Studying the New Star Rating for Funds

It’s been more than two years since we changed the Morningstar Rating for funds. Naturally, we want to use a number of different measures to monitor how the rating is doing. Because most mutual funds are designed for the long haul and we only have two years’ worth of data to measure; it’s far too early to draw strong conclusions. Some trends have appeared, but we can’t yet say if they are simply noise. However, we can still get some indication on how the star rating is working and therefore how it should be used.

How the Rating Works Now
In June 2002, we changed the rating groups from four broad asset classes to specific categories. We did this to ensure that funds would earn high ratings only when their managers added value, rather than when their category performed well. We also wanted a better reflection of funds’ risk levels, so we adopted a utility function of risk measure that penalizes funds for volatility in either direction. In addition, we gave the new risk measure a greater weighting in the rating calculation to prevent it from being overwhelmed by returns.

Some things remained the same. The star rating is still a quantitative measure that’s updated monthly. It is still graded on a curve, with 10% of funds getting 1 star, 22.5% getting 2 stars, 35% getting 3 stars, 22.5% getting 4 stars, and 10% getting 5 stars. We still adjust for loads.

How We Ran the Study
We grouped funds by their star ratings and measured how they performed relative to their category peers over the two years following our adjustment to the rating. We also looked beyond performance to examine fund characteristics such as cost and risk when grouped by star rating.

Does the Rating Predict Returns?
The star rating is designed to provide a quick summary of a fund’s past risk-adjusted performance. However, we recognize that what most investors care about is whether it can predict future performance.

The early indications are that the rating does have some predictive abilities, though it’s important to recognize that we’re looking at a short time span. Nonetheless, in all four major asset classes, 5-star funds outperformed lower-rated funds from the same category from July 1, 2002, to June 30, 2004. Likewise, 1-star funds in each asset class underperformed.

How the Star Rating Fared on Performance and Costs
Asset Class	Star Rating	2-Yr Avg % Rank	Expense Ratio	Front Load	Deferred Load
Domestic Equity		45	1.39	0.66	0.64
		50	1.34	0.74	0.73
		51	1.44	1.11	0.96
		53	1.67	1.51	1.27
		52	1.95	1.67	1.26
International Equity		32	1.66	1.00	0.72
		47	1.58	0.75	0.73
		53	1.75	1.15	0.86
		60	2.14	1.51	1.45
		56	2.32	1.73	1.24
Taxable Bond		38	0.77	0.22	0.23
		48	0.86	0.42	0.34
		51	1.10	0.83	0.78
		59	1.33	1.36	1.39
		52	1.44	1.51	1.73
Muni Bond		22	0.58	0.38	0.10
		37	0.81	1.34	0.34
		45	1.06	1.72	1.05
		59	1.31	1.32	1.83
		64	1.50	0.87	2.34
Data through June 30, 2004.

Among U.S.-stock funds, 5-star funds outperformed 55% of their category peers on average, while 1-star funds finished a little below average. The gap was wider in international equity, where 5-star funds beat 68% of the competition, but 1-star funds only outperformed 44% of their peers on average. In taxable bonds, 5-star funds beat 62% of their peers on average, while 1-star funds lagged slightly. The star rating fared best of all in the muni-bond category, where 5-star funds trounced 78% of their peers and 1-star funds outperformed only 36% of their peers. As you can see, the star rating could have helped you enjoy better returns, but it certainly wasn’t a guarantee of huge returns. If this advantage were to hold up over time, however, that modest advantage would compound into a very large edge in the long run. Still, we don’t know if that’s how it will work.

Where the Star Rating Wasn’t Predictive
Although 5-star funds beat 1-star funds most of the time, there were a few notable exceptions. We noticed that the rating didn’t predict future returns as well when high-risk strategies were paying off. That makes a lot of sense, because the star rating is risk-adjusted. We believe investors do better with funds that produce good returns and a smoother ride.

If you think back, 2002 was the year of credit fears. Scary implosions like WorldCom’s spurred panic selling in junk bonds. That selling soon spread to even decent corporate credits like A rated  Ford Motor (F). This selling wasn’t much more rational than the Internet buying binge of 1999. The bottom for credit came in October 2002—not too long after the beginning of our study period. The following year, bond investors regained their senses, and lower-quality credit enjoyed a powerful rally.

Because of those forces at work, the star rating didn’t work too well in lower-quality categories, such as high-yield and multisector bonds. In those categories, the funds with the least credit risk dominated the 5-star slots at the beginning of our study. A subsequent reversal in market direction meant those same funds lagged over most of the following two years. Thus, 5-star high-yield and multisector funds were the worst places to be. For high-yield bond funds, 3-star funds were the best, as they, on average, landed in the category’s top 40%. In multisector bonds, 1-star funds led the way, with returns averaging in the top 37% of the category.

We saw a rather similar effect in technology and large growth. June 2002 was also close to the nadir for tech stocks and high-P/E stocks. Thus, tech funds and large-growth funds with high price risk or tech exposure received low ratings, while more-conservative funds received higher ratings. As you know, tech came roaring back in 2003—a boon to those same high-risk funds. Thus, 5-star tech funds slightly underperformed, while the average 1-star tech fund landed in the category’s top 39% over the ensuing two years. The average 5-star large-growth fund outperformed only 44% of its peers, while 1-star large-growth funds robustly outperformed 63% of their rivals.

By definition, risk-adjusted returns will vary from unadjusted returns by the amount of risk, so these results aren’t a big surprise. It will be interesting to see if lower-risk funds regain ground over the long haul. Certainly, tech risk was punished in the third quarter of 2004, after this study’s time period concluded.

Where Did the Rating Do Best?
In a word: bonds. Municipal bond and high-quality corporate-bond funds with high ratings did much better than those with low ratings. The fact that the star ratings for munis on the whole did better than those for taxable bonds is a further indication that the success of high-risk strategies can undo the ratings in the short run. Unlike taxable bonds, munis didn’t suffer from a credit meltdown. Therefore, there was no snap-back for lower-quality funds. 

The same was the case for government-bond funds where credit quality isn’t much of an issue. In the intermediate-government category, 5-star  funds outperformed 79% of their peers in the two-year time period, while 1-star funds outperformed just 27%. Intermediate-term bond (government debt plus corporate debt) 5-star funds beat 70% of the competition.

Among stock funds, the star rating worked particularly well in the large-value and mid-value arenas. Five-star large-value funds outperformed 64% of their category, while 5-star mid-value funds topped 72%. Considering how unpredictable small growth is, it was a little surprising to see the star rating work so well there, as 5-star funds from that category beat 60% of their peers.

Do 5-Star Funds Have Lower Costs than 1-Star Funds?
In nearly every category the answer is yes. It’s not a big surprise that 5-star funds have lower costs, as expense ratios are the best predictor of returns that we know of. They’re also good predictors of risk in the bond world, because managers of high-cost bond funds sometimes take on more risk to keep their yields competitive with lower-cost funds.

In addition, 5-star funds have lower front-end and back-end fees as well. The average 5-star domestic-stock fund charges a front load of 0.66%, compared with 1.67% for 1-star funds. Put another way, when you buy a 5-star fund, you are usually buying a lower-cost fund.

What’s the Risk Profile of 5-Star Funds?
As you might expect from a risk-adjusted return measurement, lower-risk funds earn better star ratings than high-risk funds. As we’ve found in past studies, high-risk funds are much more difficult for investors to use because they may get in near the top and then sell at the bottom when the pain becomes too great.

How You Can Use the Star Rating
The star rating is a great way to quickly gauge whether a fund has added value over time on a risk-adjusted basis. So far, there are signs that it has some predictive value, but it’s too early to draw conclusions. Even if these early indications are borne out over time, though, the star rating should only be used as one of many data points that enable you to select and monitor funds.

After all, the rating measures past performance and therefore doesn’t immediately reflect fundamental changes. Say, for example, a manager leaves or a fund’s expense ratio is cut significantly. These are both events that alter a fund’s long-term prospects, so you’d want to factor them into your evaluation--even though the star rating wouldn’t reflect the change.

One good way to use the rating is as a negative screen. With thousands of funds out there, you could begin by tossing out 1- and 2-star funds. Sure, you’d lose a few good funds by doing so, but screens are designed to narrow the field--not capture every possible idea. Likewise, you could watch to see if any funds you own slip to 1 or 2 stars. If one does, I’d suggest digging further to see why it has underperformed. It might be that the fund’s style is out of favor or that poor returns by previous managers are hurting returns, and you should stick with it. Or it could be that the fund just isn’t executing like you’d hoped, and it’s time to cut it loose.

At this point, using the rating as a positive screen--that is, screening for 4- and 5-star funds when searching for a fund to buy--seems only warranted for bond funds, as the margin of success for 5-star stock funds is rather slim.

In summary, the Morningstar Rating for funds can help, but it doesn’t free you from doing your homework on a fund’s fundamentals. That’s one reason we have a Fund Analyst Picks list based on fundamental criteria and significant legwork by our analysts. We hope the star rating will help you reach your goals when you make it a part--not the heart--of your process.

A version of this article appeared in the November 2004 issue ofMorningstar FundInvestor.