While several courts have accepted this method to date, this article argues that this method is not reliable.
In several recent antitrust class certification cases, plaintiffs have used a novel econometric method to claim empirical evidence of class-wide antitrust impact. This is a two-step method referred to by its proponents as “in-sample prediction.” The first step involves the estimation of an aggregate overcharge via a regression. The second step uses the coefficients estimated by this regression to predict a but-for price for each at-issue transaction by class members. Proponents of this method claim that it can establish impact for individual observations (i.e., transactions). They identify an at-issue transaction as impacted when the actual price is more than the predicted but-for price.
By carefully analyzing the way this method predicts individual impact, we show that the method does not predict the causal impact of the challenged conduct.
While several courts have accepted this method to date, coauthors Nikhil Gupta and Matthias Lux argue in this article that this method is not reliable. By carefully analyzing the way this method predicts individual impact, we show that, contrary to claims of the method’s proponents, the method does not predict the causal impact of the challenged conduct on a specific transaction.
This article was originally published in the Antitrust Source in November 2025.
Boston 
