Victor Augias (University of Bonn)

Optimal Selection Design with Investment

Abstract. We study how to optimally design selection mechanisms, accounting for agents’ investment incentives. A principal wishes to select agents in a heterogeneous population. The principal commits to a possibly random selection rule that depends on a one-dimensional and intrinsically valuable characteristic of the agents. Agents have a strict preference for being selected by the principal and may undertake a costly investment to improve their characteristic before it is revealed to the principal. We show that if the agents’ cost is convex in the amount of investment and the principal wants to select in the upper tail of the characteristic distribution, then deterministic “pass-or-fail” selection rules are optimal. This result provides a firmer foundation as well as a potential explanation for the use of such simple selection rules in environments where agents’ investment incentives are a first-order concern.

Joint with Eduardo Perez-Richet.