The Post-Devaluation Time Profile of Reserves and Prices under Neoclassical Assumptions

Abstract

The re-introduction of neoclassical assumptions into balance of payments theory was done by Hahn (1959) and Kemp (1962). Both of these writers employed a Patinkin-type model specialized to two countries, two commodities, two individuals and two fiat moneys. By making the assumptions that all goods were gross substitutes and normal, they were able to demonstrate the proposition that in the period in which a small devaluation by the home country takes place, (a) the balance of payments of that country improves; (b) the prices of the two commodities, expressed in home currency, increase but not proportionately with the price of foreign exchange; and (c) the foreign currency prices of the two commodities decrease, but not proportionally with the fall in the price of the home country's currency. In the first years of this decade, a further theorem was discovered which provided the rationale for the Hahn-Kemp results. It is that (1) a devaluation is equivalent, in its effects on the model's real variables and monetary variables expressed in home currency, to an equiproportional increase in the foreign money stock of the same percentage amount as the increase in the price of foreign currency; and (2) a devaluation is equivalent, in its effects on the real variables and the monetary variables expressed in foreign currency, to an equiproportional decrease in the home money stock of the same percentage amount (see e.g., Kemp, 1970; Kuska, 1972). Now (1) implies that the world money stock, measured in home currency, increases but proportionally not by as much as the price of foreign exchange; and (2) implies that the foreign currency value of the world money stock decreases, but not by as much proportionally as the devaluation. Therefore, in the long run, on price-specie-flow considerations, one would expect the Hahn-Kemp results to hold; and that this is so was demonstrated in Kuska (1972, p. 315) for the general case in which there are arbitrary numbers of commodities and individuals in the world economy. The Hahn-Kemp results just state that these long-run properties also hold in the initial post-devaluation period in the two-commodity, two-individual case-if the goods are gross substitutes and normal. Even aside from the fact that the Hahn-Kemp theorem has been demonstrated only for a restricted case, the resulting situation is not entirely satisfactory. We know that, in the long run, providing the system is stable, a devaluation will increase the home country's reserves, and we know what it will do to prices. If we make the Hahn-Kemp assumptions, we also know that qualitatively similar results happen in the initial period for their specialized model. We do not know, however, whether their results hold in the general case; and further, we do not know what occurs in the interim