Morningstar’s star-rating system is widely used to compare mutual funds. Should investors rely on these rankings? The following discusses why investors should not simply gaze at the stars. When choosing mutual funds, many investors go straight to the fund’s rating from Morningstar. The logic is simple: five-star funds are the best, so these funds should be the obvious choices. However, these investors would be wise to step back before jumping into a fund based on its rating.

Numerous studies have been conducted on the predictive power of the Morningstar ratings system, with most reaching the conclusion that the system has little (if any) predictive powers. Generally speaking, the system can identify the funds that have consistently poor performance (either by poor stock selection, bad market timing efforts, high expenses or combinations thereof), but cannot identify future outperformers with relative consistency. And most investors would not be happy with funds that only used to perform well.

A New Day?
In 2002, Morningstar reconstituted its rating system in hopes of improving its predictive powers. Previously, Morningstar placed funds in only four categories. Under these groupings, funds that focused on the hottest sectors could gain an edge in the rankings. For example, in mid-2000, three-quarters of funds that invested in large, fast-growing companies garnered four or five stars. At the same time, only six of the 123 funds that specialized in small, bargain-priced stocks earned such a ranking.

Today, Morningstar uses 48 groups —such as small-cap value, foreign large-cap growth and global real estate—with the intention of better classifying funds and providing a more apples-to-apples comparison.

Also, Morningstar used to judge the riskiness of a fund by simply comparing its average performance to that of the 90-day T-bill. If the fund beat the T-bill, it was considered safe, so even funds with highly volatile returns could be considered safe if they simply beat the benchmark. According to this standard, many technology funds were considered safe in 1999, even though they ended up suffering significant losses in the 2000–2002 bear market.

According to Morningstar, now “risk is measured as the amount of variation in the fund’s performance, with more emphasis on downward variation.”

Recent Study
A study by professors Matthew Morey and Aron Gottesman found that the new system did indeed rank funds in a better manner than the previous system. That is, five-star funds did better than four-star funds, which did better than three-star funds and so on.

But before simply concluding that Morningstar “got it right,” a closer look at the data is warranted. First, the time period chosen is not only a short period, but almost entirely encompasses an up market.

It is logical to think that better performing funds will continue to outperform peers when things are going well. It is the down markets where the star ratings should have the most value. Indeed, the changes made to Morningstar’s rating system were aimed at providing more accurate readings during down markets, so the true value of the reconstituted system has not yet been measured.

An interesting finding of their study was that the alphas measured in each star-rating class (with one exception) were negative. This implies that investors would likely do better by purchasing index funds no matter what the star rating.

While the number of fund categories has been expanded to provide more accurate comparisons, several studies have shown that mutual funds are often miscategorized. This may indicate that funds are compared with others that have lower expected returns. For example, a large-cap value fund categorized as a large-cap neutral fund would have an advantage over its peers, as it has a higher expected return. Conversely, this would mean that funds actually belonging to this class may have a lower rating simply due to being compared with funds outside their class.

Perhaps an Improvement, But Still Flawed
To be fair, Morningstar itself has said that “too often investors use the star rating as the main criterion in evaluating funds, only considering those with four or five stars and casting off the rest as subpar. That’s not always the best strategy, though, because the star rating 3 doesn’t necessarily tell the whole story.” Morningstar provided a few examples of why the star system does not always provide an accurate picture.

Funds are ranked based on their three-, five and 10-year track records, with more weight given to older records. Thus, funds that haven’t been around as long may have an advantage over older funds. Consider two funds: One created in 1999 and one created in 2003.

Since the newer fund hasn’t been around for 10 years, only its three-year and five-year records will be considered when assigning a ranking. However, the older fund would also have its 10-year record considered, and that record would receive more weighting in determining its ranking. The older fund would have experienced the 2000–2002 bear market, which would certainly count against it. Thus, the newer fund that began after the 2000–2002 bear market is likely to receive a higher rating regardless of the quality of the funds.

This means that stars with poor ratings may be good funds, but received poor ratings due to the methodology. Consider the case of Longleaf Partners International. The fund held a two-star rating for a significant amount of time due to ranking near the bottom of its category in returns for 2004 and 2006. The fund was not yet 10 years old, so only its threeand five-year records were considered.

However, the instant the fund hit the 10-year mark, its rating shot up to four stars due to its 10-year record, which beat 97 percent of the other funds in its category.7 Summary Investors who use Morningstar’s star system as the determinant for choosing funds may be approaching their investment choices the wrong way. There are more important factors to consider when building a portfolio, such as how a fund fits into an investment plan.