A recent case out of the Northern District of California, Lanovz v. Twinings North America, Inc., highlights the evolving landscape that litigants are facing in class certification under Federal Rule of Civil Procedure Rule 23, including courts subjecting plaintiffs’ damages models to a more rigorous analysis.  Drawing upon the Supreme Court’s holding last year in Comcast v. Behrend, Judge Ronald M. Whyte’s decision in Twinings, along with Judge Pregerson’s decision in In re Pom Wonderful LLC, No. ML 10-02199 DDP (RZx), 2014 WL 1225184 (C.D. Cal. Mar. 25, 2014), provide important lessons for companies facing potential class certification in consumer fraud class actions.

Lanovz v. Twinings North America

In Lanovz v. Twinings North America, Inc., the plaintiff brought claims on behalf of a purported class of all individuals in California who purchased allegedly “misbranded” Twinings’ green, black, and white tea produced since May 23, 2008.  No. C-12-02646-RMW, 2014 WL 1652338, at *1 & n.3 (N.D. Cal. Apr. 24, 2014).  The plaintiff claimed that she paid a premium for Twinings’ tea and would not have purchased the tea but for Twinings’ unlawful labeling, which described Twinings’ tea as a “Natural Source of Antioxidants” or as a “natural source of protective antioxidants.”  Id. at *1.  The plaintiff alleged that the labels were illegal under California law and labeling regulations of the U.S. Food and Drug Administration (“FDA”), which California law had adopted. 

The plaintiff moved for class certification.  As Judge Whyte explained, to be certified as a class, the plaintiff had to meet the four conjunctive criteria of Federal Rule of Civil Procedure 23(a): (i) numerosity, (ii) commonality of issues, (iii) typicality of the representative plaintiffs’ claims, and (iv) adequacy of representation.  Fed. R. Civ. P. 23(a).  In addition, the plaintiff had to satisfy one of the three disjunctive requirements of Rule 23(b) for types of class actions.  Fed. R. Civ. P. 23. The plaintiff alleged that the class action could be maintained under either Rule 23(b)(2) or Rule 23(b)(3).  Rule 23(b)(2) allows for class treatment when the party “opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole.”  Id.  Rule 23(b)(3), in contrast, permits class treatment when a court finds that “questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.”  Id.

Judge Whyte held that the proposed class met Rule 23(a)’s requirements for maintaining a class action because the plaintiff had satisfied the criteria of commonality and typicality and the parties did not contest that the class was sufficiently numerous or that the plaintiff would be an adequate representative.  Judge Whyte certified the class under Rule 23(b)(2) for injunctive relief only, explaining that the plaintiff had requested an injunction (i) requiring Twinings to cease selling “misbranded” products, (ii) enjoining Twinings from continuing to market, distribute, or sell the products, and (iii) ordering Twinings to take corrective action.

Significantly, Judge Whyte denied the plaintiff’s request to certify the class under Rule 23(b)(3), holding that the plaintiff had not presented a sufficient damages model.  The plaintiff’s damages expert had three proposed models for measuring damages: (i) refunding the entire purchase price; (ii) a price premium model; or (iii) a regression analysis.  The court rejected each of these models.  Along with Judge Pregerson’s decision in Pom, Judge Whyte’s rejection of these proposed models provides useful guidance.

Damages Models in Comcast, Pom, Twinings, and Implications

Twinings and Pom relied in part on Comcast in denying the plaintiffs’ requests for a Rule 23(b)(3) class.  Both district court decisions represent some of the latest cases carrying out Comcast’s dictate to engage in a “rigorous analysis” in enforcing Rule 23(b)’s predominance requirement and reinforce that a plaintiff’s damages model in a class action must be tied to the defendant’s liability.


The Supreme Court’s Comcast decision appeared to raise the hurdle for plaintiffs seeking to certify a class.  In Comcast, the plaintiffs alleged four different theories of antitrust impact against a cable television provider.  The trial court found that three of the four theories were not suited to classwide resolution, but one was capable of class-wide proof.  The plaintiffs’ damages expert calculated an amount of damages but used a model that did not isolate damages resulting from any one of the four theories of antitrust impact.  After the district court and Third Circuit certified a plaintiff class, the Supreme Court disagreed, concluding that because the plaintiffs’ damages model assumed the validity of all four of the plaintiffs’ theories of antitrust impact, the model failed to tie damages to the lone remaining theory of liability.  As a result, the Court explained, the plaintiffs had not established that damages were capable of measurement on a classwide basis, and without such a showing, could not demonstrate that classwide issues of fact predominated over individual questions and satisfied Rule 23(b)(3).  The Court wrote that a plaintiff’s damages model “purporting to serve as evidence of damages” in a class action “must measure only those damages attributable to [the defendant’s conduct].  If the model does not even attempt to do that, it cannot possibly establish that damages susceptible of measurement across the entire class for purposes of Rule 23(b)(3). . . . [A]ny model supporting a ‘plaintiff’s damages case must be consistent with its liability case.’”  Comcast, 133 S.Ct. 1426, 1433 (2013) (citations omitted).

The impact of Comcast is evident in both Pom and Twinings and their rejection of plaintiffs’ proposed damages models.  In Pom, the plaintiffs alleged, on behalf of a class of consumers, that the defendant misleadingly advertised that certain Pom juice products provided various health benefits and that research demonstrated these benefits.  The defendants moved to decertify a damages class comprised of persons who purchased a Pom Wonderful 100% juice product between October 2005 and September 2010.  Judge Pregerson explained that Comcast held that although the plaintiff need not necessarily demonstrate through common evidence the “‘precise amount of damages incurred by each class member,’” the plaintiffs “‘must be able to show that their damages stemmed from the defendant’s actions that created the legal liability.’”  2014 WL 1225184, at *2 (citation omitted).

Similarly, Judge Whyte interpreted Comcast as emphasizing that damages “must be capable of determination by tracing the damages to the plaintiff’s theory of liability.”  Twinings, 2014 WL 1652338, at *5.  A tour of each of the rejected damages models highlights the analysis that courts may apply against plaintiffs seeking to certify a class.

Full Refund Model

Both the plaintiffs in Twinings and Pom presented a “Full Refund” model that used the full retail price paid as the measure of damages and assumed that consumers would not have purchased the defendant’s product if not for the alleged misrepresentations.  Both courts rejected the model for similar reasons.

As Judge Pregerson explained in Pom, the relevant California laws under which the plaintiffs sought relief authorized restitution for victorious litigants, but restitution “requires evidence of the actual value of what the plaintiff received,” and such a party “must generally return any benefit that it has received.”  2014 WL 1225184, at *3 (citations omitted).  The plaintiffs argued that the “Full Refund” model could be appropriate if a jury determined that consumers did not receive their desired benefit from the defendant’s juices.  Judge Pregerson rejected this argument and explained the inadequacy of such a model:

In determining the appropriate amount of restitution, however, the question is not whether a plaintiff received the particular benefit he sought or what the value of that benefit was or would have been.  Plaintiffs do not cite, nor is the court aware of, any authority for the proposition that a plaintiff seeking restitution may retain some unexpected boon, yet obtain the windfall of a full refund and profit from a restitutionary award.  Nor can Plaintiffs plausibly contend that they did not receive any value at all from Defendant’s products.  Because the Full Refund model makes no attempt to account for benefits conferred upon Plaintiffs, it cannot accurately measure classwide damages.

Id. at *3.

In Twinings, Judge Whyte encountered a similar model, which was based upon refunding the entire purchase or “register” price of the tea.  For the same reasons as Judge Pregerson, Judge Whyte rejected the model because it was not the proper measure of damages.

Both Twinings and Pom imply that damages models seeking a full refund will be rejected as not accurately measuring classwide damages where the available monetary remedy is restitution.

Price Premium Model

Both the Twinings and Pom plaintiffs presented second models, price premium models, which were also rejected.  A “Price Premium” model quantifies damages by comparing the price of the defendant’s product with other similar products of the same size and assumes that, absent the alleged misrepresentations, demand for the defendant’s product would have been less and the market price would have been lower.

Judge Pregerson found the plaintiffs’ price premium model untenable.  The plaintiffs were unable to show that their damages stemmed from the defendants’ actions that created the legal liability, or, in other words, that the alleged misrepresentation caused plaintiffs to pay a “price premium” more than they otherwise would have paid in absence of the misrepresentations.  Instead of providing a survey or other evidence of consumer behavior, the plaintiffs’ expert merely “observed” that Pom juices were more expensive than certain other juices and assumed that not a single customer would have chosen Pom juice over some other juice if not for the allegedly deceptive advertising.  Such a “model” did not “comport with Comcast’s requirement that class-wide damages be tied to a legal theory” nor was there anything of substance to allow the court to conduct the required “rigorous analysis.”  Pom, 2014 WL 1225184, at *5.

Judge Whyte similarly held in Twinings that the model did not align with Comcast’s requirement “that class-wide damages be tied to a legal theory.”  He explained that the model was legally insufficient because the plaintiff’s expert had “no way of linking the price” difference to the antioxidant label “or controlling for other reasons why ‘comparable’ products may have different prices.”  2014 WL 1652338, at *6.

Twinings and Pom reveal that a “price premium” model should not pass muster under Rule 23(b)(3) where the model lacks evidence of consumer behavior and therefore impedes a court from engaging in a rigorous analysis.

Regression Analysis Model

The plaintiff’s expert in Twinings presented an additional model, an “econometric or regression analysis,” which the court also rejected.  Such a model would estimate the “portion of sales gleaned by Twinings as a result of the false and misleading use of the term ‘Natural Source of Antioxidants’ on its tea products.”  Twinings, 2014 WL 1652338, at *6.  Judge Whyte noted that the method “might arrive at a legally relevant damages analysis” because it can be translated “into the percentage of sales attributed specifically to the claims” asserted.  Id.  The plaintiff’s expert, however, abandoned a regression analysis as a method of calculating damages in the case because the antioxidant claims had been on labels over the entire class period.  It would be impossible therefore to invoke a regression analysis approach because of the “lack of any variable in sales or units sold attributed to the antioxidant claims.”  Id.  Because the plaintiff could not present any damages model capable of estimating the price premium attributable to Twinings’ antioxidant labels, the plaintiff failed to satisfy the requirements for class certification under Comcast.

Twinings indicates that a regression analysis damages model based upon a defendant’s sales may not withstand Rule 23(b)(3) where the defendant has sold products with the alleged mislabeling throughout the entire class period.

Final Lessons

Recent district court decisions out of California reflect that the holding in Comcast raises the hurdle for certifying a class in consumer fraud class actions.  Plaintiffs’ damages theories should be subjected to a “rigorous analysis” under Rule 23, and various types of damages models may falter under the predominance requirement of Rule 23(b)(3).

For additional information, please contact Michael Klebanov or Joe Orlet.