More on Executive Compensation: The Economist's Special Report

Over the weekend, I read The Economist's 20 page special report on executive pay, in the current issue of the British news weekly. The report contains the usual excellent mix of historical information, empirical data and research, and sensible reasoning that I've come to expect from the magazine's reporting.

Of particular interest to me, however, was this passage, found on page 9 of the special report,

"As the average firm size increases, so each company must pay its top executives more. When managers control more assets, they can make more of a difference to absolute profits. Hence, in a competitive market full of bigger companies, boards will be prepared to spend more on talent. Using a schematic mathematical model, two American-based European economists, Xavier Gabaix and Augustin Landier, concluded that the sixfold increase in the size of American firms between 1980 and 2003 may account for much of the sixfold increase in managers' pay during that period. Tiny differences between the abilities of top managers could explain large differences in pay. Chart 2, from another study, shows how in historical terms the level of pay is still relatively low in relation to the size of companies."

To be honest, this particular piece of data was news to me. I suppose the continuing drumbeat of mergers should have alerted me to the trend, but I had no idea that the asset sizes had increased to this extent.

Shortly thereafter in the article, the report states,

"In fact, the typical chief executive of an S&P500 company, who earned just under $7MM in 2005, according to the Corporate Library, must think he is in the wrong job.....senior investment bankers stood to earn bonuses of $20MM-25MM in 2006 and top traders $40MM-50MM. To qualify for Alpha magazine rankings of the top 25 hedge-fund managers in 2005, you had to earn $130MM."

This puts some things into perspective, does it not? Essentially, we are faced with a simple conundrum which is going to make a lot of people very unhappy. Large corporations have gotten very much larger in the past 20+ years, and, thus, the CEO of such a company has a much larger effect, good or bad, upon the total return of those swollen asset sizes. While the lower-level workers may have experienced productivity gains as well, it's doubtful that they are on a scale of what may be possible for an effective shareholder-wealth creating CEO of a very large-cap US company.

Thus, we have William McGuire, late of UnitedHealth Group, making over a billion dollars based upon the peak value of his options.

On one hand, this aggregation of assets for the average large cap company means more opportunity to add value for shareholders. And, thus, a basis for higher compensation for those CEOs who achieve that via consistently superior total returns. For those that do not, however, it means larger compensation packages for even larger opportunity costs.

Ironically, as the company size has risen so dramatically, the sector which is most-often involved in that activity, investment banking, pays its people even better than the CEOs of the merged companies it creates, as the second quoted passage notes. I suppose this is yet another symptom of industrial sales people typically earning more than the CEOs of the firms for which they work, so it's not, per se, part of the 'problem.'

Returning to the CEO compensation issue, the Economist report also notes, later on in the section, that many boards reward senior executives with options, as if they are free when, in fact, they become costly and dilutitive, on the bases on which they are typically granted.

This brings me to a more detailed idea for CEO compensation, based upon my proprietary research. If a CEO must be paid with options, why not make the maximal value of the options available be the product of the difference of the company's 5-year average total return with the S&P500, and the recent market value of the firm, multiplied by some reasonable fractional value, such as 10%. Thus, the CEO of an outstanding firm which had consistently out-earned the S&P500 would stand to gain 10% of that difference, multiplied by the value of the firm's equity.

Another topic which was discussed in the Economist's report is that of risks borne by CEOs, relative to the compensation they receive for a job that is experiencing a decreasing average tenure. However, from my observations, and discussions with a few CEOs, I believe the report misses an important behavioral aspect of this issue. Most CEOs of large companies would likely do the job for half of what they are paid, simply to enjoy the control of all those assets, as well as the high public profile that often comes with the territory of being a large company CEO. And, as I have contended in prior posts, if a company needs an effective CEO, offering an incentive-rich compensation package with a low fixed component should attract candidates who are confident they can deliver performance for shareholders. Does a board really want a CEO who has to be guaranteed immense riches in order to come aboard and attempt to enrich shareholders at a superior rate to the S&P?

We see far more consistent underperformance than outperformance among S&P500 CEOs, yet, to my knowledge, none of the former return the money as a consequence of their failure. Even GE's CEO, Jeff Immelt, who now boasts that 75% of his compensation is tied to performance, made upwards of $30MM before he took that risk. And, the tie-in is not as draconian as he would have others believe.

What is the price, and for what time period, that boards should pay, on behalf of shareholders, in order to suffer an ineffective CEO? It's as easy to pay little for failure as to pay a lot. Yet, if properly constructed, an effective CEO should, with the same compensation package, stand to make a considerable sum by succeeding for the shareholders.

To me, it seems that running ever-larger companies implies three things. First, it's more important than ever for boards to get compensation policies and levels 'right,' as failure is ever-more costly.

Second, CEOs who do consistently outperform the S&P500 on behalf of their shareholders, are doing a much tougher job, with larger asset bases, than their counterparts of 20-odd years ago were, and deserve much more income for doing so. Those who fail should be both paid much less, until they demonstrate success. If a board still needs to give a CEO 3-5 years to prove her/himself, the least they can do is withhold lavish compensation until the long-term performance record of a CEO is clear.

Third, perhaps the growth in average asset size of large companies is a mistake, viewed from a performance perspective. Should we have expected more consistently superior total return performances from the larger firms, if those mergers were value-creating? If not, perhaps the current outrage over CEO compensation levels is derative of the challenge of running companies that are, in fact, too large to be effectively led to consistently superior total returns.