Traded Intermediate Products, and the Theory of Devaluation
构建了一个包含贸易中间产品的两国货币模型，弥补了以往贬值理论忽略中间产品和货币效应的缺陷，为理解汇率变动和国际收支调整提供了更完整的分析框架。
The literature on devaluation so far has mainly centered on external dependence through trade in finished products and through capital flows. Very few attempts (e.g., Coppock [1971] and Shea [1976]) have been made to incorporate intermediate products which indeed constitute the bulk of international trade into the theory of devaluation. Coppock considers a two-country model with fixed factor proportions; but as he admits himself, his model is very special in the sense that income effects are assumed away and prices of domestic factors are kept unchanged. Shea develops a Keynesian macromodel, but his is also special because of his use of Cobb-Douglas production function and that within a onecountry framework. The main result of these works is that in the presence of traded intermediate products a modified or at least a reinterpreted MarshallLerner condition is needed for a successful devaluation. These papers however suffer from two basic limitations. First, as indicated above, they lack a full two-country general equilibrium framework. Second, they ignore the monetary effects or the real balance effects which have recently been recognized to be crucial to the theory of exchange rate changes and balance of payments. It may be remarked here that Findlay and Rodriguez [1977] do introduce a money market in their analysis of monetary and fiscal policies in the presence of intermediate imports. But the real balance effect is ignored there because the stock of money holding does not appear either in the consumption function or in the demand function for money. It is possible in their model to examine the effects of devaluation, but it is very likely to yield results the same as Shea's (see Findlay and Rodriguez [1977, p. 211]). In fact the only essential difference it has compared to Shea's model is the Cambridge-type money market equation. It is also a one-country model. The purpose of this paper is to present a two-country monetary model with traded intermediate products, that make up for these deficiencies in the earlier analysis. It belongs to the category of papers by Hahn [1959], Kemp [1962, 1969], Negishi [1968, 1972], Amano [1972] and Dornbusch [1973a, 1973b] which constitute the modern approach to devaluation and balance of payments. Within this approach itself two types of models are discernible: one with the assumption of gross substitutability (Hahn [1959], Kemp [1962, 1969] and Negishi [1968]); and the other with separability hypothesis (Negishi [1972;