A Principal-Agent Model of Sequential Testing
This paper analyzes the optimal provision of incentives in a sequential testing context. In every period the agent can acquire costly information that is relevant to the principal’s decision. Neither the agent’s eﬀort nor the realizations of his signals are observable. First, we assume that the principal and the agent are symmetrically informed at the time of contracting. We construct the optimal mechanism and show that the agent is indiﬀerent in every period between performing the test and sending an uninformative message which continues the relationship. Furthermore, in the ﬁrst period the agent is indiﬀerent between carrying out his task and sending an uninformative message which ends the relationship immediately. We then characterize the optimal mechanisms when the agent has superior information at the outset of the relationship. The principal prefers to oﬀer diﬀerent contracts if and only if the agent types are suﬀiciently diverse. Finally, all agent types beneﬁt from their initial private information.
Gerardi, Dino and Maestri, Lucas, "A Principal-Agent Model of Sequential Testing" (2008). Cowles Foundation Discussion Papers. 1994.