Knowledge that Transforms

The history of balance-of-payments theory since the early 1930's has been one of successive of increasing degrees of theoretical sophistication. Five stages of analysis (conceptually if not always chronologicallv) may be distinguished: the simple approach following the classic paper by Joan Robinson, the approach, the Keynesian multiplier approach, the Keynesian policy approach pioneered by James Meade, and most recently the approach stemming from the work of Robert Mundell. Differences between these approaches have occasionally been the focus of sharp controversy, most notably in the case of the elasticity and absorption approaches, and recentlv in the case of the monetary approach as contrasted with other approaches that have in common an emphasis on elasticities or the influence of exchange rate changes on trade flows via relative price changes and international elasticities. The purpose of the present note is to bring out the key differences between these alternative approaches, as exemplified by a simple case that can be illustrated by a simple diagram. The simple case is that of devaluation by a single country in a world economy so large that macro-economic repercussions of devaluation on real incomes, world money demand relative to supply, and the prices of imported goods can be ignored. A further simplification is the assumption that export supply is perfectly elastic in response to currency price (cost of production is constant) short of employment, interpreted as a specific level of total output, after which point supply is perfectly inelastic. For simplicity, also, where the analysis involves full-emplovment conditions the initial equilibrium point is assumed to coincide with exact full employment. Finally, international security transactions are assumed absent, all capital movements taking the form of money flows; and all money is assumed to be international money, to avoid problems (important in reality) of substitution between international reserve assets and domestic credit. The last assumption raises the problem that a devaluation alters the amount of money valued in foreign currency, and vice versa; this problem is ignored until the end of the exposition. Figure 1 graphs income earned from export sales X plus purchase of homeproduced goods cE against expenditure E, both measured in unit values of product (at some point below it will be convenient to assume measurement in terms of foreign currency unit

this paper contains the development of the model to be tested, while Section II provides an interpretation of the model and spells out the testing framework. Section III contains a description of the sample, while Section IV contains the empirical results and an example of their use in analyzing a particular bargaining situation. Section V provides a summary of the results, a discussion of some of their limitations, and suggestions for further research

Interest rates; Inflation (Finance)

In a recent paper in this Review, Nancy Gordon, Thomas Morton, and Ina Braden (G-M-B) presented a model of facultv salary determination and results of an empirical test of their model. In this paper I present results of a replication of their study using data from another institution and results from tests of alternative models. Faculty salary differentials can be explained in part by differences in individuals' characteristics. Jacob Mincer has stated that differences among individuals in their stock of human capital (for example, level of education, years of work experience) explains much of the variation in their earnings. College and university administrators claim that faculty salary depends on productivity in the areas of teaching, scholarship, and service. Differences in demand conditions among disciplines suggest including department or school among the factors thought to determine faculty salaries.' There is no consensus among economists on a definition of labor market discrimination. G-M-B have an implicit definition of discrim;nation contained in their model. Their defiritioni of labor market discrimination is that the sex differential in salary which is unexplained by differences in individual characteristics of age, seniority, education, rank, race, and discipline is due to sex discrimination. I propose a different definition of discrimination in that I omit rank as one of the relevant characteristics. Sex discrimination may occur through slower promotion rates for females than for males, in which case rank itself would reflect discrimination. 2 I had access to a set of data from another institution, comparable to the G-M-B data. The two institutions are both large Ph.D. granting universities with similar faculty size. The variables in both studies are the same. In Table 1, empirical results of a test of the G-M-B Model I using the new data set are presented alongside the G-M-B empirical results. Model I is as follows: