Donate to Links

Click on Links masthead to clear previous query from search box

GLW Radio on 3CR

Recent comments


Syndicate content

The debate on the rate of profit

By Michel Husson

July 2010 -- International Viewpoint -- A polemic on the rate of profit has developed over the last few months. This article seeks to review this debate which turns around four essential questions. [1]

The four questions are:

  1. an empirical question: what has been the evolution of the rate of profit since the early 1980s in the big capitalist countries?
  2. a theoretical question: what is the status of the tendential fall in the rate of profit in the Marxist analysis?
  3. a “semi-theoretical” question: what is the nature of the crisis?
  4. a programmatic question: what is the impact of this discussion on the proposals advanced in the period opened by the crisis?

The evolution of the rate of profit

The entry point to the debate concerns whether the rate of profit has risen or not since the early 1980s, notably in the United States which most contributions focus on. Graphic 1 below summarises and updates the results of previous works [2]. Both in the USA and in the three main European countries, we can clearly distinguish two periods: a fall in the rate of profit until the early 1980s, then a rise. It can be noted that the fluctuations are most marked in the USA where the rate of profit falls from 2007 onwards, and this before the crisis moreover (Husson 2009b). But the tendency is certainly there.

To this “bull” position is opposed a “bear” position (to employ the most neutral terms possible, here borrowing the language of the stock market) which disputes this schema and advances other evaluations of the rate of profit, which do not show an upward tendency since the early 1980s. The spectrum is moreover fairly broad, going from a smaller net increase in the tendential fall, via a flat encephalogram.

The reasons for these differences [3] concern the measurements of profit and capital used to calculate the rate of profit. So far as profit is concerned, two questions are raised: should we take into account, in the case of the USA, enterprises which do not have corporate status? And, more generally, should we take into account the profit of financial companies?

In the USA, the statistics of the BEA (Bureau of Economic Analysis) distinguish profits (corporate profits) calculated for companies (domestic corporate business) and incomes from ownership (proprietors’ income). The latter concern enterprises which have a status other than that of company (corporate): partnerships, sole proprietorships and tax-exempt cooperatives). In incorporating them, we find the concept of net operating surplus for the whole of the private sector and that is why I have employed this enlarged definition of profit. However, this choice modifies the level of the rate of profit but leaves its evolution practically unchanged.

On the other hand, the evolution of the rate of profit is not the same according to whether or not one takes into account the profits of financial companies (banks, insurance and so on). If we exclude these from the definition of profit, the rate of profit (still in the USA) tends to stay flat. This can be understood: the share of profit taken by the financial companies represents a growing fraction of the profits realised by the private sector as a whole.

The argument invoked is that these amount to virtual profits which correspond to the valorisation of fictitious capital. But it seems to me that there is here a confusion between company accounting and national accounting; for the latter, the profit of financial companies is defined more or less as the difference between interest received and interest paid. This flow measures the capacity of the banks and insurance companies to capture a part of the value created every year. As to the valorisation of assets, it appears elsewhere in the balance sheets ; the same goes moreover for the non financial companies and for households whose income does not incorporate the valorisation of their assets (shares, houses and so on) which is recorded in another account, that of their “wealth” [4]. The value added of financial companies (and thus their profits) is a component of GDP of which the real counterparts are consumption, investment and the trade balance. Not considering the profits of the financial sector as real flows would lead to breaking accounting equality between the two “optics” of measurement of GDP.

Then there are the problems of measurement of capital. Some value it in “gross” rather than “net” terms, in other words not taking account of depreciation, or in Marxist terms, of the transmission of the value of fixed capital to commodities. But the main controversy concerns the mode of valorisation of capital: either at current prices – as done by most contributions – or at historic cost which is, according to Andrew Kliman (2009b), the only correct method. I have discussed this position in a text called Les coûts historiques d’Andrew Kliman, republished in part in La hausse tendancielle. The response by Kliman (Masters of words, 2010) does not seem to me to change the terms of the debate on this point nor to fundamentally challenge the arguments I have advanced. In truth, this choice (historic cost or current prices) does not have enormous empirical implications. The true difference resides in the corrections subsequently made by Kliman to measure the rate of profit in value which lead to tendentially falling rates of profit over the last 50 years (Husson 2009c).

Two readings of the neoliberal period

This note of reestablishment of the rate of profit takes place within a reading of the period which emphasises several “stylised facts”. This is a term we owe to the economist Nicholas Kaldor (1961) who explains his method thus: "facts as recorded by statisticians, are always subject to numerous snags and qualifications, and for that reason are incapable of being summarized” and hence theorists “should be free to start off with a stylised view of the facts – i.e. concentrate on broad tendencies, ignoring individual detail” (Nicholas Kaldor (1961), ’Capital Accumulation and Economic Growth.’ In: Lutz/Hague (eds.): The Theory of Capital, London, pp. 177-222). Among the stylised facts identified by Kaldor, we already find the stability of the rate of profit, the capital-output ration and the share of labour in total income. This method is still valid today and it is possible to identify four stylised facts which characterise capitalism in its neoliberal phase:

1. a fall in the share of labour

2. an increased rate of profit

3. stagnation of the rate of accumulation

4. an increase in the share of dividends.

One of the characteristics of capitalism (seen through “bull” spectacles) is that the rate of profit increases but does not lead to an increased rate of accumulation. This is not to say that the rate of accumulation falls, but that it does not increase as much as the rate of profit. In the “bear” version this scissor effect between profit and accumulation obviously doesn’t exist: the rate of profit stagnates, the rate of accumulation also, so they are in synch.

A supplementary argument is however advanced by Louis Gill, here following a suggestion by François Chesnais: what if the investment was realised elsewhere than in the imperialist metropolises? Domestic investment would be relatively stagnant but would be broadly compensated by investment abroad, particularly in the emergent countries.

This objection raises a more general problem. Capitalist globalisation tends to dissolve the economic significance of national frontiers: the map of the multinationals corresponds increasingly less to that of countries. To take only one example, the US trade deficit is linked to a great extent to imports from emergent countries like China, but which also correspond to investment and transfers of production by US companies. The usual accounting tools are increasingly less appropriate to this globalised world. That said, the more rapid growth of international investment is not a sufficient objection. Although it reduces the “scissors” between profit and accumulation, it would be necessary to show that its profitability is lower than that of domestic investment, which is highly improbable.

A supplementary difficulty is that the data on international investment have difficulty in distinguishing “real” investment as in financial investment from movements inside groups. A recent study shows that in the case of France, a stricter definition of investment flows leads to a perceptible decrease in their size: in 2008, French investment abroad would be more than 80 billion euros against 137 with the traditional method of calculation. In the opposite sense, foreign investment in France would only be 10 billion instead of 66 (Nivat and Terrien 2010).

It amounts to a vast worksite which goes beyond the question of the rate of profit alone. One can nonetheless mobilise rare data relating to the groups as a whole, and we dispose of some precious information based on a Thomson Financial data base. It brings together the results of 215 non-financial companies appearing in the index of the 250 biggest French companies ranked according to their stock market capitalisation. These enterprises are highly internationalised, inasmuch as they realise 60% of their turnover outside of France (32% in Europe and 28% in the rest of the world). These consolidated accounts thus relate to quantities which are not reduced in France, whether in terms of wages or investment.

We find then the stylised facts mentioned above, in first place the increase in the rate of profit, measured here as the relationship between profit and total capital (graphic 2). This upward evolution is more marked than for the non-financial companies taken as a whole, which means that the big groups register higher profit rates, in part thanks to their internationalisation. Note in passing that capital is here measured in the manner that capitalists measure it in their balance sheets, namely at historic cost, which shows that the increased profitability appears even with this measurement that Kliman stressed was the only one possible.

A more detailed examination of the data (table 1) shows that capitalist intensity remains virtually constant, which implies that the increased profitability is essentially due to a considerable decline in the share of wages in value added: it fell by 11.6% over the period in question.

Despite this leap forward in profits, gross investment only increases by 1.1% and thus we see the famous “scissors" between profit and investment appear: the difference is then 8.4%. This non-invested profit is devoted to debt reduction inasmuch as external financing falls by 4.2% and thus to the increase in dividends paid which go from 2% to 6.2% of value added. Even subtracting interest (falling) and taxes (upwards) from profit, the share of gross profit, in other words profit after taxes and interest (but before depreciation), increased by 9.5%.

In short, the set of stylised facts is found in this study which has, once again, the dual interest of starting from the private accounting of the big groups and of not confining itself to the national level. This confirms in particular that foreign investment does not compensate for the weak dynamism of domestic investment.

The “scissors” between profit and accumulation of capital is certainly a fundamental characteristic of contemporary capitalism which I have stressed for a long time (see for example Husson 1999). And this note is broadly shared. Thus, the book by Patrick Artus, Le capitalisme est en train de s’autodétruire, speaks of a “capitalism without project”. A study by the UN explicitly stresses this phenomenon (Husson 2008b). An entire literature of “post-Keynesian" Inspiration takes this phenomenon for granted and seeks to explain it. One of the most interesting authors in this current, Engelbert Stockhammer (2006), poses thus the question of the investment-profit puzzle: “We face then an interesting enigma: the ratio of investment to profit shows a downward tendency [and] all countries show a similar tendency. Although this is a logical consequence of the relative evolution of investment (downward) and profits (upward), that raises interesting questions which, bizarrely, have received little attention until now. The first is knowing why companies do not invest their profits? You could call this the Marxist question”.

John Bellamy Foster and Fred Magdoff (2008) point to the “spectacular decoupling between profits and net investment" in the USA; the latter is falling “significantly” in percentage of GDP, as well as the share of profits in GDP reaching “a level never observed since the beginning of the 1970s” (graphic 3).

In total, there are two readings of the configuration put in place by neoliberal capitalism, which can be summed up in the table below.

The four stylised facts of the “bull” analysis are compatible with each other and lead to a coherent “recital” of the period. The same is not true for the “bull” analyses which, moreover, only rarely offer an overall view.

To show this, let’s start from the share of dividends (uninvested profits) in total profit: it seems hard to deny that it has strongly increased over the period (table 3). In addition there is agreement in saying that the rate of investment has not increased. That implies that the counterpart of the increase in dividends (still in proportion of profit or of value added) is a fall in the share of wages. But in this case the rate of profit should have increased.

A first way of emerging from this logical contradiction would be to argue that the rate of profit has remained constant (or fallen) in spite of the fall of the share of wages, because the organic composition would have increased. But no partisan of this thesis has produced such a demonstration, and for a reason: the movements of the income/capital ratio (an approximation of the organic composition of capital) are small and somewhat upwards.

The only way out is to argue that the share of wages has not fallen, but it is at the price of a debatable affirmation, even in the case of the USA. Louis Gill (2010) argues that "the share of profits in value added has remained more or less constant from 1980 to 2008 in the United States” and produces graphic 4A reproduced below. One could counter this with graphic 4B which shows on the contrary that the share of profits has increased by 5-6% between the early 1980s and today.

Another incomprehensible debate between statisticians? No: the two graphics are constructed on the same data. The graphic on the right uses a different scale and deals with a longer period which gives a good overview of the parallelism between the share of profits (more or less the rate of exploitation) and the rate of profit: down before 1980, upward tendency since.

Kliman (2009c) adopts the same position: “wages receive a relatively constant share of the GDP produced by companies, throughout the post war period. This division has no major effect on the rate of profit”. He specifies that this "ratio of income from ownership to the remuneration of wages is similar to what Marx calls the rate of surplus value”. Indeed, it is a well know fact that the median wage has become disconnected from the productivity of labour, which should lead to a fall in the share of wages.

The solution to this apparent contradiction is to be found on the side of the wages of managers which have significantly increased. Therein lies another debate: what is the nature of very high wages whose share has constantly increased in the US? Kliman makes them a category apart which is neither surplus value, nor variable capital, and stresses that these incomes escape the enterprises. It is a very debatable argument: on this account, the dividends paid to the shareholders would not be surplus value either, since, by definition, these profits are not retained by companies. It is enough however to exclude one per cent of the highest wages to find a fall in the share of wages as marked in the US as in Europe (Husson 2010d).

Let us reformulate the reasoning in the opposite direction and admit that the rate of profit, the rate of accumulation and the share of wages are constant. The investment-profit enigma does not exist. But if all is constant in proportion of value added (profits, wages and investment), we do not see how the share of dividends should increase. In short, the "bear” reading leads logically to denying either the increase in the share of profit distributed to shareholders, or the fall in the share of wages, which are yet two characteristic features of contemporary capitalism.

The law of the tendential fall of the rate of profit

In the Marxist vulgate, the trajectory of the rate of profit depends on the relative evolution of its two components, the rate of exploitation — to the numerator — and the organic composition of capital — to the denominator. One says then that the rate of exploitation can increase up to a certain point only, whereas the organic composition increases continuously. Consequently the rate of profit ends up by falling (it is a tendential law). The law is then at the end of the day a law of the tendential increase of the organic composition: the accumulation is reflected by an increase of dead labour (capital) in relation to living labour, the sole creator of surplus value. In these conditions, surplus value tends to increase less quickly than the capital advanced, hence the fall in the rate of profit. The notion of “tendential law” means that the rate of profit does not fall always and everywhere but at the end of the day the tendency triumphs over the counter-tendencies.

This classic presentation is not correct because it does not sufficiently break down the factors which determine the evolution of the rate of profit and does not bring out the dual influence of the productivity of labour, both on the numerator and the denominator. In the numerator it is clear: when productivity accelerates and the real wage does not follow, the rate of exploitation increases. This is what Marx calls relative surplus value.

But the productivity gains also have an effect on the denominator, in other words on the organic composition: they bring down the price of the “machines” which make up fixed capital and can thus compensate for the increase of their number. It is necessary to distinguish two notions: technical composition and organic composition of capital. Nobody disputes that the technical composition of capital (the number of "machines” in relation to the number of employees) increases tendentially. But that does not imply an increase of the organic composition (in value): the productivity gains intervene here also by bringing down the value of the "machines” which make up fixed capital and can thus compensate for the increase of their number.

It is precisely one of the " causes which counteract the law” (of the tendential fall of the rate of profit) laid down by Marx : “the same development which increases the mass of the constant capital in relation to the variable reduces the value of its elements as a result of the increased productivity of labour, and therefore prevents the value of constant capital, although it continually increases, from increasing at the same rate as its material volume i.e., the material volume of the means of production set in motion by the same amount of labour-power. In isolated cases the mass of the elements of constant capital may even increase, while its value remains the same, or falls". Thus the same influences which tend to make the rate of profit fall, also moderate the effects of this tendency”. (Capital, Volume 3, Chapter 14).

But precisely because we are talking about the “same influences” (the productivity of labour), there is no reason a priori to think that the tendency will systematically outweigh the counter-tendency. Once again, the productivity of labour is liable to compensate at the same time, in a perfectly symmetrical manner, for the increase in real wages and the increase in physical capital. Or else, it would be necessary to postulate that the productivity of labour increases systematically less quickly in the section of production goods than in that of consumer goods, but it is the opposite configuration which is the most frequent.

The evolution of the organic composition is then in reality undetermined, as Marx suggests: “In isolated cases the mass of the elements of constant capital may even increase, while its value remains the same, or falls". The numerator and denominator of the rate of profit can then remain constant, and consequently the rate of profit itself.

If one wishes to study the conditions of evolution of the rate of profit, it is necessary then to abandon the classic binary breakdown (rate of surplus value/organic composition of capital) for a ternary breakdown bringing in wages, the productivity of labour, and the efficiency of capital, that is the income/capital ratio [5] We obtain then the following result: the rate of profit increases if the increase of the real wage is lower than that of the “global productivity of factors” defined as the weighted average of the productivity of labour and the efficiency of capital. In simple terms, the gains of the productivity of labour could compensate both for the increase of real wages and that of physical capital per head. The error of the canonical presentation of the law consists in forgetting this possibility by confusing the organic composition of capital (in value) with its technical composition.

Since the argument was used by Chris Harman (2010b), it should be stressed that this analysis has nothing to do with “Okishio’s theorem”. This theorem was supposed to demonstrate that the rate of profit cannot fall because the capitalists would never introduce technologies liable to make it fall. This demonstration forgets obviously the competition between individual capitals and the uncertainty which accompanies any investment project. But the reference to this theorem is outside of the subject: there is then indeed a clear difference, which relates to the most elementary logic, between saying that the rate of profit cannot fall, and affirming as Okishio does, that the rate of profit cannot fall. These are two different debates.

It is necessary then to place Marx’s statement in an analysis of the dynamic of capitalism and to present a long term version of the law which could be put thus: “The conditions of functioning of capitalism can be met for a fairly long period, but the mechanics ensuring this are not stable or in any case cannot be durably reproduced” (Husson 1996). At the end of a given time, the increase in capital per head no longer produces the same productivity gains. It is the double decrease in the productivity of labour, in relation to capital per head but also in relation to wages which initiates the fall in profit. It is the structural contradictions of capitalism (search for the maximum profit, competition between capitals) which lead tendentially to this fall. But this schema only applies on the long horizon, that of the theory of long waves.

In addition this global efficiency is indissociable from the matching of needs which do not emerge from technology: it is necessary moreover that the real wage acts on the “good” commodities from the viewpoint of the productivity gains that they allow. The rate of profit should then be conceived as a synthetic indicator of the dynamic of capital which reflects all the terms of the reproduction of capital, on the side of production (creation of surplus value) and on the side of realisation (outlets).

The time of capital and the nature of crises

One of the reasons for the divergence of views among Marxist economists relates to a problem of method: can we extend the analysis of the cycle to a general comprehension of the dynamic of capitalism and its different levels of crisis? Can we mobilise the same tools that were suitable for periodic crises to analyse the current crisis? One could perfectly well defend the idea that this distinction is not really employed by Marx, who did not dispose of a very long period of observation, nor of a completely finalised theory of crises. But, apart from thinking that Marxism is reduced to the writings of Marx alone, it is necessary to take into account subsequent developments which have allowed a better comprehension of the capitalist dynamic.

Kondratiev’s work on "long cycles” inaugurated a new perspective seeking to periodise the trajectory of capitalism. But his fault was precisely to have conceived these long movements as cycles and this is the main subject of the critique Trotsky made of him. It is necessary to reread Trotsky’s short article, “The curve of capitalist development”, published in 1923, because it provides the bases of an understanding of the long time of the history of capitalism: “We observe in history that homogeneous cycles are grouped in a series. Entire epochs of capitalist development exist when a number of cycles is characterised by sharply delineated booms and weak, short-lived crises.

As a result we have a sharply rising movement of the basic curve of capitalist development. There are epochs of stagnation when this curve, while passing through partial cyclical oscillations, remains on approximately the same level for decades.” This line of thought would be taken up by different authors and would be later systematised by Mandel with the theory of long waves. The Marxist analysis should then integrate this contribution by adopting a historic approach to capitalism and taking up the notion of “productive order” (Barsoc 1994). Concretely, it means accounting for the differences between the neoliberal capitalism established since the beginning of the 1980s and the so-called “Fordist” capitalism of the post war boom period. In both cases, it was always and still is capitalism, but its dynamic, mode of reproduction, and social effects are different. In other words we are talking of different historic periods: “The long waves are more than simple upward and downward movements of the growth rates of the capitalist economies. They are, in the full sense of the term, specific historic periods” (Mandel 1995).

Marxist theory is then confronted with a double task: stressing that the fundamental social relations are unvarying but, at the same time, that their implementation is not the same, according notably to different social relations of force. In the absence of this work being done, the absence of historic periodisation leads to certain amount of driftage, the first being to apply to long waves the tools of analysis adapted to short cycles.

The modelling of the cycle is an essential contribution by Marx who was a precursor of it and has never really been transcended. Simplifying in the extreme, his schema is the following: during boom periods, the rise in the rate of profit and competition leads capitalists to anticipate the pursuit of the movement and to invest too much. But these excesses of capital no longer succeed in valorising themselves: the rate of profit falls and the economy enters into recession. The mechanics of the cycle rest then on the specific time frame of the accumulation of capital with an over-reaction, upwards and downwards, of investment to outlets: demand is a flow, capital is a stock. This phenomenon is called the “accelerator” by macroeconomists. It is accentuated by the cyclical evolution of the share of wages: in general it tends to rise when the economy slows up because wages do not react immediately to the slowing up of productivity. Fluctuations in the rate of exploitation thus combine with those in investment to give a strongly cyclical character to the rate of profit.

In total, the dynamic of the cycle results from the behaviour relating to accumulation of capitalists subject to competition: this mechanism is incorporated in the “normal” functioning of capitalism [6]. And this same mechanism guarantees in some way an automatic exit from recession. This schema was taken up by a non-Marxist economist, Patrick Artus, in the case of the so called cycle of the “new economy” of the second half of the 1990s in the USA (Artus 2002). But, precisely, it was a cycle and the current crisis cannot be reduced to a cyclical crisis.

Another consequence of the absence of historic approach consists in not understanding the sequence of the mechanisms which lead to crisis. In the current crisis, the rate of profit has obviously fallen and even began to do so a little before the outbreak of the financial crisis (Husson 2009b) but that had nothing to do with a prior “over-accumulation”. The latter only appeared with the outbreak of the crisis, in the form of the excess capacities of production “revealed” by the crisis.

We can illustrate this point starting from the case of the USA by comparing the volume of consumption, the rate of profit and the rate of use of capacities. The latter measures the relation between effective production and the potential production which would result from a full use of capacities (which is never 100%): it is then a good indicator of the over-accumulation of capital, in the sense of excess capacities of production. The three curves evolve in a parallel manner (graphic 5), confirming thus the link which exists between outlets, rate of use of capacities and profitability of capital. The slowing up of outlets leads to a low use of capital and brings down the rate of profit. We can see the outline of the end of the “high tech” cycle, with the recession of 2001-2002, followed by a recovery: here we see the cyclical functioning of the economy. Then comes the crisis: we see then a virtually simultaneous fall of great breadth, of consumption. rate of use and rate of profit. In other words, under-consumption, over-accumulation and the fall in the rate of profit are closely linked and it is their interaction which needs to be understood. It is not by opposing two interpretative frameworks (over-accumulation versus under-consumption) that one can come to a better comprehension of the empirical facts. But again it is necessary to understand the terms.


We should first clarify a possible confusion between overaccumulation of capital and flow of fictitious capital. The latter designates, as its name indicates, a set of potential drawing rights on surplus value. Their volume increases with the swelling of the financial bubble but this is an overaccumulation as fictitious as the capital of the same name. Its “exuberant” growth prevents it from being able to claim an effective profitability equal or superior to the average rate of profit, because the sum of the virtual financial income exceeds the capacity for extraction of surplus value. This capital is then fictitious to the extent that the available surplus value does not allow it to use its apparent profitability.

Then we should specify terms. Most of the orthodox analyses tell us that the rate of profit falls because there has been overaccumulation of capital. But this is a pure tautology. For Marx, overacumulation is defined indeed by relation to the inability to obtain the average rate of profit. One cannot then make of this concept a principle which explains the fall in the rate of profit which is the form under which it manifests itself.

In the current sense of the term, over-accumulation suggests that too much has been invested. But, again, in relation to what? In reality the response can only relate to outlets. The empirical process observed in the crisis follows this sequence: fall in outlets → over-capacities of production → fall in rate of profit. Concretely, the fall in outlets leads to the under-utilisation of capacities and the stagnation or fall in the volume of profit. In other words, the same capital advanced produces less profit and the rate of profit falls.

The classic exposition of the tendential law of the falling rate of profit rests on another sequence:

investment → increase in organic composition → fall in rate of profit.

But this is again a determination which is not suitable for the analysis of this crisis which is of another nature than a periodic crisis. One can convince oneself of this by comparing the gap in the rate of profit in its medium term trend and the output gap defined as the relation between effective production and potential production which would correspond to a normal use of the capacities of production. These two indicators are closely correlated (graphic 6). In other words the evolution of the conjuncture explains very well the fluctuations of the rate of profit around its tendency.

All these observations point to the difference between two lines of interpretation of the crisis. Either it amounts to a cyclical crisis, especially strong, but capable of being analysed with the tools usually mobilised in the analysis of the cycle: profit falls (and thus there is a crisis) because of over-accumulation. This viewpoint is summed up in a simple turn of phrase by Carchedi (2010) who writes: “crises (lower rates of profit) ...” establishing thus the equivalence which underpins this type of analysis:

crisis = fall in rate of profit.

Thus, if the crisis conforms to the theory, it is necessary that the rate of profit falls, all the more in that this validates the law of the tendential fall. One can invoke also the chronic instability of capitalism and thus reassure oneself on the relevance of the Marxist toolkit.

Patrick Artus has tried again to apply a Marxist schema to the current crisis (Artus 2010): “It certainly is a Marxist reading (but one in accordance with the facts) of the crisis: over-accumulation of capital hence tendential fall in the rate of profit”. It amounts perhaps to a “Marxist” reading but it should be noted that it is not “in accordance with the facts”. Inasmuch as the previous one was convincing because it applied to a cycle, in the same way this new exercise misses the systemic character of this crisis.

The period of neoliberal capitalism is not characterised by a tendency to over-accumulation and we should rather speak of under-investment: the accumulation of capital, throughout this period, does not follow the reestablishment of the rate of profit. One finds then this essential idea: we should distinguish two temporal horizons to which one cannot apply the same tools of analysis.

The current crisis represents a major rupture in the neoliberal “productive order” whose elements, put in place over a long period, are coming apart globally. The difficulty of analysis here is that the “great crises” (to take up the expression of Robert Boyer) are always unleashed during a cyclical recession: it is necessary to go beyond the note of the breadth of the recession (stronger than usual) and analyse it, not as a periodic crisis, but as an entry into crisis of the essential features of the period.

The crisis of 1974-75 had been a crisis of “Fordism”, in other words of the correspondence which was established between productivity gains and purchasing power. The current crisis can be analysed fundamentally as a crisis of the solutions developed to the previous crisis around a central question, that of realisation. Taking account of the tendential fall of the share of wages in income, capitalism risked colliding with a lack of outlets. The solutions to this contradiction were found through finance. To simplify, finance allowed three things: 1) recycling of uninvested surplus value into the consumption of restricted social layers; 2) feeding the over indebtedness of households and sustaining their consumption ; 3) adjusting global imbalances, mainly between the United States and the rest of the world. The financial implosion challenges these three elements and consequently the coherence of the neoliberal order.


“The ultimate reason for all real crises always remains the poverty and restricted consumption of the masses as opposed to the drive of capitalist production to develop the productive forces as though only the absolute consuming power of society constituted their outer limit.” In spite of this principle clearly affirmed by Marx [7], the dogmatic version of Marxism rests at basis on a binary opposition between two analyses of the crisis: the first, centred on the concept of over-accumulation, would be the only authentically Marxist one; the second, characterised as under-consumptionist would not be Marxist but Keynesian.

This not very dialectical interpretative framework shows an incomprehension of an essential feature of capitalism: it is a mode of production that seeks to obtain the highest rate of profit possible but must also sell its commodities. This dual demand generates a permanent contradiction which manifests itself with a particular vigour during crises. We find here the error, signalled by Mandel, which consists in “arbitrarily splitting that which is organically linked, at the very heart of the capitalist mode of production (…). To wish to explain the phenomenon of crises exclusively by what happens in the sphere of production (the production of an insufficient quantity of surplus value to ensure to all capital an acceptable rate of profit), disregarding phenomena of realisation of surplus value, that is of circulation, thus of the market, is in reality to disregard a fundamental aspect of capitalist production, that of a generalised commodity production” (Mandel 1982).

Most of the time, the accusation of “underconsumptionism” rests on other quotations from Marx criticising theories which make insufficient consumption the mainspring of crises. But this respect for orthodoxy forgets one of the essential contributions of Marx, namely his study of the conditions of reproduction of capital. It is however a key question that can be summarised thus: who buys what is produced by the exploited employees? It is all very well (for an employer) to exploit their workers but the profit drawn from it remains virtual so long as it is not realised by the sale of commodities. This question is posed during the cycle, but it is posed in a structural manner over the long term. The upwards tendency of the rate of exploitation observed since the early 1980s poses a problem from the viewpoint of realisation. If the share of the consumption of employees falls by relation to the new wealth produced, the question is who will buy the rest.

To say that capitalism in its neoliberal phase faces a chronic realisation problem does not amount to support for the so-called under-consumptionist theses. The theorists of underconsumption, from Sismondi to Baran and Sweezy via Rosa Luxemburg, argued that capitalism was structurally incapable of realising profit and that it had need of external outlets. Nobody supports this thesis to this degree of generality, but the fall of the share of wages highlights a manifest problem of realisation that is met by actually existing capitalism and that it resolves through the consumption of the rich and through indebtedness.

To sweep aside this question by saying that only over-accumulation and overproduction are the causes of the crises, through the fall in the rate of profit, amounts to forgetting that over-accumulation and under-consumption are both aspects of the same reality, as formulated very well by Chesnais (2010) : “Over-accumulation has automatically for an “opposite”, so to say, under-consumption”. And vice versa.

The chronology of the crisis

That the crisis emerged in the financial sector, nobody can deny. It is even emerged on a relatively narrow segment of world finance inasmuch as the point of departure was a fraction of the US mortgage market, the famous subprimes. This is what allowed certain analysts to think (at its beginning) that the impact of this crisis would be limited. But it has extended like a trail of powder to the whole of global finance, “thanks” notably to the famous derivatives. There is moreover there a field of research of a sociological order, to explain the blindness of the finance specialists who believed in their innocence or who did not see the dangers. One will find without doubt a curious cocktail of ignorance, of mathematic dogmatism, cupidity and pure delinquency.

The fact that the system had cracked on the finance side does not allow us then to conclude that it amounts to a financial crisis. That would only be possible in postulating a strict separation between finance and that which one is accustomed to call the real economy. Indeed, these two “spheres” are closely intertwined and financialisation developed as a response to contradictions appearing in the real economy for a long time. The manner in which the financial crisis extended to all the segments of the economy give concrete indications on this mode of articulation: credit crunch, the so called effect of negative wealth on consumption, falling investment and so on. And above all, there is globalisation which has transmitted the recession to the whole of the world economy, pulverising the thesis of decoupling according to which the emergent countries would be spared.

The programmatic stakes

To paraphrase a somewhat provocative formula of Claudio Katz, during a seminar in Amsterdam: the debate between the two positions has no direct programmatic implication. Such an assertion comes up against a certain tradition, according to which a good analysis of the conjuncture leads necessarily to a good political position. A leader like Ernest Mandel could combine the two skills, but it is necessary to be wary of a posture which would give Marxist economists a role which generally is beyond their capacities.

In the debate which interests us here, the viewpoint adopted as to the evolution of the rate of profit says nothing on the political implications of the note. Certainly, there is a great temptation to establish facile connections. For example, those who, like the author of these lines, hold that the fall of the share of wages is a fundamental cause of the crisis, see themselves accused of Keynesianism or “underconsumptionism”. If they say that wages are too low, they are arguing for a wage-led recovery which would save capitalism. This form of rhetoric is reversible: those who think that the main cause of the crisis is the fall in the rate of profit could be accused of being in an underhand way in favour of a lowering of wages so as to re-establish profits.

Another line of demarcation separates those who analyse this crisis as a financial crisis and those who consider it as a crisis of the system itself. And that leads effectively to different orientations: regulationist-reformist in the first case, anti-capitalist in the second. Again we could discuss this opposition: after all, one could very easily hold that this crisis is financial and be at the same time anti-capitalist, even if this position does not exist in practice. More generally, anti-capitalism is not indexed on the rate of profit. The reasons that we have all criticised this system are not located in the evolution — upwards or downwards — of the rate of profit.

That is why Claudio Katz’s recommendation is useful: we should carefully distinguish theoretical debates from programmatic ones, and avoid thinking that the theoretical analyse of the conjuncture supplies us mechanically with the key to the strategic issues. That capitalism as a system is the target is a point of agreement, which should not be spoiled by polemics which play on words. To take an example, the idea according to which capitalism would be increasingly less capable of satisfying social needs was mocked by Gill (2010) as an absurdity indeed an illusion on my part, since this is not the objective of capitalism. However the latter must sell its commodities and it could not do it if they were deprived of use value, in other words did not respond to any social need. To fulfil this necessary condition, it shapes needs and the allocation of incomes. But it seems to me that one of the characteristics of contemporary capitalism is that this is increasingly difficult: the gap is growing between profitable supply and social demand, and capitalism tends increasingly to reject the satisfaction of elementary needs in the name of its criterion of profitability. There is here a critical line which touches the very bases of the system — that which I call the mode of capitalist satisfaction of social needs — and goes much further than the study of the rate of profit.

A second reproach targets my critique of the irrationality of the system. Understanding the rationality of this system is not relevant, says Louis Gill. But, bizarrely, he refers to an article by Chesnais which evokes… “the fundamental irrationality of capitalism” in an article which is apparently — all the same — “at the antipodes” (sic) to my own analyses. It would be however possible to agree that capitalism has its own logic, but that it is increasingly irrational from the viewpoint of humanity (and of the planet). And that allows once again criticism of the system on other bases than its chronic instability.

If one leaves aside the reformist witch trials, the debate is of the strategic order. As always it concerns the articulation of immediate slogans and the socialist perspective. The crisis is exacerbating the tension between these two political levels. On the one hand, its immediate effects are equivalent to a growing social regression and, on the other, its breadth demonstrates the fragility and growing illegitimacy of the system. The construction of a transitional approach is then all the more necessary, but in a more difficult sense. It is necessary both to fight tooth and nail against the measures for “exit from crisis” and open a radical, hence anti-capitalist alternative perspective. It seems to me that the question of the division of incomes is a good point to hang around the principle “we won’t pay for their crisis”. That has nothing to do with a "wage led recovery" but with a defence of wages, jobs and social rights on which there should not be any dispute. Then comes the idea of control over what they do with their profits (pay dividends or create jobs) and our taxes (subsidising the banks or financing public services). The issue is to pass from defence to control and it is on the basis of this switch that the challenge to private property (the real anti-capitalism) can acquire a mass audience.

This approach can be discussed and should be worked on, but it is counterproductive to rule it out as reformist, or regulationist, opposing to it the sole revolutionary posture which would be to call for the overthrow of the system without having a precise idea of what roads mobilisations can take and the concrete targets they should seek. On a more tactical level, the “razor sharp” delimitations seeking to separate the good anti-capitalist wheat from the anti-neoliberal chaff, represent very often a useless expenditure of energy. In the current conjuncture it is enough to fight to the end for a just and clearly defined demand, to come directly up against the lines of defence of the system.

[This article first appeared in International Viewpoint, the magazine of the Fourth International. Michel Husson is an economist, in charge of employment at the Institut de recherches economiques et sociales (IRES) in Paris. He is member of the Fondation Copernic, a left-wing think tank, and of the Scientific Council of ATTAC. He has just published Un pur capitalisme, Lausanne 2008, Éditions Page Deux. You can consult his writings on]


These references are available online here.

Artus P. (2002), Karl Marx is back, CDC, Flash n°4, January.

Artus P. (2005), Le capitalisme est en train de s’autodétruire, La Découverte.

Artus P. (2010), Une lecture marxiste de la crise, Natixis, Flash n°2, January.

Barsoc C. (1994), Les rouages du capitalisme, La Brèche.

Bihr A. (2009b), Pour une approche multidimensionnelle des crises de la production capitaliste, juin.

Bihr A. (2009a), A propos d’un « excès de plus-value », March.

Carchedi G. (2010), Zombie Capitalism and the origin of crises, International Socialism 125.

Chesnais F. (2010), Crise de suraccumulation mondiale ouvrant une crise de civilisation, Inprecoe n°556/557.

Ciprut D. (2009), Sur la baisse tendancielle du taux de profit, February.

Foster J.B. et F. Magdoff (2008), Back To The Real Economy Monthly Review, December.

Freeman A. (2009), What makes the US profit rate fall?

Gill L. (2009), A l’origine des crises : surproduction ou sous-consommation ? May.

Gill L. (2010), Les faux pas d’Alain Bihr, les dérives de Michel Husson Carré rouge n°43, March.

Harman C. (2007a), The rate of profit and the world today International Socialism 115.

Harman C. (2007b), Le taux de profit et le monde d’aujourd’hui.

Harman C. (2010a), Not all Marxism is dogmatism: A reply to Michel Husson, International Socialism 125.

Harman C. (2010b), Crise et taux de profit : une réponse à Michel Husson, Inprecor 556/557.

Husson M. (1996), Misère du Capital, Syros.

Husson M. (1999), Après l’âge d’or. Sur le Troisième Age du capitalisme in G. Achcar (dir.) Le marxisme d’Ernest Mandel, PUF.

Husson M. (2008a), La hausse tendancielle du taux d’exploitation Inprecor 534/535, January-February.

Husson M. (2008b), La tendance à la baisse de l’investissement note hussonet n°1, 18 September.

Husson M. (2009a), Le dogmatisme n’est pas un marxisme A l’encontre, June.

Husson M. (2009b), Etats-Unis: la baisse de la profitabilité a précédé la crise financière, note hussonet n°8, September.

Husson M. (2009c), Les coûts historiques d’Andrew Kliman, December.

Husson M. (2009d), Taux de profit : et pourtant il monte ! note hussonet n° 9, décembre.

Husson M. (2010a), La hausse tendancielle du taux de profit, January.

Husson M. (2010b), Le profit et la crise, January.

Husson M. (2010c), La crise et les marxistes, Regards, February.

Husson M. (2010d), Le partage de la valeur ajoutée en Europe, to appear in La Revue de l’Ires.

Kaldor N. (1961), Capital Accumulation and Economic Growth in Lutz F.A. and Hague D.C. Eds., The Theory of Capital, Macmilllan.

Kliman A. (2009a), “The Destruction of Capital” and the Current Economic Crisis 1st Draft, January 15.

Kliman A. (2009b), The Persistent Fall in Profitability Underlying the Current Crisis 2nd (incomplete) draft, October 17 data.

Kliman A. (2009c), La crisis económica, sus raíces y perspectivas, Conferencia, December 15, debate.

Kliman A. (2010), Masters of Words. A reply to Michel Husson on the character of the latest economic crisis February 19.

Johsua I. (2009), La crise, entre sous-consommation et sous-accumulation Tout est à nous! Revue n°4, October 2009.

Louçã F. (2003), Ernest Mandel et la pulsation de l’histoire.

Mandel E. (1982), L’explication des crises capitalistes.

Mandel E. (1986), Traité d’économie marxiste, Christian Bourgois.

Mandel E. (1995), Long Waves of Capitalist Development. A Marxist Interpretation, Verso.

Marx K. (1867), Capital, Volume 3.

Nivat D. and Terrien B. (2010), Application de la règle du principe directionnel étendu aux statistiques d’investissements directs, in Lionel Fontagné and Farid Toubal, Investissement direct étranger et performances des entreprises, CAE, March 2010.

Permanent Revolution (2009a), Kliman and Choonara review 2009.

Permanent Revolution (2009b), Joseph Choonara Marxist accounts of the current crisis , 2009.

Roelandts M. (2009), Le cadre méthodologique de la théorie des crises chez Marx et sa validation empirique July 2009.

Shaikh A. (1978), An Introduction to the History of Crisis Theories.

Stockhammer E. (2006), Shareholder value orientation and the investment-profit puzzle Journal of Post Keynesian Economics, vol.28 n° 2, 2006.

du Tertre R. and Guy Y. (2009), Les traits stylisés des grandes entreprises cotées en France à l’ère du capitalisme financier, La Revue de l’Ires n° 62, 2009/3.

Trotsky L. (1923), La courbe du développement capitaliste.


[1] Being myself a participant in this debate -- which justifies (for once) the use of the first person -- I cannot pretend to a perfect impartiality.

[2] The data base of the European Commission that I used previously contained errors which falsified the résults – above all before 1980 – and I have abandoned it in favour of national sources (see Husson 2010a).

[3] For a more “technical” discussion see La hausse tendancielle du taux de profit (Husson 2010a).

[4] In the USA, these data are produced by the Flow of Funds Accounts of the Federal Reserve Bank.

[5] For a more detailed demonstration, see La hausse tendancielle du taux de profit, article cited, 2nd part.

[6] For an explanation of great clarity which incorporates the notions of multiplier and accelerator within a Marxist logic, see chapter 11 of Marxist Economic Theory by Ernest Mandel.

[7] Karl Marx, Capital, Volume 3, Chapter 30.

Powered by Drupal - Design by Artinet