Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
This article originally appeared on ETF.COM here.
There is a considerable body of evidence demonstrating that the post-hire performance of active managers tends to be disappointing relative to their pre-hire performance. Specifically, managers’ performance tends to regress toward zero after adjusting for expenses, risk, exposure to common factors and survivorship bias. For example, the research has found that:
- While 20 years ago, about 20% of actively managed funds were generating statistically significant alpha, that figure is now about 2% as the market has become more efficient, the competition has gotten tougher and, as described in my book, “The Incredible Shrinking Alpha,” academics have been converting what was once alpha (a source of outperformance) into beta (a common factor that can easily be replicated).
- There is little evidence of performance persistence among hedge funds and actively managed mutual funds.
Problems with performance chasing
Despite the evidence, strong past performance is the prerequisite for manager selection by individuals as well as institutional investors. Robert Ferguson, Anna Agapova, Dean Leistikow and Joel Rentzler contribute to the literature on the persistence of performance among active managers with the paper, “Chasing Performance and Identifying Talented Investment Managers,” which appears in the Spring 2018 issue of The Journal of Investing. Their focus is on explaining why investors are likely to get poor results from performance-chasing.
The authors begin by noting that while choosing managers on the basis of historical performance might make intuitive sense, “it takes an impractically long time to differentiate talented from untalented managers – far longer than the five years or so that many investors believe is sufficient.”
They provide a mathematical example involving a world that contains 20 untalented managers (who have a normal distribution of returns with a mean alpha of 0 and a standard deviation of 2.68%) and one talented manager (who generates 3% alpha, which is very large, with a 3% tracking error).
Ferguson, Agapova, Leistikow and Rentzler demonstrate that we should expect, randomly (purely by luck), one of the 20 untalented managers will generate an alpha about 50% greater than 3% over the next 20 years. In fact, they write, “the probability that the talented manager beats all 20 untalented managers over the five-year period is only about 14.8%.”
Thus, performance-chasers end up with the untalented manager 85.2% of the time. And that is using a five-year horizon. Many institutions use three-year horizons, in which case the odds of hiring the talented manager would be even worse. Even at 10 years, the odds are just 36% that the talented manager will outperform all the untalented ones.
Making matters worse is that the probability the talented manager will beat all 20 untalented managers is surprisingly small for all lengths of measurement period. Even after 15 years, it only rises to about 55%. To achieve a 95% (99%) probability that the talented manager will outperform, you need a 38-year (54-year) historical performance record.
Such lengthy records are rare. Even if they existed, the likelihood is that the manager would have received so much cash flow that the hurdles to generating alpha would have become virtually insurmountable. That’s the curse of success – it sows the seeds of its own destruction.
Summary
The authors illustrated how the probability that a talented manager’s performance will exceed the best in a pool of untalented managers can be surprisingly small, even over long periods. It helps explain why performance-chasing has produced such poor results for individual and institutional investors alike. A better use of historical performance relegates it to a secondary role, the primary focus being an a priori analysis of a manager’s investment process (security selection and trading strategy).
Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.
More Smart Beta Topics >
