Acasual reader who picked up Piero Sraffa’s Production of Commodities by Means of Commodities might be fascinated by the crystalline style in which it is written, but he would be puzzled to know why it is considered to be a contribution to economic theory of the first importance, or even to make out what question is being discussed. To attempt to give an account of the significance of the book in broad terms is to rush in where angels fear to tread, for Sraffa himself gives no indication of the conclusions to be drawn from it. The sub-title is A Prelude to a Critique of Economic Theory, and the critique is still to come. Meanwhile, however, socialist readers may have an impression that Sraffa’s argument supports their political convictions. It might be helpful to try to show what bearing the one has upon the other.

The question of the determination of the relative prices of commodities in conditions of perfect competition is one of the most technical and formalistic departments of economics, yet it has always been impregnated with ideological passion because of its association with the theory of value—the contention that value is created by labour alone, hotly opposed by the contention that capital also deserves part of the credit.

To Adam Smith it appeared obvious that in ‘that original state of affairs’ when the worker had ‘neither master nor landlord to share with him’ commodities exchanged at prices which corresponded to the relative amounts of labour time required to produce them.

Ricardo set out to find the laws which regulate the distribution of the produce of the earth between ‘the proprietor of the land, the owner of the stock or capital necessary for its cultivation, and the labourers by whose industry it is cultivated’. (He was particularly concerned to show that landlord’s rent is an incubus on society, but that aspect of the problem does not concern us here.) In order to discuss the distribution of the product of industry, Ricardo had first to find a unit in which to measure it. Labour-time was the obvious unit to take, but then he came up against a puzzle. In competitive conditions, the rate of profit on capital must be the same in all lines of production. Different products require different proportions of capital to labour, for technical reasons; thus there must be different shares of profit, for different products, in net output (that is, in value added—gross proceeds minus the replacement cost of raw materials and depreciation of equipment). A difference in the rate of profit then entails a different pattern of relative prices—those products with a high capital/labour ratio falling in price when the overall rate of profit falls and those with a low ratio rising. Thus it appears that a change in distribution changes that which is to be distributed. Ricardo was struggling with this puzzle to his dying day.

Marx took over the notion that the prices of products are proportional to the labour-time required to produce them and read into it a new and striking interpretation—all products exchange at their values and this is true also of labour-power itself—for labour-power is ‘produced’ by the labour-time required to provide the subsistence of the worker. Since output exceeds the wage, labour produces more value than it receives. Thus the theory of exploitation was derived from the theory of value.