MICRO/THEORY: Data-Driven Contract Design; Professor Justin Burkett (Georgia Institute of Technology)

Abstract

We propose a prior-free model of incentive contracting in which the principal's beliefs about the agent's production technology are characterized by revealed preference data.  The principal and the agent are each financially risk neutral and the agent's preferences are understood to be quasilinear in effort. Prior to contracting with the agent, the principal observes the output produced by a population of identical agents in best response to finitely many exogenously-specified contracts.  She views any technology that rationalizes this data as plausible and evaluates contracts according to their guaranteed expected payoff against the set of all such technologies.  This paper does four things.  First, we characterize the set of technologies that are consistent with the revealed preference data.  Second, we show that robustly optimal contracts are either empirical contracts or equity bonus contracts that supplement mixtures of contracts from the data with equity payments.  Third, we provide conditions under which these optimal contracts append equity payments to only a single contract from the data.  Fourth, and finally, we show that all of our results generalize without complication to a setting in which there might be arbitrary forms of unobserved heterogeneity within the population of agents.

Click here to view paper.