Latin American Trade Policies in the 1970's: Reply

Abstract

I am very pleased indeed with Professor Balassa's full agreement with my policy recommendation for Latin America.' At the same time, I am rather perplexed at his reaction to my introduction of the concept of an inefficiency illusion. The essence of my argument is that, when international competitiveness is used as an index of efficiency in the context of a multiple exchange rate system, deflation by the financial exchange rate will overstate noncompetitiveness. Instead, costs should be deflated by the exchange rates affecting the different inputs concerned. Since the input mix varies across sectors, each sector's cost exchange rate will normally also be different. Balassa argues for deflation by a hypothetical free trade rate; however, such a rate is inappropriate because it simply does not reflect the exchange rates actually affecting the inputs. Indeed, it is not at all clear what each actual nominal tariff rate deflated by a hypothetical free trade exchange rate would mean. It certainly would not measure the incentive given the particular commodity involved that depends on all the other tariffs, since we must look to distortions in relative prices for effects on factor allocation. Moreover, relative price distortion could be read off undeflated data as easily as off deflated data. It also does not measure the cost of protection, since again that is a function of the distortion in relative prices, not of a single price in comparison to some free trade exchange rate. Thus, again undeflated data would do perfectly well. In sum, all each deflated nominal tariff would measure is the absolute price increase (decrease) resulting from the imposition of the tariff system. And that does not tell us much of anything.2