As children, we are taught to assess risk. Crossing the street against a red light is possible, but risky—doing so in rush hour traffic is even more risky.
The same logic underpins Modern Portfolio Theory’s (MPT) Capital Asset Pricing Model (CAPM) and mean-variance efficiency, or the efficient maarket hypothesis (EMH), which suggests higher risk will equal higher return. Long-term observations of capital market returns confirm that stocks outperform long-term bonds, and that those bonds, in turn, outperform Treasury bills.
This logic gave portfolio managers the framework to build diversified portfolios, just as pension and investment assets exploded with baby boomer savings. But, in 1972, A. James Heins and Robert A. Haugen released a working paper, “On the Evidence Supporting the Existence of Risk Premiums in the Capital Market,” that confounded investment professionals. The research suggested that, with equities, the idea that higher risk equalled higher return was not only questionable, but false. In subsequent work, the experts found that in all developed and emerging markets (those with sufficient data), low-volatility stocks (low standard deviation) consistently outperform high-volatility stocks.
In a 2012 paper, Nardin, Baker and Haugen measured the standard deviation of stocks in 21 developed and 12 emerging markets over 24 months, covering 99.5% of the capitalization of each country. They ranked them by deciles and calculated their performance, repeating monthly for data from 1990 to 2011. Chart 1 shows the results for the U.S. market. Note the lowest-volatility stocks in decile 1 had the highest returns, compared with the highest-volatility stocks in decile 10 with the lowest returns.
CHART 1: Low-volatility stocks had the highest annualized returns
Chart 2 shows risk versus return for each country’s lowest-risk decile, compared with highest-risk decile: note the clustering of lowest-risk stocks has generally higher returns than those with the highest risk.
CHART 2: Risk versus return
The implication is that, on a risk-adjusted basis, investors are not rewarded for holding riskier stocks. For many advisors, the implication of the evidence is good enough: buy low-volatility stocks. But understanding why higher risk does not equal higher return is important, because it may have other implications.
Baker and Haugen summarized the research. Professional analysts and portfolio managers favour popular stocks so they look good to their clients and can deflect blame if decisions are wrong. As a result, these stocks are overrepresented in institutional portfolios and tend to be overvalued. Chart 3 shows value, growth, small-cap and minimum-volatility indices in a Canadian context.
CHART 3: Value, growth, small cap & minimum volatility
With extended valuations and continuing European economic malaise, some advisors may be starting to think about defensive strategies. Ioulia Tretiakova, director of quantitative strategies at our firm, notes that from September 30, 2008 to December 31, 2008, growth outperformed value. Also, minimum volatility outperformed all other major styles during the meltdown.
One explanation for the minimum volatility anomaly is volatility drag, the compounding effect that impedes returns. Say for instance, the market is up 50% today and down 50% tomorrow. You’ll end up at 75% of the starting value. The 25% that’s lost is the result of volatility drag—the higher the volatility, the higher the drag. Table 2 (below) shows the standard deviation of various styles from 2001 to 2015 in Canada. To some extent, low volatility may help explain why the value style has been successful.
So, whether or not minimum volatility is the Holy Grail of investing, it is worth pursuing.
Table 1: How different strategies performed from Sept. 30, 2008 to Dec. 31, 2008
|Canada Minimum Volatility (CAD) (Large + Mid Cap)||-19.2%|
|Growth (Large + Mid Cap)||-19.4%|
|Small Cap Value||-20.9%|
|Canada (Large + Mid Cap)||-22.9%|
|Value (Large + Mid Cap)||-26.7%|
|Small Cap Growth||-34.0%|
Table 2: Standard deviations, 2001-2015
|Small Cap Value||15.8%|
Table 3: TSX-traded low-vol ETFs
|BMO Low Volatility||ZLB||0.40%|
|Powershares S&P/TSX Composite Low Volatility||TLV||0.34%|
|iShares MSCI Canada Minimum Volatility||XMV||0.33%|
|First Asset MSCI Canada Low Volatility||RWC||0.60%|
|Powershares S&P 500 Low Volatility CAD||ULV||0.36%|
|iShares MSCI USA Minimum Volatility||XMU||0.34%|
|BMO Low Volatility US Equity||ZLU||0.36%|
|First Asset MSCI USA Low Risk Weighted||RWU/B||0.60%|
|First Asset MSCI USA Low Risk CAD||RWU||0.60%|
|International and Global Equities||Symbol||Expenses|
|iShares MSCI EAFE Minimum Volatility||XMI||0.37%|
|iShares MSCI Emerging Markets Minimum Volatility||XMM||0.41%|
|iShares MSCI All World Minimum Volatility||XWD||0.46%|
|First Asset MSCI Europe Low Risk CAD||RWE||0.60%|
|First Asset MSCI Europe Low Risk Weighted||RWE/B||0.60%|
|First Asset MSCI World Low Risk CAD||RWW||0.60%|
|First Asset MSCI World Low Risk Weighted||RWW/B||0.60%|
|PowerShares S&P International Developed Low Volatility Index ETF||ILV||0.40%|
|PowerShares S&P Emerging Markets Low Volatility Index ETF||ELV||0.44%|
Originally published in Advisor's Edge Report
Read this article and full issues on the iPad - click here.