Housing, Household Portfolio, and Intertemporal Elasticity of Substitution

Abstract

This paper investigates the impact of the inclusion of housing in a household portfolio on household's intertemporal decision making. Residential housing is one of the principal assets households hold, and thus changes in housing return can affect household consumption over time. We assess whether the inclusion of housing in the household portfolio affects one of the important parameters of the intertemporal choice, the intertemporal elasticity of substitution (IES). The IES measures how a change in asset return affects household's consumption growth. Since the use of aggregate time series data presents potential aggregation problems, we estimate a consumer model using household-level data, in particular the Consumer Expenditure Survey (CEX), and thus account for household heterogeneity and demographics. Moreover, utilizing a household-level data set, we estimate IES parameters for different groups of assetholders: stockholders, bondholders, and homeowners. Our results indicate that a higher housing return positively affects consumption growth, and housing is an important asset to account for in the household portfolio. The estimation with the portfolio return that includes housing results in the IES of about 0.3, which is lower than that obtained using the Treasury bill rate. The estimation is also more robust to alternative sets of instruments and for different groups of assetholders.