Traded Goods, Nontraded Goods, and the Balance of Payments

Abstract

HISTORICALLY, analyses of the determinants of a country's balance of payments position have centered on (1) relative price effects (the elasticities approach), (2) income effects (the absorption approach), and (3) asset effects (the portfolio balance approach). In their recent discussions of monetary and fiscal policy, McKinnon [8] and Mundell [9] examined the income and asset effects. Their small country assumption (given world prices of tradable goods) ruled out relative price effects. In contrast, recent discussions of currency depreciations and optimal currency areas have emphasized the importance of relative price changes by relaxing the small country assumption and by introducing nontraded goods. In his classic paper on optimum currency areas, McKinnon [7] defined nontraded goods as follows. 'The ratio of tradable to nontradable goods' is a simplifying concept which assumes all goods can be classified into those that could enter into foreign trade and those that do not because transportation is not feasible for them .... This overly sharp distinction between classes of tradable and non-tradable goods is an analytically simple way of taking transportation costs into account. By tradable goods we mean: (a) exportables, which are those goods produced domestically and, in part, exported: (b) importables, which are both produced domestically and imported.2 McKinnon's definition of an optimal currency area depended crucially on the relative size of the nontraded goods sector. Pearce's [11] analysis of a currency depreciation represents possibly the most comprehensive attempt to incorporate nontraded goods. Using a model with two traded goods and one nontraded good, Pearce concluded that It may well be that the success of exchange depreciation as a policy rests more upon its power to reduce the price of non-traded goods relative to those traded than upon its power to affect the real terms of trade.3 Despite this promising start, A. Krueger [4] noted in her review of Pearce's paper that To date, to this writer's knowledge, no other payments models have