Recently, the Wall St. Journal, the most wisely read financial newspaper in the U.S. ran a long research article very critical of the Morningstar rating rating system that awards funds 1 to 5 stars. The ratings are very widely followed and are the basis for a great deal of investment into mutual funds, from individuals to professional investment managers to institutional investors.
What the Journal found was that of the funds Morningstar rated 5 stars, only 14% were 5 stars five years later and that there were more of them rated 5 years later as 1 star than 5 stars. They quoted Morningstar as saying that the star rating was never intended to be predictive. But, it would be hard to argue that they are not used that way or intended to be used that way.
So, how should we take that? Is Morningstar worthless? Worse, does it actually increase the chances of faring poorly?
First, I have noticed myself the tendency for funds that carry high performance through the current year to stumble the next year. For that reason, I consider short-term data to be not that useful.
But, the Morningstar database allows advisors like myself to get a great deal of data about a fund. It tells us how volatile it is likely to be, how consistent it has been in the past, what style the fund has, how it tends to correlate with different factors, whether there is a new manager, if the manager is significantly invested in his own fund, the stewardship of the fund family and a number of more esoteric bits of information that I find useful.
Here’s a chart in that WSJ article that I think is very instructive.
Yes, funds from every rating tend to on average converge in future performance, with most of the convergence happening by the 3rd year. But, just as importantly, the average rank of each star rank group stays the same. The 5-star group remains the top group and the 1-star group stays the lowest group on average. Amazingly, the same is true for all five groups.
Second, the 1 through 3-star group tends to keep dropping in relative performance but the 4-star group not so much and the 5-star group tends to keep its relative outperformance.
Those who have seen how performance converges to the mean are not surprised by all this. But, the study actually vindicates the ranking system. The only thing it really says is that the confidence in how every fund will perform in the future has been too high for many investors.
This is why I so value consistency of returns, both in actual results and in style. The group of funds that stays consistently better than its peers is not a large group, hence the strong move toward indexing, especially with ETFs.
I have my own ideas on how best to predict that. I would like to see more studies that look at factors like team management of funds vs star managers and the effect of lower expenses on predictability. The frequency with which a fund has stayed in the top two half for fees and performance should be another positive factor. The effect of being bought by another firm or having a manager change, especially in a star manager system, should be generally negative if past performance had been very good.
Bottom line – the Morningstar rating system is still relevant, though not as predictive as people not paying close enough attention might have believed. Morningstar data has also greatly improved transparency and lowered fees. Future study would be very helpful on determining what other factors help predict good future performance. And, since the star groups as wholes keep their relative rankings, you are usually better off with a 4 or 5-star fund than the others.