Mutual fund advertisements are far too effective. Fund companies often promote actively managed funds that have generated high returns, and investors flock to such funds. Unfortunately for these investors, there is little relationship between high past returns and high future returns.
Why doesn't strong past performance continue? The primary reason is that luck is a major factor in fund returns, and luck generally does not persist. Investors tend to overlook the role of luck in fund returns. There are thousands of actively managed equity funds, so even if
2010 Investment Guide
The Average Investor Is His Own Worst Enemy
SEC Issues Worthless Warnings On Fund Performance Ads
all fund managers were randomly picking their portfolios by throwing darts at a stock page, a large number of funds would still soundly beat market averages.
In a new study, finance professors Eugene Fama of the University of Chicago Booth School of Business and Kenneth French at Dartmouth's Tuck School of Business quantify the role of luck in fund returns. They find that the strong returns of actively managed funds are almost always due to luck, not the stock-picking skill of fund managers. The study will be published in the Journal of Finance.
Fama and French examine the returns from 1984-2006 of actively managed funds that invest primarily in U.S. common stocks. They compare funds' actual returns to the returns of comparable standard passive benchmarks, such as Fama and French's so-called three-factor and Carhart's four-factor asset pricing models.
Fama and French first examined the returns of funds before management fees but after commissions and other trading costs. They found that these returns for actively managed funds, as a whole, are about the same as those of their passive benchmarks. In other words, overall, active fund managers have little stock-picking skill. (For related reading, check out Your Mutual Fund: It's Riskier Than You Think.)
Think about that for a minute: Investors pay well over $10 billion annually in fees to managers of actively managed funds who, as a group, have only enough skill to cover their trading costs. Because funds have other costs, such as management fees, which are included in fund expense ratios, active funds' returns actually trail their passive benchmarks by approximately the level of the funds' expense ratios (around one percentage point per year).
The fact that active managers as a whole have little stock-picking skill doesn't mean that every manager has little skill. In fact, there might be many great managers whose stock-picking skills are being offset by poor stock pickers.
To estimate the frequency of skilled managers, Fama and French conducted 10,000 simulations of the effect of luck on fund returns. They used past fund returns to simulate the distribution of the returns of funds in a world in which all managers have just enough skill to cover all their costs, including management fees.
Because of luck, individual funds' simulated returns often differed greatly from the expected returns of the funds' benchmarks. Many funds had good luck (and thus exceeded their benchmarks), and many funds had bad luck (and thus fell short of their benchmarks). The results of these simulations indicate how much variation across fund returns is likely to occur due to luck alone.
They next compared the distribution of actual fund returns between 1984 and 2006 with the results of the 10,000 simulations. If many fund managers have more than enough stock-picking skill to cover all their costs, then there should have been many more strong performers than occurred in the luck-based simulations. This is because such skilled managers should have been more likely to generate high returns than did managers in the simulations, all of whom had only just enough skill to cover their costs. (Learn about some of the very best fund managers in The Top 5 All-Time Best Mutual Fund Managers.)
The results are disheartening. Very few fund managers had sufficient stock-picking skill to cover their costs. For example, using a standard measure of expected returns as the benchmark (the three-factor asset pricing model), only the actual best-performing 2% of funds, as a group, outperformed the simulations. In other words, as a group, the top 2% of funds had higher returns than the top 2% of the simulated fund returns. Funds did even worse when the expected returns of the four-factor asset pricing model were used as the benchmark.
These findings indicate that only a very small percentage of fund managers have more than enough skill to cover their costs. The vast majority of strong-performing managers are lucky rather than sufficiently skilled.
Furthermore, even in this small number of managers who, as a group, have more than enough skill to cover their costs, the amount of skill was unimpressive. Fama and French found that these managers are unlikely to noticeably outperform a large, efficiently managed index fund in the future. (For more on this type of investment, read The Lowdown On Index Funds.)
The case against active management is even stronger when one realizes that Fama and French's analysis doesn't take into account all the costs of investing in actively-managed funds. Because active managers generally trade more often than index fund managers, investors who own actively-managed funds outside of tax-advantaged accounts (such as 401(k)s and Individual Retirement Accounts) typically must pay higher capital gains taxes than do investors in index funds.
Fama and French's study illustrates the folly of chasing funds with high past returns. Investors do not realize that they are generally chasing luck, not skill. Thus, it should be no surprise that successful actively managed funds generally don't continue their strong performances. So, the next time you see an advertisement touting a fund's market-beating returns, remember that the fund's manager was probably just lucky.