Can investors assess the performance of an active fund manager without access to the fund’s detailed trading records (especially trades not evident from quarterly holdings reports)? In the February 2009 update of his paper entitled “Active Alpha and Active Beta – Detecting the Unobserved Actions of Portfolio Managers”, Anders Ekholm presents a new methodology for indirectly measuring the effects of a fund manager’s trading that relies exclusively on portfolio returns. His approach decomposes fund tracking error into two aspects of active management: stock selection (idiosyncratic risk, or active alpha) and general market timing (systemic risk, or active beta). Applying this methodology to daily returns for a sample of actively managed U.S. equity mutual funds over the period 12/31/99-3/31/08, he finds that:
- On average, the sampled funds generate a negative four-factor (market, size, book-to-market, momentum) Jensen’s alpha (underperform their passive benchmarks).
- Mutual fund managers engage in both security selection and market timing to greatly varying degrees, with widely varying performance results. For specific managers, the level of emphases on these two aspects of investing is very stable.
- On average, stock selection (market timing) helps (hurts) fund performance. Specifically, portfolio managers with a past high level of security selection (market timing) activity outperform (underperform) in the future.
- Benchmark tracking error obscures the elements of active fund management, because it combines the empirically offsetting effects of stock selection and market timing.
In summary, mutual fund investors are more likely to find outperformance in funds that emphasize stock selection rather than general market timing.
Could poor general market timing in this study be a consequence of investment policies (minimal cash-on-hand) and the (mis)timing of investor purchases of fund shares rather than fund manager market timing skill?
