Have researchers unfairly treated actively managed mutual funds by using non-investable benchmarks? In their February 2012 paper entitled “Another Look at the Performance of Actively Managed Equity Mutual Funds”, David Blitz and Joop Huij evaluate the performance of actively managed equity mutual funds against a set of passively managed market, small capitalization, growth and value index funds rather than modeled risk factors. They select four Vanguard funds as passive benchmarks: 500 Index Investor (VFINX); Small Cap Index Investor (NAESX); Value Index Investor (VIVAX); and, Growth Index Investor (VIGRX). At the beginning of every month, they rank active funds into deciles based on 12-month lagged returns and calculate average fund returns by decile over the next month. They then relate the decile series returns to: (1) the commonly used market, size, book-to-market (value) and momentum factors; and, (2) factor-mimicking combinations of the Vanguard passive index funds. Using monthly returns for those of 6,814 living and dead U.S. equity mutual funds which are not obviously index funds from April 1993 (when all four Vanguard index funds become available) through March 2010, they find that:
- Relating the passive index funds to the corresponding risk factors:
- VFINX efficiently captures the overall equity risk (market factor) premium.
- NAESX fails to capture much of the size factor premium.
- In tandem, VIVAX and VIGRX do not capture the value factor premium, since these funds have nearly identical average returns.
- A strategy that each month invests in the one of these four funds with the highest 12-month lagged return fails to capture the momentum factor premium.
- The top 10% of actively managed mutual funds based on past performance beat the bottom 10% of funds by 8.8% per year, with the momentum factor explaining most of the gap. However, the top decile earns a statistically insignificant four-factor annual alpha of only 1.4%.
- Relating actively managed mutual funds to the passive index funds, the top 10% of active funds:
- Beat VFINX by 5.0% per year, with outperformance not attributable to market beta.
- Earn a significant three-factor annual alpha of 3.5% relative to a portfolio of passive index funds that mimic the market, size and book-to-market factors (driven mostly by NAESX exposure). Incorporating a momentum strategy that each month invests in the passive index fund with the highest return over the past 12 months does not substantially affect this alpha.
- Beat an equally weighted, monthly rebalanced portfolio of the four passive index funds by 4.3% per year.
In summary, evidence from benchmarking against real investment alternatives rather than factor models does not support belief that investors are better off buying low-cost index funds instead of actively managed mutual funds with strong past performance.
Cautions regarding findings include:
- As confirmed by one of the authors, active fund returns do not account for any loads (front-end and back-end), which may materially reduce net returns to investors. He pointed out that previous research finds similar performance persistence results for funds with and without sales loads, so restricting the active fund universe to no-load only should produce similar results.
- As confirmed by one of the authors, mutual fund trading restrictions may interfere with any implementation of monthly ranking. He noted that the study is not intended to represent a realistic trading strategy, and that harvesting the potential alpha found remains an open question.
