I’ve been trying to work my way through reader questions that also would be good material for inclusion on this stock, and I reached an intelligent question from a reader that wanted to know why I don’t write about small-cap index investing on this site all that often, given what he called the proof of their permanent superiority as investments. The “proof” he was referring to is this: According to research conducted by Ibottson & Associates measuring the 1926 through 2013 time period, an indexed collection of small American companies delivered returns of 12% per year. A similarly indexed collection of large American firms delivered annual returns of 10% per year over that same time frame.
When I’ve discussed this topic before, I mentioned that within the top tier of large American firms, you can doably get results that equal the 12% figure cited by the small-cap basket.
But there is another thing I should point out but did not previously do so because I was unaware: The superior performance generated by small-cap stocks happened in short, concentrated bursts. There were five years that were highly, highly important for you to be invested in small-cap stocks if you sought to benefit from significant outperformance. You had to be invested from October 1990-October 1991, when small-cap stocks delivered returns of 51%. You had to be invested October 2002-October 2003, when small-cap stocks returned a little over 60%. You had to be invested October 1966-October 1967, when small-caps delivered returns of 75%. You had to be there March 2009-March 2010, when returns were over 90%. And then there is April 1942 through April 1943, when small-cap returned just shy of 150%. And then the big cahuna: From June 1932 through June 1933, when the small-caps returned 315%.
The years 1990, 2002, 1966, 2009, 1942, and 1933 had a tremendously outsized impact on the performance of small-cap companies. If you missed those six years and were around for the other 87, you saw your 12% annual returns become less than 7%. In particular the Depression year of 1932 and the War year of 1942 loom large in our understanding of the historical data. It’s not that it doesn’t count; after all, the sharp declines that preceded those comeback years count, but it’s important to recognize how it happens: It’s not, oh, small-cap stocks swim along nicely at 12% per year. Rather, revenue declines send small-cap indices down sharply, and the slightest pickup in revenues during a deep recession leads to a very fast comeback.
To use an example that actually takes into account the pre-eminent small-cap index fund, let’s take a peek at the Vanguard Small-Cap Index Fund. Since its inception in 1960, it has net returns of slightly above 10% when you take into account the long-term fees. Fifty-four years is going to be an entire investment lifetime or more for many of your reading this, and you’d be looking two percentage points less than what the advertised rate of return would be for small-cap stocks. It’s a reminder of how we’re creature of the time period we choose. Over the past thirty years, the Vanguard Small-Cap Index Fund has returned 8%. That’s a four percentage point slide from the 1926-2013 figure.
That’s why I choose to focus my writing on what makes intuitive sense—if you find companies growing earnings in the 8-12% range (and occasionally, you find gems like Visa that do better than that), and you get a dividend thrown into the returns, you can have somewhat clear expectations about what you’re going to achieve. Just blindly adopting the strategy of investing in small-cap index funds rely on the measuring period: Is it near a century—okay, then you’re looking around the 12% range. If you’re looking at 50+ years, you’re looking at 10-11%. If you look at the past thirty, you’re looking at 8%. It’s entirely possible that 1984-2014 is the entirety of someone’s investing life, and small-cap stocks would have performed four percentage points below historical expectations. This isn’t a dig at small-cap index funds; they have a valuable role to play in terms of diversification, and all things be told, they do quite well. But I also want to be as clear in my understanding as possible, and that means recognizing the outsized role that six years in the past 80+ had towards affecting the long-term Ibottson data.