Archive for April 2nd, 2018

Marx’s law of value: a debate between David Harvey and Michael Roberts

April 2, 2018

This is going to be a long post, so bear with me.  First, you will have to read a paper (attached below) by Professor David Harvey, then a critique by me below – and finally a reply to my critique by Professor Harvey.  And then it’s up to you readers to see what you make of it: is this like a medieval religious debate about how many angels there are on the head of a pin; or is it a debate that leads to something really worth knowing?

For more on the nature of Marx’s law of value and its relation to crises, see my new book, Marx 200 

David Harvey’s misunderstanding of Marx’s law of value (Michael Roberts)

Recently, Professor David Harvey (DH) sent out an email to several people, including me, attaching a short paper for discussion (see Harvey paper).  The paper outlines DH’s view that Marx’s theory of value in capitalist economies has been badly misunderstood.

Just in case you are unaware (difficult to believe), Professor Harvey is probably the most eminent Marxist scholar alive today with a host of books, papers and educational videos to his name on Marxist economic theory.  The short paper circulated expressed succinctly his view of Marx’s value theory that he has recently outlined more expansively in his latest book, Marx, Capital and the madness of economic reason.[i]

In the paper, entitled Marx’s refusal of the labour theory of value, DH argues that Marx did not have a ‘labour theory of value’ at all.  His theory of value was distinctive from that of the classical economist, David Ricardo.  Instead, according to DH, Marx argued that value was a reflection of labour embodied in a commodity which is only created/revealed in exchange in the market.  As DH puts it: “if there is no market, there is no value”.  If this correct, then it is in the realization of value as expressed in money that value emerges, not in the production process as such.

DH then goes on to argue that if wages are forced down to the minimum or even to nothing, then there will be no market for commodities and thus no value – and this is the “real root of capitalist crises”.  And thus it follows that a policy for capital to avoid crises would be by “raising wages to ensure “rational consumption” from the standpoint of capital and colonizing everyday life as a field for consumerism”.  This is the consequence of a correct view of Marx’s value theory, according to DH.

DH points out that this interpretation of value theory “is far beyond what Ricardo had in mind and equally far away from that conception of value usually attributed to Marx.”  It certainly is.  But is DH right in his interpretation of Marx’s value theory and, even if he is, does such interpretation have any empirical validity?  I would answer both these questions with: ‘no’, ‘non’, ‘nein’, to use Marx’s three best known languages.

DH starts by saying that “It is widely believed that Marx adapted the labour theory of value from Ricardo as a founding concept for his studies of capital accumulation” and “since the labour theory of value has been generally discredited, it is then often authoritatively stated that Marx’s theories are worthless.”  It is not clear who DH is referring to here.  Clearly bourgeois mainstream economists consider Marx’s law of value as discredited.  The neoclassical marginalists have long rejected the concept of labour-value by labelling it ‘metaphysical’.  Neo-Ricardian, post-Sraffian, and post-Keynesian economists, in particular, are also strongly inclined to dismiss any notion of ‘value’ as an ideological mystification.

But most Marxist economists are aware of the distinction between Marx’s value theory and Ricardo’s.  And the difference is not what DH says it is, namely, Ricardo had a ‘labour theory of value’ and Marx did not.  The difference is that Ricardo had a theory of (use-) value based on ‘concrete labour’ (physical amounts of labour) measured in labour time.  Marx’s law of value was based on ‘abstract labour’ (value measured in labour time when ‘socially’ tested on the market).

Under capitalism, human labour power itself is a commodity to be sold on the market.  Indeed, this is a key characteristic of the capitalist mode of production where the majority has no means of production and so must sell their labour power to the owners of the means of production.  So, just as with other commodities, labour has a dual property. On the one hand, it is useful labour, that is, expenditure of human labour in a concrete form and for a specific purpose and with this property creates use values. On the other hand, it is abstract labour, that is, expenditure of human ‘labour power’ without specific characteristics which creates the value of the commodity in which it is represented.  Thus Marx made the distinction between labour and labour power, a distinction that is absolutely crucial for the understanding of the source of profit.

This was the great advance in Marx’s law of value. The labour time embodied in the commodities normally purchased by the worker for the reproduction of himself and his family in a day is less than the labour time that a worker actually offers to the owner of capital during the same time period. The result is that for any given time period, the worker produces more value than the wage equivalent which is paid by the owner of capital for the use of the labour power. This difference, Marx calls “unpaid labour” and “surplus labour”- or surplus value.  Marx’s value theory of abstract labour exposes the exploitative nature of the capitalist mode of production, while neither Ricardo’s nor Adam Smith’s labour theory of value does.

DH mentions just once (and in passing) this vital discovery of Marx (i.e. abstract labour) that distinguishes Marx’s law from the classical labour theory of value.  And that is because DH wishes to press on to his interpretation of Marx’s theory as one where value is created/realized only in exchange, and not in the process of production using labour power.  DH says that “value is initially taken to be a reflection of the social (abstract) labour congealed in commodities.”  But “as a regulatory norm in the market place, value can exist, Marx shows, only when and where commodity exchange has become “a normal social act.” So, without money, there is no value.

Yes, but the value of a commodity is still the labour contained in it and expanded during the production process before it gets to market.  Value is expended physical and mental human labour which is then abstracted by the social process of production for the market.  Value is not a creature of money – on the contrary.  Money is the representation or exchange value of labour expended, not vice versa.  I think Marx is clear on this crucial point.  He says in Capital Volume One: ‘The value of a commodity is expressed in its price before it enters into circulation, and it is therefore a pre-condition of circulation, not its result.”[ii]

Murray Smith in his new and forthcoming edition of his book, Invisible Leviathan[iii], provides a concise explanation of the difference between Marx’s law of value and DH’s interpretation.  Marx said that: “Money as the measure of value is the necessary form of appearance of the measure of value which is immanent in commodities, namely labour-time.” Smith comments that this “is certainly inconsistent with the idea that value can be created in the act of exchange. ..It is precisely because exchange effects a process of ‘equalisation of products of lab our on the market’ (that is, involves a real abstraction) that production oriented toward exchange must take account of the fact that ‘physiological labour’ is both utility-shaping and value-creating – that is, both concrete and abstract at one and the same time. To try and argue that that value is created ‘not in production but at the articulation of production and circulation’ is a notion replete with circular reasoning and requiring the most robust of mental gymnastics to entertain…. The problem with this approach is that if one accepts that abstract associated labour has no substantial existence apart from the value form, money, then commodity values appear to be severed entirely from any determination in the conditions of their production, and the way is paved for an effective identification of value and price.”

Instead, Marx’s law of value is based on the view that the labour involved in the production of commodities produces value, while exchange realises it in money-form. It is only because of this that Marx can distinguish between the amounts of value and surplus-value created in commodity production, and the generally different amounts realised through exchange.

Contrary to the view of the mainstream and neo-Ricardian economists, there is no ‘mystification’ here.  Value is objective and real and not just expressed in money.  Marx’s law of value, where abstract labour (measured in labour time) explains exchange value and prices, can be empirically validated.[iv]

There is reason behind DH’s interpretation.  If value is created only at the moment of exchange for money and ‘money rules’, then it will be (effective) demand that will decide whether capitalism smoothly accumulates without recurring crises.  To show this, DH describes in some detail the impact of capitalist accumulation on the conditions and living standards as capitalists strive to raise relative surplus value through the introduction of machinery.  He uses some of the graphic examples provided Marx in Chapter 25 of Volume One.  DH emphasises that capitalist accumulation aims to minimize the value of labour power – even to the point of pauperism.

DH concludes that “If this is a typical outcome of the operation of the capitalist law of value accumulation, then there is a deep contradiction between deteriorating conditions of social reproduction and capital’s need to perpetually expand the market.  As Marx notes in Volume 2 of Capital, the real root of capitalist crises lies in the suppression of wages and the reduction of the mass of the population to the status of penniless paupers.”  So the ‘real root of crises’ is found in the “suppression of wages” and the “reduction of the mass of population to the status of penniless paupers”.  This is an underconsumptionist theory of crises.

There are several points here.  First, Chapter 25 entitled, The general law of capitalist accumulation, does not just refer to the pauperization of the working class.  DH leaves out a very important aspect of that general law: the tendency for the organic composition of capital to rise[v].  This is what drives up relative surplus value but is also a key factor in the tendency of the rate of profit to fall (developed in Volume 3), ‘the most important law of political economy’[vi], which lays the basis for Marx’s theory of crises.  DH ignores this aspect.

But DH goes further in his underconsumptionist interpretation.  “Value depends on the existence of wants, needs and desires, backed by ability to pay in a population of consumers……It also means that the diminution of wages to almost nothing will be counterproductive to the realization of value and surplus value in the market. Raising wages to ensure “rational consumption” from the standpoint of capital and colonizing everyday life as a field for consumerism are crucial for the value theory.”  Thus DH argues that capitalism goes into crises because wages are suppressed; and so raising wages, ensuring ‘rational consumption’, would provide the ‘ability to pay’ and so end the crisis.

This underconsumptionist interpretation of Marx’s crisis theory has been firmly dismissed – by Marx himself – in the famous note in the same Volume 2 that DH refers to (underlines are my emphasis).

It is sheer tautology to say that crises are caused by the scarcity of effective consumption….That commodities are unsaleable means only that no effective purchasers have been found for them.   But if one were to attempt to give this tautology the semblance of a profounder justification by saying that the working-class receives too small a portion of its own product and the evil would be remedied as soon as it receives a larger share of it and its wages increase in consequence, one could only remark that crises are always prepared by precisely a period in which wages rise generally and the working-class actually gets a larger share of that part of the annual product which is intended for consumption. From the point of view of these advocates of sound and ‘simple‘ (!) common sense, such a period should rather remove the crisis.”[vii]

In my view, Marx rejected both the law of value as DH interprets it and also the conclusion that crises caused by an inability to pay for the ‘wants, need and desires’ of people.  But Marx could be wrong and DH right on the cause of crises.  Empirical evidence does not support DH, however.

Let me cite just three facts.  The first is that workers’ consumption is not the largest sector of ‘demand’ in a capitalist economy; it is productive capital consumption.  Gross domestic product or expenditure is a measure of annual demand for ‘wants, needs and desires’.  In the US, consumption would seem to constitute 70% of GDP.  However, if you look at ‘gross product’ which includes all the intermediate value-added products not counted in GDP, then consumption is only 36% of the total product; the rest constitutes demand from capital for parts, materials, intermediate goods and services.  It is investment by capitalists that is the swing factor and driver of demand, not consumption by workers.

This is shown in the second fact.  If we analyse the changes in investment and consumption prior to each recession or slump in the post-war US economy, we find that consumption demand has played little or no leading role in provoking a slump.  In the six recessions since 1953, personal consumption fell less than GDP or investment on every occasion and does not fall at all in 1980-2.  Investment fell by 8-30% on every occasion.

Percentage change in US real personal consumption (PC), investment and GDP

The third fact relates directly to wages and DH’s claim that raising them would help capital.  Carchedi finds that of the 12 post-WWII crises, 11 have been preceded by rising wages and only one by falling wages (the 1991 crisis)[viii].  That confirms Marx’s view in the note in Volume 2  above.

I conclude from DH’s short paper that he aims to establish an argument that class struggle is no longer centred or decided between labour and capital at the point of production of surplus value. Instead in ‘modern’ capitalism, it is to be found in other places in his ‘circuit of capital’ that he presents in latest book and in various presentations globally.  For DH, it is in the point of realisation (ie over rents, mortgages, price gouging by pharma firms etc) or in distribution (over taxes, public services etc) that the ‘hotspots’’ of class struggle are now centred.  The class struggle in production is now less important (even non-existent).

In my view, to support this, DH presents a series of theoretical confusions in this paper.  First, Marx did not have a labour theory of value.  Second, value is only created in exchange (in realisation). Third, the rate of profit (or even profit alone) is irrelevant to crises: what matters is the driving down of the value of labour power to the minimum (or even zero!) so that workers are unable to meet their ‘wants, desires, etc’.  This becomes a crude underconsumption theory – cruder than Keynes.

DH deliberately ignores the difference (and duality) between concrete and abstract labour, and its counterpart, use value and exchange value.  The dual nature of value in a commodity, as Marx discovered, is reduced by Harvey to the lack of the ability of workers to buy their use values.  Use value (wants and desires) is the key, not exchange value in value, for DH.  Marx’s theory of crisis (based on insufficient surplus value) is replaced with insufficient use values for workers as consumers.   Overaccumulation is replaced by underconsumption.  The class struggle becomes not workers versus capitalists; but consumers versus capitalists or taxpayers versus governments.

It’s not Marx’s view.  More important, the whole approach is confusing to a class analysis and strategy for the working class struggle.

And now here is David Harvey’s response to my critique of his paper on Marx’s value theory.

The misunderstandings of Michael Roberts (David Harvey)

There are, obviously, some serious points for discussion over Marx’s value theory and I hope that some dialogue with Michael Roberts can help clarify them.  Before getting to them, I need to remove a number of misreadings and misrepresentations of my position in Roberts’ response.  Let me be clear. Value is always created in the act of production.  But it is realized in the moment of market exchange. I therefore think of value in terms of what Marx calls “the contradictory unity of production and realization.”  Value cannot be produced through market exchange. But it cannot be realized outside of market exchange.  Marx is clear enough about that.

The essence of value is abstract labour or, as I prefer to refer to it, “socially necessary labour time”.  Roberts is obviously correct to say that Marx’s definition is entirely different from the concrete labour time that Ricardo postulated.  No matter whether we say “abstract labour” or “socially necessary”, however, the onus then falls on how the abstraction is made and how socially necessary is to be understood. The answer to such questions has to be grounded in material processes and not constructed through idealist exercises.  So by what materialist process is value constructed if it is not “immanent” in commodities but historically created.

The answer is given in Marx’s starting point in Capital which is the idealized material act of commodity exchange.  If the capitalist takes a commodity to market and there is no want, need or desire for it, then the labour congealed in it is socially unnecessary and it therefore has no value (this is what Marx says at the end of the first section of Capital – p.131 in Penguin/Vintage edition).  This does not mean that value is created in the market (which Roberts incorrectly accuses me of saying). But – and here this may be my peculiar way of looking at it – I take the value created in production to be only a potential value until it is realized.  An alternative way would be to say that the value is produced but then the value is lost if there is no demand for it in the market.  In which case, we would need to construct a strong theory of devaluation to account for what happens in the market place. Devaluation rarely appears in Roberts’ accounts and has no role in his response.  Given my interest in the relation between value and not-value or anti-value this latter formulation might also work for me.  But in either case I think it undeniable that the state of wants, needs and desires backed by ability to pay has an important role to play in sustaining the circulation of capital. This does not mean, as Roberts again and again infers, that this is the only relevant factor in crisis formation.  I have gone out of the way many times to say that this is just one important moment in the circulation of capital where devaluations (sometimes but not always of crisis proportions) can occur.

But again and again Roberts loves to relegate me into that pejorative category of underconsumptionist whenever I mention such matters.  It was Marx, not me, who said “the real root of crises” lies in the diminished purchasing power of the working classes and if I cite Marx on that point it is because it is a neat antidote to all those who endlessly cite the falling rate of profit.  Crises come in many shapes and forms, I have argued.  The falling rate of profit or the collapse of consumer demand are two of many other explanations (I note in passing that Marx in his comments on the crises of 1847- and 1857 – crises that had an uncanny resemblance to 2007-8 – described the crises as commercial and financial crises without any mention of either falling profit rates or insufficient consumer demand).

My objection to any exclusionary productivist interpretation (to cite a matching pejorative characterization!) is that it casts to one side the whole history of creation of wants, needs and desires (let alone the mechanics of ensuring an ability to pay) in the history of capital accumulation.  I think we should pay much more attention to this aspect of things. This does not mean I downplay, deny or refute all the work that has been done on the labour process and the importance of the class struggles that have occurred and continue to occur in the sphere of production.  But these struggles have to be put in relation to struggles over realization, distribution (e.g. rental extractions, debt foreclosures), social reproduction, the management of the metabolic relation to nature and the free gifts of culture and nature. These have all figured large in recent anti-capitalist movements and I insist that we take them all seriously along with the more traditional focus on the Marxist left favoring class struggle at the point of production as the key moment for struggle.  This is why I think the diagram I offer of circulation and the definition of capital as value in motion is so important.  Strange to have it all dismissed in the citation from Murray Smith as “circular reasoning”!!

This perspective opens up some interesting lines of enquiry and points of difference.  Marx’s account over struggles over the working day and the forces that drive technological and organizational changes in the search for relative surplus value all depend upon the “coercive laws of competition”.  That term comes up at various key points in Marx’s argument throughout Capital.  Where is this force mobilized and most clearly felt?  In the market of course!  We cannot understand what goes on in the realm of production (or social reproduction for that matter) without market forces playing their part. It is the coercive laws of competition in the market that mandate capitalist reinvestment and the lengthening of the working day etc.

But this tracks back to how Marx sets up how the abstraction of value – which, by the way, is in Marx’s view, a social relation hence “immaterial but objective” and not “immanent” and “real” as the quote from Murray Smith proposes (“not an atom of matter enters into the objectivity of commodities” says Marx in Capital– p.138).  Value arises not as a product of thought but as a product of a historical material process.  Marx’s study of equivalent and relative forms of value leads into the generalization of exchange which underpins the rise of value as a regulatory norm operating in the market and it is this regulatory norm of value that then returns to dominate behaviors not only in the market but also in the realm of production and social reproduction.  This is a very dialectical move that Marx make but quite commonly encountered in Marx’s work.  Only in this way, for example, can we understand how it is that workers make the capital which then returns to dominate them and how we can all become prisoners of our own products (academics beware!!).

Finally, let me comment on the empirical example on which Roberts reduces final demand to 30 from 70 percent.  To be sure, there is a complicated question of how to deal with value relations across commodity chains (there is an interesting piece by Starosta on Commodity Chains and Marx’s Value Theory in Antipode for 2011).  But imagine a situation where iron ore is mined and the mining company produces value and surplus value that is realized through a sale to a company that produces steel which realizes more value and surplus value through a sale to an auto company that produces yet more value and surplus value that is realized by a sale of autos to final consumers who want and need an auto and have the money to buy one.  The value of the auto is all the accumulated past abstract labour applied.  Suppose for some reason the final consumers cannot pay or get fed up with autos. Then all the accumulated value is lost (devalued).  In practice, as Marx observed, the chain of payments might take a while to work through but when it does then all value production in the chain disappears.

Of course, all sorts of other scenarios can be imagined.  But the point here is that no one apart from crazy people and speculators will want to accumulate steel in the absence of a market for it.  So what happens to value in all of this becomes problematic and Robert’s account makes it seem as if investment in the production of means of production is independent of final demand and can occur without any mind for final market conditions.  Of course, there are certain kinds of investments with all sorts of time lags (fixed capital and infrastructures) like the Chinese overproduction of cities funded by doubling down on indebtedness, where things get very complicated (as I outlined in the final chapter of the madness of economic reason book).  But Roberts’ empirical example makes no sense to me whatsoever as an elucidation as to why realization and the politics of realization are irrelevant or at best a sidebar to the main action at the point of production.

All this and we have yet to get to the thorny questions of money and the politics of distribution along with the circulation of interest-bearing capital in relation to value theory.  Can banks produce value?  They are clearly producing representations of value hand over fist…..Are they a mere side-bar too?

For more on the nature of Marx’s law of value and its relation to crises, see my new book, Marx 200 



[ii] Capital Volume One, p260 trans. Ben Fowkes, New York: Vintage 1977

[iii] Murray Smith, Invisible Leviathan, Historical Materialism, forthcoming 2018

[iv] Cockshott and Cottrell broke down the economy into a large number of sectors to show that the monetary value of the gross output of these sectors correlates closely with the labour concurrently expended to produce that gross output.  Anwar Shaikh also did something similar.  He compared market prices, labour values and standard prices of production calculated from US input-output tables and found that on average labour values deviate from market prices by only 9.2 per cent and that prices of production (calculated at observed rates of profit) deviate from market prices by only 8.2 per cent. Lefteris Tsoulfidis and Dimitris Paitaridis investigated the question of price-value deviations using the input-output Table of Canada. They found for the Canadian economy the results are consistent with Marx’s law of value. And G Carchedi, in a recent paper, showed that the validity of Marx’s law of value can be tested with official US data, which are deflated money prices of use values.  He found that money and value rates of profit moved in the same direction (tendentially downward) and tracked each other very closely.

[v] “The accumulation of capital, though originally appearing as its quantitative extension only, is effected, as we have seen, under a progressive qualitative change in its composition, under a constant increase of its constant, at the expense of its variable constituent.”  Capital Vol 1, Chapter 25

[vi] Grundrisse p748

[vii] Capital Volume 2, Chapter 20