Jonathan Berk: Are Mutual Fund Managers Skilled, Or Just Lucky?

Image

Peter Lynch, the former manager of the Fidelity Magellan Fund, in 2011 | Reuters photo by Brian Snyder

In 1973, Princeton economist Burton Gordon Malkiel famously pointed out that a blindfolded money manager throwing darts at a newspaper’s financial pages could select a portfolio that would do just as well as one carefully selected by experts.

He was translating into layman’s terms research that suggested that most investors would do better investing in index funds than in actively managed funds. His conclusion was accurate, but new research from Stanford Graduate School of Business explains why no one — especially financial policymakers — should jump to the conclusion that active-fund managers have no superior investment skills. Over time, people have used the truism about index funds to mistakenly conclude that mutual fund managers have no skill, and that it is impossible to ever beat the market.

In fact, research by Jonathan Berk of Stanford Graduate School of Business and Jules H. van Binsbergen, formerly of Stanford and now at Wharton, suggests that the typical mutual fund manager is persistently skilled, and that top performers are especially good. It’s just that the market is so hypercompetitive that most investors can’t benefit from the skill — it is competed away too quickly as money pours into emerging managers’ funds. The managers and their companies, rather than investors, capture the value of the total market earnings and fees charged to investors.

For policy makers, the research suggests that mutual fund managers have been unfairly castigated. If we confuse the questions of how skilled mutual fund managers are with how much individual investors can benefit from their skill, we risk making poor decisions about how to regulate and set policy in finance.

Measuring Investment Skill

Though conventional wisdom holds that mutual fund managers are unskilled, they are some of the most highly compensated members of our society. The researchers began looking into how that could be.

The basic economic principle of rents holds that someone cannot earn a “rent” — a wage above costs, in this case — unless they possess a desired skill in short supply. Though there can be distortions in the market, such as government incentives or penalties that might explain the high incomes, it seemed impossible that mutual fund managers would earn such high wages without possessing any skills at all.

“We then asked ourselves a basic but crucial question: Could we be measuring skill incorrectly?” Berk says.

Many people have used gross alpha — the industry term for returns above a benchmark of diversified stocks, such as the S&P 500, and before fees charged to investors are deducted — as a proxy for investment skill. The researchers offer the example of Peter Lynch to show why looking at skill this way could be a mistake.

In his first five years managing Fidelity’s Magellan Fund, Peter Lynch had a 2% monthly gross alpha on average assets of about $40 million. In his last five years, his gross alpha was only 20 basis points per month, but on assets that ultimately grew to more than $10 billion.

“Based on the lack of persistence in gross alpha, one could mistakenly conclude that most of Peter Lynch’s early performance was due to luck, rather than skill,” Berk and van Binsbergen write.

But the skill is still there, which you can see when you take into account how much money Lynch actually made from the funds he invested. The value he extracted from financial markets went from less than $1 million per month in his first 5 years to over $20 million per month in the last year.

If not gross alpha, then how ought skill to be measured? Berk says it is important to first recognize how a manager makes money. First, she buys low and sells high, to make money for investors. She then charges fees to investors for the returns.

The money made by the manager — and the better representation of her skill — is the return she earns over her benchmark plus the fees that investors are willing to pay her, says Berk. As with the Lynch example, the measure needs to take into consideration the percentage fee charged and the size of the fund upon which the percentage is charged.

Funds Analyzed

Berk and van Binsbergen looked at a universe of 5,974 mutual funds from 1969 to 2011 and compared their results to comparable Vanguard index funds, which are alternative products that investors can actually buy. When they divided the mutual funds into 10 groups based on the amount of money managers have made in the past, as described above, the researchers found that the funds that made the most in the past also made the most in the future. That is, the ability to make money is persistent. The researchers also calculate that the average fund manager added $2 million in value each year.

Who Benefits?

If the higher earnings are persistent, why can’t individual investors benefit more from them?

The market responds very quickly when a new manager with skill emerges, rewarding her with more assets to invest. As the fund grows, it is harder for the manager to make money for a variety of reasons: For instance, placing trades in large enough quantities for all the investors becomes more difficult.

Over time, returns are lower on the larger funds. But the amount of money the skilled managers earn remains high, based on this analysis.

Does it matter that mutual fund managers are skilled if investors don’t benefit from the skill? Consider other professions. The Army, for instance, would not rate a doctor only on her rate of cures without regard to the difficulty of her cases, the number of people she is required to see every hour, and whether she is operating in a war zone. Conflating skill with results might lead to poor policy decisions. If mutual fund managers have no skill, then it follows that their high pay could be the result only of marketing — or worse, chicanery.

This research found the opposite story: Mutual fund managers walk an ever-narrower ledge in a highly competitive industry.

The research revealed another intriguing result: The mutual fund manager’s current compensation from aggregate fees and the value he or she added to the fund predicted the fund’s future returns even better than past value added. That suggests that investors pick up on tiny signals in the market to evaluate the potential for managers to outperform in the future. It’s possible that neither investors nor mutual fund managers have been as foolish as they have been portrayed.

Jonathan Berk is A.P. Giannini Professor of Finance at Stanford Graduate School of Business. Jules H. van Binsbergen, formerly of Stanford GSB, is now an associate professor of finance at Wharton, University of Pennsylvania.