BackgroundWe investigate the importance of managerial structure on mutual fund metrics. More specifically, using solo- and team- managed funds, we examine the role of gender on return performance and risk levels. The motivation for our study comes from the literature on gender related differences in risk taking behavior, investment strategies, and financial decision making. This paper relates to the summer research paper of finance Ph.D. student Yanfei Sun who examined the effect of management structure and team size on the performance of mutual funds (Hilliard and Sun, 2016). In the following paragraphs we give a brief overview of current research in finance and psychology and develop our hypothesis. At the end of 2015, the assets under management in the mutual fund industry reached almost 16 trillion dollars (2016 Investment Company Fact Book). Large amount of these assets are managed by funds with active management. Managers of these funds make decisions about asset allocation and security selections. The results of these decisions are reflected in fund performance. Before 2000, the majority of actively managed mutual funds were managed by solo-managers. In recent years more and more funds have shifted toward a team-managed structure. In fact, in the first quarter of 2016 more than 77 percent of actively managed funds were team-managed funds accounting for more than 80 percent of assets under management. One may argue that the team-managed funds are more suitable for large funds but funds size does not seem to be the most important determinant of management structure. The average fund size of the solo-managed funds is higher than the average fund size of funds with up to five managers. Solo-managed funds and team managed funds seem to differ in their realized returns as well as in their risk levels. A large body of literature has been devoted to the study of mutual fund performance with different managerial structures. However, empirical evidence on the benefits of team-managed funds is not conclusive. Some researchers argue that team-managed funds perform better because of their ability to better aggregate and evaluate information. This view is consistent with research on small groups indicating that group performance is quantitatively and qualitatively superior to the performance of an individual. Teams benefit from broader knowledge resources as well as from mutual correction of errors introduced by individual members. (Sharpe 1981, Shaw 1932). Studies confirming better performance of team-managed funds include, for example, those of Patel and Sarkissian (2016) and Adams, Nishikawa, and Rao (2015). There is some evidence that team decisions reflect a compromise among the opinions of all members (Sah and Stiglitz 1984) and are less extreme than decisions of individual members (Bar, Kempf, and Ruenzi 2010). Therefore solo- and team-managed funds may differ not only in their absolute performance but also in their level of risk. Some empirical evidence is provided, for example by Bliss, Potter, and Schwarz (2008) and Patel and Sarkissian (2016). There are other well documented theories. For example, the group shift hypothesis (Moscovici and Zavalloni 1969, Hogg, Turner, and Davidson 1990, Kerr 1992) suggests that the opinion of members of the team shifts toward the opinion of the dominant member. This makes the final decision of the team identical to the decision of the dominant person. Consistent with this view, Prather, Middleton, and Cusack (2001) find no difference between the outcomes of team- and solo-managed funds. Yet others argue that team-managed funds may suffer from a free rider problem. The performance of an individual in a team-managed fund cannot be separately observed and therefore a team member may become a free-rider. This may lead to the underperformance of the team-managed funds as empirically documented by Chen et al. (2004), Massa, Reuter, and Zitzewitz (2010), Stein (2002), Goldman, Sun, and Zhou (2016), and (Bar, Kempf, and Ruenzi (2010). The empirical evidence on gender-related differences between the investment strategies of males and females points to higher risk aversion in women (Powell and Ansic 1997). Women tend to exhibit lower confidence about their investment decisions than men (Estes and Hosseini 1988, Barber and Odean 2001). Contrary to this view, Atkinson, Baird and Frye (2003) do not find any significant differences in performance or risk-levels between male and female managers of taxable fixed income mutual funds. They conclude that differences in investment behavior that are attributed to gender may be related to other constraints such as investment knowledge and wealth. Yet another study of actively managed funds finds that female fund managers take less risk than male managers (Niessen-Ruenzi and Ruenzi 2015). Other studies that directly link to our proposed study are studies on corporate board gender diversity. These studies suggest that presence of females on the board may have a positive effect on firm value (e.g., Carter, Simkins and Simpson 2003, Adams and Funk 2012, Berger, Kick and Schaeck 2014). Contrary to the generally accepted view of higher risk aversion in females, the presence of a female on the board does not necessary translate to more risk-averse corporate decisions (Adams and Funk 2012, Berger, Kick and Schaeck 2014). Psychological literature also suggests that males and females differ in their competitiveness and cooperation within the group. For example, men more frequently engage in competitive between-group interactions (Pemberton, Insko and Schopler 1996) but are also more likely to engage in intergroup rivalry than women. On the other hand women are more interpersonally oriented (Baumeister and Sommer 1997) and their cooperation is relatively unaffected by intergroup competition (Vugt et al. 2007).ProposalBased on the previous literature we expect that there are gender-related differences in risk-taking behavior. We also expect that the significance of these differences will not be the same in solo-managed versus team managed mutual funds. Patel and Sarkissian (2016) argue that differences in the empirical findings on managerial structure of mutual funds may be largely attributed to discrepancies in managerial data reported in some databases. We, however, presume that these differences may be also a result of gender-related differences in risk-taking behavior and group cooperation. Therefore we propose to examine the effect of manager’s gender on the risk levels and performance of solo and team-managed mutual funds.DataWe use Morningstar Direct and CRSP-mutual fund databases. Both databases are free of survivorship bias. In addition, the Morningstar Direct database is generally considered to be Gold Star source of data on the personal characteristics of managers. In our sample, we include only actively managed mutual funds whose stated objectives are “Aggressive Growth”, “Growth”, “Growth & Income”, or “Equity Growth”. We do not include passively managed index funds because the decision making process of these funds is not crucial in portfolio selection. We use data from the years 1992 to 2016 covering some 12,300 mutual funds. Morningstar Direct collects data on managers but does not directly state whether the manager is a male or a female. Therefore we will assign the gender to each manager’s using unambiguous given names, suffixes and hyphenated names. Ambiguous gender will be resolved using additional sources on a name-by-name basis. From previous research we expect about 10 percent of managers will be female (Niessen-Ruenzi and Ruenzi 2015).MethodsTo measure the risk of the fund, we follow the Bar, Kempf, and Ruenzi (2010) and use three different measures of risk: total risk, defined as standard deviation of the past 12 months’ returns, systematic risk, defined as the beta from the CAPM model, and idiosyncratic risk, defined as the standard deviation of fund i’s residual fund returns from the same model. We evaluate the effect of managerial structure on fund performance using following model:\(Risk_{i,t}=\beta_0+\beta_1FS_{i,t-6}+\beta_2Gender_{i,t-6}+\beta_3FChar_{i,t-6}+\beta_4MT_{i,t-6}+YD+e_{i,t},\)where Gender is a dummy variable that takes the value of one if the manager is female. For the Fund structure (FS) we construct two sets of measures: (1) solo- versus team-managed, and (2) the number of fund managers. The fund characteristics (FChar) variables include fund age, size, expense ratio, turnover ratio, family size, and objective dummy. To account for non-linearities, we use the natural logarithm of fund size, family size, fund age and manager tenure (MT) variables. In the regression, we also include the year xed (Y D) effect. To avoid a contemporaneous effect, we take 6-month lags for all independent variables. The psychology literature documents gender-related dierences in cooperativeness and competitiveness within the group. Therefore, we also allow for the interaction between the manager's gender and fund structure:\(Risk_{i,t}=\beta_0+\beta_1FS_{i,t-6}+\beta_2Gender_{i,t-6}+\beta_3Gender_{i,t-6}FS_{i,t-6}+\beta_4FChar_{i,t-6}+\beta_5MT_{i,t-6}+YD+e_{i,t}.\)
