The Accumulation of Contradictions or The Economic Consequences of Rosa Luxemburg

Originally published in RP6 (First Series, 1976). Re-published as a "web special" for RP43 (Third Series, 2007).

The revolutionary basis of Marxism rests on its ability to prove that capitalism is not only a crisis-ridden system, but that as a mode of production it also faces objective limits to its own self-expansion. On this rests the view that communism is not only desirable (the position of utopians, past and present) but an absolute necessity. In the era of decadent capitalism, in this century, this necessity has become increasingly urgent, as the method by which capitalism today resolves its crises through the physical destruction of value (i.e. through war), threatens to extinguish humanity itself.

One writer who had the merit of realising the need for an objective understanding of the limitations of capitalism was Rosa Luxemburg. Her whole life was devoted to a struggle against utopian and reformist elements in the German Social Democratic Party and she was among the first, if not the first, to realize that the era of national liberation struggles had definitely closed. In a polemic of 1896 she concluded that Polish capitalism was inextricably linked by "chains of gold" to Russian capital and that therefore Polish independence would not lead to a further development of capitalism. Thus revolutionaries had no further interest in supporting it, as Marx and Engels had done (and indeed as Engels still did in the 1892 introduction to the Polish edition of the Communist Manifesto), nor had they any interest in supporting "national liberation" in any state in the era of imperialism, as any new state would only be at the mercy of one of the imperialist powers. Because these states could not develop independently, capitalism had now ceased to be progressive and proletarian revolution was the only aim of all revolutionaries. It was Luxemburg, rather than Lenin, who represented revolutionary clarity on the national question during the post-war revolutionary wave. She condemned the nationalities policy of the Bolsheviks which allowed the break-up of the old Russian empire on the basis of "national self-determination" as "nothing but hollow petty bourgeois phraseology and humbug" (The Russian Revolution p.49) because it delivered Finland and the Ukraine over to German imperialism (which helped the weak local bourgeoisies to murder their respective proletariats).

This revolutionary clarity on the national question (1) cannot be attributed to the economic theories which Luxemburg propounded in the latter part of her life. Guy Sabatier (2) in an article in Spartakus No.64 (November-December 1975) tries to tell us however, that the economic basis for Luxemburg's views are to be found in her major economic work, The Accumulation of Capital (1913). But even his own account of Luxemburg's political evolution shows that her understanding of the national question pre-dates by nearly a quarter of a century the appearance of The Accumulation of Capital. Nor can Sabatier argue that Luxemburg held those views on economics before that date but merely had not got around to publishing them. In Social Reform or Revolution, written in 1899, she shows that she originally held a Marxist position on the understanding of the cause of capitalist crises.

It is the threat of the constant fall in the rate of profit, resulting not from the contradiction between production and exchange, but from the growth of the productivity of labour itself ... (which) has the extremely dangerous tendency of rendering impossible any enterprise for small and middle-sized capitals. It thus limits the new formation and therefore extension of placements of capital. (3)

Quoted in Looker, Selected Political Writings of Rosa Luxemburg

Luxemburg's theory that capitalist collapse was caused, not by the rising organic composition of capital, but by the saturation of markets emerged as a response to the situation in German Social Democracy at the turn of the century. Marx had always argued that the central law of capitalist development was the falling rate of profit.

This is in every respect the most important law of modern political economy, and the most essential for understanding the most difficult relations. It is the most important law from the scientific standpoint ... hence it is evident that the material productive power already present, already worked out, existing in the form of fixed capital... that the productive forces brought about by the historical development of capitalism itself, when it reaches a certain point, suspend the self-realisation of capital, instead of positing it ...

Grundrisse p.748-9

But by use of empirical arguments Bernstein and the other revisionists pointed out that capitalism did not appear closer to catastrophe but on the contrary, because of the success of the trade unions, was even passing on some of the benefits of its "success" to the workers. (4) The logic of this was continued reformism with the assumption that socialism would arrive through the gradual socialisation of the capitalist economy and political power would be obtained via the ballot box. From the very beginning Luxemburg was the most bitter (and coherent) opponent of the revisionists. Against their Kantian idealism she retorted:

If one admits, with Bernstein, that capitalist development does not move in the direction of its own ruin, then socialism ceases to be objectively necessary.

Social Reform or Revolution, op.cit., p.60

So far so good. Luxemburg fought valiantly for the revolutionary core of Marxism at this juncture. However for whatever reason, whether it was the empiricist arguments of the revisionists, or the work she herself was doing in the Party school of the SPD on economics, Luxemburg abandoned the Marxist explanation (5) (though she never admitted this) of the capitalist breakdown theory. This did not mean that she abandoned the breakdown theory. On the contrary, she looked for a better, more powerful theory than that of Marx and which would better match the reality of her time more clearly. This theory was finally revealed in The Accumulation of Capital. Here she positively abandoned value theory by asserting that the fall in the rate of profit could not be the cause of the capitalist crisis.

The increasing productivity of labour ensures that the means of production grow faster in bulk than in value, in other words means of production become cheaper ... This phenomenon amongst others, checks the actual decline in the rate of profit and modifies it to a mere tendency, though our example shows that the decline of the rate of profit would not only be retarded but rather completely arrested.

The Accumulation of Capital pp.337-8

And again, in the Anti-Critique (in reply to the critics of The Accumulation of Capital) she equates the validity of Marxist economics with the extinction of natural life.

Or else we are left with the somewhat oblique comfort provided by a little "expert" from the Dresdener Volkzeitung who, after thoroughly destroying my book, explains that capitalism will eventually collapse because of the falling rate of profit ... However, what may be this comfort is unfortunately dispelled by a single sentence by Marx, namely the statement that "large capitals will compensate for the fall in the rate of profit by mass production." Thus there is still some time to pass before capitalism collapses because of the falling rate of profit, roughly until the sun burns out.

ibid pp.76-77

If Luxemburg's economic understanding had been as good as her polemical talent she would never have written the above; nor, for that matter could she have written her book in the first place. It is more a commentary on the victory of revisionism in German Social Democracy that only a "little expert" stands up for Marxism but it shows that Luxemburg and most of her critics were agreed on one thing - Marxist economics were finished.

To begin with Marx cited many ways in which the fall in the rate of profit could be counter-acted, including the increased intensity of exploitation, depression of wages below the value of labour power, relative over-population and foreign trade (see Capital Vol. III Chapter XIV), but he is quite clear that they do not halt the fall in the rate of profit. On the contrary, they help to explain why "this all is not greater and more rapid" (Vol.III, p.232). Second, increased productivity (not cited by Marx as a counter-acting influence) is also a further condition of the fall in the rate of profit.

The rate of profit does not fall because labour becomes less productive, but because it becomes more productive. Both the rise in the rate of surplus value and the fall in the rate of profit are but specific forms through which the growing productivity of labour are expressed under capitalism (Capital Vol. III p.240). Luxemburg's basic error here is only part of her greater failure to fully understand the significance of the rising organic composition, and by implication, value theory itself. However, this can be best illustrated by turning to her actual theory.

Luxemburg's basic argument in The Accumulation of Capital was, that in a world composed only of capitalists and wage workers (the world posited by Marx in Capital), the finite limits to the consumption of the capitalists (physically) and the proletariat (due to its restricted purchasing power) would clash with the unlimited capacity of capitalism to expand production. Capitalist expansion had hitherto been possible only because huge areas of the world had been given over to pre-capitalist modes of production which had served as markets to realise that part of the surplus value destined for further capitalisation. According to Luxemburg, when all these pre-capitalist areas have been integrated into capitalism so that the market for capital expansion is saturated, overproduction is thus the order of the day. The permanent capitalist crisis has arrived. Since she considered this situation had been largely brought about by 1913 she was thus able to offer a logically deducible explanation of imperialism - the struggle for markets between the Great Powers.

Marx and the problem of expanded reproduction

Taking Luxemburg's argument in its various stages, she begins by pointing to what she sees as Marx's failure to solve the "problem" of reproduction. In Volume II of Capital Marx sought to show that expanded reproduction (i.e. accumulation) of capital was possible in a world composed solely of capitalists and workers. In Volume I he laid out this framework;

If we are to examine our topic in its integrity, without disturbances by subsidiary circumstances, we must, for the once, treat the whole world as one nation, must assume that capitalist production is everywhere established, and has got possession of all branches of industry.

Capital Vol. I p.638, quoted in The Accumulation of Capital p.136 and 331 in a slightly different translation

and in Volume II he turned to the process of circulation of commodities within this capitalist world. For the sake of clarity we will briefly outline the arguments of both volumes here.

Marx did not discover the law of value out of thin air but rather built on the work of earlier writers like Smith and Ricardo, who by virtue of their commitment to the continued existence of capitalism, were unable to follow the logic of their own analysis to its ultimate conclusion.

The rise of a new wealthy, dynamic society in (principally) 18th Century Britain led to an enquiry conducted by the educated representatives of the new class into the origin of wealth in general. They not only wanted to know how wealth came to be created but why a certain quantity of one commodity cost more or less than the same quantity of another commodity. Adam Smith was the first to realise that (for example) the mere possession of underground coal was not the possession of wealth until someone had brought that coal to a point where it could be used. Similarly, the possession of the most fertile field was worth nothing until someone had worked on it. Labour was thus seen to be the source of all value. The difference in value between a ton of coal and a ton of bread is the difference between the amount of labour time which it takes on average to produce the two different commodities. This in its simplest form, is the basis of the labour theory of value.

Like coal or bread, labour power is itself a commodity. It has to be sold by the propertyless (i.e. the proletariat) to the owners of the means of production (such. as factories, farms, mines, etc.).

However, according to the labour theory of value, the value of a commodity is fixed by the amount of labour that is, on average necessary for its production; so what is the value of labour itself? The answer is the same as for all other commodities. It is fixed by the amount of time the worker has to work in order to provide himself with the necessities of life (see Capital Vol. I pp.230 and 563).

This is the key to understanding the meaning of exploitation. Under capitalism the capitalist is thus able to buy the worker's labour power at its value but the worker produces far more commodities than the equivalent of the value of his labour. He thus produces surplus labour, which, in the shape of the commodities created by this labour, means surplus value for the capitalist. This surplus value is the source of the capitalist's profit. The law of value poses its own problems for the capitalist system when we look at accumulation. Taking the whole world as a single capitalist, each cycle of accumulation has to involve the same elements. These are: machines and raw materials (Marx called constant capital or c), labourers, to whom the capitalist pays the price of their labour power in the form of wages (Marx called this variable capital or v); and the final outcome of the process is a product incorporating a value greater than the sum of its original components (because labour has been expended). This is the surplus value, or s. Thus our total product contains elements of all three and its formula is therefore: c + v + s.

Obviously, from the point of view of the capitalist system as a whole, the capitalists should try to keep v as high as possible in relation to c because men can create new value but machines cannot. But in the real world of capitalist competition this makes less sense to the individual capitalist who is only interested in making a greater profit than his rivals. To do this he must be constantly on the look-out for labour-saving devices, enabling him to produce more commodities at less cost and thus increasing the productivity of labour. In value terms, he must use most of his surplus to expand his constant capital (c). Thus we find that the organic composition of capital (the ratio c / v) rises. But as labour is the source of all value the reduction in v (in relation to c) leads to a reduction in s in relation to c + v, thus s / (c + v) (the rate of profit) has a logical tendency to fall. This is the central contradiction of capitalism. It must expand or die, each new expansion makes death more certain.

However, this has already taken us beyond Volume I and into III of Capital. In Volume II, the volume concerned with the "circulation of commodities", Marx divided total production (c + v + s) into its two branches: Department I (production of means of production) and Department II (production of means of consumption). Obviously this distinction can be blurred (e.g. bricks for factories are in Department I whilst bricks for houses are in Department II), but a further, and more fundamental relation exists in terms of the expansion of production. The total annual product of society can be summarised in diagrammatic form in the following fashion.

Department Product
Dept. I Ic + Iv + Is
Dept. II IIc + IIv + IIs
Total social product

If we examine the relationship between the two departments it is clear that in order for the cycle of production to begin anew there must be some exchange of commodities between the two departments. The machinery or machine tools produced in Ic can be redistributed in the same department, but workers cannot be paid with the means of production represented by Iv + Is. Similarly, the constant capital in Department II (IIc) cannot remain in the form of consumer goods if it is to set in motion a new cycle of reproduction in Department II. On the other hand, the variable capital and surplus value (IIv + IIs), can be consumed by the workers and capitalists in that Department. From this it can be seen that the key to reproduction on this level is the exchange of the constant capital of Department II (in the form of consumer goods) for the variable capital and surplus value of Department I (in the form of machine tools). In other words the condition of equilibrium under conditions of simple reproduction (i.e. where there is a replacement of the total social product but not an extension) is that: Iv + Is = IIc But this is simply a heuristic device used by Marx to explain the workings of the capitalist economy. In real life all capital accumulation takes place under conditions of expanded reproduction (i.e. where there is an increase in the total social product).

Nevertheless, as Marx maintained, the argument still holds good so long as the correct proportionality between the increased production of both departments is maintained.

In production on the basis of increasing capital, I (v+ s) must be equal to IIc plus that portion of the surplus product which is re-incorporated as capital, plus the additional portion of constant capital required for the expansion of the production in II; and the minimum of this expansion is that without which real accumulation, i.e. real expansion of production in I itself, is unfeasible.

Capital Vol. II p.521

Here Luxemburg takes issue with Marx. Marx was trying, at this point, to simply show how accumulation was possible. He did not say that there would not be crises caused by a temporary disproportionality between departments (e.g. such as in the development of new lines of production or agricultural crises such as harvest failures) but he did show that the central contradiction of the capitalist mode of production, its historical contradiction, could not be found in its process of circulation. Luxemburg could not agree to this because she did not accept that part of Marx's analysis where he said the contradictions of capital did occur in the production process. Because she could not accept the tendency for the rate of profit to fall as the motive for capitalist crisis she had to go on to found a new breakdown theory.

There can be no doubt that under capitalist conditions Department II is dependent on Department I in so far as accumulation is determined by the additional means of production. Conversely the accumulation in Department I depends on a corresponding quantity of additional consumer goods being available for its additional labour power. It does not follow, however, that so long as both these conditions are observed, accumulation in both departments is bound, as Marx's diagrams make it appear, to go on automatically year after year. The conditions of accumulation we have enumerated are no more than those without which there can be no accumulation.

The Accumulation of Capital p.131

Needless to say, Luxemburg went on to try to undermine these diagrams - an attempt, which had it succeeded would simply have proved that accumulation was theoretically impossible. She proceeded to take several steps away from Marxism by asking: for whom does expanded reproduction take place?

Who, then realises the permanently increasing surplus value? The diagram answers; the capitalists themselves and they alone - and what do they do with this increasing surplus value? The diagram replies: they use it for ever greater expansion of their production. These capitalists are thus fanatical supporters of an expansion of production for production's sake.

ibid. pp.334-5

Why this should be such a mystery to a self-proclaimed Marxist we cannot see. First, Marx (and not just Tugan-Baranowsky as Luxemburg would have us believe) did actually agree with the above statement which Luxemburg made sarcastically.

The industrial capitalist ... as personified capital he produces for the sake of production; he wants to accumulate wealth for the sake of the accumulation of wealth.

Theories of Surplus Value Volume I p.282

But this is not all. Luxemburg herself was well aware of the "goad" (Marx) of competition in driving on the capitalists to further accumulation.

Capitalist methods of production do more than awaken in the capitalist this thirst for surplus value whereby he is impelled to ceaseless expansion of reproduction. Expansion becomes in truth a coercive law ... thus as soon as a few capitalist enterprises have been enlarged, competition itself forces all others to expand likewise.

The Accumulation of Capital p.40

But this insight into competition by Luxemburg seems like an afterthought rather than an essential element of the nature of capitalist relations of production. After all she has already discounted the law which is the motor force of capitalist competition. What forces capitalists to constantly revolutionise the means of production when each development further strangles the source of capital renewal?

...a fall in the rate of profit calls forth a competitive struggle among capitalists and not vice-versa.

Capital Vol. III p.256

Thus in chapter twenty-five of The Accumulation of Capital she adopts such an aggregative approach in her discussion of the relations between Department I and Department II, and so overlooks the factor of competition, that she makes some totally incorrect assumptions. She assumes that surplus value cannot be realised by capitalists within the same department. But in fact there is no reason why this cannot happen. For example, a capitalist producing food realises surplus value if a proletarian from a textile factory buys his products. Further, her denial of the movement of capital between departments and her failure to understand the nature of capitalist competition is a total denial of Marxism. For if, as Marx argued, the rate of profit tends to be equalised throughout the capitalist economy by competition, how can this occur in a system where there is neither competition nor the movement of capital between departments? (6)

The method by which Marxist economic enquiry is carried out is by making valid abstractions from capitalist reality, but Luxemburg's abstraction is invalid as it fails to take into account the essential nature of the world she is supposed to be describing. This is because Luxemburg's method is not Marxist. She gives notice of this when she says that "desire" to accumulate is not enough, there also has to be an increase in "effective demand" (see page 131 of The Accumulation of Capital) for commodities. "Desire" is a strange enough piece of subjectivism for a Marxist but the search for "effective demand" puts Luxemburg in a new camp - that of the Keynesians.(7)

The flaw in Marx's analysis is, in our opinion, the misguided formulation as a mere question as to "the sources of money", whereas the real issue is the effective demand, the use made of goods, not the source of the money which is paid for them.

The Accumulation of Capital p.155

Luxemburg here is misrepresenting Marx's views. He is not interested in where the money comes from in order to realise surplus value but where the money comes from to exchange the total of commodity values. There is little excuse for Luxemburg here for Marx specifically states his aim.

The only assumption essential here, that in general there is money enough for the exchange of the various elements of the mass of the annual reproduction, is not affected in any way by the fact that a portion of the commodity-value consists of surplus-value ... The question is therefore only: where does the money come from to make possible the exchange of this total of commodity values? It is not at all: where does the money come from in order to turn the surplus value into money?

Capital Vol. II pp.477-8

The preceding quotation of Luxemburg's raises a number of questions (apart from why Luxemburg misrepresented Marx) concerning her treatment of the money question. Marx, as we have shown, was merely asking where does the money material come from, therefore his answer is simple enough. It comes from the capitalists of the gold industry. Luxemburg treats this with scorn, because she herself has misunderstood the question as asking where the effective demand comes from. We will deal with Luxemburg's preoccupation in a moment but first let us examine more closely her views on money.

She quotes with glee Marx's words in Capital (Volume II p.494) that "money itself is not an element of real reproduction" but she fails to understand the significance of the remark. Money in itself is not an element of reproduction because it neither creates new value nor is it used up in the production process. Money is merely a particular form which commodities take in the process of circulation. "The common whore of mankind" (Marx quoting Shakespeare in Capital Volume I p.113), it is "the vehicle not only of its actual value but also of its ten-fold value, because it has turned over ten times a day, realised ten different commodity values." (Volume II p.494).

Luxemburg seems to think that for every increase in the commodity value there must be an equal increase in the money value. Yet Marx is quite specific that at any given moment not all commodities produced will be transformed into the money form of capital. Thus a limited amount of money (a tenth, in Marx's example above) can serve to circulate commodities far in excess of its value by the same money passing through many hands and exchanging itself for numerous commodities. It is the speed of circulation of a certain mass of commodities which determines the mass of money required.

The velocity of circulation, hence the number of repetitions of the same function as means of purchase and means of payment by the same pieces of money in a given term, the mass of simultaneous purchases and sales, or payments, the sum of the prices of the circulating commodities, and finally the balance of payments to be settled in the same period, determine in either case the mass of circulating money, of currency.

Capital Vol. III p.445

But Luxemburg erroneously attributes to Marx the view that the "entire social surplus value" (see The Accumulation of Capital p.302) would have to be equalled by the production of the gold diggers so she is able to present us with an absurd picture of a massive heap of gold being produced to equate itself to the surplus value in the heap of commodities produced by the other branches of production. Luxemburg's error is obvious. As Bukharin concluded:

The entire amount of surplus value which is repeatedly produced must never be identified with the newly increased sums of money, since the process of realisation has no need of such a sum; equally the accumulation of capital must never be confused with the accumulation of money capital. (8)

Imperialism and the Accumulation of Capital p.185

Having exposed Luxemburg's inadequacies in this respect let us return to her Keynesian question of where the effective demand lies. Luxemburg's solution to this non-question is as bizarre as anything written claiming the title of Marxism. She says that the increased consumption cannot be from the capitalists because accumulation proceeds because of their abstention from consuming the vast majority of the surplus value produced; and the increase cannot be provided by the workers, because,

From the capitalist point of view it is plainly absurd to produce more consumer goods merely in order to maintain more workers.

op.cit. p.32

Actually it is not absurd for the capitalists to increase production out of desire for personal gain. Though it is not the purpose of accumulation, and though their consumption decreases relative to the increase in capital values, this consumption increases absolutely. Equally it is not absurd for the capitalist to expand variable capital (i.e. the working class) because accumulation could not proceed without it (unless all realisable surplus value was turned into constant capital - plainly an absurdity in itself). In reality capitalists man the additional capital by employing additional workers who also produce an additional demand.

But to follow Luxemburg's logic to the end, she finally reveals to us who the "real" buyers are. She tells us that both a realisation of the surplus value and accumulation are impossible in a purely capitalist society and that we must look outside the capitalist relation to pre-capitalist societies or pre-capitalist enclaves (like the peasantry). How these pre-capitalist areas can realise in money terms the surplus value produced by the capitalist Luxemburg circumspectly manages to avoid telling us. Her book is full of truisms concerning how these pre-capitalist areas first have a commodity economy forced upon them (that is if they did not have one already) and then a capitalist commodity relationship. But at the point when they play a role in the capitalist world market they themselves have become capitalist. It is not pre-capitalist tribesmen with cowrie shells or a subsistence peasantry that can solve the question of where does the money come from to realise surplus value; for what money do they have? (9) The commodity is still in the form of goods and so the C-M metamorphosis is not complete. Luxemburg's solution to her own question is thus no solution at all.

Underconsumption and the law of value

Hitherto we have only hinted at the fundamental error in Luxemburg's method as we have been attempting to criticise her argument. The remaining four sections of this article will be an attempt to spell out the consequences of non-adherence to the law of value.

To begin with Luxemburg regarded herself as putting forward a reasonably orthodox Marxist view (despite her obvious failure to accept the premises of Volume II of Capital). After all Marx had made statements which, when taken in isolation, would appear to reveal him as an underconsumptionist. For example, in the Communist Manifesto he talks of a crisis of a new type.

In these crises there breaks out an epidemic that in all earlier epochs would have seemed an absurdity - the epidemic of over-production.

page 48

And again in Volume III of Capital,

The last cause of all real crises always remains the poverty and restricted consumption of the masses as compared to the tendency of capitalist production to develop the productive forces in such a way, that only the absolute power of consumption of the entire society would be their limit.

But these remarks (and others in similar vein) have to be placed in their true context of the capitalist relation of production.

There are not too many necessities of life produced, in proportion to the existing population. Quite the reverse. Too little is produced to decently and humanely satisfy the wants of the great mass.

Volume III p.257

Capitalism is not interested in necessaries of life per se, but in the maximisation of profit.

On the other hand, too many means of labour and necessities of life are produced at times to permit of their serving as means for the exploitation of labourers at a certain rate of profit.

ibid. p.258

Profit, the rate of profit and hence the law of value thus assert the restrictions of capitalist development. There are not too few labourers or too many productive forces in the world but there are too few productive labourers to create the necessary mass of profit needed to offset the decline in the rate of profit; and there are too many productive forces to be able to sell commodities profitably. The rise in the organic composition, the diminution of v in relation to c is the clearest expression of this relation. As Marx said when discussing the creation of the world market:

The stupendous productivity developing under the capitalist mode of production relative to population, and the increase, if not in the same proportion, of capital values (not just their material substance), contradict the basis, which constantly narrows in relation to the expanding wealth ... hence the crises [our emphasis].

ibid p.266

Clearly this has little to do with Luxemburg' s theory as she was looking outside the capital-wage labour relationship, beyond the realms where the law of value rules supreme, in order to find her saturated markets, her failure of the consumer.

To look at this empirically; Luxemburg clearly regarded the world market as saturated in 1913 because those pre-capitalist areas which were left were too small for world capital. This was in fact the basis for her explanation of imperialism.

Its ultimate aim, that is to say, is to establish the exclusive and universal domination of capitalist production in all countries and for all branches of industry ... For capital the standstill of accumulation means that the development of the productive forces is arrested and the collapse of capitalism follows inevitably, as an objective historical necessity. This is the reason for the contradictory behaviour of capitalism in the final stage of its historical career: imperialism.

op.cit. p.417

But in 1913 the whole of India, China and huge areas of the rest of the globe were still largely pre-capitalist (i.e. in Luxemburg's terms they were outside the wage-labour - capital relationship though we would say that they are formally dominated by capital) and even today over half the world's population is not contained within the wage-labour relation. How then can Luxemburg's theory explain the growth of capitalism after 1918 or after 1945? Only by saying that markets were not saturated in 1914, and therefore that capitalism is not yet decadent. Pannekoek, when he became a Luxemburgist took the view to its logical conclusion by arguing in the 1940s that the revolution was, as yet, still some time away because there were still too many areas outside capitalism.

The hundreds of millions crowding the fertile plains of China and India once drawn within the confines of capitalism, its chief work is accomplished ... Then its further expansion is checked ... then the mutual fight of capitalists for world domination becomes fiercer, with new world wars impending.

Workers' Councils p.93

"Saturated markets" is thus a rather unconvincing explanation for the reality of capitalist decadence. With this view markets appear saturated (as in 1914) but then they are not (as in 1945). This is hardly a materialist analysis and it is clear that whilst a theory which is as concrete to grasp as Luxemburg's could be useful polemically (if it was not so contradictory), it obviously cannot explain the real movement of capital. A value analysis can explain why accumulation can be arrested, and why, after a crisis, it can then proceed. Each crisis leads to a devaluation of constant capital, thus raising the rate of profit and allowing the cycle of reconstruction - boom - slump - war to be repeated once again. (10)


Luxemburg's theory of imperialism will be dealt with more fully in a text to be published later this year on imperialism in general. Here we will confine ourselves to a brief summary of the main points.

First, her rationale for imperialism based on "saturated markets" is extremely weak and inadequate. If, as Luxemburg admitted (see, for example p.358 of The Accumulation of Capital) the capitalist metropoles still contained pre-capitalist enclaves (e.g. serfs, peasants) why does capitalism have to expand overseas and away from the capitalist metropoles from the very beginning of its existence? Why doesn't it first bring all the areas closest at hand within the capital - wage-labour relationship if it merely seeks for new markets? The explanation is to be found, not in the need for new markets but in the search for essential raw materials and the maximisation of profit. Second, Luxemburg's theory implies that imperialism is a permanent characteristic of capitalism. As capitalism, for Luxemburg, has always sought to extend the market in order to accumulate, her theory cannot distinguish between the original expansion of trade and money economies at the dawn of capitalism in Europe and its later imperialist expansion. In her view Portugal and Castile must have been the most advanced capitalist powers in 1620. Mercantile capital was necessary for the original accumulation of capital but this is a qualitatively different phenomenon from the capitalist drive to accumulate once it is established as the dominant mode of production. But for Luxemburg,

Trade capitalism and mercantilism, industrial capitalism and liberalism, finance capital and imperialism - all these phases of capitalist development disappear or dissolve into capitalism as such.

Bukharin op.cit. p.253

Third, on an empirical level, whilst it is difficult to see how colonial plunder of the 17th and 18th centuries and the increasing volume of trade between developing European capital and the colonies (largely luxury goods or slave traffic) can be the result of overproduction, the evidence for "saturated markets" as the rationale for late 19th century imperialism is equally weak. Luxemburg claimed that "internal capitalist trade can at best realise only certain quantities of value contained in the social product" (op.cit. p.366) yet the following table clearly shows that at the time she was writing the vast majority of capitalist trade was with other capitalist countries.

- To Europe Aus/NZ/Canada USA
Britain 35 18 n.a.
USA 70 15 n.a.
France 60 n.a. n.a.
Germany 45 n.a. n.a.
Japan 20 n.a. 30
Table of exports 1900-10s (%) - Source: Barratt-Brown The Economics of Imperialism pp.168 & 190

Further, in the chapters in which Luxemburg seeks (we presume) to validate her theory by historical examples (see Chapters 27-30) she gives us plenty of moral outrage but no proof of the value of the areas mentioned as markets. Indeed, on pages 412-3 imperialism in South Africa is after "labour power", "diamonds", and "gold".

Finally, Luxemburg's theory of imperialism falls down on the money question. Having apparently realised in Chapter 30 that pre-capitalist tribesmen and peasants do not have any money to realise the surplus value contained in the commodities looking for realisation, she decides to let the capitalist first give it to them in the form of international loans (see The Accumulation of Capital pp.427-9). She says that these loans are repaid either by taxation or, as in the case of the Suez Canal, by forced labour. But this is no solution. First, where do pre-capitalist elements get money to pay taxes and second, how does forced labour create a market? Nor are we told of any previous sale of commodity values by the pre-capitalist elements to the industrial capitalist. According to Luxemburg therefore, capitalism is buying its own surplus value with its own money, or more pointedly, it is simply giving away commodities to extend an ideal mode of production. If capitalism was as generous as this view makes out it would seem to be a mystery why it has not succeeded in developing the world through international loans.

Only the law of value can explain why capitalism is today in its death throes and yet it has failed to develop the vast majority of the globe. Why aren't these areas furnishing a pre-capitalist market? The answer to this lies in the arguments we have posed above. Capitalism is no longer interested in developing such areas because to do so would be unprofitable.

Capital stagnation cannot have physical causes, for the the existing material forces of production and labour power are not altered by the crisis. Nor can it find its cause in a material overproduction by the means of production, for in this respect the world is obviously undercapitalised; not enough means of production exist to satisfy even the minimal needs of the world's population. The turn from prosperity to depression can only be explained as a shift from a sufficient to an insufficient profitability of capital.

Mattick, Marx and Keynes p.70

But Luxemburg was not interested in the shift in value relations; not only did she make the bulk of Marx's work appear irrelevant but she also made it quite clear that empiricism was enough for her.

Whatever the theoretical aspects the accumulation of capital as an historical process depends in every respect upon non-capitalist strata and forms of organization [our emphasis].

op.cit. p.366

Having already abandoned the law of value it now seems as though Luxemburg wants to abandon theory as well. It therefore comes as no surprise to learn that her inspiration was derived less from Marx than from the early nineteenth century economist Sismondi. Lenin's description of Sismondi's theory will have a familiar ring to Luxemburgists.

Sismondi asserted that as a result of the development of large-scale enterprise and wage labour in industry and agriculture, production inevitably outruns consumption and is faced with the insoluble task of finding consumers; that it cannot find consumers within the country because it converts the bulk of the population into day labourers, plain workers and creates unemployment, while the search for a foreign market becomes increasingly difficult owing to the entry of new capitalist countries into the world arena.

A Characterisation of Economic Romanticism p.7

Whilst she described him as having "the interests of the proletariat at heart", Lenin and Marx (as Luxemburg knew) were unanimous in denouncing him as a "reactionary and utopian" (Communist Manifesto p.81).

State capitalism

The CWO has written on a number of occasions on the tendency towards statification which is the tendency of global capitalism in its decline as a social system, and which must be located within the trajectory of capitalist development itself (11). We therefore do not intend to repeat all the points made in these articles. Rather we will look at the compatibility of both Marx's theory and Luxemburg's theory to deal with a reality which neither of them lived to see.

The Marxist explanation was put forward in Revolutionary Perspectives 1 but stating the case briefly, we see that an active state intervention in the economy is needed when the lack of surplus value becomes chronic; the mechanisms for the equalisation of the rate of profit throughout the capitalist economy go into disarray and the industries with the highest organic composition of capital (coal, steel, etc.) which are vital to the existence of the economy, have a tendency to collapse, putting the whole capitalist economy in peril. In fact one could go as far as to say that the non-functioning of the mechanism for the equalisation of the rate of profit is the main feature of decadent capitalism:

Totally incorrect too is the assertion that under our present economic system ... the law of value regulates the "proportions" of labour distributed among the various branches of production.
If this were true it would be incomprehensible why our light industries, which are the most profitable, are not being developed to the utmost, and why preference is given to our heavy industries which are often less profitable, and sometimes altogether unprofitable.

Economic Problems of Socialism in the USSR J.V.Stalin p.27

These economies, whether in the fully statified form of the East or the Keynesian form of the West, are an attempt to solve the problems created by the law of value within value production itself. In no way is the commodity character of labour power altered and hence exploitation takes exactly the same form as it always has done under capitalism.

Such a clear and cogent explanation of state capitalism is not possible on the basis of Luxemburg's theory, and it is not surprising that attempts by Luxemburg's heirs today to argue her case have been either feeble or non-existent. We will look at these briefly below but first we will have another look at Luxemburg's views.

The central viewpoint that Luxemburg clings to is her view that all commodities must pass via the market before all the surplus value contained in them can be realised and accumulated. To quote an example,

This surplus value can be of no use to the capitalist so long as it remains hidden in the commodity form of the product. Once that commodity has been produced, it must be realised, it must be converted into a form of pure value, that is into money.

op.cit. p.38

Once again Luxemburg reveals her confusion on the nature of money. It is not "pure value" but a special case of the commodity form, the universal commodity which effects the exchange between all others; gold, and not super-tankers operate as money for reasons of convenience, not because it is pure value.

After this pure nonsense, Luxemburg's theory cannot cope with a system where the state functions as the "collective capitalist", and where exchange can take place via equivalent values and without each commodity passing through the C-M cycle or, as Marx would have said, presupposing the "ideal" existence of money. Thus in Russia, for example, though the product of Department II passes through the C-M cycle, the bulk of those of Department I do not, and the state delivers means of production in exchange for equivalent commodity output. If we deny that these are commodities, or that they contain surplus value realised and capitalised without assuming the money form, we are denying the commodity character of the bulk of production in Russia et. al. and thus are accepting the Stalinist view that state capitalism equals socialism.

However, our Luxemburgists do not base their understanding of the capitalist nature of Russia on any internal dynamic like the law of value, but on Luxemburg's view that:

Internal capitalist trade can at best realise only certain quantities of value contained in the social product... That part of the surplus value which is earmarked for capitalisation must be realised elsewhere.

Luxemburg op.cit. pp.366-7

Thus, the U.S. publication of the International Communist Current (ICC) claims that:

... the impossibility of realising the capitalisable portion of surplus value within ones own borders compels the state capitalist society to expand beyond its frontiers and seek outlets on the world market.

Internationalism 2 p.38

In the first place, as we argued at greater length in Revolutionary Perspectives 1 (pp.13-14) the accumulation of capital in Russia between 1921 and 1941 proceeded precisely within the closed system which the Luxemburgists deny. In this period Russia's economy grew quicker than at any other time, yet foreign trade at this time amounted essentially to imports of manufacturing and agricultural machinery which was paid for by interest-bearing credits and the export of gold, jewels and art treasures. The export of goods of Department I and II via the credit system was so tiny as to be negligible yet accumulation proceeded. (12)

In the second place, Luxemburgists could fall back on the defence that Russia's rapid accumulation in this period was due to the large number of peasants within the country acting as an "external market" to Russian capitalism. But if they do this, then once again, they are denying that state capitalism is the clearest manifestation of decadent capitalism and ergo that capitalism is not decadent even in 1939. If decadence, in the Luxemburgist schema comes about when all "pre-capitalist markets" are exhausted on a global level then equally this must hold true for every national capital, Russia included. Luxemburg's argument about "external" trade applies to global capital, but under decadence her argument must also apply to national capitals. Obviously, under ascendant capital, "pre-capitalist" outlets might have been found within the national capital (e.g. peasants, tribesmen) and accumulation in the Luxemburgist sense could have proceeded internally. But if "pre-capitalist" outlets for the individual units of global capital can still be found within their respective national capitals, clearly capitalism cannot be decadent.

Therefore a corollary of decadence for the Luxemburgist must be that "internal" accumulation is now impossible for each national capital taken individually. Its historical occurrence would prove either a) the existence of "pre-capitalist" areas on a large scale and hence no decadence or b) the inadequacy of Luxemburgist economics.

In reply to this we have heard nothing from either of the avowedly Luxemburgist groups, the ICC or PIC. The former have attempted to disguise their theoretical nakedness by equating the growth of statification in the twentieth century with that of decadent societies in all epochs (see their pamphlet The Decadence of Capitalism p.8 and p.12), so that in a thirty-three page pamphlet we are not offered a single economic reason for the appearance of this "striking manifestation" (13). The PIC are far more honest in their Luxemburgism, but their barrenness on this question is shown in Jeune Taupe (Nos. 5 and 6) where in two articles on Comecon we are given lists of facts without any kind of theoretical framework. Neither tendency has proved the intrinsic state capitalist nature of Russia et. al. on the basis of Luxemburgist economics and their writings have been confined to windy rhetoric.

Waste production and value theory

If there are those who have followed the argument so far but still cling to the illusion that Luxemburg somehow did hold to value theory, then the last chapter of The Accumulation of Capital should dispel it. The chapter is entitled "Militarism as a Province of Accumulation". In it she says "we have again and again refused to consider the state and its organs as consumers" (page 454). This we would assume to be logical - of the total product it is obvious that the support of the state is a cost of production, its income deriving from the pool of surplus value. However, though she refused to consider the workers as consumers before, she now says that taxation of v (the variable capital apportioned to the workers) provides extra surplus value for capitalisation.

But as yet no opportunities for such capitalisation have come into being, no new market, that is to say for the surplus value that has become available, in which it could produce and realise new commodities. But when the monies concentrated in the exchequer by taxation are used for the production of armaments the picture is changed.

Luxemburg, op.cit. p.456

Armaments production is thus for her a "third buyer",

it is a pre-eminent means for the realisation of surplus value; it is in itself a province of accumulation.

page 454

But what new value has Luxemburg suddenly discovered that she never saw before? The answer is that she has not discovered any at all. Luxemburg has mistaken revenue (in this case taxation) for the creation of new value. It is paid for by taxation (in that respect she is correct) but this must be taxation of the already existing surplus value of other industries.

To illuminate this point we must understand what productive labour is. Productive labour, according to Marx, is that labour which produces surplus value for capital as a whole, is labour whose product can be incorporated in a new cycle of production, either as means of production (e.g. machine tools) or as means of consumption (e.g. food, clothing) which maintain the working class.

If accumulation is to take place, part of the surplus product must be transformed into capital. But short of a miracle only those things can be transformed into capital which are utilisable in the labour process (i.e. the means of production), and in addition such articles which are suitable for the maintenance of the worker (i.e. the means of subsistence). Consequently, part of the annual surplus labour must have been applied to the production of supplementary means of subsistence, over and above the quantity that was requisite for the replacement of the capital advanced. In a word, surplus value is only convertible into capital, because the surplus product whose value it is, already contains the material constituents of new capital.

Capital Vol. I p.638

But whilst machine tools and food production already contain "the material constituents of new capital", what part can arms play in the formation of new capital? They can neither be used to create new means of production, nor can they be eaten by the proletariat. True, guns used for hunting food is a productive use of weapons but this is marginal to their production. (14) Thus the Marxist definition of waste production does not depend on a moral judgement but on the scientific measure of whether new value for total social capital has been created.

Only bourgeois thick-headedness, which regards the capitalist forms of production as absolute forms, hence as natural, eternal forms of production, can confuse the question of what is productive labour from the standpoint of capital with the question of what is productive labour in general, and consequently fancy itself in everywise in giving the answer that all labour which produces anything at all, which has any kind of result is ipso facto, productive labour.

Theories of Surplus Value, ed. Burns, 1951 p.178

This is not to deny that the individual arms firms, like Vickers, does not make a profit when it sells tank recovery vehicles to Iran, but the payment from the Iranians comes from the surplus value created by the Iranian working class. In this sense the Vickers production has not only not increased the surplus value at the disposal of capitalism but has even caused one section of world capital to squander surplus value on commodities which cannot re-enter the productive cycle. Luxemburg therefore, has found new value where there is none.

Turning to our present-day defenders of Luxemburg, we would expect them to also hold to Rosa's view. With the PIC this is true. In their document La crise du capitalisme et ses implications (written for meeting with the CWO in October, 1975) they produce an exact replica of Luxemburg's argument, claiming that arms production is productive for capital. Paradoxically, however, they state that this production which is supposed to be a way out for capital "provokes a super-exploitation of the workforce" (via taxation) which will thus revolt against capital. This nonsequitur is obviously stuck in to avoid the criticism that by seeing arms production as productive for capital, the PIC is preparing for an unlimited period of reformist activity. In practice the PIC's activities have depended on a voluntarist perspective which says that the working class will be (or should be!) so appalled by the consequences of expansion on the basis of militarism that they will seek to revolt in order to avoid a third world war.

So much for the PIC, but what of the ICC? In their major work on decadence (cited above) we find that Luxemburg provided "the clearest statement about the fundamental origins of the decadence of capital to this day" (page 3). In fact, so clear do the ICC find her that she is quoted only once (in the introduction) and never referred to again. Indeed, the theoretical authority used is Fritz Sternberg, a Keynesian, whom the introduction apologetically tells us was a supporter of the Labour Party. Sternberg is excused in the following terms:

But while a recognition of the decay of capitalism is absolutely necessary to any communist practice today, it alone is not enough.

page 6

To which we wholeheartedly agree - as we hope we have shown so far no explanation of capitalist decadence can be adequate unless it is based on value theory itself. And in the last analysis the ICC seems to agree, for when we reach the section on militarism, accumulation and productive and unproductive labour, we find a clear change of method. Gone is the half-hearted empiricism whose sole proof for capitalist decadence on the basis of “saturated markets” lies in quotations from eminent spokesmen of the bourgeoisie like Roosevelt, Hitler and Goering. In place of methods of pure empiricism we have an excellent analysis, which not only returns to Marx (for the first and only time) but also proves that armaments are "a pure loss for global capital" (p.25) and that arms production "is not a solution to crises" (p.30). Indeed, they provide an excellent case against Luxemburg herself.

Its role (i.e. arms production) as an economic stimulant has made certain analysts conclude that the development of the armaments industry was the result of the conscious decision on the part of the capitalists to create an artificial outlet invented to fill the needs of an economy constantly threatened by suffocation due to lack of markets.

op.cit. p.30

The ICC do not tell us who these analysts are; nor do they admit that these positions, correct in themselves, are totally opposed to the framework of Luxemburg. Instead we are given a dishonest, eclectic account which bases itself on shallow empiricism, abandoning it as soon as it becomes inadequate. We do not simply uphold the view that capitalist crises are the product of the tendency for the rate of profit to fall as against the alternative of saturated markets as posed by the Luxemburgists. In a larger sense, we hold all manifestations of the real world can only be understood by an analysis of their relation to the production of value.

Our insistence on the necessity for an understanding of the law of value, is no mere piece of arid scholasticism, nor is it simply a defence of Marx's graven tablets (though we find them eminently defensible, having a coherence based on the law of value). Rather, it is an essential pre-condition of communist politics. Whilst theories about saturated markets, or the capacity for resistance of the working class have a greater immediate appeal, and are therefore superficially easier to comprehend, there can be no royal road to an understanding of capitalist contradictions. This understanding is the only basis for our actions if we are not to sink into voluntarism or despair. We must first understand how the law of value operates before we can begin to destroy it.

Communist Workers' Organisation, 1976

Note on Sources

As the argument is based on textual interpretation in many places these can be checked in the following editions.

  • Luxemburg - The Accumulation of Capital (Routledge Kegan Paul 1963)
  • Luxemburg - The Accumulation of Capital: An Anticritique (including Bukharin's Imperialism and the Accumulation of Capital) (Monthly Review Press 1972)
  • Luxemburg - What is Economics? (Merlin 1954)
  • Luxemburg - The Russian Revolution (Ann Arbor 1972)
  • Marx - Grundrisse (Pelican 1973)
  • Marx - Capital Volume I (Everyman 1972)
  • Marx - Capital Volume II (Lawrence and Wishart 1971)
  • Marx - Capital Volume III (Lawrence and Wishart 1974)
  • Marx - Theories of Surplus Value Volume I (Lawrence and Wishart 1969)
  • Mattick P. - Marx and Keynes (Merlin 1971)

(1) On the national question, see “Marxism and the Irish Question” in Revolutionary Perspectives 2.

(2) Luminary of the French group Pour une Intervention Communiste (PIC) See below.

(3) And the other Luxemburgist group, the International Communist Current (ICC - British sibling, World Revolution) believe that Luxemburg held her saturated markets views in Social Reform or Revolution. See their pamphlet The Decadence of Capitalism.

(4) Not understanding how Marx himself had said the tendency would act.

We have thus seen in a general way that the same influences which produce a tendency in the general rate of profit to fall, also call forth counter-effects, which hamper, retard, and partly paralyse this fall. The latter do not do away with the law, but impair the effect. Otherwise, it would not be the fall of the general rate of profit but rather its relative slowness, that would be incomprehensible. Thus the law acts only as a tendency. And it is only under certain circumstances and only after long periods that its effects become strikingly pronounced.

Capital Vol. III p.239

(5) She excused this by maintaining that Volume III of Capital was only a series of rough notes put together by Engels (See pp.168-9 of The Accumulation of Capital). What she never asks is why the ailing Marx felt it necessary to press on with the rough notes for Volume III (which deals with the falling rate of profit) whilst leaving incomplete the volume devoted to the circulation of commodities.

(6) Luxemburg quite often pays lip service to Marx’s basic postulates and then proceeds to ignore them in practice. Not only on the question of competition (The Accumulation of Capital p.40) but also on the question of the equalization of the rate of profit. On page 79 of the above she acknowledges its significance but on p.34O she denies its possibility.

(7) It is no accident that Joan Robinson, a Keynesian should find Luxemburg's theory “of the greatest interest” or that Luxemburg's main defender in the 20's (Sternberg) should become a Keynesian in the 40's.

(8) This is the best critique of Luxemburg's economics known to us despite the wretched conclusion which is a non-sequitur to the text.

(9) She seems aware of this contradiction (see p.368 op.cit) but never attempts to solve it.

(10) See The Economic Foundations of Capitalist Decadence in Revolutionary Perspectives 2 for a fuller explanation.

(11) See Theories of State Capitalism in RP 1; RP 2 pp.43-50; RP 7 will also contain an analysis of capital's crisis in Eastern Europe.

(12) We do not mean here primitive accumulation - Russia had already completed this stage and in 1914 was one of the leading capitalist states.

(13) Though we are told on page 12 that state capitalism in “young capitalisms” today is caused by the internal contradictions of the system. Precisely what these internal contradictions can be for a Luxemburgist we are never told.

(14) And the crackpot scheme to create harbours in Australia by the use of atom bombs has been shelved until a “clean” bomb can be found.

Revolutionary Perspectives

Journal of the Communist Workers’ Organisation -- Why not subscribe to get the articles whilst they are still current and help the struggle for a society free from exploitation, war and misery? Joint subscriptions to Revolutionary Perspectives (3 issues) and Aurora (our agitational bulletin - 4 issues) are £15 in the UK, €24 in Europe and $30 in the rest of the World.