Our study finds that managers who hop around the style box perform no better or worse than style purists.
In the winter issue of Morningstar Advisor magazine, in an article titled "Fashion Change," we discussed the currently popular belief that so-called flexible U.S. stock fund managers outperform those managers who have more-restrictive investment approaches. This mindset contrasts sharply with the conventional wisdom of a decade ago, when "style-pure" funds were praised for their discipline and their opposites were criticized for being "drifters."
As the article demonstrated, the evidence in favor of flexible managers is scant. Yes, there are several studies that purport to show the superiority of managers who follow flexible investment approaches, but upon closer examination, these studies aren't particularly relevant for advisors who are monitoring funds by how they move among the squares of Morningstar's Style Box. The existing studies either use different definitions of flexibility than style-box consistency, or they are unrealistic, straw-man exercises that do not reflect how Morningstar places funds in the style box.
So, until now it has not been known if flexible managers tend to outgain style-pure managers.
The answer is, they don't. They are not worse overall as a group, nor are they significantly better. They are essentially the same, so say the numbers generated for this publication.
Here's what we did. For all diversified U.S. equity funds (counting only one share class per fund), we tested the three-year period through March 31, 2008. Each fund tested had 12 quarterly portfolios. If all 12 portfolios registered in the same square of the style box, the fund scored as "low flexibility." If one of the portfolios moved to a second square of the style box and did not move again, the fund registered as "moderate flexibility." Finally, if the fund moved twice or more during the three-year period, the fund was called "high flexibility."
Each of the nine style-box squares was sorted by these three flexibility groupings, so that 27 buckets were created in total. Then, average total returns for each bucket over the trailing three-, five-, and 10-year periods were calculated. So, too, were average standard deviations.
The results are pretty straightforward to interpret and can be seen in the accompanying charts. The figures shown represent the margin of out- or underperformance, per annum, by a given flexibility grouping for a given fund category. For example, over the three-year period, the high-flexibility large-growth funds outgained the large-growth average by 13 basis points per year, the moderate-flexibility funds from the same category trailed by 52 basis points, and the low-flexibility funds led by nine basis points.
To assist in interpreting the results, flexibility groupings that performed at least 50 basis points better than each of the other two groups are marked in green. Conversely, groupings that performed at least 50 basis points worse than their peers are marked in red.
Click here for the related image.
This visual pattern tells the story quite clearly --no story. For the trailing three- and five-year periods, high-flexibility funds were marginally better than moderate, and moderate slightly superior to low flexibility. The margin of victory, however, was tiny indeed. For the three-year period, high flexibility had two greens, no reds, and seven uncoloreds among the nine style-box categories, while low flexibility had one green, two reds, and six uncoloreds. The five-year results were similarly tight.
Over the 10-year period, however, the high-flexibility funds reversed course and showed marginally worse results. Moreover, over the entire sample size of the three time periods, the moderate-flexibility group had the greatest number of green results. If flexibility is a true factor (for good or ill), then it is quite odd to see funds with a middle-of-the-road approach to that factor finish the best. Therefore, it seems safe to say that in aggregate the degree of a portfolio manager's flexibility is not relevant to an advisor when determining the merits of his or her fund.
This precept also holds true for the funds' risks. We could publish a similar series of charts for the flexibility groupings' average annualized standard deviations, but that would require using a great deal of space to tell the same story: The level of flexibility really does not matter. Some time periods, for some categories, the high-flexibility funds were the most volatile, other time periods the moderates and lows were. In almost all cases, the margins were very small, well under 1 percentage point of standard deviation.
Neither was there a visible effect on lists of the 10 best- and worst-performing funds. It seems logical that the high-flexibility funds might make up a bigger percentage of both the winners and losers, because flexible managers are taking chances that will get them to stand out, one way or another. The numbers do not demonstrate that effect, however. Indeed, most of the very top winners and very top losers come from the low-flexibility camp. But this finding occurs primarily because there are more low-flexibility funds, thus more candidates for the top and bottom lists. Once this effect is scrubbed, the numbers are, once again, very similar.
But Expenses Make a Difference
There is one area of significant differences: expenses. For several categories, the average expense ratio declines powerfully as one moves from high to moderate to low flexibility. For example, high-flexibility large-blend funds check in with an average expense ratio of 1.08, moderate flexibility registers 0.9, and low flexibility is at 0.71. A sharper case yet is with the mid-growth category, where the expenses cascade south from 1.83 to 1.35 to 1.19 as flexibility moves from high to low.
This occurs for two reasons. First, lower-priced institutional funds are likelier than most funds to score as low flexibility. This is because the institutional marketplace, in general, has been unimpressed with arguments by mutual fund managers that they should have broad mandates. Instead, institutions tend to use funds that are tightly defined. Second, at least in some cases, funds that are highly flexible charge higher fees, because they are perceived (at least by their own managers) as adding more value.
In summary, manager flexibility is an issue of choice for the advisor, not of best practices. Just as it is a matter of taste of whether to select funds that are quantitatively or qualitatively managed; that are headed by teams or by individuals; or whether to invest in boutiques or with major multinational fund families, it is also a matter of taste as to whether you prefer high- or low-flexibility funds.
John Rekenthaler is Morningstar's vice president, research and new product development.