Events

Prof. Mira Frick
Associate Professor of Economics
Yale University

We consider moral hazard problems where a principal has access to rich data
about an agent’s action. We characterize the optimal rate at which the principal
can achieve the first-best payoff as the amount of data grows large. Our results
suggest a novel rationale for the widely observed binary-payment contracts, by
showing that such simple contracts can achieve the optimal convergence rate.
In contrast, contracts that display rich wage variation (e.g., linear contracts)
approximate the first-best at a highly suboptimal rate. Our analysis also yields a
robust ranking over monitoring technologies that does not require the principal
to know the agent’s specific utility function. We discuss how this ranking sheds
light on the difference between comparing information structures for incentive
provision vs. learning.