CEO Pay by the Numbers

Keith Kefgen | CEO PERFORMANCE STUDIES, COMPENSATION, CORPORATE GOVERNANCE
On November 17, 2008, Goldman Sachs CEO Lloyd Blankfein, announced that Goldman executives were giving up their bonuses. In the days following, many troubled financial firms including Citigroup, Morgan Stanley and UBS declared the same. Will the same hold true for the hospitality industries (hotel, gaming, and restaurant)? Just twelve months ago, private equity firms were buying up hospitality companies at multiples never seen before. With stock prices and real estate values tanking, how will CEO pay be affected? We submit that if compensation programming was done correctly, change or declarations such as Blankfein's would be unnecessary. How is this possible? In a well conceived pay-for-performance model, executives should make superior pay when they outperform the market and inferior pay when they under-perform the market. The "at risk" components of compensation (bonus & stock) should never be treated as a guarantee. In advising compensation committees, Aethos™ has developed various pay-for-performance models that speak to the heart of the issue. Let's look at how one of these models can be applied to pay and performance in the hospitality industry. Pay-For-Performance Our CEO pay-for-performance model quantifies the performance of a given chief executive relative to his or her peers. The model incorporates standard business indicators, including EBITDA/FFO growth, market capitalization and stock appreciation, and compares that to total compensation. The result is a pay-for-performance index that determines how much a CEO was over or under paid. In applying the model, we analyzed 2008 proxy statements for CEOs in the hotel industry. We separated compensation into four categories: base salary, short-term incentives (bonuses), long-term incentives (stock options, restricted stock grants), and all other compensation. We then calculated the growth in stock price and EBITDA/FFO over a three year period. We used three years to avoid yearly anomalies and reward long-term thinking. Finally, we calculated market capitalization at year end and populated our formula, Aethos™ Value Index™ = (M + S + E)/C. The index had a standard deviation applied, with an average score of 100. In other words, a score of 100 means that the CEO was paid exactly what they deserved (see attached chart). The formula is defined as Aethos™ Value Index™ = (M + S + E) / C, where:
  • M quantifies the difference in running a large versus small cap company. M is calculated by comparing the market capitalization of the subject company to the average market cap of the survey peer group.
  • S measures stock appreciation, calculated by comparing the stock appreciation of the subject company to that of the peer group over a given period of time (three years for our study).
  • E measures EBITDA growth, and is measured by comparing the EBITDA growth of the subject company to the average EBITDA growth of the peer group over a given period of time (In the case of a REIT, FFO is used).
  • C defines the Compensation Benchmark, which quantifies the degree to which a chief executive earns more or less than his peer group. It includes base salary, short-term incentives, long-term incentives and other compensation as required by SEC regulations.
The Aethos™ Value Index™ is akin to a multiple regression as we have found that linear regressions fail to provide any notable conclusions for predicting pay and performance. For example, a linear regression tells us that generally speaking, as a company gets bigger, CEO pay increases. Unfortunately, a big company is just as likely to have inferior performance as any other company. The question that must be answered is, "did the CEO earn his/her pay"? Our model was designed to tie multiple elements together and answer this question. Pay-For-Performance Applied As previously mentioned, a CEO who receives an Aethos™ Value Index™ of 100 was paid exactly what he or she deserved. Let's look at Robert Alter, CEO of Sunstone Hotel Investors, to demonstrate how the model can be effectively applied. Over the survey period, Sunstone had a stock appreciation of 24.16%, which was greater than the peer average of 16.89%. Additionally, Sunstone had positive EBITDA growth of 2.41%, compared to the peer group, which saw negative growth of -4.80%. Sunstone's market cap of $1.8 billion is smaller than the peer group average of $5.9 billion. Finally, we compared Alter's total compensation to the CEO average and came up with a compensation benchmark of .58, which means he was paid lower than the average CEO total of 1.0. When combined, these factors measure how Alter performed in relation to his peers. According to the Aethos™ Value Index™, Alter should have been paid $2,740,000 in total compensation. That compares favorably to his actual pay of $2,698,000; an excellent job by the Sunstone compensation committee. On the extreme of over paid is Joe Martori, CEO at ILX Resorts who should have made $2,000 instead of $471,000. On the extreme of under paid is Jon Bortz, who should have made $4,028,000 instead of his actual pay of $1,901,000. Compensation Planning It appears that the Sunstone compensation committee did an excellent job of programming executive pay. Let's compare our results to the compensation philosophy at Sunstone to see if it was good design or luck that played a role in Mr. Alter's actual pay. The Sunstone compensation philosophy states that, "the compensation program is designed to reward performance relative to financial and other metrics that result in favorable total shareholder returns, both in terms of absolute appreciation in the value of our shares, and in terms of relative performance as compared to our peers, taking into consideration our competitive position with the real estate industry and each executive's long-term career contributions to the company". Each component of Mr. Alter's pay has a philosophical approach. Annual base salary is benchmarked against a peer group (Hotel Real Estate Companies). The annual incentive bonus is based on metrics of return on investment, adjusted FFO per share and subjective and objective individual performance. We prefer less subjective metrics but it looks as if Sunstone got it right. Sunstone's annual equity incentive awards are "based on the achievement of corporate and individual performance criteria and made in the next calendar year". Because the awards are granted only if performance criterion is met, total annual compensation is highly correlated to company performance. It is clear that proper program design and not luck is attributable to Alter's pay. Tying Performance to Compensation Components As we have suggested throughout, company performance is measured by assessing financial and other measurable metrics. However this analysis is not enough. If, for example, the market suffers from an economic downturn where earnings and stock values plummet throughout an entire industry, a company's measurable performance may be the worst it's been in several years. But what if said company's value fell 10% while the rest of those in its peer group plummeted 50%? By comparison, this company performed quite well. For this reason, we submit that proper planning ties each component of executive compensation (aside from base salary) to a peer group or industry norm, much like the program at Sunstone. By definition, a base salary represents the floor and bears no relationship to company performance. As we have proven in past regressions, base salaries tend to be directly tied to the size of the company. This is because as the company gets larger, the responsibilities for a given CEO increase as well. Best in class short-term incentive plans are based on measurable performance metrics such as earnings per share, net income, revenue growth, and so forth. Best in class long-term incentives are based on performance over a longer time horizon (usually 3-5 years) and have a relationship to company performance. The following example illustrates the necessary care in designing long-term incentives:
  • Company A awards it's CEO 100,000 restricted shares at $1 per share, which grows ten fold to $10 over three years, leaving the CEO with $1 million in stock value. Company B, on the other hand, awards it's CEO 1 million restricted shares which grow from $1 per share to $2 over the same time, leaving this CEO with $1 million as well. It is evident that the CEO of company A performed better, but because he had one tenth of the shares he did not earn more than the CEO of Company B. This highly simplified example shows that long-term incentive plans should also account for overall and peer performance in order to ensure fair executive pay.
CEO compensation continues to be one of the most scrutinized topics in corporate America. Shareholders are increasingly attempting to regulate multi-million dollar compensation awards to under-performing CEOs. Large shareholders have actively petitioned corporate boards regarding CEO compensation as they want a greater say on CEO pay. Whether it is the job of the compensation committee, a third party or the entire board of directors, planning CEO compensation can be a complicated process. However, an appropriate pay-for-performance model ensures that a CEO is properly compensated based on his/her performance and accounts for fluctuations in the market and overall economy. The Aethos™ Value Index and similar programs built on this theory don't have to abruptly amend bonus plans, but rather can withstand the intense scrutiny applied to executive compensation that often accompanies large changes in the economic climate.