The Losing Trade In The Stock Market Is High Risk

| About: PowerShares S&P (SPLV)

Summary

It has been axiomatic in finance that higher risk should be compensated with higher expected returns.

Digging back into a long-run stock market data set, this article examines portfolios formed based on their trailing volatility.

In reality, the highest risk stocks have not only failed to deliver higher returns - they have not delivered even positive returns over long-time intervals.

Over the last fifty years, introductory finance classes have universally introduced the Capital Asset Pricing Model (CAPM) to students. The model is used to determine a theoretically appropriate return for that asset's non-diversifiable risk when added to a well-diversified portfolio. In layman's terms, the model suggested that higher expected returns were a function of higher risk. This sensitivity to market risk is often represented by the quantity beta. Owners of higher beta stocks should expect higher average returns as compensation for taking more risk. Conversely, lower beta stocks should expect lower average returns for their less risky profile.

Over a long period of time that nearly mirrors the post-CAPM period in domestic equity markets, higher risk stocks have actually experienced lower returns. Market participants may have been expecting to be paid higher returns in exchange for withstanding the higher volatility of riskier stocks; they have experienced the volatility but not the returns.

In The Graph All Dividend Investors Should See, I examined a dataset from Dartmouth professor Kenneth French and illustrated that investors should prefer dividend paying companies that pay healthy and sustainable levels of dividends. Over a data period stretching nearly ninety years, stocks that paid no dividends or stocks that paid amongst the highest dividend yields in the market produced inferior returns with above market volatility. Digging back into the French datasets, I wanted to highlight another long-run study that illustrates an important market anomaly that I have sought to describe to Seeking Alpha readers.

This additional dataset breaks the domestic stock market down into sub-portfolios based on the variance of trailing returns over the past sixty days. This portfolio formation allows the researcher to look at the return profile of securities that have experienced different levels of risk - as measured by the stock's variability - over long time periods.

I have buried the lede in setting up this analysis. The title of this article reflects the curious fact that from 1964-2015, investors who put money into the highest variance decile of the U.S. stock market produced an average annualized total return that was negative. A losing trade indeed. Not only were investors not paid for this higher risk as expected in CAPM, they were not paid a return at all!

The graph and tables below illustrate the risk-adjusted returns of the portfolios formed based on realized variance.

Click to enlarge

Note that the most volatile decile above returned an average annual return of -0.54% with volatility more than double that of the stock market. You can see the cumulative return profile of these portfolios over the sample period (1964-2015) below:

The line graph above illustrates the underperformance of the two most risky deciles, but may abscond the tremendously lagging returns of the highest risk decile. Below, I have taken the same information and graphed it logarithmically to highlight this underperformance.

Low Volatility Anomaly

Regular readers know that I authored a multi-part series on the Low Volatility Anomaly in July 2015, and have added some more articles recently on this topic. These articles included an introduction to the concept, a theoretical underpinning for the anomaly, cognitive and market structure factors that contribute to its long-run performance, and empirical evidence that demonstrates the outperformance of low volatility strategies across markets, geographies and long time intervals.

I have depicted the relative outperformance of low volatility strategies using the graph below which shows the cumulative total return profile (including reinvested dividends) of the S&P 500 (NYSEARCA:SPY), and the indices underpinning the PowerShares S&P 500 Low Volatility Portfolio ETF (NYSEARCA:SPLV), and the PowerShares S&P 500 High Beta Portfolio ETF (NYSEARCA:SPHB) over the past quarter century plus. The volatility-tilted indices are comprised of the one-hundred lowest (highest) volatility constituents of the S&P 500 based on daily price variability over the trailing one year, rebalanced quarterly, and weighted by inverse (direct) volatility.

Click to enlarge

Source: Standard and Poor's; Bloomberg

With the high beta index and low volatility indices tracking the one hundred highest and lowest volatility stocks, we can compare this to the French data by combining the ten deciles into five quintiles.

Click to enlarge

Like in the Standard and Poor's indices, the most volatile quintile lags the broader market materially. The least volatile quintile still produces slightly above market returns, and has the best risk-adjusted returns of the five cohorts. In fact, the relationship between return per unit of risk falls in each quintile as you move out the risk curve.

The S&P dataset and the French dataset have slightly different populations (S&P is 500 Index constituents and French included 3,531 companies in the 2015 portfolios.) S&P forms their volatility tilts based on trailing one-year returns while the French uses trailing sixty days. The S&P index returns are volatility-weighted and the French data is value-weighted.

However, the takeaway for Seeking Alpha readers should be clear. This is an additional dataset that demonstrates the relative outperformance of low volatility stocks over long time intervals. The driver of this outperformance is in part just not owning the most volatile stocks in the market. Not only have those stocks not delivered higher returns consistent with their higher risk - they have not delivered positive returns at all over the past fifty years!

Disclaimer: My articles may contain statements and projections that are forward-looking in nature, and therefore inherently subject to numerous risks, uncertainties and assumptions. While my articles focus on generating long-term risk-adjusted returns, investment decisions necessarily involve the risk of loss of principal. Individual investor circumstances vary significantly, and information gleaned from my articles should be applied to your own unique investment situation, objectives, risk tolerance, and investment horizon.

Disclosure: I am/we are long SPLV, SPY.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.