How stock-option grants were gamed and what to do about it
The compensation of top American corporate executives has soared during the past 15 years. Measured in 2005 dollars, the average annual compensation of the CEOs of the large companies in the Standard & Poor’s 500 almost tripled from 1992 to 2005, growing from $3.7 million to $10.5 million. This growth in executive pay has been accompanied by a parallel increase in public interest in the subject, and has prompted heated debate about executive compensation—its appropriateness and effectiveness—and corporate governance.
In this context, the opportunistic timing of executive stock-option grants, via backdating or otherwise, has attracted a great deal of news coverage, regulator attention, and public debate since the media first focused on it in the spring of 2006. The U.S. Senate’s banking and finance committees held hearings on the subject. The Securities and Exchange Commission and a small army of private law firms hired by companies themselves have been intensively investigating past grant practices in many companies. More than 150 firms have thus far come under scrutiny, dozens of executives and directors have been forced to resign, and many companies have announced that they will have to revise their past financial statements.
But our understanding of option-grants manipulation remains incomplete. What circumstances and factors led to opportunistic timing of grants in some companies but not in others? To what extent has such timing resulted from systemic problems in corporate governance? What lessons should investors and firms draw? In two recent studies, Yaniv Grinstein from Cornell, Urs Peyer from the INSEAD business school, and I explored these questions by investigating empirically the incidence, causes, and consequences of option-timing practices.
In the debate over executive pay, I have sided with those critical of existing arrangements. In Pay without Performance: The Unfulfilled Promise of Executive Compensation, Jesse Fried of Berkeley and I offered a detailed account of the flaws in pay arrangements—in particular, the decoupling of executive compensation from corporate performance and the lack of transparency about both the amount of such compensation and the extent to which it is sensitive to performance. We also argued that the flaws have resulted from the absence of true arm’s-length contracting between boards and executives, and that such flaws are likely to remain unless we adopt governance reforms that improve directors’ incentives to focus on shareholders’ interests. The evidence about backdating reinforces these concerns.
During the past 15 years, the most important component of executive pay packages, and the one most responsible for the large increase in the level of such compensation, has been stock-option grants. The increased use of option grants was justified as a way to align executives’ interests with shareholders’. For various tax, accounting, and regulatory reasons, stock-option grants have largely comprised “at-the-money options”: rights to purchase shares at an “exercise price” equal to the company’s stock price on the grant date. In such at-the-money options, the selection of the grant date for awarding options determines the options’ exercise price and thus can have a significant effect on their value.
Earlier research by financial economists on backdating practices focused on the extent to which the company’s stock price went up abnormally after the grant date. My colleagues and I focused instead on how a grant-date’s price ranked in the distribution of stock prices during the month of the grant. Studying the universe of about 19,000 at-the-money, unscheduled grants awarded to public companies’ CEOs during the decade 1996-2005, we found a clear relation between the likelihood of a day’s being selected as a grant date for awarding options, and the rank of the day’s stock price within the price distribution of the month: a day was most likely to be chosen if the stock price was at the lowest level of the month, second most likely to be chosen if the price was at the second-lowest level, and so forth. There is an especially large incidence of “lucky grants” (defined as grants awarded on days on which the stock price was at the lowest level of the month): 12 percent of all CEO option grants were lucky grants, while only 4 percent were awarded at the highest price of the month.
The passage of the Sarbanes-Oxley Act in August 2002 required firms to report grants within two days of any award. Most firms complied with this requirement, but more than 20 percent of grants continued to be reported after a long delay. Thus, the legislation could be expected to reduce but not eliminate backdating. The patterns of CEO luck are consistent with this expectation: the percentage of grants that were lucky was a high 15 percent before enactment of the law, and declined to a lower, but still abnormally high, level of 8 percent afterwards.
Altogether, we estimate that about 1,150 CEO stock-option grants owed their financially advantageous status to opportunistic timing rather than to mere luck. This practice was spread over a significant number of CEOs and firms: we estimate that about 850 CEOs (about 10 percent) and about 720 firms (about 12 percent) received or provided such lucky grants. In addition, we estimate that about 550 additional grants at the second-lowest or third-lowest price of the month owed their status to opportunistic timing.
The cases that have come under scrutiny thus far have led to a widespread impression that opportunistic timing has been primarily concentrated in “new economy” firms. But while the frequency of lucky grants has been somewhat higher in such firms, more than 80 percent of the opportunistically timed grants have been awarded in other sectors. Indeed, there is a significantly higher-than-normal incidence of lucky grants in each of the economy’s 12 industries.
Finally, we estimate that the average gain to CEOs from grants that were backdated to the lowest price of the month exceeded 20 percent of the reported value of the grant. On average, such gains increased the CEOs’ total reported compensation for the year by more than 10 percent.
The backdating of executive grants naturally prompts questions about the role, if any, played by companies’ outside directors. In his opening statement at the Senate Finance Committee hearing on backdating in September 2006, then-chairman Charles Grassley expressed concern that “boards of directors were either asleep at the switch or, in some cases, willing accomplices themselves.” Both scenarios represent failures by directors in their oversight function.
Although such concerns have been expressed with respect to directors’ possible failure to prevent the backdating of executives’ grants, little attention has been paid to the option grants awarded to outside directors themselves. It has been assumed that such grants have been appropriately timed. But our research indicates that opportunistic timing has not been limited to executives’ grants; rather, it has been present to a significant degree in outside directors’ grants as well.
Examining the universe of about 29,000 events in which public companies awarded option grants to (one or more) outside directors, we found that about 9 percent were “lucky” events taking place on dates with stock prices at a monthly low. We estimate that about 800 of these events reflect opportunistic timing, not mere luck. This opportunistic timing was spread over about 460 firms (about 7 percent of the total). As with CEO options, we find that the opportunistic timing of director grants was more common before the adoption of Sarbanes-Oxley, but that it continued afterwards. We also find that such timing has been present in 11 of the economy’s 12 industries (the utilities industry, where director grants are infrequent anyway, is the only exception).
Furthermore, there is a link between directors’ and executives’ luck. Overall, in companies in which opportunistic timing of option awards took place, luck tended to lift the boats of both executives and outside directors.
Backdating or Use of Inside Information?
Excessive incidence of lucky grants can result not only from backdating but also from “spring-loading”—the selection of dates based on the use of inside information that suggests the stock price is about to rise. But our research enables us to conclude that higher-than-random incidence of lucky options grants to CEOs and outside directors has been to a substantial extent the product of backdating:
- The use of inside information is unlikely to lead insiders to differentiate between two stock prices that are very close to one another. We find, however, that a day with the lowest price of the month was substantially more likely to be selected for an option grant than a day with the second-lowest level even when the difference between the two price levels was less than 1 percent. Of course, if the date of a grant event is set retroactively, when the whole distribution of stock prices is known and available to choose from, those selecting the date might well choose to take advantage even of such small price differences.
- Furthermore, spring-loading, which is based on information possessed by insiders at the time the grant is awarded, does not depend on the ability to delay reporting the grant. However, we find that both directors’ and CEOs’ grant events were significantly less likely to be lucky when they were reported in the same month in which the grants took place than when they were reported after the grant month.
- Finally, insiders’ private information might well enable them to predict the future direction of the stock price of their company relative to the market, but it is unlikely to enable them to predict the future direction of the stock market as a whole. We find, however, that days with low stock market price were more likely to be chosen as a grant date, even when the dip in stock was produced by market-wide movements rather than firm-specific ones.
Backdating and Corporate Governance
Although our research identifies a significant incidence of opportunistic timing, I would like to stress that most companies have not engaged in such timing in awarding options. The evidence does not support claims that “everyone was doing it.” And it enables identifying factors that were systematically associated with higher likelihood of lucky grants.
CEO and director grants of options were more likely to be lucky grants when the potential payoffs from such luck were relatively high. For any given firm, the odds of a grant being lucky increased when the gap between the lowest and the median price in the grant month was higher. Thus, grant manipulation appears to have reflected rational economic decisions influenced by payoffs—not to have resulted from thoughtless application of a habit or internal norm.
In addition, opportunistic timing was correlated with increased influence of the CEO on the company’s internal pay-setting and decisionmaking processes. Lucky grants were more likely to occur when the board lacked a majority of independent directors, as well as when the CEO had been in place for a long time.
The Wall Street Journal’s Holman Jenkins has suggested in a series of columns that even though backdating involved “technical” violations of legal rules, it has been used by firms in shareholders’ interest as a rational and sometimes tax-advantaged substitute for other forms of necessary compensation. This argument suggests that, other things being equal, companies awarding lucky grants should tend to pay a receiving executive less (relative to peers) through other forms of compensation. But CEOs benefiting from lucky grants tended to receive more, not less, through other forms of compensation. Furthermore, the fact that many outside directors were themselves recipients of lucky grants reinforces the view that opportunistic stock-option awards were produced by governance failures, not business decisions made rationally and in good faith to serve shareholder interests.
The Benefits—and Limits— of Independent Directors
Outside directors play a key role in the structure of public companies with dispersed ownership. They are relied on to monitor, supervise, and set executives’ compensation. Increased reliance on outside directors has been advocated by many financial economists and legal scholars, and increasing the power and role of independent directors was a key element of recent corporate-governance reforms. Director independence, in short, is now viewed as the foundation on which the American corporate-governance system can safely rest.
Our research demonstrates that director independence can indeed produce benefits. Companies with a majority of independent directors have been associated with lower incidence of opportunistic option timing. But it also shows that director independence is not a panacea. Board independence reduced but did not eliminate opportunistic timing, which still took place among companies with a majority of independent directors. Furthermore, our finding concerning the opportunistic timing of outside directors’ grants indicates that “agency” problems might arise not only between boards and executives but also between independent directors and shareholders.
The mere classification of directors as outside directors does not guarantee that they will generally be guided solely by shareholder interests. We should seek to put in place the conditions under which outside directors can be expected to best perform their critical role. In this connection, I should note that opportunistic timing of outside directors’ grants was more likely to occur when the firm had more entrenching provisions (weaker shareholder rights) that protect insiders from the risk of removal.
Back to the Future
It might be argued that past backdating practices do not provide grounds for current concerns because they are unlikely to continue: with firms expected to comply promptly with filing requirements in the future, and with heightened attention by investors and regulators to grant-date selection, insiders can be expected to steer clear of grant-timing games. But this is hardly reassuring.
The past patterns of insider luck reflect persistent, widespread, and systemic governance problems: the existence of incentives to provide executives with increased performance-insensitive compensation below the radar screen; the use of compensation schemes that can be gamed by insiders; the prevalence of pay-setting processes not geared to maximize shareholder value; and the failures of internal monitoring systems. These problems do not go away just because regulators and increased outside scrutiny have shut off one (especially troublesome) practice produced by them.
Indeed, the problems of camouflage, decoupling of pay from performance, and failure of internal pay-setting processes—for which the backdating scandals have provided especially striking and vivid examples—have already had other significant manifestations, amply documented in Pay without Performance. Thus, while backdating may be a problem of the past, the backdating cases highlight the importance of continuing to seek ways to address basic problems in pay-setting processes and directors’ incentives.
Investors should press boards to make pay arrangements fully transparent, sensitive to performance, and not subject to gaming. There is little reason for executive compensation to depend on the particular choice of date for awarding equity incentives. Thus, the exercise price for options could be set at the average price level of the company’s shares during the month of the grant (or an even longer period). Similarly, executives should not continue to have broad freedom to unload equity incentives and to time the unloading based on their private information.
Because the design of options can easily be improved, investors should not be tempted to endorse the move by some companies from stock options to bonus compensation. The problem with stock options is not inherent to this instrument, but rather lies in how it has been used. To the extent that those designing pay arrangements do not have the right incentives, bonus payments are as susceptible—probably more susceptible—to gaming than stock-option plans. Indeed, bonus payments have historically been only weakly linked to performance. And insiders’ ability to game bonus compensation is currently greatly facilitated by companies’ common use of constantly shifting short-term performance metrics whose specifics are generally not disclosed to shareholders.
In the end, however, there is a limit to what imperfectly informed outsiders can do, which is why the most fundamental solution lies in improving the incentives and performance of corporate directors. To reduce the influence that executives have on the setting of their own pay arrangements, one possible approach is to require that the compensation of top executives, or at least of the CEO, not only be approved by the compensation committee but also ratified by a super-majority of the independent directors. Such a requirement would ensure that arrangements unable to gain widespread support among the company’s independent directors would not be adopted. Furthermore, directors’ own recognition that a small number of them can make a difference in pay decisions may counteract whatever factors might otherwise induce them to go along with flawed compensation arrangements.
Most important, to improve executive compensation we need to provide directors with stronger incentives to focus on and serve shareholder interests. Legal rules and corporate arrangements have long weakened shareholder rights and insulated the board from shareholder involvement. The rules governing corporate elections should be reformed to provide shareholders with a viable power to replace directors. And barriers to shareholders’ ability to place changes in governance arrangements on the ballot and adopt them should be dismantled. Even though significant backdating may belong to the past, its underlying causes are problems with which the corporate governance system must continue to wrestle. Reforms that make directors more accountable and more focused on shareholder interests are long overdue.
Lucian A. Bebchuk, LL.M. ’80, JF ’83, S.J.D. ’84, A.M. ’92, Ph.D. ’93, Friedman professor of law, economics, and finance and director of the Corporate Governance Program at Harvard Law School, is coauthor, with Jesse Fried ’88, J.D. ’93, of Pay without Performance: The Unfulfilled Promise of Executive Compensation (Harvard University Press, 2004). This article draws on his “Lucky CEOs” and “Lucky Directors,” both coauthored with Yaniv Grinstein and Urs Peyer, as well as on the reform proposals developed in his “The Myth of the Shareholder Franchise” and “The Case for Increasing Shareholder Power,” all available on line at www.law.harvard.edu/faculty/bebchuk.