Not all Marxism is dogmatism: A reply to Michel Husson

Posted: 19 October 09

Chris Harman

Michel Husson has criticised a number of Marxist economists, including myself, in the most stringent manner.1 He writes:

The crisis has given rise in recent months to a series of contributions characterised by a counterproductive and discouraging dogmatism… What these contributions have in common are references to the orthodox interpretation of the law of the tendency of the rate of profit to fall.2

This, he claims, ignores the elementary fact that “there has been a very clear tendency for the rate of profit to rise in the principle capitalist countries. This evolution is sufficiently pronounced that one cannot expect it to be affected significantly by more or less appropriate correctives.”

He then provides a graph based on figures provided by the European Union for the US, Europe and the G8 that purport to justify this argument.

But there are problems with these figures. First, he does not explain where the European Union got its figures or detail the methodology used to construct the rate of profit from them. (All he says is that they “relate the next excess from exploitation to the stock of capital”.) Without such information it is impossible for those of us accused of “dogmatism” to subject his argument to scientific scrutiny.

Second, his graphs differ markedly from numerous other attempts to calculate the path of the rate of profit. This is most clearly the case with the US. Husson claims that in the decade 1998-2008 it not only recovered from a fall in the years 1968-82 of about a quarter, but rose to nearly about 30 percent higher than its pre-1973 level.3

By contrast, Robert Brenner, Fred Moseley, Simon Mohun, Alan Freeman and Andrew Kliman have each provided figures which, although not identical, each show a genereal pattern that differs from Michel Husson’s.4 On the one hand, they show a bigger decline up to 1982 than does Husson; on the other hand, they show the recovery since to have been much more limited than Husson—not even reaching, let alone going 30 percent above the figure for the late 1960s. Moseley, who shows the biggest recovery of profit rates in recent years, provides a pattern that nowhere approaches Husson’s figures, while Brenner, Mohun, Freeman and Kliman all show recent profitability to have been substantially lower than during the 1960s. These figures are all based upon checkable statistics provided by the US Bureau of Economic Analysis in its NIPA tables.

Arnaud Sylvain only provides figures up to 2000 but his graph certainly does not show the massive rise above the figures for the 1960s that Husson claims. Rather, it shows their peak in the late 1990s as being only about the level of 1973 and well below the level of the mid- to late 1960s.5 Dumenil and Levy’s figures up to 1997 are similarly different to Husson’s.6 Figures by Goldman Sachs only start in the 1980s and so do not show the decline in profits from the late 1960s. But the pattern of the recovery of US profits is again different to Husson’s, with the level in 1997 and 2007 only about 10 percent higher than in 1988.

The same problems with Husson’s US figures are to be found with calculations for other countries. Brenner and Arnaud Sylvain show a long-term fall in the rate of profit for Japan, as do Arthur Alexander7 and Fumio Hayashi & Edward C Prescott.8 Mehmet Ufuk Tutan has carried out a meticulous study of profit rates for Germany. It only goes up to 1987 but shows a much less marked recovery than would be implied by Husson’s graph for the four biggest European economies.9 I have come across three studies of profit rates for China. One shows a sharp drop up of more than a third between 1978 and 2000;10 the second shows an even sharper drop of around 40 percent from the 1980s until 2003;11 the third shows a fall for manufacturing until 1999 but considerable rise after that.12

All these figures, it should be noted, are subject to enormous qualifications on at least two fronts.

First, there are problems deciding on how to measure capital investment—and then on finding the appropriate figures. Firms calculate the profitability of any investment by adding together the initial capital outlay on structures and equipment, adding the annual expenditure on raw materials, components and wages, and then dividing their net profits by this total. Ie they divide their profit over a number of years by what they laid out in investment over those years. So with “conventional accounting procedures… The value of the capital stock and of capital consumption are measured at historic cost”.13

But aggregating the different investments made at different times over a particular period is a necessarily complicated procedure, and most attempts to measure national rates of profit use a different procedure—that of current cost accounting. The profit made in a given year is measured against the market value (ie the replacement cost) of the structures and equipment used. This necessarily leads to a distortion in the figures, since any increase in productivity since the investment was made will mean that its current value less than what was laid out on it (even if the market price increases due to inflation). The rate of profit will appear higher than it actually was. The greater the speed of technological innovation, the greater the discrepancy will be. This is very important in recent years, given the rapid increase in computer-related productivity.14 One would expect current cost calculations of the rate of profit to show recent profits relatively high compared with those previously.

A second sort of distortion can affect the figures for historic costs calculations of the rate of profit during periods of rising prices of stocks of goods held by companies. Profitability can be magnified by the apparent increase in the value of these holdings as a result of inflation.15

Andrew Kliman has attempted to provide figures that eliminate both sorts of distortion using historic cost calculations that attempt to remove the effects of inflation.16 The overall pattern of his graphs is not completely dissimilar to that of Brenner, Mohun and Moseley—except that they show a much small recovery of profitability from the low point of the early 1980s.

A third distortion to profitability figures may come from asset bubbles. These lead to firms asserting that the value of the assets they hold has risen by more than is justified by any increase in real value production (this shown clearly in the US Flow of Funds accounts)—and in the case of financial transactions the increase can be recorded as part of the net increase in national output. Since profits in the national economy are usually calculated by deducting wage costs from this net national product, its increase leads to profits seeming to increase more than they actually have. The shock of the financial crash of the last two years is now leading some bourgeois economic commentators to recognise that there were “fictitious profits”—and with them “fictitious economic growth”—in the mid-2000s, if not earlier. Most calculations of profitability try to circumvent this problem by restricting themselves to non-financial corporations (or, sometimes, the non-financial business sector). But many major non-financial corporations such as General Electric (the US’s biggest manufacturing corporation), Ford and General Motors became increasingly dependent on financial operations from the 1990s onwards.

The Financial Times has reported that Andrew Smithers, the economic analyst, “has argued for some time that this is seriously distorted by inflation of asset values. In 2007, he calculates, changes in real estate values alone added a daunting $1,000 billion plus to non-financial corporate balance sheets”,17; Smithers calculates that asset appreciation of this sort “accounted for a rather staggering 22 percent of total earnings from non-financial corporations” in the past decade.18 General Electric was fined several million dollars in August for falsely inflating its profit figures

There is every reason to think that Husson’s figures will reflect all these distortions. The European site he refers to certainly estimates the value of investment undertaken by firms in terms of its replacement cost, not its historic cost. It also seems likely that the figures are for the whole economy, including all of the bubble-inflated profits for the financial sector. But without a detailed account of his methodology and sources it is difficult to be more precise.

Faulty understanding of the theory

It is not only Husson’s figures that are questionable, so is his attempt to explain them in Marxist terms. He writes of Marx’s “tendency of the rate of profit to fall”:

There is no a priori reason to think that the tendency systematically overcomes the countertendency. The productivity of labour is able to compensate, in a perfectly symmetrical manner, for rises in real wages and the increase in physical capital.

The accumulation of means of production does not have to mean, he insists, an increase in the organic composition of capital:

The increase in the productivity of labour permits a reduction in the costs of machines. And this countertendency can compensate for the increase in the number of machines so that the evolution of the organic composition is indeterminate.

Under such circumstances an increase in the rate of exploitation of each worker can lead to a rising rate of profit. And this, he argues from his figures, is what has been happening over the past quarter of a century: “The numerator and denominator of the rate of profit can remain constant, and, consequently, the rate of profit.”

There is, however, a gap in his reasoning. He ignores an important point which has been made by various Marxists in controversies over the rate of profit in phe last 40 years.19 The controversy was aroused by the theorem of the Japanese Marxist Okishio, who argued that because capitalists would not introduce new technology unless it raised the rate of profit, rising capital investment could not reduce the rate of profit. Its effect would in fact be to raise productivity and so reduce the cost of new investment, producing a general rise in profitability. The only thing that could then reduce profitability would be a fall in the rate of exploitation (ie rise in the share of output going to workers).

There is a simply and conclusive counterargument, which has been put in different ways by Robin Murray,20 Ben Fine and Lawrence Harris,21 Guglielmo Carchedi,22 Alan Freeman, Andrew Kliman23 and myself.24 It is that the effect of increased productivity in reducing the cost of future investments does not help individual capitalists profit from existing investment. As the saying goes, “You cannot build the houses of today with the bricks of tomorrow.” The fact that new machines will cost less to buy in a year or two’s time does not somehow reduce the amount you have already spent on your existing ones. In fact, the more rapidly technological innovation takes place and productivity rises, the more rapidly the machines suffer from “moral depreciation” and become obsolete. There is increased pressure on profitability as a result, not reduced pressure.

There is only one way this can lead to reduced downward pressure on profit rates. That is if the cost of existing investments can somehow how be removed from the book (if capitals are devalorised). But the individual capitalist cannot simply shrug off what they have spent already. If they do so they face a reduction in profitability whatever way they look at it.

As I put it in my latest book, Zombie Capitalism:

Investment…takes place at one point in time. The cheapening of further investment as a result of improved production techniques occurs at a later point in time. The two things are not simultaneous… When capitalists measure their rates of profit they are comparing the surplus value they get from running plant and machinery with what they spent on acquiring it at some point in the past—not what it would cost to replace it today… [The rate of profit] necessarily implies a comparison of current surplus value with the prior capitalist investment from which it flows. The very notion of “self-expanding values” is incoherent without it.25

This means there is only one way in which the falling cost of new investment can overcome problem of profitability. It is if some capitals bear the losses due to devalorisation and are driven out of business while others benefit by buying up their structures, equipment and raw materials at less than their value. That is, the crisis, by creating conditions under which some capitals cannibalise others, provides the conditions under which falling costs of new investment can serve to counter the long terms downward pressure on profit rates.

However, one important empirical feature of economic crises of the past 40 years has been a relatively low level of business failure. The concentration and centralisation of capital means that the biggest firms are able to protect their less profitably divisions from going bust—a point already made by Preobrazhensky writing in 1931.26 And fear by other capitals and states of the damage which would follow if the big really firms themselves simply went bust has led in each crisis in recent decades to states stepping to prevent that happening—what mainstream economic commentators now call the “too big to fail” problem.

A study of bankruptcies in the US concludes that there were rare indeed until the 1990s. There were more in the short crisis of 2000-2 (Enron and WorldCom were the best known) but the turn to state bailouts since the collapse of Lehman Brothers others—including the massive involvement of states in preventing a simple collapse of General Motors and Chrysler—shows the limits on the devalorisation of capital through crisis.

It is this which explains the dilemma of neoliberal economists faced with the crisis. A pure Hayekian or New Classical position would mean allowing many of the giant firms to go bust so as to provide a lease of life to the rest. But the reality is that the concentration and centralisation of capital has reached a point where the different elements of the system are so intertangled that unprofitable firms going bust can damage, rather than help, the remaining profitable ones.

The real roots of the crisis

Such reasoning enables us to see that roots of the crisis do in fact lie in downward pressure on profitability since the end of the 1960s. The attempts to deal with this have included all the means referred to by Husson—attacks on wages, the social wage, working conditions and so on—which increase the rate of exploitation. A variety of different sources show an increase if the “share of capital” as opposed to wages for all the major capitalist countries. But in the absence of massive bankruptcies of the giants firms, this has not been enough to restore the rate of profit to its old level. The result has been a long-term slowdown in the rate of productive accumulation, even when the rapid accumulation taking place in China is taken into account.

In passing, it should be said that this downward trend in accumulation has one side effect—the reduced tempo of accumulation can, at least at time, reduce the upward pressure on the organic composition of capital.27

But the most important effect of decreased accumulation at a time of an increasing rate of exploitation is to open up a gap between the capacity of the system to produce goods and the capacity of the market to absorb them. This “overproduction” is not a result of “under-consumption” as such, but of the failure of accumulation to take place on the scale necessary to replace lost consumer demand by an increased demand for investment goods.

The expansion of finance has taken place against this background. On one side it has constituted an attempt by capitalists to obtain rates of profit higher than they can from productive investment. This can work for some individual capitalists, but not for the system as a whole, since ultimately surplus value comes from productive investment. On the other side it amounted to workers and members of the middle class being granted loans to buy things, which created a short-term increase in demand for the goods that could not have been sold otherwise. This “privatised Keynesianism” was positively encouraged by Alan Greenspan as head of the US Federal Reserve in the aftermath of the collapse of the telecoms-dotcom bubble of the late 1990s and the panic at the time of 9/11. But this too could not work beyond a certain point because workers could only have afforded interest payments on the loans sufficient to provide a continuing high level of profits to the finance sector if wages had been raised—which would then have reduced the rate of profit through the system as a whole. As even bourgeois economic commentators such as Martin Wolf have recognised, the loans provided the means by which an important section of the US population became “the consumer of last resort” for the rest of the world system—and especially for Germany, China, and, via Chinese demand, for Japan, the rest of East Asia and the Latin America.

The crisis can then be seen as product of the “law of the tendency of the rate of profit to fall and its countervailing tendencies”—providing the countervailing tendencies are seen as limited in their effects, raising the rate of profit from the very low level of the early 1980s but not sufficiently to boost accumulation on the scale needed to absorb everything produced by the system.

This resolves a mystery in Husson’s picture—showing why his supposedly high level of the rate of profit (higher than in the long boom of the “glorious 30 years”) has not led to a level of productive investment sufficient to pull the whole global economy forward. Instead, for three decades we have had spells of accumulation in particular parts of the world system for a number of years at a time (sometimes sharp spells, as in Brazil in the late 1970s, China over the past decade or the US in the mid-1990s), but not sustained global boom. His explanation seems to be that there are political forces preventing the system undertaking the Keynesian measures that would be good for it, and instead leading to a diversion of accumulated money capital from productive investment into finance. It makes more sense to see the flow of money capital into finance as a response to productive profit rates that are deemed to be too low—something which is confirmed by all the calculations of profit rates except for the ones Husson supplies.

It should be added that Husson’s own figures leaves another question unanswered. What happens to the money capital that goes into finance? For only a portion of the money capital that flows into finance can stay there since finance (by definition) is not mainly concerted with converting money capital into commodities. Some goes on building financial offices and on salaries, but most flows back into the rest of the system. The financial system is network of pipes connecting different parts of the system that do convert money capital into commodities—or at best, a sink, where money capital can only absorb a certain amount of money capital without overflowing. The housing boom will have converted some of the money capital into commodities, but not as much as might be thought since most of its was absorbed by rising house prices not by rising house construction (this was particularly clear in Britain, where house prices quadrupled in 12 years, while housing construction remained at a historically low figure). In fact, the figures suggest, as least of the US, that it was only in the mid-2000s that finance did not eventually flow back into productive investment. In the mid- to late 1990s there was net borrowing by industrial corporations. The problem was that even in this period, capital globally did not feel confident enough about profitability to undertake investment on a scale necessary produce a sustained boom.


1: This paper was written for a recent conference in Amsterdam organised by the Fourth International.

2: Michel Husson, “Le dogmatisme n’est pas marxism”,

3: The figures I give here are based on Husson’s graph, since he does not provide other statistical detail.

4: The relevant graphs are shown at the end of this article.

5: Arnaud Sylvain, “Rentabilité et profitabilité du Capital: Le cas de six pays industrialisés”.

6: Gérard Duménil and Dominique Lévy, “The Profit Rate: Where and How Much Did it Fall? Did It Recover? (USA 1948-1997)”.

7: Arthur Alexander, “Japan in the Context of Asia”.

8: Fumio Hayashi & Edward C Prescott, “The 1990s in Japan: A lost decade”, September 2001.

9: Mehmet Ufuk Tutan, The Falling Rate of Profit in German Industry.

10: Phillip Anthony O’Hara, “A Chinese Social Structure of Accumulation for Capitalist Long-Wave Upswing?”, Review of Radical Political Economics, 38 (2006), pp397-404.

11: Jesus Felipe, Edith Lavina and Emma Xiaoqin, Fan, “Diverging Patterns of Profitability, Investment and, Growth in China and India during 1980-2003”, World Development 36:5 (2008), p748.

12: Zhang Yu and Zhao Feng, “The Rate of Surplus Value, the Composition of Capital, and the Rate of Profit in the Chinese Manufacturing Industry: 1978-2005” , paper presented at the Second Annual Conference of the International Forum on the Comparative Political Economy of Globalization, 1-3 September 2006, Renmin University of China, Beijing, China.

13: Bank of England Quarterly, 1975.

14: See, for instance, Stacey Tevlin & Karl Whelan, “Explaining the Investment Boom of the 1990s”, March 2000.

15: Bank of England Quarterly 1975.

16: He does so using normal mainstream inflation figures and his own approach based on the “monetary equivalent of labour time” (MELT).

17: Financial Times, 26 November 2008.

18: Tony Jackson, Financial Times, 30 March 2009.

19: In English, the controversies first developed in Bulletin of the Conference of Socialist Economists in the early 1970s, with contributions on different sides from Glyn, Murray, Himmelweit, Rowthorn, Armstrong, Fine and others.

20: Robin Murray, Bulletin of Conference of Socialist Economists 1973.

21: Ben Fine and Lawrence Harris, Rereading Capital (Macmillan 1979).

22: Guglielmo Carchedi, Frontiers of political economy (Verso, 1991).

23: Andrew Kliman, Reclaiming Marx’s Capital (Lexington Books, 2007).

24: Chris Harman, Explaining the Crisis (Bookmarks, 1984), pp20-34; Chris Harman, Zombie Capitalism (Bookmarks 2009), p 68-75.

25: Zombie Capitalism, pp74-75. The first version of this argument I have come across was that put by Roger Murray, using a corn model of production, in the Bulletin of Socialist Economists.

26: Evgeny Preobrazhensky, The decline of capitalism (ME Sharp, 1985), p137.

27: Figure for capital-output ratios and for capital-labour rations suggest this has been the case at various times since the mid-1970s.