Disclosing a Random Walk

Abstract

We examine a dynamic disclosure model in which the value of an asset follows a random walk. An agent learns with some probability the asset value at any point in time in and decides whether to disclose it. He maximizes market perceptions of the asset’s value that reflect the information that was disclosed. We focus on the case in which the agent can disclose only timely information and compare it to the case when he can also reveal stale information. We show that when the agent discloses only timely information he discloses also negative information but argue that overall there is less disclosure. (joint with Andy Skrzypacz and Amnon Schreiber)