Testing under information manipulation

Abstract A principal makes a binary decision based on evidence that can be manipulated by privately informed agent. The principal’s objective is to minimize the expected loss associated type I and II errors. When commit an acceptance standard, optimal test features ex-post inefficient standards, internalize agent’s manipulation incentives. We provide conditions for set soft or harsh is, lower higher respectively, than standard. misaligned (i.e., low type) dominant, sets standards when prior probability candidate relatively small. In contrast, aligned high large. both scenarios, these result in non-commitment equilibrium outcome Pareto dominated under commitment. also revelation mechanism dominate commitment agent