Capital's inherent contradictions

I think it’s fair to say since the 1970s debates on explanations for capital’s historic crisis have been pretty much polarised been supporters of the ‘saturation of markets’ versus the falling rate of profit (FRoP) or Luxemburg v Grossman/Mattick.

Through discussions, especially with Link, I’ve recently come to accept that Luxemburg’s theory of capital’s need for non-capitalist buyers is wrong because it is incompatible with Marx’s analysis of capital’s contradictions, and looking back at your lengthy critique of Luxemburg written in 1976 (!) I can see that many of your criticisms were perfectly correct. But, apart from her lasting contribution as a revolutionary, I do still believe she was trying to address a real problem identified by Marx.

Looking at Chapter 15 in Capital Vol 3, which as you well know is on the workings of the FRoP, Marx also talks about the problem for capital of realising the surplus value produced - the famous ‘second act’ of the capitalist production process. Briefly, the fundamental problem for capital is that its only motive and purpose is its own self-expansion, the production of an ever-expanding mass of surplus value; but for accumulation to take place the mass of commodities produced must be sold and the largest possible portion of profit reconverted into capital, and here capital faces barriers due to its own relations of production, which restrict society’s capacity to consume while at the same time generating an ever-growing mass of commodities…

For Marx, capital can only attempt to resolve the inherent contradiction between the conditions for the production of surplus value and its realisation by continually extending the market. However, the growth of productivity continually threatens to outstrip the capacity of the available market, re-creating the same problem in conditions of expanded production; the more capital develops, the more it comes into conflict with its inherent limits on consumption. In other words, periodically, and at a higher and higher level, “Too many commodities are produced to permit of a realisation and conversion into new capital of the value and surplus-value contained in them under the conditions of distribution and consumption peculiar to capitalist production.” (Ch 15).

If this is right I can only conclude that the circulation or consumption of commodities, far from being a secondary question as is sometimes argued by supporters of the FRoP, is not only integral to the ability of capital to accumulate, but also the source of inherent contradictions as a result of the capital-wage labour relationship itself: the more capital develops, driven by the falling rate of profit, the more it comes into conflict with the inherent limits on society’s capacity to consume. And the inescapable problem for capital appears to be that if there are no set limits to the expansion of the internal market, there are definite geographic limits to the expansion of the external market.

Of course whether the problem of realisation or the falling rate of profit is the key factor in explaining the entry of capital into its historic crisis – or whether as is more likely it is an interplay of these and other factors – will probably still be being debated well after capitalism’s revolutionary overthrow… But, to avoid false arguments, is it possible to agree that capital’s problem of realisation, and therefore of finding markets for its ever-growing mass of commodities, is an inherent contradiction of the mode of production and at least potentially a decisive factor?

I would be very interested in comrades’ views.


My critique of Luxemburg, plus a contribution from Link, can be found here


Marky - Just spotted this. Cannot open the link to you and Link which it would be good to read before replying.

Yes I seem to have problems getting links to work; here'a another go:

Mine: Link's:


Also your 1976 text for reference:

Thanks for your comment and sorry not to reply earlier. The links in your second post worked but then the links from your blog back to our articles didn't. We actually addressed this directly in one of our replies to C McL (who was then charged with taking up the cudgels with us by the ICC - we see today he now denies that capitalism is decadent).

"Market gluts and over-production are an obvious feature or, as Marx says in Theories of Surplus Value Volume II, “phenomenon” of capitalist society, and set it apart from all other previous societies. In Capital, Marx sought to explain how this society capable of creating vast quantities of cheap commodities was suddenly thrown back on itself, and despite the self-evident poverty all around, was, in certain crisis conditions, incapable of maintaining, let alone extending, its market. We believe he found these in his analysis of the operation of the law of value, and it is this which determines the cycle of expansion and contraction of both the market and the forces of production. This is totally different from the wretched Say (and other capitalist apologists like McCullough, the editor of Adam Smith’s works) who tried to deny that there were crises and that there was overproduction under capitalism. The question is not whether these take place (as they self-evidently do, we can move on from red herrings like James Mill, McCullough and Say, etc.), but to explain how they come about." This is in our article from 2007 on The Dynamic of Capitalism. The law of the tendency of the rate of profit to fall is not just an explanation of periodic crises but also explains the drive of capitalism and its tendency to constant expansion. As Marx wrote in Chapter 15 "On the other hand, the rate of self-expansion of the total capital, or the rate of profit, being the goad of capitalist production (just as self-expansion of capital is its only purpose), its fall checks the formation of new independent capitals and thus appears as a threat to the development of the capitalist production process. It breeds over-production, speculation, crises, and surplus-capital alongside surplus-population." The question is why is capitalism capable of expanding the market at one point only for it to be thrown back into crisis in another. What is the mechanism which produces both these conditions? It has to lie in the way the law of value operates. Marx seems clear that it is the fall in the rate of profit which determines the process of both expansion and contraction and after a lengthy analysis in which he shows that the mass of capital can keep on growing whilst the rate of profit is falling until at a certain point a collision occurs because "The real barrier of capitalist production is capital itself." Even when Marx says that in the last resort the cause of all real crisis is the poverty and restricted consumption of the masses he goes on in the next page to say this is not because too many commodities are produced to satisfy human wants but indeed people are living at a low level - its just that capitalism won't provide for them at a cost which they can afford (as capitalism is driving down their wages as the rate of profit falls). In short it is the "antagonistic social relations" which drive this overproduction but these rest firmly on the question of producing valorisable capital. I don't really get your distinction between external and internal for our time since the geographic limits to the external market have not changed in a hundred years i.e. the world is totally capitalist (or as near as makes no difference) so we come back to the same question - why is it that capital can expand at one time and then collapses into periodic crisis at another. This produces among other things "overproduction" but this is a phenomenon not the cause which is hidden from view and harder to grasp.

Thanks for the clarification of your position. I have read your response to CMcL but wanted to be clear what your current views are. I'll consider carefully what you have said and respond in due course...


As MH says we have been having questioning everything and our discussions about the errors in Luxemburg’s theory of accumulation and moved on to issues of overproduction and rate of profit.

I find the argument is often presented that overproduction is about being purely market related and purely related to the working class’s ability to purchase. This seems to me to be quite clearly an insufficient argument, its impossible by definition and so of no help at all in understanding capital’s problems. Overproduction must relate to both production of constant capital as well as variable and as constant capital grows faster and larger and so it seems to me then that the focus of overproduction must be about the amount of constant capital in existence.

I think I remember you criticising Grossman but I cant remember why but having looked at some of his writing again I discovered he says much the same in that he focusses on over-accumulation as the problem for capital. This seems to make a lot of sense to me and over-accumulation of capital seems a close relation of FROP in that its the same thing or a product of the same process of accumulating capital. He says correctly I think that the a specific figure for the RoP is not what generates crises and sees an interaction between this and the mass of surplus value produced as being a factor in producing such crises. As you yourself paraphrased - A high rate of profit may generate a small mass of surplus value whereas a low rate of profit can generate a high mass of profit.

This does seem to describe well how capital has progressed over the periods of ascendancy and obsolesence and it does integrate the issues of production and circulation as MH is suggesting.

MH and I came to the same conclusion some time ago that neither decadence or obsolescence were based on economic criteria but rather on political/social/historical factors. So neither the actual rate of profit nor the proportion of pre cap markets could be a determining factor in the onset of crisis (I being kind there, pre cap markets have nothing to do with it anyway). This suggests that other factors must come into play to interact with TROPF to provoke outright crisis and as you add and ’... but these rest firmly on the question of producing valorisable capital’ relates it to circulation issues.

We would all agree that contradictions of capitalist relations of production are key. As C says: "The real barrier of capitalist production is capital itself’

And this is where I think C’s comments on causal and phenomenonal factors are useful way to approach things but im not sure you are saying this but that sounds to me like an interpretation would be that capital or capitals relations of production is the source and that the TROPF is peripheral alongside overproduction?

Anyway i suppose i am asking here for a wider explanation of cause and phenomenon?

I have a separate but related point to raise if you don’t mind. What role does waste production play in all this in the past century? It a burden on the wc generated by sv that in not capitalised yet it has increased enormously. Its not possible to quantify and does not impact on say rate of exploitation and rate of profit, yet it keeps an awful lots of wheels turning in social terms. In these terms it sounds like the market must absorb overproduction if capital is to stay healthy so I wonder what impact the capacity to absorb waste production can have. As a peripheral factor, how do you see it impacting on crises?

Link I am not really sure who your questions are directed towards. All I can say is that all arguments about the falling rate of profit, the organic composition of capital and the cycles of accumulation have to be understood in terms of the system as a whole. Thus you can find that waste production will keep the wheels turning for some capitalists (at best recycling the surplus value created elsewhere). Arms production is a classic example. Its product cannot re-enter the reproduction process either as Dept I or Dept II but it undeniably makes a profit for those companies and states who sell arms to other states. It is certainly not " a province of accumulation".

In response to Cleishbotham's first reply to me:

Yes, the ICC article by CMcL accused you of defending the same position as those bourgeois political economists like Say and Ricardo who saw the production and realisation of capital as identical and therefore denied the possibility of overproduction. But you clearly accept the existence of overproduction as a phenomenon; the question for you is how does this phenomenon come about?

You argue that it is Marx’s analysis of the law of value that determines the cycle of expansion and contraction of both the market and the forces of production. It is therefore the rate of profit that results in periodic crises and it is these periodic crises that produce the phenomenon of overproduction.

I would argue that for Marx there appears to be something more fundamental at work here than the periodic appearance of a glut of commodities on the market that the working class cannot afford to buy: a tendency of capital towards overproduction which is the result of inherent limits on capitalist society’s ability to consume; the workers, after all, must by definition produce more than they can themselves consume, otherwise there would be no profit, while the capitalists cannot simply consume the total mass of commodities themselves: “The consuming power is furthermore restricted by the tendency to accumulate, the greed for an expansion of capital and a production of surplus value on an enlarged scale. This is a law of capitalist production imposed by incessant revolutions in the methods of production themselves…” (Capital 3 Ch 15, my emphasis)

So the available market must always tend to be outstripped by the needs of accumulation. This is why "The real barrier of capitalist production is capital itself". The market must be continually extended; but this only results in the further growth of capitalist production so the problem of realising surplus value is simply recreated at a higher and higher level.

I don't really get your distinction between external and internal for our time since the geographic limits to the external market have not changed in a hundred years i.e. the world is totally capitalist (or as near as makes no difference) so we come back to the same question.”

That was actually my point; capital reached the geographical limits to the extension of the external market about a century ago, which logically must precipitate some kind of crisis of the system for the reasons described above. The history of the last 100 years surely demonstrates the result: chronic overproduction and increasing destruction of capital. For me this points to the problem of realisation as at least a potential cause of capital's historic crisis.

My intention in raising this was not to go over (very) old ground again but I do think the problem of realisation in the past has tended to be conflated with a defence of Luxemburg’s theory, which has had the effect of disguising the extent to which it is based on Marx’s analysis of capital’s contradictions, distinct (albeit intimately connected to) the falling rate of profit.

I have read this a few times but honestly I really do not understand the last paragraph.

At least the rest of it is clear then? ;-) You mean the very last para? I was simply referring to the fact that this debate has always tended to be posed as Luxemburg v Grossman/Mattick, in which the defenders of Luxemburg's theory tend to argue she was simply taking up Marx's analysis of the problem of realisation (as in the case of the ICC article you refer to). I don't think this is the case. It is therefore possible to agree that Luxemburg's theory of capital's need for external buyers is wrong - but at the same time I believe it is necessary to recognise that capital's problem of realising surplus value, as analysed by Marx, is indeed an inherent contradiction due to the capital-wage labour relationship.

The problem is one of valorisation (verwertung) which implies not just that things are sold but are sold with added value, "Realisation" as a translation does not carry this connotation so clearly (see the Penguin Classics Volume 1 of Capital). The question though that the pure overproductionists cannot answer is why at one point in time is it possible for valorisation to carry on so happily and then at another the process starts to slow down. This can only be because as accumulation proceeds the organic composition of capital rises and thus the rate of profit falls. This can lead to new investment by some capitalists who can then sell above value but such new investment only increases the ration of dead to living labour and restricts the possibility of increasing the rate of profit further down the line. And the tendency for profit rates to equalise across the economy leads eventually to a general slowdown. Trying to claw back surplus value via wage cuts and producing more commodities in an attempt to sell each with a greater value component are the twin roots of overproduction - but only a value analysis can explain how it comes about. I have nothing to add to the paraphrase of Chapter 15 of Capital Volume 3 above.

As an aside there is some controversy about the geographical question of the absorption of the entire planet into the world market. Marx and Engels thought it had occured in about 1850 but Bukharin later argued that a "world economy" only formed in the late 19C (his explanation for imperialism). Pannekoek in later life came round to the theories of Luxemburg but concluded in the 1940s that there were still huge areas outside the capitalist world market (he cited China and India) so considered that capitalism could happily expand for some time yet.

The question though that the pure overproductionists cannot answer is why at one point in time is it possible for valorisation to carry on so happily and then at another the process starts to slow down. This can only be because as accumulation proceeds the organic composition of capital rises and thus the rate of profit falls.”

I’m not sure who the “pure overproductionists” are and in any case I think we agree it is not the periodic appearance of overproduction per se that is the issue here but what this reveals about the deeper contradictions of the system as a whole and its historical trajectory.

For you, following Marx, the contraction and expansion of the market is determined by the law of the tendency for the rate of profit to fall, which is the explanation for capital’s historic crisis. For me, following Marx (hopefully!), although I don't in any way dismiss the role of the FRoP, the fundamental contradiction of capital is that it produces without regard for the market, which is inherently restricted due to the capitalist-wage labour relationship, and it is the resulting tendency towards overproduction that precipitates its historic crisis.

The main arguments are hopefully clear enough by now so I’ll leave it there.

Except I’m not entirely clear what your view is on the ‘status’ of this question. You have previously described Luxemburg’s theory as outside of Marxism but does this extend to the problem for capital of realising surplus value, ie. of markets?

To put it another way, is the debate we've been having here one essentially between different interpretations of Marx’s analysis of capital’s contradictions within the Communist Left? Could you be a member of the CWO, for example, and defend a different analysis to the FRoP?


Overproductionism (and its twin underconsumptionism) are both epiphenomena. Capitalism's distinctness as a mode of production is precisely that, a mode of production. If the problem was simply markets Capital would have been as slim a volume as the History of Swiss Naval Victories. Or as Marx said in the introduction "if things and their essence were the same what need would there be for science?". The secret to to the special crises of capitalism lies in its unique mode of production. The market is not unique to capitalism (although its role is mightily enhanced by the dynamism of such a productive system). And when we look at value relations under capitalism we are well aware that the law of the tendency of the rate of profit is "the most important law" for explaining not just capitalism's periodic crises but also capitalism's extraordinary dynamism. The markets expand and contract according to the ability to invest but the propensity to invest only comes from the recognition that it is worth it ie, the rate of profit makes it worthwhile. This explains the cycles that capitalism operates in, explains the historical process of the centralisation and concentration of capital which in turn explain the tendency to monopoly which leads to the operation of the state to keep the system going as well as the transfer of rivalry from that of the firm to that of the nation-state. This is the science. The other "interpretations" don't even get close to that.

Look at the fate of those who have tried to deny the analysis of value relations as the basis of capitalism's complex story. The ICC originally argued that 1914 was the end of all expansion (the first version of their Decadence of Capital pamphlet wove elaborate statistics to try to prove this). For them the crisis which opened at the end of the 60s (which was more political than economic initially but then merged with the economic crisis after 1971) was the final crisis. This led to the perspective which they maintained until c 1990 that the way to revolution was open. Collapse of that perspective led to the post-modernist adoption that all was chaos and decomposition (the only germ of truth in that is that the 4 decade long crisis is deepening as it approaches its 5th decade). Unclear whether the class war is over or what. Unclear whether we are still trying to create the conditions for a new international. Those that have split have ended up in even worse places (like C McL) because they still refuse to accept value analysis.

The reference to Rosa Luxemburg putting herself outside Marxism was simply to say she was telling lies when she reinterpreted Marx's discussion of the "outlying fields of production" as really meaning "the outlying fields of consumption". I don't think it was intended to mean that Luxemburg had abandoned Marxism in every sense and whatever errors she made (as we all do) she was a defender of the working class to the end of her days. That is the class line.

But theory is important. Our whole framework as an ICT rests on our understanding of how capitalism functions. It was the basis for our formation and it has served us well. We predicted the bursting of the speculative bubble in 2007 because of it. Our Platform is underpinned by it. Difficult then to see how anyone rejecting our analysis of how capitalism functions (and fails to function) could be comfortable in the ICT.

“Capitalism's distinctness as a mode of production is precisely that, a mode of production. If the problem was simply markets Capital would have been as slim a volume as the History of Swiss Naval Victories.”

Well we know that Marx never completed his projected volume on the role of the world market but he certainly saw the establishment of the world market as one of the three “cardinal facts” of capitalist production and unless we think Engels did a particularly bad editing job on Capital Volume Three we also know he said quite a lot of things about the importance of markets for capital, even in his chapter on the workings of the falling rate of profit.

In any case, no one ever said the problem was “simply” markets and I’ve been very careful throughout to emphasise that the problem of realising surplus value is intimately connected to the falling rate of profit which helps to drive capital’s inherent tendency towards overproduction.

“Capitalism's distinctness as a mode of production” is surely that it is not production to satisfy human needs but production for capital as an end in itself, which creates the unique problem, as you say above, of valorisation, “which implies not just that things are sold but are sold with added value.”

Very true. In order for valorisation to take place surplus value must not only be extracted from the worker; the resulting total mass of commodities must be sold. “Capitalism’s distinctness as a mode of production” is therefore that it is entirely dependent on markets in order to achieve its only purpose of valorisation. The specific problem for capital is that production and the market are independent of one another and the latter is restricted by society’s ability to consume.

Markets therefore are not only a vital necessity for valorisation to take place but they are inherently problematical for this distinct mode of production. To talk of valorisation without addressing the problem of markets is surely to address only one stage of the total process that Marx describes. We need to see it as a totality.

We don't talk of the problem of valorisation without talking about markets but you still have to explain why the markets expand happily at one time and then suddenly cannot. That is the totality of the question and it is underpinned by the law of value.

I didn’t try to pursue this debate at the time, partly because there seemed no particular urgency to clarify what is obviously a long-standing theoretical question, but also because there were clearly some frustrations on both sides that our positions were not really being understood, and I think it’s worth recognising that such frustrations have been a real factor, and continue to be a factor, in discussions between comrades who still share the same basic political positions.

That said, I also think fior me this exchange did succeed in moving the debate forward and finding some common ground, which was in a recognition that the process of valorisation must be viewed as a totality: as you say above, “The problem is one of valorisation (verwertung) which implies not just that things are sold but are sold with added value.” While you still insist that it is the fall in the rate of profit that explains why overproduction occurs at one time and not another, you do also appear to affirm that the problem of valorisation includes the problem of markets.

Is this a fair summary?

If so, while there are still clearly real differences here I think this moves us away from the problem I identified at the start of the thread: that since the 1970s debates on explanations for capital’s historic crisis have tended to be polarised between supporters of the ‘saturation of markets’ versus the falling rate of profit (FRoP) or Luxemburg v Grossman/Mattick.

So I’ll leave it there, except I must pick you up on your accusation above that Rosa Luxemburg was “telling lies when she reinterpreted Marx’s discussion of the “outlying fields of production” as really meaning “the outlying fields of consumption”.

Did you really mean to accuse Rosa Luxemburg – someone you recognise as “a defender of the working class to the end of her days” – of telling deliberate falsehoods in order to deceive people about what Marx actually wrote? Or was this just a bit of polemical excess? Personally I think Luxemburg was mistaken but I don't think she was deliberately trying to deceive people. Can I make a plea that we all avoid this kind of accusation unless we are prepared to back it up with evidence? It just gets in the way of comradely debate within the Communist Left.


I do not see how the debate has moved on at all since we have always agreed with your statement. "While you still insist that it is the fall in the rate of profit that explains why overproduction occurs at one time and not another, you do also appear to affirm that the problem of valorisation includes the problem of markets."

Capital is a market oriented system which depends on it for the sale of its commodities. Marx asked in the preface to Capital Vol 1 that "if things and their essence are the same what need is their for science?" And our question has always been the same. The market is a given, a constant, in the system so why is it that at one time it expands furiously and at another it contracts? As we have said this is a phenomenon which needs explaining and it has to be found in the shift in value relations such as that brought about by the law of the tendency of the rate of profit to fall.

And let's also be clear Rosa Luxemburg did not talk about "saturated markets". She talked about eh absence of the "third buyers" from outside the system. It was a gloss put on it by the ICC. If the third buyers were now gone by 1914 then the condition of the system was one of permanent saturation of markets and a great deal of effort was expended to try to show that this was the case (an effort that was later abandoned). Without an understanding of how the cycle of accumulation operates it is impossible to explain why capitalism expands dramatically at one point and then contracts rapidly at another. The ICC in the past simply denied there were cycles which was one reason why their perspectives constantly insisted on proletarian revolution/upsurge being on the agenda. If the proletariat was only "demystified" of support for leftist reformism then the road to revolution would be open. This took them from a basis in Marxism to some form of idealism.

On Rosa Luxemburg's lie - yes I meant she lied to herself but as she is dead I am sure she took no offence. To a living person it would have been unhelpful but we have no gods however much we value their contribution in other ways.

“I do not see how the debate has moved on at all since we have always agreed with your statement. "While you still insist that it is the fall in the rate of profit that explains why overproduction occurs at one time and not another, you do also appear to affirm that the problem of valorisation includes the problem of markets."

I started this thread in order to better understand your view on the problem identified by Marx of the realisation of surplus value, trying to avoid the whole issue of Luxemburg's theory that has dogged debates in the Communist Left for so long. We appear to agree that without the realisation of surplus value there can be no valorisation, and that markets are a problem for capital.

So your criticism of Luxemburg is quite unnecessary here. You’re quite right, she didn’t talk about "saturated markets", but neither did I. This rather proves my point that the false polarisation between supporters of the ‘saturation of markets’ versus the falling rate of profit (FRoP) has served to obscure the fact that she was trying to address a real problem.

That said, I’m not sure you really do agree with my statement since in the same post you refer to markets as “a given, a constant”, which seems to contradict the idea that they are inherently problematic for capital?

"Without an understanding of how the cycle of accumulation operates it is impossible to explain why capitalism expands dramatically at one point and then contracts rapidly at another”.

Yes and in response I have agreed that the fall in the rate of profit is a key driver. But we clearly don’t agree on the emphasis you put on this. For me, understanding the ‘cycle of accumulation’ does not in itself get us to an understanding of why, in the real social and historical conditions of capitalism, the problem of realising surplus value is a contradiction that helps drive it towards its ‘dissolution’. Marx actually described accumulation as becoming an ever-increasing circle or spiral, which seems to me a more accurate description of its continual attempts in reality to overcome the insoluble contradiction created by its own antagonistic social relations.

A last point:

On Rosa Luxemburg's lie - yes I meant she lied to herself but as she is dead I am sure she took no offence. To a living person it would have been unhelpful but we have no gods however much we value their contribution in other ways.”

This seems a bit disingenuous to be honest. Whether your target is living or dead, I think you should be more careful before accusing someone of lying. That’s all.

That said, I’m not sure you really do agree with my statement since in the same post you refer to markets as “a given, a constant”, which seems to contradict the idea that they are inherently problematic for capital?

This is the nub of the discussion. Markets are not "inherently problematic" but only problematic at certain points - cue: when the value relations have shifted in the course of cycle or the rate of profit is insufficient for the given mass of capital. We then get the contradiction that too many commodities are produced to be valorised (and yet needs remain unsatisfied as the creators of new surplus value have only a restricted power of consumption).

Agreed re the spiral since the cycle of accumulation brings about an ever increasing centralisation and concentration of capital which tends to monopoly, state capitalism and imperialism. It has always underpinned our analysis of why capitalism's decadence begins from the early years of the last century. It also explains why each cycle (including the one we are now living through) has different characteristics.

I'll revive this thread not to return to its precise arguments, but to post a comment in general connection with the dispute about Rosa Luxemburg's theory (which comrade Link criticises). I know, I know, you must be thinking, not this damn dispute again! My comment is hopefully also of significance, beyond that specific dispute.

My thought is as follows:

It's sometimes argued (eg by Link) that the sheer obvious fact of any GDP growth (now matter how small) existing (after the almost complete conquest by capitalism of the world), already proves that accumulation takes places, and so Luxemburg's theory is dead false, QED.

GDP doens't tell us anything about surplus value, so one has to find a Marxist alternative concept, which can be called eg the annual "Total Value", from which the percentage of surplus value can be read. For the year 2000, the share of surplus value in the US "Total Value" was 39,73%. I came up with this number from the interesting chapter "Estimating gross domestic product with surplus value" by Victor Kasper (pp. 121-183 in Confronting 9-11, Ideologies of Race, and Eminent Economists, 2002). Btw, professor Kasper uses gold as the price unit (the book can be found on a free book download site).

Since obviously US GDP didn't grow 39,73% (from 2000 to 2001), not all of that surplus value was productively invested or "accumulated". Comrade Link insisted only on the fact that at least "some" accumulation took place, but Rosa Luxemburg is interested in the possibility of the realisation+accumulation of "all" surplus value. The annual GDP growth figure doesn't show that all surplus value has been accumulated.

US goverment expenditure (I just looked up for eg 2022) is 38,1% of GDP. Even if one would assume that the government distributed this collected surplus value only to productive sectors, the fact remains that GDP growth rate doesn't annually increase by that much. As for the other candidate, like luxury consumption, Luxemburg wrote:

But this is a conjuring trick, seeing that the capitalists consume the surplus value in so far as they spend it on their luxuries – they do not accumulate at all. My question is about the possibility of accumulation, not whether the personal luxuries of the capitalists are possible. Answer this clearly, if you can, or else go play with your wine and tobacco, or go to blazes for all I care.

Even if we assume the share of surplus value (39,73% of annual Total Value) was all privately consumed by capitalists (and sustained the workers in the luxury industries), the fact remains that this didn't cause GDP to grow 39,73%.

So where did all the surplus value go to then, if it not was towards the increase of GDP? Was it thrown in the seas? I don't know where it went. But I think, based on the GDP growth figures, I must conclude, that it didn't accumulate. Rosa Luxemburg's claim is that (absent a third-buyer), surplus value couldn't accumulate, and the sheer fact of existing GDP growth rates prove her right, that the majority of the surplus has not been accumulated.

Let's assume that the share of surplus value was one-third (33,3%) of the annual Total Value (or "GDP") for the past century or so in the world. If surplus-value was entirely accumulated each year, the absolute GDP mass would double in less than 3 years from the start (this is even without calculating the compound growth). Clearly, we haven't seen such phenomenal GDP growth. If one objects here, that wars and crises don't simply pause or slow accumulation, but actively detract from it, and so cause for some periods a negative GDP growth rate, the detraction would need to be above the range of 33,3%, to begin to have any negative effect at all.

As I said, the significane of estimating the share of surplus-value (in total annual production) lies beyond just this dispute. For example, the share of surplus-value is perhaps an indication of the superfluous labour time now spent in capitalism, that would be reduced in socialism.

For sake of disclosure, I had been engaging with comrade Link in the dispute on Luxemburg's theory on the former forum of the ICC. Just one point that I want to repeat here (if I can still correctly remember it), namely from my exegisis of the militarism chapter in her book, – a long complex argument, which gets misinterpreted often (even in the ICT's text from 1976):

If you look closely at Luxemburg's model (of spending on armaments), her result is that the total value of the social product actually decreases (the "GDP" falls, if you will), and it is the surplus-value that increases (absolutely). It is only logical that the total value decreases in her model, because, as the ICT indeed pointed out, militarist spending on arms is unproductive. But how can Luxemburg then use her model as proof for her argument? The conceptual point here (and in general important for the concept of "accumulation") is, that the capitalist class is not interested in increasing the total value of the social product; it is rather interested in increasing the surplus value. Luxemburg wrote:

The social product is smaller and less capital is now employed – but then, the object of capitalist production is not simply to employ as much capital as possible, but to produce as much surplus value as possible.

In Luxemburg's model, the surplus-value increased (and therefore militarism is sort to speak "productive" for accumulation).

The way this happens for Luxemburg is twofold (and quite specific); in one part, by a conditional increase of the exploitation (really switch to a super-exploitation) of the whole (not just in arms industries) working class through inflation/tax, which decreases their wages below the value of their labour-power. The second part, is just taxes on non-capitalist source (peasants/artisans). These two sources are not unlimited, and so once these conditions or methods expire, militarist spending is no longer a way to accomplish accumulation.

Not to be obnoxious, but I noticed the translation of an (IMO conceptually important) sentence in Luxemburg's book had been totally wrong in the old English version (by Agnes Schwarzschild). It has been largely corrected in the new edition (2015), translated by Nicholas Gray. It's in the first chapter (I correct the sentence here further by the word "thus"):

Capitalist production is not production for the purpose of consumption, but the production of value. Value relations govern the entire production process as well as the reproduction process. Capitalist production is not the production of consumer goods, nor is it merely the production of commodities: instead it is the production of surplus value. Expanded reproduction, in capitalist terms, means the extension of surplus value production. Surplus value production does indeed proceed in the form of commodity production, and thus, in the last instance, as the production of consumer goods. During the course of reproduction, however, discrepancies between these two facets are continually caused by variations in the productivity of labor. As productivity increases, the same amount of capital and of surplus value represents a progressively larger volume of consumer goods. The expansion of production in the sense of the production of a larger mass of use-values [thus] does not necessarily imply expanded reproduction in the capitalist sense. Conversely, capital may, within certain confines, extract a greater surplus value through an increase in the rate of exploitation (e.g. through depressing wages) without producing a larger volume of goods. However, in both of these cases, the elements of expanded reproduction are produced as capital, since these elements are surplus value both as a magnitude of value and as a quantity of material means of production.

^ The highlighted sentence is important because it stresses, that the mere fact of an increase in the production of material/physical goods, as you would find reflected in statistics like GDP growth (or a manufacturing production index), does not yet indicate that accumulation (which interests us as Marxists) has even taken place at all, given that we remained stuck at "the same amount of capital and of surplus value". I think it should also caution us against quick reliance on the notion of a "surplus product" (cf. the recent use in some paper of the concept of a "Baran ratio", applied to centuries of English society up to the 1860s).

In the next sentence ("Conversely...") btw, Luxemburg envisions a scenario where accumulation does take place, that is, an increase only of the surplus value and so as a consequence (I infer) also an increase of "total value", but just without an actual increase of goods (GDP). It seems similar to her later case of armaments spending (that I mentioned above), but not really, because there "total value" certainly decreased (as well as a decrease of goods/GDP, namely subsistence goods for workers).

Finally, the last sentence ("However...") is a bit difficult to understand for me, given my interpretation of the preceding sentences. In German it read:

Aber in diesem wie in jenem Fall werden gleichermaßen die Elemente der erweiterten Reproduktion im kapitalistischen Sinne hergestellt. Denn diese Elemente sind: Mehrwert sowohl als Wertgröße wie als Summe von sachlichen Produktionsmitteln.

It seems Luxemburg here regards both of her two scenarios as examples of expanded reproduction (accumulation), whereas, as explained, her first case is not an example of accumulation. So it seems Luxemburg is here unclear (ie incorrect about her own cases). And this error by Luxemburg I think (partly) could also have caused Agnes Schwarzschild then to so oddly (totally) mistranslate the first sentence, namely as an attempt to make the whole passage more consistent:

Expanding production, understood as the creation of a greater amount of surplus value, need not therefore necessarily imply expanding reproduction in the capitalist meaning of the term.

Btw, in the same first chapter Luxemburg makes clear that "accumulation" is not just realisation (in money) of the surplus value, but an increase of the surplus value (what she calls the "fourth condition"). Without an increase of surplus value, there is no capital accumulation.

I'm far from considering the hitherto silence from others here as an indicating their agreement with the Luxemburgist understanding/definition of the concept of "accumulation" as an extension of surplus value, that I posted above (perhaps only Marky would agree, since he wrote in his first post about it being "an ever-expanding mass of surplus value"). I consider the definition is controversial. It will be clear why this definition leads to some pretty outlandish conclusions, if I further spell out its practical meaning with an example. I already mentioned how fast GDP growth would have looked, if there was full capitalisation of the surplus value, making up one-third (33,3%) of total value of the social product: GDP would double within 3 years even without compound growth. It would increase 10 fold within 30 years. GDP would increase 40 fold within 120 years (this is close to, in fact a bit greater, than real GDP growth from 1900 up to 2020). The outlandish conclusion though is, that according to Luxemburg's definition, this isn't actual accumulation, because each year (for 120 years) the added/invested surplus value was the same (absolute) figure, namely 33,3% of GDP in 1900, ie $1,3 trillion (world GDP being estimated at $4 trillion in 1900). Each year, total value would increase by the same $1,3 trillion (almost like in the movie Groundhog Day). The condition for this to hold, is that this figure is invested only in additional means of production (not in additional workers). Yet surplus value could still increase (and so accumulation in Luxemburg's sense take place), even without additional workers, though, namely through a reduction of the value of labor-power, that is a rise in the rate of exploiation, so a further condition for the scenario to hold (of no accumulation for 120 years), is that there be no technological improvement (in production of wage goods). Absent this latter condition, the value of variable capital would fall, surplus value rise correspondingly, and so there is accumulation (in Luxemburg's sense), but without there have being an increase of workers (I think this is also an important conceptual point in itself), or an increase of the figure of annual invested surplus value (and there would be no change in total value/GDP I think).

Also, to take more normal conditions: if we allow invested surplus value to also go into additional workers (eg increasing the workforce 5 fold, over 120 years), and if technological improvement occurs, reducing the value of labor-power (eg also 5 fold), then it is as if all surplus value (during 1900-2020) had been invested only in additional means of production, for the figure of the value of variable capital has remained the same (from what it was in 1900). By the way, one can even imagine the value of variable capital fell (if eg the workforce increased merely 3 fold, and value of labor-power fell 5 fold). The conceptual point here is, that the mere same figure (of the value of) variable capital in the scheme (in 1900 as in 2020) doesn't need to correspond with the same number of workers (body count or hours), so there can be accumulation in Luxemburg's sense (increase of surplus-value), even if the figure of variable capital shows no increase (or even fell), that is, it is as if surplus-value has not been invested in hiring additional variable capital.

Above I gave a concrete implication from Luxemburg's definition of accumulation, as the extension of surplus value, that, under certain conditions (eg invested surplus value goes only into means of production), even the situation where we have a 40-fold increase in GDP over the course of 120 years, would not count as accumulation, according to her definition (ie because there is no extension of surplus value).

Perhaps that apparently absurd situation I sketched, is even the scenario Luxemburg envsioned in this famous passage (ch. 25):

Who, then, realizes the constantly increasing surplus value? The schema replies: the capitalists themselves, and they alone. What do they do with their ever-increasing surplus value? The schema responds: they use it to constantly increase their production. These capitalists are thus zealots of the expansion of production for its own sake. They have ever more machines built in order to build yet more machines. The result of all this is not the accumulation of capital, but the increasing production of means of production without any purpose, and it is also characteristic of Tugan-Baranovsky’s audacity and penchant for paradoxes that he assumes that this never-ending merry-go-round in midair is a true theoretical reflection of capitalist reality and that it can actually be derived from Marx’s own theory.

If the capitalsist invest, each year, the same amount of surplus value into new means of production (+ a couple of other conditions I gave above)), there is no extension of surplus value, and so no accumulation. But there is indeed an extension of GDP (or total value), namely in the rising mass of the value of machines.

I think this is the real meaning, and the only way to make sense of her counter-position. Again:

The result of all this is not the accumulation of capital, but the increasing production of means of production without any purpose

The definition of accumulation she uses is the extension of surplus value. A mere increase in (the value and material mass of) means of production, would not extend surplus value.

So the question is, do we understand (as Luxemburg does) accumulation to be the ever-increasing amount of surplus value, or do we understand accumulation to be a mere (repetitive) addition of the same (determined) amount of surplus value?

Another point that Luxemburg could be making here (in ch. 25) when she wrote that Marx's "diagram does not permit of any other interpretation than that of production for production’s sake", seems so simple that I might have overlooked it (and perhaps even the ICC). Her point is I think that even when we accept the idea that "cap­italists can simply exchange their products among themselves" (to quote from an ICC text), this does not consitute accumulation, for there is no extension of surplus value. At the moment when the capitalists managed to re-invest all the surplus-value by purchasing new means of production (/raw materials) from each other, they have not invested any of their surplus value in the purchase of more of the commodity labor-power (the only commodity that produces value), simply because capitalists (in either department) don't produce the commodity labor-power. (We still assume the condition that the value of labor-power hasn't fallen).

The arguments (eg made by comrade Link) along the lines that only capitalists can buy means of production, etc. turn against Luxemburg's critics, for precisely when we accept that all the products (containing surplus value) have been bought by the capitalists themselves, ie when the market creates its own demand so to speak, the question arises more prominently, what money do the capitalists have left to buy more of the crucial commodity labor-power (if they already had to spent all their money to realise the surplus-value-containing products from each other)? And without that purchase of the commodity labor-power, there is no accumulation (defined by Luxemburg as extension of surplus value).

I can't find textual evidence in Luxemburg's book to settle the issue (I raised in my first post) on whether she believed, that none of the surplus value can be realised (as comrade Link and probably the majority of commentators assumed), or whether she merely believed, that not the entire surplus value can be realised. I think it's important to note this unclarity in the dispute, which might be due to Luxemburg herself. I have suggested that it is the latter. And, if my further reasoning has been correct so far, we can determine the part/percentage of surplus value that can be realised (and the part that can't be realised). For that is the next question.

The amount of the surplus value that gets re-invested in new means of production, by definition doesn't extend surplus-value (ie doesn't count as accumulation), but Luxemburg's theory is able to concede, that this amount of value so re-invested, can be realised (as it is has been realised once before). Luxemburg's theory is able to concede (the basic point of most of her critics, eg of Mandel in his intro to Capital Volume II, about the fact of the existence of multiple capitals, at different stages in their production cycle), that there is indeed a certain reserve of money available in society, that lets the capitalists realise each others' surplus value products, of a fixed size in one cycle of production (which can be repeated forever and cumulatively grows the total value, but only at the same size each cycle). But there is not enough money to also purchase more of the commodity labor-power. The money that is spend on purchasing additional labor-power is a deducation from the available money reserve, that was meant to realise the products containing surplus value.

So I have determined the part of surplus value which for Luxemburg cannot be realised (by capitalism itself): it is equal to the amount of value of the purchased additional labor-power.


Let me be so cocky as to propose why until now we (and even its defenders) haven't understood Luxemburg. I think it's because we involuntarily tend to look at the problem of accumulation from an individual level. From an individual capitalist perspective, investment in new machinery increases a firm's profit mass (the capitalist doesn't care about extension of his individual surplus value). But the invidual firm which invests (only) in new machinery, and as a result makes a greater profit, actually doesn't accumulate. There is mere redistribution of already existing surplus value, not extension of total surplus value. One must thus treat the issue of accumulation (extension of surplus value) on the social level, as an abstract theoretical problem like Luxemburg does. By the way, on the individual level there arises the illusion that (due to the increased profit) value is created thanks to machinery.

If we remain stuck on the individual level and don't treat it as an abstract issue, it becomes indeed counter-intuitive to our ordinary understanding, or perhaps totally incromphensible, for would why a capitalist need to seek to extend the individual amount of his surplus value to an ever-increasing size, ie hire more workers (ie why he would need to accumulate, to exist as a capitalist)?

On an old thread comrade Cleishbotham made the correct point, that competition is the effect of the LTRPF (not its result, as apparently Mattick thought, – I take the comrade's word about Mattick's views), and the LTRPF, in its turn, is a result of the increase of the composition of capital, which, finally, in its turn we often assume (and it's this assumption I like to focus on) can only be the result of productivity growth.

Marx (ch. 15, vol III) seems to imply, that an increase of the composition of capital can be the result also of something else (and so doesn't require the assumption of productivity growth/investment in new machinery):

When would over-production of capital be absolute? ... As soon as capital would, therefore, have grown in such a ratio to the labouring population that neither the absolute working-time supplied by this population, nor the relative surplus working-time, could be expanded any further (this last would not be feasible at any rate in the case when the demand for labour were so strong that there were a tendency for wages to rise); at a point, therefore, when the increased capital produced just as much, or even less, surplus-value than it did before its increase, there would be absolute over-production of capital; i.e., the increased capital C + ΔC would produce no more, or even less, profit than capital C before its expansion by ΔC. In both cases there would be a steep and sudden fall in the general rate of profit, but this time due to a change in the composition of capital not caused by the development of the productive forces, but rather by a rise in the money-value of the variable capital (because of increased wages) and the corresponding reduction in the proportion of surplus-labour to necessary labour.

^ If variable capital rises in terms of money (but not in terms of actual labour hours), and so surplus value stays the same, then indeed the general rate of profit (s/c+v) falls. Since v has apparently become greater, the CC (that is c/v, with v having grown, and c remaining the same) is altered, by which, at first sight, I was tempted to guess Marx means, CC has decreased. But surely we know that a fall in the profit rate is due to rather an increase of the CC. So I try to seek a more plausible meaning: as v has risen (in terms of money), perhaps Marx implies no additional actual v will be hired, or the number of wage workers even decreases, that is anyway, certainly v has not risen. One must actually assume that Marx means v has fallen, as this is the only thing here that can cause an increase of the CC (which in turn causes a fall in the general rate of profit). If v (actual labour power employed) has fallen, there is less creation of surplus value, As said, the rate of profit still will decline, and this will cause competition to break out. What counts with regard to the concept of accumulation is, that the mass of surplus value has not merely not been extended, but even declined.

My point is, that all of this happened without the assumption of productivity growth (ie without a change in the value of c)!

The comrade had stressed the point, that competition is not the cause, but the effect of the LTRPF. I suggest the reason apparently (I take the comrade's word on it) people like Mattick believed the LTRPF to be a mere result of competition, was their assumption that the LTRPF can be nothing else than the result of productivity growth (which is indeed the result of competition). As I here tried to show, for Marx the LTRPF can (at least also) occur as a result of a rise in wages (ie without productivity growth, and so without competition).

Marx's reference to the increase of wages, which I take as a decrease of variable capital (in terms of hours), to define what a cessation of accumulation means, might seem like either a simple tautology, or a totally problematic claim. It might not seem historically realistic. And there lurks the assumption of a rise in the value of labor-power (whereas one would think it would decline via the path of extending relative surplus value).

I did the above exegesis, just to address the possible objection, that the assumptions of Luxemburg (and of myself, for the purpose of understanding her theory) are too unrealistic:

- They (Marx and Luxemburg) share the assumption of no productivity growth.

- They share the assumption of no competition.

- Luxemburg (or my interpretation of her) assumes the condition that the value of the individual commodity labour-power which is hired in additional fashion (by the investment of surplus value), remains constant, while Marx's assumption was similar if not even more unrealistic, namely it rises. (or its rise to an extent greater than what could be compensated for by the new value created by additional workers – this is just my intuitive attempt to derive a determination from Marx's line on a "reduction in the proportion of surplus-labour to necessary labour").

(to clarify the last sentence in my previous post:

As the LTRPF is a result of an increase of the CC, and there is no competition (no systemic inner capitalist logic), a capitalist, from the individual perspective, has no necessity to buy additional labour-power. His extension of surplus value would be just from personal greed or fortuitous. I know (unfortunately) no way yet other than to view the issue of accumulation, ie the extension of surplus value, from the social level, as an abstract issue. )

Correction to my first sentence:

On an old thread comrade Cleishbotham made the correct point, that competition is the effect of the LTRPF (not its CAUSE, as apparently Mattick thought, – I take the comrade's word about Mattick's views),

This thread:

Mattick not only at one point in Marx and Keynes clearly states that competition is what causes the falling rate of profit (we know it because we originally [c. 1973] believed him but will dust down the old notes if you need a precise reference)

I stated in my last couple of posts, that the issue of accumulation, ie the extension of surplus value, from the social level, is a difficult because abstract issue. I was begging the question, what is the necessity of accumulation (in that technical sense of extension of surplus value)? But now perhaps I think I'm missing a deeper critical point, which would be that there is no (impersonal-historical system-level) necessity of accumulation. Wouldn't that be an ultimate point of Marx, to show that there is not any necessity of accumulation?