The Zero Alpha Group
is making a name for itself by taking on Morningstar. Wednesday, the advisor group put forth its evidence that Morningstar fund performance data systematically overstates returns for actively managed funds. For its part, Morningstar downplayed the criticism and said that it has been working on a fix.
The Zero Alpha Group (not to be confused with the more prominent Alpha Group), is a network of advisors that manages nearly $5 billion in assets. The firms in the network have adopted passive investment strategies for their clients.
At the heart of ZAGs critique of Morningstar's fund performance numbers is the effect of "survivorship bias" on the fund trackers database.
ZAG member Savant Capital Management
released the study
examining Morningstar figures that tracked the performance of actively managed funds over the ten-year period from 1995 to 2004. The study claims data for this up-and-down period "systematically and significantly" overstated the returns posted by mutual funds in all but one of Morningstar's 42 categories.
The "survivor bias" results from merged, folded, or otherwise "dead" funds that Morningstar removed from the database, while stronger performers are retained. According to ZAG's analysis, discounting weak links in this way boosted apparent returns by an average of 1.6 percent per year in the period studied.
Survivor bias is not a new concept within the industry, but with the release of this study, ZAG members have signaled an intention to raise awareness of the issue among the broader population.
"We hope that continued focus by [the media] will educate investors, and will promote some pressure; that more thorough analysis is done by a number of groups, so that investors can make wise decisions," Glenn Kautt
, president, chairman, and chief executive officer of The Monitor Group
in McLean, Virginia -- the newest ZAG firm -- said during the call.
"The vast majority of investors are completely clueless as to the existence of this problem," remarked Pat Beaird
, founder and president of Dallas-based BHCO Capital Management
. He added that the report "will allow us to take [investor awareness] to the next level." Amy Barrett
, co-author of the report and a Savant portfolio manager, expressed the hope that a new system of managing data can be created in the future.
In fact, Morningstar managing director Don Phillips
told the MFWire
that "survivorship-adjusted category averages" are in the pipeline, having been proposed in a corporate initiative earlier this year.
"We're working through some of the methodology questions," he said.
However, he is unsure such figures will have broad relevance for investors. "The way we're calculating is really more germane to the questions most investors are asking," he said. He likened Morningstar's existing system to the way a marathon operator might average the results of all race finishers. In that analogy, ZAG is advocating a more complex system in which the race is divided into segments and averages are calculated piecemeal, to account for runners who dropped out.
"Despite all the hyperbolic language they used .. it's simply a matter of how you calculate the category average," said Phillips.
Indeed, the strong words of the ZAG report's sponsors are sure to generate some attention-grabbing headlines.
"Active managers don't have a leg to stand on," contended Beaird, pointing out that only small growth managers outperformed their index after survivor bias was accounted for, and then only in the bull market of 1995 to 1999. "This analysis is the first to strongly suggest that Morningstar data -- the main data source relied upon by individual investors and many financial professionals -- effectively overstates real historical returns, and does so on a significant basis."
Yet the misleading nature of data "is not so much a Morningstar issue as it is an industry issue," Beaird later added. Data from all corners, he charged, is "contaminated" with survivor bias.
"They do sort of hint at this vast, industry-wide conspiracy," said Phillips, but added this interpretation does not make sense. Reintroducing dead or damaged funds to the mix would allow lower-than-average funds to claim to be average, he said, meaning the calculations ZAG advocates would if anything benefit the fund industry.
Interestingly, the bias is not great enough to make actively managed funds appear even equal to indexed ones. In the words of Brent Brodeski
, Savant managing director and the report's other co-author, the numbers only work to make actively managed funds appear "somewhat competitive with indexing."
"In general, the managers failed regardless. When you add the bad managers in, they failed by a larger margin," he said.
Asked why he believes Morningstar, or competing tracking firms, would resist eliminating survivor bias, Brodeski replied that "data providers, frankly, are in the business of allowing fund managers to not only track their data but also to provide data in the format that can be used to encourage the sales of funds and the purchase of funds." Though he praised the services Morningstar provides, he observed that data free of survivor bias "may or may not further the cause of the fund industry in general."
Phillips agrees that the survivorship-adjusted averages are useful in comparing active to passive aggregate results within a category, but suggested such a comparison serves narrow interests.
"To capture the investor experience," he said, ZAG might have called for asset-weighted averages -- which would be "less flattering" to index funds.
During the conference call, the ZAG member representatives were also asked whether they have seen their investors pushing for the option of actively managed funds. Beaird was emphatic that BHCO has not.
"We have received no pressure whatsoever from clients to depart from an indexing strategy," he said, adding that his firm is always concerned with questioning the appropriateness of benchmarks.
Stay ahead of the news ... Sign up for our email alerts now