Time-Varying Relative Risk Aversion : Theoretical Mechanism and Empirical Evidence

This paper explores the issue of understanding time-varying relative risk aversion with household-level data. First, we derive an analytic form for a parsimonious portfolio choice model with the preference given by Greenwood, Hercowitz and Huffman (1988, GHH), and then, the analytical solution identifies four partial equilibrium effects in our model with the GHH preference on risky shares through two channels and two net effects whose signs hinge on the value of a key structural parameter. Based on household-level data, our empirical results from both mean and quantile regression models show that wealth negatively affects risky shares and the estimated effects are statistically significant and robust, which is in line with the theory. Finally, we conclude that the GHH preference alone is not sufficient in explaining how risky shares respond to labor income in the household-level data about portfolio choices