COMMENTARY : In Game of ‘I Win, You Lose,’ CEOs Are Getting Richer Faster Than Workers Can Say ‘Logic’ : Growing the Pay Gap

Is the market for chief executive pay: a) becoming more and more rational or b) as crazy as ever?

If you chose “b,” go to the front of the class.

I have just analyzed the 1994 pay packages of the CEOs of 424 of the nation’s largest companies. Here are a few tidbits to get your juices flowing:

* About three-quarters of the CEOs had been in their jobs at least three years. On average, each member of this group of 292 executives earned $2.7 million in 1992 (counting base salary, bonus for annual performance and various long-term incentives, including the estimated present value of stock option grants).

* The average American worker earned $18,900 in 1992. So the average CEO that year earned 145 times the pay of an average worker.

* One year later, in 1993, the same 292 CEOs running the same companies increased their total direct compensation by 20.6%, to $3.3 million. The average American worker, however, managed to increase his pay by only 2.6%. So in 1993, the average CEO was making 170 times as much as the average worker.

* Last year, the same 292 CEOs running those same companies increased their total pay by a further 12.8%, this time to $3.7 million. The average American worker eked out a 3% gain, to an even $20,000. So in 1994, the CEOs made 187 times as much as the average worker.

There are three rational explanations for this ever-increasing gap between the pay of CEOs and that of their workers.

One is that there has been a huge increase in demand for CEOs compared to workers in general, but no increase in supply. Wrong! There are only so many major companies in the United States, and the number doesn’t appear to be increasing.

Another explanation is that there has been a lowering in the supply of capable CEOs but no lowering in demand. Wrong again! There are more highly trained managers in America than ever before--including the women managers whom, the data indicates, these companies have not even begun to promote to top posts. There were only 25 women listed among the top five executives of the 424 companies last year--and none of them were CEOs, according to the firms’ proxies.

The last reasonable explanation for the huge increase in CEO pay would be that the economy is in an up cycle. Perhaps there is something to be said for the argument that CEOs are entitled to more pay in good times--at least in a few companies. But how is it that the average pay among major-company CEOs rose 12.8% between 1993 and 1994, while shareholders in those same companies managed to achieve only a 2% return on their investment in 1994, counting both stock price appreciation and dividends? Maybe we should get those CEOs to incorporate and go public, so we can buy a share of their future earnings instead of the future earnings of the companies they run.

So there just doesn’t seem to be any rational explanation for the rapid rise in the pay of CEOs compared to the pay of their own workers.

Of course, one might respond: So what? CEOs and their shareholders are playing an adult board game called “I Win, You Lose.” If CEOs--being smarter than your average shareholder--end up with all the money, who is hurt, except another group of well-off people?

But the phenomenon of economic polarization in America--where the number of people in poverty is rising, the number of people earning millions of dollars is also rising and the middle class is disappearing--contains all the ingredients of future societal destabilization. Put it another way: If I extrapolate CEO-to-worker pay ratios of 145, 170 and 187 out to, say, the year 2010, I figure by then that the average CEO will be earning a pay premium versus his workers that last was seen in 1789, in the days of Louis XVI of France. And you know what happened to Louis XVI.

In this year’s study of 424 CEOs, on the low end I found the $308,000 earned by Leon Hess of Amerada Hess Corp. in New York and the $313,000 collected by Warren E. Buffett of Berkshire Hathaway Inc. in Omaha. At the other end of the spectrum, I found the $24.5 million earned by Ralph J. Roberts of Philadelphia’s Comcast Corp. and the $30.5 million earned by Lawrence A. Bossidy of Morristown, N.J.-based AlliedSignal Inc.

What might explain why one CEO earns a paltry $300,000 while another earns $30 million?

As I ran the numbers through my computer, I found that three factors were successful in accounting for at least some of the variance:

* Company size. I gave equal weight to three different measures of company size: revenue, invested capital and the number of employees. By these standards--and other things being equal--the larger the company, the more the CEO earns.

* Company performance. The measure used here was shareholder return during the CEO’s tenure--though not more than 10 years. Again, other things being equal, the better investors fared during the CEO’s tenure, the more the CEO earned in 1994.

* Sensitivity of the CEO’s pay package. I measured the extent to which the CEO’s pay package would rise or fall under extreme swings in shareholder return. Included here were not only the elements of direct compensation, but also the actual shares owned by the CEO. By this measure too--other things being equal--the more sensitive the CEO’s pay package, the more he earned.

*

There does, then, appear to be some rationality in the CEO market after all, you say. But consider that these three factors, in combination, were successful in explaining only 34% of the variation in CEO pay levels. Fully 66% of the variation seems to come under the category of “your guess is as good as mine.”

There are, however, a couple of non-rational factors that help explain the variations.

The first is where the CEO works. If his office is in the New York metropolitan area, then--other things being equal, such as company size and performance--he earns 29% more.

Why is that non-rational? Isn’t New York a high-cost-of-living area? It sure is, but so are Los Angeles and San Francisco, and CEOs in those high-rolling cities don’t earn a dime more than their counterparts in, say, ultra-cheap St. Louis or Minneapolis.

The second non-rational factor concerns the number of long-term incentive plans in which the CEO participates. There are three types of plans in use: stock option plans, restricted stock plans (where the CEO gets shares and has to do nothing to get them, save remain with the company) and performance share plans (where the CEO gets shares but has to do something to get them, such as boosting earnings or return on equity).

In a perfectly rational market, if one CEO participated in one long-term incentive plan, a second CEO participated in two plans and a third participated in three plans, all three CEOs--assuming all else was equal--would earn the same level of compensation. The CEO with three long-term incentive plans would simply receive a lower dosage of each long-term incentive drug.

But in the real world of CEO pay, for each additional plan in which the CEO participates, he earns an average of 54% more.

The reason: Consultants love to design new incentive plans, because they can earn lots of fees. But getting the CEO to reduce the dosage level in the long-term incentive plans he already has is about as challenging as prying a piece of prime filet from the jaws of a Doberman pinscher who hasn’t eaten in a week. So the consultant swallows hard and adds the new long-term incentive plan on top of everything else the CEO already has, and the board’s compensation committee obligingly goes along.

The result? Higher and higher compensation. And when other CEOs begin to take notice of those higher levels of pay--through the surveys brought to them by the very same consultants--they too demand to participate in yet more incentive plans.

To be sure, there are some companies out there that have made a serious effort to play the pay game fairly. They have held back on base salary growth for the CEO and other senior executives. They have made bonuses swing more in relation to performance. And they have pretty much forced top executives to hold on to a lot of company shares so that--for the first time--they can, to use President Clinton’s favorite phrase, “feel the pain” of shareholders when times turn bad.

But that progress has not, it seems, come without a stiff price tag--namely, a continuing, rapid rise in overall CEO pay levels.

Reapportioning the slices in the pay pie to emphasize pay for performance is a heartening development. Now institutional shareholders--indeed, all shareholders--need to take steps to bring the diameter of the pie under control.

(BEGIN TEXT OF INFOBOX / INFOGRAPHIC)

Getting Paid What They’re Worth?

Thirty CEOs stand out in Graef Crystal’s survey of the 1994 pay of 424 executives. The first table below shows the 15 most striking combinations of low performance--in terms of total shareholder return during the CEO’s tenure--and high pay. When it comes to paying for performance, Crystal says, “the boards of these companies must all be considered finalists for an ‘unclear-on-the-concept’ award.” The second chart shows the 15 most dramatic combinations of high performance and low pay. Crystal notes that some of the lesser-paid CEOs on this list--such as Warren Buffett and Bill Gates--supplement their incomes with extensive stock holdings.

Low Performance, High Pay

Competitive Actual pay, pay, Company Execuitve in millions in millions Bear Stearns Cos James E. Cayne $15.6 $3.5 Equitable Cos. Richard H. Jenrette 8.6 3.0 Union Pacific Drew Lewis 10.0 4.0 Corp. Novell Inc. Robert J. Frankenberg 5.1 2.0 Rubbermaid Inc. Wolfgang R. Schmitt 4.3 1.8 Western Atlas Alton J. Brann 4.6 2.1 International Inc. State Street Marshall N. Carter 3.5 2.0 Boston Corp. J.P. Morgan & Co. Dennis Weatherstone 7.4 4.3 Ford Motor Co. Alexander J. Trotman 11.9 7.7 Shawmut National Joel B. Alvord 3.2 2.2 Corp. Burlington Thomas O’Leary 2.5 1.8 Resources Inc. Ingersoll-Rand Co. James E. Perella 3.4 2.5 Newmont Mining Ronald C. Cambre 0.998 0.726 Corp. Baker-Hughes Inc. James D. Woods 2.8 2.0 Tenneco Inc. Dana G. Mead 4.0 3.0

Company Deviation Bear Stearns Cos 346% Equitable Cos. 186 Union Pacific 150 Corp. Novell Inc. 148 Rubbermaid Inc. 134 Western Atlas 124 International Inc. State Street 74 Boston Corp. J.P. Morgan & Co. 70 Ford Motor Co. 54 Shawmut National 43 Corp. Burlington 39 Resources Inc. Ingersoll-Rand Co. 38 Newmont Mining 37 Corp. Baker-Hughes Inc. 35 Tenneco Inc. 34

*

High Performance, Low Pay

Competitive Actual pay, pay, Company Execuitve in millions in millions Berkshire Hathaway Warren E. Buffett $313 $12.3 Inc. Microsoft Corp. William H. Gates III 458 11.1 Wm. Wrigley Jr. William Wrigley 506 5.2 Co. Tellabs Inc. Michael J.Birck 606 4.3 Franklin Charles B. Johnson 764 4.3 Resources Inc. McDonnell Douglas John F. McDonnell 1,647 6.4 Corp. Gap Inc. Donald G. Fisher 2,207 7.7 Oracle Systems Lawrence J. Ellison 2,410 8.1 Corp. Home Depot Inc. Bernard Marcus 2,771 8.6 Lowe’s Cos Leonard G. Herring 1,733 5.0 Ralston Purina Co William P. Stiritz 1,708 4.3 Cisco Systems Inc. John P. Morgridge 1,317 3.1 LDDS Communications Bernard John Ebbers 2,264 5.3 -CL A National Media Jeffrey C. Barbakow 1,856 4.4 Enterprises Inc.* Kohl’s Corp. William S. Kellogg 1,708 3.7

Company Deviation Berkshire Hathaway -97% Inc. Microsoft Corp. -96 Wm. Wrigley Jr. -90 Co. Tellabs Inc. -86 Franklin -82 Resources Inc. McDonnell Douglas -74 Corp. Gap Inc. -72 Oracle Systems -70 Corp. Home Depot Inc. -68 Lowe’s Cos -66 Ralston Purina Co -60 Cisco Systems Inc. -58 LDDS Communications -57 -CL A National Media -57 Enterprises Inc.* Kohl’s Corp. -54

Note: “Actual pay” is the sum of base salary, annual bonus, restricted stock grants, performance share and unit payouts, estimated present value at grant of stock options and miscellaneous compensation. “Competitive pay” is the amount an average-paying company would offer to the CEO if it were to have this company’s size, performance and CEO-pay-package sensitivity. “Deviation” is the percentage by which actual pay differs from competitive pay.

* Now Tenet Healthcare Corp.

Sources: Standard & Poor’s Compustat; Crystal Report