August 4, 1997

Debate: Duking It Out over EVA Gary Hamel; Al Ehrbar

In a recent FORTUNE story ("How Killers Count," June 23), corporate strategy guru Gary Hamel argued that EVA, or economic value added (which measures how much a company's profits exceed its cost of capital), isn't an adequate way to measure wealth creation. Al Ehrbar, a senior vice president of Stern Stewart, the consulting firm that coined the term "EVA," strongly disagrees with Hamel's opinion. Here's his argument and Hamel's rebuttal.

EHRBAR: We won't quarrel with Hamel's pronouncements on strategy (that's his specialty, not ours), but he betrays a woeful misunderstanding of EVA and of how you determine whether a corporation is creating wealth. Dismissing EVA ("kindergarten" is his word), Hamel argues that a better way ("graduate school") of determining whether a company is winning is to look at what happens to its percentage share of the total market value in its industry. If the market value of an industry increases and a company's share grows larger, it is a superior wealth creator. But consider what happened in the U.S. auto industry from the end of 1986 to the end of 1995. Over that period, Ford's share of the combined market value of the Big Three jumped 8.7 percentage points to 28.5%, and its own market capitalization rose by $29.5 billion. Pretty good, right? Wrong. In those same years, Ford added $36 billion to its capital base so that, on balance, it destroyed $6.5 billion of shareholder wealth. By Hamel's arithmetic, Chrysler didn't do as well as Ford. Its share of the industry's market value rose 6.6 points to a paltry 14.5%. Yet Chrysler created $2.8 billion of shareholder wealth over the period. Hamel also is wrong when he asserts that "no one measure can capture all the dynamics of corporate performance...." There is just such a measure, and it is called MVA, for market value added. MVA is the market value of a company's debt and equity, minus all the capital that lenders and shareholders have contributed over time. In other words, MVA is the difference between cash in (what investors have contributed) and cash out (what they could sell their claims for today). EVA, in contrast, isn't a measure of wealth creation and doesn't purport to be one. EVA is a measure of the amount by which profits exceed or fall short of the cost of capital in any one period. Yet while EVA doesn't measure wealth directly, it is an invaluable guide for management because it correlates better than any other measure with changes in MVA. Remember, it isn't the level of EVA per se that really matters. What counts most is changes in EVA: Continuous increases are rewarded with increases in MVA, while declining EVA is punished with declining MVA. Hamel's failure to understand this crucial point probably accounts for his complaint that a company can be earning more than its cost of capital while getting clobbered by competitors. His prime example is IBM, which had a handsome 17% return on capital last year and yet has suffered mightily over the last decade. So what? The real question is what happened to IBM's EVA ten years ago: it plummeted. As the 1996 return on capital suggests, IBM's EVA has now advanced smartly under Louis Gerstner. And Big Blue's shareholders aren't complaining.

HAMEL REPLIES: Mr. Ehrbar agrees with my argument in all its essential elements. We both agree that "EVA...isn't a measure of wealth creation." We also agree that, in the long term, a company cannot raise its market capitalization if it doesn't have a positive EVA. And MVA, like EVA, is merely a measure of capital efficiency. But these accounting tools reveal nothing about a company's relative capacity to create new wealth within its industry. In Silicon Valley today, several semiconductor companies have a positive EVA. But as an investor, my money has been on Intel. For Intel, besides earning its cost of capital, has been capturing the lion's share of the new wealth created within its industry. Let me build on this. To simply say that IBM had a negative EVA in the late 1980s and early 1990s is to seriously understate the extent to which IBM missed out on opportunities to create new wealth in one of the fastest-growing industries in the world. If IBM had maintained its share of market capitalization in its industry over the past decade, its current stock market valuation would amount to nearly $210 billion, as opposed to its actual valuation of about $96 billion. It was, by and large, new and more revolutionary competitors that captured that $140 billion difference. (The market cap of Microsoft alone is now $155 billion). Further, it would be ludicrous to suggest that IBM's most substantial problem over the last decade was a lack of capital efficiency. IBM had a time-to-market problem, a complacency problem, and more besides. So if the goal is to create new wealth, earning one's cost of capital is merely kindergarten. With respect to Ford and Chrysler, my calculations reveal a different picture. A definition of the auto industry broader than the Big Three shows that, while Ford's share of market capitalization fell from 36.1% to 30.5% between 1989 and early 1997, Chrysler's share of market capitalization increased from 8.8% to 19.1%. I'm sure Chrysler strives for a positive EVA, but thankfully its ambitions go far beyond this. It was innovation--from minivans to macho trucks to cab-forward design--that drove the majority of Chrysler's wealth creation, not capital efficiency. Finally, in the emerging knowledge economy, capital efficiency is even less a wealth driver. In the industrial economy, capital was everything. In the knowledge economy, it often means literally nothing, especially to companies like Microsoft and Amgen whose assets walk out the door every night. Look beyond cost of capital. The goal should be to build new markets and change industry rules.