What Fewer Stocks Means for Factor Premiums
Fewer stocks today may mean stock picking is dying, but will it doom factor premiums as well.
Larry Swedroe, Director of Research, The BAM Alliance
Jason Zweig, financial columnist for The Wall Street Journal, is one of the truly “good guys” in the world of financial media (by which I mean that he looks out for the interests of investors by providing knowledge instead of the typical hype). If he writes it, it’s worth reading. Thus, I read with interest his recent column on why stock picking is a dying phenomenon.
Zweig began by noting: “There were 7,355 U.S. stocks in November 1997…. Nowadays, there are fewer than 3,600.” He added: “A close look at the data helps explain why stock pickers have been underperforming.”
He then explained: “Several factors explain the shrinking number of stocks … including the regulatory red tape that discourages smaller companies from going and staying public; the flood of venture-capital funding that enables young companies to stay private longer; and the rise of private-equity funds, whose buyouts take shares off the public market.”
Zweig, in part quoting Michael Mauboussin, investment strategist at Credit Suisse Group AG, writes: “For stock pickers, differentiating among the remaining choices is ‘an even harder game’ than it was when the market consisted of twice as many companies …. That’s because the surviving companies tend to be ‘fewer, bigger, older, more profitable and easier to analyze,’” thus “making stock picking much more competitive.”
Enduring Result
Zweig then moved on to his second major point: “The evaporation of thousands of companies may have one enduring result … and it could catch many investors by surprise.”
Specifically, he writes that the historical outperformance of many investment factors, such as size and value, may “have been driven largely by the tiniest companies, exactly those that have disappeared from the market in droves.”
He warns: “Before concluding that small stocks or cheap ‘value’ stocks will outrace the market as impressively as they did in the past, you should pause to consider how they will perform without the tailwinds from thousands of tiny stocks that no longer exist.”
Deeper Dive
I received several requests from clients to comment on the article, so I’ll share my thoughts on it here as well.
To begin, I agree that the fewer the number of stocks available, the less opportunity there should be for stock pickers to generate alpha (outperform on risk-adjusted basis).
I also agree that, because the smallest companies are the ones disappearing, all else being equal, it’s logical to conclude the small-cap premium should shrink. Indeed, a virtually monotonic relationship has existed between returns and size, with returns increasing as you go from the first decile (the largest stocks) to the 10th (the smallest, or micro-cap, stocks).
Historical Results
With all that said, let’s review the historical results to see if they support the article’s hypothesis.
Thanks to data from The World Bank, we find that, at the end of 1974, there were 2,670 stocks listed on the U.S. exchanges, far fewer than existed in 1999 (the year the dot-com bubble burst), when there were more than 7,000 stocks. That’s also far fewer than the 4,331 stocks listed at the end of 2016.
If the thesis regarding the shrinking pool of small stocks is correct, then the size premium should have been much smaller pre-1975 and certainly much smaller post-1999. However, the size premium (SMB, or small minus big) from 1927 through 1974 was an annual average 2.54%, very close to the 2.75% size premium from 1975 through 1999.
One would think that, with an almost tripling of the number of stocks—most of which would be considered very small—the premium should have been much larger. Yet it was only 0.21 percentage points greater.
Furthermore, from 2001 through 2016, when the number of stocks fell by about one half—with most of that disappearance coming from the smallest stocks—the size premium actually rose to 3.58%.
Other Logical Reasons
What’s more, as my co-author Andrew Berkin and I point out in our new book “Your Complete Guide to Factor-Based Investing,” there are other logical reasons to believe that the ex-ante size premium should now be now lower.
For example, trading costs have decreased substantially over time with the elimination of fixed commissions, and the decimalization of prices and the presence of high-frequency traders have led to smaller bid/offer spreads.
In addition, mutual fund and ETF expense ratios are also much lower, providing investors with access to the size factor at much lower costs. Because implementation costs have fallen, investors can capture more of the size premium. Thus, again, all else being equal (such as economic risks), the premium itself should be expected to be lower.
These factors have led my firm’s investment policy committee to apply a 25% haircut to historical risk premiums (such as size and value) other than market beta when forming forward-looking return expectations. For market beta, we use a valuation-based approach, taking the inverse of the Shiller CAPE 10 (the earnings yield).
Unexpected Outcomes
There’s an important lesson you can take from the above: If you had the perfect foresight to predict the shrinking pool of small stocks, combined with the lower implementation costs, you almost certainly would have forecasted, back in 2000, a smaller size premium. Yet the premium actually rose.
This is just another example of why it’s so hard to time premiums. The best that investors can do in a world of uncertainty is to diversify broadly across premiums (factors) that meet the criteria we establish in our aforementioned book.
Specifically, before investing in a premium/factor, you should be convinced by the evidence that it demonstrates persistence (the factor historically delivers reasonably reliable returns); pervasiveness (it, on average, delivers these returns in a variety of locales and asset classes if such tests apply); robustness (it shouldn’t be dependent on one very specific formulation but fails to work if other, related versions are tested); intuitiveness (it makes sense to us rather than just going by historical performance); and, finally, investability.
Aside from investability, all of these criteria come down to some aspect of whether we believe the historical results are real and not just the product of data mining. Investability is the one that differs, implicitly asking the all-important question: Even if we believe the factor is real, can a practical investor really make money from it after costs?
An Observation
Before closing, I have one other interesting observation. While the number of listed stocks in the U.S. has declined sharply over the past 18 years, the number of listed stocks outside the U.S. has grown dramatically.
According to a March 2017 paper by Dimensional Fund Advisors, “Going Global: A Look at Public Company Listings,” the “number of firms listed on US, non-US developed, and emerging markets exchanges has increased from about 23,000 in 1995 to 33,000 at the end of 2016.” Many of these are small stocks.
If we follow the thinking in The Wall Street Journal article, it should have become much easier for active managers to outperform and the size premium should have increased. Yet the evidence on the success of active management outside the U.S. has been just as dismal as it has been domestically over this period.
And the size premium has been virtually unchanged—the world ex-U.S. small-minus-market factor premium (MSCI weighted) was an annual average 2.15% from 1975 through 1995, and 2.26% from 1996 through 2016.
Increased Difficulty
Finally, as my co-author Andrew Berkin and I explain in our book, “The Incredible Shrinking Alpha,” there are explanations for the increased difficulty stock pickers (active managers) have experienced—over the past 20 years the percentage of active managers generating statistically significant alpha has shrunk from about 20% to about 2%, and that’s even before taxes—that go well beyond the shrinking pool of stocks hypothesis discussed in Zweig’s article.
The four explanations we present are that academics have been busy converting what was once alpha into beta (a common factor, such as size, value, momentum, profitability and quality); the pool of victims needed to exploit has been shrinking at a rapid pace; the skill level of the remaining competition has greatly increased; and the amount of assets chasing the shrinking pool of alpha has grown dramatically.
This commentary originally appeared June 30 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
