The disclaimers regarding Morningstar's rating system have always been there for investors.

Advisors have long warned that the company's famous star-based system is misleading in its simplicity, and academic researchers have concurred, saying their data show the stars are poor predictors of investment performance. Morningstar itself has continuously cautioned that its ratings are mere starting points for serious research, not ends in themselves.

But millions of investors, unable or unwilling to turn away from the prospect of easy answers, have ignored the cautionary notes. Time after time, clients insist their advisors buy five-star wonders only to watch them fall short of the lofty expectations.

As part of an effort to address the problem, Morningstar revised its system a year ago. Is the new approach more reliable than the old? Preliminary results are mixed.

“The new method is better, but advisors still cannot rely on the star system,” says Pran Tiku, a principal with Peak Financial Management, a registered investment advisor in Wellesley, Mass.

Stakes Rising

Whether the new system will prove helpful seems particularly important now because Morningstar is extending its ratings to separate accounts. If it hasn't happened already, you can bet that your clients will soon come racing to your office insisting on buying five-star separate accounts.

When it was first introduced in 1985, the Morningstar approach represented a big step forward. For the first time, individual investors had easy access to comprehensive data on funds. Morningstar directories soon became among the most popular volumes at public libraries.

A large part of the system's appeal, of course, is simplicity. Under its original approach, Morningstar divided the fund universe into four categories, domestic stocks, foreign stocks, taxable bonds and municipals. Funds were graded by risk-adjusted returns, and the top 10 percent received five stars, while the bottom 10 got one star.

The star system seemed well suited for an era in which many funds were generalists, holding broad collections of stock or bonds. But in the 1990s, funds became more specialized, and flaws in the star system grew more pronounced. For instance, because all domestic funds were lumped together, small-cap value specialists were compared to large-cap growth. With the markets favoring large-cap growth in 1999, nearly all funds in that category had four or five stars. At the same time, nearly all small-value funds were rated three stars or lower. When the market turned down in 2000, it became clear that investors who followed the stars were making exactly the wrong moves.

Style Points

To repair the system in 2002, Morningstar began awarding stars according to style categories. That way, small value funds would only be compared with each other. And no matter what sector the market favored, the top 10 percent of all small value funds would have five stars.

“The old ratings just told you what part of the market was hot,” says Todd Trubey, a Morningstar analyst. “Now you will see more stable and predictable results.”

A year after the new system went into effect, Morningstar analyzed the results and painted a mixed picture of the outcome. If you bought the average four or five-star foreign fund in June 2002, you recorded above-average results for the next 12 months. High-rated bond funds did particularly well, probably because five-star fixed-income funds tend to have low fees — a consistent advantage in the bond world.

On the other hand, top-rated mid-cap blend funds underperformed their low-rated competitors by nearly 2 percentage points. Overall, Morningstar declared, it was too early to draw any conclusions, and “no trends emerged.”

Some observers predict that the new system is not likely to prove much more effective than the old one. As in the past, Morningstar takes the returns of a fund and then deducts points for the amount of risk taken. So riskier funds have a harder time obtaining top rankings. The problem, say observers, is that Morningstar focuses too much on past performance, a questionable predictor.

“To do proper due diligence, you have to really understand portfolios and then make an educated guess about how they will do in the future,” says Donald Robinson, chief investment officer of Lockwood Advisors, an investment advisor in Malvern, Pa.

Star Attraction

Robinson says Morningstar's new separate account ratings are likely to have some of the same sorts of limitations as the mutual fund evaluations. But he says Morningstar's expansion could help the separate account business grow more rapidly.

“It is a well-known name, and my guess is that Morningstar will encourage some advisors to try separate accounts,” he says.

In the separate account field, Morningstar faces established competitors, including Effron's PSN and Mobius Data from CheckFree Investment Services. These enable an advisor to screen for, say, a large-cap manager with high alpha. But the dominant services do not attempt to rate managers with stars or other measures as Morningstar does.

Separate Accounts That Earn Their Stars
Product Category Morningstar Star Rating 12-Month Return 3-Year Return 5-Year Return
Dodge & Cox
Fixed Income
Intermediate-Term
Bond
5 6.4% 10.1% 7.5%
Heartland Advisors
Small-Cap Value
Small Blend 5 65.6 25.3 23.0
Lord Abbett
Large Cap Value
Large Value 4 28.8 2.6 9.3
NWQ Investment
Large-Cap Value
Moderate Allocation 5 20.5 5.7 N/A
Sands Capital
Large-Cap Growth
Large Growth 4 36.4 -13.3 4.9
Source: Morningstar. Returns through 9/30/03.

Current separate-account winners of five stars include Dodge & Cox's intermediate-term bond product and Heartland Advisors' Small-Cap Value account. Morningstar also is introducing a new flexibility rating, which scores separate-account managers on their willingness to provide clients with services, such as access to daily portfolio holdings and tax reports. Managers rated “highly flexible” include Rittenhouse Asset Management and Roger Engemann & Associates.

Loyal users of Mobius and PSN say they see little need for stars. “Mobius gives you terrific data on risk scores, and everything you need to make decisions,” says B. J. Webster, a partner with Wharton Business Group, an investment advisor in Malvern, Pa. “I think it will be difficult for Morningstar to gain market share.”

But Morningstar can't be counted out so easily. Many of its competitors focus on high-end products costing $6,000 or more annually that are aimed at institutions and big broker/dealers. Morningstar targets those customers, but it is also introducing Morningstar Principia Separate Accounts, a disc that costs $1,395 a year. For dedicated users of Principia Mutual Funds, the new disc will be instantly familiar. Even inexperienced advisors can quickly get data on returns and risks — and the star ratings. That could encourage long-time fund loyalists to try separate accounts.