Andrew Bongjune Choi

Abstract:  A principal wishes to retain an agent only when the state is good and privately receives information about the state over time. The principal wants the agent to stay so that he might retain her, but the agent faces a decreasing outside option and would rather leave if she does not expect to be retained. The principal optimally induces the agent to stay by committing to commit, that is, by committing today to tell the agent tomorrow about what the principal will do the day after. When the agent has a high initial outside option, with some probability, the principal retains the agent regardless of his information---even if he realizes early that the state will turn out to be bad. As a result, the agent stays too long and is retained too often relative to the first-best benchmark. The principal may ask the agent to stay until he fully observes the state, only to fire her; this does not necessarily mean that the principal leads the agent on. We apply our results to worker retention, relationship-specific investment, and forward guidance.

Presentations: Econometric Society Meeting (Tokyo, Shenzhen), Young Economist Symposium (Yale)

Abstract:  We introduce search friction into the two-sided matching model with transfers by requiring that a worker-firm pair can match only if they are acquainted. Each worker may conduct a costly search to acquaint himself with new firms. A matching is stable if there are no blocking pairs among acquainted agents and no worker wishes to conduct a search. A matching is efficient if it maximizes total surplus given the acquaintance of agents, and no searching can increase total surplus net of search cost in expectation. All stable matchings are efficient, but not all efficient assignments can be made stable.

Presentations: Econometric Society Meeting (WUSTL, CUHK)