Prudent Cold Skew
Are small companies riskier than larger companies? And do smaller stocks earn higher returns as a result?
A new paper by French researchers offers a fresh perspective on this heated debate. They argue that it’s not size by market capitalization but by average daily trading volume that matters. Volume measures both the liquidity and the level of investor interest and is, they argue, a superior way of thinking about size.
They tested across 10 countries and found consistently that “cold” low-volume stocks significantly outperform “hot” high-volume stocks—and do so far more reliably than small stocks outperform large stocks.
The researchers argued that trading volume is superior to size because low market capitalization is highly correlated with high-beta (low-premium) stocks while volume is not. This heavy loading on the market factor “pollutes the performance of SMB [small-minus-big] portfolios” and generated spurious results. Volume-based CMH (cold-minus-hot) portfolios don’t have this tendency. Measured this way, they confirm the size effect is “alive and well” and has been quite persistent and robust internationally.
This is consistent with a large body of research by Robert Stambaugh, who, along with Lubos Pástor, recently published an excellent review of the literature on the excess returns of less liquid stocks.
They find that smaller stocks have a significantly higher positive skew compared to larger stocks. Skewness measures how a distribution of stock price outcomes “leans.” Positive skew describes a distribution where there is a long “right tail” of high-return outcomes, resulting in a mean stock price outcome higher than the typical outcome. Over time, portfolios designed to capture these long right-tail distributions have earned a premium over portfolios that avoid such stocks.
Figure 1: Skewness of Stock Returns by Market Cap and Average Daily Volume
Source: Ciliberti et al.
Other recent research has noted this lopsided payout and distribution pattern as well.
Figure 2: Annual Stock Return Distributions by Size Deciles
Source: Bessembinder. US CRSP data since 1926.
But what’s most interesting about the French paper is not their identification of this illiquidity effect but rather their work on how illiquidity and risk interact.
Traditional efficient-markets rationalizations for the size premium often posit that this positive skew and the higher average returns in small stocks is compensation for the riskiness of the portfolios.
But what the French researchers found was that extreme return outcomes in the SMB and CMH portfolios were actually highly concentrated in the larger stocks. Quite contrary to conventional theory, their analysis suggests that “the ‘risk’ of such portfolios is not at all due to the exposure to small stocks, but rather to the supposedly safer large stocks.” In other words, the risk of being horribly wrong wasn’t showing up in the low-volume stocks so much as the bigger stocks that were in the CMH portfolios.
Figure 3: Extreme Portfolio Returns by Stock Size
Source: Ciliberti et al. 20 years of the best 100 and worst 100 days of profit and loss by low to high volume.
They conclude that, in our elusive quest for asymmetrical long-term absolute return strategies, “’prudence’ (i.e., aversion for negative skewness and appetite for positive skewness) should favor small cap/ADV stocks, in contradiction with the idea that SMB or CMH are risk premia strategies.” In other words, there is a small free lunch of cold skew to be had.
The main problem with this free lunch is that the premium can only be exploited by small, nimble funds. Most funds over $200M will put in place minimum daily volume requirements that essentially remove the “cold skew” stocks from consideration.
One of the most robust premia in public equity markets—with the highest statistical significance—is simply buying thinly traded stocks that other funds, by virtue of their asset-gathering tendencies, can’t profit from.