Can investors usefully predict the short-term direction of the stock market by contrasting the outlooks implied by out-of-the-money (OTM) and at-the-money (ATM) market index options. In the October 2011 update of their paper entitled “Implied Volatility Spreads and Expected Market Returns”, Turan Bali, Ozgur Demirtas and Yigit Atilgan investigate the relationship between stock market index implied volatility spread (slope of the volatility smile) and future stock market return. They consider several measures of the implied volatility spread, such as the difference in implied volatilities between the S&P 500 Index OTM put option and the ATM call option that have the highest open interest or trading volume each day. They define moneyness as the ratio of strike price to stock price, with ATM (OTM) having moneyness between 0.95 and 1.05 (from 0.8 to 0.95). They exclude options with time to expiration less than 10 days or more than 60 days, options priced less than $0.125 and options with missing or anomalous data. Using daily closing prices for S&P 500 Index options and S&P 500 Index daily opening and closing levels from January 4, 1996 through September 10, 2008, along with contemporaneous firm and economic data used in robustness tests, they find that:
- On average over the entire sample period, implied volatilities for S&P 500 Index OTM put options are about 8%-9% higher than those for ATM call options.
- The spread between the implied volatilities of S&P 500 Index OTM put options and ATM call options relates negatively to future S&P 500 Index returns (in excess of the risk-free rate). In other words, when S&P 500 Index OTM put options are exceptionally expensive (cheap) compared to ATM call options, next-day open-to-close stock market returns tend to be exceptionally low (high).
- The predictive power of the implied volatility spread persists for several weeks. Specifically, a 1% increase in the daily volatility spread reliably indicates a 0.02% to 0.10% decrease in next-week stock market return, depending on the spread calculation method. A significant negative relationship between implied volatility spread and future stock market return extends to an horizon of one month, but clearly vanishes by three months.
- The power of index volatility spread to predict stock market return is notably stronger when: (1) S&P 500 constituent firms announce earnings; (2) there is important cash flow or discount rate news; and, (3) the University of Michigan consumer sentiment index is extremely high or low.
- This predictive power persists after controlling for conditional volatility, variance risk premium, macroeconomic variables (default spread, term spread and aggregate dividend yield) and lagged S&P 500 Index return, and after correcting for the slightly later close of the options market relative to the stock market, small sample biases and non-normality in the return distribution.
In summary, the spread in implied volatilities between out-of-the-money put options and at-the-money call options on the S&P 500 Index relates negatively to future U.S. stock market return, with flow of private information from the options market to the stock market the most plausible driver of this predictive power.
Cautions regarding these findings include:
- The study does not explore any trading strategies (e.g., useful small and large implied volatility spread thresholds) to exploit the discovered stock market predictability.
- The study reports gross, not net, returns. Incorporation of reasonable trading frictions may alter findings.
- Reported returns are for an index rather than a directly tradable asset. Use of a tradable fund or derivative would boost confidence in economic significance of findings.
- It would be interesting to know the performance of implied volatility spread as a predictor during the extreme market conditions just after the end of the sample period.
For similar research on individual stocks rather than a broad market index, see “Smirking Because They Know Something?”.
