Traditional asset allocators have relied on correlations to control volatility in the portfolio since Modern Portfolio Theory was first introduced in the 1950s. By combining risk assets with low or negative correlations (along an Efficient Frontier), investors hope to achieve a ‘diversification effect’, resulting in a portfolio with lower expected volatility than the weighted average volatility of its underlying constituents.
But the correlation of returns between risk assets has been notoriously unstable and trending higher for decades, reducing the effectiveness of traditional asset allocation techniques. Many cite the globalization of capital markets and the rise of indexing as contributing factors. Regardless, their reliability in terms of delivering an expected diversification effect has come under closer scrutiny in recent years, particularly in down markets when stable correlations would be most useful but have proven most elusive.
A more interesting application for correlation is the relationship to volatility. It is a well-established phenomenon that the returns of most risk assets have a negative correlation to their own volatility. Many believe that behavioral factors are at play as markets tend to slowly melt up but quickly melt down due to under and over-reactions by market participants. This inverse relationship between returns and volatility holds across risk assets ranging from equities to high yield to commodities and REITs.
But, more important, this inverse relationship also holds between the returns of these same risk assets and volatility in the market. In other words, when market-based volatility (i.e, systematic risk) is rising, the returns on most risk assets are falling. As an example, the chart below looks at the correlation of the rolling three year returns for various risk assets to the rolling volatility (measured by standard deviation) of the S&P 500 Index (as a proxy for market risk) dating back to 1991.
Source: Data from Morningstar. Correlations are based on the rolling 3 year returns to the rolling 3 rear standard deviation of the S&P 500 index from 1991-2015. Indexes are S&P 500 Index, MSCI EAFE and MSCI Emerging Markets indexes, Russell 2000 index, Nareit All-Equity REIT index, Bloomberg Commodity index, Barclays High Yield, Barclays Muni, Barclays L-T Treasury and Barclays Aggregate Bond Indexes respectively.
As you can see, the returns on many risk assets used to construct portfolios are (strongly) negatively-correlated to the volatility in the market with the exception of duration-sensitive fixed-income (munis, treasuries and investment grade bonds). The latter (i.e, duration) has been reduced in many portfolios in recent years in anticipation of rising rates (and replaced with higher-yielding, credit-sensitive fixed-income), removing an important source of ballast to portfolios. At the same time, volatility in the market has been unusually low over the past few years but is beginning to show signs of rising recently. The result may be a portfolio with many pie wedges that provides a level of diversification from one perspective but has a concentration in risk from another perspective.
© Van Etten Consulting and vanettenconsulting.com, 2016. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Keith Van Etten and vanettenconsulting.com with appropriate and specific direction to the original content.