First of all, it is not the case that Sraffa has only just acquired an important place in discussions of economic theory. In an article he wrote in 1926 on the laws of variable return, he was already pointing towards a critique of the dominant theory of value.footnote14 At the time, however, more interest was shown in some of his remarks on the theory of the firm—remarks that were later developed by Joan Robinson in her theory of imperfect competition. But the work by Sraffa which is today at the heart of discussion belongs to the period since 1951. I am referring to the edition of Ricardo which he produced for the Royal Economic Society, and his book Production of Commodities by means of Commodities. These works are central to the two closely-linked themes that characterize the present situation of theoretical research: namely, the critique of modern value and distribution theory, and the revival of the very different approach to these problems adopted by the British classical economists down to Ricardo, and taken over by Marx for his ‘critique of political economy’.

I mean theory based on the marginal method that has held almost undisputed sway over economic thought since the last quarter of the nineteenth century. At its heart lie the twin concepts of ‘marginal utility’ (i.e. the increment of satisfaction derived from a unit-increment of consumption of a particular good) and ‘marginal product’ (i.e. the increase in output associated with a unit-increment of the ‘factor of production’ applied). By correlating a decrease in utility and marginal product with an increase in the good and the factor of production respectively, this theory has sought a rational foundation able to sustain the notion of ‘demand’ for ‘factors of production’ (traditionally, labour, capital and land)—demand which is supposed to determine, by coming together with the corresponding ‘supply’, the return on the various factors of production. This coming together or ‘equilibrium’ of the demand and supply of factors of production involves, in turn, similar equilibria on the output markets, thereby determining the price of various products.

This theory underpins the idea, so often employed even in recent discussions of economic policy, according to which free market prices reflect the ‘scarcity’ of the ‘factors of production’ and the products in question. Such prices are then supposed to guarantee—in a way which we cannot discuss here, but which depends on the validity of the theory itself—the ‘efficiency’ of the economic system based on competition. Indeed, it is argued that the system of free competition would have to be copied even in a socialist economy.

Yes, marginalist theory, with the whole of its complex analytical structure, has entered into crisis as a result of two developments in theory. The first goes back to the late thirties and is the result of Keynes’s work, which was itself not dissociated from the great economic crises of those years. It was then recognized that the tendency of labour supply and demand to ‘balance out’—a tendency to which marginalist theory referred in its determination of wages, profits and prices—could not in fact be assumed to operate, at least in the short run. However, Keynes’s critique did not attack the premises of orthodox theory; on the contrary, he explicitly accepted them.

It was the second development in theory which was to rebel against these premises, by criticizing the conception of capital as a ‘factor of production’ to be measured, in the last analysis, as a quantum of value. This critique has already demonstrated the false nature of some basic marginalist propositions—for example, that which postulates a direct relationship between ‘capital intensity’ and real wage-rates. It has even shown that a production technique which has been abandoned because of a rise in wages may profitably ‘return’ after wages have undergone a further rise. This possibility, of course, directly contradicts the proposition according to which wage-rises lead to the introduction of more capital-intensive techniques, thereby reducing the level of employment. And it was these propositions that were supposed to provide the rational basis for declining ‘demand curves’ of labour (and the other factors of production), as well as for the notion that such curves ensured ‘stable equilibria’ and thus bore out theories explaining distribution of the social product in terms of supply and demand.