read article and answer question

Reflections on CEO Compensation by John C. Bogle

W ith all due respect—and with the benefit of hindsight—I take issue with the positive appraisal of executive compensation ex-

pressed by University of Chicago professor Steven N. Kaplan. In his testimony before the Committee on Financial Services of the U.S. House of Rep- resentatives, Kaplan makes three principal argu- ments in defense of executive compensation. I’ll take them in order, and make my disagreements clear.

Kaplan Argument 1: The U.S. Economy Has Done Well

F irst, Kaplan argues that “during the past 15 years, the period in which CEO pay has been criticized, the U.S. economy has done ex-

tremely well.” Of course that’s true. But corporate profits have, over time, grown at about the rate our economy has grown—no more, no less. So it’s hard to see that the CEOs of these corporations, as a group, have added much value. My own conclu- sion on the subject was expressed in my book The Battle for the Soul of Capitalism.1

In 1980, the compensation of the average chief ex- ecutive officer was forty-two times that of the average worker; by the year 2004, the ratio had soared to 280 times that of the average worker (down from an astonishing 531 times at the peak in 2000). Over the past quarter-century, as Table 1 shows, CEO com- pensation measured in current dollars rose nearly sixteen times over, while the compensation of the

average worker slightly more than doubled. Measured in real (1980) dollars, however, the compensation of the average worker rose just 0.3% per year, barely enough to maintain his or her standard of living. Yet CEO compensation rose at a rate of 8.5 percent annually, increasing by more than seven times in real terms during the period.

The rationale was that these executives had “cre- ated wealth” for their shareholders. But were CEOs actually creating value commensurate with this huge increase in compensation? Certainly the average CEO was not. During that twenty-four-year period, corporations had projected their earnings growth at an average annual rate of 111⁄2 percent. But they actually delivered growth of 6 percent per year— only half of their goal, and even less than the 6.2 percent nominal growth rate of the economy. In real terms, profits grew at an annual rate of just 2.9 percent, compared to 3.1 percent for our nation’s economy, as represented by the Gross Domestic Product. How that somewhat dispiriting lag can drive average CEO compensation to a cool $9.8 million in 2004 is one of the great anomalies of the age.

To be sure, during the 15-year period selected by Kaplan, corporate profits have grown at a much faster rate than our economy. From 1992 through 2007, earnings of the large companies represented in the Standard & Poor’s 500 Stock Index have grown at an 8.7% rate, compared to a 5.4% growth rate for U.S. GDP. But that relationship appears to be period-dependent. For example, in the decade since 1997, S&P earnings have grown at a 5.2% nominal rate, actually a hair short of the GDP growth rate of 5.3%. Yet that is the era in which CEO compensation went through the roof.

The fact is that corporate profits have consti- tuted a relatively stable percentage of our econ- omy. Figure 1 shows that since 1940, aggregate corporate profits have, on average, constituted some 6% of our GDP, ranging from lows of about 4% to highs of about 8%.

Further, while corporate profits reached 8.2% of GDP in the first three quarters of 2007 (the

This article is in response to Steven N. Kaplan’s Congressional testi- mony on empowering shareholders on executive compensation and H.R. 1257, the Shareholder Vote on Executive Compensation Act before the Committee on Financial Services of the United States House of Represen- tatives, on March 8, 2007. The ideas expressed in Mr. Kaplan’s article in this issue are based largely on this testimony. Due to timing issues, we asked Mr. Bogle to respond to Mr. Kaplan’s testimony, not to the article that appears in this issue. To access the full testimony, visit www.house.gov/ apps/list/hearing/financialsvcs_dem/htkaplan030807.pdf.

Note: The opinions expressed in this article do not necessarily repre- sent the views of Vanguard’s present management.

1 Yale University Press, 2005.

John C. Bogle ([email protected]) is founder and former CEO, The Vanguard Group.

2008 21Bogle

Copyright by the Academy of Management; all rights reserved. Contents may not be copied, e-mailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download, or e-mail articles for individual use only.

most recent data available), the decline in profits in the fourth quarter, carrying over into 2008, will result in a reversion toward the long-term mean of roughly 6% that is evidenced in the chart. Over time, then, our corporations as a group have pro- vided a steady share of GDP, not a growing share that might justify some rise in CEO compensa- tion—to say nothing of the leap from 42 times the average worker’s compensation to 280 times re- flected in the earlier table. (Interestingly, my fig- ure of $9.8 million of compensation for the aver- age CEO in 2004 is actually lower than Kaplan’s figure of $11.9 million.)

Kaplan Argument 2: CEOs Are Not Alone in the Compensation Explosion

I f CEOs are, as a group, average, what explains their soaring pay scales? Kaplan’s position is that the rise in CEO compensation “appears to be

largely driven by market forces. . . . The pay of other talented and fortunate groups . . . for exam-

ple, hedge fund investors [sic] . . . and professional baseball, basketball, and football players . . . [has] increased by at least as much.” Here’s how I re- spond in The Battle:

It is said that soaring CEO compensation was in part a reflection of the enormous, and increasingly public, compensation paid to star athletes, entertainment personalities, and movie stars. Such comparisons are absurd. These celebrities are essentially paid by their fans or the owners of teams or networks out of their own pockets. CEOs are paid by directors, not out of their own pockets but with other people’s money, a clear example of the “agency problem” in our invest- ment system.

The fact is that corporate shareholders have played little, if any, role in setting executive pay. In 1949, Benjamin Graham (see Graham, 2005) got this issue exactly right. He observed that in terms of legal rights, “the stockholders as a class are king. Acting as a majority they can hire and fire managements and bend them completely to their will” (pp. 207–208). But in terms of the

Table 1 Annual Salaries, the Average CEO and the Average Worker

Current Dollars 1980 Dollars CEO Worker CEO Worker

1980 $625,000 $14,900 $625,000 $14,000 2004 $9,840,000 $35,000 $4,500,000 $15,900 Total increase 1,147% 136% 614% 7% Annual rate 12.2% 3.6% 8.5% 0.3%

Sources: John A. Byrne, “Executive Pay: The Party Ain’t Over Yet,” Business Week, April 26, 1993; Claudia H. Deutsch, “My Big Fat C.E.O. Paycheck,” New York Times, April 3, 2005; and author’s estimate.

Figure 1 After-Tax Corporate Profits* as a Share of GDP

*With inventory valuation and capital consumption adjustment.

22 MayAcademy of Management Perspectives

assertion of these rights in practice, “stockholders are a complete washout. . . . [T]hey show neither intelligence nor alertness . . . and vote in sheep- like fashion for whatever the management recom- mends no matter how poor the management’s record of accomplishment may be” (pp. 207–208). That was true when he wrote those words in 1949; it is no less true today.

This agency problem permeates corporate gov- ernance and bears a major responsibility for the rise in CEO compensation. Three primary factors have driven the growing lack of accountability of corporate boards to shareholders:

1. The indifference of the institutional money managers (themselves highly paid), who in the aggregate now hold effective voting control of corporate America. (The 100 largest managers alone now own 58% of all stocks.)

2. The conflicts of interest faced by these manag- ers, in which their fiduciary interest in repre- senting the mutual fund shareholders and pen- sion beneficiaries they are duty-bound to serve seems to have been overwhelmed by their fi- nancial interest in gathering and managing the assets of these mutual funds and pension funds. (Managers, unsurprisingly, don’t wish to offend their large corporate clients.)

3. The fact that most institutional shareholders no longer practice long-term investing (which logically demands attention to corporate gov- ernance issues). They have turned instead to short-term speculation, in which they hold corporate shares for an average of a year or less (which logically leads to indifference about governance issues.)

Kaplan Argument 3: Higher Compensation Is Tied to Higher Returns

I have no reason to doubt Kaplan’s conclusions

(reflected in his Exhibit 13) that CEOs manag- ing companies whose stocks have provided

higher returns have received systematically higher compensation. After all, given the heavy role played by stock options in compensation pack- ages, it would be little short of astonishing were that not the case. But I take issue with the heavy reliance on stock prices as the principal basis of

CEO compensation. The short-term and momen- tary price of a stock, as we must now know, is as illusory as it is precise. CEO performance should be based on the long-term and enduring building of intrinsic value, which is as real as it is imprecise. (Now there’s a paradox.)

Yes, stock prices correlate nicely with business results, but only over the very long run. In the short term, correlation seems random at best. Sim- ply put, stock prices are a flawed measure of cor- porate performance. Prices (using Lord Keynes’s classic formulation) involve both enterprise—the yield on an investment over the long term—and speculation— betting on the psychology of the market.

During the 1980s and 1990s, for example, earn- ings of the corporations in the S&P 500 grew at about 5.9% a year, well below their growth of 7.7% during the previous two decades. Yet stocks performed well during that period, simply because the price-earnings multiple on the S&P 500 soared from 8 times to 32 times, adding about 7.5% per year of speculative return to stock price returns, a nonrecurring event that dwarfed the actual earnings growth rate. This market-emo- tions– based “lottery effect” suggests, at the mini- mum, that options prices should be adjusted to reflect changes in the general level of stock prices.

But basing compensation on increasing the in- trinsic value of business would be a far better way of rewarding executives for durable long-term per- formance. For example, CEO compensation might be based on corporate earnings growth, corporate cash flow (even better, for it is far more difficult to manipulate), and dividend growth, and on return on corporate capital relative to peers and relative to corporations as a group (say, the S&P 500). Such measurements should be taken only over an extended period of time, and only after deducting the corporation’s cost of capital. Of course, those standards are challenging, but that is what real business success is all about.

Much of the responsibility for our flawed sys- tem of CEO compensation, I believe, can be at- tributed to the rise of the compensation consul- tant. First, it must be clear that compensation consultants who consistently recommend lower pay or tougher standards for CEO compensation

2008 23Bogle

will likely not be in business for long. To make matters worse, the well-known methodology of consultants— grouping CEOs into peer groups measured in quartiles—leads inevitably to a ratchet effect.

It has been observed— correctly, I believe— that boards typically fall in love with their CEOs (at least until something big goes wrong). When a board finds that its own CEO’s pay reposes in the fourth quartile, it raises his (or her) compensation to bring it into, say, the second quartile, which, of course, drops another CEO into the fourth quartile. And so the cycle repeats, onward and upward, over the years— almost always with the encouragement of an ostensibly impartial consultant.

Such a methodology is fundamentally flawed and has the obvious effect: The figures in these compensation grids almost always go up—for the group (so far)—and almost never go down. War- ren Buffett (2005) pungently describes the typical consulting firm by naming it, tongue-in-cheek, “Ratchet, Ratchet, and Bingo” (p. 15). Until we pay CEOs on the basis of corporate performance rather than on the basis of corporate peer groups, CEO pay will, almost inevitably, continue on its upward path.

About That Hindsight

N ow here’s where the hindsight comes in. Since the date of Kaplan’s Congressional testimony, astonishing turbulence in our financial mar-

kets has revealed additional serious flaws in com- pensation practices. Recent events have made it clear that incentives that are strictly backward- looking in nature have tended to encourage CEOs to take outsize risks with the corporation’s assets and strategies, the better to maximize reported corporate profits and stock prices during their reign. It now seems obvious that CEO compensa- tion should be spread out over an extended period of time, well beyond the CEO’s term in office. Consider the compensation of three well-publi- cized CEOs, all in the financial sector, during the recent turbulence.

1. Charles Prince, CEO of Citigroup, took office in October 2003, with Citi stock selling at $47.

While the bank did well for a few more years, it created a highly risky investment portfolio that fell to pieces within five years, with write- offs (so far) of some $21 billion. Citi’s earnings fell from $4.25 per share in 2006 to $0.72 for 2007, and the stock, at this writing, is at about $20. Mr. Prince was paid $138 million for his efforts when times were good, but incurred no penalty for the disaster that followed. (Prince resigned on November 4, 2007.)

2. The experience of Stanley O’Neal, CEO of Merrill Lynch, was similar. The risks assumed by the firm in its risk-laden investment portfo- lio came home to roost late in 2007, with $19 billion of write-downs (with more likely to come). Merrill reported net losses for the year of $10.73 per share, and its stock tumbled by more than half, from $95 to $37. Yet Mr. O’Neal’s compensation of $161 million be- tween 2002 and 2007 was not affected, and the retirement plan package he received on his resignation in October 2007 was paid in full by the board—another $160 million, for a total of $321 million.

3. James E. Cayne, CEO of Bear Stearns, was paid some $232 million between 1993 and 2006 as the stock price of this investment banking powerhouse rose from $12 per share to $165. But the firm’s risky and largely illiquid portfo- lio, along with its high leverage (assets of about 35 times capital), brought Bear to the edge of bankruptcy. The Federal Reserve was required to guarantee the value of much of the portfolio before JPMorgan Chase agreed to buy the com- pany for a price of $2 per share, ultimately raised to $10 —measured from the high, a loss of some $25 billion of shareholder capital. But Mr. Cayne’s multimillions of dollars in com- pensation had already been paid. (In fairness, his investment in Bear, once valued at $1 bil- lion, had dropped to $60 million when he sold his shares in March 2008.)

The point, now obvious, is that CEO compensa- tion should have a contingent component. Incen- tive pay should be spread out over an extended period of years, and stock options should be phased in as well—for example, 50% exercisable on the first

24 MayAcademy of Management Perspectives

exercise date, with 10% exercisable annually over the subsequent five years. There should also be, as I wrote in The Battle, “clawback” provisions for returning incentive compensation to the company if earnings are restated. (I had understood that Section 304 of the Sarbanes-Oxley Act provided effective clawback provisions for equity-based ex- ecutive compensation when restatements occur. However, such clawbacks are limited to restate- ments resulting from “misconduct,” and the SEC has yet to pursue a single case.)

Conclusion

T he evidence is that, contrary to Kaplan’s con- clusion, CEO compensation is seriously out of line, and too often has provided excessive and

unreliable lottery-type rewards based on evanes- cent stock prices rather than durable intrinsic

corporate value. While I share Kaplan’s concern that allowing nonbinding shareholder votes on executive compensation “may not generate appre- ciable benefits over the current system,” I con- clude that it is a policy that should be tried. Its implementation costs would be relatively modest, and it would force institutional shareholders to consider compensation issues with greater care. Anything that draws the institutional owners who now control corporate America into acting as responsible corporate citizens should benefit our society at large.

References Buffett, W. (2005). Chairman’s Letter, Berkshire Hathaway

2005 Annual Report. Graham, B. (2005). The intelligent investor. New York: Har-

perCollins. (Original work published in 1949)

2008 25Bogle