Seeking Alpha
Value, momentum, portfolio strategy
Profile| Send Message|
( followers)  

In honor of yesterday's announcement that Eugene Fama will share this year's Nobel Prize for Economics with Lars Peter Hansen and Robert Shiller, I thought it would be appropriate to share some of Fama's previous work on the nature of investment returns to the Seeking Alpha community.

In a 2009 paper entitled "Luck Versus Skill in the Cross Section of Mutual Fund Returns," the University of Chicago's Eugene Fama and Dartmouth College's Kenneth French illustrated that taking the aggregate of actively managed mutual funds produces an alpha, or abnormal return, of just 0.13% per year. This finding demonstrates that even before sometimes hefty expenses, active managers produce only de minimis excess returns; after expenses, investors, on average, simply transfer their wealth to the managers in the form of management fees. After expenses, alpha was on average negative 0.81%. The aggregate portfolio of these managers was highly correlated to the capitalization-weighted market as a whole. A simpler way to get a broad exposure to a cap-weighted index would be just to purchase SPY. The 9.5bp expense ratio on this passively managed index seems a lot cheaper than the 81 bps paid to managers. For readers "seeking alpha," excess returns will prove elusive on average for investors purchasing actively managed mutual funds.

One could posit that there must be managers whose keen insight and alpha-generative processes allows them to outperform over a long sample period, and that the aggregate market performance is simply skewed downwards by underperforming managers. Fama and French tested this hypothesis by running simulations of fund returns and comparing the results versus the actual dispersion of mutual fund returns over a period ranging from 1984-2006. By comparing actual versus simulated results, the authors could determine if there was a presence of very good managers (or very bad managers) outside the simulated dispersion of results. If alpha has been essentially zero on average, potentially the average alpha could have been skewed downwards by a slew of bad managers in discontinued funds, drifting true alpha higher absent these underperformers than its insignificant historical average. This comparison of actual versus simulated results also lets us know if there are truly great managers, who are producing alpha over time, or if the fat tails of investment returns simply lead some managers to outperform as often as expected in the simulation due to the fat tails of investment returns in general.

If you have a bulk of your investment wealth in actively managed mutual funds, the results are discouraging. Upwards of 90% of the funds produced returns lower than what was expected via the simulation. For these funds, true alpha is likely negative, and the managers are not producing enough relative performance to offset their expenses. The top three percentile of performers fared better, but did not still did not populate the tail in great enough numbers to far surpass the numbers of funds which should be in this higher returning strata simply by chance.

If I asked the Seeking Alpha community to name one manager that had managed to outperform the broad market over long time intervals, the most oft cited name would be Warren Buffett. After all, the Oracle of Omaha turned a failed textiles manufacturer in Berkshire Hathaway (BRK.A, BRK.B) into a powerhouse conglomerate that employs 288,000 people. As I demonstrated in "Making Buffett's Alpha Your Own," the stunning performance of Berkshire Hathaway can be boiled down into two factors - the Low Volatility Anomaly and leverage - and not through long-run stock picking.

Fama and French, the aforementioned article's authors, are of course the luminaries behind modern finance's Fama-French Three Factor Model that observes that small-cap stocks tend to outperform large-cap stocks and low market to book stocks tend to outperform high market to book stocks. Adding these observations to the Capital Asset Pricing Model better describes stock market performance than beta alone. Regressing the aggregate mutual fund industry's returns in the sample period showed that the industry as a whole had a market-like exposure, but little exposure to the size and value elements in Fama and French's model.

I am not a die-hard proponent of the Efficient Market Hypothesis that Fama has championed. In many past articles, I have illustrated that alpha can be generated through momentum, a violation of the Efficient Market Hypothesis. While there are many ways to generate excess returns, including the size and value premia, Fama has proven one sure way to generate negative alpha on average - paying a manager high fees to try and beat the market over time.

Source: A Nobel Prize Winner On Alpha