Strategic Testing and Screening Equivalence

We study selection contests in which initially uninformed candidates strategically choose the difficulty of pass-fail tests, and a principal selects a candidate based on observed test outcomes. We show that the expected rank of the selected candidate in the symmetric equilibrium is the same whether or not the principal observes failed candidates' test choices. Even more surprisingly, an optimally chosen uniform test for all candidates yields the same expected rank as these decentralized environments. As a corollary, restricting the set of admissible tests cannot improve the equilibrium selection value. By contrast, if the principal can discriminate across candidates, asymmetric testing procedures strictly improve selection.

(with Ilan Kremer, Avi Lichtig, and Andrzej Skrzypacz)

Link to the paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6959119

Amnon Schreiber's homepage: https://amnonschr.wixsite.com/my-site