In the paper, CEO Education, CEO Turnover, and Firm Performance, which was recently made publicly available on SSRN, we analyze the effects of CEO education on CEO turnover (firing and replacement) and firm performance. Our primary interest is on the role that CEO education plays in a firm’s decision to replace its current CEO, the role that it plays in selecting a new CEO, and on whether CEO education significantly affects performance.

We use six main measures of CEO education: whether or not the CEO attended a Top-20 undergraduate school, whether or not the CEO has an MBA, law or masters‟ degree, and whether or not the MBA or law degree is from a Top-20 program. Our study includes more than 14,500 CEO-years and more than 2,600 cases of CEO turnover from 1993-2007.

…continue reading: CEO Education, CEO Turnover, and Firm Performance

In the paper New Dogs New Tricks: CEO Turnover, CEO-Related Factors, and Innovation Performance, which was recently made publicly available on SSRN, we examine the association between CEO turnover and innovation performance in the sample period 1993-2005. We find strong empirical support for the notion that CEO turnover is associated with higher levels of innovation, which we measure with patent counts and citations. After controlling for the endogeneity of CEO turnover, we present evidence consistent with CEO turnover increasing a firm’s patent counts, patent citations, patents per research and development dollar, and citations per patent in the subsequent three and five years. We also find that internal new CEOs create more innovation than external new CEOs. Moreover, we find that relatively overconfident CEOs, CEOs with higher option compensation, and an environment with relatively high information asymmetries are associated with more (and higher quality) innovation. Finally, we find that stock-option compensation and information asymmetries are negatively associated with subsequent innovation around the time of CEO turnover.

…continue reading: CEO Turnover, CEO-Related Factors, and Innovation Performance

In the paper, CEO Turnover and Retention Light: Retaining Former CEOs on the Board, forthcoming in the Journal of Accounting Research, unlike prior CEO turnover literature which characterizes the board’s decision as a choice between retaining versus replacing the CEO, we focus instead on the CEO’s decision rights and introduce a third option in which the incumbent CEO is removed but retained on the board for an extended period. We call this Retention Light.

Firms may benefit from Retention Light because former CEOs possess unique monitoring and advising abilities, but the former CEO could also exploit available decision rights for personal benefit. A Retention Light CEO’s decision rights generally exceed those of CEOs who exit the firm entirely but fall short of the rights of a retained CEO.

…continue reading: CEO Turnover and Retention Light

In the paper, Why Are CEOs Rarely Fired? Evidence from Structural Estimation, which is forthcoming in the Journal of Finance, I evaluate the forced CEO turnover rate and quantify effects on shareholder value by estimating a dynamic model. The model features costly turnover and learning about CEO ability. To fit the observed forced turnover rate, the model needs the average board of directors to behave as if replacing the CEO costs shareholders at least $200 million.

I find three main results. First, the empirical forced turnover rate is low, in the sense that the model needs large turnover costs to fit the data. Second, these costs mainly reflect CEO entrenchment rather than a real cost to shareholders. According to the model, eliminating this entrenchment would raise shareholder value by 3%, assuming we could hold all else constant.

…continue reading: Why Are CEOs Rarely Fired?

Shareholder representation by the board of directors is a fundamental component of corporate governance. A great deal of research has focused on the characteristics of corporate boards, yet we know little about uncontested director elections. The subject is particularly important in today’s environment. Congress, stock exchanges, and individual firms have instituted dramatic governance changes. Moreover, shareholders, activist organizations, the New York Stock Exchange (NYSE), and the Securities and Exchange Commission (SEC) have proposed and debated additional changes to the method by which directors are elected. Apart from directors, shareholders do not have representation in the companies they own. If shareholder impact on director elections is weak, so is the link between owners and managers.In our forthcoming Journal of Finance paper, Electing Directors, we examine the determinants as well as the efficacy of uncontested director elections on a large sample of firms in the post-Sarbanes Oxley Act (SOX) era. We test several hypotheses relating performance at both the firm and director levels to the votes directors receive. We also examine whether votes matter to subsequent performance, compensation, or governance.

Our sample consists of 13,384 director elections at 2,488 different shareholder meetings during 2003 to 2005. We find that while director and firm performance as well as corporate governance characteristics affect how shareholders vote, the resulting differences in the level of votes are trivial. In general, the differences in votes are statistically significant but economically minor. At both the firm and director levels, votes exceeding 90% are the norm even for poorly performing firms and directors. There are two exceptions: directors attending less than 75% of board meetings or receiving a negative ISS recommendation receive 14% and 19% fewer votes, respectively. However, even though the variation in director votes is small, we find that fewer votes for compensation committee directors significantly impact subsequent abnormal CEO compensation, and fewer votes for independent directors impact subsequent CEO turnover. Also, the removal of poison pills and classified boards is significantly linked to director votes. Nevertheless, lower levels of votes appear to have little impact on the election of directors themselves or on subsequent firm performance. Directors also do not appear to suffer reputational effects from low votes.

The full paper is available for download here.

My paper, co-authored with Julian Atanassov of the University of Oregon, was recently accepted for publication in the Journal of Finance. This paper investigates how labor and investors’ relative influence and firm level variables interact to affect corporate governance. A key conclusion is that weak investor protection combined with strong union laws are conducive to worker-management collusion harmful to investors.

Specifically, we analyze restructuring decisions when firms suffer a sudden, sharp deterioration in operating performance. We proxy for stakeholders’ relative influence at the country level by the strength of legal protection of investors and labor. We consider three types of restructuring measures: large scale employee layoffs, top management turnover, and major asset sales. Our sample consists of 9,923 firms (10,947 firm-years) at the onset of sharply declining operating performance in 41 developed and emerging economies over the period 1993 to 2004.

We find that poorly performing firms in stronger investor protection countries are more likely to undertake large-scale worker layoffs and replace top management than those in weaker investor protection countries. These restructuring actions are followed by superior operating performance in all legal environments. Major asset sales are different, however. We observe more asset sales when investor protection is either very strong or very weak. Asset sales in strong investor protection countries are followed by superior operating performance, whereas asset sales in weak investor protection countries are followed by inferior subsequent operating performance.

The likelihood of value-reducing asset sales increases as collective bargaining and labor relations laws grant more power to labor unions, suggesting that these asset sales are countenanced by workers. In addition, underperforming top managers in low investor protection countries are more likely to retain their jobs as union power increases. These results point toward management-worker alliances motivated by a mutual desire to retain jobs. For such an alliance to work, management needs funds to minimize layoffs and wage cuts. Lacking other means to raise the necessary funds, poorly performing firms sell assets to forestall layoffs even when doing so hurts subsequent operating performance. Indeed, asset sales in weak investor protection countries do not lead to layoffs, whereas in strong investor protection countries asset sales predict layoffs.

We also find that strong union laws are less effective in preventing layoffs when financial leverage is high, indicating that financial leverage is an effective instrument with which investors counter the power of workers.

Overall, our results highlight the importance of interaction among management, labor, and investors in shaping corporate governance.

The full paper is available for download here.

The Program on Corporate Governance has recently released our paper CEO Tenure, Performance and Turnover in S&P 500 Companies.  As Dennis Berman noted recently, our study identifies two groups of CEOs, ”owner-CEOs” and ”manager-CEOs,” and shows that manager-CEO retirements increase substantially during the fifth year of the CEO’s tenure.  The abstract of the piece is as follows:

The centrality of the CEO is reflected in the empirical literature linking CEO turnover to poor firm performance.  However, less is known about the institutional and personal correlates of CEO turnover.  In this study, we find two CEO characteristics interact with turnover: tenure and ownership.  We interpret our results as indicating that CEOs of S&P 500 firms divide into two groups with different tenure patterns–”owners” (who have large personal shareholdings) and ”managers” (who have smaller holdings).  The tenure of manager-CEOs (as opposed to owner-CEOs) exhibits a term structure loosely similar to the one produced by the tenure process at academic institutions.  Turnover of all kinds is low during a CEO’s first four years on the job.  In contrast, once a CEO reaches his fifth year, retirements begin a multi-year increase and exits via merger exhibit a large one-year spike.  These term effects are strongest for relatively young CEOs, and appear to be independent of such factors as firm performance or retirement norms.  We also find that deals and retirements are partially related, but partially distinct, modes of CEO turnover in other respects, which are similar along some dimensions but sharply different along others.

The full Article can be downloaded here.