BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Big News From AQR: Size Matters, If You Control Your Junk

This article is more than 9 years old.

A whiz-bang paper has just been released that basically solves one of the longstanding mysteries in financial economics: the outperformance (or not) of small company stocks. It’s called, “Size Matters, If You Control Your Junk.” The authors are all lettermen from the varsity squad at AQR Capital: Cliff Asness, Andrea Frazzini, Ronen Israel, Toby Moskowitz, and Lasse Pedersen.

First, the backstory: In 1981, Rolf Banz wrote a paper announcing that over the past forty years smaller firms had displayed higher risk-adjusted returns than larger firms. Right from the start, Banz was hip to some important qualifications to this pronouncement. He saw that this did not apply to all small companies equally: only very small companies got the outperformance. He also wondered whether this was due to their small size per se, or whether it was because of some other X factor that just happened to be correlated with size. The size factor (as it came to be known) received the Papal blessing when it crashed like a Buick into the classic Fama-French 3-Factor Model. When company size was combined with value and market exposure, it explained 90% of the variation in investment returns. The view was that the premium from investing in small company stocks was not a violation of basic market efficiency, but a compensation for taking on the added risk attendant upon their more diverse prospects.

Dimensional Funds opened a microcap fund in 1981 to capitalize on the new finding. Others followed. Then a funny thing happened on the way to this fandango. The small cap premium vanished like the grin of a Cheshire cat.

Academics filed their teeth and pounced: The size effect was only of marginal statistical significance, they claimed. The prices used to establish it were stale and then suffered from survivorship bias as the losing stocks got delisted. It was hard to replicate the data overseas. It seemed to be completely dependent on the time period you examined – sometimes you saw it, sometimes you didn’t. It only worked in January. It got arbitraged away by greedy investors who hunted it to extinction once the word got out. It was not really an effect at all: the real factor was illiquidity (because small stocks were thinly traded). Or maybe it was really just a quirky subset of the better-established value factor.

Whatever the reason, the small cap premium became the bastard red-headed stepchild in the factor zoo. Then, just to completely flummox everybody, the snarky small cap effect reappeared again in 2000 and stuck around thereafter. Yes, we love small cap! Give us more!

With this new paper, Asness et al sort the whole mess out. But to understand it, we need to wrap our brains around another factor – one that has been in the news recently: the Quality Factor.

The idea that buying high quality companies is the right way to go comes from Berkshire Hathaway's co-chairman Charlie Munger. Warren Buffett, Berkshire’s chairman, famously had been a value investor in the tradition of Ben Graham. He looked for undervalued stocks. The determinant was price: could a company be purchased at a substantial discount (including a margin-of-safety) to its intrinsic value?

Munger, though, had a different idea. In 1972, he recommended to Warren that Berkshire buy a little outfit in Los Angeles called See’s Candies. See’s was a small but super high quality company.

Warren loved the candy but choked on the price. The catch was that he would have to pay full retail for the business - $25 million. With Munger’s prompting, he closed his eyes and wrote a check.

What happened next became investment history. Berkshire discovered that the demand for See’s Candy was inelastic. They could raise prices every year and no one cared. See's became a cash cow, returning $1.35 billion to its owners.  The new mantra became: a high quality business at a fair price. This led to subsequent purchases like American Express and Burlington Northern. Ben Graham did not eschew value -- he generally looked for stocks with a Standard & Poor's rating of 'B' or better. But he thought name-brand companies like Coca-Cola were for "retail" investors, i.e. overpaying suckers.  Graham was uncomfortable making "qualitative" judgments about a business. Munger and Buffett weren't. They built the investment fortress that is Berkshire Hathaway on the back of the quality + value premia.

It’s A Small World After All

What this new paper finds is that once you control for the quality of the companies, the size effect is real: the returns from size stack up like cordwood, with the smaller, high quality companies having progressively higher returns than those in the larger company deciles next door.  'Quality' here is not a subjective judgment -- it means profitable businesses that display profit growth, safety, and have a high payout. By these measures, high quality small companies outperform by a significant margin and consistently across time, industries, and markets. It is only the small cap stocks with junk in the trunk – the more distressed and illiquid securities – that disappoint. The trick is to order the combo plate: high quality plus small market capitalization.

Naturally, the paper raises new problems. The riskiest small cap stocks underperform. This torpedoes the risk-based story for the small cap premium. These doggy stocks are also the ones that are less liquid, so there goes the "illiquidity" explanation, too.

In my view, this exposes the charade of retrofitting risk-based or “behavioral” explanations on to empirically-derived factors. The human brain is so constituted that it can come up with a post-hoc explanation for anything. To make up just-so stories about them feels reassuring, but only adds noise. These explanations are chewing gum for the mind.

Finally, the paper is so well written that it breaks the complexity barrier; a lay reader can go through the opening pages and get a comprehensible overview refreshingly free of academic cant. Small is beautiful; so is the writing.