Courts Must Referee Expert Battles In Deciding Class Certification Question
Mark’s Analysis of Class Certification Issues
The Minnesota Court of Appeals recently addressed the question of whether a court could weigh competing evidence at the class certification stage. In deciding that issue, the Court also clarified the standard of proof necessary to certify a class. Whitaker v. 3M, 764 N.W.2d 631 (MN Ct. App. 2009). In the past, many courts had been reluctant to enter the fray of competing expert opinions when deciding whether to certify a class, particularly if those opinions also went to the merits of the lawsuit. Many courts had also held class proponents to a relatively low standard when assessing whether they had met the class certification requirements set forth in the rules of civil procedure.
Finding that neither the Minnesota Supreme Court nor the Court of Appeals had yet addressed these issues, the Court of Appeals joined a growing number of state and federal courts which have held that the certification requirements of Rule 23 of the Minnesota Rules of Civil Procedure must be established by a preponderance of the evidence. Id. at 638. The Court went on to hold that that standard requires a district court to “resolve factual disputes relevant to rule 23 certification requirements, including expert disputes.” Id.
In Whitaker, the Court of Appeals reversed a trial court order certifying a class of employees claiming age discrimination. Both plaintiffs and defendant below had submitted expert evidence of the results of statistical analyses performed by their respective labor economists. Plaintiffs’ experts argued that there were significant differences in “labor market outcomes” for older workers. For example, they claimed that older workers did not receive their fair share of promotions and compensation increases. Defendant’s experts conducted their own analyses but took into account other relevant factors, such as performance or tenure, that could account for Plaintiffs’ findings. Defendant’s experts concluded that there were no significant differences in these labor market outcomes that were attributable to the employees’ ages.
They also took Plaintiffs’ experts to task for failing to establish any baseline against which to measure the relative significance of any differences they claimed to find, especially in light of generally accepted labor economics theories demonstrating that, even in the absence of discrimination, differences in such employment outcomes are to be expected. In fact, many courts have recognized that age, which changes over time, is not amenable to analyses such as those done by Plaintiffs, which look only at specific points in time, as might be done in a race or gender case. This unique aspect of age discrimination cases often renders those typical race or gender analyses unreliable. Thus, Defendant’s experts argued, Plaintiffs’ analyses provided no support for class certification, much less any claim of discrimination.
Under the preponderance of the evidence standard, a fact finder must decide that the evidence on one side of a question outweighs the evidence on the other. Using that standard, a court may not certify a class based solely on “some” evidence, or a “suggestion” drawn from the evidence, or evidence that is merely not “fatally flawed.” Rather, the preponderance standard adopted from the federal case law requires a district court to “resolve the differences among the experts to the extent that they are related to, and for the limited purpose of, determining whether class certification requirements were met.” Id. at 639.
Under the standard adopted in Whitaker, when deciding a motion for class certification, a court must find that the requirements of Rule 23 have all been met by a preponderance of the evidence. Most importantly, Rule 23 requires: (1) that there be questions of law or fact common to all members of the class; and (2) that the named plaintiffs’ claims or defenses are typical of those of the class members. This means that a district court must examine alleged defects in the plaintiffs’ evidence, including evidence introduced through an expert. In the Whitaker case, for example, the court would have to determine the validity of and weight to be assigned to 3M’s concerns about: (1) Plaintiffs’ failure to establish a baseline of expected, non-discriminatory differences; (2) the problems 3M identified with doing specific point-in-time analyses; and (3) the alleged inappropriate use by Plaintiffs’ expert of certain statistical controls in her analyses in light of predicted non-discriminatory correlations between age and employment outcomes. Id. at 639.
Going forward, in a typical employment class action, which will rely heavily upon expert statistical evidence, resolution of conflicting expert opinions often will be an integral part of the class certification decision process. A district court will need to address and resolve even those questions which relate to the merits, and referee the battle of the experts – to the extent necessary to determine whether the evidence is sufficient to demonstrate that Rule 23’s requirements have been proved by a preponderance of the evidence. Class proponents will need to ensure that their evidence meets that standard. Given the importance of certification in the life of a class action, it is a burden reasonably imposed.