Experts in consumer protection class action matters increasingly propose to apply a technical marketing research tool known as a conjoint survey, sometimes combined with an economic tool often used in merger analysis in which “but-for” market outcomes are simulated using mathematical models of competition, to estimate damages from alleged misrepresentations.
Market simulations that we have seen used in consumer protection class action litigation apply what is known as the static Nash Bertrand model of competition to define the competitive interaction between firms in the market, a critical assumption that is often not explicitly disclosed, let alone examined and justified. While incorporating a model of the supply side of the market is a necessary step for predicting a but-for price, the static Nash Bertrand model of competition is not necessarily suitable for any given market because it makes strong assumptions about competitive behavior.
Several studies in the peer-reviewed economics literature have demonstrated that simulations based on this model frequently fail to predict real-world prices with even reasonable accuracy when tested empirically. Retrospective merger studies have shown that static Nash Bertrand-based simulations can substantially overestimate or underestimate actual price changes, making them unreliable for quantifying damages—an application for which numerical accuracy is of high material importance.
Identifying a model of the competitive process that is applicable to a given market requires market-specific empirical analysis, as different markets exhibit different competitive dynamics that significantly affect price predictions.
The full working paper, written by CRA’s Jarrod Welch and Debra Aron, is available here.
