DON'T WISH UPON THE STARS TO PICK FUNDS
By JASON ZWEIG

(MONEY Magazine) – For a fund company, getting the top five-star rating from the oracles at Morningstar is better than being blessed by the Pope. Last year, according to Financial Research Corp. of Chicago, 87% of all new money going into U.S. stock funds--some $97 billion--went into funds rated four or five stars by Morningstar, the fund rating firm also based in Chicago. That statistic suggests that too many investors--not to mention brokers and financial planners--are using Morningstar as the ultimate shortcut: They count the stars and buy the fund. As if this star-crossed tactic weren't rampant enough, soon you'll be seeing even more stars. In May, Morningstar announced that it will begin providing one-year star ratings in addition to its three-, five- and 10-year scores. Morningstar took the step reluctantly, to enable funds to comply with advertising rules set by the National Association of Securities Dealers.

With at least 9,000 funds from which to choose, investors can be forgiven for feeling they need a handy screening tool. But the star system is a clumsy one at best. To their credit, the folks at Morningstar say as much themselves. "We've consistently conceded the limitations of our star ratings," says Morningstar president Don Phillips. "The star rating is not a conclusion; it's a starting point." Both Phillips and Morningstar publisher John Rekenthaler say they've owned several one-star funds and that the bulk of their assets are in three- and four-star funds. A look at how stars are born will help explain their shortcomings. Morningstar computes each fund's total return (income plus or minus price changes) and its risk (the average amount by which the fund underperforms the 13-week Treasury bill). It then subtracts any sales loads and compares each fund's figures with the average of all funds in its broad category--equity, "hybrid" (which includes funds that own both stocks and bonds), taxable bond or tax-free bond. Morningstar divides the resulting range of scores into five tiers. The top 10% of funds in each category land in the highest tier, winning five stars; the next 22.5% form the second tier, earning four stars; the following 35% get three stars; and the remainder get one or two stars.

There are several flaws in this system. First, the stars tell you quite a bit about a fund's past but very little about its future. The average five-star fund sustains that rating for only six months. And every once in a while a nova collapses into a black hole. Piper Jaffray Institutional Government Income, for instance, had five stars at the end of 1993. Four months later, after the fund lost 20% of its value, it had only one left.

Second, the stars tell you more about fund categories than about individual funds. A remarkable 52% of all U.S. aggressive growth funds had four or five stars as of the end of March; all that really says, though, is that the sector is booming. Conversely, only 5% of all foreign funds now carry four or five stars, down from 31% at the end of 1994. But that slump merely reflects the fact that the U.S. market has been stronger than foreign ones since then--because Morningstar rates foreign funds against all stock funds (most of which invest at home). Likewise, Morningstar compares short-term and long-term-bond funds on the same scale--even though they react very differently to changes in interest rates.

Phillips defends Morningstar's broad rating scales, saying that extra refinements would simply cause other distortions. He adds, however, that in the future, users of the firm's software may be able to customize their own fund baskets for star ratings. For example, you could compare Scudder International with other foreign funds, rather than with all stock funds.

Meanwhile, how should you gaze at the stars? For openers, do it as little as possible. Ignore the new one-year rating altogether; the period is too short to tell you anything useful. Even before counting stars over a longer period, ask: Does a fund own U.S. or foreign, large or small, low-priced or high-priced stocks? Does a bond fund favor short term or long term, investment grade or junk? Does it overlap with funds you already own? How much did it lose in the market drops of 1987 and 1990? Do its multiyear "average" returns come mainly from one or two great years or from steadily good results? Are its expenses nice and low? Does it generate a high or low income tax bill?

The wealth of data available from Morningstar (plus services from Lipper Analytical, Value Line and MONEY itself) can help you answer these vital questions. Only after you've narrowed your search to three or four worthy candidates should you look to the stars to help you make a final choice.

Morningstar's Rekenthaler suggests using the stars to determine which asset categories are in favor and which are not. Depending on your risk tolerance, you may want to buy a fund in a hot asset class on the gamble that it will stay sizzling--or a cold one, on the bet that it will soon defrost. (Personally, I much prefer the second approach.) The shortage of high-rated foreign funds, for example, suggests that international stocks may be unfairly neglected right now. This approach turns the stars into a contrary indicator, and that makes a lot of sense to me.