Motivated Misspecification

Abstract: 

I propose a model of expectation management to investigate how the type of misperception is determined in a principal-agent framework. In my model, the principal controls the agent’s expectation of a project's potential, and the agent exerts effort over time. An unrealistically high expectation stimulates effort in the short run but potentially backfires and lowers the agent’s effort in the long run. I characterize circumstances under which the principal makes the agent overly optimistic or pessimistic about the project. The key intuition is that, to sustain excessive effort, the principal should downplay factors that affect project output independent of the agent’s effort. Such manipulation makes the agent misperceive that the output is highly sensitive to her effort, therefore stimulating more effort by the agent. My work thus provides a novel approach to induce effort (perception manipulation), complementary to the usual monetary or informational incentives studied in the principal-agent theory. I apply my results to understand manipulation in a wide range of interactions such as mentorship and abusive relationships.