Are female executives more risk-averse than male executives?(Author abstract)(Report)

Introduction

Are female executives more risk averse compared to their male counterparts? This paper addresses this issue by comparing stock trading behavior between male and female executives in response to stock option awards. A number of prior studies show that women are more risk averse than men. Hersch [1996] finds that women make safer choices than men when it comes to making risky consumer decisions such as smoking behavior, seat-belt use, preventive dental care, and having regular blood pressure checked. Pacula [1997] observes that women exhibit greater risk aversion to drug use than men.

In the area of financial decisions, a number of prior studies document that females are less likely to take risk than males. Most of these studies examine the allocations of pension assets by gender and find that females invest less in risky investments such as common stock [Bernasek and Shwiff, 2001; Bajtelsmit et al., 1999; Sunden and Surette, 1998; Bajtelsmit and VanDerhei, 1997; Hinz, McCarthy, and Turner, 1997]. In Bernasek and Shwiff [2001], for example, gender is the most significant factor explaining the allocation of the defined contribution pension to stocks. They find men allocate more of their pension fund to stocks than women.

In addition to pension asset allocations, propensity of risk taking by gender is measured by examining household wealth allocations and psychological traits. Jianakoplos and Bernasek [1998] in their survey paper examine the role of gender between household wealth and other socioeconomic factors and the proportion of risky assets held. They find that the magnitude of decrease in risk aversion due to an increase in household wealth is smaller for single women than for single men. Also, 63 percent of single women were not willing to take any financial risk with their investments versus 43 percent for single men. Overall, single women are relatively more risk averse than single men. Using household data, Barber and Odean [1999] show that women invest in less risky positions than men where risk is defined in terms of volatility of portfolio and individual security returns, beta of the security, and size of the investment. Sexton and Bowman-Upton [1990] examine the psychological traits of growth-oriented male and female entrepreneurs and find that females score significantly less on traits related to risk-taking. In computerized laboratory experiments, Powell and Ansic [1997] observe that women are less risk seeking than men. Overall, these studies support the view that women prefer less risk than men in their financial decision-making. (1)

A recent study by Ofek and Yermack [2000] shows that those managers who already own a large investment in the firm's stock will start selling their shares when new stock options are awarded to them. Such stock selling provides diversification benefits to the executive and is an indication that he is risk averse. According to Ofek and Yermack [2000], the magnitude of executive's stock selling is closer to the optimal hedge ratio of 60 percent. That is, the executive sells 600 shares that he already owns for every 1,000 shares of new stock options that he receives.

Following prior empirical studies that women are generally more risk averse than men, the authors hypothesize that female executives will sell more shares of the firm's stock than male executives in response to stock option awards. The authors also predict that it will be the female executives who sell shares in proportion to the hedge ratio of 60 percent. Contrary to the prediction, it is found that the female executives sell a smaller number of shares than the male executives once new stock options are awarded. This is true even after controlling for the levels of executive stock ownership. Male executives with high level of ownership seem to sell shares based on hedge ratio of 60 percent. The findings are consistent with those in Bliss and Potter [2002] that women take more risk than men. Their empirical results show that women fund managers hold portfolios with marginally higher risk than men.

One of the important features of this study is that it directly examines the attitude of the male and female executives towards portfolio diversification. While most of the prior studies examined gender differences in allocations of pension and household wealth, none of the studies investigated the gender effect on adjustment in an individual's personal wealth due to addition of risky security (e.g., stock options). Upon new stock option award, if female executives sell shares to hedge their portfolios and male executives do not, or they do to a lesser degree, then it is an indication that female executives are less willing to take additional risk while male executives are risk takers.

The remainder of the paper is organized as follows. Data and sample are described in the next section followed by empirical results and concluding remarks.

Data and Sample

All data for the empirical analysis are obtained from the 2000 version of Standard and Poors ExecuComp database. There are 20,622 executives listed on ExecuComp database (on 'person' file), of which 905 are female executives and 19,717 are male executives. There are 75,751 person-year observations from 1992 to 2000 with at least one new stock option award, option exercise, or restricted stock award transaction. Of the 75,751 person-year observations, 72,895 transactions are by male executives and 2,856 are by female executives. All share quantities are adjusted for stock splits and are stated in common year 2000 units in this study. Since the empirical analysis is based on annual change in shareholdings, the final sample does not include data from 1992. After deleting observations with data that are preloads and not meaningful, the final sample includes 69,767 person-year observations of which 67,004 transactions are by male executives and 2,763 are by female executives.

Panel A of Table 1 provides descriptive statistics for male and female executives. On average, male (female) executives sold 7,180 (7,130) shares upon receipt of 136,150 (111,520) new stock options each year during 1993-2000. This indicates that there is no difference in annual changes in shares between the male and female executives. Both the mean and median values indicate that these two groups exhibit similar trading patterns. With respect to current stock ownership in the firm, male executives seem to have higher ownership (0.80 percent) than the females (0.37 percent). Male executives also received more stock options (136,150) and exercised them more (126,903) than female executives (111,520 and 117,480). Cash compensations were also higher for male executives. On average, male (female) executives earned $596,910 ($407,560) annually as salary and bonus. Overall, the total value of cash and equity-based compensation is higher for male executives than female executives.

To address the problem of unequal variance in data analysis resulting from different sample sizes between males and females, a randomly selected sample of 2,763 observations for the male executives was created from the full sample of 67,004 observations. This sub-sample will be used along with the full sample in the analysis. The descriptive statistics for this smaller sample are shown in Panel B of Table 2. The data exhibit similar characteristics as the full sample. There is no statistical difference in annual changes in shares between the male and female executives. Also, the male executives own a higher percentage of the company's stock than the female executives as with the full sample. They received more equity-based compensation and exercised more stock options. The amount of cash compensation is higher for the male executives than for the female executives as well.

Empirical Results

The empirical analysis is based on changes in executive's existing shares in response to new stock option awards. A decrease in shareholdings would indicate that the executive has reduced the risk of his personal portfolio due to additional risk associated with the new stock option awards. A decrease in shareholdings is, therefore, consistent with the notion that the executive is reluctant to take additional risk resulting from the stock option awards. A reduction in already owned shares by the female executives and not by the male executives would indicate that the female executives are risk-averse while the male executives are not.

Table 2 presents the result of the findings for the full sample of person-year observations. There are four panels in Table 2. Panel A presents the results of the impact of stock option awards on the changes in shareholdings of executives without considering executive's stock ownership and gender. The authors test the null hypotheses that stock option awards have no impact on executive shareholdings. The hypothesis is tested by regressing annual changes in shares owned by the executives on their stock option awards. The regression equation is as follows:

[DELTA][S.sub.t] = [[beta].sub.0] + [[beta].sub.1] x [C.sub.t] + [[beta].sub.2] x [R.sub.t] + [[beta].sub.3] x [I.sub.t] (1)

where:

[DELTA][S.sub.t] = change in shares owned by an executive in year t,

[C.sub.t] = new stock options awarded in year t,

[R.sub.t] = current year's (year t) dividend-adjusted stock returns, and [I.sub.t] = 1 if financial services industry, 0 otherwise

[beta]s are the parameter estimates.

The null hypothesis that [[beta].sub.1] = 0 is tested. A positive value of coefficient [[beta].sub.1] indicates an increase in executive shareholdings due to an increase in new stock options and a negative value indicates a decrease in shareholdings due to an increase in new stock options. Since new stock options do not increase shareholdings, the null hypothesis is that there is no change in shareholdings after receiving new stock options. The alternative hypothesis is based on the premise that stock option awards increase the risk of the executive who will hedge the risk through stock selling. To achieve an optimal hedge, the executive will sell shares equal to the number of new stock options times the change in option value per unit change in stock price. Ofek and Yermack [2000] find such hedge ratio to be 0.60, i.e., the executive sells 600 shares for every 1,000 new stock options.

Print Get the Mag Weekly Updates