The image on Martijn Cremers' computer looks like a psychedelic mountain range. One band of brightly colored peaks rises above another. On the horizon, the sky is a tomato red.
What appears to be a backdrop from the Beatles’ animated film Yellow Submarine is actually a chart with a point to make about increasingly risk-averse mutual-fund managers.
The two bands at the top of the chart—the red “sky” and the tallest line of “mountains,” colored in a medium blue—represent the percentage of large-cap mutual-fund assets that are invested in funds whose holdings are 80-100 percent different from the fund’s benchmark, such as the S&P 500 or Russell 1000.
The left side of the chart represents the situation 30 years ago. Back then, such adventurously managed funds held almost half of all assets. But by 2011 (the extreme right side of the chart), the situation had changed dramatically. Only about 10 percent of total mutual-fund holdings were in funds whose portfolios differed by 80 percent or more from their benchmark indices.
The story here is the creeping ascension of what Cremers, a finance professor and noted researcher whose studies have been published in top academic journals, and other experts call “closet indexing.” The term refers to mutual and other stock funds whose managers are supposed to be buying and selling stocks based purely on the profit opportunities they uncover but whose portfolios tell a different story.
Cremers and a former colleague invented the measurement plotted in the psychedelic chart. It’s called Active Share, and it simply shows the percentage of a fund’s weight-adjusted portfolio that differs from its benchmark index. The measure has attracted plenty of attention lately—including the cover story in the Jan. 14, 2013, Barron’s magazine—because of what its results imply.
Investors in some managed funds aren’t getting their money’s worth.
Managed funds typically charge 1 to 2 percent in fees, which cover management costs, including a salary for a cracker-jack stock picker. The chief competition for managed funds is automated types of funds that track the performance of an index, such as the S&P 500. Indexed funds, which now account for about 25 percent of the mutual-fund market, typically charge much lower fees, about one-tenth to one-third of 1 percent, Cremers says. Indexed funds don’t claim to be able to outsmart other traders or beat a benchmark. They simply replicate what it would be like to “own” the index.
What Cremers’ Active Share research shows, however, is that 30-40 percent of total investments in mutual funds are riding along in funds that claim to be actively managed, but are doing little more than tracking their benchmark. Like a low-cost indexed fund.
“I would argue that funds that have a lower Active Share should generally be cheaper,” said Cremers, “but in many cases they aren’t.”
Active share has come to be regarded as an important barometer in investment circles. It’s already been adopted by the well-known fee-based investor information service Morningstar Direct. Barron’s reported that Lipper Analytical Services is planning on incorporating the measure in its ratings of funds, too.
As Michael Herbst, direct of active-funds research for Morningstar, told Barron’s: “If nothing else, Active Share pulls back the kimono on managers who charge high fees for piggybacking the index.”
Why are more and more fund managers apparently piggybacking? No one knows for certain. Cremers believes it’s a conscious strategy.
A closet indexer will never dazzle Wall Street with eye-popping returns, but neither is a fund manager likely to be fired for consistently performing only a little above or below the benchmark. Likewise, returns that mirror a benchmark may be seen as boring, but they’re unlikely to trigger a painful mass exodus of investors. A fund manager’s compensation is typically based on fee income. The more money the fund has under management, the more the manager makes.
Cremers developed Active Share with colleague Antti Petajisto when they were both junior faculty in the International Center for Finance at Yale University’s School of Management. Their 2009 article “How Active is Your Fund Manager: A New Measure that Predicts Performance” (Review of Financial Studies) has been downloaded more than 14,000 times from the online Social Science Research Network. That makes it the 70th-most-downloaded article out of more than 382,000 titles.
Petajisto has since moved on to become a vice president of active investment strategies at the global investment firm BlackRock, Inc. Cremers joined Notre Dame last summer as a full professor of finance with a concurrent appointment in the Notre Dame Law School.
In terms of his academic career, Cremers has come a long way in a short time. Since earning his Ph.D. in finance from New York University’s Stern School of Business in 2002, he’s published 18 journal articles with another 11 in the works.
He’s also come a long way geographically. The youthful scholar grew up in the southern Netherlands village of Boxmeer, the middle of three children. It’s no great surprise that he speaks perfect English. His father was a high school English teacher, his mother a nurse who worked in a nursing home. He learned English not just from his father and school, but from watching American TV shows such as Cheers. In the Netherlands, English-language programs are not dubbed but subtitled, so Dutch children hear a lot of English growing up, he said.
At college in Amsterdam, he majored in econometrics, which he describes as “80 percent statistics and 20 percent economics.” Cremers met his future wife, Liesbeth, an anthropology major a year ahead of him, during his second year. The couple has since grown into a large family with five children, ages 9, 7, 5, 3 and 1. Cremers says Notre Dame’s Catholic character was a major selling point in his decision to join the faculty, along with the rare opportunity to be hired in as a full professor. At Mendoza, he teaches two MBA courses on fixed-income securities plus an undergraduate course on corporate governance and Catholic social teaching.
His study and research leading up to his doctorate and first job post-graduation at Yale focused on methodology. Then his priorities changed. “I slowly learned to think first about economically interesting questions and I put methodology second,” he says.
Closet indexing was one of those economically interesting questions. The phenomenon had been talked about on Wall Street for years, but Cremers and Petajisto weren’t satisfied with results from the conventional tool of detection, tracking-error volatility. Whereas Active Share looks at holdings, tracking error compares a fund’s returns to that of the fund’s stated benchmark.
One problem with tracking error, Cremers says, is that it relies on past data. Another is that if the entire stock market moves in a particular direction, tracking error can cast suspicion of indexing upon a genuinely active fund manager. The best closet-indexing detector is a combination of tracking error and Active Share, Cremers says.
Incidentally, in Cremers’ view, only funds with an Active Share of 80 percent or higher should be considered truly actively managed. Any fund at 60 percent or below (that does not self-identify as an indexed fund) is a closet indexer, he believes. Funds with an Active Share between 60 and 80 percent he considers moderately active.
Cremers’ pursuit of big-picture questions has led him to explore issues of corporate governance. One of his most recent papers, co-authored with Allen Ferrell, a professor of securities law at Harvard Law School, looks back at the wave of hostile takeovers and leveraged buyouts in the 1980s and the invention of the so-called poison pill. The paper is forthcoming in the Journal of Finance.
The “pill” is a defensive tactic devised by boards that prevents hostile-takeover bidders from negotiating share purchases directly with shareholders. They have to go through the board instead. Cremers and Ferrell compared firm-valuation effects of shareholder rights before and after the poison pill was judicially approved in 1985. They concluded that shareholder rights became more important afterward, because weakening the power of hostile takeovers made the board more central.
Cremers’ poison-pill paper followed another successful research collaboration with a Harvard Law scholar, Lucian A. Bebchuck. Like his Active Share work, this investigation required inventing a simple measure to analyze a familiar and complicated issue. In this case, it was executive compensation.
Together with Urs C. Peyer of Europe’s INSEAD graduate school of business, the scholars looked at whether it’s healthy for a company to pay its CEO a lot more than its other top executives.
In “The CEO Pay Slice” (Journal of Financial Economics, May 2011), the researchers looked at records from more than 2,000 companies and added up the total compensation of a company’s top five executives. They then divided the pay of the CEO alone by the total for the top five. If all five executives were paid the same—which almost never happens—each one’s “pay slice” would be 20 percent. They found that the typical CEO pay slice was 30-40 percent.
They next compared CEO pay slice to firm performance. They found that firms with a high CEO pay slice typically had lower value and lower profits. Such firms were also more likely to make large acquisitions to which the stock market reacted negatively. And they were more likely to have granted scandalous backdated stock options to executives.
The bottom line, says Cremers: “Firms seem to do better when then have a good overall top five team or a board that appreciates all five [top executives] rather than just the CEO.”