Outperformance In Theory, Not Practice
Earlier this week, I reviewed several studies on active share and explained why it is not an indicator of outperformance. Today I’ll look at additional studies focusing on active share and fund turnover, as well as whether active share is predictive of future performance in emerging markets.
Active Share And Turnover
One of the more interesting findings on active share comes from co-authors Martijn Cremers and Antti Petajisto in their original 2009 study, “How Active Is Your Fund Manager? A New Measure That Predicts Performance.”
Cremers also co-authored with Ankur Pareek the study “Patient Capital Outperformance: The Investment Skill of High Active Share Managers Who Trade Infrequently.” This 2016 paper looks at the metric of active share from a different perspective, differentiating between those with patient trading strategies (low turnover) and those with active trading strategies (high turnover).
Among the findings in the study, which covered the 19-year period from 1995 through 2013, was that mutual funds that had stock-holding durations of less than two years generally underperform, regardless of active share.
The authors found that after adjusting the benchmark-adjusted net returns for exposure to five factors they considered (market, size, book-to-market, momentum and liquidity), the mutual funds with short fund duration and high active share produced an alpha of about -2% per year, which was statistically significant at the 5% level. Thus, we can conclude that high active share alone isn’t a predictor.
It’s also interesting to note that in footnote 6, the authors state that the findings of the original 2009 paper by Cremers and Petajisto don’t hold up in the out-of-sample period from 2004 through 2013. During this period, the top active share quintile produced a negative alpha of 2%.
Cremers and Pareek did find that “among high active share funds, patiently managed portfolios have been most likely to outperform. Patient funds are those which trade relatively infrequently, i.e., funds with long holding durations or low portfolio turnover.”
For example, they found that after adjusting the benchmark-adjusted net returns for exposure to the same five factors, on an equal-weighted basis the high-active-share, high-fund-duration (longer holder period) mutual funds outperformed by 2.3% per year (t-stat of 3.1). On a value-weighted basis, the alpha was 1.9% (t-stat of 1.9).
However, when fund turnover was used instead of fund duration, the data isn’t as compelling. On an equal-weighted basis, funds in the highest active share quintile and lowest turnover quintile produced an alpha of 0.9% with a t-stat (1.2) that wasn’t close to being statistically significant at the 5% level. On a value-weighted basis, the figures are better, with an alpha of 1.44 and a t-stat of 1.9. A t-stat of 1.96 is the cutoff for being statistically significant at the 5% level.
Betting Against Beta
The authors also found that after accounting for two additional factors, the betting against beta (BaB) factor proposed by Andrea Frazzini and Lasse Pedersen in their “Betting Against Beta” paper published in the January 2014 issue of the Journal of Financial Economics, and the “quality minus junk (QmJ)” factor, proposed by Clifford Asness, Andrea Frazzini and Lasse Pedersen in their 2013 working paper, “Quality Minus Junk,” most of the outperformance of the patient and high active share managers of mutual funds can be explained—what was alpha in the five-factor model had been converted into beta in the seven-factor model.
This finding led the authors to conclude that the managerial skill from the most active and patient mutual fund managers is quite similar to that manifested by Warren Buffett, as documented by Andrea Frazzini, David Kabiller and Lasse Pedersen in their 2013 working paper, “Buffett’s Alpha.” The trio documented that Berkshire Hathaway, managed by Warren Buffett, as well as Berkshire’s public equity holdings identified from quarterly 13F filings, have significant positive alphas that become insignificant when controlling for their exposure to BaB and QmJ.
Cremers and Pareek state: “Specifically, the clear majority of the outperformance of the patient and active mutual fund managers seems due to their picking safe (low beta), value (high book-to-market) and quality (profitable, growing, less valuation uncertainty, higher payout) stocks, and then sticking with their convictions and holding on to those over relatively long periods.”
They concluded: “Our results thus indicate that Warren Buffett’s skill seems generally shared by mutual fund managers in the top Active Share and Fund Duration quintiles.” (Note that Cremers and Pareek did not find this explanation held for institutional funds and hedge funds with high active share and low turnover, which also outperformed—the alpha remained when using a seven-factor model.)
As mentioned above and as explained in my book, coauthored with Andrew Berkin, “The Incredible Shrinking Alpha: And What You Can Do to Escape Its Clutches,” when academic research uncovers additional factors that add further power in terms of explaining the differences in returns among diversified portfolios, it converts what was once alpha into beta (loading on a systematic factor such as size, value, momentum, profitability, etc.).
Of course, this shouldn’t detract from how we view the performance of the active managers who benefited from exposure to these factors prior to the research being published. After all, they had employed these strategies before factors were added to standard asset pricing models. However, now many low-cost funds provide exposure to these common factors in a systematic way and do so in a tax-efficient manner.
Active Share In Emerging Markets
Aron Gottesman and Matthew Morey, authors of the January 2016 study “Active Share and Emerging Market Equity Funds,” examined whether active share was predictive of future performance in emerging markets over the six-year period from 2009 through 2014. The study included 67 actively managed emerging markets funds. Following is a summary of their findings:
- The most active quintile of funds had active share of at least 78.3%. The average active share was 83.2% and the highest was 94.5%. For the middle quintile, the figures were 69.2%, 71.2% and 72.3%, respectively. For the lowest quintile, the figures were 61.3%, 49.1% and 23.6%, respectively.
- There’s a positive, and significant, relationship between the average level of a fund’s active share and fund performance. More active funds have significantly better performance than less active funds.
- A fund’s standard deviation of active share is a negative, and significant, predictor of fund performance. Highly active funds that change their active share over time have significantly worse performance than funds that keep their level of activeness consistent over time.
- Expenses are negatively and significantly related to fund performance.
- Active share and expenses are positively correlated. Funds with higher active share are more expensive.
- Because fees for actively managed emerging market funds are significantly higher than for passively managed ones, active managers must overcome a high hurdle. Thus, closet indexing (funds with low active share) in emerging market equity funds robs investors of the opportunity to outperform. About 16% of the funds studied were closet indexers (active share of 60% or less).
Gottesman and Morey reached the following conclusions: “We have found that the level of activeness matters for performance. More active funds outperform. In addition we have found that the consistency of the fund’s activeness matters also for performance. Funds that keep their active share consistent over time outperform funds who attempt to alter their active share over time.”
Before you draw any conclusions, however, let’s review the authors’ findings on returns. They ranked funds by active share and placed them into quintiles. The table below shows the monthly returns of those quintiles over the period 2009 through 2014.
While the two most active quintiles had the highest returns, the least active quintile outperformed the middle quintile and second-least-active quintile. If active share were predictive, that shouldn’t be the case.
We can now compare the return of the most active quintile with the returns of three emerging market benchmark indices provided by Dimensional Fund Advisors (DFA):
First, note that all three DFA indices outperformed the most active quintile. We can also note that small and value stocks outperformed during this period. Thus, the relatively strong performance of the top quintile might be explained not by stock selection skills but by their exposure to the size and value factors.
Another reason to be cautious about drawing conclusions from the study by Gottesman and Morey is that it covered only a brief, six-year period and only 67 mutual funds. Thus, each quintile held only 15 stocks. That’s a very small sample from which to draw strong conclusions.
Based on these facts, at least, it seems hard to make a compelling case that active share is a predictor of future outperformance. One thing we can say is that unless you have funds in the highest active share quintile, your odds of outperforming are poor. That’s surely a loser’s game. Additionally, given the relatively narrow outperformance of the top quintile, it seems unlikely that taxable investors could benefit from using active share even as a predictor of future performance.
Success Sows The Seeds Of Its Own Destruction
Another issue to consider regarding active share is that, if it is predictive, it sows the seeds of its own destruction. Here’s why. Jonathan Berk, in his paper, “Five Myths of Active Portfolio Management,” suggested the following: “Who gets money to manage? Well, since investors know who the skilled managers are, money will flow to the best manager first. Eventually, this manager will receive so much money that it will impact his ability to generate superior returns and his expected return will be driven down to the second best manager’s expected return.
At that point investors will be indifferent to investing with either manager and so funds will flow to both managers until their expected returns are driven down to the third best manager. This process will continue until the expected return of investing with any manager is the benchmark expected return—the return investors can expect to receive by investing in a passive strategy of similar riskiness. At that point investors are indifferent between investing with active managers or just indexing and an equilibrium is achieved.”
Berk went on to point out that the manager with the most skill ends up with the most money. He added: “When capital is supplied competitively by investors but ability is scarce only participants with the skill in short supply can earn economic rents. Investors who choose to invest with active managers cannot expect to receive positive excess returns on a risk-adjusted basis. If they did, there would be an excess supply of capital to those managers.”
Active Share Can Fall
There’s yet another reason why active management (and active share as a useful tool) sows the seeds of its own destruction. As a fund’s assets increase, either trading costs will rise, or the fund will have to diversify across more securities to limit trading costs. In other words, its active share will fall.
However, the more a fund diversifies, the more it looks and performs like its benchmark index, becoming the dreaded closet index fund. And we saw the results from closet index funds in the study by Gottesman and Morey.
Also keep in mind that the publication of papers about active share spreads the word and thus speeds up the process that Berk described. In addition, the publication of papers can impact the behavior of fund managers who want to attract more assets.
Knowing that investors are looking at active share as a measure of skill, they may reduce the number of stocks in their portfolio to look more skilled. For investors interested in how the publication of academic research impacts return predictability, I suggest reading the 2015 paper by R. David McLean and Jeffrey Pontiff, “Does Academic Research Destroy Stock Return Predictability?”
The research has found that the original findings on active share have not held up out of sample. It’s also been found that high tracking error funds don’t outperform low tracking error funds once exposure to common factors was taken into account.
In addition, it was found that only the high active share and low turnover funds outperformed. High active share and high turnover actually produced large and statistically significant negative alphas.
And unfortunately, once we account for the additional factors of BAB and QMJ, most of the alpha of the high active share/low turnover funds could be explained by the seven common factors. And of course, once a successful fund experiences the inevitable cash inflows, the hurdles to generating alpha become significantly greater.
Finally, in the very asset class that one would expect active share to perform best (emerging markets—it’s supposedly an inefficient asset class), there’s little to no evidence it has outperformed well-designed, passively managed structured portfolios. It doesn’t seem that much, if anything, is there.
Of course, this doesn’t mean there isn’t skill in active management. It just means that, given the efficiency of the market (due to the high skill level of the competition), it’s very difficult to generate alpha after expenses.
And as Andrew Berkin and I point out in our book, “The Incredible Shrinking Alpha: And What You Can Do to Escape Its Clutches,” the competition is getting tougher by the day. In addition, as academic research continues to convert what was once alpha into something better (exposure to a common factor), the sources of alpha are drying up.
The bottom line is, the only thing that has changed since Charles Ellis wrote his book in 1998, “Winning the Loser’s Game: Timeless Strategies for Successful Investing,” is that active management has become even more of a loser’s game. There’s no holy grail that will allow you to improve the odds sufficiently enough that it becomes a game that is prudent to play—unless you place a high value on the entertainment.
This commentary originally appeared April 26 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