Food for Thought: Werdebaugh v. Blue Diamond Growers

In Werdebaugh, plaintiff brought a class action against Blue Diamond, alleging that its package labeling was unlawful, deceptive, and misbranded in violation of California law. Specifically, plaintiff alleged that the packaging of several Blue Diamond products listed evaporated cane juice as the sweetener used when the ingredient was actually sugar. Also, the packaging included the statement “All Natural” when, in fact, the products contained synthetic ingredients.

The court initially certified a damages class. In doing so, it accepted the regression model proposed by plaintiff’s expert as an appropriate damages model under Comcast Corp. v. Behrend, 133 S.Ct. 1426 (2013). Thereafter, plaintiff submitted two different regression analyses conducted by his expert, Dr. Capps. Plaintiff also submitted an alternative damages figure based on a separate study. Defendants then moved to decertify the class, arguing that the damages models did not satisfy Comcast. The court agreed.

A regression model purporting to serve as evidence of damages in a class action must measure only those damages attributable to the defendant’s conduct. Plaintiff was thus required to present a damages methodology that could determine the price premium attributable to Blue Diamond’s use of the “evaporated cane juice” and “All Natural” label statements. Preliminarily, the court approved Dr. Capps’ proposed “before-and-after” regression analysis, which would have compared the differences in sales of Blue Diamond’s products before-and-after it began using the labeling language at issue. However, Dr. Capps later concluded that it would be impossible to conduct a before-and-after analysis. Instead, his first and second damages models relied on a hedonic regression analysis, which considers the price impact associated with various product attributes—including product labels. In addition to the hedonic price regression analysis, Dr. Capps presented an alternative damages figure based on a 2007 study conducted by Professor Jeffrey Anstine. The Anstine study found a price premium of roughly 40 percent for yogurt labeled “All Natural.”

Defendant argued, in part, that the regression models used by Dr. Capps were flawed because (1) the model conflated the effect of the alleged mislabeling with the value of Blue Diamond’s brand; (2) the model failed to control for other key factors impacting price; and (3) Dr. Capps’ reliance on the Anstine study failed Comcast.

The court found that the first damages model submitted by Dr. Capps improperly conflated the alleged misleading label with the value of Blue Diamond’s brand. The report assumed that none of Blue Diamond’s competitors used the challenged labeling claims. Dr. Capps corrected for problems caused by his assumption by collapsing the Blue Diamond “brand” and “label” into a single variable. As a result, the court concluded that the first model could not isolate the harm attributable only to the labeling claim, but instead reflected the joint effect of both the labeling claims and the value of Blue Diamond’s brand. Accordingly, the court found that Dr. Capps’ first model failed to establish that damages could be determined and attributed to plaintiff’s theory of liability on a class-wide basis.

Furthermore, when the court granted plaintiff’s motion for class certification, Dr. Capps represented that his regression model would control for variables such as advertising, seasonality, income, and regional price differences. But when the damages models were submitted and Blue Diamond moved to decertify, the court found that plaintiff failed to show how Dr. Capps’ damages models controlled for advertising. Dr. Capps’ first model asserted that it controlled for advertising and promotions “via the quarterly and year dummy variables.” But Dr. Capps failed to explain how these variables control for the impact of advertising on the price premium. In a separate declaration, Dr. Capps argued that hedonic regression analysis did not need to control for advertising because he was trying to measure the price premium, and not how advertising impacted the price premium. The court, however, stated, “it is precisely because advertising may impact the price premium that Dr. Capps’ damages model must control for advertising.” Thus, the court found that both the first and second damages models failed to control for advertising.

The court also considered the alternative damages model based on the Anstine study. It rejected Dr. Capps’ attempts to rely on the study and explained that nothing in the study purports to measure the price premium attributable to Blue Diamond’s use of “evaporated cane juice” and “All Natural” labels on the products at issue. The court ultimately found that there was no connection between the Anstine study and Blue Diamond’s liability.

Finally, the court addressed plaintiff’s argument that the issues with Dr. Capps’ damages model could not defeat class certification. Plaintiff argued that the only issue at the class certification stage is the soundness of the methodology used by the expert. The court disagreed, stating, “[t]o the contrary, the court is obligated to do more than rubberstamp a proposed damages class merely because a plaintiff’s expert purports to have used a peer reviewed methodology such as a regression analysis.” For further support, the court pointed to Comcast. There, the Supreme Court noted that the mismatch between the damages model and the plaintiff’s liability case made class certification inappropriate under Rule 23(b)(3).

Plaintiff also relied on two Ninth Circuit opinions, Leyva v. Medline Inst. Inc., 716 F.3d 510 (9th Cir. 2013) and Jimenez v. Allstate Ins. Co. 765 F.3d 1161 (9th Cir. 2014), to argue that individual questions as to damages could not independently defeat class certification. However, the court explained that Leyva and Jimenez, when read in conjunction with Comcast, set forth the principle that “so long as the damages can be determined and attributed to a plaintiff’s theory of liability, damage calculations for individual class members do not defeat certification.” The court did not decertify the damages class because of the need for individualized damages calculations. Rather, it decertified the class because plaintiff failed to put forth evidence that damages could be determined and attributed to plaintiff’s theory of liability on a class-wide basis. As such, the court concluded that the predominance requirement was not satisfied.

Read more significant court decisions affecting the food industry in Food for Thought: 2014 Litigation Annual Review.