Working Paper: NBER ID: w10912
Authors: Pedro Santaclara; Shu Yan
Abstract: We use a novel pricing model to filter times series of diffusive volatility and jump intensity from S&P 500 index options. These two measures capture the ex-ante risk assessed by investors. We find that both components of risk vary substantially over time, are quite persistent, and correlate with each other and with the stock index. Using a simple general equilibrium model with a representative investor, we translate the filtered measures of ex-ante risk into an ex-ante risk premium. We find that the average premium that compensates the investor for the risks implicit in option prices, 10.1 percent, is about twice the premium required to compensate the same investor for the realized volatility, 5.8 percent. Moreover, the ex-ante equity premium that we uncover is highly volatile, with values between 2 and 32 percent. The component of the premium that corresponds to the jump risk varies between 0 and 12 percent.
Keywords: risk; options; equity premium
JEL Codes: G1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| filtered measures of ex-ante risk (diffusive volatility and jump intensity) (C58) | average risk premium (G22) |
| realized volatility (G17) | average risk premium (G22) |
| perceived risks embedded in options (D81) | average risk premium (G22) |
| perceived risks (D81) | ex-ante equity premium (G12) |
| jump risk component (G12) | ex-ante equity premium (G12) |
| perceived risks (D81) | equity premium puzzle (G12) |
| filtered risk measures (C46) | risk premium (G19) |