Abstract
Consider the relationship between a regulator (the principal) and anexperimenter (the agent) such as a pharmaceutical company. The pharmaceuticalcompany wishes to sell a drug for profit, whereas the regulator wishes to allowonly efficacious drugs to be marketed. The efficacy of the drug is not known tothe regulator, so the pharmaceutical company must run a costly trial to proveefficacy to the regulator. Critically, the statistical protocol used toestablish efficacy affects the behavior of a strategic, self-interested agent;a lower standard of statistical evidence incentivizes the agent to run moretrials that are less likely to be effective. The interaction between thestatistical protocol and the incentives of the pharmaceutical company iscrucial for understanding this system and designing protocols with high socialutility. In this work, we discuss how the regulator can set up a protocol withpayoffs based on statistical evidence. We show how to design protocols that arerobust to an agent's strategic actions, and derive the optimal protocol in thepresence of strategic entrants.