The S&P 500 has clawed back nearly all its losses since hitting its COVID-19 induced low late March. This remarkable rally of over 40% is historical considering it also ranks as one of the best 50-day runs of all time.
It is also extraordinary from a volatility perspective, and what we know about risk premiums. The volatility risk premium is compensation investors receive for providing “insurance” against changes in market volatility. Financial theory dictates that risk-averse investors dislike changes in vol, and are willing to pay a premium to protect against it.
This notion manifests itself in the options market as implied vol trading at levels above historical volatility, pictured in Graph 1. The VRP (Implied Vol – Historical Vol) is on average positive, indicating the profitability of the short vol trade. A larger VRP implies the larger the premium and profitability of a short vol position. However, implied vol occasionally undershoots actual vol, which represents losses to assets exposed to short vol.
Source: OptionMetrics
The VRP level turns out to be a strong predictor of overall market returns at an intermediate horizon. The economic intuition behind this is that implied vol embeds uncertainty regarding macro fundamentals, for which data materializes over a quarterly cycle. Graph 2 plots quarterly returns (overlapping) of the S&P 500, and the VRP level.
Source: OptionMetrics
It is important to acknowledge that any regression model only utilizing a single predictive variable for the aggregate market along with overlapping returns are going to contain inference issues. However, the simple histograms below demonstrate how different the return distribution become for positive and negative VRP values.
Source: OptionMetrics
Source: OptionMetrics
It is evident that for negative VRP values, the distribution of the following quarters’ returns becomes significantly more skewed towards negative values. The skewness of returns when the VRP is negative is -1.04 versus -0.53 for positive. Simply put, downside risk becomes larger over the next quarter when the VRP enters negative territory.
However, the negative magnitude of the VRP clearly matters as well. The market has never posted a positive quarterly return when the VRP is below -0.2, until now. Extremely negative values are associated with widespread systemic financial distress. In these cases, financial intermediaries are completely tapped out of risk bearing capacity, i.e. they are unable to take on short vol positions due to risk constraints.
There are four notable historical cases when the -0.2 threshold was breached; the selloff of 2002, the Financial Crisis of 2008, the Sovereign Debt Crisis of 2011, and the most recent Covid-19 selloff. These previous events were consistent with extreme economic and financial turmoil, along with lasting damage for institutions.
This time is extraordinary for the VRP. The VRP bottomed out on 4-17 at its lowest level of -0.59, but the quarterly return is on track to be strongly positive. A possible explanation is the limited number of extremely negative observations produced during the sample period may not be representative of the true distribution of quarterly returns following catastrophic events. Regardless, if the market continues on its upward trajectory, the Covid-19 selloff will remain a prominent exception to pricing of variance risk.