Printer Friendly

EVA & MVA as performance measures and signals for strategic change.

The increasing frequency with which the business environment demands strategic change elevates the role played by performance measures in assessing alternative business strategies. Traditional accounting measures of performance have long been criticized for their inadequacy in guiding strategic decisions. Two alternative measures of business performance, EVA (economic value added) and MVA (market value added) have been attracting much attention of late. According to a recent article in Fortune, EVA is employed by a large number of firms, including Coca-Cola, AT&T, Quaker Oats, Eli Lilly, Georgia Pacific, and Tenneco.(1) Unlike traditional accounting measures of performance, EVA attempts to measure the value that firms create or destroy by subtracting a capital charge from the returns they generate on invested capital. In addition to their use as performance measures, EVA and MVA are recommended by some as metrics for executive compensation plans and the development of corporate strategies.

Despite this wide interest in EVA and MVA, little is known empirically about the advantages of these measures over traditional accounting measures. Conceptually, EVA and MVA are superior to accounting profits as measures of value creation because they recognize the cost of capital and, hence, the riskiness of a firm's operations. However, conceptual superiority does not always translate into practical advantage.

To shed empirical light on the subject, we have investigated the effectiveness of EVA and MVA as measures of performance, as signals of strategic change, and as metrics relevant to strategic development. The study followed 241 firms over the period of 1987 to 1993. We analyzed the relation between various performance measures and stock returns, which generally are considered to be the best benchmark for a firm's performance; the turnover of chief executives, which is an indicator of strategic change; and the desirable extent of diversification to be pursued by firms, which has been the subject of considerable debate.

EVA and MVA Defined

EVA and related measures attempt to improve on traditional accounting measures of performance by measuring the economic profits of an enterprise - after-tax operating profits less the cost of the capital employed to produce those profits. Details on how to estimate the parameters necessary for the computation of economic profits are contained in the books of Stewart, Copeland et al., and Rappaport.(2,3,4) Some of the ways in which EVA can be raised include increasing operating profits without new capital, lowering the cost of capital, increasing investments in projects that generate rates of return greater than their cost of capital, and curtailing investments in projects that generate rates of return less than their cost of capital.

EVA is closely related to MVA - the difference between the market value of a firm and the economic value of the capital it employs. MVA effectively measures the stock market's estimate of the net present value of a firm's past and expected capital investment projects. Theoretically, a firm's MVA at a given point in time is equal to the discounted present value of the yearly EVA it is expected to generate. Hence, the theoretical underpinnings of both EVA and MVA are found in the net present value concept used by most firms in their capital budgeting decisions.

The increased interest in EVA, MVA, and related performance measures reflects a heightened awareness by corporate managers that their task is to create value for shareholders. Corporate managers have had a legal duty to maximize shareholder value since the advent of the corporate form in the 1800s. In recent years, various market mechanisms have evolved to discipline managers who stray from this goal in favor of other goals such as maximization of size, earnings, earnings growth, earnings per share, and market share. Since the early 1980s, dozens of seemingly profitable Fortune 500 companies have become targets of hostile takeovers that ultimately resulted in the dismissal of chief executives and their management teams. In the past several years, large institutional investors such as Calpers and TIAA-CREF have become more active in challenging management teams that fail to create value; in many cases their activism has also resulted in the dismissal of senior executives.

In their quest for value, managers increasingly have recognized the limitations of traditional accounting measures of performance such as return on assets (ROA), return on equity (ROE), and return on sales (ROS). Although useful for other purposes, these measures suffer from a major limitation as a measure of value creation, because they ignore the cost of capital investments required to generate earnings. The sidebar describing an analysis of XYZ Company demonstrates the limitations of ROA in measuring value added. (See page 36.)

Stern Stewart & Co. coined the terms "EVA" and "MVA" and owns a trademark on both terms. In The Quest for Value, Stewart discusses the theory underlying EVA and MVA, and he describes how to estimate these two measures empirically. Although the terms EVA and MVA are unique to Stern Stewart, others use different names for basically the same concept. For example, Copeland, Koller, and Murrin refer to the "economic profit model," while Rappaport refers to "shareholder value creation."

Measuring Performance and Predicting CEO Firings

Despite the wide interest in EVA and MVA, little evidence exists on the efficacy of using these measures versus more traditional accounting measures to evaluate firm performance. To test the validity of using these new measures, we asked two simple questions:

* How do EVA and MVA relate with stock performance - a well established market measure of performance?

* In terms of an internal effect of performance, are CEO firings related to EVA and MVA?

We collected data on EVA and MVA that Stern Stewart & Co. has published in various sources for 241 large U.S. companies for four years: 1987, 1988, 1992, and 1993. Although we would have liked to have found EVA and MVA data for the intervening years, we do not believe our analysis is seriously impaired by these data limitations. Roughly two-thirds of the sample consists of firms in manufacturing industries.

For each firm, we computed six performance measures for each of the four years: three accounting rates of return (ROA, ROE, and ROS); stock returns; and EVA and MVA, both expressed as returns on equity value. Using the relation of a measure with the stock returns as a test of the effectiveness of the measure, we found that all six measures are positively correlated with stock returns. This suggests that EVA and MVA, like the traditional measures, are effective measures of performance. Moreover, even though not by a large difference, the correlation of EVA with stock returns is higher than the correlation of any of the other five measures with stock returns, providing the EVA with a slight edge as a performance measure.

Our next step was to study the consequences of corporate performance, as measured in terms of EVA and MVA, for chief executives. We began with a search of the business press and other sources to determine whether any of the firms' chief executive officers had been fired during the period 1988-1995. If chief executives left for reasons other than health, death, normal retirement age, or another job opportunity, we assumed they were dismissed for reasons related to performance. We refer to these as cases of "CEO turnover."

We detected 34 cases of CEO turnover, including the highly publicized dismissals of Robert Stempel at General Motors, Kay Whitmore at Eastman Kodak, John Akers at IBM, and Paul Lego at Westinghouse. The frequency of CEO turnover increased over time, with two cases in 1988, three in 1989, three in 1990, five in 1991, five in 1992, seven in 1993, seven in 1994, and two in 1995. The incidence of CEO turnover during the period is about 14 percent.

To determine whether the incidence of CEO turnover is related to a given performance measure, we first ranked the entire sample of 241 firms by their average level of performance over the four years for which data were available. We divided the sample into two subsamples consisting of firms with performance above and below the median level of performance. The incidence of CEO turnover was then compared across the two subsamples. If a given performance measure is used by the market to assess managerial performance, the incidence of CEO turnover should be significantly higher for firms with lower levels of the performance measure.

Exhibit 1 contains the findings of our analysis. The incidence of CEO turnover is significantly related to both MVA and EVA. The CEO turnover rate is 8.3 percent for firms with above median MVA and 20.0 percent for firms with below median MVA. That is, chief executives with below median MVA performance are fired roughly two-and-a-half times more frequently than chief executives with above median performance. Similarly, the CEO turnover rate is 9.0 percent for firms with EVA above the median and 19.3 percent for firms with EVA below the median. The evidence strongly suggests that chief executives who produce high MVAs and EVAs face significantly lower rates of dismissal than their counterparts who produce low MVAs and EVAs.

Not surprisingly, stock returns are also highly correlated with CEO turnover. The CEO turnover rate is 9.6 percent for firms with stock returns above the median and 19.0 percent for firms with stock returns below the median. This suggests that stock price performance is a significant determinant of CEO turnover.

The accounting measures generally are less correlated with CEO turnover. Neither ROA nor ROE is significantly associated with CEO turnover, suggesting that managers are better served to focus on stock returns, EVA, and MVA. The incidence of CEO turnover is, however, more highly associated with ROS, but the data do not permit much confidence in this finding.

These results support the arguments made by proponents of EVA and MVA. The market for chief executives acts as if it uses MVA, EVA, and stock returns more than traditional accounting measures to judge chief executive performance. Failure to perform in terms of EVA and MVA appears to have serious consequences for top management. Because the dismissals of chief executives usually are associated with strategic change, EVA and MVA appear to provide a signal of these changes as well.

The Relationship between EVA, MVA, and Corporate Focus

To examine the effect of focus on corporate success, we evaluated firms in terms of their performance on the six measures and then compared them in relation to one specific strategy - the decision to concentrate activities in one core business rather than to form conglomerates in the hope of exploiting economies of scale through size and synergy across businesses, that is, the extent to which the activities of firms are focused or diversified.

To measure corporate focus, we used a revenue-based Herfindahl index, which is equal to the sum of the squared percentage of a firm's revenues derived from different lines of business. For firms in only one line of business, the Herfindahl measure takes the value of one. For firms in more than one line of business, the Herfindahl takes on a value of greater than zero and less than one; the more diversified a firm, the lower its Herfindahl. For each firm in the sample, we computed its Herfindahl in each year and then found the four-year average. We ranked the sample by the average Herfindahl, and categorized firms as having a focus that was above or below the median.

We found considerable variation in the focus of firms in the sample. The least focused firms are General Electric (Herfindahl of 0.113), followed by Tenneco (0.183), Textron (0.200), DuPont (0.204), and ITT (0.219), while 60 firms are focused in one line of business (Herfindahl of 1).

To assess the relation between focus and performance, we undertook an analysis similar to our examination of the relation between performance and CEO turnover. We compared the performance of firms of above median focus with those of below median focus as shown in Exhibit 2. We found that firms with above median focus earn an average stock rate of return of 31.2 percent per year, while firms with below median focus earn 25.0 percent. The difference in rates of return is sizable (6.2 percent). Our findings support the view that firms with narrower focus experience better stock price performance. Similar conclusions can be drawn regarding performance based on MVA. Firms with above median focus generate MVA returns of 44.1 percent, while those with below median focus generate MVA returns of only 25.6 percent - a significant difference. The performance in EVA terms is also better for the more narrowly focused firms, although the data do not permit us to draw that conclusion with reasonable confidence. These results are similar to those found by Stern Stewart, as reported by Judith Dobrzynski in her New York Times article, "Why the Market Likes Johnny One-Notes and Is Skeptical of One-Man Bands."[5]

The accounting measures also show that the performance of more highly focused firms is significantly better in terms of average ROA (12.74 percent for more focused firms versus 9.97 percent for less focused firms), but not in terms of individual ROE and ROS.


Although EVA and MVA have received considerable attention in recent years and are used by many prominent U.S. firms, there has been limited empirical study of these performance measures. For our sample of 241 firms during the period 1987-1993, we find that EVA and MVA are significantly positively correlated with stock price performance, attesting to their effectiveness as performance measures. Furthermore, affirming the importance attached to these measures, we found an inverse relation between performance in terms of EVA and MVA and CEO turnover. Finally, we found that firms with greater focus in their business activities have significantly higher MVA than their less focused counterparts. Collectively, the results suggest that EVA and MVA are effective performance measures that contain information about the quality of strategic decisions and serve as signals of strategic change.


1. "Creating Stockholder Wealth," Fortune, 11 December 1995, 105-114.

2. G. Bennett Stewart III, The Quest for Value (New York: Harper Collins, 1990).

3. Tom Copeland, Tim Koller, and Jack Murrin, Valuation: Measuring and Managing the Value of Companies (New York: John Wiley & Sons, 1995).

4. Alfred Rappaport, Creating Shareholder Value: The New Standard for Business Performance (New York: Free Press, 1986).

5. Judith Dobrzynski, "Why the Market Likes Johnny One-Notes and Is Skeptical of One-Man Bands," The New York Times, 8 October 1995, Section 3, p. 4.

RELATED ARTICLE: XYZ Company; An Example

XYZ is a telecommunications firm with two units - a local telephone exchange network and a cellular unit, each accounting for 50 percent of the firm's operations. The local telephone unit generates ROA of 8 percent, and the cellular unit generates ROA of 10 percent; therefore, the firm as a whole generates an ROA of 9 percent. Without accounting for the opportunity cost of the capital that investors have provided to the firm, it is impossible to determine if an ROA of 9 percent for this firm creates value.

The local telephone unit is less risky than the cellular unit, and investors, therefore, require a lower return on capital invested in the telephone unit than they do on invested capital in the cellular unit. The required returns (i.e., the cost of capital) for the local telephone and cellular units are 6 percent and 12 percent, respectively. Therefore, the return that investors require on the company as a whole (i.e., the company's cost of capital) is 9 percent. With an ROA of 9 percent, the company is neither creating nor destroying value. It is simply providing investors with a return that they could have achieved by investing in other assets with equivalent risk.


The limitations of using ROA and other accounting measures of performance to evaluate the performance of different units within a firm also can be illustrated with this example. Looking only at ROA, it appears that the cellular unit is outperforming the local telephone unit by two percentage points. However, the ROA of the local telephone unit exceeds its cost of capital by two percentage points, while the ROA of the cellular unit falls short of its cost of capital by two percentage points. Hence, the local telephone unit is creating value while the cellular unit is destroying value, even though the telephone unit has a lower ROA.

In this example, we have ignored the fact that, in practice, ROA - as an accounting artifact - may not measure economic returns. Permitted accounting choices compound the problem, making the ROA incomparable with even the appropriate cost of capital number.

Kenneth Lehn is Professor of Business Administration and Director of the Center for Research on Contracts and the Structure of Enterprise in the Katz Graduate School of Business Administration at the University of Pittsburgh. Anil Makhija is Associate Professor of Business Administration in the Katz Graduate School of Business Administration at the University of Pittsburgh.
COPYRIGHT 1996 Emerald Group Publishing, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:includes related article; economic value added; market value added
Author:Lehn, Kenneth; Makhija, Anil K.
Publication:Strategy & Leadership
Date:May 1, 1996
Previous Article:Strategic process improvement through organizational learning.
Next Article:Strategic planning for rapid & profitable growth.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters