For this commentary on asset allocation I’ll start with the widely understood justification for asset allocation and then move on to a less well-known concept that has some important implications for those using relative strength strategies.
A fundamental justification for asset allocation is the notion that different asset classes offer returns that are not perfectly correlated, hence diversification reduces the overall risk in terms of the variability of returns for a given level of expected return. Therefore, having a mixture of asset classes is more likely to meet the investor’s wishes in terms of the amount of volatility and possible returns. Asset classes such as stocks, bonds, real estate, commodities, and currencies are typically employed to construct an asset allocation.
However, many financial advisors stop there. Many think of US stocks as one asset class, bonds as one asset class, real estate as one asset class and so on. Such thinking leaves a lot on the table. What if US stocks can be broken down into several viable asset classes? Does this not have the potential to further improve the benefits of asset allocation? For example, this must-read white paper by AQR Capital makes it clear that value and relative strength are two complementary strategies. Remember that both strategies profiled in that white paper select securities from the same universe of U.S. stocks.
Anyone who goes to our website sees the following:
High relative strength stocks have historically provided high returns, but they often do not correlate very well with the broad market. For that reason, high relative strength stocks frequently appear to act like a separate asset class. From an asset allocation perspective, there is significant value in a high-return asset class that is uncorrelated with most other stocks and bonds. Non-correlated performance can help smooth out the returns in a diversified portfolio.
To demonstrate this point, consider the R-squared of our Systematic Relative Strength Aggressive portfolio (which invests in U.S. stocks) to the S&P 500.
Remember that the R-squared statistic gives the variation in one variable explained by another. It is computed by squaring the correlation coefficient between the dependent variable and independent variable (S&P 500 in this case). As shown in the table, 60% of the variation in our Systematic RS Aggressive portfolio can be attributed to the S&P 500 (therefore, 40% of it is not). This means that even though we are fishing from the same pond (although this portfolio can also invest in mid-cap US stocks) the variation of the performance is quite different from that of the S&P 500. In fact, this portfolio has a lower R-squared to the S&P 500 than the Russell 2000 (small cap US stocks), MSCI EAFE (developed international stocks), MSCI Emerging Markets, and the Dow Jones Real Estate Index! This is not a typical result—where the S&P 500 is more correlated to foreign markets than it is to another portfolio composed entirely of domestic securities! A more typical result may be found when looking at the five most popular equity funds, where their R-squared averages 0.96, according to Morningstar.
How can this happen? Well, relative strength identifies those securities that have strong intermediate-term relative strength out of a universe of securities. Often, those securities with the best relative strength are not the stocks that have the biggest influence on the movement of the S&P 500.
All financial advisors are in the asset allocation business. Some do it well. Some do it poorly. I would suggest that including relative strength strategies in your asset allocations has the potential to be very helpful from a performance and diversification perspective.
Click here to receive the brochure on our Systematic Relative strength portfolios. Click here for disclosures.
[...] Attention Asset Allocators « Systematic Relative Strength [...]