Can managers' characteristics explain European bond mutual fund performance?

Finance Research Letters 58 (2023) 104626 Available online 21 October 2023 1544-6123/© 2023 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Can managers’ characteristics explain European bond mutual fund performance? Pablo Dur´ an-Santomil a , * , Luis Otero-Gonz´ alez a , Renato Domingues b , Paulo Leite c a Department of Finance and accounting, Economics Faculty, Universidade de Santiago de Compostela, Avda. Burgo de las Naciones, 15782 Santiago de Compostela, Spain b Department of Business Science, School of Management, Polytechnic Institute of Tomar, Quinta do Contador, Estrada da Serra, 2300-313, Tomar, Portugal c Applied Management Research Unit (UNIAG), School of Management, Polytechnic Institute of C´ avado and Ave, 4750-180 Barcelos, Portugal ARTICLE INFO JEL classification: G11 G12 J16 Keywords: Bond mutual funds Performance Manager characteristics Gender diversity ABSTRACT This study aims to discover whether the characteristics of European bond fund managers affect mutual fund performance and risk. Based on 7,930 fund-year observations, gathered from the Morningstar database for the period 2005–2019, we have compared several performance and risk measures among team-managed and single-managed funds. We have also analysed the impact of gender diversity on fund performance and risk. Our results indicate that funds managed by teams do not exhibit statistically significant differences in performance or risk when compared to singlemanaged ones. Management teams with greater gender diversity do not significantly impact fund performance or risk in relation to all-male or all-female teams. Overall, our findings indicate that the underrepresentation of women in the fund industry, or gender bias, does not make economic sense. 1. Introduction Compared to funds managed by individuals, the number of funds and investor flows managed by teams has increased (e.g., Bliss et al., 2008). Team-managed funds have theoretical benefits like improved decision-making and greater productivity (see a debate in Adams et al., 2018). Also, “gender bias” affects professionals’ investment choices (Sehrish et al., 2023) and investors. Niessen-Ruenzi and Ruenzi (2019) have observed significantly smaller inflows to equity funds with female managers, and Atkinson et al. (2003) indicated the same for fixed-income funds managed by women. The mutual fund industry has not made any progress on gender representation over the last two decades according to Morningstar data. 1 Compared to male fund managers, their female counterparts have shorter tenures and are less likely to be promoted (Barber et al., 2017). These empirical observations should be supported by evidence that funds managed by teams or male managers achieve better performance than those managed by individuals or women. The literature does not show consistent results of better performance in favor of male managers or teams, so the fact that they have higher popularity is puzzling. Furthermore, since practically all studies focus on equity funds, our paper tries to cover a gap in the literature for the impact of bond fund managers’ characteristics on performance and risk. Regarding the effects of gender, though most studies have found no significant difference in performance between maleor female- * Corresponding author. E-mail address: [email protected] (P. Dur´ an-Santomil). 1 See https://www.morningstar.com/articles/1029482/the-percentage-of-us-female-fund-managers-is-exactly-where-it-was-in-2000 Contents lists available at ScienceDirect Finance Research Letters journal homepage: www.elsevier.com/locate/frl https://doi.org/10.1016/j.frl.2023.104626 Received 4 July 2023; Received in revised form 18 September 2023; Accepted 20 October 2023 Finance Research Letters 58 (2023) 104626 2 managed funds, they have shown evidence of higher inflows in the former. Using responses from fund managers, Beckmann and Menkhoff (2008) found that females exhibit higher risk aversion. According to B¨ ar et al. (2011), they demonstrate less extreme performance because they adopt less extreme investment styles. Bliss and Potter (2002), Babalos et al. (2015), Morningstar (2018), Niessen-Ruenzi and Ruenzi (2019) and Rau and Wang (2022) have found no evidence of differential performance and risk based on gender. Barber et al. (2017) found no differential performance between male and female fund managers, but female managers’ returns were considered to exhibit lower volatility. Instead of using male or female single managers, several papers have focused on gender diversity or mixed teams. Mixed teams outperform male-only and female-only teams according to Citywire’s Alpha Female report 2022 2 and Vanguard research (Lawrence, 2022). B¨ ar et al. (2011), Qiong et al. (2014) and Grossmass et al. (2019) have discovered that the higher the percentage of females a management team has, or the more gender diversity there is, the more negative the impact is on performance. However, as far as we are aware, in the case of fixed-income funds, only Atkinson et al. (2003) have explored the link between gender and mutual fund performance and risk. They noticed no substantial difference in performance or risk between US funds managed by men or women, but fund flows were shown to be higher for the former than for the latter. Regarding management team characteristics, several papers have focused on team size, without achieving consistent results. Teammanaged funds underperformed individually-managed funds in the studies of Chen et al. (2004) and B¨ ar et al. (2011), outperformed and behaved more conservatively in Han et al. (2017) and displayed no difference in performance in Prather and Middleton (2002, 2006), Bliss et al. (2008) or Dass et al. (2013). However, teams exercise less extreme investment styles, so they have a lower likelihood of demonstrating extreme performance outcomes (B¨ ar et al. 2011). Qiong et al. (2014) saw that having more members in a team has a negative effect on performance. Chen et al. (2020) have shown that equity funds managed by teams with higher diversity in terms of gender or age have significantly lower performance. Our paper aims to answer some unexplored questions in the European fixed-income fund literature: Do management teams obtain better performance than funds managed by individuals? Does increased gender diversity on the management team influence performance or risk? To our knowledge, this paper is the first to explore these research topics for European fixed-income funds. We have organised this paper as follows: the first section establishes the hypotheses; the next one contains a brief description of the data and the variables, and the key statistics for these are shown; next, we explain the empirical analysis and, in the final section, we propose our conclusions. 2. Hypotheses Understanding the relationship between management structure, gender diversity and fund performance in the area of fixed-income mutual funds is essential for investors and researchers. However, studies have traditionally focused on equity funds, some finding that team-managed funds perform better than their single-managed counterparts (Han et al., 2017). This better performance is often attributed to the diverse expertise and collaborative efforts among team members. In contrast, other studies have suggested that there is no significant difference in performance (Chen et al., 2004; Bliss et al., 2008). Some have even indicated that team-managed funds may underperform single-managed ones (B¨ ar et al., 2011; Karagiannidis and Booth, 2022) due to coordination costs or expenses. Given these varying results, we have proposed the following hypothesis: H1.: Team-managed and single-managed bond mutual funds exhibit similar levels of performance and risk. Female managers often tend to adopt investment styles that lead to lower volatility in the funds under their management (Barber et al., 2017). However, in terms of performance, no significant differences have been detected when comparing female managers to their male counterparts (Atkinson et al., 2003; Niessen-Ruenzi and Ruenzi, 2019; Rau and Wang, 2022). The topic of gender diversity in mutual fund management has garnered increasing attention in recent years. The literature presents a heterogeneous landscape, with certain studies suggesting that mixed-gender management teams outperform both male-only and female-only teams (Lawrence, 2022). Conversely, others indicate that gender diversity may have an adverse impact on fund performance (B¨ ar et al., 2011; Qiong et al., 2014; Grossmass et al., 2019). In light of these variations, we have formulated the following hypothesis: H2.: Gender diversity in bond mutual fund management teams does not have a significant impact on fund performance or risk. 3. Data, variables and statistics 3.1. Data Our survivor-bias-free data covers the major subcategories of European bond mutual funds (namely, the Euro Government, Euro Corporate and Euro Diversified bonds), for the period 2005–2019, obtained from Morningstar Direct. We have an unbalanced panel with 7,930 fund-year observations and a number of funds that vary from 310 in 2015 to 659 in 2019. 2 See https://uk.citywire.com/alpha-female P. Dur´ an-Santomil et al. Finance Research Letters 58 (2023) 104626 3 3.2. Variables 3.2.1. Dependent variables a) Mutual fund performance We have used a variety of performance metrics, including yearly net returns, Sharpe ratios, and 1-factor alphas, all of which are from the Morningstar Database. Yet, as bond funds can exhibit diverse investment strategies, we have also assessed performance using 4-factor and 6-factor alphas. The 6-factor model, which includes Bond (BD), Default (DF), Option (OP), Equity (EQ), Term Spread (TS) and Government Spread (GS) factors, is based on Otero-Gonz´ alez et al. (2022) and is as follows: ri,t= α i+β1iBDt+β2iDFt+β3iOPt+β4iEQt+β5iTSt+β6iGSt+ ε i,t(1) where r i,t is each fund’s excess return, BD t is the Bloomberg Euro Aggregate / Treasury / Corporate index excess return, DF t is the return spread among the Intercontinental Exchange (ICE) Euro High Yield index and the Bloomberg Euro Treasury index, OP t is the return spread among the ICE Euro Asset-Backed and Mortgage-Backed Securities index and the Bloomberg Euro Treasury index, EQ t is the MSCI European and Monetary Union (EMU) index excess return, TS t is the difference in return between the Bloomberg Euro Treasury 10+and the 1–3 years indices, GS t is the difference in return between the ICE Greece, Ireland, Italy, Portugal and Spain (GIIPS) and the ICE Euro Government excluding GIIPS indices and ε i,t are the regression residuals. The risk-free rate corresponds to the 1-month Euribor. 3 The 4-factor variant is estimated without the TS and GS factors, in line with Elton et al. (1995) and Leite and Cortez (2017), among others. b) Mutual fund risk A fund’s maximum loss during a given time period was measured by its value at risk (VaR), which was computed with a 95% confidence level. c) Mutual fund performance extremity To measure the performance extremity of a manager we employed the extremity measure of B¨ ar et al. (2011), PE i,t , computed as the absolute difference between a fund’s performance, P i,t , and the average performance of all funds in the same year and segment (Morningstar category), Pi,t. Following these authors, we normalized the absolute difference by dividing it by the average absolute difference of all n funds in the relevant category and year 4 : PEi,t= Pi,t−Pi,t  i/n⋅∑n j=1 Pj,t−Pi,t  (2) 3.2.2. Independent variables The Morningstar Database contained the manager tenure and manager history for each fund manager. Information on the managers’ gender was hand-collected for all managers who were members of a management team. Next, based on this information, the following additional variables were constructed for each fund: the number of men and the number of women in the management team, the dummy Team i,t (a binary variable with the value of 1 in the case that it was a management team and 0 for a single manager) and Femaleratio i,t (the proportion of female managers). Finally, we measured gender diversity using the following approach (see, for example, Qiong and Yang, 2014 and Lawrence, 2022): GenderDiversityi,t=∑ j −pj⋅ln(pj)(3) where: j =1, 2 because there were 2 categories (female and male). The proportion of team members belonging to one category (for each fund i and year t), p j , was then computed. 3.2.3. Controls We used the following control variables: fund age (years active), fund size (the log of each fund’s total net assets – TNA), expense ratio (as a percentage of TNA), turnover ratio (as a percentage of portfolio holdings that are replaced in a given year) and manager tenure (years of experience). Previous performance was controlled by the Morningstar Stars rating (Otero et al., 2019; Otero-Gonz´ alez et al., 2022) and by the R-Squared (the percentage of a mutual fund movement that can be explained by movement in a benchmark index), an indicator of active management that has been used in several mutual fund papers (e.g., Amihud and Goyenko, 2013). 3 Firstly, regression (1) was estimated for each fund, in each calendar year, using weekly data. Then, we converted the weekly alphas ( α w ) into yearly alphas ( α y ) with expression α y =(1 + α w ) tw −1, with tw equalling the annual number of trading weeks. 4 A fund with an average performance extremity has a PE i,t =1. P. Dur´ an-Santomil et al. Finance Research Letters 58 (2023) 104626 4 3.3. Descriptive statistics Euro diversified bond funds made up most of our dataset (54.2%), while Euro corporate and Euro government bond funds accounted for the remaining 25.2% and 20.6%, respectively. The average annual return, as shown in Table 1, was approximately 2.76%; however, average fund alphas were negative, reaching yearly values of -0.07%, -0.40%, and -0.47% with the 1-, 4and 6-factor models, respectively. Additionally, there were differences between funds in terms of active management (R 2 ), the ratio of net expenses was relatively low (0.77%) and the average rating was 3.16 stars. The average fund age was 20.3 years, and managers had an average experience of 10.2 years. Regarding fund management, the average team size was 1.6 managers (ranging from 1 to 11 managers). Management teams were less common (39.6%) than individual managers (60.4%). The proportion of female managers was 17.9% across the whole sample, but this percentage had increased since 2005 (15.4%) to 2019 (19.5%). Therefore, women were underrepresented in the European fixed-income industry, just as they were in the global mutual fund industry. Atkinson et al. (2003) reported only a 5.6% share of women in their sample, but, using more recent data, Morningstar found that 14% of all portfolio managers were women. 5 Single male-managed funds represented 50% of the total, followed by all-male teams (25.2%) and mixed teams (13.2%). All-female teams were rare (1.2%), similar to the 0.7% reported by Niessen-Ruenzi and Ruenzi (2019) and single female-managed funds represented 10.4%. 4. Empirical analysis 4.1. Is it wise to invest in team-managed funds or in higher-gender-diversity funds? We related mutual fund performance and risk metrics (net returns, 1F-alpha, 4F-alpha, 6F-alpha, Sharpe ratio, volatility, VaR) to the fund’s management structure (team vs. single management) and fund age, manager tenure, fund size, expense ratios, R 2 s and Morningstar Stars ratios in a regression framework, 6 as shown in the following model: Pit = ω +β1Teami,t−1+β2Tenurei,t−1+β3Agei,t−1+β4logSizei,t−1+β5Expensesi,t−1+β6R2 i,t−1+β7Starsi,t−1+ ε i,t(4) where P it corresponded to a performance (or risk) metric of fund i in year t and ω was a constant. To address concerns about possible reverse causality and endogeneity, explanatory variables were lagged by a year. Team i,t −1 was a dummy with a value of 1 if the fund was managed by a team, or 0 otherwise (single manager). Tenure i,t −1 was the mean management tenure of each fund. Stars i,t −1 was a dummy with a value of 1 if the fund’s rating was 4 or 5 stars (best-rated funds). We also used other control variables like expenses, age, Table 1 Descriptive statistics. Variable Obs. Mean Std. Dev. Min. Max. Performance and risk metrics 1F-alpha 7732 -0.068 2.341 -21.474 15.641 4F-alpha 7735 -0.402 1.827 -13.877 10.667 6F-alpha 7440 -0.474 1.878 -18.427 42.539 NetReturn 7727 2.763 4.258 -30.279 32.779 Sharpe 7392 1.070 2.006 -22.956 8.346 VaR 7735 -0.402 1.827 -13.877 10.667 Control variables Age 7970 20.339 9.124 0.282 53.970 Stars 6699 3.160 1.037 1.000 5.000 Tenure 7744 10.175 6.439 0.250 38.830 Expenses 6819 0.767 0.395 0.004 3.997 R 2 7731 58.094 29.629 0.000 99.971 Turnover 1417 88.629 129.031 -252.630 1230.890 Size 7930 1.503 0.742 1.000 11.000 Management team variables Team 7970 0.396 0.489 0.000 1.000 FemaleRatio 7970 0.179 0.340 0.000 1.000 This table shows key statistics for each variable in the dataset. 1F-, 4Fand 6F-alphas are the 1-, 4and 6-factor alphas (see Eq. 1). Yearly net returns (NetReturn), Sharpe ratios (Sharpe) and Value at risk (VaR) have also been reported. In terms of control variables, we have included fund age (years in activity), Stars (3-year Morningstar Stars Rating), tenure (manager’s experience, in years), expense ratio (as a percentage of TNA), turnover and fund size (log of TNA). R 2 is the coefficient of determination of the 1-factor alpha regression, i.e., the percentage of a mutual fund movement that can be explained by the Morningstar category benchmark. The management team variables are Team (a dummy that takes a value of 1 for funds managed by teams) and FemaleRatio (the proportion of females on a fund’s management team). 5 See https://www.morningstar.com/articles/1029482/the-percentage-of-us-female-fund-managers-is-exactly-where-it-was-in-2000 6 Since the turnover ratio reduced the sample size, it was excluded from the analysis and was only incorporated in a robustness test. P. Dur´ an-Santomil et al. Finance Research Letters 58 (2023) 104626 5 size and R 2 s. 7 Table 2.1 shows the coefficient estimates of Eq. (4). The team variable was not statistically significant, so there was no overall persistent difference in investing with teamover single-managers in terms of mutual fund performance or risk. This result was in line with Prather and Middleton (2002, 2006), Bliss et al. (2008) and Dass et al. (2013) for equity funds. In relation to the control variables, the expense ratio was significant in several estimations: the higher the expenses, the lower the 1-, 4and 6-factor alphas and the Sharpe ratio. Funds with higher expenses were also riskier. Funds with a higher difference from their benchmark (lower R 2 ) obtained better performance but assumed more risk. Funds managed by more experienced managers performed slightly better in terms of alpha. The best-rated funds obtained higher 1-factor alphas and Sharpe ratios. Fund age did not have significant effects on performance or risk. To assess the influence of gender diversity on performance we firstly related the mutual fund performance and risk metrics to the fund’s entropy definition of gender diversity (Eq. 5), in line with the following regression: Pit = ω +β1GenderDiversityi,t−1+βjControlsi,t−1+ ε i,t(5) using fund age, tenure, size, expense ratios, R 2 and Morningstar Stars ratios as control variables. Table 2.2 reported the results of this estimation, where we could see that the gender diversity variable was not statistically significant. Afterwards, we repeated the previous regression analysis using a different gender diversity measure, namely the female ratio (the proportion of females on a management team), and included a (non-linear) squared term for the gender composition variable, as shown in the following expression: Pit = ω +β1FemaleRatioi,t−1+β2FemaleRatio2 i,t−1+βjControlsi,t−1+ ε i,t.(6) The results can be seen in Table 2.3. We found that the percentage of females in the team, as well as its squared-term were Table 2 Performance and risk models. Variable 1F-alpha 4F-alpha 6F-alpha NetReturn Sharpe VaR Table 2.1 – Team vs. single manager (Eq. 4) Team t −1 0.045 -0.011 -0.0031 0.091 0.0088 0.0048 Tenure t −1 0.0086* 0.0071** 0.0075* -0.0026 0.0005 -0.0015 Age t −1 -0.001 -0.0043 -0.0045 0.0044 0.0036 -0.0008 Size t −1 -0.0221 -0.0152 -0.023 0.0358 0.0105 0.0109** Expenses t −1 -0.5629* -0.9127*** -0.7230*** -0.4546 -0.3621** 0.0786*** Stars t −1 0.1899 0.2155 0.2662* 0.2609 0.1284* 0.0446* R2 i,t−1 -2.0249** -1.7312** -1.4796** 1.5727** -0.1682 0.2416*** Cons 2.0348*** 1.6935*** 1.4642*** 0.967 2.2127*** -0.2696** N 5211 5211 5211 5209 5181 5211 R 2 0.2036 0.1879 0.1547 0.5581 0.5829 0.5312 Table 2.2 – Gender diversity (Eq. 5) GenderDiversity t −1 -0.1013 0.0483 0.0702 0.0091 -0.0349 0.0145 Tenure t −1 0.0098* 0.0066* 0.0070* -0.0016 0.0008 -0.0016 Age t −1 -0.0011 -0.0043 -0.0046 0.004 0.0036 -0.0008 Size t −1 -0.0213 -0.0154 -0.0229 0.0381 0.0106 0.0110** Expenses t −1 -0.5588* -0.9137*** -0.7232*** -0.4459 -0.3612** 0.0791*** Stars t −1 0.1905 0.2154 0.2663* 0.2624 0.1285* 0.0447* R2 i,t−1 -2.0027** -1.7385** -1.4861** 1.6006** -0.1629 0.2419*** Cons 2.0227*** 1.6958*** 1.4633*** 0.9367 2.2109*** -0.2716** N 5211 5211 5211 5209 5181 5211 R 2 0.2036 0.188 0.1548 0.558 0.5829 0.5312 Table 2.3 – Female ratio (Eq. 6) FemaleRatio t −1 -0.1625 0.2215 0.2462 0.1202 -0.1551 0.054 FemaleRatio2 t−1 -0.1457 -0.4795 -0.475 -0.4087 0.1411 -0.0725 Tenure t −1 0.0110* 0.0067* 0.0071 -0.0001 0.0012 -0.0016 Age t −1 -0.0007 -0.0034 -0.0039 0.0044 0.0034 -0.0007 Size t −1 -0.0262 -0.0207 -0.0283 0.0352 0.0098 0.0103** Expenses t −1 -0.5704* -0.9039*** -0.7262*** -0.473 -0.3670** 0.0763*** Stars t −1 0.192 0.2199 0.2687* 0.2617 0.1327** 0.0429 R2 i,t−1 -1.9570** -1.6729** -1.4439** 1.6143** -0.1394 0.2395*** Cons 2.1499*** 1.7914*** 1.5811*** 1.0303 2.2173*** -0.2528** N 5211 5211 5211 5209 5181 5211 R 2 0.2017 0.1863 0.1547 0.5554 0.5838 0.5315 This table provides the coefficient estimates for regressions (4), (5) and (6). See the footnote of table 1 for variable definitions. We have shown the constant (Cons), number of observations (N) and adjusted R-squared (R 2 ) but have excluded the dummies that account for years and funds’ investment styles. Significance at the 10%, 5%, and 1% levels have been indicated by the symbols *, **, and ***. 7 All estimations included category dummies with fixed effects and year-by-year fixed effects to control for unobserved heterogeneity both over time and in the cross-section. In addition, we used two-way clustering by fund and year. P. Dur´ an-Santomil et al. Finance Research Letters 58 (2023) 104626 6 statistically insignificant. These results were in line with Niessen-Ruenzi and Ruenzi (2019), and contrary to B¨ ar et al. (2011), Qiong et al. (2014) and Grossmass et al. (2019). 4.2. Performance extremity We extended the previous regression analysis using performance extremity (PE i,t ), calculated for each performance measure (see Eq. 2), as the dependent variable. The independent variables were the team dummy and the same set of controls, as specified in regression (7) below: PEit = ω +β1Teami,t−1+βjControlsi,t−1+ ε i,t.(7) We ran the panel data regressions including timeand segment-fixed effects. As can be seen in Table 3.1, we reported that the least active funds in terms of R 2 displayed less extreme performance. Funds with higher expense ratios showed more extreme performance, as did funds with more Stars. However, contrary to B¨ ar et al. (2011), we found no significant differences between individual managers and teams. We also used two variants of Eq. (7) that employed the gender diversity metric, or the female ratio and its squared term, instead of the team dummy, as follows: PEit = ω +β1GenderDiversityi,t−1+βjControlsi,t−1+ ε i,t(8) PEit = ω +β1FemaleRatioi,t−1+β2FemaleRatio2 i,t−1+βjControlsi,t−1+ ε i,t.(9) Our results, presented in Table 3.2, showed that the effects of gender diversity in the management teams were insignificant. Hence, funds with higher gender diversity did not obtain less extreme performance than those with lower gender diversity. Using the female ratio, shown in Table 3.3, similar conclusions were drawn. Furthermore, we estimated the previous models for all funds (whereby individual managers did not have gender diversity by definition) as well as those for funds managed by teams. Our findings remained Table 3 Performance extremity models. Variable 1F-alpha 4Falpha 6F-alpha NetReturn Sharpe VaR Table 3.1 – Team vs. single manager (Eq. 7) Team t −1 0.0521 0.0241 0.0015 0.0950** 0.0344 0.0887 Tenure t −1 0.0019 -0.0008 -0.0004 0.0006 0.0037 0.0025 Age t −1 -0.0057** -0.0059** -0.0057*** -0.0060** -0.0095*** -0.0062 Size t −1 -0.0297* -0.0068 -0.0148 -0.0288* -0.0534*** -0.0243 Expenses t −1 0.3793*** 0.4524*** 0.4549*** 0.2366*** 0.2081*** 0.1832 Stars t −1 0.1641*** 0.1438*** 0.1493*** 0.1015** 0.0604* 0.1175 R2 i,t−1 -0.7092*** -1.0049*** -0.9067*** -0.8184*** -1.1511*** -1.1625 Cons 1.8240*** 1.5807*** 1.6808*** 1.9640*** 2.4397*** 2.139 N 5198 5197 5197 5195 5167 5198 R 2 0.0591 0.0817 0.0782 0.0537 0.0832 0.0622 Table 3.2 – Gender diversity (Eq. 8) GenderDiversity t −1 0.0007 -0.0294 -0.0724 0.0931 -0.0389 0.1254 Tenure t −1 0.0025 -0.0003 0.0001 0.0011 0.0045 0.0027 Age t −1 -0.0059** -0.0060** -0.0056*** -0.0064*** -0.0095*** -0.0067 Size t −1 -0.0284* -0.0063 -0.015 -0.0263* -0.0527*** -0.0218 Expenses t −1 0.3842*** 0.4546*** 0.4549*** 0.2457*** 0.2113*** 0.1918 Stars t −1 0.1650*** 0.1442*** 0.1493*** 0.1032** 0.0610* 0.1192 R2 i,t−1 -0.6929*** -0.9950*** -0.9005*** -0.7960*** -1.1376*** -1.1447 Cons 1.8064*** 1.5733*** 1.6820*** 1.9299*** 2.4295*** 2.1062 N 5198 5197 5197 5195 5167 5198 R 2 0.0583 0.0816 0.0786 0.0517 0.083 0.0613 Table 3.3 – Female ratio (Eq. 9) FemaleRatio t −1 -0.0096 0.01 -0.101 0.3387 -0.0918 0.3451 FemaleRatio2 t−1 0.0232 -0.0232 0.093 -0.3401 0.2057 -0.3236 Tenure t −1 0.0021 -0.0007 0 0.0007 0.004 0.0022 Age t −1 -0.0057** -0.0057** -0.0056*** -0.0062** -0.0098*** -0.0068 Size t −1 -0.0272* -0.0055 -0.0139 -0.0255* -0.0496*** -0.0222 Expenses t −1 0.3840*** 0.4466*** 0.4496*** 0.2470*** 0.2123*** 0.1876 Stars t −1 0.1618*** 0.1428*** 0.1512*** 0.1048** 0.0603* 0.1173 R2 i,t−1 -0.6847*** -0.9906*** -0.8988*** -0.8012*** -1.1381*** -1.1329 Cons 1.7683*** 1.5546*** 1.6562*** 1.9145*** 2.3493*** 2.1008 N 5198 5197 5197 5195 5167 5198 R 2 0.0574 0.0805 0.0777 0.0522 0.0833 0.0607 This table provides the coefficient estimates for regressions (7), (8) and (9). See the footnote of table 1 for variable definitions. We have shown the constant (Cons), number of observations (N) and adjusted R-squared (R 2 ) but have excluded the dummies that account for years and funds’ investment styles. Significance at the 10%, 5%, and 1% levels have been indicated by the symbols *, **, and ***. P. Dur´ an-Santomil et al. Finance Research Letters 58 (2023) 104626 7 unchanged. 5. Robustness tests Three robustness tests were carried out. In the first one, in spite of including turnover ratios as an additional control variable, we obtained similar results. All the variables in the preceding section were winsorized at the 99% and 1% levels to reduce the effect of outliers. Therefore, as a second robustness test, we repeated the entire analysis, the complete data set having equivalent results. Finally, we also estimated the previous models using other specifications on the residuals, specifically panel-corrected standard errors (PCSE) and Newey-West, no major changes to the results being found either. 6. Conclusions Our results support the premise that investors’ decisions regarding European bond funds should not be affected by the gender characteristics of the management team or by whether a fund is soleor team-managed. Controlling for the effects of expenses, manager tenure, R 2 , turnover, fund size and fund age, we showed that individually-managed and team-managed funds obtained the same performance and risk levels in statistical terms. In addition, a higher gender diversity in the fund management team does not bring significant changes to its performance or risk. Overall, our findings indicate that the underrepresentation of women in the fund industry, or gender bias, does not make economic sense. CRediT authorship contribution statement Pablo Dur´ an-Santomil: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Validation, Writing – original draft, Writing – review & editing. Luis Otero-Gonz´ alez: Conceptualization, Project administration, Funding acquisition, Resources, Software, Supervision, Writing – review & editing. Renato Domingues: Conceptualization, Data curation, Investigation, Writing – original draft. Paulo Leite: Conceptualization, Funding acquisition, Investigation, Methodology, Writing – original draft, Writing – review & editing. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data availability The authors do not have permission to share data. Acknowledgments Luis Otero and Pablo Dur´ an acknowledge the funding of the Galician Regional Government [ED431C 2020/18] co-funded by the European Regional Development Fund (ERDF/FEDER) within the period 2020–2023. Paulo Leite acknowledges the funding of FCT – the Portuguese Foundation for Science and Technology under the project UIDB/ 04752/2020. References Adams, J.C., Nishikawa, T., Rao, R.P., 2018. Mutual fund performance, management teams, and boards. J. Bank. Finance 92, 358–368. https://doi.org/10.1016/j. jbankfin.2016.09.006. Amihud, Y., Goyenko, R., 2013. Mutual Fund’s R2 as Predictor of Performance. Rev. Financ. Stud. 26 (3), 667–694. https://doi.org/10.1093/rfs/hhs182. Atkinson, S.M., Baird, S.B., Frye, M.B, 2003. Do female mutual fund managers manage differently? J. Financ. Res. 26 (1), 1–18. https://doi.org/10.1111/14756803.00041. Babalos, V., Caporale, G.M., Philippas, N., 2015. Gender, style diversity, and their effect on fund performance. Res. Int. Bus. Finance 35, 57–74. https://doi.org/ 10.1016/j.ribaf.2015.02.020. B¨ ar, M., Kempf, A., Ruenzi, S., 2011. Is a team different from the sum of its parts? Evidence from mutual fund managers. Rev. Finance 15 (2), 359–396. https://doi. org/10.1093/rof/rfq014. Barber, B.M., Scherbina, A.D., & Schlusche, B. (2017). Performance Isn’t everything: personal characteristics and career outcomes of mutual fund managers. Available at SSRN: https://ssrn.com/abstract=3032207 or 10.2139/ssrn.3032207. Beckmann, D., Menkhoff, L., 2008. Will women be women? Analyzing the gender difference among financial experts. Kyklos 61 (3), 364–384. https://doi.org/ 10.1111/j.1467-6435.2008.00406.x. Bliss, R.T., Potter, M.E., 2002. Mutual fund managers: does gender matter? J. Bus. Econ. Stud. 8, 1–17. Bliss, R.T., Potter, M.E., Schwarz, C, 2008. Performance characteristics of individually-managed versus team-managed mutual funds. J. Portfolio Manag. 34 (3), 110–119. https://doi.org/10.3905/jpm.2008.706248. Chen, J., Hong, H., Huang, M., Kubik, J.D., 2004. Does fund size erode mutual fund performance? The role of liquidity and organization. Am. Econ. Rev. 94 (5), 1276–1302. http://www.jstor.org/stable/3592823. P. Dur´ an-Santomil et al. Finance Research Letters 58 (2023) 104626 8 Chen, J.J., Xie, L., Zhou, S., 2020. Managerial multi-tasking, Team diversity, and mutual fund performance. J. Corporate Finance 65, 101766. https://doi.org/ 10.1016/j.jcorpfin.2020.101766. Dass, N., Nanda, V., Wang, Q, 2013. Allocation of decision rights and the investment strategy of mutual funds. J. Financ. Econ. 110 (1), 254–277. https://doi.org/ 10.1016/j.jfineco.2013.06.004. Elton, E., Gruber, M., Blake, C., 1995. Fundamental economic variables, expected returns, and bond fund performance. J. Finance 50 (4), 1229–1256. https://doi.org/ 10.1111/j.1540-6261.1995.tb04056.x. Grossmass, L., Lin, M.-T, & Poon, S.-H. (2019). Does gender diversity impact performance and corporate social responsibility investment decisions in mutual funds?. Working paper. Available at http://wp.lancs.ac.uk/mhf2019/files/2019/05/MHF-2019-023-Ming-Tsung-Lin.pdf. Han, Y., Noe, T., Rebello, M, 2017. Horses for courses: fund managers and organizational structures. J. Financ. Q. Anal. 52 (06), 2779–2807. https://doi.org/10.1017/ s0022109017000795. Karagiannidis, I., Booth, G.G., 2022. The influence of management design on mutual fund performance. Multinational Finance J. 26 (1-2), 1–26. Lawrence, S. (2022). Diversity matters: the role of gender diversity on US active equity fund performance. Available at SSRN: https://ssrn.com/abstract=4081494 or 10.2139/ssrn.4081494. Leite, P., Cortez, M., 2017. The conditional performance of Euro bond funds: evidence from Portugal during the debt crisis. Spanish J. FinanceAccount. 46 (2), 212–226. https://doi.org/10.1080/02102412.2016.1265708. Morningstar (2018). Fund managers by gender: through the performance lens. Niessen-Ruenzi, A., Ruenzi, S., 2019. Sex matters: gender bias in the mutual fund industry. Manag. Sci. 65 (7), 2947–3448. https://doi.org/10.1287/mnsc.2017.2939. Otero, L., Dur´ an, P., Domingues, R., 2019. Do investors obtain better results selecting mutual funds through quantitative ratings? Spanish J. Finance Account. 49 (3), 265–291. https://doi.org/10.1080/02102412.2019.1622066. Otero-Gonz´ alez, L., Leite, P., Dur´ an-Santomil, P., Domingues, R., 2022. Morningstar Star ratings and the performance, risk and flows of European bond mutual funds. International Review of Economics & Finance 82, 479–496. https://doi.org/10.1016/j.iref.2022.07.003. Prather, L.J., Middleton, K.L., 2002. Are N+1 heads better than one? The case of mutual fund managers. J. Econ. Behav. Org. 47 (1), 103–120. https://doi.org/ 10.1016/S0167-2681(01)00172-X. Prather, L.J., Middleton, K.L., 2006. Timing and selectivity of mutual fund managers: an empirical test of the behavioral decision-making theory. J. Empiric. Finance 13 (3), 249–273. https://doi.org/10.1016/j.jempfin.2005.10.00. Qiong, L., Yang, K., Zhou, Y, 2014. The effect of management team characteristics on performance and style extremity of mutual fund portfolios. J. Ind. Eng. Manag., 10 (1). https://doi.org/10.3926/jiem.987. Rau, P.R., & Wang, J. (2022). Do Investors Pay less attention to women (Fund Managers)?. Available at SSRN: https://ssrn.com/abstract=3926970 or 10.2139/ ssrn.3926970. Sehrish, S., Naeem, M.A., Karim, S., Yarovaya, L., 2023. Gender and mutual fund liquidity. Brit. J. Manag. https://doi.org/10.1111/1467-8551.12727. In press. P. Dur´ an-Santomil et al.