4th Mitsui Life Symposium on Global Financial Markets Corporate Governance in Japan and the United States “Revitalizing American Firms: Shareholder Value Creation and EVA”
The Fourth Mitsui Life Symposium on Global Financial Markets
EVA AND SHAREHOLDER VALUE IN JAPAN
Sponsored by the Mitsui Life Financial Research Center
At the University of Michigan Business School
Keidanren Kaikan, Tokyo
May 10, 1996
JOURNAL OF APPLIED CORPORATE FINANCE
VOLUME 9, NUMBER 4, WINTER 1997
E. HAN KIM: Welcome to the Fourth Mitsui Life Symposium on Global Financial Markets. My name is E. Han Kim, and I am Director of the Mitsui Life Financial Research Center at the University of Michigan Business School. It is my pleasure to introduce the two speakers who will open the proceedings of this symposium.
We are privileged to have in attendance the President of Mitsui Mutual Life Insurance Company, Mr. Koshiro Sakata, whose generous gift has endowed the Mitsui Life Financial Re search Center at the University of Michigan. President Sakata will be followed by Joe White, the Dean of the University of Michigan Business School. I would like to thank both President Sakata and Dean White for finding time in their very busy schedules to join us this afternoon.
KOSHIRO SAKATA: Thank you very much, Professor Kim. Today we are very pleased to welcome so many participants to the Fourth Mitsui Life Symposium. The relationship between Mitsui Life Insurance and the University of Michigan dates back six years to when Professor Kim of the University of Michigan and Professor Wakasugi of the University of Tokyo proposed this program. The University of Michigan, needless to say, is one of the top universities in the United States. I have had the opportunity to visit the University several times, and have seen many students with promising futures engaged in their studies and research. In addition to the scholarly atmosphere, I was also impressed by the natural setting of the campus, with all its greenery and flowers.
It was in 1990 that the professors Kim and Wakasugi came to us to propose a plan designed to raise the level of research on financial systems and capital markets in different countries around the world. We wholeheartedly endorsed their proposal and agreed to contribute the funds, and so the Mitsui Life Financial Research Center was established at the Business School.
In 1991 the first symposium was held at the University of Michigan. The second symposium was held in this same hall we are in today at Keidanren. For the third symposium, the setting was once again the University of Michigan, and the meeting was attended by many prominent scholars from the United States. I am pleased to learn that the research results supported by the Center have been widely reported in the financial and business communities as well as in academic circles.
The theme of today’s symposium is EVA and shareholder value creation. Although it is a new concept to me, EVA is now considered to be one of most basic themes of capitalism, and it is indeed a timely topic for the symposium. So I am confident that today’s discussions will be useful and meaningful to all of you. I would like to thank all of the participants for taking the time to attend the symposium. In fact, I have been told that there are also participants from our neighboring country Korea–and we are very pleased to have them. I would also like to express special thanks to Dean White and the other members of the University of Michigan as well as the panel members and lecturers.
JOSEPH WHITE: Thank you, Mr. Sakata, both for your kind remarks and your consistently generous support of the activities of the Mitsui Life Center at the University of Michigan. I would also like to express my gratitude to Professors Tak Wakasugi of Tokyo University and Han Kim of the University of Michigan. As co-directors of the Mitsui Center, they have presided over an important program of research, visiting scholars, and symposiums over the last six years. And today’s event should be an impressive addition to this record. Professor Wakasugi is the ideal person to speak to us on EVA from the Japanese economic perspective. I am also grateful for the participation of Mr. Toru Mochizuki, Mr. Mark Newburg, and Mr. Virgil Stephens, the chief financial officers, respectively, of Coca‑Cola Japan, NCR Japan, and Eastman Chemical Company. They will provide their senior management perspective on this important matter.
Finally, I am also grateful to Mr. Joel Stem, managing partner of Stern Stewart & Company for his participation today. His work on the subject of EVA has gained international recognition. EVA is playing a major role in the successful efforts of many U.S. chief executive officers to revitalize the financial performance of their companies.
This leads to interesting and important questions: How and why does EVA work? Why don’t all companies adopt it? Is it culturally limited–that is, applicable only in the U.S. or in the West, but not in Japan and Asia? Will EVA turn out to be just another management fad, or will it prove, like total quality management, to be an enduring management focus and tool? With Mr. Stern, Professors Wakasugi and Kim, and our executive panelists, we have just the right people to answer these questions. I look forward to a very stimulating and informative symposium.
KIM: Thank you, President Sakata and Dean White, for the kind remarks. The topic of the symposium, as mentioned, is “Economic Value Added and Shareholder Wealth.” To discuss this topic, we have assembled a distinguished group of speakers, representing both academic and corporate perspectives that reflect Japanese as well as U.S. experience.
At our last symposium here in Tokyo three years ago, the theme was corporate governance in Japan and the United States. In selecting that theme we were fortunate in anticipating the rise of interest in corporate governance in Japan. One issue that is closely and directly tied to corporate governance is the issue of shareholder rights as owners of corporations, and how their ownership interests are represented in managerial decision‑making. Economic Value Added, or EVA, is a management tool designed to encourage managers to do a more effective job of representing the shareholder interests. In recent years, EVA has been adopted by a large number of American companies, and it has been given a lot of credit for increasing the shareholder value of many of these firms.
Professor Wakasugi, who is the co-director of the Mitsui Center as well, will start off this symposium by discussing the current state of Japanese corporate governance, and the potential role of EVA in improving Japanese corporate performance. Professor Wakasugi is a well‑respected international scholar in financial economics. He has frequently served in an advisory capacity to the Japanese Ministry of Finance, the Ministry of Posts and Communications, and the Government Policy Investment Council. I will now turn the floor over to Professor Wakasugi.
EVA and the Japanese Corporate Governance Problem
WAKASUGI: Thank you, Professor Kim. And, before I begin, let me acknowledge a very large debt to Professor Kim, for much of what I am now about to tell you reflects his insights into U.S. and Japanese corporate governance issues.
As Dean White and Professor Kim have mentioned, EVA is now central to the thinking of American management. EVA is a management tool–more precisely, an internal performance measurement and incentive compensation system–that can be used to represent the shareholder’s viewpoint within the organization. In the United States over the past several years, the EVA approach has become very popular. Many of the companies that have adopted EVA as an internal performance measure and as a basis for incentive compensation have experienced significant operating improvements and increases in stock price.
Japan’s experience with corporate governance is one in which-‑historically at least-‑the interests of shareholders have been subordinated to the competing demands of other corporate stakeholders. In recent years, however, there has been increasing concern about the issue of shareholder representation. Indeed, we are now, I believe, on the threshold of a national acknowledgement that greater concern for shareholder interests may be the only way to reinvigorate our depressed capital markets.
So, before discussing EVA and its potential for increasing shareholder wealth, I will review some of the historical factors in Japan that may prove instrumental in leading Japanese companies to adopt this new American management tool. Please keep in mind that I will be presenting EVA from a Japanese perspective. That is, my version of EVA and that which will be presented by the American managers on the panel may involve some slight differences ‑‑ differences that reflect differences in the corporate cultures of the U.S. and Japan.
Most of you will recall that, in the 1980s, Japanese companies became very prosperous, very rich. They used their newfound wealth to diversify into new business areas. The primary focus of Japanese corporate managements was on growth–growth in sales and assets, growth in earnings, and growth in market share. New business ventures and rapid expansion of existing activities were two common ways of pursuing such growth. Moreover, in both domestic and foreign markets, the scope of Japanese companies’ investment activities was expanded to include investments in real estate and financial assets as well as the more traditional investments in plant and equipment. Growth was the watchword, as the managers of these companies were convinced that large scale and market dominance would be the keys to success in the future.
Implicit in this strategy was the assumption that capital was an unlimited and free resource. In pursuing growth at any price, no consideration was given to the cost of capital. This way of thinking is easy to criticize in hindsight, but recall that at that time Japanese managers won plaudits not only at home but also in foreign markets for their long term planning horizons and their willingness to sacrifice current earnings for future growth and market share. From the vantage point of these managers, moreover, it was easy to fall into the trap of treating capital as if it had no cost–for it certainly gave every appearance of being free. Only a tiny fraction of internally generated cash flows were distributed to shareholders as dividends and yet share prices continued to scale new heights year after year. Additional external capital could be raised with convertible bonds issues offering rates as low as two or three percent; and if such bonds were issued in foreign denominations, the yen appreciation sometimes brought the effective cost down to zero or less. So, the perception of free capital helped to justify the long time horizons of Japanese companies–planning horizons that became known, and were indeed widely celebrated, as “Japanese-style management.” Whether this was all correct or not is something that we now have to reconsider.
At the same time the Japanese management style was commanding worldwide respect, managers of U.S. companies were being criticized for focusing on cost-cutting and short-term earnings performance. As the globalization of product markets increased, the competition from imports faced by U.S. firms intensified. Achieving production efficiencies and paring product prices became crucial to survival for many U.S. firms. Thus, while Japanese companies were on an expansion binge, American companies were encouraged, if not compelled, by U.S. capital markets to engage in downsizing, divestitures, and decentralization. Inside these companies, individual units began to be treated as profit centers, and each of the units was required to reach adequate levels of profitability.
This new sense of urgency in restoring efficiency and profitability required that American managers turn away from traditional performance measures such as revenue growth or earnings. Each unit had to justify its existence and to do so required that they also justify their ongoing use of the companies’ capital resources. EVA, unlike popular performance measures like earnings per share and EPS growth, recognizes that capital has a cost and that a business unit can justify its use of capital only if what it earns exceeds that cost. Thus, the competitive pressures that led to demands for decentralization and greater operational efficiency were also the primary catalyst for the adoption of the EVA methodology by American firms.
Where does this leave Japan? In the 1990s Japan has experienced a serious and protracted recession–and its stock market has stagnated along with the general economy. There has been a growing awareness that the poor performance of the Japanese stock market is attributable in large part to Japanese companies’ emphasis on growth at all costs and neglect of shareholder returns–both of which can in turn be traced to the overall governance system’s lack of adequate representation of shareholders’ interests. As one sign of this new awareness, there has been increasing interest among Japanese managers in recent years in measuring corporate performance with return on equity, or ROE.
And, in fact, a simple comparison of ROEs over the past decade can serve to highlight the current difference in performance between American and Japanese firms. In every year during the period 1985-1995, the average Japanese ROE was lower than the average ROE in the U.S. Although part of this difference can be explained by more conservative Japanese accounting methods, much of it reflects the difference in managerial priorities. For example, whereas in the U.S. we observed an increase in corporate ROEs throughout the 1990s, in Japan there has been a significant decline of ROEs from peak levels in 1989. At present, the average company in Japan is reporting an ROE of only about 2%.
As Joel Stern will tell you later, ROE is not the ideal economic yardstick for evaluating managers’ performance–and the EVA measure is designed in part to overcome the shortcomings of ROE. But ROE nevertheless provides a good indication of whether capital is being used efficiently inside organizations–and, on this score, the difference between Japanese and U.S. firms has become painfully clear.
And such diverging trends in operating profitability, as mentioned, have been clearly reflected in the divergence of stock prices in Japan and the United States. Throughout the 1980s, Japanese stock prices increased in systematic fashion. In the first years of 1990s, however, the Nikkei index lost close to half its value, a plunge that many observers attributed to a speculative “bubble.” Perhaps even more troubling, however, has been the stagnation in prices that has continued since the large drop in the early ’90s. The Nikkei index continues to trade in a 20,000 to 22,000 range, as compared with a peak of nearly 40,000 at the end of 1989.
In contrast, the Dow Jones Industrial Average in the U.S. continues to break new records every year and currently stands at 5500. During the 1980s, when comparing the progress of the Nikkei average in Japan with the Dow average in the U.S., the rule of thumb was that the Nikkei would be ten times the Dow average. That is, if the Dow was 3000, the Nikkei would be in the 28,000 to 32,000 range. Using that rule of thumb today, the Nikkei should be above 50,000. But today the index is well below half of that level.
So, if you compare the two countries in the past decade, we see that the Japanese approach of limitless expansion has been accompanied by a very large decline in shareholder wealth. In contrast, the American strategy of reliance on capital management tools to guide restructuring in the face of global competition has produced a significant increase in shareholder wealth.
Countervailing Forces to Shareholder Value Maximization in Japan
Despite such differences in managerial focus and behavior, however, Japanese and American corporations are fairly similar in terms of their legal organizational structures. And, given such legal or structural similarities, it is pertinent to ask what accounts for such significant differences in behavior and performance. For whom, or for what purpose, is the Japanese corporation being managed?
In terms of corporate governance systems, the Japanese company, like its American counterpart, in principle belongs to the shareholders (or, at least, the shareholders have claim to the “residual” value after all other claims have been satisfied). Therefore, as in the U.S., the logical objective of Japanese management should be to attempt to maximize the wealth of the shareholders. In practice, however, there are several countervailing influences in the Japanese system that allow or even encourage managers to sacrifice the interests of shareholders in order to satisfy the competing claims of other stakeholders in the corporation.
For example, Japanese corporations are noted for the practice of crossholding of shares between companies. The stated purpose of such crossholdings is to build and maintain business relationships between, say, manufacturers and their suppliers. But, in practice, the crossshareholders do not behave like value-maximizing shareholders, and they exert very little influence on corporate management. Moreover, because such a substantial portion of Japanese companies’ outstanding shares are permanently held by other firms, these companies are not subjected to the rigors of the market for corporate control. In the American market, the shares of companies with poor managements can be acquired at depressed prices, which in turn attracts hostile acquirers who often end up replacing the incumbent managers. The Japanese system of crossshareholding prevents the takeover mechanism from operating and leaves under-performing management in place. And, so, in the absence of the threat of takeover, shareholder interests are unlikely to play a significant role in corporate decision-making.
Some corporate finance scholars have argued that there are substitute governance mechanisms that perform the disciplinary role of takeovers in the Japanese economy. For example, Japanese corporations have traditionally maintained close relationships with one or more large banks in the Japanese “main bank” system. Because banks have for a long time been the major suppliers of external financing, including equity as well as debt, it is the banks, and not the other shareholders, that exert the strongest influence over the companies’ management. But, although bank ownership may in some cases end up working to shareholders’ advantage, there is a major problem with this corporate governance solution: the banks’ interests are not entirely consistent with those of the outside shareholders. For example, it is in the banks’ interests as debt-holders that their corporate borrowers both avoid risky projects and provide an ongoing source of loan demand in periods when loanable funds are readily available. Neither of these objectives is necessarily consistent with maximizing shareholder profits.
The claims of employees are another source of conflict with the shareholders’ interest. In defining its priorities, Japanese corporate management places greater emphasis than U.S. managers on the welfare of corporate employees. In particular, Japan is noted for its system of “lifetime employment” and wages based on seniority. While both of these features can be justified as very long-run investments in employee training and morale, neither is necessarily conducive to increasing shareholder profit. And, finally, as I mentioned earlier, Japanese companies tend to emphasize market share over profitability as the key measure of operating success. In so doing, Japanese companies are said to place relatively greater emphasis on customer satisfaction.
Another important difference between the Japanese and U.S. systems–one that helps explain why shareholders are higher on American companies’ list of priorities–is the active role played by pension funds in the United States. Owing to their increasing volume of share transactions, U.S. pension funds have become a decisive factor in valuing corporations. And, as the size of pension funds has grown, they have begun to attempt to influence management decision-making more directly through various kinds of activist governance initiatives. As a result, shareholders in the United States have another means of representing their interests that is not available to shareholders in Japan.
Japan and the United States can also be distinguished from each other by a fundamental difference in attitudes about the proper social role of the corporation. This difference is based in turn on a difference between Japanese and U.S. managers’ willingness to accept the assumption that the market mechanism is working (or should be allowed to work) in every field. In the United States, most of the economy has been liberalized and the market mechanism is functioning in virtually all markets, including markets for labor as well as consumer prices. In such a free-market environment, the maximization of shareholder value is more or less accepted by American society as the primary goal of the public corporation. Stockholder interests can be pursued by U.S. corporate managers, and this pursuit is reinforced by an active market for corporate control that ensures that the system operates with reasonable efficiency.
In Japan, by contrast, we are currently far from the point where we can claim that our entire market system is functioning freely. In order for us to address the current sluggish economy and depressed stock market, it is necessary to reform our entire capital allocation system. The maximization of shareholder value is a fundamental part of that reform, and EVA has the potential to help implement it. So let me give you a brief overview of the concept.
A Potential Role for EVA in Improving Capital Allocation
As individual investors investing in equity or debt, we require a rate of return that is commensurate with the risks associated with the investment. In the case of a corporation, the risk‑adjusted return it must offer in order to attract new financial capital is called the “cost of capital.” The greater the riskiness of the firm’s business activities, the greater will be its cost of capital.
Thus, a problem arises when earnings, per share or ROE is used to measure corporate investment performance. Because neither of these traditional measures takes into account the firm’s risk‑adjusted cost of capital, it is impossible to say whether any given level of EPS or ROE is acceptable to investors. A company’s real economic profit is the amount it earns in excess of the cost of capital–a cost that appears nowhere in the balance sheet or income statement used to calculate EPS or ROE. The current demand for EVA is thus based in part on one simple idea: you simply cannot know whether your enterprise is creating wealth for your shareholders until you subtract the cost of capital from income.
From the perspective of the overall firm and its risk‑adjusted cost of capital, the relevant cost of capital is the weighted average cost of both its debt and its equity capital. In order for a company to create value for all its capital suppliers (debtholders as well as shareholders), its return on total capital must exceed the weighted average cost of capital, or WACC. EVA, then, is just the dollar amount by which a company’s pre‑interest, but after‑tax net operating income (or NOPAT) exceeds the charge for total capital (WACC multiplied by debt plus equity). For example, if a firm with a WACC of 10% earns a NOPAT of $20 million on $ 100 million of total capital, its EVA is $10 million. To the extent EVA is positive, the firm is adding value for its shareholders. But if a company’s EVA is negative, the firm is destroying shareholder wealth even though it may be reporting a positive and growing EPS or ROE.
So, the cost of capital plays a crucial role in implementing an EVA standard. The lenders who lend money to the corporation and the investors whoI purchase its equity all expect to receive at least the risk‑adjusted rate of return. Without positive EVA performance‑‑or without at least the expectation of such performance in the future-‑this is not possible.
Some people have objected that, although EVA may be superior in theory, it is difficult to apply in practice. There are three main aspects of an EVA system-‑”planning,” “execution, ” and “evaluation”‑‑each of which is necessary to make EVA an effective operational management tool. Management must begin by making a plan formulated in terms of expected EVA; the plan must then be carried out; and management’s success in meeting the plan must be measured-‑again, in terms of EVA-‑and the managers held accountable.
But, in order for this three‑part management decision‑making cycle to function effectively, there are four conditions that must be met. These four conditions are very important from the standpoint of the shareholders. First, it is important to have a well-defined managerial objective (say, earn a positive EVA, or at least increase EVA each year). Second, appropriate criteria must be used in selecting investment projects (accept only positive‑EVA projects, those expected to earn at least their cost of capital). Third, companies must evaluate the actual performance of their investment over regular time intervals-‑again using EVA to measure the resulting wealth creation or destruction from the standpoint of the shareholder. Fourth and finally, the managers responsible for the capital allocation decision must be provided incentives-‑say, in the form of year‑end bonuses tied to EVA‑-that encourage them to make decisions consistent with shareholder objectives.
This, in brief, is the management approach based on the concept of EVA. One of the great strengths of the system is the consistency it maintains throughout the three management phases of strategic planning, performance measurement, and managerial compensation. Before going onto an EVA system, many U.S. companies would use the discounted cash flow method when making their investment decisions, but then use traditional measures like EPS or ROE when evaluating performance after the investment. The resulting inconsistency in the criteria for capital budgeting and performance evaluation would then lead to conflicting priorities and internal confusion among managers. EVA, however, is a measure that can be used for each of these three functions–strategic planning, performance evaluation, and incentive compensation. By achieving consistency across these three functions, managers are provided incentives both to make value‑increasing investment decisions and to operate as efficiently as possible.
Thank you for your attention.
KIM: Thank you, ProfessorWakasugi. Now Joel Stern will tell us about his recent efforts to extend EVA outside the U.S. and discuss how it works in actual corporate settings.
An EVA Primer
STERN: Thank you, Han. It is a pleasure to return to Tokyo. Since my last visit several years ago, I have read with great interest a number of academic papers that have been published by professors here. And I hope that you find of some interest what I am about to tell you.
Economic value added, contrary to what you may have heard, is not just a U.S. phenomenon. With the help of Stern Stewart, companies have been implementing EVA in South Africa for the past ten years. We are also presently working with companies in Canada, the U.K., Germany, France, Sweden, Australia, New Zealand, Mexico, and–as I will mention briefly in closing–Singapore. Moreover, we are advising companies operating in a broad range of industries‑‑everything from consumer products and industrial companies, high‑tech and pharmaceutical companies, to regulated companies such as banks, insurance companies, and public utilities. In fact, the list has recently expanded to include a number of not‑for‑profit and government‑owned organizations.
So, our focus at Stern Stewart is not just on increasing sbareholder retums and shareholder wealth. We are interested in making sustainable improvements in the productivity of all kinds of organizations. My favorite example is the United States Postal Service, which this past year adopted EVA with our help. U.S. postal workers now have a new arrangement with the American public-‑it’s a deal that says they will receive bonuses for improvements in EVA, but with a significant portion of the bonuses deferred and held at risk to ensure that such improvements are not temporary. Everybody who works for the postal service is a partner in this effort.
But, to return to my earlier statement, EVA is not limited to America, and it is not limited to profit‑seeking corporations. EVA has the potential to help a wide variety of organizations both evaluate their own performance and do a better job of identifying value‑increasing ways to invest their scarce capital. As just one example, EVA can be used by an acquiring company when setting the maximum price it is willing to pay for an acquisition target. Our feeling is that, if you understand EVA, you will know when to walk away from value‑reducing opportunities to invest funds.
The real payoff from an EVA financial management system comes from the direct linking of a better performance measure with incentive compensation. Some companies, I have been told, are using EVA for strategic planning and evaluation purposes, but not for incentive pay. That is a big mistake. If you measure performance on one basis, but pay bonuses according to another, most people will pay lip service to your stated goal and manage the variables that affect their bonuses. What we want to do is get the right measure of performance and make that the basis for the incentive scheme.
Let me also point out that when we devise an incentive scheme, bonuses are generally based not on the absolute level of EVA produced by a manager or an operation, but rather on the year‑to‑year improvement in EVA. The financial management goal is continuous improvement in EVA. One advantage of tying bonuses to improvements in EVA is that such bonuses effectively come at no cost to the shareholders. That is, the management bonuses earned under an EVA system are effectively paid for out of the much larger increases in shareholder value that accompany such EVA improvements.
Another important feature of our incentive compensation system is what we call the “bonus bank.” We pay only one‑third of the declared bonus upon declaration; the remaining two-thirds is held at risk and is subject to loss if you don’t sustain the improvement. In other words, in order to receive the full award, you must both increase your EVA and at least maintain that new level of EVA for the next two or three years thereafter.
Besides strengthening management’s incentives in a way that does not dilute the shareholder interest, yet another important potential benefit of EVA is that it can help companies communicate more effectively with the investment community. We have found in the U.S. and in the U.K.-‑and now in Germany as well, that institutional investors want to buy shares in companies that announce they are adopting EVA. Why? Because an EVA system acts as a policeman to make sure that corporate resources are not wasted.
One of the biggest problems with highly diversified or conglomerate forms of organizations‑-and I understand there are still a great many of them in Japan today‑-is that when excess cash is generated by one business, it is very often transferred as a subsidy to others that are not doing so well. And this cross‑subsidization of negative‑EVA by positive‑EVA businesses often ends up reducing the value of the entire enterprise. This problem was identified by Harvard professor Michael Jensen as the “agency costs of free cash flow.” What does that mean? It means that the excess capital, instead of being returned to shareholders who can reinvest it elsewhere, is being reinvested inside the company at unacceptably low rates of return. The managers are supposed to act as “agents” for their shareholders by making the most efficient possible use of resources. But we know from the U.S. experience with conglomerates in the 1970s and early ’80s that managers don’t voluntarily distribute excess capital unless they are under great pressure from shareholders‑‑or they are operating under an EVA system.
And, in fact, as Professor Jensen argued in our Journal of Applied Corporate Finance as early as 1989, Japanese companies today seem to be in essentially the same position as U.S. companies found themselves in back in the late 1960s and early 70s. The U.S. firms were then flush with cash and capital from their successes in the ’60s, and they too could not resist the temptation to grow through diversification. And the U.S. firms were then forced by the tough economic environment of the 70s and the shareholder pressures to restructure, to return their excess capital and focus on their core businesses.
And this same free cash flow problem faced by U.S. companies 15 or 20 years ago appears to be one of the major problems in Japanese companies today. Their policies of very low dividends and, until very recently, no stock repurchases means that excess capital has effectively been trapped inside the companies. An EVA system will encourage managesr to find ways to return that excess capital to shareholders, who can then reinvest in more promising growth opportunities in the Japanese economy.
The Link Between EVA and MVA
But I have a confession to make. EVA is not the ultimate measure of corporate success. The most reliable measure of management’s long‑run success in adding value is something we call “Market Value Added,” or MVA. And perhaps the best way to define MVA is to start by telling you what it is not. Contrary to what you may have heard from some economists, the mission of the corporation is not to maximize its market value, or its total market capitalization. Instead, the aim of corporate management should be to maximize the dollar amount by which the company’s market value exceeds the capital supplied by the firm’s investors-‑hence, the name market value added, or MVA.
Although this is a bit of an oversimplification, MVA is essentially the difference between a company’s current market value, as determined by its stock price, and its “economic book value.” A company’s economic book value can be thought of as the amount of capital that shareholders (and, in the technically more correct version of EVA, lenders and all the other capital providers) have committed to the firm throughout its existence, including earnings that have been retained within the business.
To illustrate the calculation of MVA, let’s take the case of General Electric, which was the top U.S. performer at the end of 1994. The total market value of GE’s debt and equity at the time was $101 billion. And since the adjusted book value of that capital (and I will describe some of the adjustments in a moment) was only $46 billion, GE’s market value added amounted to $55 billion. Coca Cola, which ranked number two that year, had only $61 billion in market value. But since Coke’s management had achieved that valuation with just $8 billion of investor capital, its MVA was $53 billion, or only $2 billion less than GE’s. (And I might add that, at the end of 1995, our results put Coca Cola in first place, $10 billion ahead of General Electric.)
But, to see why total market capitalization is a misleading indicator of success, let’s now look at IBM at the same time. At the end of 1994, when IBM’s shares were trading at $48 (since then, of course, the price has more than doubled), IBM’s total market value was $52 billion, or $17 billion below its adjusted book value of $69 billion. Those people who are interested only in market value or size might say that IBM is not doing so bad; after all, it’s only $9 billion behind Coca Cola. But that would have missed the point. The point is that IBM’s investors put in $69 billion, and the market says that investment is now worth only $52 billion. Before Lou Gerstner came on board, IBM’s management succeeded in destroying $17 billion and actually much more, given the premium value the company once commanded.
Now, the calculation of MVA is not quite as simple as it sounds. Calculating a firm’s economic book value, in particular, requires a number of adjustments that I will just mention briefly. At Stern Stewart, for example, we do not use pooling accounting for acquisitions; all acquisitions are treated as purchases–that is, as if they involved outlays of cash. Further unlike the accountants, we do not write off the goodwill that arises from purchase acquisitions of companies for amounts greater than their book value. That goodwill stays on the balance sheet more or less indefinitely, unless there is a real decline of economic value. As an example of another adjustment to conventional accounting statements, we also add back the balance sheet provision for deferred taxes to book equity, because it represents a reserve of cash on which the firm is expected to earn an adequate return. And there are a variety of other adjustments that can be made to get a better estimate of how much investor capital is really tied up in a firm.
Corporate balance sheets, you see, are really designed for creditors; they are meant to provide an estimate not of going‑concern value, but of the value of the firm in the event of liquidation. Take the case of research and development. The accountants write off 100% of R&D in the year it occurs, which distorts conventional accounting statements in two ways: it understates the economic earnings of the company (because R&D is expected to have a payoff down the road) and it understates the amount of capital contributed by investors. For this reason, we recommend to our pharmaceutical company clients that they not expense but instead capitalize R&D-‑that is, put a large portion of it on the balance sheet and add it back to the income statement-‑and then write it off over a period of time that reflects the useful economic life of the investment. For Coca Cola and other consumer products companies, we suggest doing the same with advertising and promotion. We put that on the balance sheet and write if off, though more quickly, of course, than research and development.
I should also mention that MVA is the measure we use for the Stern Stewart Performance 1000 ranking that has been published in Fortune magazine in each of the past four years. This last year we also put together a Stern Stewart ranking of the largest 500 U.K. industrial companies that was run in the Sunday Financial Times. In fact, we have now performed MVA rankings for the public companies in a total of eleven countries around the world. The two most recent examples are the performance rankings for Mexico and Canada that will be published in the next few months.
EVA: The Best Internal Performance Measure
So, if we can calculate MVA, why do we need EVA at all? There are a number of reasons. MVA, for starters, cannot be calculated for privately held firms or non‑profits because they do not have traded shares. And, even for most large publicly traded companies, MVA is a useful performance measure only at the very top of the organization, at the level of the consolidated firm. As you move down into the individual units, there is no measure of MVA (unless some of the firm’s units are also publicly traded).
So the challenge is this: How can we measure the efficiency of an organization that ultimately leads to improving MVA? Or, to put the matter more directly, how does a company increase its MVA?
The short answer is, by increasing its economic value added, or EVA. EVA is the internal measure of year-to‑year corporate operating performance that best reflects the success of companies in adding value to their shareholders’ investment. As such, EVA is strongly related to both the level of MVA at any given time and to changes in MVA over time. EVA is the “residual income” left over from operating profits after the cost of capital has been subtracted. So, for example, a firm with a 10% cost of capital that earns $20 million on $100 million of net assets would have an EVA of $10 million. To the extent a company’s EVA is greater than zero, the firm is creating value for its shareholders.
EVA can also be thought of as a breaking down into “annual installments,” if you will, of the multi‑year Net Present Value that is calculated by using the standard discounted‑cash flow capital budgeting technique. (Technically speaking, MVA is equal to the present value of all future EVAs.) Like NPV, EVA measures the degree to which a firm is successful in earning rates of return that exceed its cost of capital. But EVA is a more useful all‑purpose corporate tool than NPV or DCF, even though both methods properly applied give you the same answer over an extended period of time.
We have done a number of studies to see which measures of performance are most closely linked not with market value, but with market value added‑-this premium value that’s being created by management. Measures like earnings, earnings per share, and earnings growth all have some trivial relationship to MVA. When you bring in the balance sheet as well as the income statement‑-with measures such as ROE and return on net assets; or RONA-‑the significance of the relationship improves a great deal. But the correlation is not anywhere near as strong as what happens when you use EVA. And one reason for the greater strength of that correlation is that EVA, unlike ROE or RONA, corrects for accounting distortions in GAAP income statements and balance sheets, and it specifies a minimum or required rate of return that must be earned on capital employed. To have a positive EVA, your rate of return on capital or net assets must exceed the required rate of return.
I’d like to describe very briefly what happens when you use EVA as a measurement device instead of more conventional measures. A senior manager recently asked me, “Why can’t I use just ROE or RONA as a basis for evaluating and rewarding my people?” I said, “You shouldn’t do that, for two reasons: One, if a company or a division is currently earning substandard returns, managers can increase ROE simply by taking projects with higher, but still inadequate returns. In this case, you would actually be rewarding managers for reducing shareholder value and further reducing MVA.”
At the other extreme, consider a company with a 12% cost of capital that earns a 25% ROE. In such cases, managers might be discouraged from taking on all value‑increasing projects with expected returns below 25% because that will lower their average ROE. For that reason, the firm could be passing up value‑adding investment opportunities.
Neither of these distortions of corporate investment incentives occurs under an EVA framework. In an EVA system, you improve EVA as long as you take on new projects where the rate of return on net assets exceeds the threshold.
In fact, there are three different ways to increase EVA. The first, as just mentioned, is to grow the business by taking on new investments that promise to earn more than the cost of capital. The second way to improve EVA is by not growing the business, but improving efficiency and so increasing returns on existing capital. And the third way is by getting rid of those parts of your business that offer no promise for improvement. EVA encourages managers to engage in a periodic culling of their businesses, discouraging them from wasting more time and money on clear losers.
Conventional corporate reward systems often do not encourage managers to take such steps. By making the lion’s share of managerial compensation take the form of wages and wage‑related items, such compensation plans lead managers to maximize not EVA or shareholder value, but rather their size or market share. So the question is: To what extent can we change this kind of behavior that encourages size or empire‑building at the expense of profitability? You will not succeed in changing behavior simply by changing corporate performance objectives without also changing the compensation scheme. As I said earlier, if you set the performance objective with EVA over here, and you set the incentive compensation on some other basis over there, all your people will bow down to the performance objective and then march off in whatever direction the incentive structure calls for.
So, what do you need to do? Without going into great detail at this point, let me just say that there are four properties for any successful incentive compensation system. If you don’t have these four properties, your plan won’t work.
First, you need objectivity, not subjectivity. Most corporate bonus plans set targets based on negotiations between senior people and their juniors. The result is that the juniors have a tremendous incentive to “low ball” their budgets-‑because their bonus depends just on beating the budget (though not by too much, because that casts doubt on their credibility in setting the budget in the first place). Instead of stimulating managers to stretch, negotiated budgets reward managers for their success in carving out a larger share of the existing pie for themselves, often at the expense of the shareholders.
The second essential quality of an effective compensation plan is simplicity. Incentive compensation should be carried down into the organization at least as far as the level of middle management. And, in fact, many of our clients are extending the plan to cover all salaried employees. Using something we call “EVA drivers,” EVA can in some cases be calculated all the way down to the shop floor. The farther down you go, in my opinion, the more dramatic and durable the benefits seem to be. (It is for this reason, by the way, that an EVA plan is potentially far more effective than the stock options that are given out in many U.S. companies. Such programs make no sense to me at all. They may work at the level of the CEO—and perhaps for a few people who report to that person‑-but below that level they have almost no motivational value. Stock options are, in most cases, give‑away programs.)
Third, the plan has got to be significant. That means that the bonus potentials that come from improving EVA have to be large enough to affect people’s behavior.
Fourth, the plan must be honored; it must not be subject to ex post adjustments. This means that if the board of directors and senior management find some employees doing very well and then very, very well, they must resist the inclination to pull the rug out from underneath them and change the rules. We have to permit people to do very well, provided the shareholders are also doing very well.
There are also three other distinctive features of a Stern Stewart incentive compensation program. One is that there are no caps on the upside. The more EVA people produce, the greater their reward, with no limitation on the size of the bonus awards.
At the same time, as I said earlier, our system has a hold‑back or bonus bank feature. To illustrate, we typically recommend to companies that they hold hostage as much as two-thirds of declared bonuses tied to EVA. And managers will lose the two-thirds held hostage if they don’t at least maintain the level of performance that caused the declaration of the bonus in the first place.
The consequence of this bonus bank feature is to lengthen the managerial decision‑making horizon beyond one year. It’s very important to think of this as an ongoing cycle of value creation with one‑third payable now, two‑thirds later. For a successful manager under this plan, each new year brings steadily increasing payouts, new declarations, and new hold‑backs. And if you do the calculations the way we do them, it takes you about six years to get 90% of your bonus declarations fully paid out.
We also have an unusual stock option program. Without going into too much detail, managers are asked to purchase stock options with a part of each year’s bonus payments. What makes these options unusual is that they have an exercise price that is adjusted each year for the level of the broad stock market or, if management chooses, an industry group of the firm’s primary competitors. If the S&P 500 or industry composite goes up by 10% in a given year, then the exercise price goes up by 10%. This way, shareholders have to get their rewards before the managers participate in any gains.
But consider also what happens if the market or industry goes down, even while the firm-‑or some parts of it‑‑continues to churn out positive EVA. In those circumstances, the EVA bonuses can be used to buy options at a lower price, thus insulating high-performance operating managers from adverse industry or market events beyond their control. If a company’s stock price drops because of uncontrollable market or industry factors, those people who are still creating EVA inside the firm will get their reward by being allowed to buy options with a lower exercise price.
So those are the major issues of corporate performance measurement as I see them. I’m interested not just in the performance measurement itself, but in how it motivates people to behave differently. Talking about increasing shareholder value and setting the right corporate objectives are necessary but not sufficient conditions for corporate success. Managerial incentives must also be changed along with the goals to make the system work.
Let me also mention that the amount of time it typically takes to implement an EVA system, even in very large companies, runs between nine months and a year. This tells me that people inside organizations can adapt to changes. We have special work groups that we send out to our clients, and the mission of our people is to educate our client companies’ trainers so that they can in turn show the company’s managers and employees how this entire system works. It has been a very interesting and edifying experience to see the resulting changes in organizational behavior.
And let me just close with a brief story. One of our clients is among the largest companies in Singapore. But, before we were hired for the assignment, the chief executive officer of the firm said to me, “We don’t know if we should adopt your EVA model, with its emphasis on capital efficiency, or the Japanese model, with its emphasis on growth and market share. What would you do in my position?” I said to him, “I am not going to answer your question. I do not pretend to be able to change the culture of a country. My aim is more modest: I am simply trying to improve the effectiveness of private decision‑making. I am trying to get people to act as if they actually own what they do. The most motivated people in your firm are likely to be the sales force, because they often get paid mainly on commission. They get a percentage of what they do. I want to give the rest of your employees the same thing–not a percentage of turnover, but a percentage of the discretionary value that they create.”
Thank you very much.
KIM: Thank you, Mr. Stern. That brings to an end the first half of our symposium. After a short break, we will hear from three senior financial executives who will discuss their corporate experiences with EVA.
EVA at Eastman Chemical
KIM: The first of our next three speakers is Mr. Virgil Stephens, who is the Senior Vice President and Chief Financial Officer of Eastman Chemical Company. Following the company’s spin‑off from Eastman Kodak in 1994, Mr. Stephens has helped Eastman Chemical to gain notable recognition in the U.S. investment community, in part due to the company’s active and successful implementation of EVA.
STEPHENS: I am honored to be here to speak to you today. I would like to thank the sponsor of this symposium, the Mitsui Life Financial Research Center, for inviting me.
We at Eastman have found the concepts of economic value added, or EVA, to be very useful in evaluating our performance and in determining which opportunities have the best potential to add value to our company. But before I tell you about our EVA journey, I would first like to tell you a little bit about Eastman Chemical Company.
Eastman is a global company with 1995 sales of $5 billion. We manufacture a wide range of chemicals, plastics, and fibers that our customers in turn use in producing thousands of consumer products. As Professor Kim pointed out, prior to 1994 we were a division of Eastman Kodak Company. But, at the beginning of 1994, we were spun off from Kodak and are now a separate, independent, publicly held company.
Eastman Chemical is the world leader in plastics for PET bottles, a major supplier of cellulose acetate fiber, and a leading supplier of raw materials for the coatings industry. We also make many specialty chemicals and plastics. Our products are in many items that you use everyday in your foods, in your toothbrush handles, in the paint on your home, and in automobile parts, electronic components, packaging, cigarette filters, and many others. Moreover, Eastman has had a presence in Japan since 1983. Today we have a sales office here in Tokyo, and a technical service center and research and development office in Osaka.
Why is EVA important to Eastman? You have to be careful about what you measure in large organizations, because what gets measured gets managed. Now I know Japanese companies are experts in quality management tools and measurements. Eastman is the only major chemical company to win the United States Malcolm Baldridge National Quality Award, an award that is comparable to your prestigious Deming Prize. As a quality award‑winning company, we at Eastman realize you simply must measure anything you deem important. So, it is important that you get the right measure.
EVA has been the result of an evolution of thinking that has taken place over the past 40 years or so at Eastman. Years ago I suspect that when people wanted to know how well they were doing, they simply looked at the quantity of products that they sold as a primary indication of performance. How many tons of products have we shipped today versus last year?
But, as the number and kind of products produced by the company increased over time, it became apparent that tons of product shipped was not a good measure. A ton of one product was not necessarily worth the same as one ton of another product. So then sales revenue, along with market share, became the most important measures of how well we were performing.
But, when it became apparent that just sales revenue and market share were not by themselves sufficient to represent performance, the focus turned to accounting earnings. But there turned out to be a problem with earnings, too. If you wanted higher earnings, all you had to do was pour more capital investment into the company. And, as Professor Wakasugi and Mr. Stern have just finished pointing out, too much investment can be equally effective in destroying shareholder value as too little investment.
So, we at Eastman have experienced a natural progression in our performance measures that has brought us to where we are now. Today, we believe that the best measure of how well we are performing is economic value added, or EVA.
At Eastman, moreover, every person in the company is being exposed to EVA concepts. For example, everyone is taught how we calculate EVA. There are two ways of doing the calculation, both of which give you the same answer. One way is to subtract taxes from the operating profits and arrive at net operating profit after taxes, or NOPAT. Then you subtract the company’s charge for capital from NOPAT to get the EVA.
To illustrate this calculation, in 1995 Eastman’s pre‑tax operating profit was some $978 million. Subtracting taxes of $370 million left us with a NOPAT of $608 million. From that $608 million we next subtracted a capital charge of $262 million (our 10% weighted average cost of capital multiplied by our $2.62 billion of total debt and equity), which left us with an EVA of $346 million.
Another way to calculate EVA is to take the difference between our return on capital (which in 1995 was 23.2%) and the cost of capital (100/6), and then multiply that 13.2% by our $2.6 billion of total capital. Doing it this way you come to the same answer, $346 million. Either calculation is straightforward and fairly simple–and that is a big part of the appeal of EVA. Because the calculation is so straightforward, it is easy for most people in the company to understand. And, in fact, we want all of our people to be able to pick up a financial statement and calculate our EVA.
The Uses of EVA
Now that we have looked at how we calculate EVA, let me describe how we apply the concept in our business.
First of all, EVA is the primary report card for each of our ten major business units that we produce each quarter. The EVA numbers provided in these reports tell us which businesses are creating value and how much, and which business units, if any, are reducing value. And we don’t look just at the absolute numbers, we examine the trend over time to see if progress is being made. We look at whether our businesses with negative EVA are making improvement and moving toward the positive side-‑and whether those on the positive side are at least sustaining their levels of EVA. These quarterly EVA reports, incidentally, are reviewed not only by a management committee, but also by the Board of Directors of the company.
Another use of EVA at Eastman is as the basis for incentive bonuses for all of our 17,500 employees around the world. All employees have 5% of their pay “at risk.” This means they take an initial 5% pay cut, but can earn back that 5% plus a bonus that is based on the size of the spread between our company‑wide return on capital and our cost of capital. This payout for all 17,500 employees can vary from zero to as much as 30% of annual salary.
In 1995, for example, we had a very good year, and all our employees received the top award of 30%. The first 5% of that 30% goes automatically into an employee stock ownership plan, thus making all employees shareholders of Eastman.
Tying the EVA measure to employees’ compensation and making them shareholders in the company have been important steps in educating our employees. You can imagine the increase in the level of interest when employees have this kind of earnings potential. If you could overhear the conversations that now take place in our company laboratories, on the plant floor, in offices, you would often hear lively discussions about how certain decisions or practices are expected to affect Eastman’s return on capital.
For our 600 or so higher‑level managers, we use the absolute dollars of EVA earned by the entire company as the basis for incentive pay. These managers have much more than 5% of their pay at risk; indeed, our most senior managers have as much as 40% of their pay at risk. And, last year, the variable pay at Eastman for the senior management turned out to be greater than the fixed pay.
EVA is also an important tool for analyzing our investment and divestiture opportunities. For example, we recently acquired two small companies. Our analysis of Eastman’s ability to create value with those companies in terms of EVA was a major factor in our decisions to make those acquisitions. Likewise, we recently divested ourselves of three businesses. Those businesses were not earning the cost of capital and were not likely to do so anytime in the foreseeable future. So our management team determined that the capital that those businesses were requiring could be put to much better use in other investments.
Now, what are the potential drawbacks of using EVA in your company? First, let me point out that when I say “drawbacks,” I really mean “concerns” or “cautions.” When properly used, there are no serious drawbacks, nothing that should keep you from using EVA.
One word of caution is that, if misunderstood or used improperly, EVA can create an excessively shortterm focus-‑similar to what happens if you look at year‑to‑year profits or cash flow. Even in an EVA system, if you really wanted to make one year’s numbers look good, you would be tempted to neglect such things as training for your employees or maintaining your equipment. Of course, if you neglect those things very long you won’t have much of a company left. So you must discipline yourself, and provide the appropriate incentives to make these types of investments for long‑term success.
EVA may also not be the best measure to use in evaluating a startup or new business. In the first years of such businesses, you are bound to be investing more than you are getting out of them. So what you want to look for with new businesses is improvement over time. You want to set up an agreed‑upon timetable with the operating manager that will allow you to monitor the progress of the business in creating enough value to make up for the first lean years.
Now, let me come back to the benefits Eastman has experienced by using EVA.
First of all, EVA has changed the environment in which we make management decisions. By giving managers and employees incentives to use capital and assets as efficiently as possible, we find that we no longer have to dictate decision‑making from the top of the organization. And the greater decentralization of decisionmaking allowed by EVA has in turn increased our flexibility in ways that can add value.
Let me offer an example. Given the recent popularity of “just‑in‑time” inventory systems, many companies have instructed all their businesses to reduce their inventories, Our experience under EVA, however, has shown us that some businesses may find that the best way to increase EVA is to make sure they never deplete their inventory; and so they may choose to increase their capital investment in inventory. Others of our businesses have found just the opposite‑‑‑that the best way to increase their EVA is by reducing capital invested in inventory.
The point of my story is this: The optimal decision may differ from business to business, and the best tradeoffs are likely to be made where the greatest knowledge exists. In our case, a blanket corporate‑wide decision to reduce inventories would probably have ended up reducing overall value. Our corporate charge to the business units is simply to grow EVA. How they do so depends on their strategy and their plans.
A second important benefit of EVA is that it has changed the behavior of our managers and employees in important ways‑and at all levels of the company. We have more people thinking about return on capital in their decision‑making process than we have ever had before. And EVA has been critical in aligning employee interests with those of our outside shareholders. Our employees now understand very clearly that if Eastman creates value for shareholders outside the company, those who are shareowners inside the company will share in the rewards. EVA has been integrated into the company’s overall metric system so that everyone can tie into it.
KIM: Thank you very much Mr. Stephens. Our next speaker is Mr. Toru Mochizuki, Director and Vice President of Coca‑Cola Japan. Mr. Mochizuki is a life‑long employee of Coca‑Cola and has had operating experience in both Japan and the United States, where he worked in the finance division. Besides describing Coca‑Cola’s successful experience with EVA, he will also discuss some of issues involved in internal training programs.
EVA at Coca Cola Japan
MOCHIZUIU: Thank you, Professor Kim. I have been working for a very long time with Coca‑Cola Japan. We are a fully owned subsidiary of the Coca‑Cola Company, and what I am about to say reflects the thinking of not only Coca‑Cola Japan, but the entire Coca‑Cola Company.
Coca‑Cola Company is widely regarded as the first company to use the concept of EVA. In our company, all major decision‑making is guided by EVA. According to a recent Fortune Magazine article, CocaCola is also the “most admired” company in the United States. I for one believe that these two distinctions are not wholly unrelated,
But let me be a little bit more specific. The people working for the Coca‑Cola Company have a very clearcut mission. In February of 1994, the Coca‑Cola Company issued a statement to all of its employees that was entitled, “Our Mission and Commitment. ” Under the title was written the following words: ” We exist to create value for our sbareowners on a long term basis by building a business that enhances the Coca‑Cola Company’s trademark.”
The key insight of EVA is that, to create value for our shareholders, the company must generate returns in excess of the cost of capital in all of the business segments. But this does not mean that we take a shortterm view of profitability, While we pay close attention to current EVA, we also look at all our businesses over a long‑term time horizon as well. Our financial goal is to increase our EVA steadily over time.
My definition of EVA, I should start by pointing out, is a little different from that of the previous speakers. At Coca‑Cola we refer to net operating income minus capital charges as “economic profit.” EVA, in our definition, is the period‑by‑period increase in economic profit. So, what we mean by EVA is roughly equivalent to what Mr. Stem called the “improvement” or year‑to‑year increase in EVA.
And we at Coca‑Cola have worked very hard to increase our economic profit. For example, in 1981 CocaCola’s economic profit was just under $100 million. Thirteen years later, in 1994, we were able to achieve an economic profit of $2 billion. So, over this 13‑year period, our annual average growth in economic profit was 26%. During the same period, our stock price increased from $2.90 (adjusted for splits) in 1981 to $51.50 at the end of 1994, for an annual rate of increase of 25%. (Moreover, in the past 16 months, our stock price has risen another 60% to over $80, accompanied by a large further increase in economic profit.) So, judging from our case alone, EVA and stock prices appear to be highly correlated. To create economic profit and to increase EVA are the means by which we create additional shareholder value.
At the Coca‑Cola Company we also have a “Share‑owner Value Model” that shows how various “drivers” of performance contribute to higher economic profit. According to this model, there are three principal ways of increasing shareowner’s value. One is by maximizing the rate of return in existing businesses, say, through increases in operating efficiency. The second is by continuously reinvesting in both existing businesses and promising new opportunities. The third is by minimizing our cost of capital.
Let me say a little more about each of these three factors. The first is fairly self‑explanatory. It says that existing operations should be nin as efficiently as possible so as to maximize the firm’s investment on its capital currently in place. The second guides the firm’s investment policy; it says by all means reinvest in the business, or invest in new businesses, but only when the investments are expected to generate returns at least equal to the cost of capital. The second rule also tells us to find ways to reduce operating capital‑by eliminating assets, if necessary, that do not meet the costof‑capital standard. And the third rule instructs us to manage our capital structure‑in particular, by using dividends and share repurchases to pay out excess cash and capital‑so as to minimize our cost of capital.
Let me illustrate our investment rule with an example. Last year, the Coca‑Cola National Sales Company of Japan was established through an investment by each of the 17 independent Coca‑Cola bottlers in Japan. (Nine of these independent Coca Cola bottlers, by the way, are publicly traded companies.) The Japanese market is changing very rapidly, and we came up with the idea of a national sales company to improve the ability of our bottling company partners to respond to an environment of increasing change. We have also invested heavily in vending machines, which has also contributed greatly to the success of our bottlers.
Like most companies, we focus on increasing sales volume and market share as a means of increasing profitability. But we also pay close attention to the efficiency of business operations. And the focus on EVA encourages us to eliminate underutilized assets and excess capital whenever possible. As one example of this rationalization process, over the past four years Coca‑Cola Japan has consolidated three bottling plants into one and, in so doing, substantially reduced fixed costs.
The reduction of operating capital can also be achieved through better inventory control. With the aid of computers, we have developed an advanced inventory management system that not only controls our inventory, but provides valuable marketing information as well. And, as we approach the year 2000, we are continuing to invest heavily in upgrading this and other information systems.
Now, having explained some of the efforts we have been making to increase EVA, let me talk briefly about what is involved in translating the theory behind an EVA system into effective changes in corporate actions and employee behavior. The key to the power of EVA to produce results is its ability to change the mindset of the employees. It is a question of getting the most out of the company’s human resources.
At Coca‑Cola, we have extensive employee training programs. For nonfinancial managers unfamiliar with accounting and finance, we have a two‑day intensive course covering basic principles of financial statement analysis and providing instruction in how to apply EVA in day‑to‑day operation. As is likely true of most’ companies, marketing and technical people seem to have an aversion to accounting or financial matters. But we feel that all of the people connected to the Coca‑Cola Company must incorporate EVA into their daily activities. And this is the purpose of the training program.
For newly hired finance staff, we also have a more specialized three‑tofour‑day program focusing on EVA. And, in accordance with the concept of cross‑functional teams, we have also redesigned our organizational structure in ways that spread accounting and financial expertise throughout the company. The responsibility and role of these financial people is to help the marketing and operating people in implementing EVA. As a result of this move to cross‑functional teams, our marketing and other non‑finance people have become much more interested in and knowledgeable about the balance sheet, inventory management, and operating efficiency. In short, a greater EVA awareness seems to be spreading throughout the company.
One major reason why our training sessions are so effective is because our managers and employees know that their performance evaluations and bonuses will be determined in large part by EVA. At the beginning of every year, each of Coca‑Cola’s business units set performance objectives expressed in terms of EVA. And it’s not only the business units that set goals, but also individual departments within those units, and individual managers within the departments. For middle managers and above, EVA is a major factor in determining incentive bonuses. But it is not the only factor. In Coca‑Cola Japan, for example, we are rewarded for increased sales volume and increased net profit as well as increased EVA.
And let me offer one final comment in closing. In yesterday’s newspaper, I saw an article saying that there is now a movement in the U.S. away from an exclusive focus on shareholders and toward greater profitsharing with employees. At CocaCola, we already have such a system in EVA‑and we have an employees’ stock ownership program as well. Both of these programs have given our employees a feeling of participating in the decision‑making and success of the company. In fact, CocaCola may be thought of as a ‘Japanese” as well as a U.S. company in the sense that employees really do feel that the company‑along with its commitment to increasing EVA‑is their own.
Thank you very much.
KIM: Thank you, Mr. Mochizuki. Our last speaker will be Mark Newburg, Senior Managing Director and CFO of NCR Japan. Mr. Newburg was formerly employed as the CFO of AT&T Japan until the break up of AT&T. He has lived and worked in Tokyo and Hong Kong continuously since 1989. He will present AT&T’s experience with EVA, and focus particularly on the issues of implementation and financial planning.
The Case of AT&T Japan
NEWBURG: Thank you. As the final speaker, I will skip the fine points of the theory and focus in somewhat more detail on what was done at AT&T to put EVA to work, As Mr. Stem said earlier, companies have gone from nothing to having EVA in a period of nine to twelve months. This was about the amount of time it took us to install an EVA system at AT&T in 1992. In 1993, we began to report and use EVA for AT&T’s worldwide organization.
As Mr. Stem also pointed out, measuring and reporting EVA for the company is one thing, but driving it down into the organization to change behavior is much more difficult. To get EVA to work effectively, you have to give the people who are making day‑to‑day decisions a tool they can use. You also have to allow them some time to experiment with that tool to see how it works. And, besides lots of training and support, an effective EVA system must also be linked to incentive compensation. If you don’t tie compensation to EVA, people’s behavior is not going to change.
Our first step in the implementation process was to set up six different implementation teams. The “deployment” team had the responsibility for overall management of the process. They met weekly to ensure that all the other five teams were on trackbecause once again, we were trying to implement the program in a ninemonth period. In order to do that you have to stay on top of the process all the time. You have to set up clear milestones and then try very hard to meet them.
The second team was called “training and communication,” and their job was to educate managers and employees so they understand exactly what the tool is and how it can be used to improve performance. That kind of information must be communicated clearly and continuously to all levels of the organization.
The third team is the methodology team. It would be an extraordinary event to get everything right the first time out, so this team’s responsibility was to continuously monitor and refine the process.
Fourth was the “tax” team, which was made up of corporate tax people. Because EVA requires that you look at your results on an after‑tax basis, you need to push some knowledge about taxes down to the operating levels.
Fifth was the “compensation” team. As I said, you have to link EVA performance to compensation for an EVA system to work. Figuring out the best way to make that linkage was their responsibility.
Sixth, and finally, was the “shadowing” team. Their job is to develop systems for monthly reporting of EVA for each AT&T unit around the world.
As I mentioned, these implementation teams were set up in early 1992. In May of that year, the concept of EVA was incorporated into AT&T’s planning process for the following year. (We had to incorporate the EVA implementation into the May plan of 1992 because we wanted it to have an impact on 1993 compensation.)
Beginning in June of 1992, managers of AT&T’s were given an EVA tool to be used to “self‑monitor” the effects of their own decisions on a day‑today basis to see whether they were creating positive economic value. So, for approximately six months in 1992, we said, “OK, let’s pretend we are now evaluating your performance based on your economic value added to the company.” EVA thus became the baseline for evaluating a proposal to a customer, a decision foran investment, a decision to acquire a company, or a decision to divest a company or sell a division. In short, EVA became the primary criterion for all major decision-making‑-and each unit was assigned someone whose only responsibility was to track this activity. Then, in 1993, after we had already been reporting EVA on a monthly basis for six months, we began linking EVA to compensation.
Each of our six implementation teams remained in place throughout 1993 to take care of questions and problems as they arose. Our training program also continued, both in the finance organization as well as for non‑financial people. We had comprehensive training courses designed to communicate to our people what they have to do to be effective EVA managers. Using a spreadsheet, for example, a manager could simulate the potential effect on both EVA and more familiar perforrnance variables of various decisions across a number of different scenarios. That is, for each of the different scenarios and management decisions, the model would calculate EVA as well as a number of other internal measures that we also pay attention to at AT&T.
I should also mention that this management tool was supported with a 24 hour‑a‑day on‑call EVA advisory service. As I said, we were going around the world with this implementation. And our primary targeted users, once again, were management people, not the finance team.
So, I believe that comprehensive training and support, together with a direct link to compensation, are the keys to success in using EVA. And this brings me to the compensation part of the AT&T story.
Beginning in 1993, the company announced that the AT&T performance award and merit award would be given only in those years in which the company as a whole had positive EVA results. If there was no positive EVA result, there would be no reward. If you were a member of a unit that had a very good year in terms of EVA generation, you had the opportunity to “overachieve.” And since positive EVA for the entire firm was a necessary condition for any awards to take place, it was in everyone’s own interest to make the best decision for the company using EVA as the measurement tool. Whether it was lower‑level managers, directors, or senior managers, they all became accountable for our EVA.
So, what happened at AT&T as a result of moving to EVA and changing the compensation system? What we saw was a dramatic improvement in the condition of the balance sheet and a significant increase in cash flow. EVA prompted people to begin to think what they personally could do to add value to the organization, in no small part because doing so would result in added personal value.
In closing, I feel strongly that a successful EVA roll‑out must have the following three elements: training, communication, and compensation. An EVA program must be accompanied by very detailed training; you have to devote a lot of resources to training as many people as you can. And, along with comprehensive training, you have to communicate the program to everybody over and over again, and using many different vehicles. You have to get the word out so everybody understands it, and everybody feels part of it.Finally, you have to tie the compensation to performance. For no matter how good the communication is, no matter how good the training, it is not going to work without the reinforcement of EVA‑based rewards.
EVA and Employee Welfare
KIM: Thank you, Mr. Newburg. Now we open up the discussion for a brief question‑and‑answer period among
the panelists. And, to get this part of the discussion started, I would like to comment on a statement made byJoel Stem, and then ask him to respond.
Joel, when you were describing your experience in Singapore, you cited an executive who was attempting to choose between the EVA model and the Japanese model of corporate governance. Presumably the EVA model would mean a model that would try to maximize shareholder wealth, while the Japanese model emphasized employee welfare as its highest priority. My question for Mr. Stem is this: Are these two models mutually exclusive? Must maximizing shareholder wealth necessarily mean sacrificing employee welfare? Or can the two goalsbe made to co‑exist and reinforce each other?
I raise the question for the following reason: By making more efficient use of capital, EVA is a system designed to increase the size of the total pie. My thinking is that the larger pie, or larger firm value, that results from increased productivity can then be divided in different ways among the different constituencies, including a larger share for employees. For example, I am pretty sure the panelists representing Coca‑Cola, Eastman Chemical, and AT&T would agree that implementation of EVA designed to maximize shareholdervalue also ended up increasing employee welfare.
So, the point I would like to make is that the EVA model and the Japanese model are not necessarily in contradiction with each other. In fact, because of the chance it offers employees to participate in increases in company value, I would think that the EVA model could easily be adapted to the Japanese corporate culture.
STERN: I agree with your argument, and I think most micro‑economists would agree with it, too, But I would also add to your comment that EVA has the power to transform a corporate culture just by changing employees’ attitude toward their jobs. It has been our experience that a greater sense of accountability, reinforced by monetary incentives, can do wonders for employee morale; they can truly be made to feel like partners in the enterprise.
For example, several years ago we implemented an EVA system in a state‑owned enterprise outside the U.S. And when we did, I told the chief executive that he had to become the champion of the program to make it succeed. He said, “I intend to do that. But I am going to have a big problem with risk. Although our people’s expected rewards are higher, a much larger percentage of their total compensation will also be put at risk. And our people have a different attitude toward risk‑taking than you seem to think they have.”
As things turned out, however, the employees in this state‑owned enterprise outperformed everyone’s expectations, including those of the senior management team, And, for this reason, I am very skeptical of the argument that EVA is culturally bound or conditioned. People everywhere will respond if given the proper training and incentives.
And let me say this: If senior managers in one nation believe that EVA may work in the U.S., but not in their own country, then there is only one logical outcome‑‑ever larger trade barriers to protect those people who have decided that they don’t want to adopt this type of process. Because, as I said earlier, companies in countries like Australia, New Zealand, Germany, and Scandinavia‑and in regions like Latin America and Eastem Europe as well‑are now moving in this direction.
MOCHIZUIU: I think there is suspicion about the EVA system and its possible negative effects on employees. But these concerns can be overcome by education. And when the training is combined with a proper incentive system, employee morale and participation can both be very strong. This has been our experience at Coca‑Cola Japan.
STERN: Last year, I met with Mr. Goizueta, the CEO of the Coca‑Cola Company, and I asked him, “What do you like best about EVA?” He said, “It has developed a winning attitude among our employees.” And he went on to say, “Have you ever noticed that teams that win in sports often continue to win even beyond their initial capability because they believe they can be winners? That is what EVA has done for our people.”
Finally, he said to me, “I also want you to know,joel, that I checked with the people at Equifax here in Atlanta. They have had much the the same success with your EVA program as we have had with ours. They were good before, but now they have a culture that could be called exceptional.”
Is Japan Ready for Pay‑forPerformance?
KIM: In Japan and in many parts of Asia, the group is considered more important than individuals. And the culture may not be as receptive to the idea of individual bonuses based on individual initiative as it is in the United States. So if we have this resistance to merit‑based individual compensation, how would you go about motivating the management and the operating people to change?
MOCHIZUIG: Before I answer that question, let me make a more general comment about managing cultural differences. Although Coca‑Cola Japan is a 100% subsidiary of the U.S. company, virtually all of our managers and employees are Japanese. We have worked very hard to overcome international differences in corporate culture. Our feeling about all of our employees, whether they are Japanese or expatriates, is that they must familiarize themselves with two cultures, and eventually become bilingual as well. And I don’t mean to limit this to just Coca‑Cola Japan and to the Japanese andAmerican cultures. CocaCola’s worldwide philosophy is that it is both desirable and necessary for its overseas employees to master another culture in addition to their own. And this bi‑cultural approach has contributed to the company’s success abroad.
But, having said that, let me also say that there is no debate inside CocaCola about the appropriateness or effectiveness of merit‑based pay in Japan, or anywhere else for that matter. Moreover, I do not believe that EVA has been successful at Coca‑Cola Japan because is it part of an American corporation. Whether we like it or not, it is the competitive changes that are taking place around us that are forcing all companies, including Japanese companies, to establish pay‑for performance incentive plans. The seniority system has existed in Japan for many years. But it is my view that those Japanese companies that continue to adhere to that system will lose out to their competitors. To be rewarded for one’s performance is not only an important source of motivation, it is also the fairest way-‑and most people recognize this. If it is not that way in Japan, it should be that way injapan. That is my personal view.
NEWBURG: There are a couple of comments I would make on this issue of group versus individual performance. One, an EVA system, if properly implemented, has the potential to make an organization more cohesive; it should encourage more than it discourages tearowork. To create this cohesion, we have in our company something we call “visualization stands. ” They are exhibits set up in the company cafeteria, and next to the vending or coffee machines, that show everyone how we are doing compared to our EVA commitment for this year. It helps everyone understand where we are and the work that we all have to do to achieve this year’s goals.
I also think the idea of individual merit pay is coming very rapidly to Japan. I think we see more and more companies setting up pay‑for‑performance systems. And, as I said earlier, EVA gives you a very good tool for rewarding those individuals that add value to the company.
STERN: I would like to make a number of comments on this issue. The first is that I do not view an EVA program as primarily a method for rewarding individual behavior. In most cases, that is not what it is designed to accomplish. There are one or two exceptions to that rule, such as in merchant banking where certain types of individual behavior are critical to the success of the company. But an EVA program is designed to have a measurement system that rewards group, and indeed company‑wide, behavior.
How do we do this? In most cases, we suggest to companies that about 70% of an individual business unit’s incentive award be based on the performance of that unit, with the other 30% based on the performance of other units in the firm at the same level. The reason we do that is to address two issues. One is diversification of risk. If we are not doing very well this year, we could be carried this year by somebody else and we will help them out next year or the year after. The other reason is that if you come up with a very good idea in your unit, we want you to be motivated to share that good idea with people who are quite a distance away from you. If 30% of your award is based on how everybody else is going to do, you are going to develop a communication system that enables other people to benefit from that idea.
KIM: That comment leads nicely into my next‑and final‑question: Namely, how far down in the organization should you go in calculating separate EVAs? In making that decision, you must consider that there are joint costs that are being shared, and there are joint benefits. In the technical jargon, there are synergies that would be lost if the operations were run independently. And, the farther down in the organization you go, the greater is this problem of joint costs and the shared benefits.
So, my question is: How do you trade off the benefits of decentralized decision‑making, and better defined accountability and incentives, against the costs arising from the potential loss of coordination among different EVA centers? And, if you do decide to push EVA performance evaluation well down into the organization, how do you encourage people to capture potential synergies within your system?
STEPHENS: As I mentioned earlier, at Eastman we have ten different business units and we calculate EVA for each of those business units. EVA is the report card for each of our businesses, and that is what we want to see continue to grow. But, as far as the compensation system is concemed, that is determined by the performance of the entire company. We have found that the optimal compensation plan is to pay everyone based on the performance of the entire company.
About five years ago we experimented with a 70% total‑company, 30% individual‑business unit allocation. But we found that we got suboptimal behavior under that system; people were not acting in ways that allowed us to take advantage of potential synergies among our different operations. All of our businesses are in the chemical industry, and so there are lots of shared costs and potential synergies from a more integrated approach. So we went to a 100% company‑wide compensation system, and that seems to work best for us.
STERN: I think that the issue Professor Kim raised is one of the most important with respect to EVA. I don’t have a definitive or scientific answer for you, but I am going to tell you how we try to overcome the problem that you identified
No two firms have the same EVA system. Each Program is tailor‑made to the individual company’s culture, preferences, history, and the form in which they choose to do business. Every firm that works on this EVA implementation is asked by us to set up a steering committee. The steering committee includes the CEO; he is the chairman of the committe. There is a vice chairman~ usually the head of human resources or the chief financial officer, and then everybody else on the committee is a senior operating executive.
Our program is thus driven by the operating side of the business. So, when issues of cross‑unit coordination and syne~gies come up, we at Stem Stewart are not conducting the orchestra. The steering committee meets once a month over a six‑ to eight‑month period. And, during these meetings, the operating people discuss the problems‑and we listen and offer suggestions.
At these meetings, the issue of shared costs and benefits invariably comes up. By listening to their description of how their businesses interact, we help guide them in making the trade‑offs and in deciding how far down in the organization to drive the measurement system.
And our recommendations will generally be based on how they really manage the business‑as opposed to what their annual report says. When you read an ann ‘ ual report that says a company is in the following five businesses, the company may not be managed that way at all. Rather than being managed as five distinct product lines, the company may in fact be divided up along geographic or functional lines‑and so having five different EVA performance centers based on product lines would make no sense, So, the specific design of the EVA program really comes out of this process of interaction with the steering committee.
KIM: How do you deal with such problems, Mr. Mochizuki?
MOCHIZUKI: I think better coordination of activities among business units is one of the most important challenges facing most businesses today. Activity‑based costing systems, or transfer pricing, is one possible solution to the problem. Although there are often difficulties in arriving at the right measurements, the use of ABC or transfer prices can help solve some joint cost or benefit problems inside companies. But perhaps a more promising solution to this problem is greater use of cross‑functional teams that I mentioned earlier. By combining the different skills and experience of people throughout the organization, companies may discover new ways of making the whole company worth more than the sum of the parts. At the very least, our compensation systems must find a way to encouruge‑or at least not discourage‑a fairly fluid organizational structure in which numerous teams can interact and operate across the old boundaries.
KIM: Since it has been a rather long day, let me conclude very quickly, Clearly EVA offers an important decision‑making and performance evaluation criterion. It reminds managers that capital is not free, and that one of their primary responsibilities‑if not indeed their main one‑is to represent the interest of shareholders. In the long run, moreover, corporate employees’are likely to find that their own best interests are closely tied to those of the shareholders.
Nevertheless, the varied experiences of our panelists described today show that there is no single recipe for making EVA part of the management culture. The critical first step is coming to an understanding of the potential benefits of EVA and of how to translate this understanding into a consensus of purpose within the corporation. Taking this step proves to be the greatest challenge for EVA in gaining acceptance by Japanese corporations.
So, let me bring this discussion to a close by thanking our panelists for their excellent and insightful presentations. And let me say, once again, how grateful I am to President Sakata and Dean White for their support of our activities. Last but not least, I want to thank Professor Wakasugi, my fellow co‑director at the Center, for all of his efforts in making this symposium a success. Thank you all very much.
The editor would also like to thank Dan Ebels for his help in editing the transcript of this symposium.