Only a handful of winners in the long run

Every year the top-performing investment funds are crowned for their recent successes. What do these laurels actually mean for the ordinary investor? Should you invest or switch your investments to these star performing funds? The latest research indicates that this is not necessarily such a great idea.

Mark Hilbert wrote the following article, The Prescient are Few, appearing in The New York Times on July 13, 2008:

How many mutual fund managers can consistently pick stocks that outperform the broad stock market averages — as opposed to just being lucky now and then?

Countless studies have addressed this question, and have concluded that very few managers have the ability to beat the market over the long term. Nevertheless, researchers have been unable to agree on how small that minority really is, and on whether it makes sense for investors to try to beat the market by buying shares of actively managed mutual funds.

A new study builds on this research by applying a sensitive statistical test borrowed from outside the investment world. It comes to a rather sad conclusion: There was once a small number of fund managers with genuine market-beating abilities, as judged by having past performance so good that their records could not be attributed to luck alone. But virtually none remain today. Index funds are the only rational alternative for almost all mutual fund investors, according to the study’s findings.

The study, “False Discoveries in Mutual Fund Performance: Measuring Luck in Estimating Alphas,” has been circulating for over a year in academic circles. The statistical test featured in the study is known as the “False Discovery Rate,” and is used in fields as diverse as computational biology and astronomy. In effect, the method is designed to simultaneously avoid false positives and false negatives — in other words, conclusions that something is statistically significant when it is entirely random, and the reverse.

The researchers applied the method to a database of actively managed domestic equity mutual funds from the beginning of 1975 through 2006. To ensure that their results were not biased by excluding funds that have gone out of business over the years, they included both active and defunct funds. They excluded any fund with less than five years of performance history. All told, the database contained almost 2,100 funds.

The researchers found a marked decline over the last two decades in the number of fund managers able to pass the False Discovery Rate test. If they had focused only on managers running funds in 1990 and their records through that year, for example, the researchers would have concluded that 14.4 percent of managers had genuine stock-picking ability. But when analyzing their entire fund sample, with records through 2006, this proportion was just 0.6 percent — statistically indistinguishable from zero, according to the researchers.

This does not mean that no mutual funds have beaten the market in recent years, Professor Russ Wermers said. Some have done so repeatedly over periods as short as a year or two. But, he added, “the number of funds that have beaten the market over their entire histories is so small that the False Discovery Rate test can’t eliminate the possibility that the few that did were merely false positives” — just lucky, in other words.

Professor Wermers says he was surprised by how rare stock-picking skill has become. He had “generally been positive about the existence of fund manager ability,” he said, but these new results have been a “real shocker.”

Why the decline? Professor Wermers says he and his co-authors suspect various causes. One is high fees and expenses. The researchers’ tests found that, on a pre-expense basis, 9.6 percent of mutual fund managers in 2006 showed genuine market-beating ability — far higher than the 0.6 percent after expenses were taken into account. This suggests that one in 10 managers may still have market-beating ability. It’s just that they can’t come out ahead after all their funds’ fees and expenses are paid.

Another possible factor is that many skilled managers have gone to the hedge fund world. Yet a third potential reason is that the market has become more efficient, so it’s harder to identify undervalued or overvalued stocks. Whatever the causes, the investment implications of the study are the same: buy and hold an index fund benchmarked to the broad stock market.

Professor Wermers says his advice has evolved significantly as a result of this study. Until now, he says, he wouldn’t have tried to discourage a sophisticated investor from trying to pick a mutual fund that would outperform the market. Now, he says, “it seems almost hopeless.”