In Wal-Mart Stores, Inc. v. Dukes, (2) the United States Supreme Court rejected class certification in a gargantuan gender discrimination case against Wal-Mart. The primary issue, as the Supreme Court framed it, was whether the showing by the plaintiffs-present and former female employees-met Rule 23(a)(2)'s "commonality" requirement. More specifically, could the litigation be said to present "questions of law or fact common to the class" where the plaintiffs did not demonstrate an express Wal-Mart policy or practice of discriminating against women, but instead showed, at most, only that: (1) Wal-Mart granted discretion to thousands of store managers to make pay and promotion decisions in a subjective manner; and (2) the company's corporate culture and personnel practices made it "vulnerable" to gender discrimination.(3) The Supreme Court's five-justice majority held that this was not enough to show commonality, and that the lower courts had, accordingly, erred in certifying a class. The four dissenting justices expressed the view that the majority's rejection of certification for lack of commonality "import[ed] into the Rule 23(a) determination concerns properly addressed in a Rule 23(b)(3) assessment,"(4) where "the questions of law or fact common to class members [must] predominate over any questions affecting only individual members."(5)
Dukes may well have significant implications for Title VII civil rights class actions, particularly those in which circumstantial evidence is central to proving the discriminatory policy or practice. But the Court's ruling is unlikely to have a major impact in antitrust cases. If an antitrust complaint is sufficient to plead price-fixing or (less frequently) monopolization, there is inevitably a common question. Either the defendants conspired to fix prices, or the dominant player engaged in distribution or pricing conduct, that had downstream or upstream effects. Thus, the class certification battle is unlikely to be waged over commonality, but will instead take place under Rule 23(b)(3)'s predominance requirement, alluded to by the Dukes dissent. On these issues, Dukes, a 23(a)(2) ruling, offers no real guidance.(6)
II. Dukes' Background Facts
Three current and former female employees sought to represent a nationwide class of 1.5 million female employees of Wal-Mart, the largest private employer in the world. The plaintiffs alleged that Wal-Mart violated Title VII of the Civil Rights Act of 1964 by adopting and implementing a corporate policy of gender discrimination that denied female employees equal pay and opportunities for promotion. Wal-Mart had no express policy or practice of discrimination against women. Therefore, the plaintiffs had to offer evidence from which discrimination affecting the plaintiffs and the proposed class members could be inferred.
The United States District Court for the Northern District of California certified the class, and the Ninth Circuit Court of appeals, sitting en banc, largely affirmed.(7) The Supreme Court reversed on two grounds. First, a 5-4 majority held that the proposed class did not satisfy Rule 23(a)(2)'s requirement that class members share common claims. Second, a unanimous Court determined that the plaintiffs' claims for back-pay were improperly certified under Rule 23(b)(2), which covers injunctive and declaratory claims. The back-pay claims, the Court held, were not incidental monetary relief. Accordingly, due process requires that class certification be evaluated under Rule 23(b)(3), not under Rule 23(b)(2). We discuss here only the Court's commonality ruling. The Court's Rule 23(b)(2) discussion has no readily discernible application to antitrust cases.
III. Establishing A Rigorous Standard For Satisfying Rule 23(a)(2) Commonality
Before addressing the parties' class certification contentions, the Supreme Court clarified that Rule 23 does not set forth a mere pleading standard. A party seeking class certification "must be prepared to prove" each element of Rule 23, and a court must determine, after a "rigorous analysis" that Rule has been satisfied, which "[f]requently . . . will entail some overlap with the merits of the plaintiff's underlying claim."(8) This part of Dukes has its genesis in language in Eisen v. Carlisle & Jacquelin,(9) which some lower courts had construed as precluding consideration of any aspect of the merits of a case in deciding class certification. The Dukes Court rejected this approach, noting that Eisen's facts were distinguishable, and that the relevant Eisen language was "the purest dictum and is contradicted by our other cases."(10)
The Supreme Court examined whether "questions of law or fact common to the class" existed under Rule 23(a)(2) as a threshold step in considering whether to certify under Rule 23(b)(2). The Court emphasized that for commonality, "[w]hat matters to class certification . . . is not the raising of common questions-even in droves-but, rather the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation."(11) Thus, Dukes does not hold that a plaintiff must prove its case on the merits at the class certification stage, but rather requires that the common questions of Rule 23(a)(2) be susceptible to common answers.(12) The Court confirmed that even a single common question will satisfy Rule 23(a)(2).(13)
IV. Rigorous Analysis In Determining Commonality
The Supreme Court observed that in an employment discrimination class action, such as that against Wal-Mart, "[c]onceptually, there is a wide gap between (a) an individual's claim that he has been denied a promotion [or higher pay] on discriminatory grounds, and his otherwise unsupported allegation that the company has a policy of discrimination, and (b) the existence of a class of persons who have suffered the same injury as that individual, such that the individual's claim and the class claim will share common questions of law or fact that the individual's claim will be typical of the class claims."(14) The Court noted that its opinion in Falcon(15) suggested two ways to bridge the conceptual gap. First, the proposed class representatives could offer proof that the employer used a biased testing procedure that affected the entire class. Second, they could offer "[s]ignificant proof that an employer operated under a general policy of discrimination."(16) The first approach was inapplicable to the Dukes plaintiffs. And on the second, significant proof that Wal-Mart operated under a general policy of discrimination was, the Court found, "entirely absent."(17)
The Supreme Court rejected the plaintiffs' attempt to demonstrate a common issue of company-wide gender discrimination through circumstantial evidence consisting of: (a) testimony from a sociologist on Wal-Mart's corporate culture; (b) testimony from a statistician, working in conjunction with a labor economist, on pay and promotion disparities in some of Wal-Mart's 3,400 stores; and (c) anecdotes from 120 of Wal-Mart's female employees.
By way of summary, the plaintiffs' sociological expert testified that Wal-Mart had a strong corporate culture, which made it vulnerable to gender bias. On deposition, however, the plaintiffs' expert conceded that he could not "determine with any specificity how regularly stereotypes play a meaningful role in employment decisions at Wal-Mart."(18) In light of this concession, the Court found that it could "safely disregard" the expert's testimony because "[i]t is worlds away from significant proof that Wal-Mart operated under a general policy of discrimination."(19) To the contrary, the Court found the only Wal-Mart policy that the plaintiffs' evidence established was that of allowing discretion by local managers over employment matters. "On its face," the Court concluded, "that is just the opposite of a uniform employment practice that would provide the commonality needed for a class action; it is a policy against having uniform employment practices."(20)
The Court also observed the parties' dispute over whether the testimony of the plaintiffs' expert sociologist met the standard for admissibility under Federal Rule of Evidence 702 and the Court's decision in Daubert v. Merrell Dow Pharmaceuticals, Inc.(21) The district court had ruled that Daubert did not apply to expert testimony at the certification stage. In dicta, the Supreme Court expressed skepticism with the lower court's conclusion, but left the issue unresolved because, in all events, the plaintiffs' expert testimony did "nothing to advance [plaintiffs'] case."(22)
Through their statistician and labor economist, the plaintiffs proffered regression analyses that arguably demonstrated statistically significant compensation and promotions disparities between men and women at Wal-Mart-disparities that the expert asserted could be explained only by gender discrimination. The Supreme Court held that this evidence was insufficient to establish that the plaintiffs' theory could be proven on a class-wide basis for two reasons. First, the experts' analyses of Wal-Mart's data, which were conducted at the regional level, did "not establish the existence of disparities at individual stores, let alone raise the inference that a company-wide policy of discrimination is implemented by discretionary decisions at the store and district level."(23) Second, even if the plaintiffs' experts could demonstrate that pay and promotion patterns in all of Wal-Mart's 3,400 stores differed from the national average, that difference, the Court said, still would be insufficient to prove commonality. Merely showing that Wal-Mart's discretionary policy produced a gender disparity was not enough, as the plaintiffs had failed to "identify the specific employment practice that is challenged."(24)
Finally, as anecdotal evidence, the plaintiffs offered 120 affidavits-one for every 12,500 class members. The Court held that the evidence was "too weak to raise any inference that all the individual, discretionary personnel decisions are discriminatory."(25) The reports were concentrated in only six states. Half of the states had only one or two anecdotes, and 14 states had none at all. This material, the Court concluded, was insufficient to infer a nationwide policy of discrimination.(26)
V. Dukes' Divergence from Certification in Antitrust Cases
The lower courts certified the Dukes class for injunctive relief and back pay under Rule 23(b)(2), and that provision, unlike Rule 23(b)(3), has no predominance element. Moreover, the Supreme Court held that, even before reaching Rule 23(b)(2), the plaintiffs had failed to meet Rule 23(a)(2)'s commonality requirement. The plaintiffs' theory that Wal-Mart granted discretion to thousands of managers to make pay and promotion decisions in a subjective manner, the Court reasoned, was "the opposite of a uniform employment practice that would provide the commonality needed for a class action."(27) In consequence, nothing in the Dukes decision speaks to Rule 23(b)(3)'s requirement that common questions must "predominate over any questions affecting only individual members . . . ."(28)
Rule 23(b)(3), however, is where the strike lines regularly are drawn on class certification in antitrust cases. In Dukes, the common question-whether Wal-Mart practiced gender discrimination-had to be inferred from ambiguous circumstantial evidence. On the other hand, in an antitrust case, the legal sufficiency of, for example, the plaintiff's price-fixing allegations will invariably be tested on a defense motion to dismiss. If the allegations are insufficient, the case will be dismissed, thereby obviating any class certification inquiry. If, however, the complaint passes muster, commonality itself will not be subject to genuine attack in the certification proceedings.
The existence of the price-fixing conspiracy, and the product overcharges resulting from it, present common questions that are susceptible of class-wide proof. As a leading class action commentator has observed, "[i]n an antitrust action on behalf of purchasers who have bought the defendants' products at prices that have been maintained above competitive levels by unlawful conduct, the courts have held that the existence of an alleged monopoly or conspiracy is a common issue that will satisfy the Rule 23(a)(2) prerequisite."(29) Put another way, commonality in a price-fixing case emerges with a clarity that just is not there in a case alleging employment discrimination based on circumstantial evidence of disparity among employee compensation or promotion.
With commonality virtually a foregone conclusion in a price-fixing case, class certification often hinges, instead, on whether class member injury from the violation, often referred to as "impact," is amenable to common proof.(30) Defendants regularly argue that an antitrust class may not be certified because the overcharge resulting from their illegal conduct supposedly did not impact all of the defendants' customers to the same extent. This argument is one of predominance, asserted under Rule 23(b)(3), where the plaintiff's burden is greater than it is for the commonality required under Rule 23(a)(2).(31)
Commonality under Rule 23(a)(2) can be thought of as an "on-off" element: either the case presents one or more common questions, or it does not. By contrast, predominance under Rule 23(b)(3) calls for a more fine-tuned assessment of the role of the common questions in the array of issues that the case raises, and in the relationship of the common issues to individual ones that might be raised.
It would be a rare price-fixing case, indeed, where the defense side could not put forth any debatable individual issue at all. Customers come in different shapes and sizes, and they routinely negotiate their purchases in ways that could, arguably, distinguish them from one another. A defendant, however, should not be able to defeat predominance simply by identifying an individual customer or group of customers for whom impact from the price-fixing conspiracy differs from that of class members at large. It is here that "rigorous analysis" works both ways: common questions can well "predominate" even amidst individual variation.(32)
In sum, Dukes' is not likely to have a significant effect on class certification proceedings in antitrust class actions. The court's recent certification ruling in In re Aftermarket Automotive Lighting Products Antitrust Litigation(33) is instructive. There, purchasers of aftermarket auto lights alleged a price-fixing action against the industry's leading manufacturers and their distributor-subsidiaries. The briefing on class certification was nearly complete at the point that the Supreme Court issued Dukes. The defendants thus made a post- Dukes filing, arguing among other things that "[i]ndividual inquiry into the independent factors affecting pricing decisions on a product-byproduct basis is required, just as Wal-Mart's decisions to hire male over female employees in any particular case must be evaluated on an individual, rather than classwide basis."(34) The Central District of California certified the class nonetheless:
[T]he existence of a "general policy" of price-fixing is-at least for the purpose of this motion-undisputed. The question to be resolved is . . . whether Plaintiffs will be able to present proof of impact and damage resulting from this alleged policy on a class-wide basis. Nothing in Wal-Mart Stores suggests that Plaintiffs will inevitably be unable to present such evidence in this case.(35)
Dukes did not address the requirement of predominance-much less did it brighten the analysis or heighten the evidentiary standard by which predominance must be shown. Because the Court did not enter this fray at all, on class certification antitrust plaintiffs and defendants will continue to duke it out.
1 The authors are, respectively, Partner and Associate, Labaton Sucharow LLP, New York City. Mr. Himes, also Co-Chair of the firm's Antitrust Practice Group, is the former Antitrust Bureau Chief, Office of the Attorney General of New York.
2 131 S.Ct. 2541 (2011).
3 See id. at 2549.
4 Id. at 2562.
5 Id. at 2566 (internal quotations omitted).
6 Rule 23(b)(3) also requires that the class action be "superior to other available methods for fairly and efficiently adjudicating the controversy." However, in antitrust cases if predominance is shown, the additional "superiority" element is unlikely to be an obstacle. Also, for simplicity's sake, we discuss here only price-fixing cases. Monopolization cases are not, for purposes of our discussion, materially different.
7 Dukes v. Wal-Mart Stores, Inc., 222 F.R.D. 137 (N.D. Cal. 2004), aff'd in part, remanded in part, 603 F.3d 571 (9th Cir. 2010).
8 Dukes, 131 S.Ct. at 2551 (citing Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 160 (1982)).
9 417 U.S. 156, 177 (1974).
10 Dukes, 131 S.Ct. at 2552 n.6.
11 Id. at 2551 (internal quotations omitted).
13 Id. at 2556-57.
14 Id. at 2552.
15 457 U.S. 147.
16 Dukes, 131 S.Ct. at 2553 (internal quotations omitted).
18 Id. (internal quotations omitted).
19 Id. at 2554 (internal quotations omitted).
21 509 U.S. 579 (1993).
22 Dukes, 131S.Ct. at 2554. This is, however, a substantial issue that probably will reach the Supreme Court again. In a pre- Dukes opinion, the Seventh Circuit held that "when an expert's report or testimony is critical to class certification . . . the district court must perform a full Daubert analysis before certifying the class." Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815 (7th Cir. 2010). However, in a post- Dukes ruling, the Eighth Circuit distinguished American Honda, concluding that defendant's "desire for an exhaustive and conclusive Daubert inquiry [at the class certification stage] . . . cannot be reconciled with the inherently preliminary nature of pretrial evidentiary and class certification rulings." In re Zurn Pex Plumbing Prods. Liab. Litig., No. 10-2267, --F.3d--, 2011 WL 2623342, at *5 (8th Cir. July 6, 2011).
23 Dukes, 131 S.Ct. at 255.
25 Id. at 2556.
27 Id. at 2554.
28 Id. at 2558.
29 Herbert B. Newberg & Alba Conte, Newberg on Class Actions § 3.10 (4th ed. 2002).
30 See, e.g., In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 307 (3d Cir. 2008); In re Flash Memory Antitrust Litig., No. C 07-0086, 2010 WL 2332081, at *7 (N.D. Cal. June 9, 2010); see also Hal J. Singer, Economic Evidence of Common Impact for Class Certification in Antitrust Cases: A Two-Step Analysis, 25 ANTITRUST 34 (Summer 2011).
31 See, e.g., Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623-24 (1997) ("Even if Rule 23(a)'s commonality requirement may be satisfied ... the predominance criterion is far more demanding").
32 As Judge Posner has written:[A] class will often include persons who have not been injured by the defendant's conduct; indeed this is almost inevitable because at the outset of the case many of the members of the class may be unknown, or if they are known still the facts bearing on their claims may be unknown. Such a possibility or indeed inevitability does not preclude class certification . . . .
Kohen v. Pac. Inv. Mgmt. Co. LLC, 571 F.3d 672, 677 (7th Cir. 2009), cert denied, Pac. Inv. Mgmt. Co. LLC v. Hershey, 130 S.Ct. 1504 (2010). See also In re Amaranth Natural Gas Commodities Litig., 269 F.R.D. 366, 385 (S.D.N.Y. 2010) (internal quotations omitted) ("the fact that damages must be calculated on an individual basis is no impediment to class certification").
33 No. 09-mdl-2007, 2011 WL 3204588 (C.D. Cal. July 25, 2011). The authors are among the co-lead counsel for the plaintiffs in the case.
34 In re Aftermarket Automotive, Defs.' Supplemental Br. re: Newly Decided Supreme Court Authority in Opposition to Plaintiffs' Motion for Class Certification, 9 (C.D. Cal. filed June 30, 2011).
35 In re Aftermarket Automotive, 2011 WL 3204588, at *4.