Identification of Firms' Beliefs in Structural Models of Market Competition

34 Pages Posted: 27 Jul 2020 Last revised: 16 Aug 2020

See all articles by Victor Aguirregabiria

Victor Aguirregabiria

University of Toronto - Department of Economics

Date Written: June 2020


Firms make decisions under uncertainty and differ in their ability to collect and process information. As a result, in changing environments, firms have heterogeneous beliefs on the behavior of other firms. This heterogeneity in beliefs can have important implications on market outcomes, efficiency, and welfare. This paper studies the identification of firms' beliefs using their observed actions -- a revealed preference and beliefs approach. I consider a general structural model of market competition where firms have incomplete information and their beliefs and profits are nonparametric functions of decisions and state variables. Beliefs may be out of equilibrium. The framework applies both to continuous and discrete choice games and includes as particular cases models of competition in prices or quantities, auction models, entry games, and dynamic investment games. I focus on identification results that exploit a natural exclusion restriction in models of competition: an observable variable that affects a firm's cost (or revenue) but does not have a direct effect on other firms' profits. I present identification results under three scenarios --- common in empirical IO --- on the data available to the researcher.

Keywords: identification, Non-equilibrium beliefs, Revealed beliefs approach, Structural models of competition

JEL Classification: C57, D81, D83, D84, L13

Suggested Citation

Aguirregabiria, Victor, Identification of Firms' Beliefs in Structural Models of Market Competition (June 2020). CEPR Discussion Paper No. DP14975, Available at SSRN:

Victor Aguirregabiria (Contact Author)

University of Toronto - Department of Economics ( email )

150 St. George Street
Toronto, Ontario M5S 3G7
4169784358 (Phone)


Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics