Although the cost of executive compensation has grown measurably in recent years, after adjusting for the riskiness of equity-based compensation, pay increases (measured in terms of executive value) have arguably been much more modest.
Between 1992 and 1998, median CEO pay levels in S&P 500 industrial companies increased nearly three-fold, from less than $2 million to over $5 million. That increase largely resulted from a dramatic growth in stock option grants, which grew from 23 percent of total CEO compensation in 1992 to 44 percent of compensation in 1998. Indeed, in fiscal 1998, fully 97 percent of S&P 500 companies granted options to their top executives, compared to 82 percent in 1992.
Stock options accordingly have drawn increasing attention in recent years. But a new study by Brian Hall and Kevin J. Murphy suggests that attention has not included sufficient analysis of the distinction between the cost of options to companies and the value of options to executives. In Stock Options for Undiversified Executives (NBER Working Paper No. 8052), Hall and Murphy point out that that both practitioners and academics routinely use standard option-pricing formulas, such as Black-Scholes, to approximate both the cost and value of options. Hall and Murphy, however, argue that while these formulas are appropriate for approximating the cost of options (subject to adjustments for early exercise and forfeiture), these formulas are not appropriate for measuring the value of non-tradable options held by risk-averse, undiversified executives who cannot easily hedge their holdings.
For this reason, the researchers develop an analytical framework for valuing stock options and measuring the incentives created by such options, with a special focus on distinguishing between the cost to the company and the value to the executive-recipient of options. Their risk-adjusted pay calculations use S&P 500 executive compensation data and take into account such factors as shares of company stock owned, executive non-firm related wealth, option grant size and characteristics, executive risk aversion, and company features such as stock price volatility.
Hall and Murphy derive the risk-adjusted "Executive Value" of a non-tradable option and compute the "value-cost" ratio by dividing executive value by the company's cost of options. Their results show that value-cost ratios are lower for more risk-averse and less diversified executives, and that value-cost ratios are higher for options that are "in-the-money" or have provisions allowing early exercise. Understanding the divergence between the value and cost of options, the authors maintain, casts light on virtually every stock option practice, ranging from option design to executive behavior to stylized facts about executive pay trends.
Specifically, Hall and Murphy confirm executives' claims that the Black-Scholes values are too high. Their analysis also helps explain why executives demand large premiums for accepting stock options in lieu of cash compensation. The study most markedly shows that although the cost of executive compensation has grown measurably in recent years, after adjusting for the riskiness of equity-based compensation, pay increases (measured in terms of executive value) have arguably been much more modest.
Hall and Murphy use their framework to analyze the relative merits of restricted stock versus options as incentive instruments. Their analysis suggests that restricted stock may be preferable to options under certain circumstances. Specifically, while standard at-the-money options maximize incentives when grants are an add-on to existing pay packages, restricted stock is preferred when options are granted instead of cash. Moreover, the results explain why both executives and shareholders benefit from early-exercise provisions, why executives routinely exercise options on their vesting dates, and why relatively short vesting periods are the norm.
The Hall and Murphy analysis has important implications for further research in the contentious matter of executive pay. There is a strong need, they say, for a framework for understanding and quantifying the value-cost efficiency in all forms of risky compensation, not just options. This need will continue to grow, Hall and Murphy assert, as long as companies increasingly put higher percentages of pay at risk for ever-larger numbers of employees, a trend that exists not only in the United States but abroad as well.
-- Matt Nesvisky