Fund picking

Fund picking or selection in a broader sense of the word is primarily concerned with the appropriate choice of asset class mix and expected risk and return characteristics over the holding period. For the cases when an investor is deciding between multiple similar funds, past performance has historically been used to predict future returns. Similarly to asset selection, past performance of funds has been shown to be a poor indicator of the future success of funds. In addition, better gross returns have been shown to be netted by proportionally higher management fees (Grinblatt and Titman, 1989). Moreover, the general lack of information and insufficiently long fund managers’ track record makes statistically robust fund picking difficult if not impossible. Even with nearly complete information about the fund, research mostly does not support the existence of skilled or informed fund portfolio managers (Carhart, 1997). Although individual fund picking or selection is difficult, there are certain promising facts that have been observed on a cross-section of funds and their asset returns.

It has been found that managers’ distributions of returns across the investment portfolio can reveal skill. For example, if a single asset is responsible for the majority of the outperformance, it is likely to be due to one-off lucky pick rather than skill (Kacperczyk, Sialm, and Zheng (2008), Elton, Gruber, and Blake (2011), Koswski, Timmerman, Wermers, and White (2006)). Furthermore, information on time series of fund holdings and temporal distribution of returns has shown that managers who bet more often and make persistent profits across time are more likely to be skilled (Fama and French (2010) Barras, Scaillet, and Wermers, (2010)). Last but not least, researchers have also identified additional indicators of managerial skill to supplement traditional factor analysis (Chen, Jegadeesh, and Wermers (2000), Wermers (2000), Cohen, Coval, and Pastor (2005), Kacperczyk and Seru (2007)).

By and large, investors have been found to be chasing past winning funds rather than focus on the quality of the underlying fund management. In the case of closed-end funds, investor preferences, and views of the future returns are biased as it is in the general manager selection process. This bias has been found to be systematically in one direction; causing unjustified deviation from net asset values of the closed-end fund prices. A large body of research has tried to explain this puzzling behaviour. Explanations include investor sentiment effects (see, e.g., De Long, et al., 1990 and Lee, et al., 1991); open-ending frictions (see, e.g., Brickley and Schallheim, 1985; Bradley, et al., 2010; and Brauer, 1988); agency costs (see, e.g., Barclay, et al., 1993; Khorana, et al., 2002 and Del Guercio, et al., 2003); managerial skills (Chay and Trzcinka, 1999; Coles, et al., 2000; Johnson, et al., 2006; and Berk and Stanton, 2007); and market segmentation (see, e.g., Bonser-Neal, et al., 1990; Bodurtha, et al., 1995; Gemmill and Thomas, 2002; Nishiotis, 2004; Cherkes, et al., 2009; Froot and Ramadorai, 2008; and Elton, et al., 2013).

All in all, the past return is a poor predictor of future long term performance in the fund industry. Investors should look at multiple characteristics of the fund they choose. Rather disturbingly, it has been found that 76.6% of active funds have zero alphas. 21.3% yield negative performance and are dispersed in the left tail of the alpha distribution. The remaining 2.1% with positive alphas are located at the extreme right tail. However, it is possible to benefit from looking at a cross-section of funds. Evidence suggests that momentum in mutual fund returns exists, it is feasible to earn sizable returns from selling most overvalued funds while holding most undervalued funds compared to their net asset value. However, the investor must be prepared for higher turnover as fund momentum strategies usually have short holding periods in a matter of weeks or months.

Subscribe for Newsletter

Be first to know, when we publish new content

The Encyclopedia of Quantitative Trading Strategies

Log in