WHAT’S A CEO WORTH?
Two of the more popular themes for U.S. corporation bashers are “Pay is not linked to performance” and “CEOs are paid too much!” This year we tested these hypotheses/ accusations against what 225 companies reported in their proxy statements and annual reports. What did we find? The answers may surprise you.
September 1 1988 by David Meredith
The results of this year’s study are similar to those of last year’s: In virtually two-thirds of the 225 companies in our sample, the pay of the CEO is aligned with performance. Admittedly, 35 percent of these are low-paying/ low-performing -but pay and performance are aligned. Performance is defined as shareholder return compared with industry medians; while CEO pay is reported total pay (salary, bonus, plus annualized present value of long-term incentives), relative to size-adjusted industry norms.
So, you ask, how is it that so many published reports of CEO compensation conclude that pay is not related to performance? We see two technical flaws in most approaches. The first problem is defining pay as monies received in a particular year without properly linking these amounts with the time period over which monies were earned. The picture becomes very distorted when CEOs exercise large numbers of options at once, even if these options were earned over a relatively long period of time. With invalid data, it is difficult to find meaningful correlations-in other words, garbage-in, garbage-out. Second, attempting to find performance/pay correlations across different industries is virtually impossible without using sophisticated statistical models to filter out the “noise” accounted for by pay and performance differences among industries. For instance, when looking at 225 companies across all industries, there is virtually no correlation between pay and performance, when applying curvilinear regression analysis. Yet, when the same approach is used on an industry-by-industry basis, individual correlations become highly significant (as high as 0.79).
THE PAY-PERFORMANCE LINK
Those who claim that pay is not linked to performance have a somewhat valid point. While two-thirds of our sample did indeed link pay to performance this year as was the case last year-there are, nonetheless, one-third whose pay and performance are improperly aligned. The problem: when performance goes down, CEO pay does not follow to the extent that the critics (and we) would argue it should.
Yet, when performance improves, CEO pay is often quick to follow. For example, the automotive and the farm and industrial equipment industries’ average return to shareholders fell to 4.8 and 1.1 percent, respectively. By failing to penalize poor performance, those industries permitted the greatest increase in performance/pay misalignment. On the other hand, the financial and publishing industries (10.5 and 20.3 percent return, respectively) showed the greatest gains in aligning pay with performance.
It appears to us that the spotlight on the performance/pay issue has spurred many management teams and boards to action. This year we saw much more activity than usual in the area of long-term plan adoption and amendments. Nearly one-third of the 225 companies either proposed the adoption of a new plan or proposed substantive amendments to existing long-term incentive plans.
Did these changes help? Yes! When comparing those who made changes with those who didn’t, it is obvious that the “changers” take the performance/pay issue more seriously than those who take no action. Of those who changed or tuned their plans in 1987, 68 percent started out aligned versus 60 percent of those who made no change. After the changes, the alignment gap widened to 73 percent versus 61 percent. That’s a lot of progress in one year. The moral of the story is that management teams and boards should, and can, take action to better align CEO paywith shareholder returns.
Although we are pleased with the gain in performance/pay alignment, it seems that boards are more inclined to use an alternate prescription to fix the performance /pay problem: fire the poor performer. In fact, while 10 percent of our sample companies replaced their CEOs in 1987, a full 24 percent of the CEOs from the group we deemed high risk (i.e., high paid and low performing) were replaced. Only 5 percent of the high pay/high performers were replaced. If you’re a high-risk CEO, the moral of this story is either fix your performance, fix your pay or update your resume.
CEO PAY PROTECTED
This year’s findings show that too many CEOs (and their boards) are attempting to fix the performance/pay equation by adding more to salary, or by adding more to cash incentives, or by increasing the number of options granted. Some are even lowering the performance target by replacing underwater options (those granted at a higher exercise price than the current trading price) with new, lower-priced options. I’ll bet the shareholders wish they could cancel their “underwater” purchases! Alternatively, boards of underperforming companies are much more inclined to change CEOs. It would seem that a more prudent-and humane-way to deal with the issue would be to design incentives into pay plans to provide a significant reward or a significant penalty. (See “Home Runs And Strike Outs”)
The problem seems to be that more and more of the CEO pay package is fixed-that is, not subject to significant performance-driven changes. This year, we examined how pay has changed for the CEO group from last year. Salaries increased by an average of 8.7 percent for the 43 companies reporting salaries (for the same CEO) in both years. Total cash (salary plus bonus) increased by 7.3 percent. And while the annualized present value of long-term incentives increased an average of 8.4 percent, the grant values increased by 20.2 percent. What role, if any, did the stock market crash have in the changes of performance/pay alignment? While performance did slip badly with most of our sample companies, it was not just the result of Black Monday. In fact, when we examined the changes in the aggregate market value of the companies over the two performance periods (1982-1986 vs. 1983-1987), a majority of the decline was caused by a change in the beginning value. The market value for the 217 “matched pairs” increased by 20.4 percent from 1982 to 1983, and only 1.5 percent from 1986 to 1987.
Despite an overall decline in share growth appreciation, the total performance pay (salary, bonus, plus annualized longterm pay) increased by 8.1 percent from 1986 to 1987, at a time when shareholders saw only 1.5 percent increases in their stock portfolios-7.5 percent less than a longterm treasury note. CEOs’ pay seems to have been subject to more protection than most shareholders’ portfolios.
HOW MUCH IS ENOUGH?
The press has displayed a morbid fascination with the highly compensated CEO. Is it possible that this breed can be typified as overpaid, under-performing and self-serving? Are the Eisners, Iacoccas, Petersens, Reeds and Sculleys of the world worth that much? To help shareholders answer this question, we extended our survey methodology by comparing the CEO’s absolute dollar competitive pay premium (or penalty) with shareholders’ absolute performance premium (or deficit).
As detailed on the following 14 tables and accompanying graphs, the high paid/high performing CEO gets an average pay premium of $1.06 for every $100 premium he creates for the shareholder. Yet, on average, the poor performer is penalized $.23 for every $100 he costs the shareholder in terms of value created. Because these numbers are inflated by a small number of CEOs with big swings in pay, we also calculated median numbers to check this potential distortion. The high performer receives median premium of $.19 per $100, while the median penalty for the poor performer is only $.08 per $100. Thus, the pattern is consistent: the upside potential is almost three times as great as the downside risk. I wonder if this is the same for shareholders?
The real issue is: Are shareholders getting their money’s worth for CEO pay? When one CEO was asked how he could possibly justify his rate of pay, he responded: “I can’t. But, if they don’t pay me, someone else will.” And, in today’s world of LBOs and other financial restructurings, it is clear that the public spotlight and SEC scrutiny have not hit the world of private deals. In the LBO world, it is not unusual for the management team to get 15 or 20 percent of the wealth created for the shareholders. The CEO will often get half of the package. So, at $.19/ $100 premium, or an $.08/$100 penalty, it is not surprising to find the Don Kellys (of Beatrice fame) moving to a new league where the standard payoff can be $8 to $10 per $100 created. In this light, Donald Petersen represents a relative bargain for Ford shareholders. In the past four years, Ford has outperformed the industry to the tune of $13.5 billion, which puts Petersen’s pay premium of some $7 million into perspective, as he gets about a nickel for every $100 in premium value created. Apple’s John Sculley-evidently a good negotiator-charged a $1.83 premium for each $100 of Apple’s $3 billion performance premium. In each of these cases, CEO pay seems not only to be linked with performance, but may represent a bargain in terms of return on investment.
So why all the fuss? It seems to us that the performance pay critics may really be arguing that the CEO and the shareholders are not playing on a “level field.” When performance is up, everyone is happy. Yet, when performance goes down, it is the shareholder who gets the short end of the stick. Take the case of Citicorp where pay and performance are aligned. The bank’s shareholders “enjoyed” a $3.5 billion deficit in share value relative to the industry. And John Reed? He was penalized $637,000.
METHODOLOGY AND TERMS
CEO pay data for the most recent fiscal year were analyzed from annual reports and proxy statements for a sample of 225 companies that represent a cross section of small, medium and large companies, and low, medium and high performers in 14 separate industries. In cases where the CEO changed in mid-year, we chose the CEO who served the longest during the year.
Company Performance was measured by summing the growth in stock price with all dividends for a four-year period, then calculating the compound growth rate that would yield this return based on an investment in the stock at the beginning of the period. (As a simplifying assumption, dividends were not “reinvested.”)
The Performance Factor was determined by subtracting the median industry performance from company performance.
Actual Pay is annualized present value of salary, annual bonus, and long-term incentives. The valuation of stock options illustrates this:
· Assumption: options exercised in five years.
· Exercise value is established by projecting the option price for five years at the rate the stock price increased over the previous four years.
· Future gain (exercise value less option price) discounted by 7 percent to determine present value.
Competitive Pay was calculated through regression analysis (“lines of best fit”) comparing revenues with CEO pay for each company within an industry group.
Pay Factor is the percent difference between Actual and Competitive Pay.
Shareholder Premium (Penalty) is the difference between growing the market value of the company at the Company Performance rate and growing it at the Industry Performance rate.
Pay Premium (Penalty) is the difference between Competitive and Actual Pay.
Pay Premium (Penalty) per $100 in Performance Premium (Penalty) is calculated by dividing the Pay Premium (Penalty) by the Shareholder Premium (Penalty).
NM relates to companies where pay and performance are not aligned.