Investing in talents mager characteristics and hegde fund performance

Investing in Talents: Manager Characteristics and Hedge Fund Performances Haitao Lia , Xiaoyan Zhangb , and Rui Zhaoc March 2008 a Li is from the Stephen M Ross School of Business, University of Michigan, Ann Arbor, MI 48109 b Zhang is from the Johnson Graduate School of Management, Cornell University, Ithaca, NY 14853 c Zhao is from BlackRock Inc., 40 East 52nd Street, New York, NY 10022 We thank Andrew Ang, Warren Bailey, Lauren Cohen, Jed Devaro, Ravi Jagannathan, Wei Jiang, Maureen O’Hara, Gideon Saar, Clara Vega, and seminar participants at Columbia University, Cornell University, and the University of Wisconsin at Milwaukee for helpful comments We thank David Hsieh for making the lookback straddle returns available on his Web site and Ken French for making the Fama-French factor portfolios available on his Web site We are responsible for any remaining errors Electronic copy available at: https://ssrn.com/abstract=990753 Investing in Talents: Manager Characteristics and Hedge Fund Performances Abstract Using a large sample of hedge fund manager characteristics, we provide one of the first comprehensive studies on the impact of manager characteristics, such as education and career concern, on hedge fund performances We document differential ability among hedge fund managers in generating risk-adjusted returns and flow-chasing-return behaviors among hedge fund investors In particular, we find that managers from higher-SAT undergraduate institutes tend to have higher raw and risk-adjusted returns, more inflows, and take less risks Our results provide supporting evidence to some of the assumptions and implications of the rational theory of active portfolio management of Berk and Green (2004) However, in contrast to the results for mutual funds, we find a rather symmetric relation between hedge fund flows and past performance, and that hedge fund flows not have a significant negative impact on future performance JEL: G23, G11, G12 Keywords: hedge fund performance, manager characteristics, hedge fund flows Electronic copy available at: https://ssrn.com/abstract=990753 An investment in a hedge fund is really an investment in a manager and the specialized talent he possesses to capture profits from a unique strategy.–– Sanford J Grossman, The Wall Street Journal, September 29, 2005 Hedge funds have experienced tremendous growth in the past decade According to the SEC and various hedge fund research companies, the amount of assets under management by hedge funds has grown from about $15 billion in 1990 to about $1 trillion by the end of 2004, and the number of existing hedge funds is about 7,000 to 8,000 Some industry experts even predict that hedge fund assets could exceed $3.2 trillion globally by 2009 As a result, hedge funds have attracted enormous attention from a wide range of market participants and academics in recent years Hedge funds differ from mutual funds in the ways they operate and how their managers are compensated For example, hedge funds are not subject to the same level of regulation as mutual funds and thus enjoy greater flexibility in their investment strategies As a result, hedge funds frequently use short selling, leverage, and derivatives, strategies rarely used by mutual funds, to enhance returns and/or reduce risk While mutual funds charge a management fee proportional to assets under management (usually 1-2%), most hedge funds charge an incentive fee, typically 15% to 20% of profits, in addition to a fixed 1-2% management fee Moreover, hedge fund managers often invest a significant portion of their personal wealth in the funds they manage; and many funds have a high watermark provision, which requires managers to recoup previous losses before receiving incentive fees Hedge funds also differ from mutual funds in the economic functions they perform in the economy As pointed out by Sanford J Grossman (2005) in a recent Wall Street Journal commentary, while mutual funds enable small investors to pool their money and invest in diversified portfolios, “a hedge fund is a vehicle for acquiring the specialized talents of its manager.” Grossman observes that, “Hedge funds are typically managed by an entrepreneur, and hedge fund returns are the outcome of an entrepreneur activity.” As a result, Grossman emphasizes that a “fund’s return will be no better than its management and the economic environment in which it produces its product An investor should understand the product being produced and the manager producing it.” Grossman’s observation suggests that the performance of a hedge fund depends crucially on both the investment strategies it follows and the talents of its manager(s) in implementing such strategies Though great progress has been made in understanding the risk and return properties of many hedge fund strategies,1 only limited analysis has been done on the impact of manager talents on hedge fund performances in the literature Just like any entrepreneur activity, it is entirely possible See, for example, the interesting works of Agarwal and Naik (2004), Fung and Hsieh (1997, 2001), and Mitchell and Pulvino (2001), among others Electronic copy available at: https://ssrn.com/abstract=990753 that some hedge fund managers are better than others in making investment decisions Given the billions of dollars poured into hedge funds from pension funds, endowments, and other institutional investors each year, identifying manager characteristics that lead to superior performance could be very helpful to potential investors in selecting hedge fund managers and also could have profound welfare implications In addition to the practical value of identifying superior managers, understanding the impact of manager talents on hedge fund performances also provides a way of testing some of the assumptions and implications of the rational theory of active portfolio management of Berk and Green (2004).2 For example, one important assumption of Berk and Green (2004) is that mutual fund managers have differential ability in delivering positive risk-adjusted returns However, if the theory of Berk and Green (2004) is true, then it could be difficult to identify cross-sectional differences in risk-adjusted returns in equilibrium using mutual fund data, because most mutual funds might have increased their sizes to the extent that their risk-adjusted returns have disappeared Moreover, due to established investment process and team-oriented approach to portfolio management in many mutual fund families, the impact of individual managers on mutual fund performances is likely to be small as well Consistent with this view, Chevalier and Ellison (1999a) find that although mutual fund managers from higher-SAT institutes tend to have higher raw returns, their results become much less significant for risk-adjusted returns In contrast, the unique structure of hedge funds suggests that manager talents might be more important for hedge fund performances Since a significant part of hedge fund compensation comes from incentive fees, hedge fund managers may not want to grow their funds to the extent that all risk-adjusted returns disappear In addition, many hedge funds have a high watermark provision, and many hedge fund managers have personal wealth invested in their funds As a result, inferior hedge fund returns could be really costly for these managers Therefore, even in equilibrium there might be an optimal fund size at which abnormal returns still exist In addition, the entrepreneur nature of hedge fund operations suggests that hedge fund performance should depend more significantly on individual managers Therefore, hedge fund data provide a unique opportunity for testing the theory of Berk and Green (2004) In this paper, we provide a comprehensive empirical analysis of the impact of manager characteristics on hedge fund performances We conjecture that everything else equal, a manager who is more talented and more devoted to his/her job is more likely to have better performance We use intelligence and education as proxies for manager talents We use manager career concern as Berk and Green’s (2004) model combines three elements: competitive provision of capital by investors to mutual funds, differential ability to generate high average returns across managers but decreasing returns to scale in deploying these abilities, and learning about managerial ability from past returns The theory predicts that mutual fund managers increase the size of their funds, and their own compensation, to the point at which expected returns to investors not outperform passive benchmarks in equilibrium Electronic copy available at: https://ssrn.com/abstract=990753 a proxy for manager job commitments The rationale is that a manager who is under pressure to establish his/her career at an early stage might be willing to put in more effort than a more established manager We first construct probably the most comprehensive dataset on manager characteristics based on more than 4,000 hedge funds covered by TASS between 1994 and 2003 Boyson (2003, 2004) studies hedge fund performance and manager career concerns using a much smaller sample of about 200 funds up to 2000 In contrast, our dataset covers a wide range of information on personal, educational, and professional backgrounds of managers of 1,002 hedge funds up to 2003 Specifically, we collect information on the following six characteristics of the lead manager of each fund if such information is available: the composite SAT score for the manager’s undergraduate institute (SAT), whether the manager has a CPA or CFA, whether the manager has an MBA degree, the total number of years of working (WORK), the number of years of working at the specific hedge fund (TENURE), and the manager’s age (AGE) Broadly speaking, the six characteristics can be divided into two groups: SAT, CFA/CPA and MBA dummies represent intelligence and education; WORK, TENURE, and AGE could represent working experience and career concern We also conduct a careful analysis on risk adjustments for hedge fund returns to obtain hedge fund abnormal performance Many studies have shown that due to the dynamic trading strategies and derivatives used by hedge funds, traditional linear asset pricing models could give misleading results on hedge fund performance Given that there are no well-established risk-adjustment methods for hedge fund returns, we choose a wide variety of models to ensure the robustness of our results Specifically, in addition to the traditional Fama and French (1993) (hereafter FF) three-factor model, to capture the nonlinearity in hedge fund returns, we also consider a wide variety of models that include returns on various hedge fund indices and options as factors In particular, we consider the model of Agarwal and Naik (2004) and the seven-factor model first proposed by Fung and Hsieh (2004) and used recently by Fung, Hsieh, Naik, and Ramadorai (2006) (hereafter FHNR) As a further robustness check, we consider two specific hedge fund strategies whose risk and return properties have been carefully examined and thus are reasonably well understood in the literature These are the trend following strategy studied by Fung and Hsieh (2001) and the risk arbitrage strategy studied by Mitchell and Pulvino (2001) Based on the new dataset on manager characteristics and various risk-adjustment methods, we document a strong impact of manager education on different aspects of hedge fund performances, such as fund risk-taking behaviors, raw and risk-adjusted returns, and fund flows Specifically, we find that managers from higher-SAT institutes tend to take less (overall, systematic, and idiosyncratic) risks and have higher raw and risk-adjusted returns In our analysis, risk-adjusted returns include both alpha and appraisal ratio (the ratio between alpha and residual volatility) We also find that managers from higher-SAT institutes tend to attract more capital inflows On Electronic copy available at: https://ssrn.com/abstract=990753 the other hand, we find some weak evidence that managers with longer years of working tend to have lower raw and risk-adjusted returns and take less risks These results are very robust to the different risk-adjustment benchmarks, sample periods, and types of funds (funds of funds vs regular hedge funds) we consider Although we document differential ability among hedge fund managers in generating riskadjusted returns and flow-chasing-return behaviors among hedge fund investors, we find mixed results in our tests of other implications of Berk and Green’s theory For example, unlike the convex relation between flows and lagged returns documented for mutual funds, we find that hedge fund flows react to lagged returns rather symmetrically We also find a significant and robust negative relation between hedge fund flows and both fund age and lagged fund size This suggests that there might be an optimal fund size beyond which hedge fund managers start to take less inflows Finally, in contrast to the results for mutual funds, we not find a significant negative impact of current fund flows on future fund performances for hedge funds Our paper contributes to the fast-growing literature on hedge funds by providing (i) one of the first systematic studies on the impact of manager characteristics on the cross-sectional differences in hedge fund performances and (ii) an empirical test of Berk and Green’s (2004) theory using hedge fund data Our paper also complements and extends FHNR (2006), the first study that tests Berk and Green’s (2004) theory using hedge fund data.3 While both FHNR (2006) and our paper show that some hedge fund managers are indeed better than others, our study traces superior hedge fund performances to important manager characteristics, such as education and career concern Therefore, our paper provides an economic explanation for the existence of superior performances as well as a guidance on how to identify superior hedge fund managers based on manager characteristics Our results on flow-return relation also are broadly consistent with that of FHNR (2006) While FHNR (2006) show that fund flows negatively affect the transition probability of have-alpha funds to remain in have-alpha category, the effect of flows on future risk-adjusted returns is not statistically significant Collectively, the results of our paper and FHNR (2006) suggest that the basic mechanisms of Berk and Green’s (2004) model are also at work in the hedge fund industry However, because of the unique compensation structure of hedge funds, hedge fund managers not have the same incentives as mutual fund managers in growing the size of their funds Therefore, the negative impact of fund flows on future returns for hedge funds may not be as strong as that for mutual funds, and hedge funds may still exhibit positive abnormal returns even in equilibrium.4 Our results strongly suggest that hedge funds are Using data on funds of funds, FHNR (2006) show that some hedge fund managers are able to deliver better alphas than others They further show that the alpha producing funds of funds (denoted as have-alpha funds) experience greater and steadier capital inflows than the other funds that fail to produce alphas (denoted as betaonly funds) This view is also consistent with the findings of Kosowski, Naik, and Teo (2007) Using powerful bootstrap Electronic copy available at: https://ssrn.com/abstract=990753 very different from mutual funds, and a manager’s talents and motivations should be important considerations in selecting hedge fund managers The remainder of the paper proceeds as follows In section I, we introduce our data on hedge fund returns and manager characteristics In section II, we introduce a wide variety of riskadjustment benchmarks for hedge fund returns In section III, we examine the relation between different aspects of hedge fund performance and manager education/career concerns In Section IV, we specialize our analysis to two special hedge fund strategies whose risk and return properties have been relatively well understood in the literature In Section V, we study the behaviors of fund flows and the impact of fund flows on future fund performances Section VI concludes I Data on Hedge Fund Returns and Manager Characteristics The data on hedge fund returns and manager characteristics are obtained from TASS Among all the datasets that have been used in the existing hedge fund literature, the TASS database is probably the most comprehensive one TASS builds its dataset based on surveys of hedge fund managers Funds report to TASS mainly for marketing purposes, because they are prohibited from public advertisements Overall, TASS covers more than 4,000 funds from November 1977 to September 2003 All funds are classified into “live” and “graveyard” categories “Live” funds are those that are active as of September 2003 Once a fund is considered no longer active, it is transferred to the “graveyard” category.5 The “graveyard” database did not exist before 1994 Thus, funds that became inactive before 1994 were not recorded by TASS To mitigate the potential problem of survivorship bias, we include both “live” and “dead” funds and restrict our sample to the period between January 1994 and September 2003, yielding a sample of 4,131 funds Our analysis focuses on different aspects of hedge fund performances to obtain a more complete picture These include fund risk-taking behaviors (measured by overall, systematic, and idiosyncratic risks), raw and risk-adjusted returns, and fund flows We make these choices because we believe that managers would devote their time and effort to improve performance measures that could lead to higher compensations, which could come from management/incentive fees and personal wealth invested in their funds For example, Goetzmann, Ingersoll, and Ross (2003) argue that both returns and capital flows are important for hedge fund manager compensation, although the relative importance depends on market condition and is time-varying The monthly returns provided by TASS are net of management/incentive fees and other fund expenses, and are and Bayesian methods, the authors show that the abnormal performance of top hedge funds cannot be attributed to luck and that hedge fund abnormal performance persists at annual horizons A fund is in “graveyard” because either it had bad performance or it had stopped reporting to TASS For instance, a fund might have done well and attracted enough capital, and it no longer has any incentive to report to TASS Electronic copy available at: https://ssrn.com/abstract=990753 closely related to actual returns received by investors TASS also provides data on several fund characteristics, such as management and incentive fees, whether a fund has a high watermark, and whether its managers have personal wealth invested in the fund Other than returns and fund characteristics, TASS also provides rich information on personal, educational, and professional backgrounds of managers of most funds Although the return data of TASS have been extensively studied in the literature, our paper is one of the first that examines the impact of manager characteristics on hedge fund performance Specifically, we identify a lead manager of a particular fund and construct a dataset on the characteristics of this manager.6 For educational background, we identify the undergraduate college the manager attended, the SAT score of the college from U.S News and Princeton Review of 2003,7 whether the manager has an MBA degree, and whether the manager has a CFA or CPA For professional background, we obtain the years the manager has worked (WORK) either directly from the dataset or assume that the manager started working right after MBA if he/she has one However, if neither information is available, then WORK is missing We also obtain the number of years the manager has worked at a particular fund, which we refer to as manager tenure (TENURE) For personal information, we obtain the age of the manager (AGE), which is either reported in the dataset or inferred based on the assumption that the manager was 21 upon graduation from college Generally speaking, SAT, MBA, and CPA/CFA dummies could capture either the intelligence or education of the fund manager, while WORK, TENURE, and AGE could capture the working experience and career concern of the manager Out of the 4,000 funds covered by TASS, we are able to identify most of the characteristics of the lead manager for 1,002 funds Panel A of Table provides summary statistics on quarterly returns, and fund and manager characteristics for the 1,002 hedge funds.8 For fund characteristics, we report incentive and management fees, whether the fund has a high watermark, whether the manager has personal wealth invested in the fund, the age and asset value of the fund, and the number of managers of the fund For manager characteristics, we include SAT, MBA, and CFA/CPA dummies, AGE, WORK, and TENURE To be consistent with the Fama and MacBeth (1973) regression approach used in later analysis, we report time series averages of cross-sectional distributions of each individual variable That is, at each quarter, we calculate the mean, standard deviation, minimum, first quartile, median, third quartile, and maximum of the distribution of We choose the founder of a fund as the lead manager, and for funds with multiple founders we choose the one that is in charge of investment strategies or for whom the characteristics information is available We repeat our analysis using SAT scores in 1973, 1983, and 1993 obtained from Lovejoy’s college guide and U.S News and reach very similar results The general level of SAT scores has increased from early 1970s to 2003 by about 100 points We use quarterly returns mainly because, as documented in Getmansky, Lo, and Makarov (2003), quarterly returns might be more precisely measured than monthly returns for hedge funds due to liquidity issues Electronic copy available at: https://ssrn.com/abstract=990753 each variable Then we report the time series averages of each of the above quantities over all quarters in our sample period The average raw and excess quarterly returns are 3.33% and 2.28% respectively, with a wide dispersion The lowest return is around -17% and the highest is more than 26% per quarter In terms of fund characteristics, we find that most funds charge a 20% incentive fee and a 1-1.5% management fee About 40% of the funds have a high watermark, and managers of 60% of the funds have personal wealth invested in their own funds The mean and median ages of funds are about and years, respectively The mean and median fund sizes are about $86 million and $31 million, respectively Although the majority of the funds are run by one or two managers, certain funds have more than 10 managers The SAT scores range from the lowest of 878 to the highest of 1,511 with a mean/median around 1,300 In results not reported, about 30% of the managers graduated from Ivy league universities About 17% of the managers have either a CFA or CPA, and 47% of the managers have an MBA degree, while the rest fail to report on this item.9 For many funds, the age variable is missing and in total we only have around 7,351 quarter-fund observations with age information For those funds with age information, the mean and median manager ages are about 44 and 42.5 years, respectively, with the youngest of 27 years and the oldest of more than 72 years.10 Out of the 1,002 funds, we directly observe the WORK variable for 899 funds For the rest of the funds, we construct WORK based on the finishing date of MBA degree On average, managers have close to 20 years of working experience, with the shortest of years and the longest of 50 years The average tenure with current fund is about to years, with the shortest of less than one quarter and the longest of 20 years Panel B of Table reports the correlations among fund excess returns and various fund and manager characteristics We find a positive correlation between fund excess returns and SAT, which provides preliminary evidence that managers from higher-SAT colleges are more likely to have better performance On the other hand, we find negative correlations between excess returns and fund age and several working experience variables This provides preliminary evidence that younger funds and managers with less working experience tend to have better performance We find a strong positive correlation of 0.93 between fund age and manager tenure, which is consistent with the typical structure of hedge funds: They are usually established by a few important managers who tend to stay with the fund.11 Chevalier and Ellison (1999b) argue that A zero value of an MBA or CFA/CPA dummy variable does not necessarily mean that the manager does not have an MBA or CFA/CPA, respectively It could be that the manager fails to report this information 10 We not include age in our regressions because age is missing for about 40% of the funds However, due to the high correlation between age and years of working, we will not lose much information by omitting age in our analysis 11 This result has important implications for interpreting the causality of our later finding that smarter managers tend to have higher risk-adjusted returns Although we interpret this result as evidence that smarter managers can deliver better returns, an alternative interpretation is that smarter managers are attracted to better-performing Electronic copy available at: https://ssrn.com/abstract=990753 years of working is a better proxy for working experience than manager tenure In our empirical analysis, we use WORK as a proxy for working experience or career concern, and we always include fund age and lagged fund size as fund characteristics controls We also find significant positive correlations between fund size and SAT/WORK, suggesting that manager characteristics affect not only the returns but also the sizes of hedge funds Due to the nature of currently available hedge fund datasets, most empirical studies of hedge funds potentially face various selection biases in their data.12 To minimize the impact of survivorship bias, we restrict our sample to the period between 1994 and 2003 which include both graveyard and live funds Panel C of Table provides a comparison between graveyard and live funds The summary statistics of graveyard and live funds are constructed in a similar way as that in Panel A Consistent with conventional wisdom, we find that live funds tend to have higher raw/excess returns and more assets under management Although there are some differences between graveyard and live funds in terms of fund and manager characteristics, these differences are not very significant.13 Panel D of Table compares the funds with manager characteristics with the rest of the funds covered by TASS In general, we find that funds with manager characteristics tend to be younger, have higher returns and less assets under management than funds without manager characteristics II Risk-Adjustment Benchmarks for Hedge Fund Returns The rich dataset constructed in the previous section allows us to examine the relation between hedge fund performance and manager characteristics One challenge we face in this analysis is that risk adjustments for hedge fund returns are much more difficult due to their use of derivatives and dynamic trading strategies Many studies have shown that standard linear asset pricing models fail to adequately capture the risk and return properties of most hedge funds, and it is fair to say that there is no well-established method for hedge fund risk adjustments in the existing literature Therefore, to ensure robust findings, we consider two broad classes of models to obtain risk-adjusted hedge fund returns In the first class of models, we use various hedge fund indices as benchmarks to adjust for risks in hedge fund returns The basic idea behind this approach is that these indices might be able to capture the risk exposures of average hedge funds and automatically adjust for the nonlinearity hedge funds Though this interpretation could be true for mutual funds, the 0.93 correlation coefficient suggests that the hedge funds in our sample are most likely started by their current managers 12 13 See Ackermann, McEnally, and Ravenscraft (1999) for a taxonomy of potential biases in hedge fund datasets One reason that live and graveyard funds have similar SAT scores is that graveyard funds include funds that have done poorly as well as funds that have done well and stopped reporting to TASS In results not reported, we divide the graveyard funds into finer sub-categories and find the liquidated funds on average have lower SATs than the graveyard funds that have done well Electronic copy available at: https://ssrn.com/abstract=990753 Fung, W., D Hsieh, N Naik, and T Ramadorai, 2006, Hedge funds: Performance, risk, and capital formation, working paper, Duke University and London Business School Getmansky, M., 2004, The life cycle of hedge funds: Fund flows, size and performance, working paper, University of Massachusetts, Amherst Getmansky, M., A Lo, and I Makarov, 2003, An econometric model of serial correlation and illiquidity in hedge fund returns, Journal of Financial Economics 74, 529-610 Goetzmann, W., J Ingersoll, and S Ross, 2003, High water marks and hedge fund management contracts, Journal of Finance 58, 1685-1717 Grossman, S J., 2005, Hedge funds today: Talent required, Commentary, The Wall Street Journal, September 29, 2005 Kosowski, R., N Naik, and M Teo, 2007, Do hedge funds deliver alpha? A Bayesian and bootstrap analysis, Journal of Financial Economics 84, 229-264 Mitchell, M., and T Pulvino, 2001, Characteristics of risk and return in risk arbitrage, Journal of Finance 56, 2135—2175 25 Electronic copy available at: https://ssrn.com/abstract=990753 Table Summary Statistics of Quarterly Returns and Fund/Manager Characteristics This table provides summary statistics of quarterly returns and fund/manager characteristics for 1,002 hedge funds from the TASS database between January 1994 and September 2003 Quarterly returns are calculated as percentage changes in net asset values during the quarter, net of management/incentive fees and other fund expenses Quarterly excess returns are the difference between quarterly returns and quarterly risk-free interest rate Fund characteristics include management fee, incentive fee, whether a fund has a high watermark, whether manager has personal capital invested in the fund, fund age, assets under management, and total number of managers Manager characteristics are the characteristics of a manager of a particular fund that we identify as the lead manager The variable SAT represents the composite SAT score of the undergraduate college that a manager attended from the U.S News and Princeton Review of 2003 The variable AGE represents the age of a manager and is either reported in the database or inferred based on the assumption that the manager was 21 upon graduation from college The variable WORK, which represents the number of years that a manager has worked, is either obtained directly from the dataset or is calculated by assuming that the manager started working right after MBA if he/she has one The variable TENURE represents the number of years that a manager has been with a fund and is obtained directly from the dataset The MBA and CFA/CPA dummies measure whether a manager has an MBA degree or a CFA/CPA, respectively Live funds are those funds that are active as of September 2003, and dead funds are those in the “graveyard” category Panel A Summary statistics of quarterly fund returns and fund/manager characteristics Fund Returns Quarterly Return % Quarterly Ex Return % Fund Characteristics Incentive Fee Management Fee High Watermark Dummy Personal Capital Dummy Fund Age Fund Size ($Millions) Number of Managers Manager Characteristics SAT (/100) CFA/CPA Dummy MBA Dummy AGE WORK TENURE Mean Std Dev Minimum Q1 Median Q3 Maximum 3.33 2.28 6.50 6.49 -16.14 -17.12 0.03 -1.02 2.91 1.85 6.42 5.36 26.75 25.67 17.93 1.27 0.39 0.59 3.88 86.41 2.02 5.99 0.58 0.47 0.48 3.18 180.73 1.30 0.00 0.00 0.00 0.00 0.50 0.28 1.00 20.00 1.00 0.00 0.00 1.59 9.71 1.00 20.00 1.00 0.26 0.69 2.98 30.96 2.00 20.00 1.51 0.96 1.00 5.26 90.13 2.29 33.06 4.66 1.00 1.00 19.71 2091.27 10.14 13.09 0.17 0.47 44.03 19.92 3.71 1.42 0.37 0.50 8.85 8.80 3.01 8.78 0.00 0.00 27.16 4.08 0.09 11.99 0.00 0.00 37.04 13.84 1.51 13.30 0.00 0.21 42.52 17.81 2.88 14.21 0.00 1.00 50.61 25.34 5.08 15.11 1.00 1.00 72.74 49.48 19.52 26 Electronic copy available at: https://ssrn.com/abstract=990753 Panel B Correlations between quarterly excess returns and fund/manager characteristics Fund Age Log Size SAT AGE WORK TENURE Excess Returns -0.02 0.00 0.02 -0.02 -0.04 -0.01 Fund age 0.24 0.04 0.22 0.26 0.93 Log(size) 0.09 0.02 0.07 0.27 SAT AGE WORK -0.04 0.04 0.03 0.85 0.29 0.30 Panel C Fund/manager characteristics of live and dead funds Variable Fund Returns Quarterly Return % Quarterly Excess Return% Fund Characteristics Incentive Fee Management Fee High Watermark Dummy Personal Capital Dummy Fund Age Fund Size ($Millions) Number of Managers Manager Characteristics SAT (/100) CFACPA Dummy MBA Dummy AGE WORK TENURE Dead funds Mean Std Dev Live funds Mean Std Dev 2.97 1.89 7.08 7.07 3.44 2.39 6.12 6.11 18.67 1.26 0.24 0.65 3.53 57.65 1.73 5.28 0.53 0.38 0.45 3.05 136.33 0.87 17.60 1.27 0.47 0.55 3.98 96.64 2.16 6.24 0.60 0.49 0.49 3.22 191.61 1.43 13.08 0.10 0.51 44.96 20.34 3.50 1.39 0.29 0.50 9.76 9.55 3.06 13.11 0.21 0.46 43.35 19.48 3.74 1.43 0.41 0.50 8.42 8.33 2.96 Panel D Summary statistics of funds with and without manager characteristics Variable Fund Returns Quarterly Return % Fund Characteristics Incentive Fee Management Fee High Watermark Dummy Personal Capital Dummy Fund Age Fund Size ($Millions) With Manager Characteristics Mean Std Dev Without Manager Characteristics Mean Std Dev 3.33 6.50 2.66 11.01 17.93 1.27 0.39 0.59 3.88 86.41 5.99 0.58 0.47 0.48 3.18 180.73 15.86 1.56 0.28 0.47 4.73 107.41 7.92 0.93 0.41 0.49 4.96 491.44 27 Electronic copy available at: https://ssrn.com/abstract=990753 Table Raw Returns and Manager Characteristics This table reports the results of Fama-MacBeth regressions of hedge fund quarterly excess returns and total return volatilities on manager characteristics controlling for fund age and lagged fund size Quarterly excess return is calculated as the difference between raw quarterly return and quarterly risk-free rate Total return volatility is calculated as the volatility of monthly returns of the past twelve months The variable SAT represents the composite SAT score of the undergraduate college that a manager attended from the U.S News and Princeton Review of 2003 The variable WORK, which represents the number of years that a manager has worked, is either obtained directly from the dataset or is calculated by assuming that the manager started working right after MBA if he/she has one Both fund age and lagged fund size are obtained directly from the dataset To eliminate outliers, we delete top and bottom 1% observations for each quarter We report t-statistics right below the parameter estimates in italic, where ***, ** and * entries represent significance at the 1%, 5% and 10% level, respectively The time-series averages of quarterly adjusted R2 are also reported Intercept SAT WORK Fund Age Lagged Size Adjusted R2 Quarterly Excess Return 4.562*** 2.80 0.091** 2.05 -0.027*** -3.65 -0.006 -0.23 -0.171** -2.27 2.58% Total Return Volatility 11.537*** 26.40 -0.103*** -6.14 -0.016*** -8.60 0.038*** 4.33 -0.392*** -20.34 10.46% 28 Electronic copy available at: https://ssrn.com/abstract=990753 Table Risk-Adjusted Returns and Manager Characteristics This table reports the results of Fama-MacBeth regressions of hedge fund alpha, factor loadings, residual volatility, and appraisal ratio (the ratio between alpha and residual volatility) under different benchmark models on manager characteristics controlling for fund age and lagged fund size The three models, INDEX, FoF, and STYLE, use the broad hedge fund index (a weighted average of returns of all hedge funds) provided by TASS, the index of funds of funds (a weighted average of returns of funds of funds), and style indices (the weighted average returns of all funds within each style), as risk factors, respectively The model FF is the Fama and French (1993) three-factor model, AN represents the option-based model of Agarwal and Naik (2004), and FHNR represents the seven-factor model used in Fung, Hsieh, Naik and Ramadorai (2006) The variable SAT represents the composite SAT score of the undergraduate college that a manager attended from the U.S News and Princeton Review of 2003 The variable WORK, which represents the number of years that a manager has worked, is either obtained directly from the dataset or is calculated by assuming that the manager started working right after MBA if he/she has one Both fund age and lagged fund size are obtained directly from the dataset To eliminate outliers, we delete top and bottom 1% observations for each quarter We report t-statistics right below the parameter estimates in italic, where ***, ** and * entries represent significance at the 1%, 5% and 10% level, respectively The time-series averages of quarterly adjusted R2 are also reported Panel A Cross-sectional distributions of risk-adjusted returns under different models This panel provides distributional statistics of alphas under different models At each quarter, we calculate the alpha of each hedge fund as in equation (3) using the six risk-adjustment models we consider Then for each quarter, we calculate the mean, standard deviation, and 5, 25, 50, 75, and 95 percentiles of the alphas under each model of all hedge funds The time series averages of all the above quantities are reported for each model in the table Mean Std Dev 5% 25% 50% 75% 95% INDEX 1.23 4.94 -7.14 -1.17 1.18 3.62 9.64 FoF 1.09 5.30 -8.07 -1.35 1.14 3.70 9.90 STYLE 0.77 4.97 -7.63 -1.61 0.77 3.09 9.15 FF3 1.15 5.01 -7.31 -1.16 1.14 3.56 9.65 AN 1.79 5.70 -7.33 -1.07 1.42 4.43 11.75 FHNR 2.41 4.35 -4.98 0.18 2.31 4.51 10.08 29 Electronic copy available at: https://ssrn.com/abstract=990753 Panel B Fama-MacBeth regression of risk-adjusted returns on manager characteristics Intercept SAT WORK Fund Age Lagged Size Adjusted R2 INDEX -0.237 -0.23 0.102*** 2.65 -0.021** -2.44 -0.066*** -2.85 0.046 0.84 2.20% FoF -0.696 -0.66 0.127*** 3.38 -0.018** -2.41 -0.055*** -2.99 0.051 0.99 2.00% STYLE 0.600 0.56 0.164*** 4.88 -0.027*** -3.29 -0.021 -1.10 -0.073 -1.56 2.17% FF3 -0.405 -0.34 0.174*** 3.76 -0.025*** -3.32 -0.053** -2.19 0.011 0.20 2.65% AN 2.815 1.56 0.077* 1.66 -0.013 -1.55 -0.005 -0.15 -0.093 -1.03 1.92% FHNR 0.012 0.02 0.130*** 3.57 -0.018*** -2.58 -0.071*** -3.41 0.079** 1.96 2.75% Panel C Fama-MacBeth regression of risk-taking behaviors on manager characteristics Intercept SAT WORK Fund Age Lagged Size Adjusted R2 Intercept SAT WORK Fund Age Lagged Size Adjusted R2 INDEX β 1.502*** 8.13 -0.005 -0.71 -0.002*** -3.55 0.014*** 4.09 -0.047*** -5.32 4.11% FoF β 2.459*** 9.70 -0.025** -2.12 -0.005*** -4.59 0.016*** 3.13 -0.071*** -5.52 3.68% STYLE β 1.731*** 14.06 -0.050*** -6.08 0.001 1.55 0.015*** 5.36 -0.025*** -3.83 3.18% FHNR β(MKT) 89.10*** 10.55 -1.491*** -3.52 -0.201*** -5.68 -0.116 -0.67 -2.35*** -5.57 3.62% FHNR β(SMB) 28.84*** 5.59 -1.44*** -3.64 -0.121*** -2.68 -0.399*** -2.63 0.441 1.28 3.35% FHNR β(PTFSBD) 10.317*** 5.14 -0.300*** -3.84 -0.005 -0.42 0.037 0.83 -0.348*** -4.09 4.08% FF3 β(MKT) 0.859*** 8.05 -0.014*** -3.15 -0.001*** -2.61 0.002 1.45 -0.022*** -4.52 2.79% FF3 β(SMB) -0.044 -0.40 0.009** 2.12 0.001 0.99 0.001 0.64 0.000 -0.04 2.22% FHNR β(PTFSCOM) 7.216*** 3.03 0.004 0.04 -0.017 -1.46 0.059 1.34 -0.391*** -3.19 4.72% FF3 β(HML) 0.300*** 4.23 -0.010*** -3.72 -0.001** -2.29 -0.002 -1.35 0.002 0.56 2.59% FHNR β(PTFSFX) 7.776*** 7.82 -0.127*** -2.75 -0.009 -1.32 0.067*** 3.38 -0.324*** -6.71 4.19% AN β(MKT) 1.306*** 7.81 -0.019*** -4.46 -0.003*** -3.01 0.003 1.46 -0.040*** -5.46 2.88% FHNR β(GB10y) 4.548 1.58 0.006 0.04 -0.061*** -3.84 0.036 0.86 -0.112 -0.91 3.00% 30 Electronic copy available at: https://ssrn.com/abstract=990753 AN β(OPT) 0.026*** 3.21 -0.001** -2.03 0.000 -0.95 0.000 -1.46 -0.001** -2.02 2.43% FHNR β(DEF) -3.394 -1.37 -0.134 -0.97 0.051** 2.42 -0.062 -1.09 0.149 0.97 2.66% Panel D Fama-MacBeth regression of residual volatility on manager characteristics Intercept SAT WORK Fund Age Lagged Size Adjusted R2 INDEX 10.429**** 28.26 -0.108*** -7.11 -0.015*** -8.55 0.030*** 4.44 -0.354*** -25.58 11.13% FoF 10.213*** 27.06 -0.102*** -6.80 -0.013*** -8.44 0.033*** 5.26 -0.355*** -23.65 11.70% STYLE 9.344*** 27.81 -0.049*** -3.35 -0.014*** -9.77 0.016** 2.53 -0.350*** -23.89 12.88% FF3 8.162*** 23.27 -0.075*** -5.48 -0.011*** -6.96 0.026*** 3.67 -0.287*** -16.45 12.17% AN 8.388*** 20.89 -0.055*** -5.29 -0.013*** -8.32 0.015* 1.93 -0.288*** -15.80 8.87% FHNR 7.095*** 29.97 -0.072*** -6.09 -0.014*** -13.26 0.017*** 2.69 -0.227*** -19.31 11.81% Panel E Fama-MacBeth regression of appraisal ratio on manager characteristics Intercept SAT WORK Fund Age Lagged Size Adjusted R2 INDEX -3.227*** -7.63 0.116*** 5.61 -0.008** -2.13 -0.037*** -4.88 0.169*** 6.28 4.30% FoF -3.418*** -8.95 0.125*** 6.00 -0.007** -2.11 -0.034*** -4.91 0.176*** 7.84 3.98% STYLE -2.164*** -6.18 0.133*** 6.86 -0.013*** -3.08 -0.035*** -4.76 0.089*** 5.96 2.90% FF3 -3.570*** -5.36 0.133*** 4.75 -0.014*** -2.93 -0.041*** -4.23 0.197*** 5.98 4.36% AN -2.323** -2.49 0.072*** 2.89 -0.001 -0.28 -0.022* -1.70 0.155*** 3.15 3.83% 31 Electronic copy available at: https://ssrn.com/abstract=990753 FHNR -5.829*** -11.20 0.177*** 7.34 -0.005 -0.91 -0.051*** -4.63 0.335*** 12.14 6.09% Table Further Robustness Checks Electronic copy available at: https://ssrn.com/abstract=990753 This table provides robustness checks of Fama-MacBeth regressions of excess return, total volatility, alpha, factor loading, residual volatility, and appraisal ratio (the ratio between alpha and residual volatility) on manager characteristics controlling for fund age and lagged fund size Panel A contains the results of 122 funds of funds Panel B contains the results where all hedge funds managed by the same manager are treated as one observation Specifically, we use weighted average returns of all hedge funds managed by the same manager in our regressions Out of the 607 managers, 371 managers handle only one fund, and 236 managers handle multiple funds The median number of funds managed by the 236 managers is Panel C contains results of two subperiods: Q1.1995-Q2.1998 and Q3.2000-Q3.2003 Fore brevity, we only report results based on the FHNR model, the seven-factor model used in Fung, Hsieh, Naik and Ramadorai (2006) The variable SAT represents the composite SAT score of the undergraduate college that a manager attended from the U.S News and Princeton Review of 2003 The variable WORK, which represents the number of years that a manager has worked, is either obtained directly from the dataset or is calculated by assuming that the manager started working right after MBA if he/she has one Both fund age and lagged fund size are obtained directly from the dataset To eliminate outliers, we delete top and bottom 1% observations for each quarter We report t-statistics right below the parameter estimates in italic, where ***, ** and * entries represent significance at the 1%, 5% and 10% level, respectively The time-series averages of quarterly adjusted R2 are also reported Panel A The case of funds of funds Dependent Variables Intercept SAT WORK Fund Age Lagged Size Adjusted R2 Excess Returns -3.399* -1.77 0.315** 2.47 -0.024* -1.73 -0.154** -2.02 0.116 1.02 16.87% Total Volatility 4.145*** 6.37 -0.013 -0.41 -0.013*** -4.79 -0.032 -1.57 -0.107*** -4.41 15.20% Alpha FHNR -1.899 -1.33 0.165* 1.88 -0.005 -0.42 0.035 0.76 0.111* 1.80 18.30% Res Vol FHNR 2.971*** 10.43 -0.066*** -3.12 -0.002 -0.69 0.025*** 3.94 -0.063*** -4.76 17.61% 32 Appr Ratio FHNR -0.364 -0.23 0.161* 1.75 -0.004 -0.28 0.000 -0.01 0.048 0.67 18.72% Panel B The case of individual managers Dependent Variables Intercept Electronic copy available at: https://ssrn.com/abstract=990753 SAT WORK Fund Age Lagged Size Adjusted R2 Excess Returns 4.620*** 2.81 0.103** 2.38 -0.017* -1.85 -0.024 -0.83 -0.185** -2.28 2.83% Total Volatility 10.549*** 33.14 0.001 0.05 0.003 1.27 0.046*** 8.44 -0.422*** -26.37 10.87% Alpha FHNR -0.507 -0.61 0.127*** 3.42 -0.004 -0.51 -0.090*** -4.21 0.104** 2.34 2.95% Res Vol FHNR 9.739*** 42.75 -0.025* -1.80 0.001 0.26 0.036*** 6.68 -0.384*** -34.30 11.59% Appr Ratio FHNR -2.373*** -6.37 0.069*** 3.34 -0.014*** -3.35 -0.036*** -3.80 0.156*** 6.07 4.40% Panel C The case of different sub-periods Dependent Variables Intercept SAT WORK Fund Age Lagged Size Adjusted R2 Alpha FHNR 0.295 0.25 0.183** 2.36 -0.014 -1.11 -0.062* -1.86 0.035 0.54 3.24% Q1.1995-Q2.1998 Res Vol FHNR 8.000*** 32.07 -0.076*** -3.02 -0.010*** -14.86 0.045*** 5.74 -0.300*** -30.90 19.37% Appr Ratio FHNR -6.881*** -8.80 0.230*** 5.12 -0.012 -1.23 -0.047** -2.28 0.391*** 7.24 7.53% 33 Alpha FHNR 0.628 0.44 0.063* 1.65 -0.035*** -3.64 -0.029 -1.37 0.061 0.77 2.57% Q3.2000-Q3.2003 Res Vol FHNR 5.755*** 20.44 -0.038*** -3.25 -0.016*** -12.82 0.007 0.80 -0.169*** -20.93 6.43% Appr Ratio FHNR -3.521*** -5.10 0.076*** 2.96 -0.008 -1.03 -0.030*** -3.76 0.244*** 7.98 3.91% Table Special Case 1: Trend Following Funds Electronic copy available at: https://ssrn.com/abstract=990753 Panel A reports empirical estimates of the model of Fung and Hsieh (2001) defined in equation (6) using the average returns of 90 trend following funds in TASS for which manager characteristics are available Panel B presents the results of Fama-MacBeth regressions of excess return, alpha, factor loadings, residual volatility, appraisal ratio (the ratio between alpha and residual volatility) under Fung and Hsieh’s model on manager characteristics controlling for fund age and lagged fund size Panel C presents the results of Fama-MacBeth regressions of excess return, alpha, factor loadings, residual volatility, appraisal ratio (the ratio between alpha and residual volatility) under a trend following style index model on manager characteristics controlling for fund age and lagged fund size The style index is constructed as a value weighted average of all 90 trend following hedge fund returns The variable SAT represents the composite SAT score of the undergraduate college that a manager attended from the U.S News and Princeton Review of 2003 The variable WORK, which represents the number of years that a manager has worked, is either obtained directly from the dataset or is calculated by assuming that the manager started working right after MBA if he/she has one Both fund age and lagged fund size are obtained directly from the dataset To eliminate outliers, we delete top and bottom 1% observations for each quarter We report tstatistics right below the parameter estimates in italic, where ***, ** and * entries represent significance at the 1%, 5% and 10% level, respectively The time-series averages of quarterly adjusted R2 are also reported Panel A Estimates of Fung and Hsieh’s model using returns of trend following funds Model Model Model βbond 0.018 0.017 0.021 βcurrency 0.032 0.031 0.034 βcommodity 0.049 0.047 0.048 βstock 0.024 0.021 34 βinterest -0.008 Adjusted R2 27.8% 28.0% 26.9% Panel B Trend following fund performance and manager characteristics using Fung and Hsieh model Intercept SAT Electronic copy available at: https://ssrn.com/abstract=990753 WORK Fund Age Lagged Size Adjusted R2 Excess Return 4.813 1.00 0.055 0.28 0.004 0.11 -0.033 -0.81 -0.208 -1.06 13.63% βbond 0.124* 1.67 -0.004 -1.25 -0.001 -0.81 0.000 -0.37 -0.002 -0.73 14.29% βcurrency 0.289*** 7.40 -0.009*** -4.13 -0.001** -2.47 0.001** 2.30 -0.008*** -4.62 13.23% βcommodity 0.376*** 5.41 -0.008** -1.99 0.001 0.99 0.001 0.74 -0.013*** -4.97 12.39% βstock 0.027 0.43 0.002 1.10 0.000 -0.17 0.000 -0.29 -0.001 -0.26 13.74% Alpha 0.716 0.18 0.365** 2.44 0.026 0.72 -0.078* -1.70 -0.150 -0.92 13.94% Res vol 9.474*** 9.72 0.008 0.18 0.030*** 2.91 -0.060*** -4.31 -0.424*** -9.69 18.13% Appr ratio -3.840*** -2.70 0.184*** 3.47 0.028* 1.89 -0.052*** -2.75 0.150** 2.05 17.75% Panel C Trend following fund performance and manager characteristics using trend-follower style index Intercept SAT WORK Fund Age Lagged Size Adjusted R2 β (style) 3.757*** 6.56 -0.128*** -5.25 -0.001 -0.31 0.005 0.60 -0.065*** -3.32 11.51% Alpha (style) -1.150 -0.26 0.268* 1.69 -0.001 -0.03 -0.060* -1.76 -0.073 -0.38 12.88% Res Vol (style) 12.332*** 11.56 0.036 0.97 0.010 1.10 -0.077*** -5.75 -0.569*** -11.89 19.71% 35 Appr ratio (style) -2.922** -2.20 0.107** 2.31 0.021* 1.85 -0.043*** -2.95 0.110* 1.73 15.67% Table Special Case 2: Risk Arbitrage Funds Electronic copy available at: https://ssrn.com/abstract=990753 Panel A contains annual returns (1994-2003) of (i) the average returns of the 150 risk arbitrage funds in TASS for which manager characteristics are available, (ii) the market portfolio, (iii) one year risk-free rate, and (iv) the broad hedge fund index of TASS Panel B contains empirical estimates of the model of Mitchell and Pulvino (2001) defined in equation (7) using the average returns of all risk arbitrage funds in our sample Panel C presents the results of Fama-MacBeth regressions of excess return, alpha, factor loadings, residual volatility, appraisal ratio (the ratio between alpha and residual volatility) under Mitchell and Pulvino’s model on manager characteristics controlling for fund age and lagged fund size Panel D presents the results of Fama-MacBeth regressions of excess return, alpha, factor loadings, residual volatility, appraisal ratio (the ratio between alpha and residual volatility) under a risk arbitrage style index model on manager characteristics controlling for fund age and lagged fund size The style index is constructed as a value weighted average of all 150 risk arbitrage hedge fund returns The variable SAT represents the composite SAT score of the undergraduate college that a manager attended from the U.S News and Princeton Review of 2003 The variable WORK, which represents the number of years that a manager has worked, is either obtained directly from the dataset or is calculated by assuming that the manager started working right after MBA if he/she has one Both fund age and lagged fund size are obtained directly from the dataset To eliminate outliers, we delete top and bottom 1% observations for each quarter We report t-statistics right below the parameter estimates in italic, where ***, ** and * entries represent significance at the 1%, 5% and 10% level, respectively The timeseries averages of quarterly adjusted R2 are also reported Panel A Annual returns of risk arbitrage hedge funds Year Risk Arbitrage Index MKT Riskfree Hedge Fund Index 1994 -1.41 -0.75 3.91 -4.36 1995 15.90 35.67 5.60 21.69 1996 19.52 33.61 6.90 35.27 1997 19.16 38.71 6.62 26.70 1998 3.59 38.03 6.48 -3.98 1999 27.11 32.75 7.02 30.33 2000 12.96 -19.21 7.45 2.24 2001 8.35 -26.35 4.90 3.80 2002 4.88 -17.94 2.16 4.93 Panel B Estimates of Mitchell and Pulvino’s model using returns of risk arbitrage funds Threshold= - 3% Threshold= - 4% Threshold= - 5% βhigh 0.098 0.098 0.098 βlow 0.205 0.234 0.272 36 Adjusted R2 29.8% 30.3% 31.0% 2003 7.14 28.79 0.96 14.36 Panel C Risk arbitrage fund performance and manager characteristics using Mitchell and Pulvino model Intercept SAT Electronic copy available at: https://ssrn.com/abstract=990753 WORK Fund Age Lagged Size Adjusted R2 Excess Return -0.487 -0.39 0.336*** 3.30 -0.032** -2.50 -0.024 -0.48 -0.094** -2.32 10.03% βlow 0.035 0.47 0.029*** 4.41 -0.002*** -3.53 0.006* 1.91 -0.016*** -5.15 9.59% βhigh -0.160** -2.23 0.025*** 4.28 -0.002*** -2.93 0.003 0.85 -0.003 -1.40 9.12% Alpha -0.501 -0.72 0.200*** 4.24 -0.020*** -3.11 -0.059** -2.40 -0.004 -0.11 10.47% Res Vol 0.377 0.95 0.204*** 7.14 0.002 0.40 -0.028*** -3.34 -0.104*** -5.17 13.45% Appr ratio -2.130*** -3.13 0.005 0.10 0.005 0.55 -0.087*** -3.71 0.231*** 5.37 13.41% Panel D Risk arbitrage fund performance and manager characteristics using risk-arbitrage style index Intercept SAT WORK Fund Age Lagged Size Adjusted R2 β (style) -0.173 -1.10 0.054*** 5.94 -0.001 -0.51 0.011 1.61 -0.013 -1.63 8.68% Alpha (style) -0.175 -0.15 0.231** 2.22 -0.031*** -2.62 -0.021 -0.46 -0.052 -1.43 10.19% Res Vol (style) 0.635 1.44 0.206*** 7.56 -0.001 -0.22 -0.021** -2.15 -0.115*** -5.36 14.33% 37 Appr ratio (style) -2.010** -2.57 0.013 0.18 0.007 0.60 -0.085** -2.50 0.199*** 4.32 12.53% Table Hedge Fund Flows and Future Fund Performances Electronic copy available at: https://ssrn.com/abstract=990753 Panel A reports the results of Fama-MacBeth regressions of fund flows, measured as quarterly growth rate of assets under management in equation (8), on fund and manager characteristics The explanatory variables in the benchmark regression include lagged fund flow, lagged fund size, current fund return, lagged fund return, fund age, SAT, and WORK The dummy variable regressions introduce an additional variable in the benchmark regression, which is the product of one of the explanatory variables and a dummy variable The dummy variable for SAT equals one if SAT is bigger than 1321 (median value of SAT for all funds) and zero otherwise, and the dummy variable for WORK equals one if WORK is bigger than 18.5 years (median value of WORK for all managers) and zero otherwise Panel B provides Fama-MacBeth estimates of nonlinear relation between fund flows and lagged fund returns Lag(Return-) is the product of lagged return and a dummy variable, which equals if the lagged return is negative and zero otherwise; Lag(Return+) is the product of lagged return and a dummy variable, which equals if the lagged return is positive and zero otherwise Panel C contains the results of Fama-MacBeth regressions of current fund performances, measured as raw returns or risk-adjusted returns using the FHNR model, on lagged fund flows To eliminate outliers, we delete top and bottom 1% observations for each quarter We report t-statistics right below the parameter estimates in italic, where ***, ** and * entries represent significance at the 1%, 5% and 10% level, respectively The time-series averages of quarterly adjusted R2 are also reported Panel A Fama-MacBeth regressions of fund flows on fund and manager characteristics Benchmark X Intercept Lagged Flow Lagged Size Return Lagged Return Fund age SAT WORK 50.099*** 5.40 0.119*** 4.19 -2.985*** -6.08 0.110 0.75 0.610*** 5.95 -0.828*** -6.68 0.957** 2.47 -0.132 -1.34 C=X*D Adjusted R2 11.37% D=D(SAT>1321) Lagged Flow 58.062*** 6.55 0.173*** 4.43 -2.860*** -5.96 0.112 0.78 0.620*** 5.95 -0.875*** -5.77 Lagged Size 60.806*** 6.72 0.120*** 4.29 -3.071*** -6.16 0.105 0.74 0.629*** 6.10 -0.925*** -6.33 -0.035 -0.64 11.71% 0.119*** 2.60 10.17% 59.782*** 6.65 0.120*** 4.35 -2.957*** -6.08 0.107 0.75 0.641*** 6.23 -0.903*** -6.17 Lagged Return 59.522*** 6.66 0.122*** 4.37 -2.941*** -6.08 0.120 0.83 0.631*** 5.80 -0.907*** -6.09 62.027*** 6.70 0.120*** 4.32 -3.089*** -6.14 0.103 0.72 0.629*** 6.08 -1.150*** -8.14 Lagged Flow 59.575*** 6.69 0.148*** 3.71 -2.961*** -6.14 0.103 0.72 0.630*** 5.79 -0.857*** -5.77 0.011 0.08 10.25% 0.008 0.06 10.38% 0.498*** 3.07 10.10% 0.016 0.28 11.37% Return 38 Fund Age D=D(WORK>18.5) Lagged Lagged Return Size Return 58.541*** 60.582*** 59.378*** 6.48 6.77 6.67 0.121*** 0.119*** 0.116*** 4.32 4.17 4.08 -2.828*** -3.015*** -2.936*** -5.76 -6.21 -6.07 0.103 0.185 0.104 0.71 1.28 0.73 0.623*** 0.614*** 0.813*** 6.04 6.05 6.04 -0.787*** -0.822*** -0.844*** -5.95 -5.43 -5.59 -0.160** -2.52 10.64% -0.189 -1.10 10.59% -0.397*** -2.66 10.64% Fund Age 60.062*** 6.51 0.121*** 4.34 -2.966*** -6.00 0.120 0.84 0.622*** 6.02 -0.974*** -3.85 0.088 0.35 10.35% Panel B Fama-MacBeth estimation of nonlinear relation between fund flows and lagged returns Intercept Lagged Flow Lagged Size Return Lagged Return Lagged ReturnFund Age SAT WORK Adjusted R2 Raw Return FHNR Alpha 29.730*** 3.20 0.156*** 5.20 -2.288*** -4.06 0.237* 1.83 0.427*** 3.77 0.174 0.56 -0.435*** -2.93 1.167*** 3.11 0.033 0.30 12% 24.554*** 3.06 0.147*** 5.03 -1.862*** -4.73 0.349*** 3.07 0.547*** 3.21 0.340 1.13 -0.487*** -3.45 1.007*** 2.87 -0.003 -0.03 11% Panel C Fama-MacBeth regression of current fund performance on lagged fund flows and manager characteristics Raw Return 4.910*** 3.62 -0.004 -1.28 Intercept Lagged Flow SAT*(Lagged Flow) WORK*(Lagged Flow) Fund Age Lagged Size Adjusted R2 -0.028 -1.05 -0.091 -1.23 2.39% Raw Return 4.846*** 3.56 -0.016 -0.42 0.000 0.03 0.001* 1.72 -0.032 -1.13 -0.087 -1.16 3.82% FHNR Alpha 1.445** 2.06 0.006*** 2.75 -0.071*** -2.98 0.071* 1.79 2.63% 39 Electronic copy available at: https://ssrn.com/abstract=990753 FHNR Alpha 1.420** 2.04 -0.016 -0.37 0.001 0.45 0.000 0.80 -0.075*** -3.21 0.072* 1.89 4.12% ... funds of funds invest in a group of other hedge funds By doing this, funds of funds can diversify away idiosyncratic risks in individual hedge funds and thus achieve more stable returns The incentive... In particular, we examine (i) the relation between fund flows and manager /fund characteristics and past fund returns, and (ii) the impact of current fund flows on future fund performance We investigate... between fund flows and manager /fund characteristics and past fund returns by using the general regression in equation (5) In addition to fund age and size, we also include lagged flow and current