Photo of John Sullivan

California rejected another attempt by the class action bar to extend the already questionable fraud-on-the-market theory from Basic v. Levinson, 485 U.S. 224 (1988), a securities class action, to what amount to failure to warn claims for consumer products or, as we’ve seen before, drugs and medical devices.  This time the class action plaintiffs’ bar was focused on e-cigarettes.  See In re NJOY, Inc., Consumer Class Action Litig., 2016 U.S. Dist. LEXIS 24235 (C.D. Cal. Feb. 2, 2016).  A handful of hopeful consumers claimed that they were misled by an e-cigarette’s labeling and were not warned about its ingredients or risks.  Id. at *3.  As is often the case with these types of class action claims, however, the plaintiffs did not allege an injury—well, at least not a physical injury.  They suffered no side effects.  They had no physical ailments.  The risks didn’t affect them.

So what were they claiming?  It’s rarely easy getting your arms around claims like these.  The plaintiff purchased the product.  It did what it was supposed to do.  And the plaintiff enjoyed it.  In fact, many plaintiffs would have purchased the e-cigarette even if the ingredients and alleged risks were disclosed.  So, again, what exactly is the claim?

Since we can’t figure that out, bring on the economists.  Why not? No group of professionals carries with it greater public confidence in their ability to bring clarity to confusion. Economists are easy to understand, helpful, and rarely wrong.  Right? So, they can tell us how these plaintiffs were damaged.
So . . . it turns out that the plaintiffs paid too much for the e-cigarettes.  The economists were able to determine this easily, using a conjoint analysis, a direct analysis and, of course, a Bayesian Hedonic regression. Ahh . . .
So what are these things you ask?  Well, first of all, don’t worry.  A conjoint analysis isn’t something that happens in the back of Cheech and Chong’s smoke-filled van.  It’s an econometric analysis.  Well, to be simpler, it appears to be a survey.  Direct analysis is sort of a survey too.  As applied by the economists in In re NJOY, the direct analysis involved questions to e-cigarette consumers intended to uncover what they would pay for additional safety.  The conjoint analysis asked consumers to make a series of choices between e-cigarettes with different attributes.  And the Bayesian Hedonic regression, which isn’t a drinking game in Cancun, uses pricing from the e-cigarette market, a regression analysis that measured values of certain e-cigarette attributes, and the conjoint analysis survey results.  Taking all this into account, the economists developed a value for safety and a proper safety warning. Simple, right?  I’m sure you feel confident that the economists got it right, establishing precisely how much each plaintiff overpaid.
What, you’re not?  Well, don’t feel bad.  You’re not alone.  The court wasn’t comfortable with any of it.  It held, citing the Saavedra decision about which we previously blogged, that the conjoint and direct analyses were impermissibly based on consumers’ subjective valuations:
[A] consumer’s subjective valuation of the purported safety message, measured by their relative willingness to pay for products with or without the message is not an accurate indication of restitutionary damages, because it does not permit the court to calculate the true market price of N-JOY e-cigarettes absent the purported misrepresentations.
Id at *21-22 (citing the Court’s previous decision denying an earlier motion for class certification).  Setting aside the limitations of subjective data, all of it comes from the demand side, with no data from the supply side.  Id. at *18-21.  It was an incomplete analysis, and the court rejected it.
Worst of all—at least from a trial tactics point of view—one of plaintiffs’ own experts had already conceded that the “Bayesian hedonic regression would not be appropriate in the context of the e-cigarette market” because the market isn’t sufficiently stable. Id. at *26.  That’s a “game over” admission.  The court rejected that analysis too.
Accordingly, since plaintiffs couldn’t establish that “damages are capable of measurement on a classwide basis,” the court denied class certification.  Id. at *27 (relying on Comcast Corp. v. Behrend, 133 S. Ct. 1426, 1433 (2013)).
Not that we would have been satisfied that class action treatment was appropriate if plaintiffs’ economists had created some sort of model that could measure damages classwide. This isn’t a securities class action, in which a class member would purchase an ownership share of a company.  Drugs and medical devices (and, to a similar extent, e-cigarettes) are consumer products purchased for different reasons by different people under different circumstances.  A warning might not change the behavior or decision of one putative class member, or change it only slightly, but cause another putative class member not to buy the product at all.  With a prescription product, one doctor may know the risk regardless of the warning, and another may learn the risk from the warning but nonetheless prescribe the product.  The class would be rife with members who used the product for materially different reasons under materially difference circumstances, the anti-thesis of class treatment.  These differences cannot be washed away by econometrics.
In fact, one of the named plaintiffs in the In re NJOY litigation admitted that he would have purchased the e-cigarette even if the allegedly dangerous ingredients had been disclosed.  The court, therefore, upheld a previous ruling that this plaintiff lacked standing to assert an omissions claim.  Id at 32.  This is one example of the many potential differences between putative class members, differences that would only be increased by the injection of the learned intermediary doctrine.
In any event, with a confusing, incomplete econometric analysis that never came close to even suggesting that it could produce valid results, and a putative class loaded with class-killing differences, this class action never stood a chance.  The court properly denied class certification.