One of the great mysteries in the asset management industry over the past 30 years has been the rapid growth of the mutual fund industry in the face of high fees and underperformance vs. passive benchmarks. Vanguard founder John Bogle, for example, has been an outspoken critic of active mutual funds for these reasons. Yet to his disappointment, investors still dump truckloads of money into mutual funds, making them a $10 trillion industry in the US alone.
Studies have shown that, overall, mutual funds don’t recoup their fees in the form of market out performance. That fact is often held up as proof that active management doesn’t pay. But is that really true? As we have complained on this website, many so-called “active” mutual funds aren’t active at all. They are closet indexers. So it’s possible that the truly active funds can out perform the index, but that performance is being swamped by the lackluster results of the closets indexers.
This is the question addressed in an August paper by Zheng Sun, Ashley Wang and Lu Zheng of the University of California.
To isolate the performance of truly active funds from closet indexers, the trio divided the universe of mutual funds into 10 deciles based on two different measures that have been discussed here at AllAboutAlpha.com. Firstly, they used the “Active Share” metric devised by Cremers & Petajisto (see related post). This measures the aggregate difference between the fund’s stock weightings and the index’s weightings for the same stock. You might call this a “bottom-up” analysis of activeness.
Second, they used a measure called the “Strategy Distinctiveness Index” or SDI which is based on the correlation of a funds’ returns to the returns of its peers (a measure the trio developed last year for hedge funds – see related post).
They organized funds into deciles according to both measures and found that the most active funds have some curious qualities. While the active funds performed no better than the market during economic expansions, they significantly beat the market during economic contractions.
The authors chose none other than John Bogle’s Vanguard 500 Index Fund as a benchmark for their analysis. The following table from the paper shows that the most active funds (the “10th Decile”) performed a little worse than the Vanguard fund during economic expansions. However, during economic contractions these truly active funds blew the pants of the Vanguard 500 fund – actually rising by 0.42% per month vs. a drop of 2.67% for the Vanguard fund.
Sun, Wang and Zheng found basically the same thing when they put funds into deciles based on the SDI measure.
So much for efficient markets, eh? Check out how each decile of active-ness performed in the chart below (constructed from data in Table 5 of the paper).
It’s almost as if truly active funds provide put protection against an economic contraction (a put option, we might add, that does seem to have a real cost (a premium) in the form of slight under performance during economic expansions).
But what might cause this apparent transgression of the efficient markets hypothesis? The authors of the report table a few hypotheses. One is that during periods of economic contraction, corporate management is more opaque and less willing to disclose material information to the markets. This provides opportunity for those who can eke out an informational advantage through proprietary research. (To back up this argument, the authors statistically illustrate that the source of out performance is stock picking, not market timing.)
They also debunk the theory that active managers hold more cash during good times and bad. It seems to make sense that this is not the case since performance during expansions is pretty close to the market, while performance in contractions is way better. But to remove any doubt, Sun, Wang and Zheng use a form of regression that controls for this possibility.
So are investors aware of this “put option” available through active funds? Perhaps not in individual cases. But amazingly, their aggregate behavior seems to be quite rationale. The trio finds that management fees increase monotonically from the least active decile (27 bps per quarter) to the most active decile (39 bps per quarter).
Active management “boutiques” have always charged a little more. But from an investor’s perspective, those fees might actually be justified – at least according to this research. It would appear that investors essentially pay for downside protection in two ways: firstly, in the form of lower returns in economic expansions and second, in the form of higher management fees.
A final thought on this interesting and potentially game changing study: Note that the authors measure performance in expansions and contractions of the real economy (using NBER definitions), not in the market itself. Since the real economy tends to lag the financial markets, quarters counted as “contractions” in this study, actually occurred during “up” quarters for the index. We’d be interested to see the results if the same methodology was applied to up and down months in the markets, instead of the real economy.