Dr. Pepper gets an upset tummy: Court approves conjoint analysis/price premium model in ginger ale class action

Fitzhenry-Russell v. Dr. Pepper Snapple Group, Inc., 2018 WL 3126385, No. 17-cv-00564 (N.D. Cal. Jun. 26, 2018) (magistrate judge)

Canada Dry Ginger Ale allegedly deceived consumers with the phrase “Made From Real Ginger” when, in fact, Canada Dry does not contain the type of, or amount of, ginger consumers would expect (ginger root). Instead, Canada Dry contains a ginger derivative, ginger oleoresin. Plaintiffs alleged that Dr. Pepper was wrongfully able to charge a 4% price premium on Canada Dry as a result. The court certified a class and rejected challenges to plaintiffs’ expert declarations in support of class certification on the usual California claims.

Plaintiffs’ survey expert, Dr. Dennis, looked at consumer understanding of “Made from Real Ginger” and materiality/price premiums. Respondents were asked “what is your understanding of the statement ‘Made From Real Ginger’ on the Canada Dry Ginger Ale?” and provided options:

[1.] Ginger oil, which is extracted from the ginger root using steam

[2.] Ginger root, which is part of the ginger plant, not an extract

[3.] Ginger oleoresin, which is extracted from the ginger root using a solvent

[4.] None of these

78.5% of California Canada Dry consumers answered that the product was made from ginger root, while 4.8% of consumers picked ginger oleoresin and 8.6% picked ginger oil. (The truth is oleoresin.)

Before answering the materiality question, respondents were presented with the definitions of “ginger root” and “ginger oleoresin.” Ginger oleoresin was defined almost as above, but with the additional information that a solvent, such as ethanol, could be used to extract the ginger. In the survey, 92.4% of respondents preferred to purchase a version made with ginger root over one made with oleoresin.

For the price premium survey, Dennis used a choice-based conjoint survey, which asked respondents to express preferences by choosing from a set of product profiles (i.e., choosing a product from a group of products). The price premium survey was restricted to respondents who recently purchased ginger ale, and presented respondents with mixes of six attributes.  The results were fed into Bayesian models that allegedly allowed Dennis to calculate a price premium from “Made with Real Ginger” for the marginal consumer; he found a 4% price premium.

Plaintiffs’ expert Weir opined on whether it would be possible to determine damages on a class-wide basis using common evidence, and provided a framework for/estimate of damages to the California class. Weir multiplied the total California sales of Canada Dry in the class period by 4% to calculate damages, resulting in a figure of $10,778,477.

Dr. Pepper didn’t challenge the experts’ qualifications or the reliability of their methods.  Instead, it focused on the argument that Dennis didn’t properly apply the methodologies behind his consumer understanding survey and price premium analysis.

As to the consumer perception survey, “survey evidence should be admitted as long as it is conducted according to accepted principles and is relevant.” “[T]echnical inadequacies in a survey, including the format of the questions or the manner in which it was taken, bear on the weight of the evidence, not its admissibility.”  Dr. Pepper pointed to its rebuttal expert’s “vastly different results” in his replication survey of consumers. That expert recreated Dennis’s consumer understanding survey, but changed the descriptors of ginger oil, ginger oleoresin, and ginger root. But even where “simple language” was used to describe ginger oil, ginger oleoresin, and ginger root, 40.59% of respondents still believed “Made From Real Ginger” meant that Canada Dry was made using “ginger root” as its ginger ingredient. “This is still a legally significant percentage of people who would be misled.” 

The rebuttal expert also did a replication survey using “technical” descriptors of ginger oil, ginger oleoresin, and ginger root, but this different language proved the point that Dennis’s survey needed to go to a fact-finder, who could determine if plaintiffs’ survey was unduly biased. “This is because the replication survey using technical language seems to be designed to confuse respondents, and encourage them to answer that they ‘don’t know’ or are ‘unsure’ of what the ginger ingredient behind the ‘Made From Real Ginger’ claim is. After all, which layperson has ever heard of the ginger root powder they purchase in stores being referred to as ‘triturated ginger’ [defined in the survey as ‘a coarse powder obtained from the ginger root using a bleaching process’]?”  [I agree that the “not an extract” language in the original survey has some biasing potential; I also wonder about the “not sure/don’t know” option—but at the same time “triturated ginger” isn’t how I would think about ginger root powder either.]

The court found Dr. Pepper’s criticisms of the price premium survey and simulator to be more substantive.  Dr. Pepper argued that despite Dennis’s representation in his declaration that his survey only contained six features or attributes— brand, type, flavor, nutrition facts, description on front of the package, and price—he instead used 11 factors, because the description on the front of the Canada Dry packaging wasn’t one attribute but six: (1) “100% Natural Flavors,” (2) “The Original Ginger Soda,” (3) “Barrel-Aged,” (4) “Made with/from Real Ginger,” (5) “Caffeine Free,” and (6) “Since 1904/1873.” The more attributes in a conjoint survey, “the higher risk that it simply becomes too complex for respondents.” Also, it may be that the reason a person considered the descriptors on the product packaging was because he or she thought the “100% Natural Flavors” claim was important, rather than the “Made From Real Ginger” claim.  

Dennis’s reply declaration didn’t assuage the court’s concerns: He argued that grouping the different product descriptions under one attribute “likely led to a dilution of the respondents’ attention (in contrast to showing only one product description for each product option), and therefore reducing possible risk from focalism bias.” If he hadn’t included the four product descriptions on the Canada Dry can on the price premium survey, the survey would have been criticized on that basis. The court understood that, but wasn’t convinced. Still, even if the survey overestimated the value consumers placed on “Made From Real Ginger,” it wasn’t excludable excludable. Dr. Pepper could attack it at trial.

Dr. Pepper argued that one of its other ginger ale products, Schweppes, didn’t have the ginger claim, and it sold in stores at either the same price or for less. Plaintiffs responded that legal authorities agree that using a side-by-side comparison is “bunk.” The court agreed that as a matter of common sense, such comparisons don’t account for any of the range of other possible reasons for these products to be priced so similarly. Even the possibility that another survey might have used side-by-side comparisons didn’t make this survey excludable.

Most substantively, Dennis considered willingness to pay in the conjoint survey, not a price premium; thus, Dr. Pepper argued, it couldn’t calculate restitution, which is the difference between what consumers paid and the true market price, which also takes into account supply-side factors. However, the study used past market prices, which reflect supply factors. The study can get insight into price premium by looking at the marginal consumer: the one who is indifferent between buying and not buying the infringing product. That consumer’s WTP is “equivalent to the price premium associated with the infringing level of the attribute; this marginal consumer can be identified by offering respondents a ‘no buy’ option.”  The model asks “At what price in that actual market in which [defendant] sold the offending products could [defendant] have sold the equivalent number of products without the false claim(s)?” The marginal consumer’s WTP discloses that price, tethered to the real market because the conjoint survey used actual market-clearing prices as the basis for the prices in its survey and actual competitor products.  Thus, the price premium study satisfied Daubert.

After that, the class certification discussion was lengthy, but largely foreordained.

Dr. Pepper argued that “Made From Real Ginger” couldn’t be material to consumer decisions because there was no common understanding of the term, as courts have ruled for “All Natural” and “100% Natural.” The court found the challenged phrase to be far less vague; almost 80% of people thought “Made From Real Ginger” meant that Canada Dry was made using ginger root. Plus, Dr. Pepper’s internal documents showed that Dr. Pepper thought the “Made From Real Ginger” claim was material. E.g., Dr. Pepper sought to capitalize on the alleged health halo ginger products have to consumers by encouraging people to believe that “Canada Dry Ginger Ale is a [carbonated soda drink] that fits into your healthy lifestyle because it is made from real ginger.” When respondents were asked their reasons for drinking Canada Dry five years later, the top five reasons were: (1) “I trust and respect the Canada Dry Brand (28%)”, (2) “Drinking Canada Dry makes me feel better by soothing my stomach (26%)”, (3) “Canada Dry is easy to find in stores (26%)”, (4) “Canada Dry tastes good with food (25%)”, and (5) “Canada Dry is made with real ginger (25%),” even though previously lots of people hadn’t believed that ginger ale had real ginger. Thus, “through its marketing, it orchestrated a change in consumer perceptions.” Another document claimed that the “Made From Real Ginger” program “is working: - New news - Strong POD and message relevant to target consumer - Consumer awareness and brand equity increased - Purchase frequency and volume growth escalated to +8.5%.” “Clearly, if a quarter of Canada Dry consumers were listing the ginger claim as a top five reason why they bought the product, the claim is material. Dr. Pepper cannot walk back evidence contained in its own documents.”

The most interesting argument Dr. Pepper presented against the price premium survey under predominance is that “it does not match [plaintiffs’] theory of liability. The survey purports to calculate a price premium associated with misleading consumers to believe the drink contains powdered or chopped ginger, but it does not do so—it calculates the premium associated with all possible meanings of the claim.” Canada Dry does contain traces of real ginger, in the form of ginger oleoresin. But the complaint still properly alleged that these traces weren’t “real ginger” as a reasonable consumer would understand it and that the flavorings contained none of the health benefits of real ginger.  Plaintiff’s rebuttal expert reported that the concentration in parts per million of 6-gingerol and 6-shaogal, which are ginger-derived compounds, in Canada Dry was far below what a person would be able to detect when drinking the ginger ale. Even if it is literally true that Canada Dry has ginger in it, the ginger is not what a reasonable consumer would expect.

The worth of the “Made From Real Ginger” claim would only matter in the future if a jury does find that the claim is misleading. Thus, plaintiffs’ damages model fit the theory of the case.