy.

The Ninth Circuit held that 23(b)(2) certification was appropriate because the injunctive relief sought had "superior strength" to the back-pay claims.  Dukes, 603 F.3d at 616.  Leaving aside the question of whether claims worth billions of dollars could ever be considered "weaker" than an injunction, the Ninth Circuit's holding misunderstood 23(b)(2)'s role.  Specifically, the Ninth Circuit misapplied the "cohesiveness" requirement of Rule 23(b)(2).  Because Rule 23(b)(2) does not allow members to "opt out," it actually requires a class to be more cohesive than Rule 23(b)(3).  See Kartman v. State Farm Mut. Auto. Ins. Co., No. 09-1725, slip op. at 19 n.8 (7th Cir. Feb. 14, 2011); Allison v. Citgo Petroleum Corp., 151 F. 3d 402, 413 (5th Cir. 1998).  And, as Wal-Mart has since argued to the Supreme Court, Rule 23(b)(2)'s history as a remedy for segregation means the drafters never considered how to administer millions--let alone billions--of dollars in relief without an opt-out mechanism.  If certifying a class is not appropriate under Rule 23(b)(3) because individualized facts would preclude efficient adjudication, that difficulty does not disappear just because the court calls the relief "restitution" instead of "damages."  The inquiries required to determine which class members deserve how much back pay (and on what grounds) are every bit as difficult.  (Title VII treats back-pay awards as discretionary instead of mandatory for just this reason.)

Second, the Ninth Circuit erred when it held that a court may rely on a plaintiff's statistical evidence to find commonality, even if the defendant has contested that evidence.  Dukes, 603 F.3d at 601-02. The commonality requirement exists to ensure that--if the class action is tried--the verdict can apply to everyone in the class equally.  When plaintiffs (or courts) define the common issues too generally, they risk producing a verdict that is useless to absent class members who may seek to enforce their rights later.

In this case, the plaintiffs' "common question" is not whether sex discrimination caused each hiring or promotion decision, but whether Wal-Mart's corporate culture was "vulnerable" to sex discrimination.  Id. at 601.  But a verdict that Wal-Mart's personnel decisions were "vulnerable" to sex discrimination does nothing to help a jury decide whether any particular decision was actually caused by sex discrimination. 

The Ninth Circuit reached this finding because it underestimated the need to decide conflicts between experts' evidence before certifying a class.  Three other circuits--the Second, Third, and Seventh--have ruled that a trial court should decide the admissibility of expert evidence before certifying a class if the experts disagree about what evidence constitutes classwide proof.  See Am. Honda Motor Co., Inc. v. Allen, 600 F.3d 813, 815-16 (7th Cir. 2010) (per curiam); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 323-24 (3d Cir. 2008); In re IPO Sec. Litig., 471 F.3d 24, 41 (2d Cir. 2006).  (The Fourth Circuit has aligned with the Ninth.  See Brown v. Nucor. Inc., 576 F.3d 149, 156 n.10 (4th Cir. 2009).)  The Ninth Circuit held that "[t]he disagreement" between the two experts regarding the appropriate level of aggregation "is the common question." Dukes, 603 F.3d at 609 (emphasis in original).  This ruling creates poor strategic incentives: it encourages plaintiffs to find an expert--any expert--who will support their certification arguments, regardless of whether her methods could ultimately pass scrutiny under Federal Rule of Evidence 702.

Given these issues, it is clear that the Ninth Circuit certified a class without considering whether the parties could try the case consistent with due process.  Its ruling did not consider how the plaintiffs' claims would actually be tried, or how relief could actually be administered.  In prizing rough justice over due process, the Ninth Circuit may have sacrificed both.

Andrew J. Trask is Counsel in the Complex Commercial Litigation department at McGuireWoods LLP. He is co-author of THE CLASS ACTION PLAYBOOK (Oxford Univ. Press 2010), and maintains the Class Action Countermeasures blog at classactioncountermeasures.com.