Since the groundwork behind Modern Portfolio Theory was laid fifty years ago, it has been axiomatic that riskier portfolios should expect to be compensated with higher returns. In recent articles, I have shown that a portfolio of long tenured dividend payers in both the United States and Canada have experienced higher trailing returns with lower variability of returns than their respective broader markets.
A recent paper by Nardin L. Baker of Guggenheim Investments and Robert A. Haugen of Haugen Custom Financial Systems entitled "Low Risk Stocks Outperform within All Observable Markets of the World" attempts to refute the widely held belief that investors are compensated for holding a riskier portfolio on average over long time intervals. The study covers thirty-three countries' equity markets from 1990-2011. Beginning with January 1990, the authors computed the volatility of total returns for each company in each country over the trailing 24 months, and then bucketed the stocks by volatility. Returns over the next one month were computed for each volatility bucket, and then the stocks were re-ranked by volatility and the returns for the next month were observed, and this process repeated for 264 months. The difference in volatility of the total annual return between the lowest and highest volatility deciles, the difference in total return between lowest volatility deciles and highest volatility deciles, and the difference in Sharpe Ratio (here return/standard deviation of return) for the developed markets in the dataset are graphed below.
Source: Baker and Haugen (2012)
Below is a rolling three-year difference in return between the lowest risk decile and the highest risk decile for the countries in the dataset.
The results appear revolutionary given that returns have always been believed to be compensation for bearing risk. The return profile above would run counter to the Efficient Market Hypothesis, and presume that high volatility stocks are relatively overvalued.
I do have several reservations about the data and its purported conclusions. First, twenty-one years in a very poor period for equity returns may not be an appropriately large sample set (although the authors claim that this phenomenon held in other long time periods.) Second, the author's backtested results that show that actual realized returns from lower volatility stocks can not necessarily be inferred to portend higher future expected returns for this strategy. Third, there is no justification for the two-year look back in volatility or the one-month rebalancing. Fourth, the author's assumption that high volatility stocks see higher demand because of the option-like nature of active manager's compensation runs counter to the earlier statement in the article about investors moving "trillions of dollars in equity investment into capitalization weighted equity indexes like the S&P 500" since passive indexers are not compensated in this asymmetric manner. The narrative about the typical investment meeting when stock picks are presented to the CIO is just a story to tie a hypothesis to the data. The chart that showed that stocks with high institutional ownership was more interesting, but causation and correlation here might be confused. Fifth, the article ends with a shot across the bow to academia and its groupthink, which cheapens the data in my opinion.
While I offer several negative counterpoints to the article above, I do commend the outside the box thinking that led the authors to even attempt to counter several decades of finance frameworks. I will continue to research other low volatility strategies like my articles on Dividend Aristocrat investing (replicated best in the United States through SPDR S&P Dividend ETF (NYSEARCA:SDY)). Perhaps new low volatility ETFs like Powershares S&P Low Volatiltiy (NYSEARCA:SPLV) and iShares MSCI USA Minimum Volatility (NYSEARCA:USMV) that attempt to pick less volatile stocks from a broader index like the S&P 500 (SPY, IVV) will prove to offer investors lower volatility but still sufficient returns and improved Sharpe ratios. I do not expect that any one strategy is preferable in dynamic markets, but believe that this work could have important portfolio management applications, and believe that it should be analyzed in greater detail. The pursuit of higher risk-adjusted return strategies dominates my personal and professional market pursuits, and I encourage Seeking Alpha readers to read the linked paper and offer up their own opinion and future analysis.
Disclosure: I am long SPY.