Lucky CEOs and Lucky Directors

The December issue of the Journal of Finance features our article Lucky CEOs and Lucky Directors. This study integrates two discussion papers we circulated earlier, Lucky CEOs, and Lucky Directors.

Our study contributes to understanding the corporate governance determinants and implications of backdating practices during the decade of 1996-2005. Overall, our analysis provides support for the view that backdating practices reflect governance breakdowns and raise governance concerns. (For recent expressions of the opposite view that backdating did not reflect governance breakdowns, see the recent op-ed by WSJ columnist Holman Jenkins, who argues that backdating was a “meaningless accounting violation.”)

In particular, we find that:

• (i) Opportunistic timing has been correlated with factors associated with greater influence of the CEO on corporate decision-making, such as lack of a majority of independent directors, a long-serving CEO, or a lack of a block-holder with a “skin in the game” on the compensation committee;
• (ii) Grants to independent directors have also been opportunistically timed and that this timing was not merely a by-product of simultaneous awards to executives or of firms’ routinely timing all option grants;
• (iii) Lucky grants to independent directors have been associated with more CEO luck and CEO compensation;
• (iv) Rather than being a substitute for other forms of compensation, gains from opportunistic timing were awarded to CEOs with larger total compensation from other sources; and
• (v) Opportunistic timing was not driven by firm habit but rather, for any given firm, the use of such timing was itself timed to increase its profitability for recipients.

Our analysis suggests that the existence of CEO and director lucky grants as a variable that can be useful to research studying the governance and decision-making of firms. We therefore make available on the website of the Harvard Program on Corporate Governance a dataset (available here) of CEO and director luck indicators based on our work.

Here is a more detailed outline of what our paper (available here) does:

We study a large dataset of at-the-money, unscheduled option grants awarded to the CEOs and independent directors of public companies during the decade of 1996-2005. Our investigation focuses on “lucky” grants, which are grants given at the lowest price of the month. Opportunistic timing can result in an abnormally high fraction of grants being “lucky grants.” In our sample, about 15% of the grants to CEOs and 11% of the grants to directors were lucky before the adoption of the Sarbanes-Oxley Act (SOX). We contribute to prior work by investigating several questions and hypotheses that are relevant to assessing the corporate governance significance and the determinants of opportunistic timing practices.

To begin, our study Lucky Directors, which is integrated into the current article, provided the first systematic evidence that grants to independent directors have been affected by opportunistic timing in ways that link director luck and CEO luck. Even after the existence of opportunistically timed option grants to executives was widely recognized, it has been generally assumed that independent directors, who play an important oversight role in the current model of corporate governance, have not directly benefited from such timing. Our work, however, shows that awards to independent directors have been themselves opportunistically timed.

We further show that the timing of director grants has not been a mere by-product of director grants being awarded at the same time as executive grants. In particular, for any given firm and CEO, we find that the odds of a CEO grant being lucky were significantly higher when the independent directors of the firm received grants on the same date. In addition, director grant events not coinciding with awards to executives were also opportunistically timed and, moreover, were more likely to be lucky when the CEO received a lucky grant in the same or prior year. We also show that the timing of director grants has not been a by-product of firms’ routinely timing all option grants; all of our results concerning director luck continue to hold when one removes from the data (the small number of) firms that provided lucky grants to all grant recipients. By providing evidence that timing practices were structured in a manner consistent with making independent directors beneficiaries of these practices, we contribute an important input for assessing the role of independent directors both in connection with opportunistic timing practices and more generally.

We next analyze the relation between opportunistic timing and total reported compensation. We test, but do not find support in the data for, the hypothesis that gains from opportunistic timing were a substitute for other means of compensation. Controlling for size, performance, tenure and other firm and CEO characteristics that determine compensation, CEOs benefiting from lucky grants also received a significantly higher total compensation from other sources. Also, consistent with the hypothesis that independent directors receiving themselves lucky grants are less inclined to provide a check on compensation decisions, we find total CEO compensation to be higher (controlling for standard compensation determinants) in firms that granted lucky grants to independent directors.

We proceed to identify an association between opportunistic timing of CEOs’ and independent directors’ grants and certain aspects of firm governance and management. In particular, we find that opportunistic timing is correlated with variables associated with CEO influence over the internal decision-making processes: the lack of a majority of independent directors on the board and long CEO tenure. In addition, we find that a majority of independent directors on the board is only effective at reducing CEO luck if the independent directors did not themselves receive lucky grants. We further find that, although the existence of an independent compensation committee is not itself associated with a reduced likelihood of opportunistic timing, an independent committee with at least one large blockholder on it is associated with such a reduction. Our results highlight that the effectiveness of independent directors could well depend on factors other than their formal classification as independent.

Finally, with prior work focusing on cross-sectional differences among firms, we contribute by providing a time-series fixed-effect panel data analysis of opportunistic timing controlling for unobservable firm and CEO characteristics that could be correlated with opportunistic timing. This analysis enables us to test the hypothesis that, among the firms engaged in opportunistic timing, such timing was merely the product of routine. Inconsistent with this view, we find that, for any given CEO and firm, grants to both CEOs and directors were more likely to be lucky in months in which the potential payoffs from such luck were relatively higher. Thus, the selection of the period in which to engage in opportunistic timing was itself opportunistically timed. This pattern is consistent with the view that opportunistic timing reflects an economic decision that is sensitive to payoffs rather than a practice habitually followed by some firms.

Although firms did not commonly engage in opportunistic timing in all possible occasions, we do find evidence of significant persistence. The odds of a CEO’s grant being lucky were significantly higher, controlling for CEO and firm characteristics in our dataset, when a preceding grant to the CEO was lucky as well. These results indicate that, beyond the characteristics we identify as associated with opportunistic timing, there are some additional unobservable traits of firms and CEOs that lead to a higher tendency for opportunistic timing. Identifying these traits would be a worthwhile task for future work.