It seems important to work out what Marx means by value. Joan Robinson, in Economic Philosophy, proposed that value is a pseudo-concept with no factual content, used by Marx not to explain any facts about capitalism but only to express his feelings towards it. Take away the feelings around the concept and very little is left.

David Ricardo had employed the labour theory of value to explain relative prices: commodities exchange at prices proportionate to the labour that went into them. He then struggled with the rate of profit — what most people today would call the rate of interest. To be worth producing, commodities produced using the same amount of labour over longer and shorter time periods must sell at different prices, to account for the compounding of interest.

For Marx the problem of profit becomes a problem of surplus value. When a capitalist applies capital and labour to produce a commodity and realise a profit, Marx insists, the capital doesn’t create any of the surplus value that makes up the profit. Labour is solely responsible for this, and the capitalist simply plunders some of what the workers produce. Yet, no less than Ricardo, Marx runs into the problem of the rate of interest. Competition must drive the rate of profit across industries to be equal to this, even though the capital-to-labour ratio is different in different industries. Marx therefore has to assume a transfer of surplus value from some industries to others: labour-intensive industries effectively subsidise the profits of capital-intensive ones.

But, as Mark Blaug pointed out in Economic Theory in Retrospect, once we have that assumption in place we could just as well hold that it is capital that alone creates value, while labour creates none of it:

If, instead, [Marx] had operated with a capital theory of value, attributing the whole of the surplus solely to machinery and implements […] he could have carried on transforming values in to prices in exactly the same way as he did. […] With a capital theory of value we can say that all capitalists share in a pool of surplus value, a pool created solely by the nonhuman factors of production; in the process of equating profit margins per unit of capital invested in both labor and machines, capitalists necessarily cause prices to fall below value in capital-intensive industries and to rise above value in labor-intensive industries. (237)

If we held a capital theory of value rather than a labour theory then the observed facts would be just the same: capitalists apply capital and labour to produce commodities, only one of these factors actually creates surplus value, and those industries employing more of that factor transfer surplus value to those employing less of it. A theory that ascribed the creation of surplus value to both factors would also be empirically equivalent to the other two.

You could reply here that Marx can explain why only labour can create surplus value through sociological facts. It is possible to pay a worker a subsistence wage — what Marx calls the value of her ‘labour-power’ — while she produces commodities worth more than that wage — the value added through the consumption of her labour-power. Surplus value is the difference, pocketed by the capitalist. But upon examination this explanation simply assumes the labour theory of value. If it were capital that supplied the difference between the value of the wage paid to the worker and the value of the commodity produced, then the worker would give no more value to the commodity than the value of her labour power — in other words labour wouldn’t create any surplus value.

This is what leads Robinson to accuse the concept of value of lacking real content. Entirely different theories of value are consistent with the same observed facts. Suppose a capitalist pays a worker £200 to live on while producing commodities that sell for £220, and pockets the £20 difference. A ‘bourgeois economist’ might explain it this way: Production takes time, and the capitalist supplies the time by effectively lending the worker money to live on while she produces the commodity. For this contribution the capitalist is compensated £20 — Alfred Marshall called this ‘the reward of waiting’.

Marx would call this nonsense, or worse, ideology. ‘Waiting’ is not an extra factor in production. It is the worker’s labour that creates the extra £20 of value, and the capitalist simply expropriates it by exploiting the fact that the worker has nothing to live on while exercising her own labour power. In effect, the capitalist is a usurer, charging the worker for a loan she cannot live without. It adds confusion but doesn’t change anything fundamental if capital — the contribution of time — involves the embodiment of past labour in machinery and implements as part of a production cycle.

Well, how do we decide between these claims about value? The bourgeois economist says that the capitalist surrenders £200 to the productive process and thus foregoes the use of it during that period; Marx wouldn’t disagree. The bourgeois economist points out that, things being as they are, production wouldn’t be possible if the capitalist didn’t do this; again, Marx wouldn’t deny it. The worker, says the bourgeois economist, might well have agreed to the deal: subsistence wages of £200 now and you give up a product worth £220 later — Marx would only add that she didn’t have much choice but to agree to this, and the bourgeois economist wouldn’t argue with that. So what is there to argue about?

If you want to say that the capitalist is entitled to profit, you will say that the capitalist contributes something to the creation of value, for example ‘waiting’. But what is the real difference between saying (a) the capitalist contributes to the creation of value in a commodity and (b) the capitalist is entitled to some of the proceeds from the sale of the commodity? Robinson seemed to take a very reductionist, non-cognitivist line on this: statements about value have no factual content at all; they boil down to expressions or injunctions or commands — e.g. ‘labour creates all value’ really comes down to ‘expropriate the capitalists!’ or ‘boo capitalism!’. I wouldn’t go that far. But since, again, the observable facts seem to be identical either way, I do think that the only basis for deciding whether or not to hold (a) seems to be whether or not you want to say (b). In that case it isn’t your theory of value that tells you what attitude to have; it’s your attitude that dictates what theory you should hold.

Alternatively, as many of my Twitter correspondents suggested, we could simply define value as the (socially average) labour-time that goes into the production of a commodity. But then the labour theory of value, which tells us that the value of a commodity is created by the labour that goes into it, becomes either a nonsensical statement of value’s self-creation or an empty triviality, reducing to the claim that the labour that goes into a commodity is the labour that goes into it.