Want to Predict Future Stock Market Returns? Measuring The Downside, or the Market’s Desire to Reduce Risk, Could Be Key, A Guest Post from researchers at Bank of Canada, Board of Governors of the Federal Reserve System, and University of Quebec at Montreal
Current asset pricing research accepts that equity market returns are largely predictable over long horizons. But what about predicting short term returns?
For years, many in the industry have looked to Variance Risk Premium (VRP) as one measure to provide superior forecasts for stock market returns over shorter time periods of less than a year. Variance risk premium is the difference between the market variance implied by option prices, and the actual variance realized over time. Since option-implied variance is on the average higher than realized variance, the seller of a futures contract based on the difference can demand a higher price from the buyer, in order to compensate the seller for taking on the position.
Much has been gleaned from findings on VRP. But recent research suggests that while VRP can be a valuable tool to look at the markets and measure the uncertainty around future variation (including capturing extreme events), the formula does not take into consideration one important aspect. Downside Variance Risk Premium (Bruno Feunou, Bank of Canada; Mohammed R. Jahan-Parvar, Board of Governors of the Federal Reserve System, and Cedric Okou, University of Quebec at Montreal) suggests that while VRP measures willingness of investors to pay to hedge against bad outcomes (capturing the vol of vol and the sense of the market’s uncertainty regarding future events), it does not take into consideration upside versus downside movements, critical to assess the market.
They instead propose a new decomposition of the variance risk premium. They assert that downside variance risk premium (versus upside) is the main component of the VRP, and most important to assess, since the downside is especially avoided by investors for its ability to severely increase the likelihood of severe losses. Investors tend to hedge against downward movements to avoid losing money. Conversely, they tend to gravitate toward upside movements and are willing to pay to get exposure to it and the potential for higher gains.
“Our research suggests that it is not enough to merely look at VRP on the whole. In order to have the best measure to help predict future stock returns, advisors, traders, and others need to further dissect this information to look at both the upside and the downside,” offers Okou. “Without doing this, we are mixing two opposing views of the market and risk not getting the complete picture.”
Researchers highlight the positive and significant link between the downside variance risk premium and the equity premium at short horizons, as well as between the skewness (upside minus downside variance) risk premium and the equity premium at intermediate prediction steps.
To compute risk neutral information, they extracted options information on the S&P 500 using OptionMetrics’ data via WRDS. “OptionMetrics offered clean, comprehensive data, with a thorough continuum of strikes–very important to us to help ensure reliable, risk neutral measures and the highest level of accuracy in our findings,” offers Okou.
Researchers were able to look at the options market to measure anticipation of future positive and negative outcomes, given the forward-looking nature of options overall.
What do these findings on VRP and downside risk mean for investors and others in the industry?
“Downside variance risk premium is an easy to compute risk (neutral vis-à-vis physical) expectation of real life variance. It’s something central banks, monitoring the financial system, can use to predict future excess return or risk aversion. It offers investors an assessment of the market in terms of good uncertainty related to bad at a given point in time,” says Okou.” Investors can benefit from looking at downside VRP.”