Last Updated Oct 3, 2011 3:09 PM EDT
What makes the efficient markets hypothesis so powerful is that the discovery and exploitation of anomalies (inefficiencies) causes them to shrink and eventually disappear. However, if a premium exists because of the existence of greater risk, the premium should be expected to persist -- at least before the fact.
In 1981, Rolf Banz published his dissertation, "The Relationship Between Return and Market Value of Common Stocks," which documented higher average returns for common stocks of small companies relative to large companies. Many people have since questioned the results based on the performance of small caps in the post-Banz era. Unfortunately, there's a lot of misperception out there. Let's see if we can clear things up.
First, Banz found that from 1927-1981, average returns increase monotonically from the largest stocks to the smallest stocks. The largest decile had average returns of 10.0 percent per year, while the smallest decile returned 24.3 percent per year.
In the post-Banz era, 1982-2010, the pattern of returns is similar (though flatter) than in the early period. The largest stocks have an average return of 12.5 percent per year versus 15.6 percent per year for the smallest stocks. This 3.1 percent return difference is smaller than in the earlier sample. However, there's too much noise in this estimate to say with any statistical confidence that the size premium has declined since 1981. The pattern we observe in the data can be an artifact of chance. Given that there's not enough data to conclude that the size premium has changed, using the full sample will yield the most precise estimate of the size premium. Over the period from 1927-2010, the smallest decile of US stocks outperformed the largest decile by 10.4 percent annually.
Photo courtesy of DonkeyHotey on Flickr.