The Dirty Secret of Fund Returns

It's been tough to beat the Clipper fund (CFIMX) over the last ten years. Its 12% average annual return, three percentage points better than the Standard & Poor's 500 Index, puts it in the top 6% of its peer group, according to Morningstar. Not bad.

But not every investor enjoyed such a nice ride. After Clipper was ranked the best large-cap value fund in 2002, investors snapped up its shares. The fund quickly tripled in size to $7 billion. Then Clipper's managers stumbled. Bets on stocks like Tenet Healthcare (THC), Electronic Data Systems (EDS) and Fannie Mae (FNM) didn't pan out. Since 2003 those new investors have trailed the S&P 500 by an average of 6%, one of the worst tallies in the large-cap value category. Last Christmas a new management team was installed.

It's a phenomenon that happens all too often — just ask investors at Calamos Growth (CVGRX), Janus Mercury (JAMRX), Fidelity Magellan (FMAGX) and Weitz Value (WVALX), among other funds.

Researchers track the issue using so-called dollar-weighted returns. This measurement calculates returns based on the performance of a fund's portfolio, but it also takes into consideration asset inflows. The returns during years with heavy cash infusions are weighed more heavily in calculating three-, five- and 10-year returns than those 12-month periods garnering little new money. The idea behind it: Investors should grade themselves not only on the fund they pick — but when they picked it. "People do a poor job of timing their investments," says Russel Kinnel head of Morningstar's mutual fund research.

Getting your report card won't be easy, though. Dollar-weighted returns aren't listed in prospectuses or tracked by the major mutual fund sites. (Morningstar is toying with the idea.) But to spot potential trouble you can keep your eyes open for a few red flags. Stay away from "hot" funds that are posting eye-popping numbers that seem too good to be true. (They are.) In addition to looking at a fund's performance over a broad time period, you should also study annual returns over that same span. A fund's stellar 10-year track record could be juiced by three years of outperformance followed by seven years of lackluster results. And as an existing shareholder, if your favorite fund starts to swell with new money it could signal a tough road ahead, since the cash will force a manager to increase the size of his holdings or buy stocks he may not want to. To protect yourself, dollar-cost average into your fund of choice.

Dollar-weighted returns have long been studied by researchers who analyze the stock market. These experts focus on investors who always seem to rush into hot stocks or a company that floods the market with shares to fund acquisitions — both of which can wash out returns for new investors. Ilia Dichev, an accounting professor at the University of Michigan's Ross School of Business, found the average return of the New York Stock Exchange between 1926 and 2002 was actually 1.3% less than what other researchers had published; the Nasdaq's return was 5.3% less between 1973 and 2002. The results are even more profound when you look at tech bubble companies like Cisco Systems (CSCO). "In the late 1990s it had a 45% average annual return, but if you calculated the dollar-weighted return it was more like 12%," says Dichev. "There was a lot of suffering, but it was invisible suffering."

Researchers are just now starting to use the same methods to get a clearer picture of historical mutual-fund returns. Dalbar, a fund data company, found that between 1984 and 2002 a typical fund shareholder's real returns were a paltry 2.6%, lower than the inflation rate and considerably less than the 12% earned by the S&P 500 index. The Vanguard research arm run by John Bogle thinks the number is even lower.

Last year Kinnel released a study that looked at the dollar-weighted returns for certain styles, sectors and individual funds. He found between 1995 and 2005, large-cap value funds had the most consistent dollar-weighted returns, varying only 0.4% from their published numbers. The mid- and small-cap growth categories, however, had gaps of 3%, technology funds had a difference of 13% and four funds — Weitz Value, Janus Mercury, Calamos Growth and White Oak Select Growth (WOGSX) — had an average 15% disparity between their published performance and their dollar-weighted returns. "I knew going in anything from the bear market was going to look bad," says Kinnel. "People were just collecting funds instead of building better portfolios."

There are plenty of reasons to account for poor dollar-weighted returns. The most worrisome is that investors tend to get in at the top of a rally. Between 1997 and 1999 the 20 funds with the largest gains returned 51%. Once the tech boom nosedived those same funds lost an average 32% over the next three years. Unfortunately, most of the money filtered in at the peak. According to Vanguard, the dollar-weighted returns for this group between 1997 and 2002 were a negative 12.2%. Another potential pitfall: investing in a particular style of funds that has seen a flurry of incoming assets but is due for a downturn. That's certainly the case at the moment with small-cap offerings.

But there are other legitimate reasons to consider, too. Every good fund manager can get into a slump over the long haul. That's why we would stick to funds like Weitz Value that are run by seasoned veterans. Even Clipper deserves a second look now that Davis Funds, one of the best independent fund shops in the industry, is at the helm. After all, not only is buying at the wrong time a cardinal sin of investing, so is selling too soon.