Can Output Explain the Predictability and Volatility of Stock Returns?

In this paper we have studied the ability of relatively standard equilibrium asset pricing models to explain two important empirical regularities of asset returns extensively documented in the literature: i) returns can be predicted by a set of macro variables; and ii) returns are very volatile. Those empirical regularities are relevant because they have often been used to reject market efficiency. In the analysis we have made use of the approximation technology in the solution of intertemporal asset pricing models recently developed by Campbell (1993) in the form suggested by Restoy and Weil (1997). We have obtained evidence from eight OECD economies using both quarterly and annual observations. Equilibrium models seem generally to find fewer difficulties in explaining the volatility of returns than their predictability for general output processes. In the case of the United States, for annual frequencies the observed predictability and volatility of asset returns are broadly compatible with the predictions of equilibrium models for a reasonable specification of preferences.