Morningstar responds to the various factual inaccuracies with New Constructs’ article titled, “Morningstar’s Dominance Harms Investors“, posted on ValueWalk last week. See below for a response from MorningStar on several of the points discussed in the article.
Backward looking fund ratings can only tell you where a fund has been, not where it’s going. The WSJ report found that Morningstar’s Star Rating system is not a good guide to a fund’s future performance. In fact, of the funds with a five-star rating, only 14% had performed at a five-star level over the three years measured by WSJ.
Morningstar’s own research in 2010 came to a similar conclusion regarding the predictive success of their ratings. In the report, Morningstar found that in 58% of the funds measured, the expense ratio, not the star-rating, had a higher success rate in predicting performance.
The implication that we have positioned the star rating as “technical research” that is “enough to fulfill fiduciary duties” is false.
We recognize and have often acknowledged the limitations of a measure like the star rating that’s based on past performance. But we also believe it can usefully tilt the odds in investors’ favor, especially when combined with other research and tools.
We have consistently described the rating as a potentially useful starting point for research.
Our research—which we shared in our response to the Wall Street Journal—finds that high-rated funds have been much more likely to succeed than low-rated funds while low-rated funds fail far more often. (These findings are based on the Journal’s own data.) What’s more, other analysis we’ve conducted has found the star rating points toward funds that are cheaper and easier to own, qualities that correspond with future investor success. (This is buttressed by a formal study we conducted into the star rating’s performance globally; we published it last year.)
With respect to the percentage of funds that achieved a five-star rating in the ensuing years, note that a fund chosen at random would have a roughly 10% chance of earning a five-star rating over the next three years, meaning that five-star funds are 40% more likely to achieve such an outcome than an arbitrarily chosen fund. Conversely, low-rated funds were far likelier to fail than a fund chosen at random.
We agree that our research has found expenses are important to future success, explaining why we base the star rating on a fund’s net-of-fee performance. What’s more, we adjust for the level of risk a fund has taken in amassing those results, so as to ensure that we’re not unduly rewarding or punishing funds based on their raw performance alone.
2. Morningstar’s research is conflicted.
Morningstar derives a significant amount of its revenue (73% of revenue from its “license-based” segment in 3Q17) from asset managers, both from selling data and collecting licensing fees to advertise Morningstar ratings. Not surprisingly, just like Wall Street’ stock ratings, Morningstar’s fund ratings are skewed toward more positive ratings. A study in 2014 found that only 3% of funds receive a Morningstar “Negative” analyst rating. Assigning negative ratings means fewer funds want to pay to advertise the rating. In addition, poor ratings can damage Morningstar’s relationship with that fund manager. The bottom line is that it appears Morningstar has material incentives to provide positive ratings regardless of how well those ratings serve investors.
This is false.
We publish the ratings because we believe they have investment merit, not for any financial gain.
The author’s assertions about the sources and nature of our license-based revenue are misleading and false. Here are the facts:
- Our license-based revenue in 2016 was $573 million, which is about 70% of our total revenue of $798.6. License-based revenue includes far more than the star ratings. License-based revenue includes Morningstar Data, Morningstar Direct, Morningstar Advisor Workstation, Morningstar Enterprise Components, Morningstar Research, and other similar products.
- The bulk of our licensed-data revenue comes from our software. In publicly available financial reports, we don’t break out the revenue that we derive from licensing our ratings, as this revenue isn’t material to our overall results
The star rating is a quantitative, backward-looking measure. There is no subjective input or overlay. What’s more, we issue as many 1- and 2-star ratings as we do 4- and 5-star ratings (as the ratings follow a bell-shaped curve).
The author is conflating matters by juxtaposing our licensing revenue with the distribution of the ratings that our analysts subjectively assign. Our analysts are walled off from the commercial side of Morningstar and assign ratings objectively on the merits. It is true that we assign relatively few “Negative” ratings, but this reflects the self-selecting nature of our coverages (i.e., users of our research are interested more in worthy ideas than demonstrably bad ones), and ignores the fact that we assign “Neutral” ratings to hundreds of funds that we cover (3,040 of 7,310, as of October 2017).
We also recently introduced a quantitatively-driven version of the Analyst Ratings which enlarges our rated universe to include funds that analysts don’t cover. Among the 19,090 funds that receive quantitative ratings, 15,076 were rated Negative or Neutral as of October 2017.
All told, 18,855 of the 27,222 funds in our database carry at least one low rating (i.e., a Neutral or Negative Analyst Rating, a Neutral or Negative quantitative rating, or a 1- or 2-star rating). In fact, we issue more low ratings than anyone on the Street.
The suggestion that we assign positive ratings to curry favor with fund companies is also unsubstantiated and false. We assign ratings on the merits, not for commercial benefit.
3. Morningstar’s research creates conflicts with best interests of investors.
Managers spend large sums of money and dedicate senior personnel to keep their fund ratings high. In 2014, Financial Times reported that PIMCO founder Bill Gross spent months lobbying Morningstar analysts to avoid a downgrade on PIMCO’s flagship fund. Additionally, fund managers hire executives specifically to manage relations with Morningstar. These executive salaries can translate into higher fees for investors.
Managers also pursue suboptimal performance strategies in order to cater to the Morningstar rating system. David Blanchett found in his 2011 paper, “Gaming the system: the impact of Morningstar category changes on peer rankings”, that underperforming managers will “drift” into different categories where their relative ranking looks better. These “drifting” funds saw significant inflows after the switch, but went on to underperform their new peers
This is false. We make coverage decisions independently, based on a variety of factors, including stakeholder relevance and investment merit.
It’s true that we upgrade and downgrade ratings after meeting with fund companies’ senior leaders and portfolio managers. There’s no correlation because access to senior leaders, by itself, doesn’t ensure a positive rating. Rather, our research incorporates a host of other factors as well. We welcome the opportunity to interact with senior fund company leaders, as it can deepen our understanding of organizational priorities and, in turn, yield insights into how the firm defines its core capabilities and investments it’s making in those areas. But it’s the sum total of our research—encompassing people, process, parent, performance, and price—that determines whether we upgrade or downgrade a fund’s rating, not a meeting with a senior leader.
We have no say on how organizations opt to work with our analysts. Some assign personnel to work with us, though usually these professionals serve as points of contact with other firms as well, such as investment consultants or brokerage platforms. There is no set approach to working with us and we recommend the funds of big and small fund companies alike, irrespective of whether they assign an individual to work with us or not.
David Blanchett’s 2011 study of fund category changes among U.S. equity funds focused primarily on subsequent changes in percentile ranking. Blanchett found evidence that ranks improved slightly following category change. However, when we examine the star ratings of US equity funds for the period July 2002 to October 2017, we find that rated funds that changed categories were much likelier to see their rating remain unchanged from one month (i.e., the prior month in the old category) to the next (i.e., the subsequent month in the new category), than they were to experience a ratings change. What’s more, among rated funds that changed category and experienced a ratings change from one month to the next, they were almost as likely to see a downgrade as an upgrade. We repeated this study covering the 12-month period following a month-to-month category change and the results held—funds were likelier to experience no ratings change than to see a change, and of those that saw a change there was a roughly equal likelihood of receiving an upgrade as a downgrade. Thus, contrary to the article’s assertion that the rating causes managers to “pursue suboptimal performance strategies”, the data suggests that category changes do not strongly influence ratings outcomes.
4. Morningstar ratings are NOT a substitute for diligence.
As noted in the WSJ report, Morningstar ratings are used as a safety net. “It’s a cover-your-ass type of service,” says Samuel Lee, a former strategist at Morningstar. “An adviser can say, ‘I’m going to put you in this fund, it’s a 5-star fund,’ …and if something goes wrong the adviser can shunt blame to Morningstar.” At the same time, Morningstar ratings can lure advisors into making decisions, not because they’re right, but because they’ll receive the least backlash. One former adviser quoted in the WSJ piece noted “Advisers get in trouble when they go against the grain. You isolate yourself more if you sell something else rather than just go with what research recommends."
Contrary to what’s implied, we have never suggested the ratings are a “substitute for diligence”.
With respect to the assertion that the ratings are used as a safety net, while we have taken pains to educate investors and their advisors on the best ways to apply the star rating in practice, some still take a short-cut approach.
The Journal claims that we sit idly by while investors and advisors misuse the star rating as if it was predictive, but this is false. We have always said the star rating is a starting point for research and therefore shouldn’t be the sole basis for assessing a fund. Despite that, some advisors use it that way, a practice that shortchanges clients, who are denied the benefit of a more thorough research and fund selection process.
Misuse of data points is hardly limited to the star rating, which is why we routinely examine how data points are being used in practice and whether there are ways we can help investors better succeed with them.
5. Morningstar ratings affect fund inflows/outflows.
Morningstar doesn’t only affect fund manager decisions, but also investors’ decisions. As reported in the paper “Star Power: The Effect of Morningstar Ratings on Mutual Fund Flow“ as well as WSJ’s recent article, investors pour money into top rated funds, even if their performance declines. On the flip side, investors pull money from low rated funds, even if their performance improves. The WSJ found more than a dozen cases where well-performing funds attracted few investors until they received a five-star rating.
1.The Buffalo Emerging Opportunities Fund (BUFOX) saw its assets under management quadruple in the five months after it received a five-star rating. Two years later the fund’s rating fell to two-star’s and its assets plunged to less than $100 million (from a high of $400 million).
2. The Hodges Small Cap Fund (HDPSX) had less than $100 million in assets prior to receiving a five-star rating in 2011. The fund manager paid Morningstar more than $10,000 to advertise its five-star rating, and by the end of 2014, the fund had reached $1.6 billion in assets. The fund currently earns a one-star rating and has just over $700 million in assets.
We acknowledge that the Morningstar rating™ for funds (the star rating) can influence demand, but it’s an overstatement to say the ratings alone have such an outsized impact, as the star rating corresponds with general investing attributes that drive investor demand. For example, investors have shown a strong preference for lower-cost funds that are easier to own and likelier to succeed, and our ratings tend to reward funds that possess those same attributes.