Are Corporate CEOs Paid Too Much?

Forming an opinion on whether CEOs are overpaid or not is more a matter of personal philosophy than quantitative reasoning. However, if one is going to weigh in with an opinion, then one should at least be informed with some relevant data. One can then pick the data that best supports his/her philosophy. An economist from the University of Chicago, Steven N. Kaplan, makes the case that while CEOs are highly paid, their level of pay should not be considered outrageous. He has produced a lengthy and detailed paper: Executive Compensation and Corporate Governance in the U.S.: Perceptions, Facts and Challenges. 

Kaplan’s conclusions are summarized in the abstract to his paper:

"In this paper, I consider the evidence for three common perceptions of U.S. public company CEO pay and corporate governance: (1) CEOs are overpaid and their pay keeps increasing; (2) CEOs are not paid for their performance; and (3) boards do not penalize CEOs for poor performance. While average CEO pay increased substantially through the 1990s, it has declined since then. CEO pay levels relative to other highly paid groups today are comparable to their average levels in the early 1990s although they remain above their long-term historical average. The ratio of large-company CEO pay to firm market value is roughly similar to its level in the late-1970s and lower than its pre-1960s levels. These patterns suggest that similar forces, likely technology and scale, have played a meaningful role in driving CEO pay and the pay of others with top incomes. With regard to performance, CEOs are paid for performance and penalized for poor performance. Finally, boards do monitor CEOs. The rate of CEO turnover has increased in the 2000s compared to the 1980s and 1990s, and is significantly tied to poor stock performance."

Compensation of CEOs comes in various forms. Stock options, for example, vary in value with time. Kaplan favors tallying the value of options at the time issued as the best representation of the intent of the board in its compensation decision. He refers to this as an "estimated" value (as opposed to a "realized" value). Kaplan provides data on large corporation CEOs as well as that for smaller corporations. The CEOs of large entities are more generously compensated than the others and we will focus on them. Kaplan acquires data from various sources and these large corporations are sometimes tallied via membership in the S&P 500 or some other stock index, or simply by rank in size.

This chart provides the most basic data with respect to compensation.

Pay grew rapidly in the 1990s as the stock markets and the economy grew at a rapid pace. Since the dot-com bubble burst, CEO pay has been relatively constant. Note also that the spike in pay around 2000 shows up mainly in the average, not the median data. This indicates that the data was being pulled up by those at the extreme of the compensation distribution who were gaining income much faster than most CEOs. It is the extremes that make it into the popular media.

This chart compares CEO pay to median household pay over the same time interval.

This is where the philosophy enters. Are these levels too high compared to wage earners? Again, this indicates a rather flat level of pay for CEOs relative to the general population over the past decade.

Kaplan suggests that it is useful to consider CEO compensation relative to the degree of responsibility they bear. He chooses company market value as an indication of responsibility.

When CEO compensation is normalized to the market value of the companies they lead, big spikes over time tend to damp out. The level of compensation rose in this representation from around 1970 to 1990, but after that it has been relatively flat. It is interesting to note that by this measure, CEOs were much more highly compensated in 1936-1950. What this chart indicates is that the highest paid CEOs tend to be the CEOs of the largest companies. Is that unexpected or unreasonable?

Kaplan claims that the trend in CEO compensation is representative of the trends taking place for high income earners in general.

The large spike in pay around 2000 is considerably dampened when normalized to the income of others in the top 0.1%.

Kaplan compares corporate CEOs to other professional classifications that rise to that level. The most interesting comparison was to high-performing professional athletes.

Top athletes in football, baseball, and basketball have been more generously compensated than the CEOs of top corporations over the past decade, and the difference continues to grow. Is that unexpected or unreasonable?

The argument is made that professional athletes require large pay packages because their careers are so short and subject unforeseeable events such as injury. Is there any correlation with the career of a CEO? Is it also risky, potentially short, and highly dependent on performance? Kaplan would probably claim that such an analogy could be made. In order to indicate the degree to which performance leaves CEOs at risk for termination, Kaplan provides this data:

Note that the data refers to turnover in a five-year period, and that the first quintile refers to the lowest performers. While there is no age limit applied to CEOs, as there is in athletics, the career of an executive of a poorly performing corporation is rather perilous, and could indeed be quite short. On the other hand, the highest performing companies tend to appreciate their leaders.

My progressive gut tells me that CEOs are paid too much relative to the average wage earner, but I also have a hard time justifying paying the CEO of a large corporation less than an all-star athlete.

Thanks to Steven N. Kaplan for the interesting data.