A Simple Market Cap Based Strategy For Generating Alpha
This is part 2 of a 5 part series of articles on using characteristics of stocks to generate alpha. The previous article may be found here.
Please note: Throughout this article when I am talking about small stocks, I am NOT talking about penny stocks or firms traded on the OTCBB or Pink Sheets. All of those stocks are extremely risky, and there has been extensive academic research showing that they have negative returns on average. That is, almost everyone loses money investing in penny stocks. In this article I am discussing small-cap stocks that trade on the NYSE, Nasdaq and former Amex.
As I discussed in my last article on the four factor model, there are four major characteristics of individual equities that are associated with a stock's performance over time. Taken together, these 4 factors predict more than 90% of a given stock's subsequent returns over the next 5 years. There have been literally hundreds of studies done on these four factors and variations of them, and those research findings are the basis for many of the strategies used at major hedge funds today.
This article talks about the size factor as it is commonly called by financial economists. The size factor is based on the market capitalization of a stock, i.e. the share price* the number of shares outstanding. As early as the late 1970s and early 1980s, financial economists had showed that small firms have higher returns than larger firms do. So the overall result that small firms outperform big firms overtime has been around for several decades now.
While there are exceptions and sometimes large stocks outperform the rest of their industry ((AAPL), (AMZN), and (CRM) come to mind), more often large-cap stocks underperform compared with peers. Firms like (CSCO), (PFE), (T), (GE), and (DD) have all underperformed smaller firms in their fields. Research has generally found that rather than trying to "pick" one specific stock or another, a more profitable strategy usually relies on going long a basket of small stocks, and short a basket of larger stocks. This can be done with an ETF or simply by picking a representative set of individual names.
However, classifying a "small" or "large" firm is frequently somewhat difficult since market caps change over time. In fact, what matters isn't the absolute size of the firm (i.e. $1 billion or $100 million), but rather the relative size of the firm. So what financial economists usually do to study this issue is sort all publicly traded firms by size and then break all of the firms up into 5 equal sized groups called quintiles. This leaves the smallest fifth of all firms in the bottom quintile, and the largest fifth of all firms in the top quintile.
When you do this, and then look at returns for those groups of stocks over the next 1 year, 3 years, 5 years, and 10 years, and then take the average per year returns and standard deviations of returns, you get the following (for the period 1960-2011).
|1 Yr Average||3 Yr Average||5 Yr Average||10 Yr Average|
|Smallest Quintile Return||11.20%||11.42%||11.90%||10.83%|
|Smallest Quintile Std. Deviation||4.60%||5.30%||3.70%||4.10%|
|Largest Quintile Return||4.30%||4.72%||3.99%||4.13%|
|Largest Quintile Std. Deviation||2.20%||2.60%||1.90%||2.40%|
|Smallest - Largest Return||6.90%||6.71%||7.92%||6.70%|
What this illustrates is that on average, the smallest fifth of all stocks outperform the largest fifth of all stocks by a healthy margin - about 6.8% every year on average. Over a ten year time frame this would add up to about 100% in extra returns for the small stock portfolio (using simple compounding over 10 years.) This is NOT a product of picking good stocks or luck. Both the top and bottom quintile contain hundreds of stocks. As a result, the contribution for any single stock big or small is minimal.
The other important point that comes out of this is that small stocks are "riskier" than large stocks. That is, the prices of small stocks are much more volatile than the prices of large stocks. The prices of individual small stocks fluctuate significantly more than the prices of large stocks, and the entire portfolio of small stocks fluctuates more than the portfolio of small stocks. What this tells you is that this increased risk associated with holding small stocks cannot be diversified away (eliminated) simply by holding multiple stocks.
Further, these results persist even after we take into account differences in momentum, valuation, and beta for small stocks and large stocks. (To learn how to take these factors into account, visit my blog and read my entry on regression analysis, "A New Tool for Most Investors" at investmentquant.com.)
So how can investors use this information? Well, if you are willing to take on the additional risk associated with holding small stocks, one obvious way to use this result is to hold ETFs or mutual funds that focus on small stocks. These might include funds like (JKJ), (SLY), (VTWO), (IWC), or (IWM). If you want to focus on a specific sector, like materials or telecom where the size effect is particularly strong, there are some small cap ETFs for particular sectors like (PSCM) for materials and (PSCU) for utilities. However these funds are generally small and thinly traded, so it is often better for investors focused on a specific sector to construct their own basket of small stocks.
In fact, for those willing to use short positions, one effective strategy used by many hedge funds is to short a portfolio of large stocks, while going long a portfolio of small stocks in the same industry. Using stocks from the same industry improves your day-to-day correlation in price movements between the long and short portfolios. Ideally you would pick stocks that you know are in the top and bottom quintiles of market capitalization. However for most individual investors this is difficult - (Quick is a $500 million market cap stock in the bottom quintile or second to bottom?), which is why ETFs are often a good choice. Just remember when doing this to use the same notational amount of investment in both the long and short portion of the portfolio. If you can match the betas on the long and short positions, this will also help to hedge your portfolio from intraday volatility.
Another way to use this if you invest in individual firms is to study small-cap stocks for investment opportunities rather than large-cap stocks.