Asking the wrong questions: Consumer surveys may not save deficient labeling cases

News

Asking the wrong questions: Consumer surveys may not save deficient labeling cases

02 October 2022

Food labeling class actions brought on behalf of allegedly deceived consumers have become increasingly commonplace, especially in hotbed jurisdictions like California, New York, and Illinois. Labeling cases are, not unlike portions of the food industry, a volume business. Relatively few of these cases progress beyond a defendant's motion to dismiss, and even fewer still survive summary judgment, but that failure rate is baked into the model for plaintiffs' counsel (the primary drivers of such lawsuits), for enough cases settle along the way to make the overall enterprise work. In the relatively small slice of cases that are litigated, a key hurdle for plaintiffs is to show that their interpretation of the allegedly misleading label is not idiosyncratic, but rather would be shared by the so-called "reasonable consumer."

One tool that plaintiffs at times employ to address this issue is a consumer perception survey, in which an expert will design a survey to canvass consumers about their understanding of what a product's packaging means. Plaintiffs have increasingly begun to cite to such surveys in complaints as a way to bolster their theories about how reasonable consumers interpret the challenged labeling claims.

Yet despite the increasing prevalence of consumer surveys, courts in food labeling class actions have made it clear that not just any survey will suffice. Two recent decisions in two food labeling hotspots – the Northern District of California and the Southern District of New York – demonstrate that plaintiffs who ground the plausibility of their claims on a consumer perception survey must be prepared for the court to scrutinize the survey's methodology and overall persuasive value, and potentially to reject it outright.

Kennard: Consumer surveys must probe understanding of the specific challenged label claim

In Kennard v. Kellogg Sales Co. (Northern District of California)¹, the plaintiff alleged that a variety of products sold under Kellogg's MorningStar Farms brand were misleadingly labeled with the term "veggie," because the products weren't primarily composed of vegetables. In her amended complaint, the plaintiff referred to consumer surveys commissioned for this case to support her proffered understanding of "veggie." The surveys, according to the plaintiff, demonstrated that California consumers seeking "meat-alternative" products would understand the "veggie" labeling to convey that the products were "primarily made of vegetables rather than other non-vegetable plant-based ingredients."

In the surveys, Californians who indicated that they "had purchased (or seriously considered purchasing) a meat-substitute product in the past 12 months" were given a questionnaire regarding either a "Veggie Burger" or a "Veggitizer." The initial screen identified two categories of meat-substitute ingredients: (1) "Vegetable-based," which "would include ingredients made of actual vegetables such as carrots, cauliflower, or potatoes," and (2) "Other Plant-based," which "could include ingredients made of other non-vegetables such as grains or oils." After being directed to look at examples of such packaging, respondents were asked which would best describe the types of ingredients they would expect the product to contain. The available answers included: "Entirely vegetable-based ingredients," "Primarily vegetable-based ingredients," "Primarily other plant-based ingredients," "Entirely other plant-based ingredients," and "I do not have an opinion." According to the plaintiff, "of the over 100 respondents to each questionnaire, over 80 percent were misled to believe the products are primarily or entirely made of vegetables."

Kellogg moved to dismiss the amended complaint, arguing that "veggie," as used in its products' labeling, would be reasonably understood by consumers to refer to vegetarian foods and meat substitutes generally, as opposed to food containing any particular quantity of vegetables. The court agreed, explaining that even if the use of the term "veggie" was ambiguous as to whether it referred to meat substitutes or a primarily vegetable-based product, a "reasonable consumer" could simply look to the products' packaging, which provided no indication that any particular vegetable or class of vegetables was present. Rather, as the court explained, "the majority of the photographs on the packaging show the Products clearly mimicking meat as vegetarian meat substitutes," and "[c]onsumers can also readily identify the actual ingredients in the Veggie Products from the ingredient list that complies with federal law."

As for the plaintiff's survey, the court found that it didn't support the complaint's theory of the meaning of "veggie" because it:

asked the wrong question – what plant-based ingredients the consumers believed were primarily in a product. The right question is whether use of the term VEGGIE in light of the types of products challenged and those Products' packaging conveyed that the Veggie Products were meat-alternative or whether those sources conveyed the challenged Products were made with vegetables as opposed to other ingredients.

In other words, even taking the survey responses at face value, the survey's design was simply too flawed to allow the plaintiff to overcome Kellogg's motion to dismiss.

KIND MDL: Leading, closed-ended questions in consumer surveys fail to shed light on consumer understanding of a claim

A few weeks prior to the Kellogg decision, the court in the longstanding class action related to the labeling of KIND's snack bars as "all natural," In re KIND LLC "Healthy & All Natural" Litig. (Southern District of New York),² rejected a different survey. That case, pending since 2015, had included at various times a wide array of allegations that several different representations on the packaging of KIND's snack bars had misled consumers. Eventually, however, only one type of representation remained in the case: the claim that the KIND bars were "All Natural."

KIND moved for summary judgment, arguing that the plaintiffs had failed to develop any evidence that a reasonable consumer would be misled by the company's "All Natural" representations. As part of that effort, KIND also moved to disqualify the opinions of plaintiffs' two experts – one of whom had conducted a survey regarding consumers' supposed understanding of the "All Natural" representation. (The other opined that KIND's snack bars did not fit that definition.) The consumer survey involved a mock-up of a product resembling KIND bar; participants were asked whether they agreed, disagreed, or did not know/were not sure if they agreed with a statement regarding the product's packaging. For example, with respect to the product's "All Natural" representation, the survey asked respondents whether – based on that representation – they would expect that the product (a) "[w]ill NOT contain artificial and synthetic ingredients," (b) "[w]ill contain artificial and synthetic ingredients," or (c) "[n]ot sure/[n]o expectation." According to plaintiffs' expert, 86.4 percent of consumers selected the option that an "All Natural" product would not contain "artificial or synthetic ingredients."

Ultimately, the court rejected plaintiffs' survey as biased and misleading, and as a result granted KIND's motion to exclude expert testimony based on the survey. As the court explained, rather than inquiring into a reasonable consumer's definition of "All Natural" in an open-ended manner, the expert's survey was rather "plainly designed to validate plaintiffs' theory." For example:

In his first question regarding consumer's expectations for an "All Natural" product, [the expert] asks only about one potential definition of "All Natural" – the definition that plaintiffs selected for this case – and only allows survey participants to select from finite choices agreeing, disagreeing, or not having an expectation about this definition. This limited inquiry is insufficient to determine in any meaningful sense how reasonable consumers understand the "All Natural" claim, or to test plaintiffs' theory.

As the court explained, the expert did "not contrast the plaintiffs' theory with any other possible competing theory," or ask "open-ended questions to determine the consumer's understanding of the ['All Natural'] term." Rather, as the expert admitted at his deposition, he only provided two "alternate understandings" of the "All Natural" claims because he "thought that's what the plaintiff's [sic] theory of liability amounted." That rendered the survey, in effect, an invalid "push poll," and its results inadmissible.

What's more, the absence of that survey wound up being fatal to the KIND plaintiffs' case; having failed to adduce any other evidence showing how consumers generally understood the term "All Natural," plaintiffs were left with no proof that the "All Natural" claim was false or misleading. As a result, the court granted KIND's motion for summary judgment as well, bringing a long-running class action case to a swift end.

Next steps

While Kellogg and KIND arose under two very different procedural postures, they are unified by a clear theme: courts will not merely rubber stamp a self-serving consumer perception survey in food labeling class actions. Rather, in order for the survey to have any bite, the expert must ask the right questions – regardless of what the answers might be.

The Litigation and Regulatory teams at Hogan Lovells have considerable experience in the defense of food labeling class actions. For more information about this article, please contact the authors or the Hogan Lovells lawyer with whom you usually work.