Statistics 101: Class Action Plaintiffs in California Must Have a Plan

In recent years, putative class counsel have begun to seek refuge in state court to avoid the rigors of complying with Federal Rule 23. This trend has accelerated since the US Supreme Court clarified the requirements for class certification in Wal-Mart and Comcast. But the same considerations underlying those decisions are present in state court, where the law is also developing to require more of class plaintiffs. The latest example is the California Supreme Court's ruling in Duran v. U.S. Bank National Ass'n, a wage and hour class action that applies to class actions more generally.

In Duran, plaintiffs claimed that US Bank National Association (USB) misclassified certain employees to avoid paying overtime. The trial court certified a class of 260 plaintiffs, and then, after class certification, devised a plan to determine USB's liability as to all class members by extrapolating from a sample of 20 plaintiffs. At trial, the court did not allow USB to introduce evidence regarding any class members other than the 20 in the sample. Based on testimony from the 20-plaintiff sample and expert statisticians, the court decided liability for the class, arrived at an average per-person recovery for lost overtime wages, and extrapolated it across the entire 260-person class. The California Supreme Court reversed, holding that the trial court's implementation of sampling was flawed in several ways, and set out guidelines for establishing effective trial plans at the class certification stage and for the use of statistics in dealing with individualized issues at trial.

Echoing federal court decisions, the Duran court explained that, at the class certification stage, trial courts must look at the feasibility of trying the case as a class action, and not just at whether common questions exist. Class certification is appropriate only if individual questions can be managed with an appropriate trial plan. In deciding whether to certify a class, it is just as important to assess the manageability of individual issues through the trial plan as it is to identify common questions. Accordingly, a plaintiff must set forth a realistic trial plan that includes manageable ways of dealing with individual issues. If this cannot be done, class certification may be denied.

The California Supreme Court further explained that if statistics will be part of a plaintiff's proof, the statistical method must be considered at the class-certification stage in the process of developing a trial plan. Trial courts need to vet the proposed statistical plan and decide whether the plan can effectively manage individual issues. Trial courts must keep in mind the elements of the defense that rely on individualized issues, and the statistical methods must accommodate case-specific deviations. The Court did not go so far as Wal-Mart in wholly disapproving of "Trial by Formula" but was clear that if extrapolation from sampling is to be attempted, it must be reliable.

The Duran court went on to hold that, at trial, even if a statistical model or plan is in place, parties must be able to raise individual issues outside of the statistical model. Specifically, a defendant's due process rights include the right to present relevant affirmative defenses, even when those defenses turn on individual questions.

Duran did not ultimately reach the issue of whether sampling is an appropriate tool for proving liability in a class action, but did acknowledge that "[s]tatistical methods cannot entirely substitute for common proof…there must be some glue that binds class members together apart from statistical evidence."

Putative class plaintiffs seeking a favorable state-court forum often select California because it is the most populous state in the union and is seen by many to be "plaintiff friendly." Duran, however, places a significant hurdle in front of plaintiffs hoping to breeze through class certification and pressure defendants to settle.  Following Duran, putative class plaintiffs in California will need a robust trial plan that can accommodate the complexities of the individual issues within the class. And, more importantly, they will need that at the class certification stage if they hope to proceed.

© Arnold & Porter Kaye Scholer LLP 2014 All Rights Reserved. This blog post is intended to be a general summary of the law and does not constitute legal advice. You should consult with counsel to determine applicable legal requirements in a specific fact situation.