On the Distributional Effects of Income in an Aggregate Consumption Relation

In this paper we analyze the relevance of characteristics of the income distribution in an aggregate consumption relation. In particular, we use a statistical distributional approach toward aggregation in order to model the aggregate consumption relation of a heterogeneous population, and apply it to UK-Family Expenditure Survey data for the years 1974-1993. We consider a nonparametric estimation methodology, which accounts for the presence of continuous and discrete variables, and, for comparison purposes, a linear parametric method. A bootstrap test on the nonparametrically estimated parameters suggests that, in addition to the mean, the dispersion of income is also relevant. Additionally, the time-invariance of the parameters of the aggregate relation is rejected. These findings have important implications for constructing empirically sound models of aggregate consumption expenditure.