While most mutual funds underperform a simple index (and the vast majority underperform after tax), does that mean one cannot build a metric that predicts fund performance better than random?
I was at the Morningstar ETF conference this past summer and learned a pretty amazing statistic: roughly all inflows into mutual funds go into 4/5 star rated funds or new funds. That was astounding to me. The Morningstar star ratings (background at the bottom of the post) have been measuring past risk-adjusted performance for over two decades. What they DO NOT do is offer any clues to future performance.
Don Phillips, President of Fund Research at Morningstar, stated:
“The star rating is a grade on past performance. It’s an achievement test, not an aptitude test…We never claim that they predict the future.”
Morningstar, quite impressively, actually disclosed a few months ago that simply using expense ratios was a better metric for predicting future performance that their star ratings. “Investors should make expense ratios a primary test in fund selection,” Russel Kinnel, director of mutual fund research at Morningstar, said in an article about the study. “They are still the most dependable predictor of performance. Start by focusing on funds in the cheapest or two cheapest quintiles, and you’ll be on the path to success.” (Older 2007 study here.)
It would be interesting to see Morningstar present this metric on gross and net-of-fee returns to try and isolate the impact of fees (their current ratings are net of fees so naturally include the expense ratio as a factor).
If I was Russ or Don, I would commission a study in house (or possibly with some cheap local U of Chicago PhD’s) or even open it up Netflix style to a competition. There have been numerous studies that illustrate ways in which one can pick mutual funds (maybe call it SuperStars? ha).
I’m sure there are more (email the papers to me and I’ll add them), and some of these probably overlap (ie high fees and low Morningstar ratings). A lot of these factors are successful in selecting hedge fund manages on AlphaClone as well.
Most of these links are from the fantastic blog CXO Advisory. It would be interesting to see a white paper that combines these factors into one metric.
Ways to improve your chances when picking mutual funds:
-Favor new funds. Academic paper here: “Performance and Characteristics of Mutual Fund Starts” Karoui and Meier
–Favor cheap funds. Academic paper here: “Performance and Characteristics of Actively Managed Retail Mutual Funds with Diverse Expense Ratios” Haslem, Baker, and Smith
-Favor funds with higher ownership stakes (manager skin in the game). Academic paper here: “Portfolio Manager Ownership and Fund Performance” Khorana, Servaes, and Wedge
-Favor funds with high “Active Share” (holdings very different from the benchmark). Academic paper here: “How Active is Your Fund Manager? Cremers and Petajisto
–Favor funds with low assets under management. Academic paper here: “How Active is Your Fund Manager? Cremers and Petajisto
–Favor funds with recent momentum. Academic paper here: “How Active is Your Fund Manager? Cremers and Petajisto and here “The 52-Week High, Momentum, and Predicting Mutual Fund Returns” Sapp
–Favor funds with redemption fees. Academic paper here: “Redemption Fees: Reward for Punishment” Nanigan, Finke, Waller
–Avoid funds with low Morningstar Stars. Academic paper here: “Selectivity, Market Timing and the Morningstar Star-Rating System” Antypas, Caporale, Kourogenis, and Pittis
-High conviction picks outperform. Academic paper here: “Best Ideas” Cohen, Polk, Silli
The Morningstar Rating for mutual funds, commonly called the “star rating,” brings both performance and risk together into one evaluation. Morningstar adjusts for risk by calculating a risk penalty for each fund based on “expected utility theory,” a commonly used method of economic analysis. Although the math is complex, the basic concept is relatively straightforward. It assumes that investors are more concerned about a possible poor outcome than an unexpectedly good outcome and that those investors are willing to give up a small portion of an investment’s expected return in exchange for greater certainty. A “risk penalty” is subtracted from each fund’s total return, based on the variation in its month-to-month return during the rating period, with an emphasis on downward variation. The greater the variation, the larger the penalty. If two funds have the exact same return, the one with more variation in its return is given the larger risk penalty.
Funds are ranked within their categories according to their risk-adjusted return (after accounting for all sales charges and expenses), and stars are assigned such that the distribution reflects a classic bell-shaped curve with the largest section in the center. The 10% of funds in each category with the highest risk-adjusted return receive five stars, the next 22.5% receive four stars, the middle 35% receive three stars, the next 22.5% receive two stars, and the bottom 10% receive one star.
Funds are rated for up to three periods–the trailing three, five, and 10 years and ratings are recalculated each month. Funds with less than three years of performance history are not rated. For funds with only three years of performance history, their three-year star ratings will be the same as their overall star ratings. For funds with five-year records, their overall rating will be calculated based on a 60% weighting for the five-year rating and 40% for the three-year rating. For funds with more than a decade of performance, the overall rating will be weighted as 50% for the 10-year rating, 30% for the five-year rating, and 20% for the three-year rating. The star ratings are recalculated monthly.