As explained in my latest book, “The Incredible Shrinking Alpha,” which I co-authored with Andrew Berkin, accompanying the rapid growth of the actively managed mutual fund industry, the average performance of mutual funds has been trending downward over the past few decades.
Teodor Dyakov, Hao Jiang and Marno Verbeek—authors of the study “The Trading Performance of Active Mutual Funds,” which appears in the August 2015 issue of The Journal of Financial and Quantitative Analysis—found that, in using the Carhart four-factor model (beta, size, value and momentum), the annualized abnormal return on the aggregate U.S. equity mutual fund portfolio was 1.7% during the period 1981 through 1990, but declined to 0.86% during the period 1991 through 2000, and to -0.75% during the period 2001 through 2010.
Our book provides four explanations for this declining trend: academics have converted alpha into beta, eliminating potential sources of alpha; a reduction in the supply of victims that can be exploited; increased competition from the remaining players; and increased cash inflows competing for the reduced supply of alpha.
Active Funds Lose Before Fees/Costs
Dyakov, Jiang and Verbeek contribute to the literature by establishing a striking new fact: Even before any fees and trading costs, active mutual funds tend to lose money through their trading. They explain: “Specifically, the stocks purchased by active mutual funds in the aggregate underperform their benchmark by -0.39% in the subsequent quarter up to a cumulative -0.83% in the subsequent four quarters. Even though the effects are not statistically different from zero, the changes from the previous two decades are.”
They continue: “For example, the reversal in the quarter-ahead performance of aggregate buys amounts to -0.97% and the effect that is statistically different from zero. During the four quarters after they are purchased, the aggregate buys have a cumulative risk-adjusted reversal of performance of -1.26%. The reversals in the performance of the sales are of similar magnitude. In total, the changes in the risk-adjusted performance of the stocks traded by funds across the two periods amount to -1.62% in the first quarter up to a cumulative of -2.51% over the four quarters after portfolio formation.”
The authors noted: “These trading losses stand in stark contrast to the gains from active trading achieved by mutual funds in the earlier period.” For example, during the period from 1980 to 2000, stocks with increases in mutual fund holdings (a relatively higher number of buyers to sellers) outperform those with decreases by 0.9 age points over the next quarter.
They concluded: “This finding suggests that there may be additional forces, other than simply the effect of more money chasing limited alpha, driving the declining mutual fund performance… . This result is the first evidence that the performance associated with stock-picking of fund managers might have decreased with time.”
Why Does Trading Destroy Value?
Dyakov, Jiang and Verbeek also try to answer the question of why mutual funds destroy value with their trading. They do so by measuring the secular trend in the tendency of mutual funds to herd—in other words, the tendency of funds to follow the contemporaneous or past trades of other funds.
The authors found that mutual fund herding has increased over time and that herding is an important driver of funds’ trading losses. They also found that the trading losses of mutual funds tend to be concentrated among stocks with the strongest mutual fund herding.
For example, they found that “stocks in the top quintile with intense buy herding have post-2000 risk-adjusted returns of -2.46% in the following quarter and those with intense sell herding of 0.85%. In the period between 1980 and 2000, these figures are 0.57% and -1.05%, respectively.”
They concluded: “The increased tendency of mutual funds to trade in herds appears to be an important driver of their trading losses.”
Other Possible Explanations
Dyakov, Jiang and Verbeek also examined whether, as an alternative to mutual fund herding, reduction in access to privileged information following the implementation of Regulation Fair Disclosure (Reg FD) may also be responsible for deteriorating trading performance. Reg FD, promulgated in 2000, aimed to limit selective access to firm-specific information that institutions, such as mutual funds, and analysts enjoyed at that time.
The authors’ hypothesis was that privileged access to firm information will be more pronounced for funds belonging to larger fund families. However, they did not find evidence that the trading performance of funds deteriorates more for funds belonging to larger fund families. Thus, they concluded that a “reduction in selective access to firm information following Reg FD does not appear to be the main cause of our findings.”
The authors also examined whether the rise of closet indexing offered another potential alternative explanation for the deterioration in trading performance. They found that, prior to 2001, funds with the highest active share score have the highest performance among their trades, generating 0.90% per quarter in risk-adjusted returns. However, after 2009, the performance is zero. Thus, they concluded “that the tendency of indexing among active funds is not the driver of our core results.”
They further examined whether their results could be explained by mutual fund flows. However, the authors found “that both funds with high and lows levels of flows experienced a deterioration in their trading performance in the post-2000 period.”
They also sorted the stocks that funds traded on the basis of their liquidity, a proxy for stock size. They again found that “our main results are not driven by the performance of less-liquid stocks. Overall, the results of these tests do not lend support to the flows-based story.”
Importantly, Dyakov, Jiang and Verbeek noted that their “results are not sensitive to the choice of a particular year as the turning point. In fact, the pattern of deteriorating trading performance of active mutual funds remains strong as we move 2000 to 1996 or 2004 as the turning point. The secular trend appears to be gradual rather than disrupt.” They also observed that “turnover and expense ratios remained on similar levels throughout the sample.”
Following are some thoughts to keep in mind as you consider the evidence presented here.
First, while the study examined the impact of trading, showing the effect to be value-destroying, it didn’t consider the returns of stocks the funds continued to hold. In other words, you shouldn’t conclude that actively managed funds should never trade.
Second, while the study looked at the impact on gross returns from trading, investors care about net returns. Obviously, after considering trading costs, the results would be even worse, especially for smaller, less-liquid stocks (the very stocks where active managers claim they can add the most value).
Third, if the stocks that mutual funds trade then go on to underperform on a gross return basis, who are the winners on the other side of this zero-sum game? The historical evidence suggests it’s not individual investors. Perhaps it is hedge funds (although their performance over the past decade has been so poor it is hard to accept that answer) or other nonpublic institutional investors. It’d be interesting to know the answer to that question, but I cannot think of a logical one.
Dyakov, Jiang and Verbeek contribute to the vast amount of literature on mutual fund performance by offering new evidence on their trading performance. As demonstrated in The Incredible Shrinking Alpha, the general downward trend of aggregate mutual fund performance has been well-identified in the literature. This new paper provides fresh evidence that mutual funds lose money through trading even before transaction costs.
This finding has important implications for mutual fund investors. Whether you decided to invest in actively managed funds, or you were forced to invest in them (maybe because of limited choices inside of corporate retirement plans), you should choose funds not only with low expense ratios, but also with low turnover rates. Those choices will at least reduce the very high odds against generating alpha (risk-adjusted outperformance).
This commentary originally appeared March 21 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.
© 2016, The BAM ALLIANCE