Blog - Investing Notes
November 14, 2007 - Sources
of Volatility's Predictive Power for Stock Returns
Past research finds that stocks with low (high) short-term
historical volatility tend to outperform (underperform). What
causes this relationship? In the November 2007 update of their
paper entitled "Volatility
Spreads and Expected Stock Returns", Turan Bali and Armen
Hovakimian examine the similarities and differences between
realized (historical) volatility and implied
volatility in the context of power to predict stock returns.
Using stock price/fundamentals data for a broad range of stocks
and volatilities implied by associated options with near-term
expiration dates over the period January 1996-January 2005,
they find that:
- Confirming past research, a hedge strategy that is long
(short) the 20% of stocks with the highest (lowest) one-month
lagged realized volatilities generates significantly
negative returns (an average alpha
of -1.5% per month).
- However, a similar trading strategy based on implied
volatilities derived from the prices of options generates
insignificant returns.
- The difference between realized volatility and implied
volatility reasonably represents volatility risk. A
hedge strategy that is long (short) the 20% of stocks with
the lowest (highest) realized-implied volatility difference
produces average raw and risk-adjusted returns of 0.5% to
0.9% per month.
- The difference between the volatility implied by call
options and the volatility implied by put options reasonably
reflects the expected future price change of the underlying
stock. A hedge portfolio that is long (short) the 20% of
stocks with the highest (lowest) call-put implied volatilities
earns highly significant raw and risk-adjusted returns of
1.1% to 1.4% per month.
- These results persist after controlling for size,
book-to-market,
liquidity,
analyst earnings forecast dispersion, probability
of informed trading and skewness.
In summary, volatility-based portfolio strategies derive
their effectiveness from: (1) the difference between realized
volatility and implied volatility ; and, (2) the difference
between call-implied volatility and put-implied volatility.
For related research, see Blog
Synthesis: Volatility Effects.
