Justice Carol Corrigan, California Supreme Court
Justice Carol Corrigan, California Supreme Court (Jason Doiy / The Recorder)

SAN FRANCISCO — The California Supreme Court tossed a $15 million overtime award to bank employees Thursday because of a faulty statistical sample, but justices were careful not to toss at-large reliance on statistics in wage-and-hour class actions.

Business interests had hoped the court would use Duran v. U.S. Bank National Association to limit plaintiffs attorneys’ use of statistical samples to generalize damages and liability over an entire class. Plaintiffs employment attorneys worried such limits could destroy their cases.

Writing for a unanimous court, Justice Carol Corrigan cautiously upheld reliance on statistical evidence and provided some guidelines for its proper use. She explicitly shied away from making a “sweeping conclusion” that would remove sampling and statistics as tools in class actions.

“Statistical sampling may provide an appropriate means of proving liability and damages in some wage and hour class actions,” Corrigan wrote. But “after a class has been certified, the court’s obligation to manage individual issues does not disappear” and each plaintiff must still give some proof of his or her claim. Justice Goodwin Liu wrote a concurring opinion.

Employment lawyers on both sides seemed ready to claim the ruling as a victory.

Kimberly Kralowec, who filed an amicus brief in the case on behalf of the California Employment Lawyers Association, said the court shot down U.S. Bank’s claim that it should have a right to call every class member individually to the stand. “Essentially they were arguing you never get to have a class action,” she said.

Jones Day partner George Howard, an employment defense lawyer, said the opinion would make it more challenging to certify a class. The court called on judges to consider whether a class is manageable and can withstand trial, not just whether all class members share a common issue. It also invited trial judges to decertify classes if individual issues prove unmanageable.

“I just think this is going to inject some much-needed rigor into this whole process,” said Howard, who is not involved in the case. “It’s no longer ‘see if I can find a common issue and hope that I get my case settled.’”

The 2001 suit alleged U.S. Bank had misclassified a group of business banking officers as outside sales staff exempt from overtime pay. A central issue before the lower court was whether the employees universally met the definition of outside sales staff, which requires spending more than half of the workday outside the office.

Both the majority and concurring opinions condemned the particular statistical sample employed in Duran to resolve that question. Judge Robert Freedman of Alameda County Superior Court used a sample of 20 workers to determine that the entire 260-member class had been improperly denied overtime and was entitled to $57,000 per person.

Freedman developed his own sampling plan without expert advice on minimizing the margin of error and “improperly extrapolated liability findings from a small, skewed sample group to the entire class,” Corrigan wrote.

The sample was not random, she concluded, as it included the two named plaintiffs handpicked by lawyers and allowed some class members originally selected for the sample to opt out.

Furthermore, Freedman refused to consider declarations submitted by U.S. Bank from 75 class members who said they spent more than half their time working outside the office.

“This evidence suggested that work habits among [business banking officers] were not uniform and that nearly one-third of the class may have been properly classified as exempt and lacking any valid claim against [U.S. Bank],” Corrigan wrote.

Edward Wynne of the Wynne Law Firm in Greenbrae argued the case on behalf of the class in March; Timothy Freudenberger of Carothers DiSante & Freudenberger argued for U.S. Bank.

Freudenberger did not respond to a phone message or email requesting comment.

Wynne called Thursday’s opinion more favorable than the original court of appeal decision, which had dismissed the case permanently. Corrigan’s opinion instead remands the case for fresh class certification proceedings and a new trial.

“We’re really pleased,” Wynne said. “While it’s regrettable that the judgment wasn’t upheld and the case was decertified, the court has given us valuable guidance for getting the case recertified.” Wynne hasn’t ruled out using a statistical model in round two of litigation.

“We’re going to get the case certified again,” he said. “I see no reason not to use statistical modeling again.”

Contact the reporter at mkendall@alm.com.