court certifies "C+Collagen" class even if consumers don't know that all collagen is from animals

Gunaratna v. Dennis Gross Cosmetology LLC, No. CV 20-2311-MWF (GJSx), 2023 WL 5505052 (C.D. Cal. Apr. 4, 2023)

Previously, the court denied a motion to dismiss plaintiffs’ claim that “C + Collagen” was misleading because defendant’s product had no collagen. Now, a class is certified.

Defendant argued that individual issues predominated because the court would have to probe each person’s understanding of the phrase “C + Collagen.” The Ninth Circuit previously held that it is an error of law and “per se” abuse of discretion to deny class certification for claims under the CLRA and UCL (and implicitly the FAL), based on a lack of “evidence that consumers uniformly interpret the statement in a particular manner.” Instead, “CLRA and UCL claims are ideal for class certification because they will not require the court to investigate class members’ individual interaction with the product.”

Also, plaintiffs did have common proof: a consumer survey. Defendant argued that the results weren’t uniform, but “the ambiguity (if any) here can only be about what the ‘+’ in ‘C + Collagen’ conveys, because the meaning of ‘collagen’ itself is not up for debate. Unlike the word ‘natural,’ collagen does have a single, controlling definition.” There’s just one dictionary definition!

Plaintiffs’ survey showed that, of those participants that provided an opinion, 95.2% believed that the products contained collagen after viewing images of the them. Furthermore, 51.7% of participants indicated they would be at least somewhat less satisfied if they learned that the products contained amino acids as opposed to collagen. And 49.2% indicated they would be at least somewhat less likely to purchase them again after learning the truth.

Defendant noted that only about half the sample indicated that the presence of collagen was material. That half considered it material “does not suggest to the Court a lack of materiality”:

Though the case law does not establish any uniform percentage that allows a court to conclude that the evidence shows that deception and materiality are susceptible to common proof (or sufficient to create genuine issues of fact), it would seem to the Court, that any percentage that a qualified expert determines is statistically significant should be sufficient for both certification and summary judgment.

Other courts have allowed lower percentages to show common proof on a motion for class certification, “especially where there is no question that all potential class members were exposed to the message because it was on all relevant products sold to the class.” (Citing cases presuming materiality where 37.1%, 24%, and 25% gave answers indicating materiality.)

Defendant’s own survey also provided some evidence of materiality. The survey asked participants, previous purchasers of C + Collagen, to select the most important characteristic that made them buy the product the first time. Out of the 19 possible “product characteristics,” the characteristic selected by the largest proportion of respondents (46.7%) was “C + Collagen.”

Deception and materiality were thus susceptible to class-wide proof and plaintiffs didn’t need not show individual reliance because reliance is presumed upon a showing of class-wide exposure and materiality.

The class-wide damages model was also tied to their theory of liability. In calculating restitution damages under the California statutes, the law “requires only that some reasonable basis of computation of damages be used, and the damages may be computed even if the result reached is an approximation.” The proposed conjoint model could do that. The proposed survey would test the “collagen” claim as compared to the importance attached to labels describing the desired effects of collagen, such as “anti-wrinkle;” “Leaves skin plump with moisture;” “anti-aging;” and “You want: Supple skin, intense hydration.” This would directly test how consumers value the effects of collagen as compared to the collagen claim itself. And anyway defendant’s criticisms went to weight, not admissibility. The damages model didn’t need to isolate and test various possible interpretations of a challenged term; it could assume that the plaintiff’s theory of liability (falsity/misleadingness) was true at the class certification stage. Courts might reject a damages model when plaintiffs haven’t yet shown that the theory of liability was itself capable of class-wide proof, but that wasn’t the case here.

Defendant also made the argument I’ve seen emerging for a while: that, under TransUnion, Article III precluded a class action.  But “TransUnion does not require that Plaintiffs prove standing as to all members of the class in order to certify the class. Indeed, the Ninth Circuit recently rejected the argument that a class may not be certified if it ‘potentially includes more than a de minimis number of uninjured class members.’” Instead, district courts should consider if the class is defined in a manner that will lead to the predominance of individualized issues regarding standing in light of TransUnion.

But a purchaser class was  “defined in a way that ensures that all members will have suffered a concrete economic injury in the form of a price premium, if Plaintiffs succeed on the merits.” Defendant argued that the class would include consumers who didn’t rely on the claim, but they paid a price premium anyway (if plaintiffs show that), and reliance/causation is presumed where there’s class-wide exposure to a prominent message. (I think this is the right result, but it shows the, uh, underdetermined parts of TransUnion, also on display here: (when) can legislatures decide to presume reliance?)

It doesn’t matter that people willingly paid the price, even if they didn’t care about collagen:

If anticompetitive behavior distorted the market – all consumers overpaid. Defendant does not get to price discriminate between those who understood the label and those who did not. The market price is set by supply and demand, and it is always the case that there are likely consumers who would pay more than the fair market price, but that does not mean those consumers should have to pay supra-competitive prices. A price-fixing cartel cannot claim that their inflated prices are not illegal as to the consumers who are satisfied with the value of the overpriced products. It is the distortion of the fair market value that results in injury to all purchasers of the relevant products.

Although it was “possible (if not likely)” that the court would grant summary judgment to plaintiffs on literal falsity, it wasn’t ready to do so without further briefing:

Falsity in this action is not about what consumers believe “collagen” means because the only admissible scientific evidence establishes that there is only one scientifically-accepted definition of “collagen.” Where consumers believe “collagen” comes from is simply irrelevant. Consumers often do not know the sources from which the ingredients in their products are derived. And the Court is troubled, if not exasperated, by the fact that a prominent skincare company has repeatedly taken the position that if reasonable consumers believe, based on the labeling of the Products, that the Products contain a specific ingredient, the falsity of the labeling does not turn on whether the Products actually contain that ingredient, but on whether consumers understand where that ingredient comes from. That position is untenable.

Instead, falsity turned on how reasonable consumers would interpret the “+”: containing collagen, or supporting it with vitamin C? And plaintiffs’ survey “more than suffices” to create a fact issue. Further, there was currently no admissible evidence showing that a statistically significant portion of consumers interpret the plus sign to mean “boost,” given that the court excluded that part of defendant’s survey.

So why not grant summary judgment? Plaintiffs’ survey didn’t give participants the opportunity, either through a closed- or open-ended question, to answer that C + Collagen means that vitamin C boosts collagen. Plus, merits discovery hadn’t closed yet, and defendant could take another bite at a survey on the “boost” theory. “If, at the close of discovery, no such evidence is in the record, Plaintiffs are free to move for summary adjudication on the issue of falsity.”

The court also rejected defendant’s argument that plaintiffs had to show that consumers wanted “animal” collagen:

Since animals are the only source of collagen, anyone who desires “collagen” inevitably desires “animal” collagen. Gunaratna’s testimony that she does not want to place raw animal parts on her face, does not prove that she did not want collagen in her skin cream, any more than a deponent’s testimony that they do not want to consume fish bladder would prove a lack of desire for Guinness beer, as Plaintiffs cleverly analogize.

Defendant was free to attack the reasonableness of plaintiffs’ reliance on the “C + Collagen” Claim, given their stated goals and desires for better skin; given that they did not fully understand where collagen comes from; given the “collagen amino acids” qualifier in other parts of the product packing; and given the vegan symbol on the back of the outer packaging.

The court also pointed out that, as to materiality, “the internal emails and documents are replete with communications indicating that the ‘collagen’ label was highly important to consumers, and in turn to the retailers and Defendant. For example, when the company used other names, Sephora thought they weren’t “strong” or “hard-hitting” enough. Plus, “[r]epresentations about specific ingredients’ presence or absence in a product are almost self-evidently material in that an advertiser is intending to make a consequential effect on a consumer.”

Defendant also argued that there could be no proof of damages because it set the price of the products before choosing a name. But the majority of sales are through retailers, and it’s those prices that are the most probative on price. Anyway, were that true, there would have been a lot fewer sales, making the price premium measure inherently conservative. (Both parties agreed that the market was competitive.) “[T]he fact that a defendant did not adjust its price based on the misrepresentation does not disprove the existence of a price premium.”