جنسیت و امور مالی شرکت ها: آیا مدیران مرد نسبت به مدیران زن بیش از حد مطمئن و مغرور هستند؟

We examine corporate financial and investment decisions made by female executives compared with male executives. Male executives undertake more acquisitions and issue debt more often than female executives. Further, acquisitions made by firms with male executives have announcement returns approximately 2% lower than those made by female executive firms, and debt issues also have lower announcement returns for firms with male executives. Female executives place wider bounds on earnings estimates and are more likely to exercise stock options early. This evidence suggests men exhibit relative overconfidence in significant corporate decision making compared with women.

Traditional corporate finance largely ignores the influence a specific manager has on decision making, focusing on firm characteristics instead of managerial characteristics. Nevertheless, few would argue that Apple would be the same company if someone other than Steve Jobs had been Chief Executive Officer (CEO), or that General Electric would be the same if not led by Jack Welch. We examine executive heterogeneity by focusing on one potentially important characteristic of executives: their gender. Behavioral differences in gender have been studied extensively in psychology and other fields, but not in corporate finance. In this paper, we examine whether the gender of an executive has a material impact on corporate decision making. We then examine whether the differences in behavior we identify have an impact on shareholder value. The number of female top executives in the U.S. has increased significantly. Among major U.S. corporations in 2005, 7.5% of Chief Financial Officers (CFOs) and 1.5% of CEOs were women, versus 3.0% and 0.5% in 1994, respectively. Despite this increase in female representation, previous research has not examined whether gender plays a role in corporate decisions. Examining this is important not only because it provides more insights into corporate behavior generally, but also because the representation of women in higher ranking executive positions continues to be relatively small. We test whether firms with female executives (CEOs and CFOs) make different financing or acquisition decisions compared with firms with male executives1. We then examine whether any differences in decisions identified for women are better or worse for shareholder value. The primary hypothesis we test, based on previous literature in other contexts, is that male executives are overconfident in corporate finance decisions compared with women. Using a difference-in-differences empirical framework on a hand collected data set of executive transitions, we identify several key differences for female executives relative to male executives. We find that firms with female executives grow more slowly and are less likely to make acquisitions. We also find that acquisitions made by female executives have higher announcement returns compared with those made by firms with male executives. We find analogous results for capital structure decisions. Female executives are less likely to issue debt, and announcement returns for debt offerings are higher when the firm has a female executive. Female executives do not, however, make significantly different changes to leverage overall. These main results are consistent with relative overconfidence for male executives compared with female executives, yet other explanations could also be consistent with this evidence2. We conduct a number of additional tests to distinguish the overconfidence interpretation from other explanations consistent with these key results. First, we test whether male executives are overconfident directly by examining their earnings forecasts. We find that earnings forecasts made by firms with male executives have significantly narrower bands than those with female executives. Second, we examine the likelihood that male executives are replaced relative to female executives. Overconfident executives should be more likely to be replaced, because overconfident decisions lead to non shareholder value-enhancing decisions. We find that male executives, especially CFOs, are more likely to be replaced (i.e., serving less than four years in tenure), consistent with men being overconfident. Third, we replicate measures of overconfidence based on stock option exercise decisions used in Malmendier and Tate (2005) for male and female executives. Male executives are less likely to exercise deep-in-the-money options early, which is consistent with male overconfidence. Finally, we examine the likelihood of an acquisition being acquirer shareholder value destroying. If male executives are overconfident, they should sometimes inadvertently undertake value-destroying acquisitions. Consistent with male overconfidence, we find that acquisitions undertaken by men are significantly more likely to have negative announcement returns than those undertaken by firms with female executives. The evidence from these four tests supports the interpretation that male executives are overconfident relative to female executives in major corporate decisions. While our main identification strategy of difference-in-differences around executive transitions largely excludes alternate explanations for our main results, we conduct one additional set of tests to rule out any lingering concerns. These tests rely on an instrumental variable approach, in which the instrument we use for a firm having a female executive is based on a previous study that calibrates a state's level of gender status equality (Sugarman and Straus, 1988). We conjecture that the more friendly a state is to women's equality generally, the more likely a firm located in that state is to have a female executive. We use as an instrument the state's gender status equality value for each firm based on the firm's headquarters location (this continuous measure ranges from 0 to 100). So, for example, a firm headquartered in Mississippi (the lowest scoring state) gets a significantly lower score than a firm headquartered in Oregon (the highest scoring state). We find that this measure is significantly related to the decision to hire a female executive in the first stage. Second-stage results from a two-stage least squares (2SLS) instrumental variable (IV) design confirm the results from the primary difference-in-differences approach. To our knowledge, ours is one of the first papers to study gender differences in the corporate setting. Gender has been examined in other business settings, including stock trading behavior (Barber and Odean, 2001), the mutual fund industry (Atkinson, Baird, and Frye, 2003), start-up firms (Verheul and Thurik, 2001), and competitive environments in laboratory settings (Niederle and Versterlund, 2007). Graham, Harvey, and Puri (forthcoming) use a survey-based approach to identify differences in CEO risk aversion and optimism and then relate those differences to corporate financial decisions. Sapienza, Zingales, and Maestripieri (2009) examine the impact of testosterone levels in men and women and the impact of that testosterone on career choices. Bharath, Narayanan, and Seyhun (2009) examine insider trading by female versus male executives, and Adams and Ferreira (2009) examine the impact of female board members on firm governance and stock performance. Our paper differs from previous literature as we examine different aspects of executive behavior including mergers and financing decisions, we use a difference-in-differences empirical approach, we hand collect data to obtain a larger sample of female executives, and we examine not only executive decision making but also the market's reaction to that decision making. The rest of the paper is organized as follows. In Section 2, we state our hypotheses; in Section 3, we provide our main empirical tests; in Section 4, we conduct additional tests to evaluate why gender differences exist; and in Section 5, we conclude.

Women make different corporate financial and investment decisions than men. Firms with female executives are less likely to make acquisitions and are less likely to issue debt than firms with male executives. Investors also react more favorably to significant corporate financial decisions made by firms with female executives. Announcement returns are higher around acquisitions and debt offerings when the firm has a female executive compared with when the firm has a male executive. This empirical evidence is consistent with men being overconfident relative to women. We confirm that men are overconfident by showing that men provide narrower earnings forecasts and are less likely to exercise options early. Further, we show that firms with male executives are more likely to execute value destroying acquisitions, and male executives are more likely to be removed from their position as executive. While other explanations, such as female risk aversion, are also consistent with some of the results of the paper, we conclude that overconfidence in male executives is most consistent with the breadth of the evidence. Regardless of the interpretation, the empirical evidence reveals significant differences between firms with male executives versus those with female executives, implying that models of capital structure and acquisitions that focus on firm characteristics alone miss this important factor for explaining differences in firm behavior. Because our evidence suggests that firms with female executives make decisions that are better for shareholders, the small representation of women in the executive ranks is perhaps surprising. One explanation is that since we examine only a subset of corporate decisions made by executives, it is possible that male executives are better along other dimensions, such as strategy or compliance. Another explanation is that perhaps the availability of qualified female executives is scarce, due to lower numbers of female MBA graduates or different weights placed on family time for men versus women. The increase in representation over time can then be explained by a greater supply of candidates. This argument, however, implies higher salaries for female executives due to a shortage of these highly qualified executives, but the evidence on female compensation does not seem to support this [Bertrand and Hallock (2001) find that female executives are not paid more than male executives, and we also examine compensation within our specific sample and find no evidence of differences in levels or composition of compensation for female executives]. Yet another explanation is that discrimination plays a role in the hiring of executives. Becker (1971) argues that employers with a taste for discrimination hire desired employees despite higher costs associated with hiring these employees. Employers who discriminate against women, therefore, employ fewer than the profit-maximizing number of women. This implies that women who are hired will on average be of higher quality, because they were able to overcome discriminatory preferences, so decisions made by female executives will be better for shareholder value creation. Our study could be viewed as providing empirical evidence in support of this theory (indicating that gender discrimination leads in some cases to the hiring of overconfident men). Future research could disentangle these potential explanations.