Why a Theory of Value?

The gentleman from Unlearning Economics asked me recently in response to my rebuttal of Steve Keen’s critique of Marx’s theory of value why indeed there is any need for a value theory at all. It seemed to him labor as the measure of value was simply assumed by Marxists, and even if their explanations of the economy were clearly better than others and they can rebut the critiques of Keen, Bose and others, it is still not clear why there should be such a thing as a ‘labor theory of value’ at all. I find I often run into this problem with many intelligent, critical people who are by no means unwilling to take my Marxisant analysis seriously, but who simply do not get what kind of thing a theory of value is, let alone Marx’s; and then indeed it must seem a strange and unnecessary quasi-metaphysical imposition. Now initially I thought this as well, and before I fully immersed myself in Marxist thought I was quite hostile to the notion of the labor theory of value, or even the need for such a theory at all. And indeed neoclassical economists have spent more than a century trying to refute both Marx’s theory and the need for such a theory at all. However, it is not so much just Marx’s arguments in Capital itself that convinced me, as my wider reading giving me a more historically and anthropologically grounded perspective about production and exchange in history, and I would venture to say the concept of a theory of value can only make sense if put explicitly in this wider context.

This is notwithstanding Marx’s emphasis that his particular theory of value applies to capitalist society only, and is made to explain and describe that type of society and no other. I agree that Capital is a descriptive work about capitalism, not a set of transhistorical norms or ideas (like many in the USSR thought), but in defending his theory of value Marx and Marxists have generally failed to adequately defend the need for a theory of value. To me it’s becoming more and more intuitive, but to many it’s clearly not. Let me describe how it makes sense to me, put in somewhat rough and ready terms – this is by no means a finished work, but it is a set of thoughts on the topic on which I hope to build and understand more in the future, and on which input is certainly welcome and necessary. If I were to write a guide to (capitalist) political economy, I would certainly start with justifying the need for a value theory and the historical context of capitalism altogether, so this should be seen as a first draft effort towards that end.

Briefly summarized, I think the value theory is necessary (and works in Marx’s version) because:
1) All capital reduces to labor, at least as an input. Marx notes in the “Critique of the Gotha Programme” that nature as well as labor is the ‘source of all wealth’, which is true; but I don’t think the distinction matters much as in practice any natural resources still require labour for their application. Even picking berries or whatever is still a necessary exercise, and all animals labor for their food.

2) The concept of value itself is relevant because we need an anthropological/economic/social explanation of the generalized commensurability of goods under capitalism. This is a very strange thing historically and anthropologically speaking, and by no means is it obvious that you can have *all* different kinds of goods as well as *all* different kinds of labors, with all their unique attributes, be commensurable through a process of competition, mediated by money. A society like that can only reproduce itself if particular things hold, such as the generalized production for exchange, the generalized competition between workers and between capitals, as well as the prehistory that generates both, and finally a yardstick that, as it were, ‘underwrites’ money as the form of appearance of this general commensurability. That yardstick is socially necessary labor time, which Marx calls value (and in Vol. 1, exchange value, as he equalizes them for the purposes of explanation there).

3) I would venture from this that ‘value’ in this sense is then value only under capitalism, and the value of the capitalist system of social relations, in which relations between people appear as relations between commodities – namely relations of commensurability according to this one yardstick. Few Marxists have bothered with this at all, but to me, for the concept to make sense other societies and social relations must also have systems of value as a yardstick for their reproduction one way or another – after all labor and the need for physical and social reproduction are transhistorical universals. So every society needs to reproduce itself and therefore have at least one ‘value’, that is, conceptual notions underwriting their social relations and the pursuit of which (though not necessarily as accumulation) is the determining logic of the reproduction of the existing social relations of production and exchange.

4) However, economic anthropology tells us that there are many societies, perhaps historically most of the total, in which not all goods and services are commensurable. Most social positions and roles are also not, though probably all of them have social mobility of some sort. So there may be different values for different kinds of social relations, *including* the different social relations of production and exchange a society can have. Various goods made may be made wholly separately from each other, not just by different people but also conceptually separate, and never be translatable into each other’s terms. (There are various examples from the literature I could give here if the reader is interested.) While all societies need to produce, and have some form of exchange, to survive, it is quite remarkable of capitalism that the relations of production and exchange take on not just a universally mediated form (money), but also that they are totally subsumed under this one drive for accumulating exchange value. Generalized commensurability therefore means there can be only one value in the sense described above, even if of course people have as many pursuits, values and ideological priorities as ever. One might be tempted to quote Marx’s early manuscripts here, where Marx talks about the social preconditions for developing the needs of the senses, which are the basis of all human pursuits of particular values:

Similarly, senses and enjoyment of other men have become my own appropriation. Apart from these direct organs, social organs are therefore created in the form of society; for example, activity in direct association with others, etc., has become an organ of my life expressions and a mode of appropriation of human life.

To sum up: it is only when man’s object becomes a human object or objective that man does not lose himself in that object. This is only possible when it becomes a social object for him and when he himself becomes a social being for himself, just as society becomes a being for him in this object.

For a man who is starving the human form of food does not exist, only its abstract form exists… The man who is burdened with worries and needs has no sense for the finest of plays; the dealer in minerals sees only the commercial value, and not the beauty and peculiar nature of the minerals; he lacks a mineralogical sense.


5) So why socially necessary labor time? The labor element is easiest. The reduction to a common commensurable standard happens in competition. But the very fact of all societies needing to reproduce themselves plus the fact that exchange is a zero-sum game on aggregate, plus the fact that there can be only one value because of generalized commensurability, plus the fact that it cannot be capital’s own activity as all capital reduces to labor, requires that labor in the process of production itself, the immediate sine qua non of society’s continuation, in some way furnish this yardstick. This was the point also taken by Smith, Ricardo, etc.

6) This is measured in time, because capitalism as the only mode of production that can only reproduces itself on an ever expanding scale requires an ever stronger compression of time and space in order to do so. This strikes me as a bit of a chicken-and-egg question to whether this drive comes first or the measurement in time comes first, but certainly all capitalist practice expresses itself in temporal terms – the labor contract covers a certain amount of labor-time, there is the constant drive to extending the working day, it is under capitalism that standardization of time, the use of temporal discipline in labor, and so forth became essentials for the entire productive process, and so on. As Marx puts it:

Secondly, with regard to that which forms the ground-work for the quantitative determination of value, namely, the duration of that expenditure, or the quantity of labour, it is quite clear that there is a palpable difference between its quantity and quality. In all states of society, the labour time that it costs to produce the means of subsistence, must necessarily be an object of interest to mankind, though not of equal interest in different stages of development. And lastly, from the moment that men in any way work for one another, their labour assumes a social form.


But the technological element is at least as important, as it is to some extent an exogenous input as well as an output of capitalism. I’d say capitalism, like all modes of production, can be seen as a particular application of technology in a particular social context, a transhistorical element comes into this as well. All societies can be seen as having a certain determinate amount of ‘socially productive power’, i.e. sheer numbers + ability per person to affect nature, to undertake production; and in practical experience such general capacity for production is expressed in time, that is in the time that we spend on things. The time of our lives is limited and linear, and any time spent is not only always mutually exclusive with all other time, but also cannot be undone. (Of course in physics the simple linear time is not actually right, but this is how we experience it at the social level). So time spent achieving a particular production is the first measure, and all else derived from this – of course many of us spend lots of time doing other things, and biologically too there is a limit to the working day, but all such time spent is spent in awareness of blissful non-work. This is where the fight over the working time becomes so central, and in fact has been in every society, best expressed in capitalism by the idiom that “time is money”. Surplus goods consumed by a nonproducing class are reducible to the time not spent by this class in producing those same goods under the prevailing conditions of that society. Technology is the factor introduced by humans to lessen the time spent working for a particular result in production, and for every other purpose of shortening time spent as well (which through this reduction to labor time comes down to the same thing).

7) Under capitalism then, the generalized production for exchange and the generalized competition combine with the temporal dimension of production to generate a standard not just of time, and of labor time, but of socially necessary labor time, as competition forces all capitals to work at the level that technology has placed the socially productive power of a given capitalist society at. So too all labor engaged in this production must be at the skill level, productivity etc. corresponding to the socially productive power of that society as a whole. If these levels fail to be obtained, the value of the goods will be lower than average, and if they are exceeded, they will be higher. The competitive pursuit of accumulation for accumulation’s sake will ensure that all measures will be taken to make a production process correspond to the highest norm of socially productive power capitalist society can achieve.

8) This contrasts with the neoclassical notion of utility because the utility concept is precisely an attempt to substitute for a value theory. It fits neoclassical economics’ deliberate and conscious ahistoricism, as this school of economic thought is interested only in capitalist exchange, not production, and only insofar as it can be presented as universal, not historical and particular. For this reason, a value theory or a set of different value theories for different societies cannot be entertained, for this would not be sufficiently general for the purposes of neoclassical economics – it needs a concept that is flat enough to allow itself to represent value in exchange, but to have no particular historical, social, or any other properties that cannot be mathematically represented and therefore serve tractability of this kind. Utility, which in contemporary economics no longer has its Benthamite normative meaning but simply reduces to the set of preferences along an indifference curve an individual faces in a given exchange situation, now plays this role. This is not, I would say, illegitimate per se; but it serves a much more general and therefore vastly more limited function than value theory does. Value theory is an approach more interactive with and akin to the historical social sciences, whereas neoclassical economics is for the most part applied mathematics. Neoclassical economics therefore does not so much do without a value theory as it massively reduces its own scope, application, and ambition in order to avoid confronting it. This is permissible scientifically, but it means that the two approaches are talking past each other.

9) Therefore when George Stigler writes in response to Adam Smith’s diamond-water paradox that

The paradox—that value in exchange may exceed or fall short of value in use—was, strictly speaking, a meaningless statement, for Smith had no basis (i.e., no concept of marginal utility of income or marginal price of utility) on which he could compare such heterogeneous quantities. On any reasonable interpretation, moreover, Smith’s statement that value in use could be less than value in exchange was clearly a moral judgment, not shared by the possessors of diamonds. To avoid the incomparability of money and utility, one may interpret Smith to mean that the ratio of values of two commodities is not equal to the ratio of their total utilities. Or, alternatively, that the ratio of the prices of two commodities is not equal to the ratio of their total utilities; but this also requires an illegitimate selection of units: The price of what quantity of diamonds is to be compared with the price of one gallon of water?

(3) he is clearly missing the point. Use value is the subjective valuation factor which in all societies corresponds to the needs, ambitions, desires, ideologies etc. of the individual people involved in that society, and exchange value is value in the sense I have used it above as appears in capitalist exchange as the result of the capitalist production process. Smith’s separation of use value and exchange value is not a category mistake. On the contrary, it is a category mistake only without a theory of value applicable to capitalism; that is, it would be a category mistake if capitalist society like so many societies of the past did not have generalized commensurability taking the form of money. But it is precisely because it does have this that it is possible at all to express the exchange value of water in terms of diamonds and vice versa! This phenomenon, and its consequences for the kind of society capitalism is compared to other societies and how it reproduces itself as this kind of society is what Marx’s value theory sets out to explain: “To the producers, therefore, the social relations between their private labors appear as what they are, i.e. they do not appear as direct social relations between persons in their work, but rather as material relations between persons and social relation between things”(4). In principle, it should then also be possible to do something similar for other types of societies – a task so far only undertaken, and often without this conscious aim, by economic anthropology.

1) http://www.marxists.org/archive/marx/works/1844/epm/3rd.htm
2) http://www.marxists.org/archive/marx/works/1867-c1/ch01.htm#S4
3) http://www.ppge.ufrgs.br/GIACOMO/arquivos/eco02277/stigler-1950.pdf
4) http://www.marxists.org/archive/marx/works/1867-c1/ch01.htm#S4


plus the fact that exchange is a zero-sum game on aggregate

While aggregation renders this far more readily apparent, those teaching the marginalist perspective strive to inculcate the idea that exchange is never zero-sum. I’d just like to share a thought for those who may be reading with this in mind:

Back when Marx was beginning to “click” for me, I recall the moment I realized — as per his Aristotelian reference to exchange implying equality and equality implying commensurability — that this marginalist lesson basically hinges on the exclusion of money. That is, exchanging a hat for a chair suits the “double-inequality” explanation a lot better than trading either for a given amount of an objective quantifier — especially an amount determined in the broader context of a market. In such a case, some amounts of money for a chair present themselves as less equal than others; and if the exchange is unequal, presumably one could add or subtract dollars until you reach equality. All prices cannot have identical social validity, else we couldn’t have a price system in the first place. And nothing about both parties gaining utility from a transaction precludes the exchange being equal in social terms. In fact, I would suggest that the latter may even be a prerequisite of the former; feeling as though you’ve been ripped off would probably lessen whatever you stood to gain, in utility terms.

Once you allow for one good to be quantified in such a manner, you also develop a basis for comparison with other heterogeneous goods — and from there you’re close to the basics of Marx’s theory of money (though I won’t get into that). To someone who contends that at no point is equality in exchange achieved, it’s probably fair to suggest they “check their units.”

“In principle, it should then also be possible to do something similar for other types of societies – a task so far only undertaken, and often without this conscious aim, by economic anthropology.”

David Graeber is no doubt best known for his recent book on Debt, but before that he had written a bit in this direction: Toward an Anthropological Theory of Value. It’s been a while since I read it, so it’s hard for me to say to what extent he may have succeeded in terms of the above article, but more than a few concepts from it (value not just as a social relation, but one capable of reordering other social relations; dialectics of action vs being; numerous fascinating details of Iriquois and Maori society; etc.) have stuck with me. I can at least recommend it on those grounds.

I have indeed read that book by Graeber, and second the recommendation. I liked it quite a bit better than “Debt”; I might review it at some point.

Thanks for this article, I struggled with this problem when reading the first chapter of Capital for the first time. Why did I have to accept labour as the only way to compare commodities when utility theory offered another alternative?

Also, do you think the concept value is modified under different forms of capitalism? I would think socially necessary labor time fails to describe a capitalism that is monopolistic and uncompetitive

I really appreciate the effort you put into this! I’ve read, but I still don’t get it.

When deliberating on where to start a new factory, in France, Turkey or Malaysia, capital doesn’t “make a production process correspond to the highest norm of socially productive power capitalist society can achieve”. It doesn’t act on value, but on prices. Why not just discuss those prices?

Even when it comes to describing “the kind of society capitalism is compared to other societies”, why can’t we just do this with prices?

It’s kind of frustrating to know that the guy who invented this whole historical materialism thing that I keep coming back to time after time, basically said that the most important thing to know about capital was how to juggle a theory a value that never, ever seem to get to point where it actually makes a statement that can be tested against empirical evidence.

Well, you’d think the fact he found it so important is an indication it might be 😉

But more seriously, nobody denies the immediate significance of prices, Marx no more than anyone. Marx’s theory of the value in capitalism is completely integrated with the theory of price, even if it is not a *predictive* theory of price (which is a pretty useless concept anyway). Values of the inputs into production are through the process of production transformed into prices of production, which are in turn transformed into market price. But what Marx is interested in is not whether the price of a TV is $200 or $900, or whether potatoes are cheaper than cabbage or not; if you want to know that, you can ask the shopkeeper or the wholesale dealer, you don’t need Marx to tell you that.

What is interesting to Marx, and to the classical economists generally, is what kind of *role* prices play in the process of production and exchange (distribution) of capitalism altogether, at the level of capitalist society as a whole – something an analysis of particular, existing price levels does not immediately reveal on its own. So the question is not: ‘do prices matter?’. Of course they do. The question is: ‘how do prices matter? And why in that way rather than another? And what is the role of the money form in which the prices are expressed?’.

This same critique can be leveled against neoclassical economics. After all, no capitalist actually has knowledge of the marginal productivity of an additional worker, or the marginal revenue of an additional widget. Any theory with any explanatory depth is going to posit things going on behind the backs of the agents whose behavior it seeks to explain.

Of course, you can live your life in capitalism and be concerned only about prices. You can live and die.

The labour LTV becomes only important and crucial, when you look at capitalism from the class standpoint of the worker, trying to understand the reason for the economic crisis that we are facing again.
Trying to understand why after a long period of economic growth after WWII, you face a systemic crisis again with all kinds of backward ideologies of racial, religious hate being pushed to the surface again.
Trying to understand why you are losing your job or why your salary is cut by a substantial percentage that threatens your living condition despite the fact that two or more generations after the WWII catastrophe have been working hard to improve the conditions of life.

Of course, a capitalist will have no problems looking at his prices, his profit margin only. But looking at the statistical data of price development gives you only a superficial and confusing impression of what’s going on. If you don’t understand the hidden currents below the ocean’s surface, if you don’t understand the tectonic plate dynamics, you don’t understand nothing at all.

The LTV is the key to understand the social “tectonic” forces behind the surface of capitalist commodity production. It explains why the social tensions are bound to increase and lead to disruption of the “peaceful” period of economic growth. “Value” is a category of social relations under disguise – materialized as the fetish of the commodity.

Sure, you can only look at “prices”, you can live your life and build a conventional brick house on the South East Asian “ring of fire”, but you might incur a shock when the earth quake crushes your home or when the economic crisis destroys your conditions of life.

Marx was not concerned about writing a “theory” for intellectual pastime. That’s why he did not only write the “Capital” but also the “Communist Manifesto”.

Thank you for this, as some kind of post-Marxist, I’ve followed your blog for some time now and am very pleased.

I’ve decided to use your article as a spring-board in one of my recent posts on German philosophy… completely unrelated to economics. Not sure whether or not you would enjoy it, but it’s worth notifying you.




Have you read Postone’s Time, labor, and social domination? It’s a challenging and very powerful reading of Marx that could really enrich your discussion here.

I am little confused; so for example, in Amazon, there are two groups of people working same amount of time but one group is paid more than the other group. Then how do Marxists calculate labour-power or labour-contents?

In reality, theories of value have existed for a long time, at least since exchange itself, because they are attempts at answering a very natural question: what determines in which proportions things are exchanged?

What I am about to say may sound strange, but I find this definition useful:

“The purpose of economic value theory — whether a cost, labor, or subjective approach — is to explain the prices of various goods and services in a market economy, i.e., to explain their “market value.” For example, why is it that gold bars and automobiles are so valuable, while tinfoil and tube socks are not?
“Explaining the formation and magnitudes of various market prices is not the sole task of economic theory, but it is a crucial component of it.”

Neoclassicals may make a big fuss about “value”, but they, too, have a value theory, even if they don’t know it: supply and demand and all the maximization apparatus (both in their theories of households, firms, and partial market equilibrium).

Among Post Keynesians, Joan Robinson, in her Economic Philosophy, reveals a misunderstanding of epic proportions about “value”; ironically, Keynes mentions his own value theory in his General Theory.

Precisely these different approaches (subjective vs labour) are indicative of the class interests represented in the theory: a capitalist sees economic life as an opportunity to consume, a merchant naturally sees the subjective satisfaction of his wares as basic in determining their value; a working man, on the other hand, as naturally sees the time required to produce that stuff as determining its value.


Incidentally, I say the quote above is curious, because the author of that quote is an Austrian economist:


Leave a Comment

Your email address will not be published. Required fields are marked *