Active share is hailed by some to be the holy grail in terms of its ability to identify future fund performance. My posts of April 24 and April 26 presented the evidence demonstrating that active share is likely another example of the triumph of hype and hope over facts.
Today we’ll look at two out-of-sample tests of the ability of active share to predict future fund performance. Out-of-sample tests are important, as their results reduce the risk that in-sample outcomes are random (a result of data mining).
A Canadian Study
The April/May 2017 issue of Morningstar presented the findings of its study “Active Share Doesn’t Live Up to Hype: An Unreliable Way to Forecast Winning Canadian Equity Funds.” The study covered the period January 2001 through December 2015.
Morningstar tested active share’s predictive power in two 10-year periods—2001-2010 and 2006-2015—dividing each 10-year period into two five-year periods. The first five years are the in-sample period used to calculate active share, and the second five years are the out-of-sample period.
Morningstar found virtually no differences in returns once exposure to the common factors of market beta, size, value and momentum were considered: “After accounting for those factors, funds delivered little to no alpha across active share levels.” The researchers also found virtually no difference when they examined the worst-case drawdowns.
Morningstar also noted that “higher active share many not ensure better results, but it is likely to lead to more extreme ones … The distribution of performance outcomes widened as active share rose.” In other words, investors were taking more risk of large losses (and investors are, in general, risk averse) without being compensated with higher returns—demonstrating that active share isn’t a totally useless measure after all.
Morningstar provided another out-of-sample test with its March 2016 study “Active Share in European Equity Funds: The Activeness of Large-Cap European Fund Managers Through the Lens of Active Share.”
Morningstar’s database is a set of large-cap European funds investing in European equities and encompasses the period January 2005 through June 2015. The researchers chose to include only large-cap funds to reduce the difficulties arising from benchmark selection and the impact of the small-cap effect.
They found that “funds with higher active share have delivered better investment results than the least active funds in most of our research period, but not unambiguously. Because dispersions in returns and risk characteristics become much wider as a portfolio’s active share rises towards 100%, investors should not rely solely on active share when selecting funds.”
However, investors should not care whether high active share funds outperform funds with low active share. What they should care about is whether high active share funds generate alpha, after adjusting for exposure to the common factors of market beta, size, value and momentum. After all, that’s what you’re supposed to be paying those high fees for—alpha.
Unfortunately, over the period 2006 through 2010, while funds with active shares above 90 outperformed those with active shares of less than 60, the high active funds generated monthly alphas of -0.14 (versus -0.24). And the first quartile of active share (funds with the highest active share) had monthly alphas of -0.10 (versus -0.26 for the fourth quartile funds).
The results were even worse over the second five-year period from 2011 through 2015, as the quartile of funds with the highest active share had monthly alphas of -0.25 versus -0.22 for the quartile with the lowest active share. And for funds with active share of above 90, the monthly alpha was -0.44 compared with -0.21 for funds with active shares below 60.
Morningstar also examined the relationship of risk and active share by examining the worst-case drawdowns and concluded: “The highest active share bucket clearly suffered the largest maximum drawdowns. For investors, this means that selecting a fund with a higher active share increases the possibility of experiencing larger losses.”
The bottom line is that active share has not been a predictor of fund performance for Canadian and European equity funds. If active share worked as a predictor, it should work everywhere we look—as is the case with the size, value and momentum premiums, which work virtually everywhere we look.
These results provide strong out-of-sample tests of the inability of active share to predict future fund performance. In other words, active share appears to be another case of hype over facts.
This commentary originally appeared April 28 on ETF.com
By clicking on any of the links above, you acknowledge that they are solely for your convenience, and do not necessarily imply any affiliations, sponsorships, endorsements or representations whatsoever by us regarding third-party Web sites. We are not responsible for the content, availability or privacy policies of these sites, and shall not be responsible or liable for any information, opinions, advice, products or services available on or through them.
The opinions expressed by featured authors are their own and may not accurately reflect those of the BAM ALLIANCE. This article is for general information only and is not intended to serve as specific financial, accounting or tax advice.
© 2017, The BAM ALLIANCE