It has been a slightly over a month now since the United States Supreme Court announced its blockbuster decision in Wal-Mart Stores, Inc. v. Dukes.pdf, 564 U.S. ___ (2011), and commentators have written at length about various aspects of the decision. One area that has drawn less attention, however, is a very brief portion of the opinion expressing doubt over the Ninth's Circuit holding that the standards for experts described in Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589-90 (1993), do not apply at the class certification stage of the case.
The Daubert case was important for the integrity of the federal justice system, in that it made clear that courts were not to accept expert testimony uncritically, but should act as "gatekeepers" by scrutinizing the methodology and validity of the expert's testimony and excluding testimony that is irrelevant or unreliable. Put another way, courts were not to accept "junk science" and could not simply submit incredible or baseless testimony to a jury simply because it came out of the mouth of an expert.
One of the many issues raised in the Dukescase related to the plaintiffs' use of an expert named William Bielby, who testified that any organization with a "strong corporate culture" was vulnerable to gender bias. Putting aside the fact that Wal-Mart's policies (i.e. part of its corporate culture) flatly prohibited discrimination, the district court and Ninth Circuit relied upon Bielby's testimony in certifying the class, and they dismissed Wal-Mart's objections on the grounds that Daubert standards did not apply at the certification stage.
Prior to Dukes, a minority of courts had taken the same view - that Daubert did not apply at the certification stage, tacitly permitting unreliable expert testimony to be used to cobble a class together. These courts included the Second Circuit (at least initially), and possibly courts in the Sixth and Tenth Circuits as well. See, e.g., In re Visa Check/Mastermoney Antitrust Litig., 280 F.3d 124, 135 (2d Cir. 2001); Bacon v. Honda of Am., Mfg., Inc., 205 F.R.D. 466, 470-71 (S.D. Ohio 2001) (stating that a Daubert analysis was "not warranted" at the certification stage"), aff'd, 370 F.3d 565 (6th Cir. 2007); Shook v. Bd. of County Commr's, 386 F.3d 963, 968 (10th Cir. 2004). These decisions were generally grounded in avoiding consideration of the merits at the certification phase of the case, but obviously created the risk that a case would be certified based on arguments that would never be supported by any evidence.
A majority of courts, fortunately, accepted the common sense view that Daubert applies at the certification phase of the case. Shortly after its Visa Check decision cited above, the Second Circuit reversed itself and held that Daubert standards did apply. See In re IPO, 471 F.3d 24, 42 (2d Cir. 2006). The First, Third, Fifth, Seventh, and Eleventh Circuits have also held that Daubert applies at the certification stage, finding such review to be part of the court's obligation to engage in a rigorous analysis of the certification issues and its role as a gatekeeper of evidence. See In re Polymedica Sec. Litig., 432 F.3d 1, 5-6 (1st Cir. 2005); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d. 305, 316-20 (3d Cir. 2008); Regents of the Univ. of Cal. v. Credit Suisse First Boston (USA), 482 F.3d 372, 379-80 (5th Cir. 2007); Am. Honda Motor Co. v. Allen, 600 F.3d 813, 816 (7th Cir. 2010); Sher v. Raytheon Co., Case No. 09-15798 2011 WL 814379 (11th Cir. Mar. 9, 2011); see also Gariety v. Grant Thornton, LLP, 368 F.3d 356, 366 (4th Cir. 2004) (citing Seventh Circuit authority regarding the need to review merits favorably).
The Dukes decision reflects that Daubert standards do apply at the certification stage. First, the Supreme Court emphasized that a trial court not only could review the merits, but that in some cases such a review was a necessary part of determining whether there are sufficient class-wide issues. This holding erases the analytical underpinnings of the courts that had refused to apply Daubert at the certification stage. Because courts can and should review the merits as they touch upon class issues, a Daubert analysis must also be conducted with respect to expert testimony.
The Court did not explicitly overrule the Ninth Circuit's refusal to use Daubert, in part because it found that Bielby's testimony and that of various statistical experts was not probative on its face. With respect to the lower court's legal holding that "Daubert did not apply to expert testimony at the certification stage," the Court simply stated "[w]e doubt that is so." Taking this brief, dismissive comment along with its admonition that courts engage in a review of the merits, Dukes stands for the proposition that Daubert does apply at the certification phase. As a result, defendants should not be subjected to classes based on flawed expert testimony.
The Bottom Line: The Dukes decision should aid in enforcing Daubert standards for experts at the certification stage of a case.