Financial theory suggests that a firm with strong corporate governance (e.g., an effective board of directors), keeps a CEO exactly as long as is optimal for the firm. When the firm’s board and other corporate governance mechanisms are ineffective, however, a CEO can stay on longer than is optimal, to the detriment of shareholder value (Brochet et al., 2021).
Theoretical studies (e.g., Casamatta and Guembel, 2010) indicate that a CEO’s decisions affect a firm’s future performance, potentially beyond the CEO’s tenure. However, the empirical evidence for this finding is almost nonexistent. If a long CEO tenure is indeed associated with bad firm performance (as suggested by Brochet et al., 2021), then focusing on its long-term implications for the firm’s shareholders is called for. In a new paper, we contribute to the corporate governance literature by studying whether and how the prolonged tenure of the prior CEO affects decisions and firm performance under the new CEO. We find evidence that long-tenured predecessor CEOs harm the firm in a way that makes recovery from the damage more difficult in comparison with shorter-tenured predecessors. The incumbent CEO ends up conducting more extensive “clean-ups” (restructuring and asset write-offs)and obtaining positive results takes significantly longer
We construct a sample of CEO turnovers in U.S. listed firms between 1993 and 2013 using manual data collection and careful review of the nature of the turnover (firings, sudden deaths, etc.). Constructing such a sample allows us to focus on the predecessor CEO whose tenures lasted longer than seven years (the median tenure of a CEO in our sample) and compare them with short-tenured ones. Our analyses are in a difference-in-difference format, whereby the firm performance during the three years before a turnover is compared with the performance during the three years after the turnover (as in Huson et al., 2004). Using such an empirical setup, we document that when the predecessor CEO has a tenure longer than seven years, several performance measures such as Tobin’s Q, the abnormal stock return, and return on assets (ROA) are significantly lower in the post-turnover period.
We also study cleanups under the new CEO. We find that during the first few years following the CEO replacement, the firms with long-tenured prior CEOs report significantly larger divestments and restructuring costs and bigger write-offs, which are a form of accounting cleanup that allows the firm to start over with a clean slate. Although according to prior studies (e.g., Ali and Zhang, 2015), most CEOs engage in some form of cleanups or earnings management early in their career, our results are new in that they show a positive relationship between such cleanups under new CEOs and the tenure of the predecessor. More drastic changes seem necessary to wipe the slate clean after a long-tenured CEO, as shown in Figure 1:
Figure 1. Cleanup Measures Around Turnover Events
The figure shows the mean cleanup activity (writeoffs) around the CEO turnover year (t=0). The time window is three years before to three years after the event year [-3;+3]. For each t there are 2428 observations, which is the number of turnover events in our sample. The turnover sample is divided into long- and short-tenured subsamples, and writeoffs are plotted separately for each subsample. Long-Tenured Predecessor (LTP) is a dummy equal to 1 if the predecessor CEO had a tenure longer than median tenure in our sample (which is seven years).
The above findings are robust to a number of controls, which include using sudden deaths as an instrument in a self-selection model, estimating a model where we control for the endogeneity of the predecessor tenure, and using matching estimators and entropy balancing.
Our study also looks at the dynamics that allow some CEOs to stay in office longer than optimal. According to Shleifer and Vishny (1989), some CEOs have an incentive to entrench themselves through manager-specific investments (e.g., legacy-enhancing projects that are not necessarily the most value enhancing for shareholders). Several other studies note that CEO tenure is linked to a firm’s choice of investment (Weisbach, 1995; Pan et al, 2016). We find evidence consistent with these studies in that the post-turnover performance is significantly worse, and more frequently so, in firms with weak corporate governance measured by a governance index (the elements of which capture the dual role of the CEO, the amount of dependent directors, and the weakness or strength of external monitoring through low or high institutional ownership). This result suggests that over the years, long-tenured CEOs have become very powerful and entrenched within their firms and leave the firm requiring serious restructuring by their successor. This result is also in line with the predictions from the management literature on employee psychology and productivity (e.g., Shen and Canella, 2002).
An alternative explanation for low post-turnover performance can be an unusually good performance by predecessor. In this case, the firms would have been rational in keeping the CEOs for as long as they did. We refer to such cases as high-skill CEOs and show that controlling for CEO ability does not affect the relationship between tenure and post-turnover performance. We find only weak evidence that the long-tenured CEO’s ability explains post-turnover underperformance of the firm. Put differently, the corporate governance explanation seems more plausible than that of CEO skill.
Our findings may have policy implications for tying CEOs to their firms for longer times and for succession planning. If a CEO’s performance deteriorates when tenure is prolonged, then certain tenure-related corporate bylaws are called for. Such bylaws can be imposed through the firms’ charters or board polices of CEO term limits. The policy implications may also cast doubt on the practices of CEO remuneration, at least when it comes to contracts with a purpose of keeping the CEO with the firm longer.
Ali, A., Zhang, W., 2015. CEO tenure and earnings management, Journal of Accounting and Economics 59, 60-79.
Brochet, F., Limbach, P., Schmid, M., Scholz-Daneshgari, M., 2021. CEO tenure and firm value, The Accounting Review96 (6): 47-71.
Casamatta, C., and A. Guembel, 2010. Managerial legacies, entrenchment and strategic inertia. Journal of Finance65(6), 2403-2436.
Pan, Y., Wang, T., Weisbach, M., 2016. CEO investment cycles. Review of Financial Studies 29, 2955-2999.
Shen, W., Canella, A.A., 2002. Revisiting the performance consequences of CEO succession: The impact of successor type, post succession senior executive turnover, and departing CEO tenure, Academy of Management Journal 45, 717-733.
Shleifer, A., Vishny, R., 1989. Management entrenchment: The case of manager-specific investments. Journal of Financial Economics 25, 123-139.
Weisbach, M.S., 1995. CEO turnover and the firm’s investment decisions, Journal of Financial Economics 37, 159-188.
This post comes to us from professors Gonul Colak at the University of Sussex and Hanken School of Economics and Eva Liljeblom at Hanken School of Economics. It is based on their recent article, “Easy Clean-ups or Forbearing Improvements: The Effect of CEO Tenure on Successor’s Performance,” available here.