Committee Search Design

This paper studies the design of committee search procedures. In each time period, a set of candidates of fixed size arrives, and committee members vote whether to accept a candidate out of this set or to continue costly search. We examine the implications of different sample sizes per period on acceptance standards and welfare, and we derive the welfare-maximizing number of candidates per period for small magnitudes of search costs. There is a trade-off between the expected value of a candidate conditional on stopping and the expected search costs. The resolution of this trade-off depends on the voting rule and the shape of the search cost function. In particular, we show that, for all cost functions and all qualified majority voting rules other than unanimity, welfare is increasing in the number of candidates per period if the magnitude of search costs is sufficiently small. This result stands in contrast to the classic finding for the single decision-maker case where the evaluation of multiple candidates per period does not improve welfare relative to reviewing one candidate at a time if there are no economies of scale in the simultaneous evaluation of multiple candidates.