Task Assignment, Relative and Absolute Performance Evaluation

Abstract

Screening talent to appropriately assign tasks among agents is an important organizational decision. In this paper, we compare the efficacies of absolute and relative performance evaluation systems in identifying agent talent when information asymmetry is present. We identify conditions under which the firm prefers absolute performance evaluation. We also find that relative performance evaluation schemes are more susceptible to performance manipulation activities by agents. These findings support the use of absolute performance evaluation even though such a system discretizes available data by classifying agent output into "performance buckets" such as pass or fail. We discuss empirical implications.