Pharmaceutical Advertising in Dynamic Equilibrium

Department of Economics and Decision Sciences

Speaker:  Pierre Dubois (TSE)

Room T-024

Abstract :

Direct-to-consumer advertising (DTCA) of prescription pharmaceuticals can improve access to innovative treatments, particularly for patients who do not regularly visit physicians. At the same time, DTCA may encourage excessive use and foster business-stealing behavior, shifting demand across drugs without improving overall welfare. While most countries prohibit DTCA, the US and New Zealand remain notable exceptions. This paper develops and estimates two dynamic models of pharmaceutical advertising: one that assumes firms optimize based on observed behavior without imposing equilibrium conditions, and another that incorporates Experience-Based Equilibrium (EBE), in which firm beliefs are required to align with outcomes in frequently visited states. We provide a novel empirical method to estimate and simulate Experience-Based Equilibrium in dynamic games when the curse of dimensionality in firms or actions prevents the estimation of standard Markov Perfect Equilibria. We compare the in-sample fit of both models and use the EBE framework to simulate counterfactual regulatory scenarios, including a ban on DTCA and an alternative public information campaign. Our findings provide new insight into the welfare and market implications of pharmaceutical advertising policy. 

Joint work with : Ariel Pakes