Information Design of a Delegated Search

Abstract

In this paper, we study a problem where a principal (e.g., a hiring manager, pharma company, advisor) delegates to an agent (e.g., a recruiter, biotech startup, student) a sequential search task spanning a time horizon of finite discrete periods. In each period, an agent's search generates an outcome of a priori uncertain value, whose distribution is publicly known. The agent bears the search cost and decides when to terminate the search. Once terminated, the principal and agent split the value from the search according to a pre-specified fraction. The realized value of each search outcome is only observable to the principal but not to the agent. Leveraging the information advantage, the principal designs an information policy to strategically provide the agent with some information about the search results over time.