That same day, Morningstar managing director Don Phillips posted a lengthy defense of the star rating system on the Morningstar website. (Earlier last month, my Moneywatch colleague Jane Bryant Quinn posted her own criticism of the star ratings.)
So who's right? In a way, all of them are. Used as an indicator of future performance, the ratings are worthless. But used as a way to separate the true dogs in the mutual fund industry from those with a fighting chance of earning decent relative returns going forward, the ratings can prove useful.
While Morningstar has never claimed that its rating system had any predictive value in helping investors select a fund that will outperform in the future, the results of Courtney's are rather extreme. But those results are a function of a flawed rating methodology that Morningstar revamped in 2002.
Prior to the change, all domestic stock funds were lumped together, with the top-performing decile receiving five stars, and so on. The result was that the ratings were much too heavily influenced by the performance of the market's sectors.
Thus in 1999, the year that Courtney's study starts with, five-star funds were dominated by those that were heavily invested in the technology-heavy large-cap growth sector. A small-cap value manager who managed to outperform his benchmark by one percent, on the other hand, earned a lower rating simply because the small-cap value sector trailed large-cap growth so badly.
Morningstar's 2002 revamp addressed that flaw by rating all funds by style -- large growth funds are rated against other large growth funds, and so on. But while that marks a big improvement, it doesn't change one important criticism: the ratings system is heavily weighted toward recent performance.
What's Wrong with the Star RatingsAccording to Morningstar's methodology, the overall rating for a fund with at least a ten year track record is calculated with the following weights: 50 percent for the fund's 10-year rating; 30 percent for the five-year weighting; and 20 percent for the three-year weighting. But a bit of simple math shows just how important the past three years are.
Obviously, the past three years account for 30 percent of the past ten years, which means that they account for 15 percent of the overall rating (30 percent X 50 percent). They account for 18 percent of the five-year rating (60 percent X 30 percent); and 100 percent of the three-year rating. Sum them all up, and we find that the past three years account for 53 percent of a fund's overall long-term rating. Unsurprisingly, the past three years are weighed even heavier for funds with only five years of performance, accounting for 76 percent.
As I noted in an earlier article, four- and five-star rated funds attract the vast majority of investor cash flow. And given the heavy weight the rating system applies to recent performance, those investors are likely to earn inferior returns as those hot funds eventually cool. But it would be silly to lay this performance chasing at the feet of the rating system. Investors have always chased past performance, and if the stars did not exist, I'm certain that the most recent top performers would continue to attract the lion's share of investor cash.
What's Right About the Star RatingsBecause Morningstar adjusts fund returns for sales loads and redemption fees, and because over the long term lower cost funds outperform higher cost funds, Morningstar's rating for funds with lengthy track records is likely to favor funds with lower overall expenses.
Additionally, as Don Phillips pointed out, higher rated funds are more likely to have investor-friendly characteristics (such as low turnover, long manager tenure) than lower rated funds.
But that does not mean that a portfolio composed of four- and five-star rated funds is likely to outperform the broad market. As much as the ratings might point you in the direction of funds that have slightly better odds of outperforming, the odds on achieving long-term outperformance are still decidedly against you.
Perhaps the best use of the Morningstar rating system is not as a way of finding which funds to invest in, but as a way to identify which funds to avoid altogether. Fund performance tends to revert toward the mean over the long-term -- today's winners become tomorrow's losers, and vice versa. But some funds are simply so lousy that they have essentially no hope of ever providing decent returns. These funds, obviously, are largely relegated to one- and two-star ratings.
Used best, Morningstar's ratings are simply a place to start the fund selection process, and then primarily as a method of eliminating alternatives. But investors relying on the ratings to help them construct a portfolio that will outperform the markets are destined for disappointment.