Abstract
This paper develops a methodology to test structural asset pricing models based on their implications for the multiperiod risk-return trade-off. A new measure, the term structure of risk, captures the sensitivities of multiperiod expected returns to structural shocks. The level and slope of the term structure of risk can indicate misspecification in equilibrium models. I evaluate the performance of asset pricing models with long-run risk, consumption disasters, and variance shocks. I find that only a model with multiple shocks in the variance of consumption growth is consistent with the propagation of and compensation for risk in the aggregate stock market.
Horizon effect in the term structure of long-run risk-return trade-offs