By Stanislaw Zarzycki and Kenneth C Marshall
It has been a while since the economics guru Burton Malkiel published his seminal study "A Random Walk Down Wall Street" in 1973. Professor Malkiel became the leading proponent of the idea that active managers cannot outperform their index in the long run . Over the past 40 years, the efficient market hypothesis has served to give birth to the whole index/ETF craze and to question the value-add of active fund managers.
Also over the past 40 years, managers got smarter, information flow became more efficient and data more accessible. We've seen the rise of the Internet and instant messaging and vastly more powerful analytic tools. Surely active managers have gotten better at the game and better at spotting inefficiencies.
But wait: does this mean instead that markets are now more efficient because all information, including private information, is incorporated in price? Are active managers, despite having more resources, becoming less effective and relevant?
In this age of instant information, we thought it highly relevant to retest the general hypothesis: do active managers outperform the market (maintaining their relevance) or does the "more-efficient" market hypothesis still prevail? To test this question, we used the Morning Star mutual fund database to focus on U.S. large cap blend managers over the period 1999-2012. This period encompasses 3 economic cycles and the rise of the age of instant information.
The initial task was to refine the database down to large cap managers focused on U.S. equities (> 95% of portfolio invested in U.S. equities, > 90% invested in large cap stocks). This was not a straightforward exercise as managers evidently drifted to foreign stocks or mid cap companies in search of performance. Our filter (U.S. and large cap) narrowed the list of mutual fund managers down from 27,736 to 158 funds whose performance went back to at least 1999.
We then compared the compounded return of these 158 mutual funds managers to the S&P500 total return index. Remarkably, 47 or 29.75% of these managers outperformed the index. One could conclude that, yes, about 1/3 of the active managers appear to defy the efficient market hypothesis. But this is not determinative of anything (contrary to some Wall Street analysts). One needs to take the analysis one step further and ask if there is any predictive power in these past outperformer returns as to future outperformance. That is to say, how many of these 47 consistently outperformed the market?
The common market wisdom is that around 30% of managers outperform their benchmark (according to a Wells Fargo study) which is very consistent with our study; however, the mainstream research fails to address an important question: is the outperformance serially correlated? In order to answer that question we pushed the study deeper.
We chose to slice the data into two periods: 1999-2005 and 2006-2012. We used the compounded return over the earlier period (1999-2005) to select managers that initially outperformed the index to find out if their prior record had much bearing on the future relative performance. Out of the 54 funds (34%) that outperformed over the in-sample 7 year period (1999-2005), only 7 (13%) outperformed the S%P 500 over the following 7 year period 2006-2012. If we then eliminate duplicate funds (funds managed by the same manager) in our sample, only 3.7% of the mutual funds continued outperforming the index in the out-of-sample period (2006-2012). This would indicate that 96.3% of actively managed funds underperformed the S&P 500.
The initial conclusion based on our subset of managers is that a reasonable percent of them can outperform the index over the long run. However, choosing those managers based on past performance (so-called "tail chasing"), is a losing proposition: the odds are overwhelmingly against you. If to choose managers based on their past outperformance we have a 96.3% probability of underperforming in the future. In other words, there is significant mean reversion of returns among active managers consistent with Malkiel's random walk proposition. So if one considers investing in active managers based on historical performance, statistically he is better off putting his money into an index fund like SPY.
1. We recognize that the term "efficient market" is attributable to Eugen Fama "Random Walks In Stock Market Prices", Financial Analysts Journal in 1965.
2. "Active Versus Passive Investing," Robert Whitehead, CFA, 2012