The impact of gender on voluntary and involuntary executive departure.

Becker-Blease, John R. ; Elkinawy, Susan ; Stater, Mark 等

I. INTRODUCTION

The employment of women in U.S. corporations has increased dramatically since the early 1990s, particularly in the highest ranking positions. Bertrand and Hallock (2001) document that, between 1992 and 1997, women almost tripled their participation in the top corporate jobs. Along with the increase in female representation in executive labor markets, researchers and the media are paying increased attention to the effectiveness of female executives. Despite the recent advancements of women in corporate hierarchies, relatively little is known about whether female executives are more likely to be fired than men or whether females who depart their positions do so for different reasons than men, particularly in response to poor firm performance or differences in firm governance structure. This issue has received media attention in recent years. For example, the year 2005 was characterized by Jones (2005) as "a miserable year for female CEOs of Fortune 500 companies, as female-headed companies trailed the Standard & Poor's (S&P) 500 Index for the second straight year."

The termination of powerful female executives is often a high-profile event, as in the cases of Carly Fiorina and Patricia Dunn, who were dismissed from Hewlett-Packard in 2005 and 2006, respectively. Although a few well-publicized cases cannot serve as a valid basis for broad generalizations about the executive labor market, these recent anecdotes suggest that the issue of whether there are gender differences in the circumstances of executive departure warrants investigation.

We track a sample of executive departures from S&P 1500 firms between 1996 and 2004. We classify executive departures as either voluntary or involuntary based on a careful examination of public news accounts accompanying the departures. We then examine whether the frequency and conditions of voluntary or involuntary departure differ for men and women. In general, we find systematic evidence that female executives are more likely than male executives to depart both voluntarily and involuntarily from their positions. Further analysis suggests that although executives are more likely to depart involuntarily following poor corporate performance and when monitored by more effective boards, the relative impacts differ depending on gender. Women are less likely than men to depart voluntarily and in general when firm size increases or the size of the board decreases but are more likely to depart involuntarily when the board becomes more male dominated. Thus, the evidence suggests that in spite of recent advancements by women into the executive ranks, their position at these ranks can be tenuous, particularly at smaller firms and firms with larger, more male-dominated boards.

The remainder of this article is organized as follows. Section II discusses the related literature on gender, executive turnover, and firm valuation measures. Section III describes the data and descriptive statistics. Section IV explains the hypotheses and empirical methods that underlie our analysis as well as the results of the analysis. Section V reports sensitivity tests that explore the robustness of our findings to alternative samples and classifications of departure reasons. Section VI concludes.

II. LITERATURE

Gender Diversity and Firm Valuation

Several studies find a positive relation between the presence of females in the executive ranks and firm value. The Economist (2005a) reports that research in the United States, United Kingdom, and Scandinavia shows a strong correlation between the proportion of women in executive positions and shareholder returns. Catalyst, a research organization specializing in women's career advancement, finds that firms with the highest representation of women on their top management teams outperform firms with the lowest representation of women (Catalyst 2004). Adler (2001) studies a sample of Fortune 500 firms over a 19-yr period and finds that firms with the best record of promoting women are more profitable than the median firms in their industries. Carter, Simkins, and Simpson (2003) find that gender diversity on the board of directors has a positive effect on Tobin's Q. More generally, the literature on the relationship between board diversity and firm valuation measures is extensive. (1)

Not all studies find a positive association between gender diversity among executives and firm performance, however. Shrader, Blackburn, and Iles (1997) examine management data for the 200 largest U.S. firms in 1992 and find that higher percentages of women on the top management team and the board of directors have no effect on financial performance. Carleton, Nelson, and Weisbach (1998) find that firms targeted by the Teacher's Insurance and Annuity Association, College Equities Retirement Fund (TIAA-CREF) for lack of gender and ethnic diversity on the board experience significantly negative cumulative abnormal returns surrounding the dates of board diversity targets. Lee and James (2007) find event-study evidence that announcements of female CEO appointments are viewed more negatively by the market than reactions to male CEO appointments. Similarly, Wolfers (2006) does not find that markets systematically undervalue firms led by females.

Although the effects of racial and gender diversity on firm value remain unclear, several major U.S. firms have instituted diversity programs to help underrepresented groups gain improved access to top management positions. The move toward increased diversity is not limited to the United States. The Economist (2005b) reports that "in Britain, the number of female executive directors in FTSE 100 firms rose from 11 in 2000 to 17 in 2004." According to The Guardian (2006), legislation was passed in Norway in 2003 requiring all public companies to have 40% female representation on their boards of directors as of January 1, 2006. After taking effect, the law also gave firms an additional 2 yr from that date (i.e., until January 1, 2008) to comply with the requirement or face sanctions up to and including the closure of the firm. The Scotsman (2007) reports that since 2003, the percentage of female board members in Norway has jumped from 6% to 37%. However, it is expected that many firms will not meet the quota and may be forced to shut down as a result.

Executive Turnover. The impact of gender on executive turnover has received comparatively little research attention. Stroh, Brett, and Reilly (1996) document turnover rates of male and female managers employed by 20 Fortune 500 firms and find that females leave their organizations in higher proportions than males. In contrast, Lewis (1992) examines middle managers in the U.S. federal civil service and finds insignificant differences in turnover rates between men and women. Lyness and Judiesch (2001) examine voluntary turnover for more than 26,000 managers in a financial services organization and find that the turnover rate among female managers is slightly lower than that of male managers. Elvira and Cohen (2001) study turnover differences between sexes at various organizational ranks and find that the proportion of female executives in the firm has no effect on the turnover of top-ranking women. However, we are aware of no evidence that exists on the relation between gender and involuntary dismissals.

Empirical evidence generally suggests that poor corporate performance precedes CEO departure and the departure of lower level executives. (2) Morck, Shleifer, and Vishny (1989) find that the sensitivity of the relation between performance and turnover is affected by the relative performance of the firm within its industry. Parrino (1997) finds that the performance-turnover relationship depends on the homogeneity of firms within the industry, while Mikkelson and Partch (1997) and Denis and Kruse (2000) find that it depends on the takeover intensity within the industry. Similarly, Weisbach (1988) and Denis and Denis (1995) find that board independence and external monitoring affect turnover probabilities. Thus, turnover is associated with firm-level performance, firm governance, and industry characteristics. We therefore incorporate measures of market- and industry-adjusted financial and stock market performance as well as characteristics of the board to predict the reasons for executive departure.

III.DATA AND DESCRIPTIVE STATISTICS Data

EXECUCOMP includes a listing of at least the top five executives in each firm in the S&P 1500. We begin with the EXECUCOMP universe of firms between 1996 and 2004 and extract the 98,990 unique executive-firm-year observations during this period. In addition to name, position, and compensation, EXECUCOMP reports data on gender, tenure, equity ownership, the presence of the executive on the board of directors, and, less frequently, the age of the executive. In fact, EXECUCOMP only reports the ages of about 10% of the executives. We therefore augment data on ages with hand-collected information from proxy statements, annual reports, and news stories. We are able to identify ages for an additional 36,557 executive-firm-year observations using these alternative sources as well as data from the Investor Responsibility Resource Center (IRRC) database described below, resulting in age data for roughly 50% of the sample.

EXECUCOMP lists a departure date for those executives who leave the firm. We use this field to indicate executive departures. Although EXECUCOMP also lists a reason for departure (RESIGNED, RETIRED, DECEASED, or UNKNOWN), these are not sufficient to classify the departure as voluntary or involuntary. We therefore follow the procedure outlined in Mian (2001) that uses public accounts of the departure from news stories to classify each departure as "voluntary" or "involuntary." (3) Specifically, for each departing executive, we search Lexis-Nexis for news stories related to the firm in general and the named executive in particular for up to 12 mo surrounding the departure date. (4) Based on the news and corporate events surrounding the departure announcement date, we classify departures as either voluntary or involuntary. Specifically, involuntary departures include firings that are specifically due to illegalities, fraud, or accounting manipulations; discipline for poor performance of the executive or firm; outright firings for which detailed specific reasons are not given; sudden departures where no reason was provided to suggest it was of the executive's own accord; and terminations that are related to firm restructuring or mergers. Voluntary departures include resignations that fall into the following categories and where there is no evidence that the decision was forced: exits surrounding disagreements with management and/or the board (but that are clearly initiated by the executive); exits for voluntary professional reasons (i.e., accepting a new position or starting a new business); exits due to health, family, or personal reasons; and retirements. We exclude observations in which the executive dies and those in which we can find no information regarding the conditions under which the executive departed.

Annual board-level data come from the IRRC database. We collect information on the size of the board, the proportion of independent and male directors, and, where missing from the EXECUCOMP database, the ages and equity ownership of executive directors.

Finally, firm stock and operating performance values, as well as industry values, are collected from the Compustat and CRSP databases. As described below, we require lagged values of certain performance measures to be available on CRSP and Compustat. This restriction in combination with data limitations from each of the databases reduces our final sample to 53,311 executive-firm-year observations, of which we have executive ages for 28,193.

Descriptive Statistics. In Table 1, we provide summary statistics on departure and the various reasons for departure. We report means for all executives, as well as separate means for men and women, departed executives, male executives who depart, and female executives who depart. Approximately 3.9% of the executive-year observations in the sample are departures; 1.0% of the observations are involuntary departures and 2.9% are voluntary departures. The modal category of involuntary departures is that of sudden departure with no reason given, which accounts for 14.4% of all departures (see the "Departed Execs" column), and the modal category of voluntary departure is retirement, which accounts for 41.0% of departures. Among executives of the same gender, the percent of women who depart is higher than the percent of men who depart (7.2% and 3.8%, respectively). The percent of women who involuntarily depart is also higher than the percent of men who involuntarily depart (2.9% and 0.9%, respectively) and a similar pattern holds for voluntary departure (4.3% for women vs. 2.8% for men). This suggests that women are more likely than men to depart for any reason, although these descriptive results do not control for other factors affecting departure outcomes.

Table 2 provides descriptive statistics for the characteristics of the executives and the firms in which they are employed. We report summary figures for the full sample as well as separate statistics for departures, nondepartures, involuntary departures, and voluntary departures. These variables include the firm's total assets, return on assets, and stock returns. Here and throughout, dollar values are measured in constant 1994 dollars based on the consumer price index. (5) Return on assets, measured as the ratio of operating income to total assets, is adjusted by the median among other firms in the same Fama and French (1997) industry code. Monthly stock returns are predicted based on an ordinary least squares regression model that includes industry and market factors, and then a monthly abnormal return is formed by taking the difference between the actual and the predicted returns of the firm where the executive is employed. A buy-and-hold abnormal return is calculated over a 1-yr period preceding the beginning of each year listed for each executive in the sample. (6)

The means indicate that 4.5% of the observations in the sample are on female executives. However, the proportion of departing executives who are female is higher (8.2%) than the proportion of nondeparting executives who are female (4.3%). Similarly, the proportion of involuntarily departing executives who are female is higher than the proportion of voluntarily departing executives who are female. Thus, although both departing and nondeparting executives are far more likely to be male than female (since the vast majority of the sample is male), women are overrepresented among departing and involuntarily departing executives. There are also significant differences in the means of other variables among departing and nondeparting executives as well as among involuntarily and voluntarily departing executives. For instance, departing executives are older and more experienced than nondeparting executives and also work in firms with lower return on assets, lower cumulative abnormal returns, lower fractions of male directors, and higher fractions of independent directors. Comparing involuntarily versus voluntarily departing executives, we find that involuntarily departing executives are younger and less experienced, are higher paid, work in firms with lower return on assets and cumulative abnormal returns, and work in firms with smaller, less independent boards.

IV. HYPOTHESES AND EMPIRICAL ANALYSIS

Hypotheses

In this section, we develop hypotheses for the departure outcomes. If firms make personnel decisions to maximize shareholder value, then we expect firms to be more likely to dismiss executives (i.e., involuntary departure) when the opportunity cost of doing so in terms of foregone shareholder value is low. Women should then have higher likelihoods of involuntary departure than men if firm owners have a preference for male executives (i.e., if there is discrimination in the executive labor market). Of course, a higher likelihood of involuntary departure for women is also consistent with nondiscrimination explanations, such as lower unobserved human capital or weaker labor force attachment. In light of previous studies such as Weisbach (1988) and Yermack (1996), we expect relatively high likelihoods of involuntary departure for executives who are older, less experienced (controlling for age), own smaller amounts of firm equity, work in more poorly performing firms, or work in firms with smaller or more independent boards.

The effects of some controls may differ by gender. If firms have a preference for men, or women have greater difficulty demonstrating their value in a male-dominated workplace, then age and compensation should have stronger effects for women; experience, equity ownership, and firm performance should have weaker effects. If large firms are more "female friendly" than small firms, firm size should reduce the likelihood of involuntary departure more for women. If male-dominated boards are relatively less supportive of female executives, increases in male board representation should increase the likelihood of involuntary departure more for women.

Assume executives pursue employment opportunities that maximize their utility, given their preferences and human capital. Then, we expect executives to be more likely to voluntarily depart their positions when the opportunity cost of leaving is low or the returns to leaving are high. Thus, women should have higher likelihoods of voluntary departure than men since it can be argued that even highly educated female professionals tend to have greater household production responsibilities and higher wage-earning spouses than their male counterparts. Female executives may also have lower expectations of career advancement due to perceptions of glass ceilings. Following the literature, we also expect relatively high likelihoods of voluntary departure for executives who are lower paid, own smaller amounts of firm equity, work in larger or more poorly performing firms, or work in firms with smaller or more independent boards.

Some effects may again differ by gender. If women tend to have higher nonlabor income than men (e.g., due to higher wage-earning spouses) then compensation should increase the likelihood of voluntary departure more for women. If large firms are more female friendly, firm size should reduce the likelihood of voluntary departure more for women. Firm performance should also reduce the likelihood of voluntary departure more for women if glass ceilings in the profession limit the mobility of females and increase the returns to employment in high-performing firms. Finally, male board representation is expected to increase the likelihood of voluntary departure more for women.

Empirical Model. We begin by using the full sample of executives to estimate the probability that a given executive departs his or her position in a given year without distinguishing the reasons for departure. Thus, our first set of models estimate the probability of general departure based on executive, firm, and board covariates that plausibly relate to the opportunity costs of voluntary and/or involuntary departure. Our estimation technique is random effects logit because there are observations available in multiple years for the majority of executives in our sample. Random effects is employed instead of fixed effects because the effects of time-invariant covariates are central to our analysis, and few executives in our sample are observed to depart in some years and not depart in other years, as would be required for identification under fixed effects. (7) Thus, our model of departure is:

Pr([D.sub.it] = 1) = [LAMBDA]([[beta].sub.E][x.sup.E.sub.it] + [[beta].sub.E][x.sup.F.sub.it] + [delta][T.sub.t] + - [mu].sup.F.sub.i]), (1)

where [D.sub.it] is a binary variable that equals one if executive i departs his or her position in year t and equals zero otherwise, [LAMBDA] is the cumulative logistic distribution function; [x.sup.E.sub.it] is a vector of characteristics describing the executive in year t (some of these characteristics are time invariant); [x.sup.F.sub.it] is a vector of characteristics describing the firm where executive i is employed in year t; [T.sub.t] is a vector of year dummies for each of the years in our sample excluding the final year; and [[mu].sup.F.sub.i] is unobserved executive-firm heterogeneity. This term captures all unobserved factors that are constant for a given executive during all years spent with a given firm (e.g., attitudes of the executive toward work versus leisure, innate managerial ability, educational background, attitudes of the firm about when to fire executives, etc.). We specify executive-firm heterogeneity instead of just executive- or firm-specific heterogeneity to allow the unobserved component of departure to differ across firms for the same executive and across executives for the same firm. (8) However, note that anything that is constant across all executives in the same firm (such as the firm's tastes for firing executives) will also be constant for a given executive during his or her time with the same firm and consequently will be subsumed into [[mu].sup.F.sub.i]. The random effects model is consistent and efficient under the assumption that [[mu].sup.F.sub.i] is uncorrelated with [x.sup.E.sub.it] and [x.sup.F.sub.it].

The vector [x.sup.F.sub.it] includes the gender and age of the executive as well as a quadratic in age to allow for a depreciating effect, the executive's tenure with the firm, a dummy variable for whether or not the executive is a director of the firm, dummy variables for whether the executive is a CEO, CFO, or COO, (9) the natural log of the executive's total annual compensation (which consists of base salary, bonuses, and fringe benefits), and the natural log of the dollar value of the firm's shares owned by the executive. The vector [x.sup.F.sub.it] includes the natural log of the firm's 1-yr lagged total assets, the firm's 1-yr lagged industry-adjusted return on assets (i.e., the industry-adjusted return on assets for year t-1), the firm's 1-yr lagged buy-and-hold abnormal stock return adjusted for market and industry factors (i.e., the sum of the monthly abnormal returns for year t-1), the total number of directors on the firm's board, the fraction of directors who are men, the fraction of directors who are independent, and a set of industry dummies based on Fama and French (1997) industry groupings. (10) We use lagged values of the firm size and performance measures (total assets, return on assets, and abnormal returns) to avoid potential endogeneity between currentyear values and executive departure or nondeparture. These could be endogenous because whether or not the executive departs could affect contemporaneous measures of performance. However, the same problem should not pertain to prior-year performance.

We also estimate random effects logit models of the probabilities of involuntary and voluntary departure with the same set of controls as the general departure model:

Pr([I.sub.it] = 1) = [LAMBDA]([[beta].sub.E] [x.sup.E.sub.it] + [beta].sub.F] + [x.sup.F.sub.it] + [delta][T.sub.t] + [[mu].sup.F.sub.i]) (2)

Pr([V.sub.it] = 1) = [LAMBDA]([[beta].sub.E] [x.sup.E.sub.it] + [[beta].sub.F] [x.sup.F.sub.it] + [delta][T.sub.t] + [[mu].sup.F.sub.i]), (3)

where [I.sub.it] and [V.sub.it] are dummy variables that equal one if the executive departs involuntarily or voluntarily, respectively, in year t. All other controls are defined as in Equation (1).

Baseline Results. The estimates of two specifications of each Equations (1)-(3) are reported in Table 3. For each departure category, Model (1) excludes age, age squared, and tenure, while Model (2) includes these variables. Because age and tenure are unavailable for some of the executives, the regressions for Model (2) have smaller sample sizes than those for Model (1).

The results suggest that, holding constant executive and firm characteristics, women are 3.4-6.8 percentage points more likely to depart than men, 1.3-1.5 percentage points more likely to involuntarily depart, and 1.5-3.8 percentage points more likely to voluntarily depart. While the general and voluntary departure results are consistent with women having a higher return to opportunities outside the firm, the involuntary departure result is consistent with corporations having a greater preference for male executives. However, it is also consistent with gender differences in unobserved human capital and with lower labor force attachment on the part of women executives.

Many of the other controls also affect the departure outcomes. Age has a positive but diminishing effect on the likelihood of each type of departure. Experience in the firm reduces the likelihood of general departure and involuntary departure but does not affect the likelihood of voluntary departure. Executives with higher total compensation are less likely to depart in general and less likely to depart voluntarily than those with lower total compensation but are no more or less likely to involuntarily depart. While higher paid executives clearly have a greater opportunity cost of leaving their positions (reducing the likelihood of voluntary departure), compensation may also serve as a proxy for innate executive ability, making the executive more attractive to the firm despite the lower return the firm earns on its investment when compensation increases at a given level of firm performance. Thus, compensation may have offsetting effects on the likelihood of involuntary departure. Executive share ownership has a negative effect on all types of departure, (11) consistent with the Jensen and Meckling (1976) finding that firm ownership aligns managers' interests more closely with those of shareholders.

Firm size has a positive effect on all three categories of departure, which suggests that executives in larger firms are more mobile in the market and/or under greater scrutiny from their employers. Higher returns on assets and abnormal stock returns reduce the probability of all types of departure, consistent with the findings of Huson, Malatesta, and Parrino (2004). Our findings therefore support the notion that executives and firms find the employment relationship more mutually beneficial under conditions of strong firm performance.

Executives employed by firms with larger boards are less likely than those in firms with smaller boards to depart in general and to involuntarily depart but are no more or less likely to voluntarily depart. These results are consistent with the notion that smaller boards are more effective monitors, which may enhance their ability to accurately detect managerial performance lapses that are grounds for executive terminations. The fraction of the board that is male has no effect on any type of departure, while the fraction of the board that is independent has positive effects on all types of departure, consistent with more independent boards being better monitors as well as less appealing to work with than boards that are less independent.

Results with Gender Interactions. Table 4 reports coefficient estimates from random effects logit models of general, involuntary, and voluntary departure that include a full set of interactions between the female dummy and the control variables in Equations (1)-(3). This approach allows us to test for the possibility that the effects of the controls differ for men and women. We report coefficients instead of marginal effects because it is problematic to assign a meaningful interpretation to the marginal effects of interaction terms in a nonlinear model. (12) Thus, the results that we presently report will speak primarily to the directions (rather than the magnitudes) of the differences in the effects of the control variables between men and women.

For each departure outcome, we again estimate two models. Model (1) excludes the age and tenure controls, while Model (2) includes them. For each departure category, we report the coefficients for men, which are the coefficients on the uninteracted forms of the variables (first column for each model), and the female-male differences in the effects of the controls, which are the coefficients on the interaction terms (second column for each model).

The results in Table 4 indicate that the female dummy is negative in one case and insignificant in the others. However, note that intercept differences are less meaningful in a model that also allows for slope differences. With a full set of slope differences, a lower female intercept merely indicates that women are no more likely than men to depart when the values of all controls are set equal to zero. When the values of the controls are set equal to their means, however, females have higher predicted probabilities of all types of departure, as shown in the last row of the table. (13)

The coefficients for men (those on the uninteracted variables) are similar to those obtained when the effects of the controls were constrained to be the same across gender. Indeed, the effects of all covariates except the characteristics of the board are the same in sign and significance for men as they are for men and women combined, which is unsurprising given the high male representation in the sample. However, when the age and tenure controls are included, the size of the board now has a negative effect on voluntary departure instead of an insignificant effect, which suggests men prefer to work with larger boards. The fraction of male directors now has negative rather than insignificant effects on the likelihoods of general and voluntary departure, which suggests men prefer to work with more male-dominated boards.

The estimates for the interaction terms reveal that several of the controls have different effects on the various types of departure for women than they do for men. Considering first the models for general and voluntary departure, compensation increases the likelihood of both types of departure more for women than men, which suggests increases in compensation increase the opportunity cost of leaving a position more for men. Total assets reduce the likelihood of both types of departure more for women than men, indicating women become relatively less willing to leave their positions when firm size increases. This is consistent with larger firms being more hospitable to women than smaller firms. Larger firms may, for instance, offer more generous family leave programs, on-site childcare, or more flexible scheduling.

Some board characteristics also have different effects on general and voluntary departure for men and women. The effect of an increase in board size on both outcomes is greater for women, indicating that women are relatively more willing to leave their positions as the size of the board increases (holding overall firm size constant). It may be that women find it more difficult to gain influence or respect when working with larger boards. The fraction of men on the board also has greater effects on general and voluntary departure for women (provided age and tenure are excluded), indicating that women are relatively more willing to leave their positions when the fraction of male board members increases. The fraction of independent directors reduces the likelihood of voluntary departure for women relative to men (provided age and tenure are included), suggesting that women have a greater preference than men for working with independent boards. A possible explanation is that more independent boards are less intertwined with old boy networks than are boards consisting of a higher fraction of insiders.

Most of the gender differences in the involuntary departure model are insignificant. (14) The lone exception is the percent of male directors on the board, which has a greater effect for women than men. This suggests that women are more likely than men to be dismissed when the fraction of males on the board increases, which is consistent with male-dominated boards having a greater preference for male executives.

V. SENSITIVITY TESTS

We perform a variety of sensitivity tests on our models due to the subjective nature of the classification scheme used to distinguish between voluntary and involuntary departures. Although in the interests of brevity, we do not report these results in separate tables, we discuss them briefly. Our first sensitivity test is to exclude retirements since these departures are considered voluntary (in the absence of evidence of pressure from the firm) and men are much more likely to retire than women. Thus, excluding retirements could potentially change the inference of gender differences in the circumstances of departure. However, most of the results are unchanged by the exclusion of retirements, which may be because we have included age controls and conducted intensive scrutiny of news accounts related to retirements to ensure that they were appropriately classified. Most notably, women remain more likely to depart in general, to involuntarily depart, and to voluntarily depart when retirements are excluded. However, the results for compensation are affected by the exclusion of retiring executives in that the effect of compensation on general departure becomes insignificant rather than negative when age and tenure are excluded. This may be because the exclusion of executives who retire results in a younger pool of executives and younger executives become more mobile than older executives as compensation increases. The results for the fraction of male directors in the models with gender interactions are also different when we exclude retirements in that the effect of the fraction of male directors on general departure is no longer negative for men. This suggests that old boy networks in firms with male-dominated boards are more beneficial for older executives who are a lower percentage of the sample when retirements are excluded.

In specifications that are also unreported, we adopt an alternative classification scheme that moves merger-related departures into the voluntary category. While adopting this new scheme, we bring retirements back into the sample and continue to consider them as voluntary departures. This change has little effect on the results for involuntary or voluntary departure. Again, women are more likely to depart both voluntarily and involuntarily, and the fraction of male directors again has a greater effect on involuntary departure for women than for men. This strengthens some of our prior inferences that women are more likely than men to be dismissed as the board becomes more male dominated.

In a final set of tests, we adopt a classification scheme that moves all departures except firing for fraud and poor performance into the voluntary category. Thus, we now consider as involuntary only those departures that are for the most clearly disciplinary reasons. This change in the classification scheme produces some interesting changes. We now find that age has no effect on involuntary departure, suggesting that the bulk of dismissals of older executives are for nondisciplinary reasons. We also now find that being a director has a consistently positive effect on involuntary departure, suggesting that the dismissals of directors tend to be for clearly disciplinary reasons. The effects of return on assets and board independence are now insignificant, indicating that increases in firm operating performance do not help prevent the most clearly disciplinary departures, while the additional firings undertaken by more independent boards tend to be for reasons other than outright fraud or demonstrably poor performance. The effect of return on assets on general departure also becomes significantly more negative for women than for men, indicating that better operating performance is relatively more helpful to women in reducing the chances of the most clearly disciplinary dismissals. Despite the aforementioned changes, our key results regarding the gender differences in the probabilities of voluntary and involuntary departure are robust to this reclassification.

VI. CONCLUSIONS

Through antidiscrimination policies, changing cultural attitudes, and evolving labor force participation trends, women are becoming better represented in executive labor markets over time. Nevertheless, recent widely publicized events have raised the concern that corporations are relatively quick to dismiss female executives. Because this sentiment seems to arise from intense media scrutiny of a few cases, we closely investigate the reasons for departure in a sample of male and female executives from the EXECUCOMP database. Specifically, during the period 1996-2004, we classify departures into categories of voluntary and involuntary departure and examine the determinants of these departure types as well as of general departure.

Our key findings are that women are more likely to depart, to involuntarily depart, and to voluntarily depart than men, controlling for firm performance, governance characteristics, and executive human capital. These results are robust to specifications that include age and tenure controls, that exclude retirements, and that use different classification schemes for involuntary and voluntary departures. Thus, our evidence is supportive of a discrimination hypothesis but cannot definitively rule out alternative explanations such as gender differences in unobserved human capital or labor force attachment or more media attention given to female departures. Moreover, because we find women are also more likely to voluntarily depart, the evidence also supports the notion that women have higher returns in nonlabor market activities than do men.

Our remaining findings are supportive of labor market and agency theories. Generally, the probability of involuntary departure is high when the opportunity cost to the firm of dismissing an executive is low, the probability of voluntary departure is high when the opportunity cost to the executive of leaving a position is low, and the probability of general departure is high in circumstances conducive to both voluntary and involuntary departure.

We also estimate models with gender interaction terms to test whether the effects of the controls on the various types of departure are different for women and men. We find differences in the determinants of general and voluntary departure that suggest a greater preference on the part of women relative to men for working in larger firms, firms with smaller boards, and firms with less male-dominated boards. Furthermore, the involuntary departure results provide suggestive evidence that male-dominated boards are more likely to dismiss women than men.

While we have attempted to provide rigorous evidence on the existence of gender differences in the reasons for departure, we acknowledge that there are some limitations to our analysis. Any classification scheme that codes departures into voluntary and involuntary categories is inherently subjective, so that it is sometimes difficult to state with absolute certainty whether the departure of an executive is voluntary or involuntary. In addition, age controls are available for only a portion of the sample, so a verdict on the robustness of our full-sample results must await the availability of a more complete data source on executive ages. However, the similarity of the results with and without the age controls offers some reassurance on this front. Finally, one could always desire more detailed controls for the professional and cognitive ability of the executive as well as educational background. The title, director, and compensation variables that we use here are indirect controls for qualifications and skills. Likewise, more direct controls for the outside opportunities available to the executive would allow us in some cases to draw more precise conclusions about the motivating factors behind a departure. Therefore, further work is required to definitively disentangle whether our results reflect discrimination or simply an efficient labor market where preferences and incentives differ for women and men.

ABBREVIATIONS

FTSE: Financial Times Stock Exchange

IRRC: Investor Responsibility Resource Center

S&P: Standard & Poor's

doi: 10.1111/j.1465-7295.2008.00186.x APPENDIX

Reasons for Executive Departures from Lexis-Nexis News Reports Classifications Based on Mian (2001) Involuntary Departures

(A) Firings or resignations for fraud associated with accounting irregularities or illegalities where the executive is clearly blamed.

(B) Clear firing because of poor company performance or questions of competence (but no illegal actions on the part of the executive).

(C) Clear firing but no reason directly provided or suggested.

(D) Left suddenly with no reason provided.

(E) Merger, reorganization, or corporate restructuring related.

Voluntary Departures

(F) Quit due to disagreements with management/ board of directors (departure initiated by executive). (G) Left with cause/voluntary professional reasons.

(H) Specific personal reasons such as health or family reasons.

(I) Retirement.

Other Classifications (Excluded from Empirical Analysis)

(J) Could not find. This category represents searches on the executive's name and firm in which no results were found.

(K) Death.

REFERENCES

Adams, R. B., and D. Ferreira. "Gender Diversity in the Boardroom." Working Paper, University of Queensland and London School of Economics, 2007.

Adler, R. Women in the Executive Suite Correlate to High Profits. Malibu, CA: Glass Ceiling Research Center, 2001. http://glassceiling.org/InTheNewsFolder/HBRARticle Pagel.html. Accessed November 1, 2006.

Balsam, S., and S. Miharjo. "The Effect of Equity Compensation on Voluntary Executive Turnover." Journal of Accounting and Economics, 43, 2007, 95-119.

Benston, G. J. "The Self-Serving Management Hypothesis-Some Evidence." Journal of Accounting and Economics, 7, 1985, 67-84.

Bertrand, M., and K. F. Hallock. "The Gender Gap in Top Corporate Jobs." Industrial and Labor Relations Review, 55, 2001, 3-21.

Bonn, I., T. Yoshikawa, and P. H. Phan. "Effects of Board Structure on Firm Performance: A Comparison between Japan and Australia." Asian Business & Management, 3, 2004, 105-25.

Carleton, W. T., J. M. Nelson, and M. S. Weisbach. "The Influence of Institutions on Corporate Governance through Private Negotiations: Evidence from TIAA-CREF." Journal of Finance, 53, 1998, 1335-62.

Carter, D. A., B. J. Simkins, and W. G. Simpson. "Corporate Governance, Board Diversity, and Firm Value." Financial Review, 38, 2003, 33-53.

Catalyst. "The Bottom Line: Connecting Corporate Performance and Gender Diversity," January 2004, pp. 1-27. http://www.catalyst.org/files/full/financialperformancereport.pdf. Accessed February 13, 2007.

Coughlan, A. T., and R. M. Schmidt. "Executive Compensation, Management Turnover, and Firm Performance--An Empirical Investigation." Journal of Accounting and Economics, 7, 1985, 43-66.

Denis, D. J., and D. K. Denis. "Performance Changes Following Top Management Dismissals." Journal of Finance, 50, 1995, 1029-57.

Denis, D. J., and T. A. Kruse. "Managerial Discipline and Corporate Restructuring Following Performance Declines." Journal of Financial Economics, 55, 2000, 391-424.

The Economist. "Helping Women Get to the Top," 23 July 2005a, p. 11.

--. "The Conundrum of the Glass Ceiling," 23 July 2005b, pp. 63-65.

Elvira, M. M., and L. E. Cohen. "Location Matters: A Cross-Level Analysis of the Effects of Organizational Sex Composition on Individual Turnover." Academy of Management Journal, 44, 2001, 591-605.

Erhardt, N. L., J. D. Werbel, and C. B. Shrader. "Board of Director Diversity and Firm Financial Performance." Corporate Governance, 11, 2003, 102-11. ma, E. F., and K. R. French. "Industry Cost of Equity." Journal of Financial Economics, 43, 1997, 153-93.

Farrell, K. A., and P. L. Hersch. "Additions to Corporate Boards: The Effect of Gender." Journal of Corporate Finance, 11, 2005, 85-106.

The Guardian. "In Norway, a Woman's Place is in the Boardroom." 9 January 2006. http://www.guardian. co.uk/money/2006/jan/09/business.workandcareers. Accessed April 2, 2008.

Huson, M. R., P. H. Malatesta, and R. Parrino. "Managerial Succession and Firm Performance." Journal of Financial Economics, 74, 2004, 237-75.

Huson, M. R., R. Parrino, and L. T. Starks. "Internal Monitoring Mechanisms and CEO Turnover: A Long-Term Perspective." Journal of Finance, 56, 2001, 2265-97.

Jensen, M. C., and W. H. Meckling. "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure." Journal of Financial Economics, 3, 1976, 305-60.

Jones, D. "Not-so-good Year for Female CEOs," USA Today, 22 December 2005. http://news.yahoo.com/ s/usatoday/20051222/bs_usatoday/notsogoodyear for femaleceos. Accessed January 15, 2006.

Lee, P. M., and E. H. James. "She'-e-os: Gender Effects and Investor Reactions to the Announcements of Top Executive Appointments." Strategic Management Journal, 28, 2007, 227-41.

Lewis, G. B. "Men and Women toward the Top: Backgrounds, Careers, and Potential of Federal Middle Managers." Public Personnel Management, 21, 1992, 473-91. ness, K. S., and M. K. Judiesch. "Are Female Managers Quitters? The Relationships of Gender, Promotions,

Ly and Family Leaves of Absence to Voluntary Turnover." Journal of Applied Psychology, 86, 2001, 1167-78.

McNeil, C., G. Niehaus, and E. Powers. "Management Turnover in Subsidiaries of Conglomerates versus Stand-Alone Firms." Journal of Financial Economics, 72, 2004, 63-96.

Mian, S. "On the Choice and Replacement of Chief Financial Officers." Journal of Financial Economics, 60, 2001, 143-75.

Mikkelson, W. H., and M. M. Partch. "The Decline of Takeovers and Disciplinary Managerial Turnover." Journal of Financial Economics, 44, 1997, 205-28.

Morck, R., A. Shleifer, and R. W. Vishny. "Alternative Mechanisms for Corporate Control." American Economic Review, 79, 1989, 842-52.

Parrino, R. "CEO Turnover and Outside Succession: A Cross-Sectional Analysis." Journal of Financial Economics, 46, 1997, 165-97.

Puffer, S. M., and J. B. Weintrop. "Corporate Performance and CEO Turnover--The Role of Performance Expectations." Administrative Science Quarterly, 36, 1991, 1-19.

The Scotsman. "Norway's Firms Face Gender Law." 28 December 2007. http://news.scotsman.com/world? articleid=3624498. Accessed January 7, 2008.

Shrader, C. B., V. B. Blackburn, and P. lies. "Women in Management and Firm Financial Performance: An Exploratory Study." Journal of Management Issues, 9, 1997, 355-72.

Stroh, L. K., J. M. Brett, and A. H. Reilly. "Family Structure, Glass Ceiling, and Traditional Explanations for the Differential Rate of Turnover of Female and Male Managers." Journal of Vocational Behavior, 49, 1996, 99-118.

Warner, J. B., R. L. Watts, and K. H. Wruck. "Stock Prices and Top Management Changes." Journal of Financial Economics, 20, 1988, 461-92.

Weisbach, M. S. "Outside Directors and CEO Turnover." Journal of Financial Economics, 20, 1988, 431-60.

Wolfers, J. "Diagnosing Discrimination: Stock Returns and CEO Gender." Journal of the European Economic Association, 4, 2006, 531-41.

Yermack, D. "Higher Market Valuation of Companies with a Small Board of Directors." Journal of Financial Economics, 40, 1996, 185-211.

Zahra, S. A., and W. W. Stanton. "The Implications of Board of Directors' Composition for Corporate Strategy and Value." International Journal of Management, 5, 1988, 229-36.

(1.) Adams and Ferreira (2007); Bonn, Yoshikawa, and Phan (2004); Erhardt, Werbel, and Shrader (2003); Farrell and Hersch (2005); Zahra and Stanton (1988).

(2.) Benston (1985); Coughlan and Schmidt (1985); Huson, Malatesta, and Parrino (2004); McNeil, Niehaus, and Powers (2004); Mian (2001); Puffer and Weintrop (1991); Warner, Watts, and Wruck (1988); Weisbach (1988).

(3.) Similar coding schemes are found in Balsam and Miharjo (2007), Huson, Parrino, and Starks (2001), and Parrino (1997). The Appendix includes a detailed description of the coding scheme.

(4.) The EXECUCOMP date is the actual departure date rather than the announcement date, although on occasion, these are the same. In general, we are more interested in the announcement date and rely on news stories for this information.

(5.) Bureau of Labor Statistics: http://data.bls.gov/PDQ/ Servlet/SurveyOutputServlet

(6.) For departing executives, we have also calculated abnormal returns based on the date when the departure is first announced to the public (i.e., the "event announcement date") rather than the fiscal year. The results for voluntary and involuntary departure are robust to this alternative specification of firm abnormal stock returns for departing executives.

(7.) Random effects entail the stringent assumption that the executive-specific unobserved heterogeneity is uncorrelated with the observed covariates. However, our results are robust to a simple logit model, which imposes the opposite extreme assumption of perfect correlation between the unobserved heterogeneity and the controls.

(8.) Alternatively, we also estimate the models using firm random effects and obtain qualitatively identical results.

(9.) CEO refers to "Chief Executive Officer," CFO to "Chief Financial Officer," and COO to "Chief Operating Officer." The excluded category of executives, with reference to these title variables, consists of those holding titles that are vice presidential in nature. This category includes the majority of executives in the EXECUCOMP sample.

(10.) Specifically, we aggregate the 48 Fama and French (1997) industry categories into the following broader groups: food and agriculture; entertainment and leisure; consumer and retail goods; health care services; textiles, construction, and manufacturing; drugs and chemicals; mining and energy; utilities and telecommunications; electricity; and finance, insurance, and real estate. The excluded industry category is food and agriculture.

(11.) We also estimate the models using the executive's proportional share ownership, rather than the dollar value of share ownership, and continued to find negative effects on all types of departure.

(12.) In a nonlinear model, the marginal effect of a variable incorporates both the coefficient and a probability weight. When a control variable is interacted with a dichotomous variable, the probability weights on the marginal effects of the interacted and uninteracted variables are different. This implies that the marginal effect of the interaction term does not measure the size of the difference in the marginal effects for the two groups. However, the coefficient on the interaction term still indicates the direction of the difference in the effect for the two groups.

(13.) The predicted probability for men is [LAMBDA]([x.sub.M][b.sub.M]), where [x.sub.M] is the vector of average values of the control variables for men, [b.sub.M] is the vector of estimated coefficients for men (the coefficients on the uninteracted variables), and A is the cumulative logistic distribution function. The predicted probability for women is A([x.sub.F][b.sub.F]), where [x.sub.F] is the vector of female means and [b.sub.F] is the vector of estimated coefficients for women. For each variable, the coefficient for women is the sum of the coefficient on the uninteracted form of the variable and the coefficient on the interaction of the variable with the female dummy.

(14.) This suggests the gender difference in involuntary departure documented in Table 3 may be largely due to unobserved variables, such as family responsibilities, alternative employment options, or expectations of upward career mobility. Discrimination is likewise a possibility. It is also the case that gender-varying measurement error in the dependent variable could contribute to the difference, for instance if female dismissals are more "newsworthy" than male dismissals, making it easier to classify female departures. Although the focus of our article is on whether women are more likely to depart, the question of why is an interesting area for future research.

JOHN R. BECKER-BLEASE, SUSAN ELKINAWY and MARK STATER *

* We wish to thank Marisa Danley, Rachel Horton, Chris Miller, Mary Spencer, and Esther Trefz, for excellent research assistance. Becker-Blease recognizes Washington State University-Vancouver for providing financial support. Elkinawy recognizes Loyola Marymount University for providing a research fellowship. Seminar participants at the California Corporate Finance Conference, the Southern Economic Association, the American Society of Business and Behavioral Sciences, Loyola Marymount University, and Washington State University-Vancouver offered numerous useful comments and suggestions.

Becker-Blease: Assistant Professor, Department of Finance, Washington State University-Vancouver, 14204 NE Salmon Creek Avenue, Vancouver, WA 98686-9600. Phone 1-360-546-9146, Fax 1-360-546-9037, E-mail jblease@vancouver.wsu.edu

Elkinawy: Assistant Professor, Department of Finance and Computer Information Systems, Loyola Marymount University, One LMU Drive, MS 8385, Los Angeles, CA 90045. Phone 1-310-338-2345, Fax 1-310-338-3000, E-mail selkinawy@lmu.edu

Stater: Assistant Professor, Department of Economics, Trinity College, 300 Summit Street, Hartford, CT 06106. Phone 1-860-297-2486, Fax 1-860-297-2163, E-mail mark.stater@trincoll.edu

TABLE 1
Percentages of Executives in Each Departure Category

Variables All Executives Male Executives

Departed 3.9166 (19.3993) 3.7638 (19.0322)

Involuntarily 1.0204 (10.0500) 0.9326 (9.6122)
departed

Fired for fraud or 0.0582 (2.4107) 0.0452 (2.1246)
misdeeds

Fired for poor 0.2195 (4.6796) 0.2003 (4.4707)
performance

Fired for 0.0094 (0.9684) 0.0098 (0.9908)
unspecified reasons

Left suddenly (no 0.5627 (7.4805) 0.5183 (7.1810)
reason provided)

Left for 0.1707 (4.1281) 0.1590 (3.9848)
merger-related
reasons

Voluntarily departed 2.8962 (16.7702) 2.8312 (16.5865)

Quit due to 0.0675 (2.5978) 0.0609 (2.4664)
disagreements with
board

Left for 1.1105 (10.4793) 1.0386 (10.1384)
professional reasons

Left for personal 0.1107 (3.3249) 0.0923 (3.0364)
reasons

Retired 1.6076 (12.5767) 1.6394 (12.6988)

Number of executive- 53,311 50,932
year observations

Number of unique 17,644 16,667
executives

Variables Female Executives Departed Executives

Departed 7.1879 *** (25.8342) 100.0000 (0.0000)

Involuntarily 2.9004 *** (16.7852) 26.0536 (43.9033)
departed

Fired for fraud or 0.3363 *** (5.7904) 1.4847 (12.0968)
misdeeds

Fired for poor 0.6305 *** (7.9171) 5.6035 (23.0043)
performance

Fired for 0.0000 (0.0000) 0.2395 (4.8888)
unspecified reasons

Left suddenly (no 1.5132 *** (12.2105) 14.3678 (35.0847)
reason provided)

Left for 0.4203 *** (6.4711) 4.3582 (20.4213)
merger-related
reasons

Voluntarily departed 4.2875 *** (20.2618) 73.9464 (43.9033)

Quit due to 0.2102 *** (4.5806) 1.7241 (13.0201)
disagreements with
board

Left for 2.6482 *** (16.0597) 28.3525 (45.0817)
professional reasons

Left for personal 0.5044 *** (7.0858) 2.8257 (16.5745)
reasons

Retired 0.9248 *** (9.5739) 41.0441 (49.2032)

Number of executive- 2,379 2,088
year observations

Number of unique 977 2,061
executives

Variables Male Departed Female Departed

Departed 100.0000 (0.0000) 100.0000 (0.0000)

Involuntarily 24.7783 (43.1838) 40.3509 *** (49.2042)
departed

Fired for fraud or 1.1998 (10.8904) 4.6784 (21.1795)
misdeeds

Fired for poor 5.3208 (22.4507) 8.7719 * (28.3717)
performance

Fired for 0.2608 (5.1018) 0.0000 (0.0000)
unspecified reasons

Left suddenly (no 13.7715 (34.4691) 21.0526 *** (0.4089)
reason provided)

Left for 4.2254 (20.1220) 5.8480 (23.5337)
merger-related
reasons

Voluntarily departed 75.2217 (43.1838) 59.6491 *** (49.2042)

Quit due to 1.6171 (12.6166) 2.9240 (16.8973)
disagreements with
board

Left for 27.5952 (44.7109) 36.8421 ** (48.3793)
professional reasons

Left for personal 2.4518 (15.4690) 7.0175 *** (25.6193)
reasons

Retired 43.5576 (49.5962) 12.8655 *** (33.5801)

Number of executive- 1,917 171
year observations

Number of unique 1,896 165
executives

Notes: Standard deviations are in parentheses. The number in each
cell is the percent of executives who are members of the row
category (or the mean of the row variable), conditioned on
membership in the column category. Thus, the numerator is the number
of executives in the intersection of the row and column categories
(e.g., male and departed) multiplied by 100, and the denominator is
the number in the column category (e.g., male).

Difference between adjacent male and female percentages is
significant at *** 1%, ** 5%, and * 10%.

TABLE 2
Descriptive Statistics for Firm and Executive Variable

Variables Full Sample Departed

Female 0.0446 (0.2065) 0.0819 (0.2743)

Age (yr) (a) 53.1877 (8.0043) 54.9790 (7.9660)

Tenure with firm 7.5829 (11.6895) 8.0761 (13.0068)
(yr) (b)

Director 0.3682 (0.4823) 0.4023 (0.4905)

Chief Executive 0.1940 (0.3954) 0.1346 (0.3414)
Officer (c)

Chief Financial 0.1353 (0.3420) 0.1489 (0.3561)
Officer (c)


Chief Operating 0.0821 (0.2745) 0.1145 (0.3184)
Officer (c)

Total direct 1.5023 (4.4836) 1.5779 (3.6882)
compensation ($
million) (d)

Executive equity 20.1221 (443.0559) 5.9407 (49.1781)
ownership ($
million) (e)

Total assets ($ 6.2449 (26.5105) 8.2608 (33.7033)
billion) (f)

Industry-adjusted 0.0673 (0.1225) 0.0550 (0.1203)
return on assets (%)
(f,g)

1-yr buy-hold -0.0062 (0.4530) -0.0829 (0.4764)
abnormal return (%)
(h)

Total number of 9.7241 (3.0520) 9.7553 (2.9922)
directors

Fraction of male 0.9226 (0.0853) 0.9156 (0.0865)
directors

Fraction of 0.6339 (0.1819) 0.6572 (0.1706)
independent
directors

Number of 53,311 2,088
executive-years

Number of unique 17,644 2,061
executives (i,j)

 Involuntarily
Variables Nondeparted Departed

Female 0.0431 *** (0.2031) 0.1268 (0.3331)

Age (yr) (a) 53.0888 *** (7.9949) 51.1968 (6.6113)

Tenure with firm 7.5628 ** (11.6323) 4.0345 (9.1266)
(yr) (b)

Director 0.3668 *** (0.4819) 0.3750 (0.4846)

Chief Executive 0.1964 *** (0.3973) 0.1673 (0.3736)
Officer (c)

Chief Financial 0.1347 * (0.3414) 0.1599 (0.3669)
Officer (c)

Chief Operating 0.0808 *** (0.2725) 0.1618 (03686)
Officer (c)

Total direct 1.4992 (4.5131) 1.9805 (5.1996)
compensation ($
million) (d)

Executive equity 20.7002 (451.8775) 2.9250 (14.6483)
ownership ($
million) (e)

Total assets ($ 6.1627 *** (26.1726) 10.2818 (46.8144)
billion) (f)

Industry-adjusted 0.0678 *** (0.1226) 0.0453 (0.1166)
return on assets (%)
(f,g)

1-yr buy-hold -0.0031 *** (0.4517) -0.1518 (0.5511)
abnormal return (%)
(h)

Total number of 9.7228 (3.0544) 9.2279 (2.9042)
directors

Fraction of male 0.9229 *** (0.0853) 0.9170 (0.0904)
directors

Fraction of 0.6329 *** (0.1823) 0.6408 (0.1776)
independent
directors

Number of 51,223 544
executive-years

Number of unique 17,176 541
executives (i,j)

Variables Voluntarily Departed

Female 0.0661 *** (0.2485)

Age (yr) (a) 56.2500 *** (7.9819)

Tenure with firm 9.5001 *** (13.8463)
(yr) (b)

Director 0.4119 (0.4923)

Chief Executive 0.1231 *** (03286)
Officer (c)

Chief Financial 0.1451 (0.3523)
Officer (c)

Chief Operating 0.0978 *** (0.2971)
Officer (c)

Total direct 1.4360 *** (2.9678)
compensation ($
million) (d)

Executive equity 7.0033 * (56.4915)
ownership ($
million) (e)

Total assets ($ 7.5488 (27.6260)
billion) (f)

Industry-adjusted 0.0584 ** (0.1214)
return on assets (%)
(f,g)

1-yr buy-hold -0.0586 *** (0.4447)
abnormal return (%)
(h)

Total number of 9.9411 *** (3.0016)
directors

Fraction of male 0.9151 (0.0851)
directors

Fraction of 0.6630 *** (0.1677)
independent
directors

Number of 1,544
executive-years

Number of unique 1,529
executives (i,j)

Notes: Standard deviations are in parentheses. The number in each
cell is the mean of the row variable, conditioned on membership in
the column category.

(a) The number of executive-year observations on age in each of the
columns is 28,210, 1,475, 26,735, 371, and 1,104.

(b) The number of executive-year observations on tenure in each of
the columns is 53,292, 2,088, 51,204, 544, and 1,544.

(c) The excluded group of executive titles is vice president.

(d) Total direct compensation includes base salary, bonuses, and
fringe benefits.

(e) Executive equity ownership is the number of shares owned by the
executive times the firm's stock price at the end of the previous
year.

(f) Total assets and return on assets are lagged 1 yr prior to the
year of observation listed for the executive.

(g) Return on assets is measured as the deviation from the industry
median, where industry groupings are as defined by Fama and French
(1997).

(h) 1-yr buy-and-hold abnormal returns are the deviations of actual
monthly stock returns from the predictions of a market and industry
model, summed over the year prior to that listed for the executive.

(i) The number of unique executives in the departed and nondeparted
columns sum to a number greater than the number of unique
executives in the full sample because nearly every executive who is
observed as a departure is also observed as a nondeparture in some
other year.

(j) The number of unique executives in the involuntarily and
voluntarily departed columns sum to a number greater than the
number of unique executives in the departed column because some
executives are observed as both voluntary departures and
involuntary departures (in different years).

Difference between departed and nondeparted or between
involuntarily and voluntarily departed categories is
significant at *** 1%, ** 5%, and * 10%.

TABLE 3
Random Effects Logit Models of the Reason for Departure (Marginal
Effect Estimates)

 General Departure
Independent Variables Model (1) Model (2)

Female 0.0337 *** (0.000) 0.0678 *** (0.000)

Age X 0.0052 *** (0.000)

Age squared X -0.0000 *** (0.002)

Tenure with firm X -0.0002 *** (0.008)

Director 0.0261 *** (0.000) -0.0029 (0.244)

Chief Executive -0.0212 *** (0.000) -0.0295 *** (0.000)
Officer (CEO) (a)

Chief Financial 0.0035 (0.123) -0.0062 ** (0.019)
Officer (CFO) (a)

Chief Operating 0.0053 * (0.056) 0.0009 (0.778)
Officer (COO) (a)

Log total direct -0.0024 *** (0.007) -0.0034 *** (0.002)
compensation
(millions) (b)

Log executive equity -0.0001 *** (0.003) -0.0001 *** (0.000)
ownership (millions)
(c)

Log total assets 0.0051 *** (0.000) 0.0046 *** (0.000)
(billions) (d)

Industry-adjusted -0.0328 *** (0.000) -0.0341 *** (0.000)
return on assets
(d,e)

1-yr buy-hold -0.0134 *** (0.000) -0.0154 *** (0.000)
abnormal returns (f)

Total number of -0.0010 *** (0.002) -0.0011 ** (0.010)
directors

Fraction of male -0.0108 (0.266) -0.0208 (0.102)
directors

Fraction of 0.0328 *** (0.000) 0.0412 *** (0.000)
independent
directors

Number of executive 53,311 28,193
years

Number of unique 18,404 8,222
executive firms

% correctly 96.1 94.8
predicted

 Involuntary Departure
Independent Variables Model (1) Model (2)

Female 0.0129 *** (0.000) 0.0152 *** (0.000)

Age X 0.0017 *** (0.003)

Age squared X -0.0000 *** (0.004)

Tenure with firm X -0.0002 *** (0.000)

Director 0.0018 ** (0.024) -0.0003 (0.745)

Chief Executive -0.0002 (0.852) -0.0015 (0.110)
Officer (CEO) (a)

Chief Financial 0.0019 ** (0.045) -0.0014 (0.106)
Officer (CFO) (a)

Chief Operating 0.0065 *** (0.000) 0.0027 ** (0.043)
Officer (COO) (a)

Log total direct 0.0002 (0.589) -0.0001 (0.845)
compensation
(millions) (b)

Log executive equity -0.0001 *** (0.0000) -0.0001 *** (0.000)
ownership (millions)
(c)

Log total assets 0.0008 *** (0.003) 0.0011 *** (0.001)
(billions) (d)

Industry-adjusted -0.0075 *** (0.000) -0.0070 *** (0.003)
return on assets
(d,e)

1-yr buy-hold -0.0044 *** (0.000) -0.0049 *** (0.000)
abnormal returns (f)

Total number of -0.0005 *** (0.000) -0.0004 *** (0.008)
directors

Fraction of male -0.0007 (0.832) -0.0019 (0.645)
directors

Fraction of 0.0031 * (0.063) 0.0044 ** (0.034)
independent
directors

Number of executive 53,311 28,193
years

Number of unique 18,404 8,222
executive firms

% correctly 99.0 98.7
predicted

 Voluntary Departure
Independent Variables Model (1) Model (2)

Female 0.0153 *** (0.000) 0.0382 *** (0.000)

Age X 0.0051 *** (0.000)

Age squared X -0.0000 *** (0.002)

Tenure with firm X -0.0000 (0.995)

Director 0.0238 *** (0.000) -0.0024 (0.247)

Chief Executive -0.0192 *** (0.000) -0.0246 *** (0.000)
Officer (CEO) (a)

Chief Financial 0.0010 (0.601) -0.0035 (0.107)
Officer (CFO) (a)

Chief Operating -0.0023 (0.272) -0.0028 (0.250)
Officer (COO) (a)

Log total direct -0.0027 *** (0.000) -0.0032 *** (0.000)
compensation
(millions) (b)

Log executive equity -0.0000 * (0.060) -0.0001 *** (0.007)
ownership (millions)
(c)

Log total assets 0.0041 *** (0.000) 0.0029 *** (0.000)
(billions) (d)

Industry-adjusted -0.0183 *** (0.000) -0.0170 ** (0.015)
return on assets
(d,e)

1-yr buy-hold -0.0067 *** (0.000) -0.0068 *** (0.001)
abnormal returns (f)

Total number of -0.0003 (0.335) -0.0004 (0.203)
directors

Fraction of male -0.0092 (0.276) -0.0154 (0.140)
directors

Fraction of 0.0285 *** (0.000) 0.0330 *** (0.000)
independent
directors

Number of executive 53,311 28,193
years

Number of unique 18,404 8,222
executive firms

% correctly 97.1 96.1
predicted

Notes: Estimates are marginal effects; p values are in parentheses.
Dependent variables for the three departure outcomes are dummy
variables that equal one if the executive departs, departs
involuntarily, and departs voluntarily, respectively. Each model
contains a full set of year and industry dummies; industry
categories include food and agriculture (the excluded group);
entertainment and leisure; consumer and retail goods; health care
services; textiles, construction, and manufacturing; drugs and
chemicals; mining and energy; utilities and telecommunications;
electricity; and finance, insurance, and real estate.

(a) The excluded group of executive titles is vice president.

(b) Total direct compensation includes base salary, bonuses, and
fringe benefits.

(c) Executive equity ownership is the number of shares owned by the
executive times the firm's stock price at the end of the previous
year.

(d) Total assets and return on assets are lagged 1 yr prior to the year
of observation listed for the executive.

(e) Return on assets is measured as the deviation from the industry
median, where industry groupings are as defined by Fama and French
(1997).

(f) 1-yr buy-and-hold abnormal returns are the deviations of actual
monthly stock returns from the predictions of a market and industry
model, summed over the year prior to that listed for the executive.

Marginal effect estimate is significant at *** 1%, ** 5%, and * 10%.

TABLE 4
Random Effects Logit Models of the Reason for Departure
(Coefficient Estimates with Full Set of Female Interactions)

 General Departure

 Model (1)

Independent Coefficient Coefficient
Variables on Variable on Interaction

Age X X

[Age.sup.2] X X

Tenure X X

Director 0.7805 *** -0.8386 **
 (0.000) (0.012)
CEO (a) -0.8176 *** 0.1540
 (0.000) (0.787)
CFO (a) 0.1037 -0.0018
 (0.134) (0.993)
COO (a) 0.1554 ** 0.1771
 (0.048) (0.624)
Log compensation (b) -0.0995 *** 0.3910 ***
 (0.000) (0.000)
Log executive -0.0022 *** -0.0142
equity (c) (0.005) (0.311)

Log total assets 0.1718 *** -0.2166 ***
 (0.000) (0.003)

Adjusted return on -0.9907 *** -0.4927
assets (d,e) (0.000) (0.443)

1-yr buy-and-hold -0.3845 *** -0.3413 *
abnormal returns (f) (0.000) (0.067)

Total directors -0.0386 *** 0.0968 **
 (0.000) (0.010)
Percent male directors -0.6953 ** 2.4018 **
 (0.028) (0.011)
Percent independent 1.0282 *** -0.3708
directors (0.000) (0.462)

Female -1.7265 * X
 (0.097)

Constant -3.3738 *** X
 (0.000)

Predicted prob(Y = 1) M: 0.032 F: 0.059

Executive years 53,311

Executive firms 18,404

% correct predict 96.1

 General Departure

 Model (2)

Independent Coefficient Coefficient
Variables on Variable on Interaction

Age 0.1937 *** 0.0092
 (0.000) (0.965)
[Age.sup.2] -0.0012 *** -0.0008
 (0.000) (0.723)
Tenure -0.0067 *** -0.0128
 (0.008) (0.282)
Director -0.0229 -0.8155 **
 (0.756) (0.017)
CEO (a) -0.9571 *** 0.1608
 (0.000) (0.779)
CFO (a) -0.2022 ** -0.0132
 (0.033) (0.957)
COO (a) 0.0524 -0.0002
 (0.570) (1.000)
Log compensation (b) -0.1096 *** 0.2029 *
 (0.001) (0.1167)
Log executive -0.0031 *** -0.0173
equity (c) (0.001) (0.312)

Log total assets 0.1520 *** -0.2658 ***
 (0.000) (0.001)

Adjusted return on -0.9815 *** 0.0650
assets (d,e) (0.000) (0.930)

1-yr buy-and-hold -0.4278 *** -0.2225
abnormal returns (f) (0.000) (0.277)

Total directors -0.0423 *** 0.1234 ***
 (0.001) (0.003)
Percent male directors -1.1268 *** 2.2103 **
 (0.004) (0.032)
Percent independent 1.1833 *** -0.6368
directors (0.000) (0.247)

Female 0.1959 X
 (0.970)

Constant -9.1728 *** X
 (0.000)

Predicted prob(Y = 1) M: 0.034 F: 0.092

Executive years 28,193

Executive firms 8,222

% correct predict 94.8

 Involuntary departure

 Model (1)

Independent Coefficient Coefficient
Variables on Variable on Interaction

Age X X

[Age.sup.2] X X

Tenure X X

Director 0.2830 ** 0.3267
 (0.028) (0.468)
CEO (a) 0.0171 -0.6681
 (0.917) (0.376)
CFO (a) 0.2825 ** -0.1260
 (0.036) (0.714)
COO (a) 0.8119 *** -0.6482
 (0.000) (0.234)
Log compensation (b) 0.0077 0.2008
 (0.888) (0.183)
Log executive -0.0162 *** -0.0008
equity (c) (0.002) (0.965

Log total assets 0.1191 *** 0.0088
 (0.007) (0.936)

Adjusted return on -1.2184 *** -0.2020
assets (d,e) (0.000) (0.851

1-yr buy-and-hold -0.7337 *** 0.1646
abnormal returns (f) (0.000) (0.569

Total directors -0.0979 *** 0.0843
 (0.000) (0.154)
Percent male directors -0.6040 2.6090 *
 (0.310) (0.076)
Percent independent 0.4307 0.6268
directors (0.126) (0.431)

Female -2.1720 X
 (0.181)

Constant -4.2262 *** X
 (0.000)

Predicted prob(Y = 1) M: 0.006 F: 0.023

Executive years 53,311

Executive firms 18,404

% correct predict 99.0

 Involuntary departure

 Model (2)

Independent Coefficient Coefficient
Variables on Variable on Interaction

Age 0.3049 *** 0.0719
 (0.002) (0.846)
[Age.sup.2] -0.0029 *** -0.0010
 (0.002) (0.802)
Tenure -0.0405 *** 0.0062
 (0.000) (0.758)
Director -0.0288 0.0916
 (0.849) (0.845)
CEO (a) -0.2434 -0.5169
 (0.158) (0.494)
CFO (a) -0.3102 0.1321
 (0.125) (0.738)
COO (a) 0.4370 *** -0.3769
 (0.007) (0.504)
Log compensation (b) -0.0027 -0.0369
 (0.966) (0.833)
Log executive -0.0130 *** -0.0021
equity (c) (0.006) (0.918)

Log total assets 0.1827 *** -0.0395
 (0.001) (0.740)

Adjusted return on -1.2227 *** 0.6027
assets (d,e) (0.001) (0.613)

1-yr buy-and-hold -0.8798 *** 0.3298
abnormal returns (f) (0.000) (0.300)

Total directors -0.0811 *** 0.0806
 (0.002) (0.209)
Percent male directors -1.1071 2.7094 *
 (0.138) (0.094)
Percent independent 0.5975 * 0.4589
directors (0.095) (0.594)

Female -3.3135 X
 (0.713)

Constant -11.052 *** X
 (0.000)

Predicted prob(Y = 1) M: 0.005 F: 0.035

Executive years 28,193

Executive firms 8,222

% correct predict 98.7

 Voluntary Departure

 Model (1)

Independent Coefficient Coefficient
Variables on Variable on Interaction

Age X X

[Age.sup.2] X X

Tenure X X

Director 0.9332 *** -1.6166 ***
 (0.000) (0.001)
CEO (a) -1.0563 *** 0.4848
 (0.000) (0.573)
CFO (a) 0.0425 0.0258
 (0.593) (0.926)
COO (a) -0.1119 0.5520
 (0.241) (0.223)
Log compensation (b) -0.1384 *** 0.4432 ***
 (0.000) (0.000)
Log executive -0.0013 * -0.0128
equity (c) (0.059) (0.526)

Log total assets 0.1919 *** -0.3654 ***
 (0.000) (0.000)

Adjusted return on -0.7231 *** -0.5292
assets (d,e) (0.002) (0.490)

1-yr buy-and-hold -0.2329 *** -0.5382 **
abnormal returns (f) (0.000) (0.022)

Total directors -0.0187 0.1229 ***
 (0.116) (0.009)
Percent male directors -0.6905 * 2.0145 *
 (0.061) (0.090)
Percent independent 1.2199 *** -0.7816
directors (0.000) (0.214)

Female -1.5479 X
 (0.239)

Constant -3.9092 *** X
 (0.000)

Predicted prob(Y = 1) M: 0.024 F: 0.033

Executive years 53,311

Executive firms 18,404

% correct predict 97.1

 Voluntary Departure

 Model (2)

Independent Coefficient Coefficient
Variables on Variable on Interaction

Age 0.2711 *** -0.2132
 (0.000) (0.396)
[Age.sup.2] -0.0017 *** 0.0012
 (0.000) (0.631)
Tenure -0.0007 -0.0061
 (0.783) (0.666)
Director -0.0209 -1.4980 ***
 (0.801) (0.002)
CEO (a) -1.1444 *** 0.4475
 (0.000) (0.602)
CFO (a) -0.1517 -0.0312
 (0.153) (0.918)
COO (a) -0.0925 0.1653
 (0.402) (0.744)
Log compensation (b) -0.1504 *** 0.3144 **
 (0.000) (0.031)
Log executive -0.0023 *** -0.0234
equity (c) (0.007) (0.406)

Log total assets 0.1487 *** -0.4314 ***
 (0.000) (0.0993)

Adjusted return on -0.6384 ** -0.1496
assets (d,e) (0.031) (0.865)

1-yr buy-and-hold -0.2321 *** -0.3816
abnormal returns (f) (0.004) (0.129)

Total directors -0.0287 ** 0.1584 ***
 (0.045) (0.002)
Percent male directors -1.0932 ** 1.7584
 (0.016) (0.164)
Percent independent 1.3671 *** -1.1904 *
directors (0.000) (0.077)

Female 6.5586 X
 (0.286)

Constant -12.452 *** X
 (0.000)

Predicted prob(Y = 1) M: 0.023 F: 0.044

Executive years 26,193

Executive firms 8,222

% correct predict 96.1

Notes: p values are in parentheses. Dependent variables for
departure outcomes are dummies that equal one if the executive
departs, departs involuntarily, and departs voluntarily,
respectively. Each model contains a full set of year and industry
dummies; industry categories include food and agriculture (the
excluded group); entertainment and leisure; consumer and retail
goods; health care services; textiles, construction, and
manufacturing; drugs and chemicals; mining and energy; utilities and
telecommunications; electricity; and finance, insurance, and real
estate.

(a) Excluded group of executive titles is vice president.

(b) Compensation includes base salary, bonuses, and fringe benefits.

(c) Executive equity is the value of firm shares owned by the
executive at the end of the previous year.

(d) Total assets and return on assets are lagged 1 yr.

(e) Adjusted return on assets is the deviation of the firm's return
on assets from the industry median, using Fama and French (1997)
industry groupings.

(f) Buy-and-hold abnormal returns are deviations of monthly stock
returns from the predictions of a market and industry model, summed
over the previous year. Coefficient estimate is significant at
*** 1%, ** 5%, and * 10%.