Effects of Monitoring and Tradition on Compensation Arrangements: An Experiment with Principal-Agent Dyads

Abstract

An experiment tested the theory that behavior monitoring and the presence of a tradition of noncontingent pay would interact to affect compensation agreements in principal-agent dyads. On the basis of the notion that risk takers can command premiums, we also predicted that agents who accepted contingent pay would earn more than those who did not. Finally, we predicted that the presence of a tradition of non-contingent pay would anchor both the form and amount of agent earnings. Data collected from 40 dyads supported all three predictions. Further, we observed that the impact of tradition exceeded that of monitoring and became stronger over time. The study has implications for predicting the features of compensation agreements and understanding the impact of traditions on levels of earnings.