We understand that not everyone shares our view of class actions, but we hold certain truths to be self-evident: The government, including its courts, cannot order someone to give money to someone else without due process of law, in most cases a jury trial. In addition, in order to win a judgment ordering a defendant to pay money, a plaintiff generally bears the burden of proof on each element of his or her claims, with very rare exceptions. These principles are cornerstones of the protection that our laws offer to defendants who are haled into court, or at least ought to offer.
Class actions are an exception to these rules. A limited exception, but an exception nonetheless. With a class action, a defendant can become legally obligated to pay money to people who never proved their claims, or even had to try. Instead, class representatives are allowed to stand in as proxies, and proof of their claims is deemed sufficient proof to force defendants to pay the other class members, possibly thousands of them, without any individualized proof. On the other side of the coin, absent class members can be bound by a judgment that awards them nothing, even though they did not participate in the proceeding. How can this be? The potential for abuse is obvious, and that is why there are strict requirements for certifying class actions—Federal Rule 23 and its state-law analogs—and why courts should certify classes only after a “rigorous analysis” demonstrates that the proposed class meets all applicable requirements. See Wal-Mart Stores v. Dukes, 131 S. Ct. 2541, 2551 (2011).
We wax on about these rules because a case in the Southern District of California recently departed from them in not only certifying, but expanding a class of women who allegedly purchased hormone replacement therapy products. The case isKrueger v. Wyeth, Inc., No. 03-cv-2496, 2015 U.S. Dist. LEXIS 137548 (S.D. Cal. Oct. 7, 2015), and it has been around seemingly forever. First filed in 2003, the case was transferred to the In re Prempro MDL, where the MDL judge denied the plaintiff’s motion to certify a class of California patients asserting claims for consumer fraud. Id. at **5-6. Not to be deterred, the plaintiff tried again with a different judge after the case was remanded back to California, and this time they found a more receptive audience.
The district judge in the Southern District of California initially certified this class:
All California consumers who purchased [HRT products] for personal consumption between January 1995 and January 2003, and were exposed to a representation from [the defendant], or health care providers, or read in literature in which [the defendant] advertised or provided to third parties to be disseminated under its brand or the third parties’ brand, that [HRT products] lowered cardiovascular, Alzheimers and/or dementia risk, and do not seek personal injury damages resulting therefrom.
Id. at *7. This is a mouthful, and reading this class definition alone proves that the class should never have been certified. This is a purported “consumer fraud” class, yet the court purports to treat as one all Californians who were “exposed” to a category of representations, made through various means (including through the filters of healthcare providers), over an 8-year period. How exactly are we to know who was “exposed” to any given representation, what it said, and when? How does proof of one person’s “exposure” to a particular representation prove that another person was similarly “exposed” and relied on a similar representation to her detriment? With the changing state of knowledge over time, how do we know that each representation over the course of 8 years was, in fact, misleading or that the individual class member or her doctor did not have access to additional information that affected their decision making? How do we know that each and every class member (and her doctor) would have made different decisions had the allegedly misleading representations been different?
Remember what we said at the beginning about due process and plaintiffs having to prove their claims before a defendant can be obligated to pay? This class threatens that concept because liability to the class members depends on these individual inquiries and others, but the individual class members are not required to make their proof before cashing their checks.
Perhaps the most interesting part of the class definition is the last line: “and do not seek personal injury damages resulting therefrom.” This is a no injury class. It has been winding its way through the federal courts for 12 years and has a masthead of lawyers longer than a Beethoven symphony, yet it is brought on behalf of women who experienced no complications. The whole thing exists for nothing. For all we know, every class member got exactly what she paid for—prescription drug therapy for post-menopausal symptoms that worked as prescribed without adverse effects.
The defendants understandably moved to decertify the class, but it only got worse. According to the district court, the class was ascertainable because each class member was required to produce “documentation” demonstrating exposure to representations “with a view to allowing the defendants the opportunity to challenge an individual’s membership.” Id. at **13-14. And what is the required “documentation”? Each class member has to submit an affidavit stating she is a member of the class. Period. That’s it. Under this Court’s orders, the defendants can then marshal “advertising records or sales call notes” or the like to challenge class membership. That is to say, the court shifted the burden to the defendants to disprove class membership. In those instances where “advertising records or sales calls notes” are not available—and the defendants presented evidence that existing documents could not establish which HRT users were exposed to the defendants’ representations—the defendants are simply out of luck. The individual stands to recover without ever having to demonstrate through objective evidence that she was even a class member. The order even implies that a class member does not have to submit an affidavit until post-trial, so the defendants will potentially have to try the case without knowing who is in the class, while putative class members get to wait and see. That is wrong, and this erosion of the ascertainability requirement is perhaps the most startling aspect of this order. (For more on ascertainability see here and here.)
So how did it get worse? In addition to denying the defendants’ motion to decertify, the court expanded the class definition to remove any reference to exposure to the defendants’ representations. The class now includes every purchaser, regardless of whether she ever saw or heard any allegedly misleading representation:
All California consumers who purchased [the defendants’ HRT products] for personal consumption between January 1995 and January 2003, and who do not seek personal injury damages resulting therefrom.
Id. at *40. For all its faults, at least the initial class definition was tethered in some fashion to individuals who actually might have a consumer fraud claim. The new class definition abandons all that on the theory that the defendants’ promotional campaign for HRT products was so pervasive that everyone must have been “exposed” and that individualized proof of deception and reliance is not required. Id. at **28-36. We are not so sure. It still seems to us that whether any class member perceived any alleged misrepresentation, whether it had any impact on her, and whether she experienced any resulting loss are issues that have to be decided on person-by-person basis, especially when dealing with representations made by various people in various media over the course of several years.
We are aware of the California cases stating that reliance can be presumed under California’s consumer fraud laws where the alleged misrepresentation was material. But even if those authorities are uncontested and can apply in a prescription drug setting, they do not deprive the defendants of the right to mount a defense (there are those pesky “due process rights” again.) A good example of this is the Eighth Circuit’s opinion in In re St. Jude Medical, Inc. Silzone Heart Valve Products Liability Litigation, 522 F.3d 836, 840 (8th Cir. 2008), where the court reversed certification of a consumer fraud claim based on similar presumptions because the law “does not eliminate the right of a defendant to present evidence negating a plaintiff's direct or circumstantial showing of causation and reliance.” A presumption can be rebutted, and now that this class is open to everyone who bought HRT products in California over an 8-year period, there is no doubt that the class includes thousands of people who never saw, heard, or felt anyalleged misrepresentation, or if they did, it had no impact on their decision making. And that overlooks the fact that even if any of these class members were actually deceived, none of them experienced any injury—i.e., the product worked as prescribed without complication.
Rebutting any “presumption” therefore requires individual inquiries for each putative class member, thousands of them, which destroys any semblance of predominance of common issues. The court mentions this in passing when recounting the defendants’ arguments, but it never really addresses this fatal problem with this expansive class. Instead, the court finds the claims “particularly well-suited to class-wide resolution” because they will involve complex evidence and expensive expert testimony. We are all for efficiency, but that goal can come at an unacceptable price. Here, we think the parties got the wrong result.