By LARRY SWEDROE
For investors in actively managed funds, the holy grail is finding the fund manager who will in the future be able to exploit market mispricings — buying undervalued stocks and perhaps shorting those that are overvalued. While it’s easy to identify those great-performing managers after the fact, there’s no evidence of the ability to do so before the fact.
For example, there are dozens, if not hundreds, of studies confirming that past performance is a poor predictor of the future performance of active managers, which is why the SEC requires that familiar disclaimer. These studies find that, beyond a year, there is little evidence of performance persistence (and that persistence is explained by momentum). The only place we find longer-term persistence of performance (beyond what we would randomly expect) is at the very bottom — poorly performing funds tend to repeat. And the persistence of poor performance isn’t due to poor stock selection; it is due to high expenses.
Given the failure of past performance as a stand-alone predictor, researchers have continued the quest for the holy grail — a number of studies have investigated the ability of various theoretically or intuitively motivated variables to predict future fund alphas, and a few showed some promise. One example of a potential predictor that has been thoroughly investigated is active share — a measure of how much a fund’s holdings deviate from its benchmark index. The original study on active share by Martijn Cremers and Antti Petajisto, How Active Is Your Fund Manager: A New Measure That Predicts Performance, was published in the September 2009 issue of The Review of Financial Studies. The authors found that active share did provide information on future performance. Unfortunately, the out-of-sample evidence did not support the original finding. That led to further efforts to determine if active share in combination with other metrics, such as turnover and past performance, could provide information on future returns. However, none of the follow-up studies have found active share (alone or in combination with other metrics) to have predictive value for mutual fund performance, including the study Active Share and the Predictability of the Performance of Separate Accounts (Cremers was one of the authors), published in the First Quarter 2022 issue of the Financial Analysts Journal.
As my co-author Andrew Berkin and I explained in our 2020 book The Incredible Shrinking Alpha, several trends have led to ever-increasing difficulty for mutual funds to generate alpha:
- Academic research has been converting what once was alpha into beta (exposure to factors in which one can systematically invest, such as value, size, momentum and profitability/quality). Investors can access those new betas through low-cost vehicles such as index mutual funds and ETFs. And as David McLean and Jeffrey Pontiff demonstrated in their 2016 study Does Academic Research Destroy Stock Return Predictability?, the publication of research leads to reduced factor premiums.
- The pool of victims that can be exploited is persistently shrinking. Retail investors (naive or “dumb money”), whose share of the market was about 90 percent in 1945, now account for only about 10 percent of all trading.
- The amount of money chasing alpha has dramatically increased. Twenty-five years ago, approximately $300 billion of assets was invested in hedge funds. Today that figure has increased to more than $5 trillion.
- The costs of trading have fallen sharply, making it easier to arbitrage away anomalies.
- The absolute level of skill among fund managers has increased (the paradox of skill problem — as people become better at an activity, the difference between the best and the average and the best and the worst becomes much narrower).
Evidence that the ability to generate alpha was shrinking was found by S&P Director of Global Research & Design Berlinda Liu in her 2018 study Does Performance Persistence of Active Managers Vary Over Time? Liu examined past Active versus Passive Scorecards and concluded: “A review of the performance persistence figures over time shows a downward trend over the longer-term horizon for equity funds, indicating an increasing difficulty to stay at the top.”
Christopher Jones and Haitao Mo contribute to the literature on mutual fund performance with their study Out-of-Sample Performance of Mutual Fund Predictors, published in the January 2021 issue of The Review of Financial Studies, in which they analysed the out-of-sample performance of various variables that research had shown to provide possible information on future mutual fund alphas. They assembled a comprehensive sample of mutual fund predictors by examining articles from 11 of the leading finance and economic journals published between 1960 and 2015.
The authors began by noting: “There are a number of reasons why the ability of a variable to predict out of sample may be different from its ability in sample. Data snooping, perhaps resulting from journals favouring statistically significant results, will lead to an upward bias in in-sample predictive performance, which will naturally decline out of sample.” And like the arguments Berkin and I made in our book, they noted that market conditions can differ as the skill level of the competition increases, the number of funds chasing the same styles (perhaps after publication of research) increases, and the amount of arbitrage capital increases: “Such conditions may result in a difference in the ability of investors to find non-zero alphas.”
Among the many predictors Jones and Mo examined were past performance, measures of skill, cash holdings, active share, holdings-based analysis, fund size, fund flows and expenses (fees and turnover costs). Alphas were measured against the Fama-French three-factor (beta, size and value) model and Carhart’s four-factor (adding momentum) model. Following is a summary of their findings:
- Alpha predictors largely fail, out of sample, to replicate their in-sample success — at least half of the alpha spread generated by predictors proposed in the literature disappears post-sample. In some specifications, the decline was significantly higher.
- The post-sample reduction in alphas was statistically significant in all specifications.
- The declines in alphas were primarily the result of changes in the level of arbitrage activity (as measured by the amount of hedge fund assets under management, the level of aggregate short interest and aggregate share turnover) in the market.
- Mutual fund competition adds explanatory power. All four measures of competition (aggregate industry size, sector size, number of peers in the same investment style and total similarity of holdings) were negatively related to alpha generation.
- There was little evidence that the declines were the result of data snooping, as pre-sample effects were larger than in-sample effects. This undermines data mining as a potential explanation for poor out-of-sample performance, implying that the correct explanation is likely to involve a gradual decrease in predictability over time rather than one that occurs right at the start or end dates of the sample periods of the original studies.
- Corporate bond fund performance exhibits similar dependence on measures of bond market arbitrage activity—higher activity was associated with lower alpha spreads.
Their findings led Jones and Mo to conclude: “To a potential mutual fund investor, advice from the academic literature on what mutual funds to hold may be much less useful than advertised.” They added: “Arbitrage forces appear to have strengthened over time, causing fund alphas to gradually shrink toward zero.”
Summarising their findings: “We conclude that the evidence in favour of positive investment in actively managed equity mutual funds is currently weak, at least for those investors basing decisions on the predictors we have analyzed. While some alpha may remain, it is far below historical averages and may have disappeared completely. Considering that even a well-diversified portfolio of actively managed funds adds at least several percentage points of idiosyncratic risk to a comparable passive portfolio, the risk-return tradeoff inherent in holding the high quintile portfolio appears marginal at best. … The clear implication of our findings is that investment practitioners, who are known to use at least some of these measures to guide portfolio selection, may be engaging in an exercise that is of dubious relevance.”
Their finding that increased arbitrage activity and competition is associated with declining alpha spreads is consistent with that of Laurent Barras, Olivier Scaillet and Russ Wermers, authors of the study False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas, published in the January 2010 issue of The Journal of Finance, who found a significant proportion of skilled (positive alpha) funds prior to 1996, but almost none existed by 2006.
I would add that the newer asset pricing models that include the investment and profitability (quality) factors add even more explanatory power to the cross-section of returns. Had Jones and Mo included those factors in their analysis, it seems highly likely they would have found even more deterioration in the alpha predictors they examined.
There is much evidence that not only does skill exist in the active management industry, but the level of skill is increasing. And investors are able to identify the skill and reward it with cash flows. However, just as theory suggests, the scarce resource (the ability to generate alpha) gets to keep any economic rent — skilled funds earn gross alphas, but the alphas go to the fund sponsors, not the investors, who earn negative alphas.
In addition, as Robert Stambaugh points out in his January 2019 paper Skill and Profit in Active Management, the fact that active managers are more skilful results in the market discovering and correcting any mispricings more quickly. Stambaugh concluded: “The results here show that an increase in overall skill can imply a smaller equilibrium amount of fee revenue.” He added: “If greater skill spells less revenue, an upward trend in skill represents a potential challenge for the active management industry.” That’s on top of the increasing hurdles we have discussed — including a continued reduction in the number of noise traders. In other words, to date, the decline in the overall level of active management (the increase in hedge fund assets has been more than offset by the shift away from actively managed mutual funds to index and other passively managed funds) is not resulting in more mispricing, but less.
The bottom line is that the active management industry is facing strong headwinds — research converting alpha into beta, the shrinking pool of victims, increasing skill levels of the competition, increasing supply of arbitrage capital — including the race to the bottom in fees for passive strategies. The good news is that not only is the cost of both passive and active strategies coming down, but despite the decline in the share of active investing, the markets have become more efficient, not less, as many have warned would be the case. And there doesn’t seem to be anything that will change the direction of the trends. The result is that the ability to generate alpha will continue to persistently shrink.
For informational and educational purposes only and should not be construed as specific investment, accounting, legal, or tax advice. Certain information is based upon third party data and may become outdated or otherwise superseded without notice. Third party information is deemed to be reliable, but its accuracy and completeness cannot be guaranteed. Indices are not available for direct investment. Their performance does not reflect the expenses associated with the management of an actual portfolio nor do indices represent results of actual trading. Information from sources deemed reliable, but its accuracy cannot be guaranteed. Performance is historical and does not guarantee future results. Total return includes reinvestment of dividends and capital gains. 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 websites. 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. Neither the Securities and Exchange Commission (SEC) nor any other federal or state agency have approved, determined the accuracy, or confirmed the adequacy of this article. The opinions expressed by featured authors are their own and may not accurately reflect those of the Buckingham Strategic Wealth® or Buckingham Strategic Partners®, collectively Buckingham Wealth Partners. LSR-22-263
LARRY SWEDROE is Chief Research Officer at Buckingham Strategic Wealth and the author of numerous books on investing.
ALSO BY LARRY SWEDROE
© The Evidence-Based Investor MMXXII