Do investors view (mechanical) smart beta returns from mutual funds as (skillful) alpha? In the April 2017 update of their paper entitled “Fake Alpha”, Marcel Müller, Tobias Rosenberger and Marliese Uhrig-Homburg investigate the conflation of smart beta (“fake alpha”) and true alpha (incremental to smart beta and generated by skill) by mutual fund managers and investors. In estimating smart beta returns, they consider size, value and momentum factors. Using monthly returns for 3,292 actively managed mutual funds focused on U.S. stocks and contemporaneous market, size, book-to-market and momentum factor returns during March 1993 to December 2014, they find that:
- Overall, mutual funds underperform the market by about the management fees.
- Net alpha, gross alpha, smart beta (“fake alpha”) and operating costs all tend to decrease as fund size increases. In other words, investment opportunities deplete faster than the efficiencies of scale grow.
- Mutual fund investors focus on market outperformance (apparently viewing smart beta and true alpha as equivalent), and mutual fund managers therefore accommodate by pursuing them as equivalent. Consequentially, the mutual fund industry allocates roughly 23% too much to active management.
- Over the sample period, there is negligible learning by mutual fund investors regarding the difference between smart beta and true alpha.
In summary, evidence indicates that mutual fund investors accept smart beta as alpha, and mutual fund managers respond accordingly.
Cautions regarding findings include:
- Returns calculated frictionlessly from factor models of stock returns are not achievable in practice, thereby overstating “fake alpha.”
- Extracting smart beta returns reliably (from large samples of stocks) is infeasible for many mutual fund investors, who may therefore view such returns as value-added (alpha).
- As described in “Effects of Smart Beta ETFs on Mutual Funds”, recent introduction of smart beta exchange-traded funds (ETF) may be changing the mutual fund industry. Such ETFs offer insight into the magnitudes of net smart betas.
