STOCK INVESTORS ENJOYED ANOTHER BANNER YEAR IN 1996. AND CEOs around the land-but especially in Dallas-extracted huge amounts of pay from their open-handed boards of directors.
In my just-completed study of the CEOs of America’s 1,000 largest companies, I found that the 33 Dallas-area CEOs cleaned up. The combination of salary and bonus rose 13.1 percent from 1995 to 1996 for the Dallas crowd-a much higher figure than the 7.6 percent median for all 1.000 CEOs. Total Direct Compensation-the sum of base salary, annual bonus, long-term incentives (including the present value of stock-option grants), and miscellaneous compensation-rose a stunning 38.3 percent in Dallas, more than double the 16 percent increase for the median CEO nationally. What’s interesting is the wild pattern of pay among the Dallas CEOs. On one extreme, I found Sterling Software’s Sterling L. Williams, who was paid astronomically above the market. On the other, 1 found Ensco International’s Carl F. Thorne whose pay package was so tiny in relation to his company’s size and high performance that I was almost reduced to tears.

Is Sterling Williams Making More Than He’s Worth?
Earning $45 million last year, Sterling Software’s CEO is the most overpaid in Dallas.
HIS NAME IS STERLING WILLIAMS, BUT THAT’S TOO MODEST for him. He ought to be called Platinum Williams. because-relative to his company’s size and performance-he outearned every other CEO in the Dallas area. His pay package included a base salary of $900,000, a bonus of $500,000 and options on 3.75 million shares of stock ( including options granted in the stock of a publicly traded subsidiary).
If Williams had been only an average performer, his "going rate" (determined on the basis of his company’s size) would have been $ 1.9 million. But he has been an excellent performer (96th percentile ranking), and, therefore, his going rate was upped to $2.7 million. But $2.7 million is a long way from the $45 million he actually earned in 1996. Williams’ whopping 1,592 percent overpayment topped the Dallas list this year, enabling him to pass last year’s No. I, DSC Communications’ James L. Donald (who is No. 2 this year).
Critics of high CEO pay often lay the blame on the fact that, in most companies, the CEO is also the chairman of the board. They advocate splitting the two jobs and creating what has come to be called a non-executive chairman. That way. they figure the non-executive chairman will ride herd on the CEO and prevent any compensation excess from developing.
But, as it turns out, Sterling Software has a non-executive chairman, the legendary computer entrepreneur Sam Wyly. So why isn’t Wyly riding herd like he’s supposed to be doing? Well, one reason is that he’s raking it in himself. And so is his brother. Charles Wyly, who is the company’s vice chairman. If you add together the pay of Williams and the two Wylys, the tab comes to about $94 million. That’s a lot of overhead.
Besides high pay. there’s also a problem at Sterling Software involving nepotism. Sam Wyly’s son is a vice president and director, and Charles Wyly’s son-in-law is a director. He is also a vice president of Michaels Stores, whose CEO. R. Michael Rouleau, turned out to be the sixth most overpaid CEO in the Dallas area. Interestingly, that company also has a non-executive chairman. And his name happens to be Sam Wyly. Maybe Wyly needs to be sent over to SMU’s business school to learn the value of a buck.

Is Ensco International’s Carl F. Thome Worth More?
Dallas’ most underpaid CEO, he’s given lie to those who say success requires excessive pay.
CHAMPIONS OF HIGH EXECUTIVE PAY WOULD HAVE YOU BE-lieve that high pay is the price of high performance. Or, to put it another way, without high pay, there isn’t going to be any high performance. The same critics also defend stock options on the basis that, without them, a CEO will short-term it-by cutting back on research and development and product quality to hype up current earnings.
Don’t place too much weight on either of these propositions! If there is any CEO in Dallas who gives the lie to both of them, it is Ensco International’s Carl F. Thome. Thome didn’t receive any long-term incentives in 1996, and his total pay package weighed in at only $844,300. Moreover, he hasn’t had any sort of long-term incentive grant since November 1990, almost seven years ago. His pay package turned oui to be 75 percent under the market, after adjusting for his company’s size and performance.
So how does a CEO who is incentive- and dollar-impoverished perform? Based on the conventional wisdom, we ought to see him languishing in bed, hardly able to bestir himself to call his office. But not Carl Thome. If you had invested $I00 in the S&P 500 Index on April 30, 1987 (Thorne became CEO in May 1987). reinvested dividends and then sold out your position on December 31, 1996 (the end of Ensco’s most recent full fiscal year), your profit would have been $244. But had you put the same $ 100 on Thome, your profit would have been $341. What is more, Ensco’s total returns to shareholders have outpaced the S&P 500 in every one of the other nine time windows I tested-time windows stretching from one year to nine years. indeed, during the past five years, Ensco has beaten the S&P 500 Index by a factor of almost 8-to-1.
So what are we to make of two high-performing CEOs, one-Sterling Williams--who is wildly overpaid, and another-Carl Thorne-who is wildly underpaid? Is it possible that the well-springs of their motivation to perform are not to be found in high base salaries, high bonuses, and Godzilla stock-option grants? If so. doesn’t that mean that all that high pay is being wasted on Sterling Williams?
LETS HEAR IT FOR BOB CRANDALL-MAYBE
AMK’s CEO seemed like be needed a tin cup. But m a bit of financial gamesmanship, bis board bas been slipping him bonuses after year s end.
LAST YEAR’S BEST BUY IN THE DALLAS area was AMR’s Robert L. Crandall, the CEO airline unions love to hate. We have been watching Crandall’s pay package for a number of years, and in all the more recent years, he looks to have been significantly underpaid compared to other CEOs who run companies of the same size and with the same performance. Indeed, the table accompanying this article shows Crandall to be the relatively lowest paid CEO in Dallas, not Ensco’s Carl Thorne.
But this year, for the first time, we stumbled on a little game that AMR’s board seems to have been playing for quite a few years. Last year I examined AMR’s proxy covering 1995, and I noted that the table reporting his pay showed his annual bonus to be zero.
However when I looked at AMR’s most recent proxy, the one covering 1996,1 read that his bonus for 1995 (companies must report three years of pay data in each proxy) was $678,400.
What’s going on here? How can a bonus for 1995 be zero in one document and $678,400 on another? And given that discrepancy, what about the fact that AMR showed Crandall’s bonus for 1996 to be zero? Is that number going to change, too, when the AMR Revisionist History department does its thing?
Well, this year I decided to read Crandall’s proxy more closely. And although the table covering his compensation for 1996 does indeed show a zero bonus, I found, in another section of the proxy, the following statement: "As of the date of this proxy statement, the [Compensation! Committee has not reviewed the individual performance of [Mr.] Crandall.... Based on American’s performance in 1996, the Committee expects that, after reviewing and thoroughly evaluating the performance of [Mr. Crandall], it will approve the payment of a bonus [to him]."
In case you’re wondering, the date of that proxy was April 30, 1997. By that date, the fiscal year had been over for four months, and the Compensation Committee hadn’t yet decided whether Crandall is a genius or a dog? Come on!
The way I see it, AMR is playing a sleazy game by hiding the bonus each year and then, when few people are looking, slipping it into the next proxy. And the crazy thing is, there’s no need to play such a game, because even with a handsome bonus for 1996, Crandall would still have turned out to be extremely underpaid.
It’s time for AMR’s Compensation Committee to straighten up, and like the company’s airline, fly right.
METHODOLOGY
MY DEFINITION OF TOTAL PAY INCLUDES the CEO’s base salary, his bonus for performance during 1996, the value of any free shares of company stock he received during 1996, the estimated present value at grant of stock options granted during 1996 (using the Black-Scholes option pricing model), payouts received during 1996 under other types of long-term incentive plans, and miscellaneous compensation (mainly perquisites and special executive benefits).
Defined that way, total pay among the 33 CEOs from the Dallas area ranged from a paltry $504,800 for TNP Enterprises’ Kevern R. Joyce to a smashing $45 million for Sterling Software’s Sterling L. Williams. Average pay for all 33 CEOs in the area was $4.5 million.
The major factor driving up pay last year was the stunning increase in the size of stock-option grants. For a humble analogy, consider that the yield a farmer produces from his fields is fundamentally the result of two factors: namely, the growing conditions (soil, rainfall, fertilizer) and the number of acres planted. In the world of stock options, one can liken the appreciation in the stock market to the farmer’s growing conditions. And growing conditions have been great in the last two years. In the same vein, consider option grants as analogous to the number of acres the farmer planted. And here, the acreage devoted to stock options has been rising far faster than any other part of the pay package. Indeed, the size of option grants has risen so fast that Dallas CEOs are virtually assured of making millions in the next five to 10 years, regardless of the growing conditions for their industries.
But that scenario still does not answer the question whether there is any rhyme or reason as to why one Dallas-area CEO earned $505,000 in 1996, while another earned $45 million. The answer is that there is both some rhyme and some reason, but not very much of either. Looking at all 1,000 CEOs in my study, I found that 37 percent of the variation in CEO pay was attributable to differences in company size (other things being equal, larger companies pay more than smaller companies). And a further 5 percent of the variation in pay was attributable to the wide differences in company performance (other things being equal, higher-performing companies pay more than lower-performing companies). But stated another way, 95 percent of the variation in CEO pay has little or nothing to do with company performance, and some 60 percent falls under the category of "your guess is as good as mine."
In calculating company size, I looked at three different measures, each of which was given equal weight: revenues, permanent invested capital (sum of shareholders’ equity and long-term debt), and the total number of employees in the company.
I measured company performance by considering the company’s total shareholder returns (i.e. stock price appreciation and reinvested dividends) in three different time windows: For the three years, 1993 through 1996; for the two years, 1994 through 1996; and for the single year 1996.
An easy way to determine whether a CEO is overpaid or underpaid would be to measure his actual total pay in relation to the simple average pay for all CEOs. But that approach could be quite distorting, given that there is a systematic relationship between total pay and company size and between total pay and company performance.
A better way to determine who is overpaid or underpaid is to determine, mathematically, what an average-paying company would offer the particular CEO, were the average-paying company to have, not average size, but that company’s size; and not average performance, but that company’s performance. Then a useful comparison can be made between the CEO’s actual total pay and the size-adjusted, performance-adjusted level of competitive pay. That is what the table accompanying this article does.
The bottom line, though, is this: Pay is rising too fast for CEOS, and many pay packages are simply irrational.