Unobserved Preference Heterogeneity in Demand Using Generalized Random Coefficients

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 function structure. We therefore prove identification of Generalized Random Coefficients models, defined as nonlinear or additive 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.

Click here to download paper.