Accounting for Net Performance in Managerial Compensation Contracts
Shane S. Dikolli
Duke University - Fuqua School of Business
Ludwig-Maximilians-Universität München - Faculty of Business Administration (Munich School of Management)
University of Vienna - Accounting and Control
March 5, 2012
AAA 2012 Management Accounting Section (MAS) Meeting Paper
Prior empirical compensation studies typically consider net performance (i.e., measured after deducting compensation) while related principal-agent theory is generally based on gross performance measures (i.e., before compensation). We analyze incentive rates in a principal-agent model based on net (and gross) performance. We analytically demonstrate that using gross performance in compensation regressions yields an unbiased estimate of total effort incentives irrespective of whether actual compensation is based on net or gross performance. In contrast, using net performance creates a bias in estimating total effort incentives and yields a non-zero benchmark in testing for the strong-form relative performance evaluation hypothesis. We provide empirical evidence of underestimated total effort incentives when using net performance in compensation regressions, suggesting a possible explanation for surprisingly weak CEO incentives interpreted in prior studies.
Number of Pages in PDF File: 42
Keywords: Executive Compensation, Relative Performance Evaluation, Performance Measurement
JEL Classification: J33, M40, M46working papers series
Date posted: August 17, 2011 ; Last revised: March 6, 2012
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo1 in 0.797 seconds