ABSTRACT
It has long been acknowledged in the literature (see for example Robinson 1952: 42ff; Lundberg 1967:2 – 3; Phelps Brown 1981; Brems 1986:261ff; 1989:170ff; Niehans 1990:451ff; 1995:7, 19; Samuelson 1993:522) that Gustav Cassel put forward in his Theory of Social Economy (1918, tr. 1923 and again in 1932 from the fifth German edition; see Cassel 1932:61 – 3) the Harrod-Domar consistency conditions for steady growth-that is, g=s/v and g=n, so that n=s/v, where g is the rate of growth of income, s is the proportion of income saved, v is the capitaloutput ratio, and n is the rate of growth of population and labour force (see for example Hahn and Matthews 1964:783 – 4; Solow 1988:8 – 9). It has also been suggested in the literature that Cassel shared with the so-called Harrod-Domar model the assumption that the technology of the economy has fixed coefficients.1 The notion that the Swedish economist assumed an exogenous capital-output ratio is behind the interpretations that either he had in mind an unstable economy with no mechanism to prevent increasing labour shortage or increasing unemployment (Brems 1986), or that his object was an economy on a fullemployment growth path, which however would be reached only by a fluke or by ‘accident’ (see Tinbergen and Polack 1950:125).
