Alternative Tax Systems to Correct for Externalities, and the Efficiency of Paying Compensation

Abstract

This article has two purposes. The first is to discuss the question of exactly which commodities it is correct to tax when an external diseconomy is present, but when for some reason it is impossible to tax the externality directly. The second is to clarify the question of whether the payment of compensation to the victims of an external diseconomy affects the achievement of allocative efficiency. The first matter-which commodities should be taxed-is of interest whenever there are political or technical problems that make it excessively difficult to tax an externality directly. Many examples of such situations can be given. One is aircraft noise, which, possibly because of international conditions, cannot be taxed; then the government will want to know whether allocative efficiency can be achieved by other methods-by taxing the sale of aeroplanes to operators, by taxing aircraft movements or by subsidizing the installation of double glazing in homes and offices near the airport. Again, the externalities of car exhaust fumes cannot be taxed directly, because of the cost of metering and levying taxes; is it then sufficient to tax the consumption of petrol, or could one instead tax the raw materials used in its production, or must one do both? Another example arises in connection with central city traffic congestion; it has been suggested in at least one instance (Greater London Development Plan, 1973) that this could be remedied partially by the imposition of a central area employment tax, in addition to (or instead of) the usually recommended methods of traffic restraint or road pricing. There is no doubt about the validity of correcting for externalities by means of a tax imposed on the output of the externality itself, or by direct regulation of output. What is in doubt is the practicability of these measures. When they are too difficult one might want to consider whether alternative taxes can achieve the desired result of Pareto optimality. Plott (1966) has shown that, in the simple case where the output of an externality is a function of only one factor of production, Pareto optimality can be achieved by taxing the use of that factor alone, but not by taxing either the priced output or any other factor. This paper includes some more general results of that kind. The contention that has been made by Baumol and Oates (1975) and by Baumol (1972), that the payment of compensation will prevent the externality taxes (wherever imposed) from leading the economy to Pareto optimality, would be of great concern if it were always true, but it is shown below that it is not necessarily so. It is argued that in theory there are methods that do not have this disadvantage; the difficulty lies in putting them into practice, and to be realistic one may have to accept that the payment of compensation may be feasible only at the cost of some allocative inefficiency. In Section I the basic production model is introduced and is used to show how corrective taxation can be used to achieve Pareto optimality without the