It’s not a drug or device case, but a recent per curiam from the 7th Circuit (with a panel that includes McConnell’s BFF Posner) recently caught our eye. American Honda Motor Co., Inc. v. Allen, __ F.3d __, 2010 WL 1332781 (7th Cir. Apr. 7, 2010), may turn out to be handy for anyone facing a putative class action where plaintiffs rely on experts to do all the heavy lifting proving up key elements of the class claims. How often in a RICO or consumer fraud class action do the plaintiffs bring in an economist who incants the magic words “multiple regression” or “trust me, I’m an economist,” and before you know it, the plaintiffs are claiming that this bit of flim-flam can take care of any individual causation and damages issues? Happens all the time. And how often do courts buy this hokum and certify a class based on the vague assurances of a hired gun because class certification is “not a time to decide the merits?” Happens more than it should.
But in American Honda, the court put a stop to the “Olé” style of Daubert gatekeeping. It found, at least for the 7th Circuit, that before a district court may certify a class, it first “must conclusively rule” on any Daubert challenge if that expert’s testimony is “critical” to class certification. Id. at *3. In American Honda, the critical expert testimony was supplied by an engineer who claimed all Honda motorcycles are defective because they don't adequately dampen “wobble” (yup, that’s right – a wobble expert). Id. at *1. The district court started in a good place, concluding that the supposed “predominance” of common issues – a prerequisite for showing that a class action is appropriate – rested on the validity of the theories advanced by the expert. Id. at *2. And then the district court marched through the expert report, expressing reservations and raising grave concerns about its reliability. Id. at *3. Sounds like we’re about to get a robust Daubert analysis pre-certification, right? But at the last minute, the district court proved it was Born to be Mild; it got cold feet and denied the Daubert motion without prejudice “at this early stage of the proceedings.” Id.
Not good enough, said the Posner posse. The court isn’t allowed to punt Daubert and proceed with a “provisional class” when plaintiffs' whole class action depends on whether or not the experts are going to be able to prove on a classwide basis what they said they can prove. Id. at *4. In dismantling the wobble expert, the 7th Circuit pointed out some rookie mistakes he made, including using a standard he originally developed solely for litigation – a standard that was not generally accepted and had never been tested to determine its reliability as a measure. Id. at *5. The 7th Circuit also faulted the expert and the plaintiffs’ lawyers for the testing they did perform – analysis of a whopping ONE used motorcycle, ridden by ONE test rider, and then making the Evel Knievel-type leap of extrapolating those results to conclude the entire fleet of motorcycles built over a 7-year period had the same wobble problem. Id. at *6. Kicking up gravel at the lawyers, the court concluded, “[t]he small sample size also highlights the constraints litigation placed upon [the expert’s] methods and professional judgment; [he] was not being as thorough as he might otherwise be due to Plaintiffs’ reluctance to pay for more testing.” Id. at *6.
So there you go – we recognize that future courts will be left to grapple with whether an expert is “critical” to class certification or not, but for now, we’ll put this win in our back pockets (and in our briefs). Frankly, it is a decision in keeping with the direction of class certification law generally, and the need for robust and critical early analysis of the “classwide proofs” proffered by a putative class. Any step in that direction, even a per curiam one, is a good result as far as we are concerned.