96-036

CONDUCT PARAMETERS AND THE MEASUREMENT OF MARKET POWER: THE IMPOSSIBILITY OF ASTRUCTURAL INFERENCE

Kenneth Corts

This paper examines a simple version of the conduct parameter method widely used in empirical industrial organization and argues that the conduct parameter fails to measure market power accurately. It is shown analytically and with simulations that in a dynamic oligopoly model this mismeasurement can be quite severe. A structural model that allows consistent estimation of the desired conduct parameter when firm behavior is generated as the equilibrium of a dynamic game is then developed. JEL classification: C1, L0.

C&S
39 pages

| Back to 1995-1996 Working Papers | Copyright © President and Fellows of Harvard College