"Strategic" Behavior in a Strategy-Proof Environment

We present direct field evidence of preference misrepresentation under deferred acceptance. The high-stakes admission process to graduate studies in psychology in Israel was centralized using the applicant-proposing version of deferred acceptance. Yet, a large fraction of these highly educated individuals, who had been informed about the strategy-proof nature of the mechanism in numerous ways, failed to play truthfully. Out of 704 rank-ordered lists that included a non-funded position in a program that offered funded positions, we found that in 137 (over 19%) the non-funded position was ranked higher (or the funded position was not ranked at all). This is despite the fact that the applicants had been informed that rank-ordered lists are never made public, funding is considered a positive signal of ability, and funding comes with no strings attached. Preference misrepresentation is associated with weaker applicants. We provide evidence from a laboratory experiment of a strong, causal, negative relationship between applicants? expected desirability and preference misrepresentation.