Robust Performance Evaluation

Abstract

This paper considers a moral hazard in teams model in which a principal knows that the agents she compensates are identical and independent, but does not know all of the actions they can take. In the face of this uncertainty, the principal chooses a symmetric contract that yields her the highest worst-case expected profit. It is shown, counterintuitively, that any such contract exhibitsjoint performance evaluation --- each agent's pay is increasing in the performance of the other--- and isnonlinear in team output. Hence, when robustness is a concern, nonlinear team-based incentive schemes--- such as team bonuses and employee stock options--- are justified. The analysis highlights a fundamentally new channel leading to the optimality of joint performance evaluation. The full paper can be found at the following link: https://ashwinkambhampati.github.io/RobustPerformance_EC.pdf.