Understanding the Size Factor

This is Part 4 in a series about smart beta ETFs. See below for links to other posts in the series. In this installment, we look at the size factor: the idea that smaller companies should deliver higher returns than large-cap companies.

 

The idea that small-cap stocks have higher expected returns has a relatively long pedigree. The pioneering paper was authored in 1981 by Rolf Banz of the University of Chicago, who looked at US stocks from 1936 to 1975 and found that smaller firms, on average, enjoyed higher returns than larger companies. It found that smaller firms had higher returns, on average, than larger firms. That finding was confirmed in a ground-breaking 1992 paper by Kenneth French and Eugene Fama, the same one I discussed in my previous post on the value factor.

Over long periods, size has mattered a lot: depending how you define them, small caps in the US have outperformed large caps by about 2% to 4% per year since 1927 (the year the reliable data begin). And while most of the research on the small-cap premium has focused on US stocks, a 2006 review of the literature by Savina Rizova of Dimensional Fund Advisors found evidence that it also exists in Canada, the United Kingdom, Australia, Germany and many other developed and emerging countries.

The size factor joined beta and value as part of what became known as the Fama-French Three-Factor Model. Remember the capital asset pricing model (CAPM), which suggested the expected return on a stock depends on whether it is more or less volatile (riskier) than the market? It turned out CAPM performed poorly when tested in the real world, but Fama and French’s new model did much better. You can explain more than 90% of a stock portfolio’s returns by analyzing its exposure to the overall market (beta), the price-to-book ratios of the holdings (the value factor) and the size of the companies (the size factor).

This was a big deal, because it meant you could improve the expected return on a portfolio by “tilting” it toward value stocks and small-cap stocks. It was the birth of what we today call factor investing, or smart beta.

Why have small-cap stocks outperformed?

There are several theories to explain why smaller companies have produced higher returns. The most widely accepted is that investors are rewarded for the additional risks: small companies are more sensitive to business cycles, have a higher cost of capital, are more likely to go belly-up, and tend to have more dramatic swings in their stock prices.

Others have argued that small stocks are less likely to be covered by analysts, less attractive to institutional investors, more thinly traded and less liquid, so investors expect to be compensated for these risks with higher returns.

What the critics say

While the small-cap premium is widely accepted among academics and money managers, it has also been challenged: it’s inconsistent over time, it’s not as strong internationally as in the US, and it doesn’t hold up as well if you identify small companies using measures other than market price. Indeed, one 2014 article goes so far as to argue that, “contrary to popular opinion, there is little solid evidence that stock size affects performance.”

While small-caps have outperformed over the very long term, some have argued the premium has disappeared. The most widely tracked benchmark for small-cap US stocks is the Russell 2000, and from 1979 through August 2016 this index returned 11.5% annualized. Over the same period, the large-cap Russell 1000 and S&P 500 returned 11.8% and 11.7% respectively. That’s a 35-year period of underperformance.

As Larry Swedroe has argued, more complete data from the Center for Research in Security Prices (CRSP) suggests the premium has indeed persisted since 1979. But even he admits it is likely lower than when it was first described. “Today, it’s both much easier and less costly to diversify the risks of small stocks, through mutual funds and ETFs, than it was during the period Banz studied,” Swedroe wrote in a 2015 article. “In addition, trading costs, in the form of commissions and bid-offer spreads, have come way down. Thus, we shouldn’t be surprised that the size premium may have shrunk over time.”

Moreover, in Canada (and probably in other small markets) screening for small-cap stocks can introduce unexpected risks. For example, the broad Canadian market already has about 35% allocated to the energy and materials sectors. The S&P/TSX Small Cap Index pushes that to about 55%, exposing investors to risks that have less to do with small companies and more to do with oil and gold prices.

How to access the size factor with ETFs

For Canadians seeking exposure to the size factor, the most direct route is the iShares S&P/TSX Small Cap (XCS), which holds just over 200 stocks with a median market capitalization of about $685 million. Many of these companies are too small to be included in the popular S&P/TSX Composite Index, generally considered a benchmark for the whole Canadian market.

For US small caps, the only Canadian-listed ETF is the iShares U.S. Small Cap (XSU), which tracks the Russell 2000. Note that this ETF uses currency hedging, so it is not vulnerable to the USD-CAD exchange rate. For unhedged exposure, you’ll have to use a US-listed ETF such as the iShares Russell 2000 (IWM) or the Vanguard Small-Cap (VB).

There’s no Canadian-listed ETF tracking international small caps, so your only options are US-listed funds such as the iShares MSCI EAFE Small-Cap (SCZ), which includes developed countries only, or the more broadly diversified Vanguard FTSE All-World ex-US Small-Cap (VSS), which also includes emerging markets.

The iShares Edge MSCI USA Size Factor ETF (SIZE) adds something of a twist, as it does not specifically target small-cap stocks. Instead, it offers “exposure to large- and mid-cap U.S. stocks with a tilt towards the smaller, lower risk stocks within that universe,” so its top holdings include PepsiCo, Colgate-Palmolive and Johnson & Johnson. The iShares Edge MSCI International Size Factor ETF (ISZE) takes a similar approach with overseas developed markets.

Traditional small-cap ETFs exclude all companies above a certain size, but those that remain are weighted according to market cap, so a $200 million company gets twice as much weight as one worth $100 million. An equal-weight index is another way to give more influence to smaller companies. Consider the options for those looking to invest in Canadian REITs: Vanguard and iShares both offer cap-weighted ETFs in this sector, while BMO’s offering weights its holdings equally. By doing so, BMO’s fund gives added influence to smaller REITs, and will therefore will have more exposure to the size factor.

Just keep in mind that equal-weighting does not necessarily mean you’re getting meaningful exposure to the size factor. A fund like the popular Guggenheim S&P 500 Equal Weight (RSP) gives additional influence to the smaller companies in its parent index, but these still large-cap stocks by any definition.

 

Other posts in this series:

Smart Beta ETFs: Your Complete Guide

A Brief History of Smart Beta

Understanding the Value Factor

 

6 Responses to Understanding the Size Factor

  1. Mike September 9, 2016 at 9:12 pm #

    Hi CCP,
    Interesting article. I used to hold XCS for the reasons listed above, but liquidated my holdings last year. I found that while VBR has a nice MER of .08%; XCS is .61% and thought that I had enough exposure with VCN. What are your thoughts about the potential returns of small cap ETFs compared to their usually higher MER?

  2. Peter September 9, 2016 at 11:14 pm #

    I believe a part of every person should have a part of their investments in small caps. Doesn’t have to break the bank but an ETF for this part of the pie is ideal.

  3. Victor H September 10, 2016 at 12:37 am #

    Well researched and interesting as always.

    Not to give the game away but it seems that abretail investor can’t out smart the market.

    Looking forward to the Disapline aspect.

  4. Phil September 10, 2016 at 3:23 pm #

    Here’s hoping a TSX-listed, non-hedged, Russell 2000 small caps ETF will be released soon.

  5. Bill September 10, 2016 at 8:41 pm #

    While small cap value has outperformed, you should avoid small cap growth. It has dramatically underperformed.

  6. Canadian Couch Potato September 11, 2016 at 8:01 am #

    @Mike: You’ve hit on an important point that I’ll discuss later. If you start with the knowledge that s small cap ETF costs about 0.55% more than a broad market ETF, the premium has to be at least that large before you break even. Add additional trading costs and potentially higher taxes from more turnover and you’re widening the gap even further.

Leave a Reply