First of all, assessing how much labour value each worker contributes to production is a very tricky business. Not only because skilled labour should create more value than; unskilled labour and hence because an adequate reduction procedure of complex to simple labour is being presupposed. But even more because how much value a worker contributes in a given time depends on how productive he is compared to other workers producing the same goods.And while this productivity can in principle be assessed in the case of workers who independently produce identifiable products, it cannot, even in principle, in the general case in which goods are the joint products of a large number of operations by a large number of workers. Conse- quently, it is in most cases impossible to say whether the socially necessary labour performed by a’ particular worker (or group of workers) was smaller or larger than the number of hours he actually worked, or than the value embodied in the goods he consumes 1 2. Secondly and more fundamentally, choosing socially necessary labour, rather than actual labour, as an ethical principle of distri- bution, is highly questionable. Why should someone who happens to be less skilfull than average, or to work on particularly poor land, or in a firm using obsolete machinery, be morally entitled for that reason to a smaller part of the social product? Surely, if work is relevant at all to the determination of how much a particular worker is entitled to, it should be the work he has actually performed and not the work which would have been necessary to someone with average skill to produce the same goods under average technical conditions.

Just think of a situa- tion in which there are widely different kinds of labour: some labour is pleasant, interesting and safe, while some other labour is un- attractive, boring and dangerous. Would it not be deeply unfair to reward both sorts of labour at the same rate? Surely, some way of weighting different kinds of labour must be found if “To each according to his labour” is going to be at all plausible as an ethical principle. And what criterion could be found for this purpose other than the average disutility associated with each kind of labour? The proceeds of production, so the underlying principle should go, are to be distributed according to desert, and desert is determined by how much disutility each contributor to production has had to suffer 1 3

Furthermore, even under equilibrium conditions, standard exploitation need not derive from wealth ownership or coercion. For suppose wealth is equally distributed, but preferences for leisure vary across individuals. Some members of the society concerned may be lazy, or disabled, or scornful of worldly goods. And they may therefore be content to earn a modest living out of the interest they get from lending their share of wealth to others. Despite the fact that all are equally endowed (and hence that no one can take advantage of his superior wealth position), the latter are standardly exploited by the former. Indeed, such a situation could even arise if the’ non- workers had less than average wealth. They would be standard exploiters, and still withdrawing with their per capita share of social wealth would make them better off, i.e. they would be capitalistically exploited in Roemer’s sense 1 7. This shows that (equilibrium) standard exploitation may derive from differences in preferences as well as from differences in wealth or from coercion. Even at equilibrium, therefore, standard exploitation may be present while capitalist-or-feudal’ exploitation is not. And even if it can be shown that there is something wrong with the latter, it would not ipso facto show that there is something intrinsically wrong with the former.