Randomization of Commodity Taxes: An Expenditure Minimization Approach
by
Fwu-Ranq Chang
Department of Economics
Indiana University
and
David E. Wildasin Department of Economics Indiana University Bloomington, IN 47405 USA
Abstract
Using an expenditure minimization approach, necessary and sufficient conditions for local random taxation are obtained in terms of the curvature of the compensated demand function, so that intuition from excess burden analysis can be applied. Major findings include: (1) random taxation is locally optimal if the compensated demand function is sufficiently convex; (2) horizontally equitable taxation is locally optimal if the compensated demand function is concave, and (3) local randomization is not optimal if the tax revenue requirement is sufficiently close to zero or to any local maximum. We also derive an inverse elasticity characterization of the optimal random tax structure.
