Knowledge that Transforms

EGIONAL differences in per capita income have been the object of continuing speculation, and several attempts have been made to isolate and measure the factors with which the differences are associated.' A more precise knowledge of these factors will contribute to our understanding of the forces which determine a particular income aggregate. Moreover, the policy implications of regional differences in per capita income, particularly regarding the expanding field of federal contributions to state and local government activities, requires a thorough analysis of the sources of these differences. The Census of Manufactures, 1947 required, for the first time, each manufacturing plant to supply information on production worker manhours as well as on the number of employees, wages, salaries, products, and other familiar measures of manufacturing.2 The availability of information on both man-hours and wages of production workers permits computing their average hourly earnings by detailed industrial and geographic classifications. This provides an opportunity for a more precise analysis than heretofore possible of the regional differences in manufacturing wages and in their effect on per capita income. In the present article, an attempt is made, first, to measure the differences among states in production-worker hourly earnings which are associated with differences in the composition of the state's manufacturing activity and, secondly, to measure the differences in production-worker hourly earnings associated with the differences among states in the wages paid production workers in the same industry. These differences are then related to the per capita income of each state and geographic division.

SCHUMPETER'S theory of interest, which was fully expounded in the first edition of his Theory of Economic Development (I9I2) but had been clearly foreshadowed in his first book, Das Wesen und Hauptinhalt der theoretischen Nationalikonomie (I908), is fairly well known, although it has not been widely discussed in English. Nor has it been widely accepted, but most critics recognize, explicitly or implicitly, that the extreme version of the theory of interest is not an essential part of Schumpeter's dynamic system.' A thorough understanding of Schumpeter's views on the problem of the interest rate requires, I believe, that we distinguish between an extreme and a less extreme version of his theory. The extreme version culminates in the proposition that in a stationary or quasi-stationary economy, in the Kreislaufwirtschaft, the rate of interest would be zero, and that the positive rate which we observe in reality is entirely the result of the well-known dynamic mechanism that Schumpeter has described and analyzed so brilliantly. The less extreme version admits that there would exist a positive rate of interest in the stationary economy, but insists that dynamic forces not only are likely to raise the interest rate above its stationary level but add, qualitatively, entirely new features to the static picture.2 The extreme version of his theory is hardly acceptable. Although Schumpeter spent much time and effort in defending it (e.g., in his famous controversy with Bohm-Bawerk),3 he frequently made remarks which indicate clearly that he was aware of the fact that this version was by no means essential for his dynamic mechanism. On the other hand, adherents of what might be called the ruling theory of interest B6hmBawerkians, Fisherians, Knightians, etc. (the differences between them are minor, at any rate much less important than the fierce controversies in which they were or still are embroiled would suggest) might well admit that there are few branches of static, equilibrium theory that require such drastic alterations, in order to preserve a semblance to reality under realistic dynamic conditions, as does the static, equilibrium theory of interest. Schumpeter was always keenly conscious of, and felt most uncomfortable with, the unreality of many assumptions underlying most static theorizing on the interest rate: existence of a uniform rate, absence of uncertainty, free capital market in the sense that everybody can borrow as much as he wants to at the ruling rate. These are, indeed, most unrealistic assumptions which have far-reaching implications.