NONTRADED INTERMEDIATE PRODUCTS AND THE MEASUREMENT OF PROTECTION 1

Abstract

THE purpose of this article is to consider the treatment of nontraded inputs in the calculation of effective rates of protection (ERPs). Three methods for dealing with this problem have been proposed-the Balassa, Corden, and Scott methods-and they will be compared here.2 In a model where nontraded products are used for intermediate purposes only (as considered in Part I), the Corden method will be shown to succeed in predicting changes in the allocation of resources between industries producing traded final commodities. A further advantage of the Corden method, it will be shown, is that it is practical, because it eliminates the endogenously determined price of the nontraded intermediate product. The main disadvantage of the Corden method, it will be argued, is that if we insist on including the nontraded product industries as separate industries, it can, per se, tell us nothing about resource allocation. The advantage of the Scott method is that-at least theoretically- it may be able to predict resource allocational changes brought about by changes in the tariff structure, between all the industries (i.e. industries producing nontraded intermediate products and final traded commodities). Finally, the Balassa method requires the assumption that stocks of the nontraded intermediate product exist and that the government or some other authority stabilizes the price of the nontraded intermediate product by offsetting purchases and sales. In considering a model where nontraded products are used for both intermediate and final purposes (see Part II), the main conclusion drawn is that the Scott method is superior to the other two (from both theoretical and practical viewpoints), because it succeeds in predicting changes in resource allocation between gross output industries (defined in the input-output sense) and because the endogenously determined price of the nontraded product cannot be eliminated. Part II also proposes a revised version of the Corden method which may be used to predict changes in resource allocation between net rather than gross output industries, where these terms are used in the same sense as in input-output economics.