IN SELECTING MUTUAL FUNDS, too many fund investors seem to rely less on sustained performance over the very long term (with all of its own profound weaknesses) than on superior performance over the short term. In 2016, over 150 percent of net investor cash flow went to funds rated four or five stars by Morningstar, the statistical service most broadly used by investors in evaluating fund returns.

These “star ratings” are based on a composite of a fund’s record over the previous three-, five-, and 10-year periods. (For younger funds, the ratings may cover as few as three years.) As a result, the previous two years’ performance alone accounts for 35 percent of the rating of a fund with a 10-year history and 65 percent for a fund in business from three to five years, a heavy bias in favor of recent short-term returns.

How successful are fund choices based on the number of stars awarded for such short-term achievements? Not very! According to a 2014 study by the Wall Street Journal, only 14 percent of five-star funds in 2004 still held that rating a decade later. Approximately 36 percent of those original five-star funds dropped one star, and the remaining 50 percent dropped to three or fewer stars. Yes, fund performance reverts toward the mean, or even below.

Reversion to the mean (RTM) is reaffirmed in comprehensive fund industry data.

Other data on fund returns confirm the power of RTM. Consider Exhibit 11.1, comparing the returns of all actively managed U.S. equity funds over two consecutive sets of nonoverlapping five-year periods: 2006–2011 and 2011–2016.

Note: Total number of funds merged or liquidated: 313.

We sorted the returns for each period into quintiles—the top quintile included funds with the best performance, and the bottom quintile contained those with the worst performance. We then looked at how those initial funds fared in the subsequent five-year period.

If it were easy to select funds that would outperform their peers by simply buying yesterday’s winners, we would expect to see persistence; that is, most funds that ended the first period at the top of the heap would remain there in the next period and those at the bottom would remain there. But no. As it turns out, RTM overpowers persistence.

Consider the funds that ranked in the top quintile during the first period (2006–2011). Over the subsequent five years, only 13 percent remained in the top quintile. A remarkable 27 percent of the winners from the first period ended up in the bottom quintile, and another 25 percent landed in the next-to-last (fourth) quintile. Even worse, 10 percent of the previous winners didn’t even survive the next five years.

At the other end of the spectrum, 17 percent of the first-period laggards ended up at the top of the heap in the subsequent period—even better than the first-period winners! And only 12 percent of the losers repeated their dismal performance in the second period, while 26 percent didn’t survive.

You need not be a statistical wizard to observe the remarkable randomness of returns through each of the quintiles, with steady RTM centering around 16 percent in each quintile—less than the 20 percent we started with in the first period. This lower number is because fully 18 percent of the funds from the first period went out of business before the second period ended, presumably due to poor performance.

A second study reaffirms the first study—with incredible precision.

You might be wondering if this pattern was just a one-time event, not likely to be repeated. I had the same question. So we looked at the preceding nonoverlapping five-year period, 2001–2006, and 2006–2011. The pattern held (Exhibit 11.2). Of the top-quintile winners from 2001 to 2006, only 15 percent remained in the top quintile, while 20 percent fell to the bottom. Even worse, 13 percent of the funds—45 funds—failed to survive.

Note: Total number of funds merged or liquidated: 454.

Among the bottom-quintile laggards from 2001 to 2006, 18 percent ended the subsequent period in the top quintile—once again, even better than the first period’s winners, only 15 percent of which maintained their position at the top. Only 6 percent of the lowest-ranking funds repeated their dismal performance. 152 of the bottom-quintile funds (43 percent) did not survive.

Just glance over the data in these two exhibits and you will see the recurring pattern of RTM. Like the results in Exhibit 11.1, the second-period results are essentially random. The vast majority of the funds in all five quintiles earned subsequent returns that were largely spread relatively equally over each performance quintile (between 13 percent and 18 percent in each).

From these data, we can conclude that RTM exerts a powerful force on mutual fund returns. There is remarkably little persistence in returns among the top and bottom funds alike. I don’t amaze easily. But these data are truly amazing. They dramatically belie the assumption of most investors and advisers that manager skill will persist. Most investors seem to believe that manager skill will persist. But it doesn’t. We are “fooled by randomness.”¹

The stars produced in the mutual fund field rarely remain stars; all too often they become meteors.

The message is clear: reversion to the mean (RTM)—the tendency of funds whose records substantially exceed industry norms to return toward the average or below—is alive and well in the mutual fund industry. In stock market blow-offs, “the first shall be last.” But in more typical environments, reversion to the fund mean is the rule. So please remember that the stars produced in the mutual fund field are rarely stars; all too often they are meteors, lighting up the firmament for a brief moment in time and then flaming out, their ashes floating gently to earth.

With each passing year, the reality is increasingly clear: relative returns of mutual funds are random. Yes, there are rare cases where skill seems to be involved, but it would require decades to determine how much of a fund’s success can be attributed to luck, and how much attributed to skill.

If you disagree and decide to invest in a fund with superior recent performance, you might ask yourself questions like these: (1) How long will the fund manager, with the same staff and with the same strategy, remain on the job? (2) If the fund’s assets grow many times larger, will the same results that were achieved when the fund was small be sustained when it is large? (3) To what extent did high expense ratios and/or high portfolio turnover detract from the fund’s performance, or did low expenses and low turnover enhance performance? (4) Will the stock market continue to favor the same kinds of stocks that have been at the heart of the manager’s style?

Picking winning funds based on past performance is hazardous duty.

In short, selecting mutual funds on the basis of recent performance is all too likely to be hazardous duty, and it is almost always destined to produce returns that fall far short of those achieved by the stock market, itself so easily achievable through an index fund.

It might help our understanding if we each ask ourselves just why it is so hard to recognize the powerful principle of reversion to the mean that punctuates not only mutual funds’ returns, but almost every corner of our lives. In his 2013 book Thinking, Fast and Slow, here’s how Nobel laureate Daniel Kahneman answered that question.

[O]ur mind is strongly biased toward causal explanations and does not deal well with “mere statistics.” When our attention is called to an event, associative memory will look for its cause . . . but they [causal explanations] will be wrong because the truth is that regression to the mean has an explanation but does not have a cause.