Explaining the International Pattern of Labor Productivity and Wages: A Production Model with Two Labor Inputs

Abstract

T HE productivity and wages of workers in any nation depend upon the quality of the workers and the relative quantities of complementary factors of production. In recent empirical work on the estimation of cross-national production functions it has generally been assumed that labor is homogeneous, and complemented by a single factor of production, capital. Reference is made particularly to the pioneering article by Arrow, Chenery, Minhas, and Solow, and the monograph by Minhas.' These assumptions, it turns out, lead to a very incomplete explanation of international differences in labor productivity and wages, and totally ignore the systematic variation of wages over industries. This simple capital-labor model, Professor Leontief argues, also leads to incorrect predictions about international specialization because labor quality differs substantially over nations.2 A natural step is to introduce labor quality into the analysis. Attempts in this direction face one major obstacle the absence of comparable data measuring the skills of workers in different countries. Consequently, we do not have at hand a direct test of the proposition that a capital-labor model augmented by the assumption of heterogeneous labor can account fullv for international differences in labor Droductivity and wages. Nevertheless, it will be shown that a more general model incorporating two imperfectly substitutable types of labor can offer a significant improvement in explaining the over-all pattern of labor productivity, wages, and international trade among developed countries and within manufacturing industries.