By: Frank Armstrong, CFP, AIF
In our last column, we diversified internationally using the traditional foreign asset class, Morgan Stanley’s Europe, Australia, and Far East index (EAFE). EAFE is an index of large companies, including many multinationals, in developed countries. Though an improvement, the results were distinctly under-whelming. As Rex Sinquefield pointed out in his paper “Where Are the Benefits of International Investing?” the EAFE index fails to provide the benefits that international investing is supposed to bring, namely, increased returns and/or reduced risk.
When you think about it, there is no good reason why a German ought to expect far higher returns for investing in Mercedes Benz than an American should expect from investing in Ford. Both companies share numerous common factors and traits. We should expect that such similar companies will over time have similar returns and costs of capital. Further, as both are large, multinational companies in developed countries with similar products competing in each other’s back yards, we might expect them to correlate closely. And they do. Their stock performance is not entirely determined by local economic factors. It is also strongly affected by common international and automotive trends.
To achieve measurable benefits, we are going to have to look further to find asset classes with higher returns and better diversification characteristics. It is not that international investing is a flawed idea, but that there are far better international asset classes than EAFE to help us achieve our goals.
The effects of style on investment returns
A pioneering work by Eugene Fama and Kenneth French examined the effect of investment style on returns. We can think of investment style as being the part of the market that we invest in. To see what Fama and French discovered, let’s start by dividing up the U.S. market by size of company from the very largest publicly-traded firms to the smallest. On a scale of one to ten, the S&P 500 might occupy the top three deciles, mid-caps and small companies would fall between deciles four and seven, and micro-caps hold down the bottom two deciles.
What Fama and French found was that in a period from 1963 to 1991, micro-caps outperformed large stocks by about 5% per year. As we might have suspected, small companies also have higher risk. As we move down the deciles, rates of return and risk increase rather smoothly.
Their work reveals another nice feature, however. There is a low correlation between the performance of a market’s largest companies and its smallest. They usually perform well or poorly at different times.
Now all this might have been just cocktail party trivia if the study hadn’t been replicated for other time frames and various countries, with startlingly similar results. When many studies find such similar patterns, we begin to feel confident that there is a fundamental economic factor at work.
Most small companies do not have the international exposure of their bigger cousins. Their stock performance and returns are driven more by local economic, political, and emotional factors than are multinationals’. Small companies have low correlations not only with the largest companies within their own countries, but also with companies both large and small in other countries.
As you remember from our discussion of Modern Portfolio Theory, if we mix risky asset classes with low correlations together, the resulting portfolio will have higher rates of return, but with lower risk than the average of the individual parts. It’s only fair to point out that as with all other diversification techniques, nothing works every day, and there will be long periods of over- and underperformance for the various sectors. For instance, let’s look at recent returns of the U.S. markets against the foreign markets. From 1982 to 1990, large foreign stocks (EAFE) outperformed the large domestic stocks (S&P 500) by 2.95% per year, while small foreign stocks beat small domestic stocks by a whopping 17.46%.
In the 1980s, foreign stocks trounced U.S. stocks on an annualized basis. In the 1990s, the reverse has been the norm.
However, the situation reversed after 1990. The entire foreign advantage vanished. Through September 1997, large-foreign trailed large-domestic by 10.36% and small by 22.17%. Investors exposed to all four segments fared well throughout the entire period, but their happiness was generated by different segments at different times.
So, let’s cut in half both our S&P 500 and our EAFE exposures in our model portfolio, and replace them with two new asset classes: domestic small companies and foreign small companies.
Now we observe a very gratifying increase in rate of return. Better yet, because of the low correlation of the new asset classes to the ones we already held, we have managed to capture a fair share of the higher returns offered by smaller companies while avoiding almost all of their inherent risk. Neat stuff, huh?
Fama and French have another, more surprising, gift for knowledgeable investors. There is an additional dimension to the style question that delivers even better rewards to properly diversified investors. Check in our next installment for a truly valuable lesson.