Making Sense of Low Volatility Investing
January 31, 2013
by Feifei Li
of Research Affiliates
Robert A. Haugen, who passed away this month, found that stocks with lower price fluctuations tend to outperform riskier ones. This article summarizes several hypotheses advanced to explain the persistent low volatility anomaly, reviews the performance of low volatility strategies in up and down markets, and shows how adding a low volatility component can reshape a portfolio’s risk–return profile.
My heart was saddened when I learned that Professor Robert A. Haugen passed away Sunday, January 6, 2013. When given the opportunity to write this issue of Simply Stated, I could not resist calling out to our industry to remember and to pay homage to “the father of low volatility investing.” Bob was a respected scholar who spoke passionately about market inefficiency and the low volatility anomaly over the past half century. Together with Dr. James Heins, he discovered in the late 1960s and early 1970s that, contrary to the prevailing theory, low-risk stocks actually produce higher returns. As a researcher and a product specialist in low volatility strategies, I have often cited their seminal paper,Risk and the Rate of Return on Financial Assets: Some Old Wine in New Bottles (Haugen and Heins, 1975). I have also had the pleasure of speaking and corresponding with Dr. Haugen’s long-time associate, David Fallace, who shares his passion on topics related to low volatility. Dr. Haugen, according to David, cared deeply about educating investors and helping them make better decisions. I am saddened that I won’t have the opportunity to collaborate with Dr. Haugen on my low volatility research, but I know that his research and his passion will remain a significant influence on my work.
·The leverage aversion hypothesis holds that many investors who demand high returns are leverage constrained and choose to increase their expected returns by overallocating to high beta stocks, even if the latter have demonstrably lower Sharpe ratios. (See Black, Scholes, and Jensen, 1972; Frazzini and Pedersen, 2011.)
Some market participants view low volatility investing as a special case of smart beta strategies because part of its outperformance relative to cap-weighted benchmarks comes from moving away from cap-weighting and thereby decoupling prices and portfolio weights. However, the risk-reducing mechanism of these portfolios also involves dropping growth-oriented glamour stocks, whose exclusion tends to ramp up the tracking error tremendously. Over the 138-month period from July 1991 through December 2012, the low volatility portfolios had tracking errors of 9–10% against the core equity indexes, as shown in Panel B. Other mainstream smart beta strategies typically exhibit tracking errors of approximately 3% against core equities. In this context, low volatility investing is unique among smart beta strategies and would mostly appeal to investors who seek to maximize their portfolio’s Sharpe ratio, earning a high return per unit of total portfolio volatility, and who are less concerned about the information ratio, which is dictated by tracking error relative to the benchmark.
Table 2shows that low volatility portfolios tend to have higher returns, lower risks, and higher Sharpe ratios than traditional large-cap portfolios across different measurement periods. The Sharpe ratios of low volatility strategies dominate those of large-cap core equity indexes over past 5-, 10-, and 20-year timespans in both the United States and the rest of the developed world. The only asset class which displayed a stronger risk-to-return performance is Core Fixed Income, represented by Barclays Capital Aggregate Bond Index, which has benefitted tremendously from the secular decline of rates over the past 30 years. Thus, low volatility investing appears to be an attractive option to anchor the investor’s equity allocation in the long run. In addition, low volatility portfolios have average correlations of 0.4–0.5 with other major asset classes, whereas traditional large-cap equity strategies have average correlations above 0.6. The lower correlation with other asset classes suggests that replacing the traditional cap-weighted equity allocation with a low volatility equity portfolio in your asset mix can improve the overall portfolio diversification.
A comparison of efficient frontiers vividly illustrates the benefit from greater diversification as shown inFigure 1. The efficient frontier including all the “other” asset classes listed in Table 2 is plotted in orange; the portfolio yielding the highest Sharpe ratio (0.82) is marked by an orange dot at the intersection of the orange hyperbola and the dark grey dotted line. Adding the U.S. Low Volatility and Developed ex U.S. Low Volatility strategies causes the efficient frontier to shift upward and to the left, where it appears as the solid blue curve. Investors seeking the highest Sharpe ratio (0.98) would implement the asset mix represented by the blue dot.
© Research Affiliates