The ‘transformation problem’ and Marx’s crisis theory

images-1by Michael Roberts

The annual London Historical Materialism conference is not just or even mostly about Marxist economics.  As its name suggests, the sessions are about Marxist analyses of all social phenomena.  But obviously for me, the issues in Marxist economics are what matters.  And there are two issues or themes (for me) that arose at this year’s conference.

The so-called ‘transformation problem’

They are not new and have been debated and discussed for over a hundred years.  But old issues die hard (as do older Marxists).  The first is the so-called transformation problem, namely, can Marx’s theory of labour value explain or be consistent with actual market prices in a capitalist economy?  The second is whether crises, slumps in production in the capitalist mode of production, can be explained or are caused by a problem in the production for profit (as per Marx’s law of profitability) or whether crises emerge because capital cannot ‘realise’ its production of surplus value in the market place through sales.  In other words, are capitalist crises due to insufficient surplus value or  too much surplus value that cannot be ‘realised’ in the market, i.e. ‘disproportion’ or ‘overproduction’?

This year’s HM continued yet again the debate on these issues.  On the first issue, Fred Moseley presented his new book, Money and Totality, which I have reviewed before on this blog.

In his book, Fred aims to put to bed the ‘transformation problem’.  And he succeeds.  He shows that Marx solved Ricardo’s problem: namely that market prices of individual commodities do not reflect their value measured in labour time.  The discrepancy is solved through the competition between individual capitals that leads to an equalisation of profit rates and an average rate of profit for the whole system.  Market prices fluctuate around prices of production measured in money, based on the cost of capital invested in money terms and the average rate of profit.  So capitalists start with money and invest in labour and means of production measured in market prices (which revolve around prices of production).  In the circuit of capital and the accumulation process, it is prices of production that rule, not individual labour values for commodities.  So no ‘transformation’ of values into prices is necessary.  Prices are given in money.

Moseley’s book refutes the critique of mainstream economics and what are called the ‘neo-Ricardians’ like Piero Sraffa who either say that the labour theory of value is irrelevant to market prices or that Marx’s solution needs correcting for logical inconsistency.  Moseley is not the first to do this, but his book provides a clear and comprehensive defence of Marx’s value theory.  The debate at the HM session was partly around some ‘Marxist’ solutions that Moseley rejects and about whether Marx’s prices of production can be considered ‘long-term equilibrium’ prices or not.  I won’t go into those controversies here, because I want to discuss that other non-problem, realisation.

A ‘realisation problem’?

I presented the basic ideas of my book, The Long Depression, in one HM session.  One of the discussants, Jim Kincaid, then presented his critique of my work, the substance of which was outlined in a previous post.  But Jim’s underlying criticism, and that was repeated by several senior Marxist economists from the floor, is that Marx’s law of the tendency of the rate of profit to fall cannot be considered the sole underlying cause of crises under capitalism, and in the case of the Great Recession was not the cause at all.  In particular, it is charged that I ignore the ‘problem of realisation’ of surplus value in the market and see the cause of crises only in the drive for profit in production – and yet there are two sides to the circuit of capital: production and realisation.

Jim Kincaid made this part of his critique and, in his new and excellent book, Francois Chesnais, the eminent French Marxist economist, also chided me for failing to recognise in any meaningful way the ‘realisation’ problem. I quote Chesnais from his book: “macroeconomic conditions shaping the capital-labour relations of power prevent the whole of the surplus value produced globally from being realised. Capital is faced by a roadblock at C′ of the complete accumulation process  (M-C . . . P . . . C′-M′)” …The fact that a ‘realisation problem’ exists alongside the insufficient rate of profit is now recognised somewhat reluctantly by Michael Roberts”.

Actually, I am not sure that I do recognise, even reluctantly, that there is a realisation problem as posed by Chesnais and others.  Let me explain. I also participated in a session at HM on a critique of Keynesian policies like quantitative easing and fiscal stimulus to overcome the Long Depression.  There was an excellent paper on the failure of QE by Maria Ivanova and Tony Norfield presented a lucid account of the continued build-up of global debt that threatens a new financial crash that Keynesians ignore or dismiss.

In my paper (the-crisis-and-keynesian-policies) comparing the efficacy of Keynesian fiscal stimulus solutions to a slump with the Marxist explanation (the Keynesian multiplier versus the Marxist one), I likened the Keynesian explanation of crises as due to “a lack of effective demand” as really like a weather man telling us that it is raining because there is water coming down from the sky.  The ‘lack of demand’ explanation is no explanation at all but merely a description of slump (falling investment and consumption).  For this analogy, I was picked up Pete Green for, again, not recognising the ‘realisation problem’ – namely the cause of crises is not (just) to be found in the capitalist mode of production but also in the distribution of surplus value and thus in an inability to ‘realise’ all the surplus value produced.

It’s the law of the tendency of the rate of profit to fall, folks

So it seems many wish to continue to reject Marx’s law of profitability as the main or ultimate cause of crises and seek alternative or eclectic explanations.  Chesnais reckons crises have multi-causes, following the views of David Harvey (see my debate with Harvey on this here).  “Marx discusses a range of issues, not simply the LTRPF but also the over-accumulation of capital and the accompanying overproduction of commodities, as well as raising the hypothesis of the ‘absolute over-production of capital’ which is hardly ever mentioned in today’s debates. So I agree with Harvey that there is no single causal theory of crisis formation”, p23 Finance Capital Today.

Pete Green apparently (at least according to the title of his HM paper) looks to ‘long forgotten’ theories of disproportionality between accumulation and consumption (due to the anarchy of capitalist production); between the expansion of capitalist production and the ‘limits of the market’ for a crisis theory.  This is an idea that comes originally from the 19th century Russian economist, Tugan Baranovsky (who actually argued that there was no ‘realisation problem’) and Marxist Rosa Luxemburg (who did think there was one). Jim Kincaid looks to the idea of too much surplus being created to be absorbed and a lack of ‘profitable investment opportunities’ (akin to the position of Sweezy and Baran).  Chesnais wants to combine Marx’s law with the idea of a crisis of ‘overproduction’, suggesting that falling profitability and overproduction are not connected but are separate (and combined?) causes of crises. “the financial events of 2007–8 were typically an integral part of a crisis ‘in the sphere of money and credit’, the underlying causes of which are overproduction and over-accumulation at a world level along with (my emphases) an effective play of the tendency of the rate of profit to fall.”

None of these alternative explanations or eclectic melanges is new and in my opinion has been dealt with effectively by a succession of Marxist economists.  G Carchedi looked at all these alternative explanations in his seminal work of the 1990s (now apparently ‘long forgotten’), Frontiers of Political Economy. Carchedi comments: “The disproportionality thesis submits that the root of crises lies in the difference between the technologically determined demand for specific use values as inputs of some branches and the technologically determined supply of the same use values as outputs of other branches. Marx’s answer is that those “ price fluctuations, which prevent large portions of the total capital from replacing themselves in their average proportions … must always call forth general stoppages”, due to “ the general interrelations of the entire reproduction process as developed in particular by credit” . However, these are only “ of a transient nature”.  (Marx, 1967c, pp. 483-4).  Thus disproportions can either be determined by price fluctuations, and in this case they are self-correcting and cannot explain crises, or by lack of purchasing power, and in this case it is the latter, rather than disproportions, which explain crises. The disproportionality and underconsumption theories cannot account for the inevitability of crises; but, as we have seen, these theories do account for the inevitability of temporary and self-correcting disturbances. Only the approach linking insufficient production of (surplus) value with technological innovations can provide such an explanation.”

It is no accident that ‘underconsumption’ as such has not been revived as an alternative theory to Marx’s law of profitability.  That’s because the idea that crises are caused by the inability of workers to pay for the goods they have produced has been so thoroughly discredited both theoretically and empirically.  But the theory of ‘overproduction’ beyond ‘the limits of the market’ is really just the other side of the coin of underconsumption.  Overproduction is when capitalists produce too much compared to the demand for things or services.  Suddenly capitalists build up stocks of things they cannot sell, they have factories with too much capacity compared to demand and they have too many workers than they need.  So they close down plant, slash the workforce and even just liquidate the whole business.  That is a capitalist crisis.

Overproduction is the very expression of a capitalist crisis.  Before capitalism, crises were ones of underproduction (namely famine or scarcity).  But to say overproduction is the form that a capitalist crisis takes is not to say it is the cause of the crisis.  To say that crises are like a thunderstorm does not explain why we are wet.  If it were the cause, then capitalism would be in permanent slump because workers can never buy back all the goods they produce.   After all, the difference between what the workers get in wages and the price of the goods or services they produce that are sold by the capitalists are the profits.  By definition, that value is not available to workers to spend, but is in the hands of the capitalist owners.

Marx devastatingly criticised those capitalist economists who claimed that there could never be a crisis of overproduction because every sale that a capitalist makes means that there will be purchaser.  As Marx said, that there is purchaser for every seller is a tautology, the very definition of exchange. Sure, “no one can sell unless someone else purchases.  But no one is forthwith bound to purchase just because he has sold”. The money from a sale can be hoarded (saved) and not used to buy.  That alone raises the possibility of overproduction and crisis.

But the possibility of crisis in the process of capitalist exchange using money does not mean it will happen and provides no explanation of when or how.  So Marx went further and explained that what will decide whether capitalists make purchases for investing in plant or new technology and to buy labour power to produce is the profitability of doing so.  “The rate of profit is the motive power of capitalist production.  Things are produced only so long as they can be produced with a profit”.

And this is where Marx’s law of the tendency of the rate of profit to fall comes in. Marx shows that the profitability of capitalist production does not stay stable, but is subject to an inexorable downward pressure (or tendency).  That eventually leads to capitalists overinvesting (overaccumulating) relative to the profits they get out of the workers.  At a certain point, overaccumulation relative to profit (ie a falling rate of profit) leads to the total or mass of profit no longer rising.  Then capitalists stop investing and producing and we have overproduction, or a capitalist crisis.  So the falling rate of profit (and falling profits) causes overproduction, not vice versa.

As Henryk Grossman explained so well, a falling rate of profit does not directly lead to a crisis as long as the mass of profit can rise.  When a falling rate of profit eventually leads to a fall in the mass of profit and thus overaccumulation of investment and overproduction of goods and services (that are profitable), then the crisis ensues.

As Marx put it: “the so-called plethora (overaccumulation) of capital always applies to a plethora of capital for which the fall in the rate of profit is not compensated by the mass of profit… and “overproduction of commodities is simply overaccumulation of capital”.  It is precisely when the mass of profit stopped rising that the Great Recession ensued.

Thus the so-called realisation problem is the result of the production problem.  Falling profitability and falling mass of profits lead to collapsing investment, wages and employment and then swathes of companies cannot sell their goods or services at existing prices and workers cannot buy them.  This is a crisis of overproduction and underconsumption.

Indeed, only Marx’s law of profitability can explain the cycle of boom and slump, while overproduction or disproportion cannot do so.  See Paul Mattick Jnr’s excellent account of Marx’s law and the business cycle pp48-51 in the best short account of the Great Recession so far, Business as usual.

Political implications

And there is a political implication from the discussion of alternative theories of capitalist crises.  For example, if we think capitalist crisis is caused by overproduction (or underconsumption) relative to the ability of workers to buy the goods produced, as Keynesians do, then the policy answer may be just to boost spending by government or make tax cuts (as Donald Trump plans now, it seems).  Problem solved.  Only, as our session at HM showed – the ‘problem of realisation’ is not solved by these measures, as Trump and the American people will find out precisely because it is not a problem of realisation but of profitability.

On the other hand, if we think it is caused by lack of profit, then there is only one solution for capitalism: destroying the value of existing capital (plant, machines and employees) in order to cut costs and so restore profitability.  Only that will get capitalism going again (for a while), but at the expense of the rest of us.  Thus the inherent contradiction of capitalism is exposed.  Only its abolition will stop the cycle of boom and slump.  The problem of production for profit is a real problem that cannot be resolved.

In my view, ‘too much surplus’, ‘disproportion’, ‘overproduction’ or ‘underconsumption’ are not Marx’s theory of crises.  But more important, they are very weak alternatives to Marx’s law of profitability as an explanation.  They are weak theoretically and even worse, empirically unverifiable.  What are we measuring when we look at ‘disproportionality’ or ‘underconsumption’?  Does consumption fall before a slump?  No, the evidence is clearly to the contrary, unlike profits and investment.  Will disproportionate investment growth compared to consumption lead to overproduction and periodic crises?  Well no, as Andrew Kliman has shown for the US in his book, The failure of capitalist production, chapter 8.  Historically, business investment always grows faster than workers’ consumption – that is the result of capitalist accumulation.  But this does not create a chronic slump or permanent stagnation because investment creates its own demand (capitalist demand).  Indeed, investment drives the productivity of labour and thus drives economic growth.  The problem is when investment collapses, not when it grows ‘too fast’.

Everybody in Marxist economic circles seems to agree that the crises of the 1970s and early 1980s were the result of falling profitability rather than overproduction or underconsumption.  But you see, the argument now goes, each crisis can have a different cause because capitalism metamorphoses into new forms or structures (neoliberalism or financialisation) that change the causal contradictions.  So we are now being told that, because profitability rose after 2001 up to the Great Recession, Marx’s law does not apply and we need to consider that the GR was the result of either financial instability, excessive credit, rising inequality and falling wage share, or weak demand and secular stagnation.

Well, none of these alternatives seems convincing to me.  As Alan Freeman recently said in his paper in the book on the crisis, The Great Financial Meltdown: Marx’s law remains “the only credible competitor left in the contest to explain what is going wrong with capitalism”.


Francois Chesnais has kindly sent me a rejoinder to my comments on his view of the current Long Depression and its causes:

“One is faced with a situation where has been a continuous drive by capital to raise the rate of exploitation in the course of the crisis yet the conditions of profitability and new investment have not been restored.

In the explanation offered the starting notion is that of “fresh fields of accumulation” (Luxemburg) as necessary to end a long recession, a further one the function of crisis in capitalism and the key central one that of “dearth of actual surplus value”.

One must start by looking at were the “fresh fields” that ended the two previous world long recessions or depressions. In the case of the 1880s it was a combination of an extension of the world market, a reach out to the many regions not properly include or not included at all, and of the opening for profitable investment over a long time span of the major industries of the “Second industrial revolution”. In the case of the 1930s it was as you said very clearly in one of your talks it was the Second war world.

Then one must look at whether a crisis is allowed to play its function of destroying existing productive and fictitious capital, of clearing the deck for new investment. This was the case in 1929 and the 1930s.

So what is the picture today? We have a situation which combines the fact that the crisis was not allowed to play its function of destroying existing productive and fictitious capital, of clearing the deck for new investment on any significant scale and that since a World war is not in preparation the only candidate as a “fresh field” would have to be new technologies associated with the emergence of whole new industrial sectors with strong new employment creation effects. The ones there are do not have this quality, on the contrary.

A low level of investment means a low creation of surplus value.  “Virtuous cumulative accumulation process” are one where profit expectation drives investment of a strong enough employment effect as to generate both surplus creation and demand permitting the completion of the accumulation cycle M-C-P-C’-M’.

Today not enough surplus value is being produced to re-launch the accumulation process and the amount that is serves to consolidate the accumulation of dividend and interest bearing assets by banks, funds and individuals (financial accumulation) and so the claims on this already very insufficient amount of surplus value. This has led both to the dead-end of the quasi-zero long term interest rate regime, which not simply the outcome of quantitative-easing and to the endless small shocks in the global financial system.

Of course government debt and the resulting pro-rentier, pro-cyclical austerity policies only aggravate this situation but they do not explain it and their reversal would not solve capitalism’s basic problem.”

This piece first appeared at Mike’s excellent the next recession blog, here.




  1. In In Chapter 9. Formation of a General Rate of Profit (Average Rate of Profit) and Transformation of the Values of Commodities into Prices of Production, of Capital, volume III, Karl Marx states: “If there are, therefore, n spheres of production, and if each makes a profit amounting to p, then their aggregate cost-price = k – np.” If instead of doing so, profits were not substracted but maintained, he would have obtained productions prices proportional with values according to the rate of surplus value. This had to be done with the “sectoral rates of profit” in labour values. The problem of no proportionality between total labour values and total prices and total profits with total surplus value lies only in the calculation of a common rate of profit, even if this common rate is the weighted average one.
    Besides this, in chapter 50, volume III dealing with prices, wages and profits, KM is clearly “sraffian”.

Comments are closed.