ON THE LOGICAL CONSISTENCY OF SRAFFA'S ECONOMIC THEORY: A COMMENT ON SAVRAN AND STEEDMAN

In a recent article in Capital and C/ass, Savran (1979, p. 131) argues that '... Sraffa's economics is inconsistent and untenable even within its own theoretical framework ' and '... that logical consistency requires that Sraffa's theory itself be abandoned ' (Savran, 1979, p. 137). In neo-Ricardian analysis, the wage/profit frontier for the economic system as a whole is constructed using data for the technological conditions of produc-tion in all industries (Figure 1). The prior specification of one of the dis-tributional parameters (say the wage rate, w a) determines the value of the other distributional parameter (ra) and the prices of all the commodities. Savran maintains that it is not logically possible to 'close ' Sraffa's system in this way. His main objection is that it cannot be known a priori whether a wage rate which has been so determined will be high enough to cover subsistence consumption. Steedman (1979) states that such criticism is misplaced and that Sraffa's system does not suffer from these alleged logical flaws. Whilst we agree with many of the points that Steedman makes, we feel that his summary treatment of Savran's argument is a little misleading. Steedman, quite rightly, treats Sraffa's book, Production of Com-modities by Means of Commodities, as having as its central concern:... the relationships which necessarily hold between wages, profits and prices, for given conditions of production, when the wage rate, the rate of profit and the price of each commodity are uniform throughout the economy (Steedman, 1977, p. 13)