Unrealistic Expectations and Misguided Learning∗ Paul Heidhues ESMT

We explore the learning process and behavior of an individual with unrealistically high expectations about ability (“overconfidence”) when outcomes also depend on an external fundamental that affects the optimal action. Moving beyond existing results in the literature, we show that the agent’s belief regarding the fundamental converges under weak conditions. Furthermore, we identify a broad class of situations in which “learning” about the fundamental is self-defeating: it leads the individual systematically away from the correct belief and toward lower performance. Due to her overconfidence, the agent—even if initially correct—becomes too pessimistic about the fundamental. As she adjusts her behavior in response, she lowers outcomes and hence becomes even more pessimistic about the fundamental, perpetuating the misdirected learning. The greater is the loss from choosing a suboptimal action, the further the agent’s action ends up from optimal. We argue that the decision situations in question are common in economic settings, including delegation, organizational, public-policy, and labor-leisure choices. We partially characterize environments in which self-defeating learning occurs, and show that the decisionmaker is indistinguishable from a rational agent if, and in some cases only if, a specific non-identifiability condition is satisfied. In contrast to an overconfident agent, an underconfident agent’s misdirected learning is self-limiting and therefore not very harmful. ∗We are grateful to Daniel Benjamin, Volker Nocke, Demian Pouzo, as well as seminar and conference audiences for insightful comments and feedback. Heidhues and Kőszegi thank the European Research Council for financial support under Starting Grant #313341. Part of this research was carried out while Heidhues and Kőszegi visited the WZB, which they thank for its hospitality. Kinga Marczell provided excellent research assistance.