ABSTRACT
Growth theory, like several other subjects in economics, has had remarkable ups and downs in the history of our subject. It was a major focus of attention of the classical economists from Adam Smith to David Ricardo, and then of Karl Marx. Afterwards the problem of economic growth was almost completely lost sight of at the time of the so-called ‘marginal revolution’, championed by William Stanley Jevons, Carl Menger and Léon Walras. While there were notable exceptions to the rule, around the turn of the century economists were predominantly concerned with the problem of value. The interest in the problem of economic growth was reignited by a contribution of John von Neumann in the 1930s. However, a greater direct impact on the profession as a whole came from the attempts to generalize Keynes’s principle of effective demand to the long run. It was particularly Roy Harrod’s 1937 contribution that gave rise to a large literature devoted to the study of economic growth and business cycles. The ‘instability principle’ enunciated by Harrod with regard to the process of capital accumulation was countered, in the mid 1950s, by the neoclassical economists Trevor Swan and Robert Solow, who showed that on the basis of sufficiently strong assumptions the economic system would gravitate towards a steady state, with the rate of expansion equal to some exogenously given ‘natural’ rate of growth. In these models Say’s law was assumed to hold, implying the full employment of labour and full capacity utilization. At the same time Nicholas Kaldor put forward the post-
Keynesian model of growth and distribution, which also started from the assumption of full employment and full capacity utilization. With the rate of growth of the labour supply and the rate of growth of labour productivity given from outside on one hand and the growth rate of investment given by the ‘animal spirits of the investors’ on the other, both neoclassical theory and post-Keynesian theory ascertained, via different routes, the distribution of income between wages and profits compatible with the given long-term growth rate. While the former assumed a given overall saving rate and a flexible capital-output ratio (via changing proportions of capital and labour by means of which a unit of social output could be produced), the latter assumed prima facie a flexible overall saving rate (via a changing distribution of income and different propensities to save out of wages and profits) and a fixed capital-output ratio.
