An Alternative Approach to the Analysis of Taxation

Posted: 30 Aug 2012

Date Written: December 1, 1976


This paper offers an extention of the traditional approach to the analysis of taxation capable of explaining observations attributable to taxation which were not previously explicable. Underlying the change in approach are the premises that every commodity is composed of a set of characteristics whose levels are economically determined and that no tax will be so comprehensively specified as to control all the characteristics of a commodity. Market adjustments to a tax will take place not only in the quantity of the good but also in the make-up of its characteristics. Let us briefly summarize the thrust of the argument. Commodities as transacted in the market are complex, and the margins with respect to which optimization takes place are numerous. Because commodity tax statutes will not generally cover all these margins, any tax will induce multiple changes not only in resource allocation away from the taxed commodity and into others but also in the "quality" of the commodity and how it is transacted, a substitution away from the taxed attributes and into the others. Adjustment in the face of incomplete specification of commodities implies upgrading of quality in the case of a unit tax and downgrading in that of an ad valorem tax. A test of the effect of taxes on cigarettes refutes the conventional hypothesis that quality will not change with a tax. The results are rather consistent with our model, which claims that the effects of unit and of ad valorem taxes will differ both from each other and from those predicted by the conventional model.

Suggested Citation

Barzel, Yoram, An Alternative Approach to the Analysis of Taxation (December 1, 1976). Journal of Political Economy, Vol. 84, No. 6, 1976, Available at SSRN:

Yoram Barzel (Contact Author)

University of Washington ( email )

Box 353330
Seattle, WA 98195-3330
United States
206-543-2510 (Phone)
206-685-7477 (Fax)

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
PlumX Metrics