Testing in Models of Asymmetric Information.

This paper explores the role of testing in models of asymmetric information. The authors demonstrate conditions under which testing for underlying characteristics can overcome adverse selection problems and lead to a full-information competitive equilibrium. This paper provides a more general statement of J. Mirrlees's result on the optimal use of infinite fines. Where testing cann ot fully resolve the problems associated with asymmetric information, the authors outline the source of the difficulties. In problems with asymmetric information, testing to discover an agent's chosen action or underlying characteristics may significantly reduce the cost of moral hazard and adverse selection. Copyright 1987 by The Review of Economic Studies Limited.