The financial press media are forever seeking to acclaim some mutual fund manager as the financial equivalent of the Michael Jordan of investment managers -- the "next Peter Lynch." They, along with individual investors, perform intensive searches of databases, seeking to find a fund manager with a persistent long-term record of outperformance -- while short-term outperformance might be a matter of luck, long-term outperformance must be evidence of skill. The latest to be acclaimed as the next Peter Lynch is Bill Miller, the manager of the ( LMVTX) Legg Mason Value Trust . He has managed to do what no other current manager has done -- beat the S&P 500 index 12 years in a row. Surely that cannot be luck. Certainly you can rely on that performance as a predictor of future greatness. Before you come to that conclusion, you should consider the following historical evidence. In each year from 1974 through 1984, the Lindner Large-Cap Fund outperformed the S&P 500 index. How were investors rewarded if they waited 11 years to be sure that they had found a true genius and then invested in the fund? Over the next 18 years, the S&P 500 index returned 12.6%. Believers in past performance as a prologue to future performance were rewarded by their faith in the Lindner Large-Cap Fund with returns of just 4.1%, an underperformance of over 8% per annum for 18 years. After outperforming for 11 years in a row, the Lindner Large-Cap Fund managed to beat the S&P 500 in just four of the next 18 years, and in none of the last nine -- quite a price to pay for belief in a discredited theory. Not yet convinced? Consider the case of David Baker and the 44 Wall Street Fund. Baker even outperformed the legendary Magellan Fund over the entire decade of the 1970s and was the top-performing diversified U.S. stock fund of the decade. Unfortunately, 44 Wall Street ranked as the single-worst-performing fund of the 1980s, losing 73%. During the same period, the S&P 500 grew at 17.5% per annum. Each dollar invested in Baker's fund fell in value to just 27 cents. On the other hand, each dollar invested in the S&P 500 would have grown to just over $5. The fund did so poorly that in 1993 it was merged into the 44 Wall Street Equity Fund, which was then merged into the Matterhorn Growth Fund Income in 1996.
The Demon of ChanceA basic knowledge of statistics provides us with the knowledge that with thousands of money managers playing the game, the odds are that a few, not just one, would have turned in a Bill Miller-like performance. Without this knowledge, investors are likely to confuse skill with "lady luck." If the examples of the Lindner Large-Cap Fund and 44 Wall Street have not yet convinced you, I hope that even the most skeptical among you will be convinced by the following example. Disappointed with the performance of its active manager, a multibillion-dollar pension plan decides to fire the manager and initiates a search for a replacement. It performs a thorough screen of potential candidates. Among the screens are a record of superior performance over the prior 15-year period ending in 2002, a high persistency of superior performance, tenure of the manager and turnover. The due diligence process has narrowed the final candidates to the following funds and a benchmark, the S&P 500 index.
|A Mutual Fund to Drool Over |
|Fund||Annualized Returns 1988-2002*|
|Larry Swedroe Investment Trust||14.3%|
|Legg Mason Value||14.1|
|S&P 500 Index Fund||11.5|
On the basis of its track record the winner of the performance derby is the Larry Swedroe Investment Trust. Not only has the fund outperformed its benchmark by a significant amount, but it has also done so with a high degree of persistency, outperforming the index in nine of the 15 years (67%). In addition, the fund has had the same manager in charge for the entire period, and turnover has been extremely low. After being presented with the data, the investment committee votes to award the management of the plan to the Larry Swedroe Investment Trust. At the last moment, one member of the committee suggests that as one final bit of due diligence Larry Swedroe should be brought in to explain his winning strategy. Appearing before the committee, I am first congratulated on the superior results of the fund. I am then asked to explain my investment strategy. I respond by stating that since my wife's name is Mona, my lucky letter is M. I therefore construct a value-weighted portfolio of all U.S. stocks that begin with the letter M and rebalance the portfolio annually. Skill or the demon of luck?
The above example was created by the technique known as data mining -- a technique for building predictive models of the real world by discerning patterns in masses of data. The computer was asked to find a "strategy" that delivered outperformance, and it mined the data and found one. Before ever concluding that because a strategy worked in the past it can be relied to work in the future, we need to ask if there is a rational explanation for the correlation between the outcome and strategy. Obviously, in the case of the Larry Swedroe Investment Trust there is no rational explanation, which is why no one would engage my fund as its manager -- at least no rational person would. Unfortunately, in the real world, many investment products are based on ideas that have much in common (they are the result of data mining) with the strategy of the Larry Swedroe Investment Trust. As evidenced by the examples of the Linder Large-Cap Fund and 44 Wall Street, belief in the "hot hand" and past performance as a predictor of the future performance of actively managed funds and their managers, even with 15 years of evidence, can be quite expensive. To paraphrase a famous quotation: Those who do not know their financial history are doomed to repeat it. Unfortunately, both the press and the public are quick to assume that superior performance is a result of skill rather than the more likely assumption that it was a random outcome. In other words, while there will be likely be future Peter Lynches and future Bill Millers, we have no way to identify them ahead of time. Also, unfortunately, we can only buy their future performance, not their past performance. Investors would be well served to read
Fooled by Randomness , an excellent book by Nassim Nicholas Taleb. P.S. A special thanks to Weston Wellington of Dimensional Fund Advisors for dredging up the hypothetical fund example.