Voices: It’s time for Morningstar to scrap its fund ratings
October 30, 2017 by Nir Kaissar, Bloomberg Gadfly
Morningstar’s ratings for mutual funds is being called into question after The Wall Street Journal took issue with its star-system. The system, which awards one to five stars, is based on the funds’ risk-adjusted performance relative to peers.
The problem, according to the Journal, is that investors assume erroneously that “the number of stars awarded to a mutual fund is a good guide to its future performance.” Stephen Wendel, Morningstar’s head of behavioral science, agrees that “Morningstar’s star ratings for funds are clearly used in the industry to imply that funds that performed well in the past will do so in the future.”
But rather than simply reminding investors that the stars are indicative of past performance only, Morningstar should take stronger action and completely revamp its data offering to better help investors.
Confusingly, the Journal’s analysis seems to suggest that fund ratings are predictive of future performance. It shows that five-star funds earned a higher average rating in subsequent years than all other funds and that four-star funds received a higher rating than all but five-star funds, and so on.
That analysis is flawed, however. Morningstar awards stars for beating peers, not the market. Given that the vast majority of funds lose to the market, highly rated funds may not be delivering any value to investors. Buying the best poorly performing fund is cold comfort.
The better question is whether highly rated funds are more likely to beat an appropriate market benchmark. Morningstar studied that question last year. It looked at the net performance of star-rated funds relative to a custom benchmark that adjusts for various active styles, such as value, size and momentum.
The results are inconclusive. They show that highly rated funds are modestly more likely to outperform than lower rated funds. But the results are not significant in many cases, which means they may just be the result of random chance.
Morningstar says that its star rating isn’t meant to be used as a predictive measure. But let’s be serious: Investors chase actively managed funds because they hope to beat the market.
It’s not surprising that there isn’t a reliable relationship between star ratings and subsequent performance. The average fund return is driven mostly by the market’s return. I recently compared the S&P 500’s trailing 10-year annual returns and subsequent 10-year annual returns since 1871. I found no discernable relationship between the two.
Morningstar says that its star ratings aren’t “meant to be used in isolation or as a predictive measure.” But let’s be serious: Investors chase actively managed funds because they hope to beat the market. If investors can’t rely on the stars to identify future winners, then what’s the point?
The answer is that the stars are good business. Investors love them. Fund companies are eager to play along. Every fund wants a shot at the coveted five stars, so they happily provide all the data that Morningstar requires for a rating. That data is a big reason investors subscribe to Morningstar, including me.
Still, Morningstar has apparently been rethinking its star ratings for years. It introduced a more qualitative rating system in 2011. Instead of stars, the five categories are gold, silver, bronze, neutral or negative.
But there’s no reason to believe that those ratings are any more useful to investors. And rather than bemoan the fact that investors use the ratings to “take a short-cut approach” to researching funds — as Morningstar CEO Kunal Kapoor does in his rebuttal to the Journal — Morningstar should simply remove them.
Instead, Morningstar should offer a curated menu of fund data and let investors decide which funds are best.
Confusingly, the Journal’s analysis seems to suggest that fund ratings are predictive of future performance. That analysis is flawed, however.
It should start by creating a custom benchmark for each fund that accounts for that fund’s active styles, such as value, size, quality, momentum or low volatility. It should then show the fund’s performance relative to that benchmark over reasonably long periods — no shorter than five years. It should also measure the significance of any outperformance so that investors can judge whether it can reliably be attributed to skill rather than chance. And it should identify a low-cost index fund that most closely resembles the fund’s strategy.
Yes, fund providers will miss their stars. They also won’t appreciate tougher grading or references to lower cost competitors. But as Kapoor acknowledges, as a self-proclaimed “independent voice for the investor,” Morningstar must “sometimes hold views or take stands that run counter to fund companies’ interests and imperatives.”
Some of this will initially be impenetrable to many ordinary investors. But they’ll never be better informed without better information. As long as Morningstar sells shiny stars, it shouldn’t act surprised when investors reach for them.