Wage-earners voluntarily accept to work under the control, and for the account of, firms run by entrepreneurs1; they do not decide what, how and how much, they must produce; wage-earners are not responsible for the consequences of their activities when they comply with entrepreneurs' orders12; inside the firm, wage-earners are subordinates. Outside the firm, wage-earners freely choose the way they spend their wages in the markets for commodities and services. Such is the 'stylised fact' which characterises the wage relationship in (...) our economies. Any theory of the wage relationship should account for this 'stylised fact' by deriving it from a consistent set of assumptions and propositions. Three main propositions are advocated in this paper: Mainstream economists conceive the 'wage' as being the price of a commodity ('labour power' or 'human labour') determined by a market – as for any other commodity and service. As a consequence, the wage relationship is thought of as an exchange relation ruled by equivalence; this view cannot be derived from the usual basic assumptions without violating the very logic of mainstream theory. Following this logic: either the wage relationship should be thought of as a relation between human beings having such different conditions that it cannot be interpreted as an exchange ruled by equivalence or 'human labour' is not to be found within the commodity space. In any case, the 'stylised fact' mentioned above is not accounted for by mainstream theory. 'Wage' is the name of the payment entrepreneurs give to other people with the view to making them participate in production under their control and for their account; wage payment is neither a purchase nor a sale – it just allows wage-earners to enter the market and spend in order to acquire the commodities they desire Wage-earners are, economically speaking, a means used by entrepreneurs for their own ends; exploitation (Fleurbaey's M-exploitation) is inherent in the wage relationship. Propositions 2 and 3 do not fit general equilibrium theory (inspired by Walras and rationalised by Arrow- Debreu); they do not fit modern mainstream theory either. They have no room in the theory of prices of production (inspired by Ricardo and rationalised by Sraffa). Moreover, they are not compatible with Marx's theory of labour power. An alternative approach is needed to satisfy them. Paving the way for that alternative theory requires us to show that 'human labour' cannot belong to the commodity space from the very point of view of mainstream theory; a sharp incompatibility between assumptions and conclusions would follow, and justify resorting to an alternative approach. This is our first step. A second step is to effectively resort to a monetary analysis – the alternative approach to real analysis, according to Schumpeter in his History of economic analysis. In a nutshell, we have to discard the commodity space postulate, and to think of economic relations as money mediation. Instead of a commodity space, we presuppose money to be a unit of account ($) combined with a means of payment. Instead of a permutation of goods and services exchanged amongst people, economic relations are payments. Individuals are no longer endowments and preferences, but accounts into which payments write down quantities of $. Conditions of people may differ according to the form of money circulation – this view allows us to deal with different types of economic relations; for what matters here, the wage relationship involves a specific form of circulation not reducible to that of exchange. The final step is to suggest that exploitation is inherent in the wage relationship; such exploitation has nothing to do with justice, since there is no norm whatsoever for the level of wage; it is also quite different from the exploitation Marx thought he had unveiled. This exploitation is simply due to the special situation of wage-earners vis-a-vis the market, which a specific form of circulation makes clear. 1 'Entrepreneur' here means, following Coase, 'the person or persons who, in a competitive system, take the place of the price mechanism in the direction of resources' (Coase, 1937, p. 388). 2 David Ellerman distinguishes between factual/de facto and legal or de jure responsibility. What I mean here is economic responsibility only. Another way at looking at it may use David Ellerman's propositions. The 'Invisible Judge' market gives the entrepreneur the property of raw material and labour – and also of the finished product sold in the market. The market gives nothing to the wage-earner inside the firm; outside the firm it gives her the property of consumption of goods she acquires out of her wage. Inside the firm, the wage-earner is not economically responsible. Arguing that if she kills someone, the wage- earner is responsible has nothing to do with the 'Invisible Judge'! (shrink)

In a letter to Engels (24 August 1867), Marx says that the best of his book (Capital) are (i) the “dual character of the labour embodied in commodities” and (ii) the surplus value theory. Marx's vindication of first point is the subject of the present article.1 We contend that the « dual character of the labour embodied in commodities » is a fundamental and specific property of a commodity society. Marx is right when he calls our attention to it......