How Morningstar Categorizes Funds

Case Study
From a letter to Mutual Fund Observer--

"Osterweis (OSTFX) is a mid-cap blend fund, according to M*. But don't say that to John Osterweis. Even looking at the style map, you can see that the fund covers all of the style boxes, and it has about 20% in foreign stocks, with 8% in emerging countries. John would tell you that he has never managed the fund to a style box. In truth he is style box agnostic. He is looking for great companies to buy at a discount. Yet M* compares the fund with others that are VERY different."

What we have here is a failure to communicate.

I've written about this subject previously, but not very clearly. This effort, I hope, will be more direct.

1) "Osterweis (OSTFX) is a mid-cap blend fund, according to M*."
Immediately comes the critical point.

Morningstar does not say what a fund is. Morningstar assigns funds to categories. The two activities are quite different.

Consider  Vanguard Managed Payout Growth & Distribution . The fund invests a globally diversified portfolio and promises to distribute 4% of its assets per year. That fund could be placed into the world-allocation Morningstar Category, which consists of funds with similar asset-allocation schemes but no income targets, or into the retirement-income category, which consists of funds that invest less in equities but that are similarly focused on income.

Neither fit is good. The fund is not a world-allocation fund, nor is it a conventional retirement-income fund. It is something different, one of a very small cadre of funds that promises a managed payout--a cadre at this stage too small to warrant its own category. Whatever choice Morningstar makes (retirement income, as it turned out), it is not because Morningstar is confused and believes the fund to be something other than what it is. It is because the fund must go somewhere.  

Similarly, Osterweis is not a mid-cap blend fund. It is assigned to the mid-cap blend category.

2) "But don't say that to John Osterweis."
Politely, the point of the Morningstar Category system is to avoid the views of portfolio managers. Different people use language in different ways. The Morningstar system, as with any independent approach that has the goal of comparability, is to impose a single consistent framework. It might be a dumb framework--but that is another matter.

3) "Even looking at the style map, you can see that the fund covers all the style boxes, and it has about 20% in foreign stocks, with 8% in emerging countries."
Both true. Most U.S. stock funds sprawl across multiple squares of the Morningstar Style Box, but Osterweis is particularly catholic in its approach. The Morningstar Ownership Zone of where 75% of its stock assets fall covers nearly the entire grid.

  - source: Morningstar Analysts

So, too, does the Ownership Zone of one of Osterweis' category peers,  Fidelity Low-Priced Stock (FLPSX).

  - source: Morningstar Analysts

Clearly, if Morningstar is to remove Osterweis from the mid-cap blend group, it should do the same for Fidelity Low-Priced Stock. 

Do the two funds belong in the same new category? Unlike the Osterweis fund, Fidelity Low-Priced Stock has a mandate that demands that it purchases stocks that cost less than $35 per share. In addition, while Osterweis has 20% of its equity assets in foreign securities, Fidelity Low-Priced Stock has 40%.

Seemingly, if Morningstar follows the advice of the letter writer, it cannot put the two funds into the same group. After all, if 20% in foreign securities prohibits accurate comparisons between Osterweis and a conventional mid-cap blend fund, then by the same coin Fidelity Low-Priced Stock's 40% stake prevents it from being compared with Osterweis. 

Should Morningstar then create an initial flexible fund category for U.S. stock funds that have little foreign-stock exposure, a second category for those from 10% to 30%, and a third for those with more than 30%? What about flexible funds that are mostly value or growth in their approach? Should there be nine such categories, for the three levels of foreign exposure and the three investment styles?

Perhaps. There's no question that Morningstar would improve the accuracy of peer-group comparisons if it had 156 categories rather than 106. Better yet would be 206, and better still would be 1,006. That proliferation, of course, would come at the cost of user confusion and the removal of comparisons between funds that are largely alike but which are separated by the ever-finer category system.

It is a trade-off. That the author disagrees with where Morningstar drew its lines is fine, but that the trade-off exists must be acknowledged.

4) "John would tell you that he has never managed the fund to a style box. In truth he is style box agnostic. He is looking for great companies to buy at a discount."
None of these affect the fund's category assignment.

5) "Yet M* compares the fund with others that are VERY different."
The payoff.

As it turns out, Osterweis has performed very much like the typical mid-cap blend fund in seven of the past 10 calendar years. It was somewhat more conservatively positioned in the middle of the period, so that it showed relatively well during the 2008 market crash before giving back those relative gains during the next two years' rally, but otherwise it has been almost spot on the category average.

Here is the table demonstrating that:

And here is the picture. Except for the previously mentioned three-year stretch, the orange and blue lines hug:

  - source: Morningstar Analysts

Finding a better home for Osterweis is not an easy task. For one, its current home isn't bad. For another, the group of self-identified flexible U.S. stock funds that call themselves "all-cap" is not an improvement. 

The year-by-year performance comparison of Osterweis versus all domestic-stock funds that label themselves as "all-cap":

That table looks very much like the previous table. Once again, the main gap occurs from 2008 to 2010. From the perspective of performance, then, the major difference between Osterweis and other domestic-stock funds would not seem to be either its broad-ranging investment style and its use of foreign securities but rather its conservative positioning for a three-year stretch. It's difficult to see creating an investment category around that attribute.

In summary: Showing that a fund differs in key aspects from those in its investment category is not enough to justify moving that fund or changing the system. Neither are references to the manager's intent or appeals from funds that perform much like its existing category average.

A sufficient critique is one that comes from a fund that truly does not behave like others in its category, that contains a proposal for a modification to the existing category system, that does not lead to rampant category proliferation, and that results in a significantly closer performance comparison between the fund and its new category. In such cases, Morningstar will consider the request carefully--and sometimes make the suggested change. 

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.