Reconciling Conflicting Evidence on the Elasticity of Intertemporal Substitution: A Macroeconomic Perspective

In this paper we reconcile two opposing views about the elasticity of intertemporal substitution in consumption (EIS). Empirical studies using aggregate consumption data typically find that the EIS is close to zero (Hall, 1988). Calibrated models designed to match growth and fluctuations facts typically require that the EIS be close to one (Lucas, 1990). This apparent contradiction is resolved when two kinds of heterogeneity are acknowledged: One, the majority of households do not participate in stock markets; and two, empirical evidence indicates that the EIS increases with wealth. We introduce these two features into a standard real business cycle model. First, limited participation creates substantial wealth inequality as in the U.S. data. Consequently, the properties of aggregates directly linked to wealth (e.g., investment and output) are mainly determined by the (high-EIS) stockholders. At the same time, since consumption is much more evenly distributed in the population, estimation from aggregate consumption uncovers the low EIS of the majority (i.e., the poor).