Seminario 23/08 – Santiago Sánchez Pagés (King’s College London) – Braggart or humble? The effect of self-reports on performance

Información

An extensive literature has studied both in the field and in the lab the effects on performance of providing absolute and relative verifiable feedback, finding mixed results. Another branch of the literature has studied the effects of non-verifiable centrally-provided feedback, also with mixed results.  But, to the best of our knowledge, no study has explored the effects of performance information provided in a decentralized manner. In this paper, we experimentally study the effects on performance of self-reports, that is non-verifiable reports made by peers about their own performance. In our main treatment, participants perform several rounds of a real-effort task and send a public report about their own performance at the end of that round. Our results show that most self-reports are truthful; only about 15% of self-reports are untrue, including both instances of over and underreporting, although the latter is less frequent. These are significant misreporting rates given that lying about one’s performance has no direct monetary implications. This suggests that misreporting is used to manage social image. We also observe that self-reports do not have performance effects compared to the control treatment where self-reports are absent. However, the content of the self-reports is relevant. More misreporting is detrimental to performance; lies seem to neutralize the positive effects of having truthful performance feedback. In addition, seeing higher reports by others are detrimental to own performance.