In the debate of active versus passive management, some literature content that the long-term net performance of active managers is roughly similar to or worse than that of benchmark returns, on average. However, a recent study found that Active Share (the difference between a fund’s holdings and its benchmark’s holdings), coupled with the duration of fund holdings are important factors in determining ex ante the outperformance of mutual funds (1).
High active share with low turnover
In short, among high Active Share mutual funds, those with patient investment strategies (holding durations of more than two years) appear to outperform their benchmarks by an average of 2.05% a year under the five-factor model used in the study.
At the other end, irrespective of how dissimilar their portfolios are relative to their benchmarks (as measured by active share), highly traded mutual funds with short durations underperform, with average annual five-factor net returns of negative 1.44%.
Factors exposure
The authors investigate the skill level of high Active Share and long-duration fund managers by measuring their exposures to seven factors: market, size, book-to-market ratio, momentum, systematic liquidity, low versus high beta (a long–short portfolio in low-/high-beta stocks), and earnings quality (long profitable, growing, higher-payout stocks and short the opposite).
The authors find that the last two factors explain the outperformance of high Active Share patient managers of both mutual funds and institutional portfolios.
Research parameters
The study has been conducted for mutual funds on a sample of actively managed all-equity US retail mutual funds from the CRSP survivorship-bias-free mutual fund database.
The authors conduct 5 × 5 independent double sorts into holding duration quintiles and Active Share quintiles. Performances are compared along both dimensions. Mutual fund performance is examined over the 1990–2013 period with respect to net returns and holdings-based gross returns. To notice that the study was conducted by including Institutional portfolios.
Fund duration, fund holdings turnover, and the self-declared turnover ratio are the three proxies the authors use to determine how long funds hold stocks in portfolios.
The authors evaluate net mutual fund performance using a five-factor model (market, size, value, momentum, and liquidity) and an index-based seven-factor model. The authors evaluate holdings-based gross returns while controlling for size, book-to-market value, and momentum characteristics.
Outcome
This study demonstrates that contrary to popular belief, over long periods mutual fund performance may, under certain circumstances, have the potential to outperform the benchmark. However, this research focus solely on US equities and we understand the results may be different, in other markets either developed or emerging. Furthermore, if we consider the US equity market as one of, if not the most efficient market, we guess that the potential for excess returns might stand higher in the other countries. For instance, experienced fund selectors are well aware that it is easier to find active funds with successful long-term track record in the European equity market compare to the US equity market.
In conclusion, this study comfort us in our empirical findings that an active manager investing over the long term with conviction increases his/her chance to succeed. Of course, it is not a sufficient condition, the ones that deliver with the appropriate level of skill set and the right execution timing will truly generate that excess returns over the long run. At WSP we use Active share as well as portfolio turnover as key indicators in our quantitative analysis. We pay special attention to the source and methodology used to supply both measure to make them relevant.
Source: Nicholas Tan, CFA, CFA Digest, April 2017, Volume 47, Issue4
(1) : “Patient Capital Outperformance: The Investment Skill of High Active Share Managers Who Trade Infrequently”, Martijn Cremers and Ankur Pareek, Journal of Financial Economics, Vol. 122, No. 2 (November 2016): 288-306