Interrelated Performance Measures, Interactive Effort, and Optimal Incentives

Abstract

This study uses principal agent analysis to investigate how the principal’s use of performance measures in the agent’s compensation contract are affected by (1) links between performance measures and (2) substitute and complementary characteristics of an agent’s efforts. We show that the directional effect of changes in performance measure interrelations on linear incentive weights depends on how the agent’s tasks interact with each other (i.e., substitute or complementary interactions). For example, increases in performance measure interrelations do not necessarily require higher incentive weights on more sensitive and precise performance measures. If efforts are substitutes for each other, the costs of effort are relatively high and the principal induces lower levels of total effort by offering lower incentives. We also show that differences in the combination of performance measure interrelations and effort interactions affect profits in distinctly different ways. When efforts are substitutes for each other, increases in the sensitivities of profit to the other performance metrics (i.e., increased interrelations), and thus to effort, may actually lead to lower profits.

Keywords: Principal-agent Theory, Incentive Compensation, Balanced Scorecard

JEL Classification: D82, J33, M40, M46