The amount of power to grant a CEO is one of the most important decisions corporate boards must make. CEO power has crucial implications for firm behavior. Yet, we still do not fully understand how CEOs become powerful over time. Under the optimal view, corporate boards reward CEOs with power for good firm performance. However, firm performance can be decomposed into a firm-specific and a luck component. Luck represents observable factors of firm performance that are outside the CEO’s control, such as market- or industry-wide conditions. In my recent paper, I ask the following question: are CEOs rewarded with power for luck?
Whether CEOs are rewarded power for luck or purely for firm-specific performance is an important question of the efficiency of corporate boards. Hermalin and Weisbach (1998) show how a CEO’s bargaining power depends on the board’s assessment of his/her ability. Boards learn about the CEO’s quality through delivered firm performance. Given that CEO ability does not depend on aggregate economic fluctuations in their model, boards are expected to filter out these random factors from firm performance while they evaluate the quality of the CEO (Jenter and Kanaan, 2015). It follows that in theory, efficient boards do not grant CEOs more power in good times than in bad times.
Empirical Methodology and Results
To test this prediction, I look at a sample of 3,873 CEOs who manage firms in the S&P 1500 index from 1996 to 2018. I consider CEOs as powerful if they can affect corporate decisions on a constant basis, despite possible resistance from other board members and/or executives. CEOs holding multiple titles and having high equity stakes are expected to have more say in a company’s decision-making process. Thus, I construct a CEO power index which is equal to the sum of four binary variables: chairman, president, founder, and high-ownership CEO. A high-ownership CEO owns more than 5% of the firm’s equity. For instance, a CEO who is the chairman of the board, president and founder of the firm, and owns 2% of the firm’s equity will have a power score of 3 out of a possible 4. Next, I estimate the impact of luck on CEO power by following a two-stage regression approach. In the first-stage regression, I decompose firm stock return into a luck and a firm-specific component using a simple market model with market-wide return as the single luck factor. Therefore, luck refers to the market-induced component of a firm’s stock return that is not attributable to CEO ability or actions. For example, a company may be doing well just because the overall market is performing well. Then, in the second-stage regression, I examine the sensitivity of the CEO-power index to luck and firm-specific stock return.
Across more than 80 different specifications, I find that CEOs are rewarded with power for luck. In the main specification, a one standard deviation increase in firm performance due to luck leads to a 3% increase in CEO-power relative to the median. In contrast, my results suggest that CEO power is not as sensitive to firm-specific performance compared with luck.
Why are CEOs Rewarded with Power for Good Luck?
CEOs may have incentives to strategically increase their power following good firm and market performance. Why would CEOs follow this strategy? Given that shareholders lack a filtering mechanism, they are more likely to misattribute luck to firm-specific performance, and hence, tolerate additional CEO power at good times. Moreover, shareholder attention to CEO power is arguably lower when good firm performance coincides with good market-wide performance. As a result, timing entrenchment in good times might enable CEOs to avoid a severe negative reaction to their firm’s stock price. If CEOs behave this way, it is more likely to be prevalent later in their tenure or in weakly governed firms, where they face less intervention, and as such, have a higher probability in gaining power. Governance is expected to be weaker in boards with a large fraction of captured and/or inside directors. Captured directors are those elected after the CEO assumes office, whereas inside directors are other executives who sit on the board.
My findings are consistent with predictions of strategic timing. First, increases in CEO power due to luck are primarily driven by firms with weaker governance. Consider a board with 10 members where 5 of the directors are captured. Replacing one non-captured director by one who is captured significantly increases power for luck by 15%. Similarly, adding one inside director to a board that initially had 8 outsiders and 2 insiders, increases the power for luck coefficient by 41%. Second, I find some evidence of an asymmetric effect in weakly governed firms where gains in CEO power for good luck are not matched with equivalent losses in power for bad luck. For firms with a high number of captured directors (weak governance), a one standard deviation increase in luck due to good luck leads to a gain of 5.98% in power relative to the median. For an equivalent decrease in luck when luck is bad, a CEO loses only 0.98% in power relative to the median. Finally, CEOs are rewarded with more power for luck as their tenure increases.
An alternative explanation for why CEOs are rewarded with power for luck relates to the CEOs’ outside opportunity. If the outside option of CEOs is positively correlated with market-wide performance and CEO talent is scarce, then talented CEOs will be rewarded power when markets perform well, as firms have a higher incentive to retain them. Lastly, boards could be incapable of filtering out luck from firm performance. While I cannot completely rule out the last two possibilities, I do not find strong evidence in their support.
My main contribution in this paper is to identify luck as a new channel through which CEOs gain power. Standard economic theory predicts that corporate boards filter out luck from firm performance while determining the power of the CEO. My findings stand in contrast to this prediction and imply that decisions related to CEO power, in the real-world, are impacted by luck. Overall, these results offer additional insights on how CEOs become powerful over time and on the efficiency of corporate boards.
Turk Al Sabah is a Ph.D. student in finance at Kenan-Flagler Business School, University of North Carolina at Chapel Hill.
This post is adapted from his paper, “CEO-Power and Luck: Impact of Stock Markets on Building Powerful CEOs,” available on SSRN.