In my last post I presented an outline of Joan Robinson’s critique of Marx on value, as I understand it (this is in Economic Philosophy rather than her book on Marx).

My point in all of this is that it’s very important to clarify this question of value if we want to understand Marxism rather than using it as a branding slogan.

Those who claim that value is just the price at which one commodity exchanges for others, or even the underlying determinant of this, miss the whole point of the transformation of value into price in Volume Three of Capital, which explains how commodities don’t exchange according to their values under normal conditions of capitalism.

Those who claim that ‘value’ simply means ‘socially necessary labour time’ reduce Marx’s labour theory of value to triviality: the claim that the socially necessary labour time embodied in a commodity is equal to itself. This isn’t far off the ‘absurd tautology’ Marx condemns at the start of ch.19 of Volume One of Capital.

Robinson sees value as a concept with no factual content, one that merely expresses a moral attitude. To bring out the force of this, I looked at Mark Blaug’s argument that, while Marx treats labour power as the only commodity capable of creating value, he could just as well have ascribed this power exclusively to the non-human factors of production, consistently with all the same observations. The point about fixed capital confuses the issue, so here I want to abstract it away.

Marx wants, on Robinson’s reading, to get to the conclusion that capitalism has expropriation built into it: labour alone creates value, yet the value created goes to the capitalist who has done none of the work. Her argument is that we can’t verify the claim that value is created exclusively by labour, since value isn’t an observable thing. I want to look harder at how the concept of surplus value works in Marx’s descriptive theory.

Ricardo had taken on the labour theory of value from Smith, as a way of accounting for relative prices. If A takes twice the labour to make as B, then one A will exchange for two Bs.

Then we have the issue of the interest rate, r. If both commodities take a year to produce, A employing one worker and B employing two workers, and the yearly cost of a worker is w, then the market price of A will be near to w(1+r) and that of B near to 2w(1+r). Twice as much labour gets you twice as much value. On the other hand, if another commodity, C, takes one worker two years to produce, then its price will be near, not to 2w(1+r) but to 2w(1+r)². Twice as much labour as A, over twice as long, gets you more than twice as much value. Ricardo dealt with this apparent violation of his labour theory by effectively defining r out of existence, at least at this stage of his analysis.

For Marx, r is determined by surplus value, a concept that doesn’t exist in ‘bourgeois’ economics. At the heart of capitalism, for Marx, is the famous M-C-M’ circuit — a capitalist uses money (M) to buy the commodity labour-power (C), and then selling the product for more money (M’). Since money is the external form of value for Marx an an idealised case will have M correspond to what is spent paying workers over a period (variable capital, v) and maintaining equipment (depreciation on fixed capital, c), while M’ corresponds to the same value plus the surplus, s. The rate of profit, r, is just s/(c+v) for a given period.

Marx is clear that what makes this possible is, among other things, the existence of a proletariat. This is a class that is ‘free’ in Marx’s two senses: the proletariat are free from indentured servitude and thus can sell their labour power on the market, and (here Marx’s sarcasm) they are free of any ownership of property, so that they must sell their labour power to survive.

This — the dependence of capitalism on a permanently expropriated and propertyless class — is something huge and fundamental that appears in Marx’s analysis and not in the analysis of either classical or neoclassical ‘bourgeois’ economics. Value — surplus value — plays its role in supporting this fundamental Marxist claim. And it isn’t a mere value judgment; it’s a factual claim. But to understand it we need to understand value.

Again, to keep things simple I want to take fixed capital out of the picture. Suppose that production occurs through variable capital alone. Now for Marx it is only possible to realise a surplus, to run the circuit M-C-M’, when there is a proletariat. Suppose, however, that we give the proletariat some capital, so they are no longer a proletariat. Then imagine that, from this stock of capital, they spend M sustaining themselves during a productive cycle, producing a commodity that they sell for M’. This would violate the principle that the very possibility of realising surplus value depends on a propertyless proletariat. What reasons does Marx have for rejecting this possibility?

In the first place, a ‘repropriated’ proletariat would presumably spend more on themselves than the capitalists were willing to spend on them. If the value of the final product is v+s, perhaps Marx’s view is that the value of labour of a repropriated proletariat will rise until the new v is equal to the old v+s, and s will therefore disappear. The rate of profit, s/v in this case (since we’ve taken out fixed capital), would then also be zero. Marx also calls the ratio s/v (even when there is fixed capital) the ‘rate of exploitation’. This too will be zero, which bears out the idea that there is only profit where there is exploitation, and a propertyless class to exploit.

‘Bourgeois’ economics would eventually arrive at an explanation of r in terms of ‘time-discounting’: we all discount the future so that it’s only worth spending a given amount today to get a greater amount tomorrow. This suggests that even a repropriated proletariat would only ration their means of subsistence over a productive period, rather than consuming it all right now, for the sake of some future gain. Blaug claims that ‘Nothing can reduce the rate of interest to zero except capital saturation: a state of affairs in which real incomes are so high that postponement of present consumption is painless’ (Economic Theory in Retrospect, ch.7). Until that point, r will not fall to zero and neither will s.

Robinson notes, in her book on Marx, that what she calls ‘the orthodox theory of profit’, based on time-discounting, is ‘far-fetched’ (An Essay on Marxian Economics, ch.7). If it were true it would be possible to eliminate unemployment all the time, simply by dropping the interest rate. Central banking literature would be scientifically accurate, and if that’s not a reductio then I don’t know what is.

Marx’s own theory of profit rests on the inexorable self-reproduction of the capitalist cycle, M-C-M’ under capitalism. The profit-seeking of the capitalist isn’t explained by time-discounting or anything else; it is simply what he calls ‘the unceasing movement of profit-making’. In a way, the capitalists are just the instruments by which the profit-cycle perpetuates itself, rather than the profit-cycle being the instrument of the capitalists.

This is Marx at his most Hegelian, and it appears to be a radical thesis in the context of classical economics. But it is quite a traditional view in broader philosophical terms; it’s not far off what Aristotle says in the Politics (Book 1, parts 8–9), and has distinct resonances with an important part of Spinoza, as I read him: money has a sort of enchanting power — prestige with the original connotations of praestigia — and humans can become its instruments.

Without reverting to the time-discounting theory, however, can we not imagine that even if we expropriated the bewitched capitalist classes and repropriated the proletariat, the latter might still pay themselves a wage somewhat below the value that they realised in the sale of its final product? This would be worth doing if the difference could be productively invested.

Think of Ricardo’s ‘corn farm’ model, except that seed corn can be eaten or planted, while mature corn is sold as a cash crop (maybe people like to decorate their houses with it). From a purchase of 100 bushels of seed, workers on a collective farm might eat 90 and plant 10, to grow just enough to afford another 100 bushels. Or they might prefer to eat 85 and plant 15, for the sake of growing more. They might do this for the sake of getting to the point where they can afford to eat, say, 100 bushels per period and plant enough to afford the same next time. Until then, again, r and s will both be greater than zero.

Somebody might reply that these transactions wouldn’t involve value because they wouldn’t involve exchange under the capitalist commodity-form. But honestly I don’t know what that means, and I don’t see what it changes. There is obviously some clear parallel between the case of the workers planting enough to get more bushels out of 100 bushels and the capitalist’s M-C-M’ cycle, whatever word-games we want to play about ‘value’.

And we don’t need the whole standard theory of time-discounting to imagine the above scenario. We only need Blaug’s idea, that people will pursue net investment and expanded production until they are consuming, in absolute terms, as much as they want to be consuming. This defines a point of capital saturation.

For Marx the crucial point is perhaps that in this scenario the point of saturation will come, whereas under capitalism it won’t. Capitalism, through the whole structure of social relations it builds, creates a situation in which a certain class of people pursue profit towards no higher purpose and therefore indefinitely. This isn’t possible unless there is a proletariat: a class that must sell its labour in order to survive at the start of each productive cycle and therefore cannot accumulate at all — otherwise we would approach capital saturation.

This means that surplus value can’t be defined simply as the difference between proceeds of sale and cost of production, nor in terms of an r greater than zero. It must, rather, refer to a permanent striving to hold up the level of r (this is why a falling rate of profit ‘contradicts’ capitalism, though only gradually). Value refers to a social situation that ‘aims’ to reproduce itself indefinitely.

There is still a lot that I don’t understand about this picture, and I might have it completely wrong. But I think that looking carefully at the concept of value reveals the number of intellectual commitments you need to make in order to really embrace Marx’s analysis of capitalism. Having generally left-wing sentiments, or wanting to impress others who do, is nowhere near enough, so long as you want to use the theory as a theory rather than a brand.

Lecturer in Philosophy, University of St. Andrews — personal website: