How should trial courts deal with challenges to damages models in class actions? The Seventh Circuit recently suggested a burden-shifting approach, in a recent opinion involving loss causation in a securities fraud case.
Glickenhaus & Co. v. Household Int’l Inc., 2015 U.S. LEXIS 8424 (7th Cir. May 21, 2015), involved an appeal after a jury awarded $2.4 billion to a class of shareholder plaintiffs. The shareholders had alleged that the defendants (the lender Household and several Household executives) had used “creative accounting” entries to understate the delinquency rate of Household’s loan portfolio, thereby misleading the market about the company’s performance. Id. at *3-*4. To prove that those misstatements caused injury to the shareholders, the plaintiffs presented an expert analysis from Daniel Fischel (of Compass Lexecon), who had measured the inflation in the share price attributable to the accounting fraud. Id. at *10.
On appeal, the Seventh Circuit ruled that Fischel’s model had failed to account for “firm-specific nonfraud factors” that might have influenced Household’s share price during the relevant period. Id. at *24-*26. Fischel had testified that any firm-specific nonfraud influences were inconsequential (id. at *26-*27), but the Court of Appeals held that his analysis was too undeveloped on that point, ordering a new trial on that narrow issue (id. at *31-*32). In that new trial, the defendants would bear “the burden of identifying some significant, firm-specific, nonfraud related information” that could have affected the stock price, and the once the defendants made such a showing, the burden would shift to the plaintiffs to account for that information in their damages analysis. Id. at *30-*31.
There are two lessons from Glickenhaus.
First, Glickenhaus illustrates a long-standing principle in the Seventh Circuit: damages experts must account for non-actionable factors that contributed to the plaintiff’s injuries. See, e.g., Blue Cross & Blue Shield United v. Marshfield Clinic, 152 F.2d 588 (7th Cir. 1998). In Glickenhaus, a retrial was permitted because the model appeared to be salvageable; the Court of Appeals allowed the plaintiffs a chance to rework their damages analysis to account for any factors that were proven relevant. But not all damages models can be salvaged, as the Court of Appeals noted by citing to Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013). In Comcast, the plaintiffs’ damages model had measured the loss that flowed collectively from four categories of anticompetitive conduct, but the class-wide liability trial would only involve one of those categories of conduct, which left the plaintiffs with no possible way to connect their class-wide damages to their liability theory. Glickenhaus, 2015 U.S. App. LEXIS 8424 at *33-*34.
Second, Glickenhaus describes a procedure for trial courts to use when dealing with criticisms of expert models, with specific instructions for arguments about extra-causal factors. Under the Glickenhaus approach, the burden shifts to the plaintiff only after the defendant has shown that a specific non-actionable factor had a significant causal impact on the plaintiff’s losses. While the Seventh Circuit suggested that burden-shifting approach only for loss-causation studies in securities fraud litigation, trial courts may be willing to borrow that approach when dealing with causation and damages experts in other types of class cases.
Photo Credit: John Liu