Executive compensation and corporate governance in the U.S.: Perceptions, facts, and challenges.

• Author: Kaplan, Steven N.
• Position: The 2012Martin Feldstein Lecture

In this lecture, I explore some commonly held perceptions of executive compensation and corporate governance in the United States: 1) CEOs are overpaid and their pay keeps increasing; 2) CEOs are not paid for performance; and 3) corporate boards are not doing their jobs. For example, Bebchuk and Fried have concluded that, "flawed compensation arrangements have not been limited to a small number of 'bad apples'; they have been widespread, persistent, and systemic." (1) I consider the accuracy of these perceptions today, and discuss the implications and challenges that the evidence poses for researchers, boards, and shareholders. (2)

[ILLUSTRATION OMITTED]

How is pay measured?

There are two ways to measure CEO pay. The first is estimated or grant-date pay. This includes the CEO's salary, bonus, the value of restricted stock, and the estimated value of options issued that year. This is the compensation the board awards the CEO and, therefore, the appropriate measure for board governance effectiveness. The second measure is realized pay. This includes the CEO's salary, bonus, the value of restricted stock, and the value of options exercised that year. Because it uses actual option gains (not estimated Vaues), this better measures what the CEO actually takes home. Accordingly, realized pay is appropriate for considering whether CEOs are paid for firm performance.

Facts about pay

Using estimated pay, I look at data from 1993 to 2010 for S&P 500 companies (from S&P's ExecuComp database). What has happened to average estimated CEO pay (adjusted for inflation) since 2000? Most audiences believe it has increased substantially. In fact, Figure 1 (on page 3) shows that while average CEO pay increased markedly from 1993 to 2000, it declined by over 46 percent from 2000 to 2010. Median CEO pay also increased from 1993 to 2000, but has since declined. The convergence between the means and medians suggests that boards have become less likely to award large pay packages since 2000.

There are still some outliers that receive attention and likely drive the perception that pay has increased. For example, three CEOs received over $50 million in estimated pay in 2010. The means and medians indicate that these are outliers and not the general rule.

ExecuComp also follows the CEOs of over 1,000 smaller companies not in the S&P 500. Average estimated pay for these CEOs, like S&P 500 CEOs, increased in the 1990s and declined in the 2000s. Today's average pay roughly equals its 1998 level.

Overall, then, estimated CEO pay -- what boards expect to pay their CEOs -- peaked around 2000, both for S&P 500 and non-S&P 500 CEOs. Since then, average estimated CEO pay has declined, returning roughly to its 1998 level.

While average pay has declined since 2000, it remains very high in absolute terms. In 2010, the average S&P 500 CEO received estimated pay of just over $10 million. This is roughly 200 times the median household income in the United States and undoubtedly also contributes to the perception that CEOs are overpaid.

Turnover

The average lengths of CEO tenures today are shorter than in the past. As a result, comparing CEO pay in the 2000s to CEO pay in the 1990s (and earlier) is not an apples-toapples comparison. In the 1970s, 1980s, and mid-1990s, roughly 10 percent of large U.S. company CEOs turned over each year, not counting takeovers. (3) Since 1998, annual turnover has increased to an average of 12 percent, implying a decline in CEO tenure from ten to eight years. Including takeovers, tenures have declined from roughly eight years before 1998 to only six years since.

[FIGURE 1 OMITTED]

The decline in tenure implies that the CEO's job has become riskier over time. The shorter expected tenure arguably offsets roughly 20 percent of the increase in CEO pay since the early 1990s.4 The true increase in CEO pay since then is lower than the compensation figures alone would suggest.

How does CEO pay compare to that of other highly paid people ?

Gabaix and Landier (5) argue that market forces can explain the increases in CEO pay. Using a simple competitive model, they show that CEO pay will rise as firms become larger because larger average firm size increases the returns to hiring more productive CEOs. They find empirically that the increase in CEO pay since 1980 can be fully attributed to the increase...