In Looking at the Numbers, Be Sure You Know What They’re Based On

Let’s face it: Good performance is what you’re really looking for in a mutual fund.

Low risk is nice, and moderate expenses make sense. It also helps to select funds that meet your objectives and fit nicely into your portfolio.

But performance is the bottom line. Arthur Levitt, chairman of the Securities and Exchange Commission, said as much earlier this year when he cautioned fund-company executives not to focus so much on performance in ads and marketing literature.

“I worry that the fund industry is building unrealistic expectations through performance hype,” he said. “You’re setting yourself up for millions of disappointed investors when your selling efforts focus exclusively on the recent bull market.”

Levitt was referring to the fat returns that thousands of funds have generated after several years of rising stock and bond prices. What he didn’t mention is that performance results for certain funds are misleading if not downright irrelevant. That’s partly because the SEC has relaxed its own rules on what constitutes a legitimate track record, but there are other causes too.

Here are the issues to watch if you want to be a savvy reader of fund claims:

* Related returns. A hot issue these days concerns promoting return figures based on something other than what the current fund has compiled. The Securities and Exchange Commission in recent years has allowed certain new funds to make use of “related” performance numbers that aren’t entirely authentic.

The litmus test was Elizabeth Bramwell’s receiving permission a few years ago to use the track record she built at Gabelli Funds to tout her own then-new Bramwell Growth Fund. Other examples include claims for two funds from Barr Rosenberg that cite numbers compiled for the firm’s private accounts, and claims for new mutual funds run by former Janus star manager Tom Marsico. The new Marsico funds cite the long-term track records of the Janus Twenty and Janus Growth & Income portfolios, which Marsico headed for nearly a decade.

The effect of these loopholes is that investors may assume--when such very well may not be the case--that the new fund will be run in the same manner as the fund whose track record is being used as a basis for performance claims.

“There’s no doubt that such past performance data can be useful information, but you’ve got to take it with a grain of salt,” wrote Morningstar’s Catherine Voss Sanders in a recent issue of the fund tracking firm’s Morningstar FundInvestor newsletter.

* Incubated funds. These funds are seeded with a firm’s own money and perhaps that of employees but withheld from the investing public until a track record can be established. The idea is that only those portfolios with good numbers will be trotted out.

Incubated funds, unlike regular mutual funds, operate in an environment of steady cash inflows and little or no outflows, thus their track records are not truly reflective of performance in the “real world” of mutual funds. “Putting floods of new cash to work or rushing to meet redemptions,” as regular mutual funds must do, “can add a tricky dimension to the money management game,” Sanders said.

* Misclassification. It’s amazing how good or bad a fund can look if it’s miscategorized.

For example, an investor combing data from Lipper Analytical Services might notice that Papp Pacific Rim was the top Pacific-region fund for the 12 months through March 31, with a return 36 percentage points above its closest rival. The only problem is that Papp Pacific Rim has never owned a significant number of Asian stocks, preferring to invest in U.S. companies with strong ties to Asia. Lipper later moved the fund into its growth category, where it finished near the middle of the pack for the second quarter.

* Manager departures. One of the most difficult performance questions is this: How do you evaluate a fund’s track record if the manager who compiled the numbers has left?

First, evaluate the present circumstances. For example, if a co-manager remains on the scene or the fund has a strong team of research analysts standing behind it, then perhaps the fund may remain a good buy. If you already own the fund, you may want to watch it carefully for a time--say, six months. Be especially wary of a knee-jerk decision to sell if doing so would cause you to realize a capital gain at an inopportune time.

Fidelity’s Select sector funds offer a case in point. About 21 of the 36 Select portfolios with five-year records have outperformed the average domestic-stock fund over that span. These figures have been achieved despite a never-ending game of musical chairs among Fidelity’s sector-fund managers, and thus are partly indicative of a strong research team.

* Survivor bias. If you’re buying or selling funds based on relative performance, be careful: The average long-term returns generated by broad fund categories are inflated.

Simply put, category return averages for five years or more aren’t quite as good as they seem. That’s because weak mutual funds have been liquidated or merged into healthier funds along the way. As they disappear, so do their sickly track records, thereby boosting the collective returns of the survivors.

“The effect of survivor bias on publicly traded mutual funds has been studied extensively,” reads a recent report from Goldman Sachs. “The studies concluded that the average return of surviving funds is 50 to 150 basis points [0.5 to 1.5 percentage points] higher per annum than for all mutual funds.”

Keep in mind that survivor bias doesn’t inflate the results of an individual fund. It only becomes relevant if you decide to sell a fund because its performance appears to be lagging the competition.

*

Russ Wiles is a mutual fund columnist for The Times. He can be reached at russ.wiles@pni.com.