Russel Kinnel is director of manager research for Morningstar and editor of Morningstar® FundInvestor℠, a monthly newsletter.
In examining the expense ratios and star ratings, I settled on three key measures that, to me, are closest to investors' bottom lines and help cut through all of the clutter.
While total returns are nice, they are not the whole story. Mutual fund companies kill off their funds at a rapid rate--thus, sweeping mistakes under the rug. However, your losses are just as real if your fund is liquidated. If there's a destruction bias for a data point, then you want to factor that in. The success ratio tells you what percentage of funds in a given group survived and outperformed their peers. After all, that's what success really is. Anything short is a failure; yet too often, investors and the press act as though total returns are the same thing as the success ratio. This is the strongest of the three measures because it is not affected by survivorship bias.
Everyone wants to know how any measure works at predicting total returns. Because equal proportions of each category are given 5 stars and 1 star, one can safely sum up returns across categories to see how the measure has done for an asset class as a whole.
Subsequent Star Ratings
The star rating is a measure of risk- and load-adjusted returns, so naturally I want to know whether the star rating is able to predict future risk- and load-adjusted returns. Investors have long handled lower-risk funds better than higher-risk funds because lower-risk funds don't trigger strong feelings of fear or greed. Thus, lower-risk funds with slightly lower official returns actually led to better results for investors than high-risk, high-return funds.
How We Ran the Data
We took a snapshot of star ratings and expense ratios from 2005 through 2008 and then tracked their progress through March 2010. We rolled up category level data into five broad asset classes: domestic equity, international equity, balanced, taxable bond, and municipal bond.
We then measured total returns as of the end of March 2010 for the mutual funds that survived the entire period. For the success ratio, we included funds that were merged or liquidated, as well as those that survived, in order to calculate the number that both survived and outperformed. For the star rating, we recorded the five-year star rating for the data set from 2005, as well as the three-year rating. For 2006 and 2007, we recorded the ensuing three-year rating--meaning we measured the figure in March 2009 for the class of 2006 and the rating in March 2010 for the class of 2007. For the class of 2008, we don't yet have a star rating.
For the purpose of this article, I focus on the gap between 1- and 5-star funds and cheapest and most expensive quintiles.
How Expense Ratios Performed
If there's anything in the whole world of mutual funds that you can take to the bank, it's that expense ratios help you make a better decision. In every single time period and data point tested, low-cost funds beat high-cost funds. To see the results, click here
Expense ratios are strong predictors of performance. In every asset class over every time period, the cheapest quintile produced higher total returns than the most expensive quintile.
For example, the cheapest quintile from 2005 in domestic equity returned an annualized 3.35% versus 2.02% for the most expensive quintile over the ensuing five years. The gap was similar in other categories such as taxable bond, where cheap funds returned 5.11% versus 3.82% for pricey funds. That same relationship held up dependably in the other time periods we measured. For 2008, the cheapest quintile of balanced funds lost 0.04% over the next two years, while the most expensive shed 1.13%.
The gap was also impressive as measured by the success ratio because high-cost funds are much more likely to have poor performance and be liquidated or merged away. For the 2005 group, we found that 48% of domestic-equity funds in the cheapest quintile survived and outperformed versus 24% in the priciest quintile. Put another way, funds in the cheapest quintile of domestic equity were twice as likely to succeed as those in the priciest quintile. It was a similar story in other categories, although in munis the advantage was greater than 6 to 1. The same basic relationship held up for the other years we looked at. Although I think of expense advantages as taking a long time to compound to your advantage, even the 2008 group saw low-cost funds with nearly a 2 to 1 success advantage.
Given that performance edge, you won't be surprised to hear that low-cost funds also produced better risk- and load-adjusted performance as measured by the star rating. For example, the 2005 group enjoyed a subsequent 3.23 average star rating compared with 2.66 for the priciest quintile in domestic equity. The edge grew in taxable bonds to 3.34 versus 2.3. The edge held up for predicting three-year ratings for the 2006 and 2007 groups.