“There are three kinds of lies: lies, damned lies and statistics.” The California Supreme Court could have been channeling Mark Twain when it rejected, emphatically, the unbridled use of statistical sampling to prove liability in a class action wage/hour case. In a unanimous decision, California’s high court in Duran v. U.S. Bank National Association, No. S200923 (May 29, 2014) gave the heave-ho to the kind of “trial by formula” that has become a feature of modern-day wage/hour litigation. At the same time, the court restored some sanity to class action litigation generally.


This class action was filed against U.S. Bank on behalf of 260 business banking officers (BBOs) who claimed they were denied overtime pay and meal/rest breaks. Liability turned on whether the bank misclassified the BBOs as exempt under the “outside salesperson” exemption, which applies to someone who spends more than 50% of the time on sales activities outside the branch.

Round one: liability. How would plaintiffs prove that all 260 were misclassified? The trial court turned to statistics.  It took testimony from 21 “sampled” class members who said they spent less than 50% of their time outside the branch, and concluded (by extrapolation, based on testimony from plaintiffs’ statistical expert) that all 260 BBOs had been misclassified. There was just one small problem: 75 class members—read, 28% of the class—filed sworn declarations saying they had not been misclassified. The trial court rejected that evidence and found the bank liable.

Round two: damages. At the damages phase, the trial court excluded the 75 declarations. Instead, the court took testimony about the average number of hours worked by the “21” employees, reckoned that the same was true for all, and awarded the class $15 million—including $6 million awarded to the 75 BBOs who admitted under oath they had no claim.

Judgment reversed. In 2012, the Court of Appeal reversed the judgment, concluding that the trial court’s flawed trial plan amounted to an improper “trial by formula” which deprived the employer of its due process rights because the employer could not raise individual challenges to absent class members’ claims. The Court of Appeal also ordered the class decertified.


The Supreme Court opened with a stout denunciation of the trial court’s plan, calling it “profoundly flawed.” The Court affirmed the appellate court’s reversal, and in doing so it imposed new substantive and procedural hurdles. Those hurdles have re-ordered how class actions will proceed in all California class actions, not just employment class actions.

Substance—Statistical Sampling Generally

A due process right to prove defenses.  Regardless of the kind of class action case (employment or otherwise) and regardless whether statistical sampling is used, a defendant has a due process right to litigate its individual affirmative defenses. This is fundamental. (Duran, slip opn. at 38; see also id. at 29 (“any trial must allow for the litigation of affirmative defenses...”).) “These principles derive from both class action rules and principles of due process.” (Id., at 31.)

Statistical sampling to prove liability. As for statistical sampling to prove liability, it is not banned altogether. However, it has to be carefully controlled. Said the Duran court:

We need not reach a sweeping conclusion as to whether or when sampling should be available as a tool for proving liability in a class action. It suffices to note that any class action trial plan, including those involving statistical methods of proof, must allow the defendant to litigate its affirmative defenses. If a defense depends upon questions individual to  each class member, the statistical model must be designed to accommodate these case-specific deviations. If statistical methods are ultimately incompatible with the nature of the plaintiffs’ claims or the defendant’s defenses, resort to statistical proof may not be appropriate. Procedural innovation must conform to the substantive rights of the parties. (Id., slip opn, at 38.)

Statistics can’t be the only “glue.” Moreover, evidence of liability common to each class member other than through statistics must come first. It cannot be created through, or replaced by, statistics: “Statistical methods cannot entirely substitute for common proof, however. There must be some glue that binds class members together apart from statistical evidence.” (Id., at 26.) As the Duran court said: “Class actions do not create a requirement of common evidence. Instead, class litigation may be appropriate if the circumstances of a particular case demonstrate that there is common evidence.” (Id. at 33.)

Procedure—A Manageable Trial Plan at the Class Certification Stage

A plaintiff who seeks to use statistical evidence to prove liability must meet a set of new criteria. It the trial plan will include statistical evidence, the court should consider at the certification stage whether a trial plan has been developed to address its use.

For one thing, a plaintiff must now prove manageability. This was optional pre-Duran.

Also, the bar has been raised. The trial court must conduct a preliminary assessment “to determine the level of variability in the class. If the variability is too great, individual issues are more likely to swamp common ones and render the class action unmanageable.” (Id. at 28.)

What did the Court mean by variability? “[A] defense in which liability itself is predicated on factual questions specific to individual claimants poses a much greater challenge to manageability” than individual questions regarding the calculation of damages. (Id. at 25.) “[T]he trial court could not abridge [the bank’s] presentation of an exemption defense simply because that defense was cumbersome to litigate in a class action.” (Id. at 31.) And “[i]f a defense depends upon questions individual to each class member, the statistical model must be designed to accommodate these case-specific deviations.” (Id. at 38)


The Duran court went further. It not only called into question a tool commonly used by the class action bar to bring wage/hour cases, it cast doubt on the future of misclassification cases themselves—even those that do not rely on sampling! Said the court: “[A]n employer’s liability for misclassification under most Labor Code exemptions will depend on employees’ individual circumstances.” (Id., at 33 (italics added).) By that, the Duran court seems to be saying that misclassification cases invariably raise inherently individual issues, no matter which Labor Code exemption is used.


The Duran court took a stroll through a neighborhood of class action law that has been roiling with controversy: The “overbroad class.”

Class action plaintiffs always contend that everyone in the class was harmed, and will calculate classwide damages by multiplying the number of all class members times the average damage per class member. This leads to the “overbroad class,” where persons are included in the class simply by virtue of the broad definition (e.g., all persons who bought a certain product, worked at an employer during a certain time period, or clicked on a webpage, etc.). In such cases, the damage figure is wildly inflated by the inclusion of persons who have no claim.

“No matter,” says the class action bar, “this can be sorted out in the claims process,” or in “Phase II, the damages phase, after liability is decided.” “By the way,” class counsel will say, “the damages phase is also where defendant can assert its now-meaningless due process right to bring individual defenses.” In other words, plaintiff gets to assemble his damages case by multiplication. But defendant has to dis-assemble that figure by subtraction. That is an impossible task, and effectively creates a presumption of classwide harm.

To date, no California appellate court has directly addressed this nonsensical argument. Until now. The Duran court said: “Only in an extraordinary situation would a class action be justified where, subsequent to the class judgment, the members would be required to individually prove not only damages but also liability.” (Id., at 25.)


Duran will affect every California class action, both employment and otherwise, whether on file now or still just a glimmer in class counsel’s eye. And if the claim is brought under California’s Labor Code or unfair competition law, Duran could affect class actions pending in federal court. Businesses facing class action exposure should review their case list and ask, “How has Duran changed this case?”