It should be no surprise that small cap stocks struggled last year compared to large cap stocks. While the S&P 500 enjoyed above average results earning 13.7 percent, the Russell 2000 index of small cap stocks only gained 4.9 percent (click here for a chart of major asset class results).
A little more than a year ago, I wrote that there is a long-running debate in academia about whether small cap stocks really outperform large cap stocks.
Here is a link to the original article, but the basic idea is that small cap stocks are more volatile than large cap stocks, so you need a lot of data to say whether the outperformance is statistically significant or chance.
The irony of the article, written at the end of 2014, was that small stocks had much better performance than large cap stocks, but not enough to offset what happened last year. If you look at the two years combined, large beat small by almost five percentage points cumulatively.
Given the recent underperformance, I was naturally interested to see a new research paper from the principles at AQR and professors from University of Chicago and Yale. This is a real academic submission, not a ‘white paper’ that is marketing in disguise (who would do that?).
The paper, provocatively titled ‘Size Matters, If you Control Your Junk,’ makes the case that small cap is weak when looked at independently. No surprise there – that debate is decades old. Their contribution is that small caps do have superior returns, if you focus on quality companies and avoid junky ones.
A few years ago, the same group of researchers wrote a paper, called Quality Minus Junk, that said that quality is an independent source of returns, like size, value and momentum.
In both papers, they argue that small caps work when factoring quality, but this new paper takes that idea to the Nth degree, testing the idea robustly over time, outside the US, factoring for illiquidity, and calendar effects just to name a few. It’s impressive.
The paper is beyond wonky, but the super quick and dirty (found on page 39 in hieroglyphic-like scrawl is that the confidence that the small cap premium for the period studied (1957-2012) is around 75 percent – not significant to academics.
If you control for value and momentum, the significance doesn’t get any better, but if you control for quality, confidence that the size premium exists jumps to 99.99 percent – pretty darn high. (As you would expect from a 60 page research paper, there is more to it, but you get the idea).
Their argument is sensible – within small cap stocks, the quality varies greatly, much more so than in large cap stocks. Unfortunately for small companies, the poor performance of the junky component of small cap is bad enough to erase the benefit of the whole strategy.
All you need to do, apparently, is cut out the junkiest 20 percent and small cap works (not every year, but on average over time).
We have an advantage here in that the main product that we use for our small cap exposure has a strong relationship with the quality factor that the AQR researchers created a few years ago.
We bought this fund (a non AQR product, by the way) for different reasons, outlined in more detail here, and are fortunate that this new research serendipitously supports how we already invest.
It wasn’t pure luck though – we believed in the idea of quality before their papers on the subject. It makes intuitive sense for one thing, but also, Warren Buffet has been preaching about quality decades before we set up shop.
(Incidentally, the same cabal of researchers wrote a paper explaining Buffets returns with a combination of value, quality and leverage. (Here’s the link).