We are publishing two old documents on the tendency of the rate of profit to fall, which were part of a debate in the early 1980s within the then Militant Tendency. Marx had predicted a tendency for the rate of profit to fall as a result of what he called the rising organic composition of capital. In his document, AG disagreed with Marx. He saw the fall in profits as being fundamentally caused by rising real wages biting into the surplus. This had serious implications as it led him to cast doubts on Marx’s theory. MB’s reply is mainly concerned with pointing out that Marx’s theory was fundamentally correct and that it is still a useful guide to understanding reality.
The In Defence of Marxism website has taken the decision to reprint a debate that took place nearly a quarter of a century ago. The discussion took place within the Militant, which at that time was a significant Marxist tendency rooted within the Labour Party and trade unions. Militant has since imploded, but the British journal Socialist Appeal traces its political and intellectual history back to the tradition of Militant.
The general reader may find this material difficult at first. It is not easy reading, but we believe it is worth persevering with. It might be useful to check out the article Rate of Profit and Capitalist Crisis (a critique of Brenner) on our website first. It pursues some of the same themes. There the basis of Marx’s theory, and its critique, are presented more simply.
In summary: AG presented evidence that there had been a fall in the rate of profit, which all parties to the discussion agreed on. Marx had predicted a tendency for the rate of profit to fall as a result of what he called the rising organic composition of capital. By this term he meant that the proportion of dead to living labour in the production process tends to rise under capitalism. In striving to raise the productivity of labour, capitalists introduce more and more machinery and materials behind the elbow of each worker. For Marx this is constant capital, which passes its value unchanged to the final product. Only the living labour of the worker produces new value. Yet this living labour shrinks relative to total capital outlaid. Hence the tendency for the rate of profit to fall. How this tendency works itself out in the economy is looked at in the reply to AG.
AG disagreed with Marx. He saw the fall in profits as being fundamentally caused by rising real wages biting into the surplus (conventionally divided into rent, interest and profit in Marxist analysis). He cast doubts on Marx’s theory. MB’s reply is mainly concerned with discussing whether Marx’s theory was correct and a useful guide to understanding reality. The problems in AG’s counter-explanation are dealt with more briefly at the end of the reply.
In 1981 AG submitted his document for discussion in a special bulletin, as was his right and as part of the democratic tradition of genuine Marxism. However he took a great deal for granted in his arguments, and his presentation dictated the form of debate.
For instance, in AG’s Appendix II he argues the case originally outlined in Okishio’s theorem ("willingly introduced techniques can never cause the rate of profit to fall”) without mentioning it by name. Readers should look at the more recent Brenner document (Rate of Profit and Capitalist Crisis) for a critique.
The document is only a small part of a voluminous debate on the rate of profit that was then taking place – and continues to this day.
Many readers may be unfamiliar with the mathematical formulae which spatter the material. The profit share is referred to in the debate. This is different from the rate of profit, and easier to estimate from national income statistics. First you work out what part of output may be regarded as surplus value. This is not just the declared profit of corporations, of course. Shaikh and Tonak have written a very good book (Measuring the wealth of nations, Cambridge 1994) which identifies, for instance, the income of bond dealers and corporate lawyers as surplus value diverted into unproductive expenditures. Once the Marxist has identified what categories of national income are really surplus, you just divide it by output as a whole to find the profit share. So the formula for the profit share is P/Y where P is profit and Y is national income. (National income and national output is the same thing).
But this is not the same as the rate of profit, which is calculated on the total capital invested – the formula for which is P/K. To find that we have to use another figure from national income accounting - the capital/output ratio. This is presented as K/Y. Dividing the profit share by the capital/output ratio gets you the rate of profit. P/K = P/Y x Y/K.
Perhaps this distinction is not made clearly enough in the original contributions. The rate of profit is calculated on the total capital invested. It is true that individual items of capital stock are devalorised. This is because, as productivity rises throughout the economy, less labour time is involved in producing them. But, as is explained in the critique of Brenner, this feedback is not an instantaneous result. AG is in effect treating the economy like an equation on a piece of paper.
Some of MB’s arithmetical examples do show fixed or circulating constant capital instantaneously devalorised due to advances in productivity. This is the assumption most favourable to AG’s case. The arithmetical examples are nevertheless intended to show that indefinitely staving off a falling rate of profit through the action of counteracting factors is unlikely. Of course AG is quite right to argue that arithmetical examples can’t actually prove anything.
What is the point of all this? How does it measure up against the reality? Robert Brenner has convincingly demonstrated that the decline in the rate of profit has been central to the economic slowdown and the eruption of crisis in the world economy – that movements in the rate of profit have been the heartbeat of the post war world economy.
"The radical decline in the profit rate has been the basic cause of the parallel, major decline in the rate of growth in investment, and with it the growth of output, especially in manufacturing, over the same period. The sharp decline in the growth of investment – along with that of output itself – is, I shall argue, the primary source of the decline in the rate of productivity, as well as a major determinant of the increase of unemployment.” (Brenner – The economics of global turbulence New Left review 229, 1998)
Of course there is some way from demonstrating the movement of the rate of profit as posing the abstract possibility of crisis to showing the actual course of the development of the capitalist economy over time. "The real crisis can only be deduced from the real movement of capitalist production, competition and credit.” (Marx, Theories of surplus value Vol. 2 p. 512)
The fall in the rate of profit is a fact. So the question is - why? AG was one of the outstanding representatives of the "wage squeeze” theorists. This was the dominant trend among academic Marxists at the time explaining the economic crisis from the1960s on as wages squeezing profits. The reader is referred to the conclusive critique from Brenner from hindsight on their explanation of events. Basically their argument was that the profit share was threatened by the rise in real wages. This in turn was happening because of continued full employment in the advanced capitalist countries. In effect the demand for labour outstripped supply, causing its price to rise.
MB’s reply, defending the traditional Marxist position, may look inconclusive against AG’s impressive marshalling of statistics. AG’s case is securely sandbagged behind Table 1!
But it is worth underlining that his statistical analysis is based on an incorrect method. The nub of the argument is: has the organic composition of capital increased? In Table 1 AG purports to show a big increase in the organic composition of capital. (Actually he divides fixed constant capital by the number of workers rather than the wage bill.) He argues, however, that this is the technical composition of capital. As the reply shows, this is quite wrong. The technical composition is an indicator of the mass of capital (the quantity of use values) and cannot be measured in money. AG then subtracts productivity increase to show that what he calls the organic composition of capital has shown no definite drift upwards. But he has presupposed what he had to prove! Both MB’s reply and the critique of Brenner point out that capital is not "uniformly and instantaneously devalorised” (to use Marx’s term) by the rate of productivity increase. AG does this to allow for the fact that the equipment can now be produced more productively than at the time it was made, but we believe his method to be wrong.
The other arguments made originally by Marx and later repeated in the reply are also valid. While units of capital get cheaper, a system of machinery takes its place. The fall in the rate of profit is usually accompanied by a rise in the mass of profits – though this cannot go on forever. And even when the mass of profit falls, that will express itself as a greater mass of products. This is because the rise in productivity throughout the economy means the same expenditure of labour time produces a greater mass of use values. And while the counteracting tendencies produced by the operation of the tendency for the rate of profit to fall mean that the law operates upon the economy only as a tendency, they cannot indefinitely override that tendency forever.
Perhaps it comes across as an obsession, but commodity prices were a major issue for the world economy at the time of writing. The document was written two years after a so-called "oil crisis”. The 1974 and 1979 downturns, both of which appeared to be triggered by interruptions in the supply of oil, were the most severe since the Second World War.
Various prophets then were predicting generalized rises in commodity prices as the major threat to capitalism at that time.
And, of course, whereas the accumulation of fixed constant capital is the mainspring of rising productivity under capitalism, the role of circulating constant capital is essentially passive. If productivity doubles, the worker is working up twice as much raw material as before. In other words, the organic composition of capital therefore increases.
And the fact is that the production of most raw materials is not reproducible at will by the capitalist class but is subject to the limits of nature puts limits on raising productivity in the sphere of raw material production. There were widespread fears at the time of a secular rise in the price of oil. In fact commodity prices typically rise at the end of a boom under capitalism,
MB’s critique of the limits of official statistics to illuminate Marxist categories might be seen as scattergun. But not all the points made are a direct critique of AG. It is rather a general setting out of the problems Marxists encounter in trying to apply our analysis to "the facts”. AG is invariably scrupulous in his interpretation of statistics, though we believe his method has proved faulty on occasion – such as his "double counting” on Table 1 referred to above.
In the twenty years since the document appeared, great strides have been made in "quantitative” Marxist analysis. Apart from the work of Shaikh and Tonak mentioned earlier, Moseley’s book has attempted to look at what is happening to the organic composition of capital and the rate of profit. (The falling rate of profit in the postwar United States economy, Macmillan 1991). Using a different method from AG, and basing his study over a different time frame, he comes to quite different conclusions. He records that the organic composition of capital has risen and that it has been partly responsible for the fall in the rate of profit that we all agree has occurred. But he also explores another dimension not raised in writing at this stage of the debate.
The difference between productive and unproductive is crucial for Marxist economic analysis. By productive labour Marx means labour that produces surplus value, whether or not it produces material products. So a programmer who writes a computer programme or a chef who cooks a meal can be regarded as productive, providing they perform their tasks in order to produce surplus value for the boss.
What happens to this surplus value? Part is accumulated while the rest is spent on unproductive expenditures ("luxuries”). Workers in luxury industries produce surplus value for their boss, of course. But the unproductive sector, including the workers who make their living there, is subsidised from the surplus value raised in the capitalist economy as a whole.
What sort of unproductive expenditures are there? What has been happening to them since the Second World War? Moseley argues that the runaway rise in unproductive expenditures has been the second great factor levering profit rates down. At the time of writing, for instance, there are four million workers in the shop and distribution sector in Britain compared with 3½ million employed in industry.
Moseley sums up his results for the USA: "Commercial labour…accounted for almost two thirds of the total increase of unproductive labour. The other two types of unproductive labour, financial labour and supervisory labour, each accounted for approximately half the remaining increase of unproductive labour.” (Moseley p. 150-151). Let’s take a look at what he regards as the most important unproductive sectors for capitalism as a whole:
Supervisory labour (supervision of inventories – stocks – and of labour)
‘The ratio of supervisory labour industries increased 86% from…1950 to…1980.’ (Moseley p.139)
Circulation labour. This in turn can be divided into
- Commercial circulation
"Commercial labour increased 134% from 1950 to 1980, while productive labour increased only 44%, thus resulting in a 63% increase in the ratio of commercial trade labour to productive labour.” (Moseley p.127)
- Financial circulation
"Financial labour” (in the Finance, Insurance and Real Estate industry) "increased 173% from 1950 to 1989, while productive labour increased only 44%, so the ratio of financial labour to productive labour increased 91%”. (Moseley p.133-134)
Moseley argues, and we agree, that the present epoch of capitalism is characterised by a huge increase in unproductive expenditure. He finds this a dead weight on the system, acting as an involuntary levy upon the profits of individual capitalists. And he shows unproductive spending exploding out of control over the past fifty years. He assesses this effect as significant as the rise in the organic composition of capitalism in causing the secular decline of profits since the War.
How will the distinction between productive and unproductive labour affect our analysis of the economy? National income accounting typically lumps all workers together. For bourgeois economics, there is no notion of productive and unproductive labour. But Marxists argue that some are producing surplus value, while the unproductive are being supported from a part of surplus value (that does not mean that they don’t work hard or are "parasites”, of course). Economists such as AG, who use aggregate consumption figures from employment to work out wages share, are likely to get a completely misconceived picture of wages biting into profits, when the reality is unproductive spending cutting into profits.
MB began to open up a second front on AG’s position by publishing a paper on "Productive and unproductive labour” to explore similar territory to Moseley, but by this time AG had abandoned the debate.
Though this is archive material, we believe it will help today’s generation to understand capitalism, a necessary step if we wish to overthrow it.
(October 2003)
Go to the two documents: A critique by AG and A reply by MB