Peer Effects in Subjective Performance Evaluation

We investigate the influence of peer quality on subjective performance evaluation using 75,413 ratings of 130 employees from 6,908 raters in a business school setting. We find that subjective performance ratings are lower for employees with higher quality peer groups in both randomized and non-randomized settings. Using a novel long-window setting, we observe peer effects persisting, but slowly decaying, for several months even when priming raters with the employees’ previous performance information. We find that the strength of the peer effects is greater for focal employees with weaker performance, for the peers with higher attribute similarity, and when the performance of peers is more extreme. Overall, we find strong and persistent peer effects in subjective performance evaluation