The Small-Cap Effect

Mark Hulbert in Barron’s on the Small-Cap Effect:

Just how strong is the statistical and historical support for the notion that the average small-cap stock outperforms the average large cap — the much-vaunted small-cap effect?
It certainly looks very strong: Virtually every major discussion of small caps’ supposed relative strength refers to the famous Ibbotson Associates data set, which shows that small-cap stocks have significantly outperformed large caps since the 1920s (see Electronic Q&A, “Ibbotson: Small Caps Still Have Big Bang,” Feb. 14).
What could be stronger than that? Most of the other alleged patterns about which investors spin endless tales have just a fraction of the apparent support that the small-cap effect has.
Yet, sacrilegious though it may be, I decided to take a closer look at the historical case for small caps’ relative strength, and it turns out that this case is surprisingly weak.
I reached this conclusion by analyzing data provided on the Website of Dartmouth University finance professor Kenneth French. This comprehensive data set, compiled by French and University of Chicago finance professor Eugene Fama, reflects the performances of virtually all publicly traded U.S. stocks from mid-1926 to the present. It is free and publicly available; those of you who enjoy crunching numbers will find it invaluable.
Consider first the returns of five different portfolios that Fama and French formed based on the market capitalizations of various stocks.
The first portfolio contained the 20% or so of stocks that had the largest market caps, while the fifth portfolio had the smallest stocks; the middle three portfolios comprised the roughly 60% of stocks in the middle. Fama and French rebalanced these portfolios yearly to take into account changes in stocks’ market caps.
At first blush, these portfolios’ returns provide incredibly strong support for the small-cap effect. The average return of the stocks in the portfolio containing the smallest-cap stocks was 17.1% on an annualized basis between mid-1926 and the end of 2005, versus 10.2% for the quintile that had the largest-capitalization stocks — for an impressive difference of 6.9 percentage points annually.
But, as broadcaster Paul Harvey might say, here’s the rest of the story.
It turns out that all of the small caps’ extra return comes in January.
If we focus on all months besides January, the largest-cap quintile has produced a 9.1% annualized return since mid-1926, in contrast to 7.8% annualized for the smallest-cap portfolio. The small caps’ 6.9-percentage-point advantage over the largest caps becomes instead — without January — a 1.3-percentage-point deficit!

The files at Kenneth French’s site also contain data for size deciles. From June 1932 to the end of 2004, the smallest decile, or 10% of stocks, gained nearly 10,875,000%. January was responsible for 26,170% of the total.
Here are the annualized micro-cap returns by month from July 1926 to December 2004:
January………159.82%
February………25.10%
March……………1.21%
April…………….12.86%
May……………….6.85%
June………………6.57%
July……………..22.64%
August………….8.16%
September…….-7.75%
October……….-14.60%
November……….7.97%
December………-1.94%

Posted by on February 17th, 2006 at 12:32 pm


The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.