Unobserved Preference Heterogeneity in Demand Using Generalized Random Coefficients

Abstract

We model unobserved preference heterogeneity in demand systems via random Barten scales in utility functions. These Barten scales appear as random coefficients multiplying prices in demand functions. Consumer demands are nonlinear in prices and may have unknown functional structure. We therefore prove identification of Generalized Random Coefficients models, defined as nonparametric regressions where each regressor is multiplied by an unobserved random coefficient having an unknown distribution. Using Canadian data, we estimate energy demand functions with and without random coefficient Barten scales. We find that not accounting for this unobserved preference heterogeneity substantially biases estimated consumer-surplus costs of an energy tax. JEL codes: C14 D12 D13 C21 Keywords: unobserved heterogeneity, nonseparable errors, random utility parameters, random coefficients, equivalence scales, consumer surplus, welfare calculations. The authors wish to thank Jinyong Hahn, Richard Blundell, Stefan Hoderlein, Rosa Matzkin, and anonymous referees for helpful discussions and suggestions. Earlier versions of this paper were circulated under the