On May 29, 2014, the California Supreme Court issued its long-awaited opinion in Duran v. U.S. Bank National Association, remanding the case to the trial court due to the trial court’s flawed trial methodology.
In Duran, the plaintiff asserted misclassification claims on behalf of a putative class of Business Banking Officers of U.S. Bank, a position that involved sales of bank products such as loans and lines of credit to small business customers. These BBOs had been classified as exempt based on the outside salesperson exemption, which requires that the employee in question spend more than 50 percent of the workday engaged in outside sales.
Despite the evidence submitted by US Bank, the trial court certified the class based on evidence that (1) the BBO position was “standardized,” (2) all BBOs had been classified as exempt by the employer without a close examination of each employee’s duties or work habits, and (3) US Bank did not train or monitor BBOs to ensure that exemption requirements were satisfied.
The case then proceeded to the parties’ presentation of trial management plans, where US Bank proposed dividing the class into 20 to 30 groups and having special masters conduct individualized evidentiary hearings on liability and damages. Plaintiffs instead proposed the use of surveys and random sampling to establish liability, followed by determination of aggregate classwide damages in a second phase of the trial. Faced with two competing trial plans, the trial court developed its own plan: a court-selected random sample of 20 class members — the Representative Witness Group (“RWG”) — who would testify at trial and from whom liability and damages findings would be extrapolated to the rest of the class.
After a 40-day bench trial in which the court refused to consider any evidence, including any of the 75 declarations or the deposition excerpts, from (properly classified) members of the class who were not part of the RWG, the trial court found that US Bank had not carried its burden of proof that the BBOs were properly classified and thus that the entire class of BBOs was misclassified as exempt and that all class members were entitled to overtime compensation. The court calculated the total overtime restitution due the class at $8,953,832, which with prejudgment interest as of May 15, 2009, came to $14,959,565.
On appeal to California’s First Appellate District Court, the appellate court unanimously reversed the decision and ordered the class decertified, holding that (1) the trial plan denied USB’s due process right to litigate affirmative defenses, (2) the high margin of error further implicated due process concerns, and (3) it was an abuse of discretion not to decertify the class when it became so apparent that individual issues predominated.
While the California Supreme Court’s decision does not reverse the certification order outright, as did the Court of Appeals decision, and instead remands the case to the trial court, the Order nevertheless provides much needed guidance on the standards of class certification and the need for a proper class action trial management plan.
First, the Court held that trial of a class action must permit litigation of relevant affirmative defenses. If those affirmative defenses turn on individual questions, that may well mean that a class action is not an appropriate vehicle to try the case.
Second, the Court explicitly required that courts, at the class certification stage, consider a plan for how individualized issues will be managed at trial. “Trial courts must pay careful attention to manageability when deciding whether to certify a class action. In considering whether a class action is a superior device for resolving a controversy, the manageability of individual issues is just as important as the existence of common questions uniting the proposed class.” “Trial courts also have the obligation to decertify a class action if individual issues prove unmanageable.”
The Court then reviewed the statistical methods used by the trial court. Although the Court did not hold that statistical proof could never be used to certify and then try a misclassification case, the requirement that such a plan “must be conducted with sufficient rigor” will likely give all but the most obstinate jurists considerable pause. Trial courts “would be well advised” to insist upon such a plan before granting certification “[r]ather than accepting assurances that a statistical plan will eventually be developed,” a phrase that likely sounds quite familiar to those who defend against wage-and-hour class claims, and “[i]n any event, decertification must be ordered whenever a trial plan proves unworkable.”
Here, the trial plan gave new meaning to the term “unworkable.” Without any preliminary assessment of the variability in the class, “the court forced the case through trial with a flawed statistical plan that did not manage but instead ignored individual issues.” After holding that due process requires that a defendant be allowed to impeach a statistical model or otherwise show reduction of liability based on affirmative defenses, the Court turned to the more general question of the use of statistical evidence to prove classwide liability and damages. In a section entitled “Sampling Techniques Must Satisfy Realistic Guidelines to Minimize the Risk of Error,” the Court addressed the “numerous problems” that arose when proper methodology “based on inferential statistics and probability theory” were not employed. First, the sample size was too small, in large part because the court chose a sample size of 20 without input from statistical experts who could have explained the necessity to first determine variability within the class so that the sample size will be statistically appropriate and capable of producing results within a reasonable margin of error. This error was exacerbated by a non-random sample: although the draw of 20 was random, selection bias infected the RWG when the court included the named plaintiffs and replaced other class members who had opted out of the class with alternates. The trial court also excluded two other members of the RWG: one who was an “outlier” that the court did not consider a “true” BBO, and the other who did not respond to the trial subpoena. Thus the court heard testimony from 19 class members who did not opt out, were not “outliers but were rather “true” BBOs, and who appeared to testify at trial, as well as the two named plaintiffs. It refused to consider potentially contrary testimony of 10 plaintiffs that would otherwise have been included: the four previously named plaintiffs (who testified that they spent 60 to 90 percent of their workday on outside sales), the four original RWG members who opted out (two of whom admitted that they opted out at the urging of class counsel so they could be replaced by members with stronger testimony), the “outlier,” and the RWG member who failed to appear for trial. This resulted in a plaintiff-biased “Representative” Witness Group, and led to a margin of error of 43.3 percent. As the Court noted, even the plaintiffs had to concede that this was “intolerably high.”
While plaintiffs and defendants will have to wait and see what impact the Duran decision has on the class certification standards and class action trial methodology, the Duran decision should be viewed as a monumental win for defendants, particularly as to the Court’s affirmation of defendants’ rights to present individualized defenses and directives to the trial courts to assess how those and other individualized issues will be managed at trial at the class certification stage.