Determination Of The Performance Measure Of Executive Compensation

Abstract

This paper investigates the principal-agent model of executive compensation through an empirical study of the interaction between CEO compensation and firm performance. As a multi level regression analysis that specifically shows the weight of the variance of the main independent variable, above and over the other independent variables, the stepwise multiple regression is employed to induce a statistical model of the pay-performance sensitivity. The stepwise multiple regression offers insights into the different weight assigned to the performance measure. In this respect, variances of the variables related to the change in the market value of firms are specifically weighted against each other in order to determine specific characteristics of the pay-performance relationship. The analysis is consistent with the agency theory that firm’ executives take advantage of the lack of control by firms’ owners to pursuit their personal interests. As the United States’ economy tumbles, the change in CEO total compensation does not seem to follow the accounting criteria of performance measures typically specified in management compensation contracts. The study reveals a lack of relationship between CEO compensation and firm performance. The link running from the change in the market value of firms and the change in CEO total compensation is flawed. The incentives faced by shareholders to discipline executives would be able to increase the performance of firms. It would be absurd for the compensation committee to rely on the single firms’ total assets value as the performance measure of CEO compensation. Other performance vehicles, such as returns, earnings, and cash flows should be considered in the determination of executive compensation.