MIA  >  Archive  >  Harman  >  Explaining the Crisis

Chris Harman

Explaining the Crisis

* * *

Appendices:
Other theories of crisis

THE BASIS FOR the argument set out in this book, taking Marx’s theory of crisis and extending it to explain the unprecedented boom of the 1940s to the 1970s, was developed twenty years ago by members of the International Socialists, forerunner of the present Socialist Workers Party. The argument had considerable predictive power. As early as 1960–1 we were able to predict that ‘by the early 1970s’ the world economy would be falling back into the pre-1939 pattern of crisis.

But our argument did not attain any significant hearing outside our own ranks. As the new crisis came to be taken for granted in the late 1960s and early 1970s – and as a whole new generation of intellectuals became interested in Marxism – our theory was peripheral to most of the discussions. A range of quite different explanations of the crisis appeared in Marxist publications.

* * *

Appendix one:
Wages as the cause of the crisis

THE SIMPLEST EXPLANATIONS for the onset of crisis blame rising wages for cutting into profits. In Britain the best-known version of the theory comes not from Tory politicians, but was put forward by two left-wing Oxford economists, Andrew Glyn and Bob Sutcliffe, back in 1972. [1] Their explanation concentrated on British figures, but implied that the argument was of wider, international, relevance.

British capitalism has suffered such a dramatic decline in profitability that it is now literally fighting for survival. This crisis has developed because mounting demands from the working class for a faster growth in living standards have coincided with growing competition between capitalist countries. This competition has prevented British capitalism from simply accommodating successful wage demands by pushing up prices ... And it has intensified because the other rich capitalist countries have been subject to the same pressures from the working class as British capitalism ... [2]

Bob Rowthorn, a Cambridge economist belonging to the British Communist Party, argued very much the same case, pointing to pressure from wages as one of the causes of the crisis in the USA: the rate of profit fell ‘because the share of output going to profits fell.’ Indeed for Rowthorn this ‘accounts for the entire reduction of US profitability between 1965 and 1970.’ [3]

The notion of a dramatic reversal in the balance of forces between labour and capital, leading to a squeeze on profits, also plays a role in the explanation of the crisis put forward by the Belgian Marxist economist, Ernest Mandel. He has argued that in the late 1960s internationally, the pool of unemployed workers – what Marx called ‘the reserve army of labour’, because they were always there in reserve, to be called into production if needed – this pool had been used up by capitalist expansion. The resulting competition among capitalists for scarce labour, he said, pushed up wages, so cutting into profits. The ‘limits of the reserve army’ were reached, ‘and a pronounced increase in real wages started to roll back the rate of surplus value.’ [4]

There are, however, decisive counter-arguments against the claim that wages have caused the crisis.

Firstly, the argument just does not fit the facts for at least two of the western economies – including the giant US economy. [5] The sources usually given to justify the argument about wages and the crisis are those of Glyn and Sutcliffe for Britain and Nordhaus for the USA. [6] Nordhaus’ figures are, for instance, quoted as authoritative by Mandel and Rowthorn. Yet they are open to considerable doubts. [7]

The radical economist Perlo has given a quite different interpretation to Nordhaus’ data. Giving figures which probably exaggerate the level of profits [8], he suggests that the ‘share of profits’ in the US rose from 19.5 per cent in 1946 to 22.1 per cent in 1974. [9] A more recent Brookings Papers analysis of US profits by Feldstein and Summers also casts great doubts on the figures given by Nordhaus. This suggests that the fall in US profits is much less than Nordhaus claimed. [10]

It can be seen that no decline in the rate of profit, on any measure, takes place until after 1970.

Glyn and Sutcliffe’s figures for Britain claim that the share of ‘wages and salaries’ rose from 73.3 per cent to 78.8 per cent between 1955 and 1970, and that there was a corresponding decline in profit’s share. But the figures are subject to many of the same criticisms as Nordhaus’s. They assume that none of the cost of financing stock appreciation is borne by company borrowing. And they give figures before tax. As Glyn himself has admitted in a later article, the share of wages and salaries does not rise, but sinks, from 55.6 per cent in 1955 to 50.2 per cent in 1970, once direct and indirect taxes are taken into account. [11]

C.J. Burgess and A.J. Webb have shown that if you take the share of national income in Britain going to all companies after tax, making deductions for stock appreciation and capital consumption, and adding on the government hand-outs to industry, you find that there is no decline in the 1950s and 1960s. [12]

But the claim that ‘a growing share going to wages’ has provoked the crisis falls down, not only on the factual evidence, but for other reasons too. Its proponents also fail to present any adequate explanation of the mechanism which might cause the alleged upsurge of wages at the expense of profits internationally.

An attempted explanation – given, for example, by Rowthorn and Mandel and by the Americans Body and Crotty [13] goes something like this. Economic expansion can take place as long as there is a ‘reserve army of labour’, prepared to enter employment at a relatively low level of wages. But in the 1960s this reserve army dried up, leading to a growth of wages (either because of upward bidding by capitalists seeking labour or by an increase in the bargaining power of unions).

The first difficulty lies in the claim that there was a ‘drying up’ of the reserve army of labour. Expansion of production leads not only to the sucking of workers into production; it also means rises in productivity which drive workers out of some areas of production and into others [14] (witness the rising average level of unemployment in the US in the 1950s and 1960s; and the near doubling of unemployment in Britain through ‘shake out’ during the Wilson government of the late 1960s).

What is more, capitalist expansion in the 1950s and 1960s had the effect of massively increasing in size the worldwide reserve army: in the metropolitan countries conditions were created in which the number of married women prepared to enter the workforce rose massively; at the same time, throughout the world there was a huge migration of workers from the countryside to the towns, so that in nearly all the ‘developing countries’ there were vast pools of labour only too eager to get employment in the ‘developed countries’. There was no sign of these pools drying up in the late 1960s. Look at the vast influx of ‘illegal’ Mexican labour into the US. Look at the growth of the ‘foreign’ population of Germany, doubling from 1.92 million in 1968 to 4.13 million in 1974, until deliberate government action was used during the crisis of 1975 to force about a million of the ‘guest workers’ out of the German economy and back to the ‘third world’.

Rather than the ‘reserve army’ drying up, something quite different happened – the crisis created problems of absorbing the reserve army. In fact, the effect of the crisis has been to extend the reserve army: it has doubled unemployment in the OECD countries and more than doubled it in the ‘developing countries’. If a ‘drying up’ of the reserve army was the cause of the crisis, then the crisis could never have happened, and should now be resolving itself as the reserve army grows still further.

This leads us straight into another difficulty with the ‘wages share’ argument – it just cannot explain why all the western economies moved into crisis at the same point in the mid-1970s. In Italy, in Britain, in Spain and in France there were important improvements in the level of working class organisation in the late 1960s and early 1970s. But there was no similar improvement in the other major economies – Japan, West Germany, and, above all, the US. Unemployment in the US grew once the Vietnam War began to wind down, and the proportion of union members in the US working class has fallen significantly since then, from 22.4 per cent of the labour force in 1965 to 20.1 per cent in 1976. [15] Furthermore, as Perlo has pointed out, for the period before the outbreak of the 1974–5 recession, ‘there was a sharp decline in real wages of non-agricultural workers from late 1972 to spring 1975, while productivity on the whole increased.’ [16]

Finally, the ‘wages share’ argument simply cannot explain the timing of the turning point from expansion to contraction. During the 1950s real wages rose, year after year, in all the major capitalist countries. By the mid-1950s even in the countries which had been devastated by the Second World War they were higher than the pre-war figure. Yet the rate of profit was not squeezed, investment grew, inflation remained at a relatively low level and there was near-full employment everywhere.

Even if it is assumed (despite our previous arguments) that in the late 1960s and early 1970s rising wages did cut the ‘share of profits’ in the output of companies and the national income, why did this happen, when it did not happen in the earlier years of the boom? The explanation can only be that the system could afford the real wage increases then, but could no longer do so. But it follows from this that it is not rising real wages that have caused the crisis, but rather the crisis which has caused real wages to cut into profits. The ‘declining share of profit’ becomes a product of the crisis, not its cause (even if it is a product of the crisis which feeds back into the economy to make the crisis worse).

Theories that purport to explain the crisis in terms of the ‘rising share of wages’ or the ‘declining rate of exploitation’ do no such thing. They force even their adherents to turn to other factors in an effort to overcome the contradictions in their own positions – factors such as government spending (Glyn and, in part, Rowthorn), the changing international structure (Rowthorn) or ‘long waves’ (Mandel).

* * *

Appendix two:
Government spending as the cause of crisis

THE NOTION THAT excessive government spending has been the cause of the crisis is very widespread. On the right it has been presented by the Oxford economists, Bacon and Eltis. [17] On the left their conclusion has been taken up, for instance, by Aglietta, who writes: ‘This analysis uses a terminology other than our own, but is governed by the same perspective’, while both Mattick and Yaffe [18] had argued before Bacon and Eltis that ‘stagflation’ – the historically new combination of stagnation and inflation – was the result of successive Keynesian exercises in deficit budgeting designed to prevent crisis. According to Mattick, efforts by government to bring together ‘labor and idle capital for the production of non-marketable goods’ lead to a ‘constant’ growth in the ‘non-profit sector’ until it ‘outweighs the profitable sector and therewith endangers the latter’s existence’. ‘Deficit financing and government induced production ... must come to an end. The Keynesian solution will stand exposed as a pseudo-solution capable of postponing but not preventing the contradictory course of capital accumulation.’ [19]

But has the ‘public sector’ been ‘squeezing’ the ‘profit sector’? The evidence for such ‘squeezing’ is meagre indeed for the period before the onset of world crisis in the early 1970s.

Bacon and Eltis crudely gave figures for employment in the public and the ‘productive’ (meaning private) sectors. These showed that the former has been growing and the latter falling – and they concluded that the rise in one had caused the fall in the other.

But they virtually ignored the fact that the rise in public sector employment was primarily drawing in people – especially married women – who were previously not in the workforce. At the same time there was a growing pool of unemployed labour which would have welcomed work in the ‘productive’ sector if there had been any available. So in Britain between March 1967 and December 1975, the number of men in employment fell by more than a million; in the same period, the number of women in employment rose by more than 900,000 – and this growth was entirely in the ‘service’ sector. [20]

Even some bourgeois economists have pointed out that in terms of resources, Bacon and Eltis do not distinguish properly between ‘public sector use of resources’ and ‘transfers made by the state between different components of the private sector, which has no direct impact on the output of that sector’ – in other words payments made by the state to individuals and companies in the private sector. [21]

In Britain, ‘In the post-war period there has been no substantial increase in the share of resources which the public sector authorities take for their own direct expenditure.’ [22] Government expenditure on goods and services was 24 per cent of the GNP in the early 1950s and 27 per cent in 1973. Certainly you cannot explain the crisis which broke in Britain in 1974 by invoking ‘excessive’ taxation to feed the allegedly insatiable demands of government: ‘the aggregate tax burden in the UK fell quite sharply between 1964 and 1973 ...’ [23] So it could not have been ‘taxation’ of the ‘productive’ sector to feed the ‘unproductive’ sector that caused the crisis.

This leaves open the possibility that the ‘productive sector’ was hit in another way, by Keynesian ‘deficit financing’, government spending financed by borrowing on the market which drew off the funds otherwise available for private industry. This, essentially, is the argument of Mattick and Yaffe – government borrowing postpones the crisis for a period, only to make it worse when it finally comes.

However, facts again put a huge question mark over any such explanation: real examples of deficit financing have been rare outside war conditions. As Matthews pointed out for Britain during the long boom:

The question is whether the high level of demands that actually occurred was due to government action or whether it was due to other forces, as a result of which government action was not needed ... The hypothesis that it was due to fiscal policy is open to a simple, basic objection. This is that throughout the post-war period, the government, so far from injecting demand into the system, has persistently had a large surplus ... [24]

It is true that there were two periods (prior to the 1974 crisis) when the government embarked upon large capital-spending programmes in order to try to give a ‘Keynesian’ boost to the economy – the Maudling boom of 1962 which was finally ended by Labour in 1966, and the Barber boom of 1972–3 that preceded the crisis of 1974. There was also in the US an important period in which the government resorted to deficits to finance an unpopular upsurge in spending (due to the Vietnam war) – in 1965–8.

But these hardly explain the elements of crisis to which these upsurges in spending themselves were a response. So both the 1971 crisis in Britain and the 1973–4 crisis internationally followed years not marked by budget deficits.

It can be seen from these two sets of figures that apart from Italy and to a lesser extent Britain, the deficits only became truly large once the crisis had broken, after 1974. But government deficits then were inevitable, unless governments were going to increase the depth of the crisis deliberately by huge cuts in their own expenditure.

As Ian Gough has noted: ‘It is not the case that the great expansion of state borrowing since World War II has been financed by government borrowing; instead taxation has grown in parallel.’ He also adds: ‘The recent growth in public sector deficits is a conjunctural phenomenon, following from the present slump not precipitating it. ’ [27]

We have seen that you cannot claim that the state sector has been ‘squeezing’ the private sector – unless somehow you see the private sector as ‘squeezed’ by a transfer of resources between its different sectors. Nor can you claim that government borrowing has destroyed the motor spring of growth in the system. But this still leaves one argument unanswered – the claim that it is the transfer of resources, via the state, between the different parts of the private sector that has brought society to the point of crisis. [28]

The most thorough attempt to do this within a Marxist framework has been by the American economist, James O’Connor (although, as I will argue later, his categorisation is grafted on to an account of the crisis quite different from Marx’s). O’Connor breaks public expenditure down into two sets of components. The first set he calls ‘social capital’. This is spending which is ‘indirectly’ productive for private capital, such as spending on roads or the cheap supply of electrical power to industry, which complements private capital in its drive to extract surplus value.

His second category consists of spending which is completely non-productive, but which is necessary for the social stability of the system. He calls these the ‘social expenses of production’. He includes here the ‘welfare system’ (such as social security payments, but not national insurance benefits, which have to be earned by work), because its function is to ‘keep social peace among employed workers’. [29] He also includes arms expenditure (since its aim is to protect and expand markets, not produce wealth) and payments to ‘non-productive’ groups such as the police and judges.

How have the different sorts of expenditure grown in relation to each other? O’Connor argues that in the US the main burden of social capital expenditure has been borne by local and state governments – which are very much influenced by local capital. The social expenses of the system, by contrast, have tended to be borne by the national capital, the federal government. ‘If federal government has earned the label of the “warfare-welfare state”, local and state governments deserve the name the “productivity state”.’ [30]

How the two sectors have grown in the US is shown by figures given recently [31]:

It would seem that the proportion of gross national product on ‘non-productive’ expenditure, by Federal government, has remained more or less constant since 1952. Indeed, remembering that 1947–9 was exceptional, in that arms spending fell drastically for the only time in the period since the Second World War, then ‘non-productive’ spending would seem to have been constant for even longer. On the other hand, ‘indirectly productive’ expenditure, by local and state governments, has increased steadily.

It is worth noting, however, that within ‘non-productive expenditure’ the proportion spent on ‘warfare’ has tended to decline slowly over the years (with the exception of the Vietnam War years of 1965–8), and the proportion on ‘welfare’ to rise.

O’Connor sees ‘productive’ expenditures as rising in order to sustain the growing needs of industry for a skilled workforce and for a modern ‘infrastructure’. But why do the non-productive, non-military expenditures also rise?

There seem to be two reasons: firstly, demographic changes which mean that there are more old people, and therefore more people likely to be sick but not insured. Secondly, the effect of the crisis itself in producing a growing ‘sub-proletariat’ of permanently unemployed – a sector that made itself felt dramatically with the ghetto uprisings of the mid-1960s.

O’Connor himself does not explain the crisis in terms of growing expenditure on these items. The title of his book is ‘The fiscal crisis of the state’, not ‘The fiscal crisis of the system’. His explanation for the crisis of the system lies elsewhere – in the explanation given by Baran and Sweezy in terms of a ‘growing surplus’ (which is dealt with later). However, the figures from the US do show clearly that there is no simple correlation between increased state spending and the elements of crisis within the system.

In the 1920s and 1930s, government spending was at a very low level and the system was in very deep crisis. In the 1950s it was about 70 per cent higher than before the war, but the system boomed. In the 1970s it was another 20 per cent higher again, but the system no longer boomed and use of industrial capacity fell to 80 per cent in 1969–73 compared with 90 per cent in the 1950s. [32] If government spending only on ‘goods and services’ is measured, and transfers between parts of the private sector excluded, the increase in spending in 1969–73 over 1951–7 is only 12 per cent – and there is a decline in the proportion of GNP going to goods and services after 1974.

What may be more significant is the shift from military expenditure to productive (and non-productive) non-military expenditure – but that is dealt with elsewhere.

Ian Gough has applied some of O’Connor’s arguments to Britain and other countries. He shows that: ‘Excluding the US (where is was boosted by military spending in the 1960s) state consumption rose by 3.9 per cent in real terms in all OECD countries from 1955–69, while GNP rose 5.7 per cent. In other words, real state consumption expenditure has fallen as a share of GNP over the last two decades’, although ‘this is more than accounted for by the decline in military spending, and social transfers have continued to rise ...’ [33] This produced anything like real problems only for those countries, such as Britain, where the rate of growth of GNP was less than the average.

Gough argues that most of the growth of state expenditure was in the ‘indirectly productive’ area:

The single most important conclusion that emerges is that an increasing proportion of the total are productive expenditures, producing inputs for the capitalist sector. The share of the social services, infrastructure and accumulation expenditures is growing, while that of unproductive luxury expenditure is declining. In it wrong, therefore, to regard the growth of the state as an unproductive ‘burden’ upon the capitalist sector; more and more it is a necessary precondition for private capital accumulation. [34]

So the biggest single area of growth of public expenditure in Britain between 1961 and 1973 was educational expenditure, which grew by 2.6 per cent of GNP. This growth was almost entirely in the sections of the educational system catering for the minority of school students who remained in the educational system after the age of 16, and was directly associated with the perception of governments that higher education had to be expanded to cope with the technological needs of British capital. Close behind the growth in educational expenditure came the growth in direct and indirect support for industry – environmental, transport (chiefly motorways) and financial grants. Welfare and social security expenditure did grow as well – but this was very much a product of the growing numbers of unemployed, a product of the crisis of the system, not a cause.

So even if you conclude that the increase in public sector spending is part of the cause of the crisis, this hardly amounts to ‘the public sector squeezing the private sector’. Rather it is that the cost of sustaining the growth of capital constrains that growth itself.

However, as I have argued earlier, it is difficult to see such spending as a cause of crisis, since it happily fitted with the needs of the system in the 1940s, 1950s and 1960s. If it seems like a burden today, it must be because something else in the system has changed. That is why those, like O’Connor and Gough, who have looked closely at the matter, see the ‘fiscal crisis of the state’ as a product of a wider crisis. For O’Connor, this is the crisis of monopoly capital as described by Baran and Sweezy; for Gough it is a crisis arising from the ‘distributional struggle’ over wages. Our explanation is rather different – but we can go along with Gough and O’Connor in rejecting public expenditure as the cause.

* * *

Appendix three:
‘Long wave’ theories of crisis

ONE SET OF theories to gain renewed popularity in recent years are those of ‘long waves’. The Russian Menshevik Kondratieff first developed such a theory in the 1920s in an effort to explain why some periods of capitalist development seem more crisis-prone than others. His ideas were taken up in the 1930s by Schumpeter, and have been ‘rediscovered’ by many bourgeois economists looking for an explanation for the world crisis since 1973. W. Rostow wrote in praise of long wave theory in 1975; one of the best-known Keynesian economists, Paul Samuelson, has said he is now attracted in this direction; and on the left, Ernest Mandel has also been associated with the revival.

Kondratieff claimed to show, on the basis of statistical data for prices, interest rates, stocks and shares, wages, levels of foreign trade and levels of national production, that as well as short-term boom-slump cycles, there were long-term ‘waves’ in economic activity. For a period of 20–25 years, he said, if you averaged out boom and slump years, you would find a rising level of economic activity – with less than average interest rates, rising prices, greater than average increases in physical output and rapidly rising foreign trade. This upward movement would then peak, and would be followed by 20–25 years of downward movement, which in turn would give way to a new upward movement.

Kondratieff calculated what the trend had been over the past 100–150 years for each of his different sets of data, and drew graphs based on deviations from the trend. He claimed that these deviations showed a clear cyclical pattern. [35] The system went through ‘waves’: for one period expanding more rapidly than the average, then more slowly, then more rapidly, and so on.

Kondratieff does seem to have emphasised one correct point: the capitalist system has expanded more rapidly in some periods than others. A modern supporter of his ideas, Ernest Mandel, provides the following figures in justification [36]:

But the sense in which these can be called ‘waves’ is very much open to doubt. Trotsky, who a year before Kondratieff had noted the existence of periods of ‘upturn’ and ‘downturn’ in the ‘curve of capitalist development’, was extremely critical of Kondratieff for labelling these periods ‘waves’. [37] Other Russian Marxists of the time was just as critical, faulting Kondratieff on both statistical and theoretical grounds. [38]

Statistically, they argued that his method was arbitrary in the extreme. Economists such as Oparin, Granovsky, Gerzstein and V. Bogdanov showed that Kondratieff’s ‘waves’ applied only to sets of figures which were dependent on the level of prices, and not to figures for real levels of production. As Granovsky put it: ‘Kondratieff only succeeded in proving that long period changes in price levels have taken place ... Except in price movements, there is no evidence of long waves ... they are the fruit of the imagination of Professor Kondratieff.’ [39] Often during Kondratieff’s ‘downswings’ there was, in reality, a rapid expansion of output – and during his ‘upswings’ there was a slowing-down in the growth of the forces of production.

But this was not all. The Russian economists also insisted that the statistical techniques he used to get his wave-like configuration for his price-based sets of figures were themselves wrong. Oparin claimed, for instance, that using different techniques he could get new curves ‘that differ considerably from those of Kondratieff ... They have completely different timing and amplitude.’ [40]

The arguments used against Kondratieff’s figures can be used against many subsequent attempts to develop long wave theories.

First, do the ‘waves’ show changes in real production, or simply in prices? Some long wave theorists hold one view, some the other. The conclusions they draw are radically different. For example Mandel, who sees waves in real production, considers the 1970s the beginning of a long ‘downswing’. By contrast Rostow, who adheres to Kondratieff’s own method and stresses price movements, sees 1972–3 as marking the end of a downward swing. ‘The world economy’, he writes, ‘experienced a sharp turning point in foodstuffs and raw material prices – a break as sharp as those of the 1790s, the 1840s, the 1890s and the 1930s ... I am inclined to believe the fifth Kondratieff upswing is upon us.’ [41]

Secondly, the objection concerning the timing and amplitude of the different ‘cycles’ is just as valid. Look for instance at the figures for industrial growth reproduced above from Mandel in his attempt to justify Kondratieff’s case. They lump together uneven numbers of business cycles to make each ‘upturn’ and ‘downturn’. The first British ‘downswing’ lasted 20 years, the first ‘upswing’ 27 years, the second ‘downswing’ 17 years, the second ‘upswing’ 19 years, the third ‘downswing’ 24 years, and so on. In the case of Germany, starting with the first ‘upswing’, the period lengths are 24 years, 17 years, 20 years, 24 years. Taking one short-term slump or boom, lasting three or four years, from one ‘period’ and transferring it to an adjacent one could, in certain circumstances, completely change the averages for the two periods, turning ‘downswing’ into ‘upswing’ and vice-versa.

These objections together destroy the statistical case for long ‘cycles’ or ‘waves’. This has been virtually admitted by one of the latest studies to defend the notion of long waves, that of Freeman, Clark and Soete, [42] who admit that ‘the evidence is totally insufficient’ to prove the existence of a ‘50-year cycle’. They later claim that it is possible to use Kondratieff’s idea of ‘long waves’ without having ‘to accept the idea of cycles as such and certainly not of fixed periodicity.’ In fact this amounts to an admission that Trotsky and the other Russian critics of long wave theory were right, against Kondratieff!

The theoretical criticisms of Kondratieff and his successors have been as powerful as the statistical arguments. Kondratieff produced a theory for the cause of his long waves in February 1926 – several years after claiming to have observed them. He argued:

Marx asserted that the material base of crises or average cycles, repeating themselves each decade, is the material wearing out, replacement and expansion of the mass of means of production in the form of machines lasting an average of 10 years. It can be suggested that the material base of long cycles is the wearing out, replacement and expansion of fixed capital goods which require a long period of time and enormous expenditures to produce. The replacement and expansion of these goods does not proceed smoothly, but in spurts, another expression of which are the big waves of the conjuncture ... [43]

R.B. Day has noted that:

The forms of investment which Kondratieff has in mind included railways, buildings and the periodic technological renovations of industry which attend the rising wave of a long cycle. A rising wave presupposes a lengthy period of saving in excess of fixed capital formation, ultimate concentration of these savings in the hands of investors and profit opportunities sufficiently attractive to induce a new wave of investment ... [44]

Kondratieff saw the new investment wave as one bringing social and political instability and a depletion of investable funds, so that the interest rate would rise, curtailing investment and producing a new declining wave.

But in the declining wave there would accumulate, on the one hand innovations which could not be brought into production until there was a new round of massive capital accumulation, and on the other investable funds, as those on fixed incomes saw the value of their incomes rise with falling prices and therefore increased their saving.

Later ‘long wave’ theorists have taken up much of Kondratieff’s account. Schumpeter stressed the role of ‘bunches’ of innovation, and this is also one of the elements stressed by Mandel. Mandel amends Kondratieff slightly, seeing the massive investment not as involving a different set of long-term capital investments to those normally undertaken in the ten-year cycle, but rather as involving a complete replacement of old capital equipment throughout industry. This massive innovation, according to Mandel, allows the system to expand for up to 25 years without encountering the Marxist crisis of the falling rate of profit.

These theoretical arguments fall apart the moment they are examined seriously.

First there is the notion that capital can be ‘saved’ during the ‘downswings’, to be invested many decades later with a new wave of innovation. The ‘downswing’ is a period in which short-term slumps are more severe than the average. But such slumps have, as one very important effect, the wholesale destruction and ‘devaluation’ of capital. What is destroyed or ‘devalued’ cannot be saved.

Gerzstein pointed out against Kondratieff that even if some ‘income receivers’ saved more in periods of depression, ‘this is certainly not true of corporate businessmen’. Oparin noted that there was no sign of any real growth in the funds of savings banks during Kondratieff’s ‘downswings’. And Garvey, in 1943, rammed the point home: ‘Kondratieff’s assumption that free loanable funds can wait as long as a quarter of a century to be re-invested, remaining unaffected by expansions and contractions of successive business cycles, is certainly one of the weakest points of his argument and one which could not withstand an empirical test.’ [45]

The criticism of Kondratieff applies just as surely to his more recent followers. [46]

There is a second, and in some ways more fundamental, theoretical objection to be made by Marxists. The theory of long waves attracts people because it seems to fit the aberrant behaviour of the capitalist system during periods of deep crisis into an apparently water-tight mathematical model. Sense appears to be made out of nonsense. The decline of the system merely serves to help restore a long-term equilibrium. The long waves, like any other simple harmonic motion, oscillate around a fixed point. Any particularly deep crisis is no more than a mechanism by which the system prepares itself for another period of expansion.

Kondratieff never hid his view that his was a theory of long-term equilibrium. Like all theories of equilibrium, it assumed that however bad things seemed, there was always a hidden hand which would put things right.

Marx, by contrast, believed that capitalism was pushed towards crises from which it would find it ever more difficult to escape. With Trotsky, we can agree that there are periods when the system has expanded more rapidly than in other periods, without falling into the trap of seeing these as part of an ahistorical, self-correcting mechanism. As Sukhanov concluded in the 1920s:

Kondratieff studies economics in the same way as an astronomer might investigate the immutable orbits of heavenly bodies. A more rational approach would be to take into account capitalism’s growth, maturity, decrepitude – and even the likelihood of death. [47]

The words are an apt criticism of the methods of Schumpeter, Rostow, Mandel and the host of bourgeois economists who have been won over to long wave theories in the past decade.

* * *

Appendix four:
A crisis of hegemony?

THE AMERICAN KEYNESIAN economist Kindelberger has described the crisis of the 1930s in the following terms:

The world system was unstable unless some country stabilised it as Britain had up to 1913 ... When every country turned to its national private interest, the world public interest went down the drain, and with it the private interests of all.

Between 1873 and 1913,

British foreign lending and domestic investment were maintained in continuous counterpoint. Domestic recession stimulated foreign lending; boom at home and it went down. But boom at home expanded imports, which provided an export stimulus abroad ... Countercyclical lending stabilised the system ... [48]

By contrast, US overseas investment in the 1930s accentuated booms and slumps: ‘US foreign lending was positively correlated with domestic investment, not counterposed ...’ [49]

It led to a boom in investment in Europe and elsewhere at the same time as the boom took off in the US – and caused investment to decline abroad at precisely the time it slumped at home. Unlike British overseas investment 50 years earlier, the slump in one part of the world was not compensated for by a boom elsewhere, but exacerbated.

Kindelberger then extends this argument to the 1970s, seeing this as a period when the ability of one great power, the US, to provide a structure for the rest of the world system broke down, very similarly to the 1920s and 1930s.

In the 1940s, 1950s and 1960s, the US economy was all powerful. Not only was the US the policeman of the world; it was also its financier. The dollar was the great safety net that prevented anyone else falling too hard. But the past ten years have seen a decline in the ability of the US to do this, as a result of the relative decline in its own economic predominance. [50]

Kindelberger’s arguments have been taken up and refined by a number of socialist economists. For example, the American radical Arthur MacEwen argues:

One of the fundamental aspects of the crisis of the US economy in the 1970s has been the disruption in the stability in the international capitalist economy. The 25 years following World War II were characterised by a continuous increase in integration of the world capitalist system. However, throughout those years, forces were building towards the destruction of stability ... By the beginning of the 1970s, those forces had come into their own, and the basis for stability – US hegemony – had been eliminated. [51]

The long expansion after 1945 rested upon the ability of the US ‘to re-establish an international order which had been lacking for half a century – since the time when other nations began seriously to challenge Britain’s pre-eminence.’ [52]

One aspect of this new hegemony was the vast expansion of US direct investment overseas – from 11 billion dollars in 1950 to 30 billion dollars in 1960, to 70 billion dollars in 1970, to 133 billion dollars in 1976.

But the very success of the new world system based upon US hegemony began to undermine that hegemony.

For both economic and political reasons, the success of the US required that it take an active role in rebuilding the war-torn areas of the capitalist system ... Consequently, throughout the post World War II period, the other capitalist nations were able to move to a position where they could challenge the US both economically and politically. As early as the late 1950s and the early 1960s it was becoming clear that Japanese and European goods were beginning to compete effectively with US products. And other nations began to grumble about the costs of supporting a world monetary system based upon the dollar ...

The contradictions were exacerbated by the Vietnam War. The war meant that the US domestic economy overheated – a process made worse by the policy of financing the war (because of its unpopularity) by deficit financing instead of by taxation. Internationally it led to a considerable reduction in US trade surpluses. Yet the flow of US investment abroad continued to expand, until the other economic powers believed the dollar was overvalued and that they were in effect subsidising the US economy to the tune of about two billion dollars a year. The chickens came home to roost in 1971 when Nixon devalued the dollar and brought to an end the stable international currency alignments that had been established at Bretton Woods at the end of the Second World War.

At the same time the collapse of US military hegemony – in part a function of the collapse of the economic hegemony that had backed up military power – meant that the US was no longer able to keep in fine the all-important Middle East oil states when they decided to force up prices in 1973–4.

The very integration of the world economy built up on the basis of US hegemony in the previous period meant that the new instability fed back into the US domestic economy – the US was dependent on imports for 40 per cent of its oil; its exports were 10 per cent of GNP by the mid-1970s as opposed to only 5-6 per cent ten years before; earnings from direct overseas investment made up 30 per cent of after-tax corporate profits; US-backed loans abroad had tripled in the years 1971–5. [53]

Sustained expansion after the 1974–5 recession was impossible within the existing international framework – it meant a massive surge of imports into the US, a balance of payments deficit and a further fall in the dollar, which in turn threatened to undermine the world monetary system still more. But destruction of the old integrated world system by a retreat into protectionism was too frightening to attempt. As a result the short-lived 1977–8 American boom came to an end – even though the European and Japanese economies had hardly pulled themselves out of the previous recession.

‘Crises of hegemony’ theories have a certain attractiveness. They certainly describe one very important aspect of what happens in periods of intensified economic instability – the break-up of the old institutional framework internationally and the decline in the stabilising power of previously predominant economies. But they do not prove that the crisis of hegemony causes the general crisis. Could it not, perhaps, be the other way round, with an internal economic crisis of the dominating power undermining its hegemony? Or perhaps both the internal economic crisis and the crisis of hegemony are the result of some third factor?

The questions become more emphatic when one looks more closely at the arguments of the ‘crisis of hegemony’ theorists, such as Kindelberger.

Take his account of British investment in the late nineteenth century. He says it sustained the stability of the world economy because savings were always invested, either at home or abroad. But simple references to Britain’s dominant international position do not explain why they were always invested. Some factor not referred to by Kindelberger must explain why British savings were always invested somewhere in the late nineteenth century, while American funds were only ever invested abroad at times of domestic boom in the 1920s and 1930s. But then this factor, or its absence, is the explanation for the deepening of the crisis in the 1920s and 1930s and must be seen as provoking the crisis of hegemony.

The same might be said for the ‘crisis of hegemony’ in the 1970s. Certainly from the early 1960s onwards the succession of crises on the foreign exchange markets flowed from the decline of American hegemony. But did these cause the general economic crisis after 1973? If so, how? Certainly not through any decline in the flow of American funds abroad. The flow into the Eurodollar market was much greater in the late 1970s than, say, in the late 1950s. It was not a drying-up of savings available either in the US or internationally that caused the deepening of the crisis. The savings were there – but were not, in general, productively invested. The Eurocurrency funds flowed into speculation, financing the burgeoning debts of the ‘third world’ and Eastern Europe – but not, by and large, into direct investment. (This does not rule out the possibility at some point in the future of a collapse of international credit – but this would be a by-product of the wider crisis, not a cause.)

You need to explain the pattern of investment in order to explain the deepened crisis – and you can’t do that simply by referring to the crisis of hegemony. MacEwen seems in part to recognise this. He does refer to ‘the domestic economic crisis’ within the US, and to the ‘fundamental contradiction contained in the capitalist relations of production’. But that is to acknowledge that you have to look elsewhere than to the crisis of hegemony for the basic causes of the general crisis. In his case ‘elsewhere’ means to the theories of monopoly capitalism put forward by Baran and Sweezy, for which see below.

* * *

Appendix five:
A raw materials crisis?

A SUB-VARIETY OF the ‘crisis of hegemony’ theory is to be found in the argument that it is an increased dependence of the advanced countries on raw material supplies located in third world countries – especially oil – that has provoked the crisis. [54] The explanation takes an extreme form, to the effect that shortages of energy sources provide an absolute limit to further economic growth. But this explanation falls down, because numerous studies have shown that there are oil reserves still in the ground that are greater than the total amount used by humanity in the whole of its history so far, and coal reserves sufficient to last another 200 years. If these have not been tapped, it is because over the past 10-15 years firms and nations have not regarded the rate of return as sufficient to justify the necessary investment: that, for intance, is the main reason that the talk of using US or Japanese technology to exploit Russia’s Siberian oil reserves came to nothing in the early 1970s.

Indeed, it is not fanciful to suggest that if the only problem facing the world was a drying up of old oil reserves, then what you might expect would be a boom, as everywhere governments and firms would spend vast sums on developing alternative energy supplies. (After all, it was precisely such periodic needs to branch out into massive new investments of a special sort that Kondratieff used to explain the upward sections of his ‘long waves’.) The international economy would then be marked by very rapid expansion, perhaps of a very unstable sort, but not a slide into stagnation.

A more limited explanation looks not at the absolute limits to growth caused by the energy crisis, but at the effects on the advanced western states of having to pay more to the OPEC countries for oil. This is said to cut into the funds available for expansion in the advanced countries themselves. (This is essentially Rowthorn’s explanation.)

But it is an explanation with a line of reasoning missing. The funds from the western countries flow into the coffers of OPEC governments and ruling groups. Why don’t they spend the funds, either on expansion of the local industrial base, on consumer goods for themselves or their subject peoples, or on investments in the advanced countries? If any of these three things happens, all that results is a redistribution of purchasing power within the world system, without any drop in overall demand, or, for that matter, in the overall level of investable funds. Some capital-owning groups (those in the advanced countries) lose out to others (the oil producers) but there should be no diminution in the possibilities of growth for the system as a whole.

We are driven back once again to ask why investable funds are not invested? If ‘recycling’ of oil surpluses does not take place, the reason must lie in some factor inside the western economies that makes OPEC governments often feel their wealth is better protected by leaving oil in the ground rather than by exchanging it for a stake in western industry.

* * *

Appendix six:
Institutional crisis theories

ONE WAY OF attempting to fill the gap in theories of the crisis of hegemony has been to consider such crises as simply one expression of a much more widespread crisis in the institutional structure in which capitalist production takes place.

For example, an American Marxist, David Gordon, starting from certain ideas derived from Kondratieff’s long waves theory, suggests that historically, capitalism has passed through a number of states of development, each based upon a different set of structures. Each stage exhausts itself beyond a certain point. The institutional structure no longer fits the needs of accumulation: for instance, all the labour available with a particular structure of the labour market is exhausted; the structure of distribution does not fit the products of some new technology; certain raw materials run out, yet institutional structures are not available to enable the use of new ones.

The changeover from one set of institutions to another cannot easily be accomplished.

Many of the institutions associated with a stage of accumulation involve mammoth infrastructural costs of organisation and construction. These primarily involve the institutions of market access – access of raw materials, intermediate goods and final consumer demand. These institutions become part of the ‘built environment’ fixed in concrete and steel. Once these institutions have been built up at enormous cost, the world economy becomes ‘fixed into these particular infrastructural forms’ while their costs are repaid. [55]

Indeed, even when these costs have been recovered, the ‘fixing’ continues – since ‘many individual capitalists acquire strong vested interests in existing institutional structures’. It takes a massive crisis to shake up these structures and to force individual capitalists into ‘abandoning their old and increasingly unprofitable ways’. [56]

According to Gordon, the breakdown of the interwar years can be seen as resulting from a breakdown in the ‘prosperous combination’ of before the First World War:

Reduced competition through mergers, new markets and reduced material costs through imperialist colonisation, and reduced class conflict through a combination of new production relations, progressive social welfare policies and aggressive union busting.

International competition had already intensified before the war, and the peace did not resolve the instabilities. As world trade became increasingly perilous, world commodity flows slackened ... Their disruption threatened production ... Once the bubble burst, prosperity could not resume until the institutional basis for a new stage of expanded reproduction had been laid ...

World War II helped pave the way for this institutional regeneration. The US emerged from the war with enormous international economic power. The dollar and American power provided a new platform for international stability. Wartime discipline, post-war anti-Communist ideology and new collective bargaining institutions helped reintegrate workers into the accumulation process. The increased concentration of corporate power promoted more stable relations of competition ... [57]

The post-war structure began to break down as each of its particular institutions ran into problems. Gordon lists, among others, over-investment in automation; increased borrowing by corporations from government to compensate for loss of earnings; the rebellion of third world peoples – especially in Vietnam – against US imperialism, and the costs to the US of countering this; increased trouble caused by unorganised workers; rising interest rates; shifting labour patterns, which make the reproductive use of the school system more and more complicated; increased protests by workers at speed-up; the burgeoning success of European capitalism making American corporations face increasingly strong competition ... and so on.

The weakness in Gordon’s argument comes out when he lists the factors responsible for the breakdown of the post-war ‘prosperity’. Each can be seen to be as much a result of a wider crisis as a cause of its intensification. Take the argument about ‘overinvestment in automation’. There is no explanation why this should take place, nor why investment should not create a market for its own products. And above all there is no explanation why investment can go on for years without running into crisis, and then suddenly run into it in the 1970s. The nearest thing to an explanation is the Kontratieff notion of large fixed investments – but this has all the problems we found in Kondratieff’s own arguments. In the same way, Gordon’s account of the crisis in the labour market exhibits the same weaknesses as those theories which see increased union pressure as the cause of the crisis.

The French economist Aglietta provides another version of the ‘structural crisis’ theory [58], and uses it to give quite a sophisticated explanation of the present crisis – and of that of the 1930s. His basic argument is that the mechanisation of labour in the past century has gone through three stages, each of which has provided the basis for a stage of capital accumulation:

1. Taylorism involved raising the productivity of labour by accelerating the speed at which tasks were done in the workplace, and cutting down the time gaps between tasks.

2. Fordism, which follows from Taylorism, saw the redesigning of the whole production process to ensure a continual flow of work, with the work process simplified and broken down until each worker performed only a few repetitive actions. The end-point of this development was the ‘semi-automatic assembly line’, leaving as little time as possible for the worker to ‘recuperate’ his or her energies during the working day itself.

Secondly, Fordism simultaneously absorbed the workers’ consumption into the production process of capitalism as a whole, by replacing earlier forms of consumption with mass consumption through the market. For example, where previously workers pre pared the vast majority of their food in their own homes, now increasing use of tinned and frozen foods turned food preparation into part of the capitalist process of production; household work became less labour-intensive due to the intervention of factory-made vacuum cleaners and such like; and the mass provision of transport brought production into yet another new area of life.

This involved ‘a revolutionising of the consumption of the working class ... Consumption was based on the individual ownership of commodities – especially motor cars and individual housing units.’ It thus provided ‘the recuperation needed for enhanced production inside the factory.’ [59]

This raised contradictions. There is a certain insecurity built into the wages system, which is necessary for capitalism if employers are to keep wages down and maintain control of production through the ever-present threat of unemployment. Yet the system was now dependent on the workers also as consumers, and needed this new market to be organised and stable. Workers had to be able to pay for houses to live in and vehicles to get to work in, even if they feared unemployment. As Aglietta puts it, it became ‘essential to limit the consequences of capitalist insecurity in the formation of the individual wage, so as not to break the continuity of the consumption process ...’

This led to the growth of social insurance funds, state unemployment benefits, legislation to cover periods of sickness and so on: ‘Fordism could regulate the evolution of private working-class consumption only by generalising the wage relation to the conditions guaranteeing the maintenance cycle of labour power: provision for the unemployed and the sick, covering of family expenses and the means of existence of retired people.’ [60]

Aglietta argues that Fordism had definite effects on the overall pattern of economic development. The expansion of production in what Marx calls Department I of the economy, where the means of production are themselves produced, depended on an increase in the output of Department II, which made consumer goods, because cars, houses and so on were now a precondition for rising labour productivity at work.

But this meant that the system could be upset if Department II were allowed to contract: on the one hand, demand for the output of Department I would decline, on the other the productivity of labour in Department I would fall as living standards fell ‘below the social consumption norm’.

So, for Aglietta, there is a sense in which Keynes is the prophet of Fordism. His criticism of neo-classical economics and his notion of ‘effective demand’ are a partial recognition of the need for production and consumption to be integrated at a certain stage of capitalist development.

3. Neo-Fordism, Aglietta’s third stage in the mechanisation of labour, has only just begun. Fordism contained within it three ‘internal obstacles’ – the lack of integration between different production cycles of differing duration; the stress produced within the individual worker by continued work pressure, leading to absenteeism and industrial militancy; and the abolition of any direct link between the effort of the individual worker and his remuneration.

Neo-Fordism attempts to deal with these problems by using electronic devices such as computers to coordinate centrally productive units that are themselves small and decentralised. The managerial hierarchy is short-circuited and small groups of workers become apparently responsible only to themselves and electronic equipment for the fulfilment of centrally set norms. Labour-only subcontracting and even the workers’ cooperative become the basis for labour voluntarily imposing on itself norms of exploitation decided by a distant central management.

Using this framework of three stages of mechanisation, Aglietta then sees the great periods of crisis of the past hundred years as being the transition phases between one stage and the next. They occur when one way of organising labour – and exploitation – has exhausted itself and the next not yet become established.

Hence the great crisis of the 1870s occurred when the form of organisation of labour typical of the first ‘machinofacture’ had reached its limits of exploitation, and lasted until Taylorism was established; the great crisis of the 1920s and 1930s occurred when the limits of exploitation within the framework of Taylorism had been reached, and lasted until Fordism had established itself; the present period of crises arises out of the exhaustion of the possibilities within Fordism and will last until the structure of accumulation is organised in a neo-Fordist mould.

The form which an ‘exhaustion’ of the possibilities of a structure of accumulation takes is a decline in the productivity increases that are to be gained by increased expenditure (whether direct, or indirect ‘non-productive’ expenditure by governments). But what causes this decline?

Here Aglietta falls back on an argument whose weakness we have already examined: he sees a long-term rise in the level of class struggle, taking place as the structure of accumulation becomes widespread, which cuts into the possibilities for increasing the rate of exploitation:

The crisis of Fordism is first of all the crisis of the mode of labour organisation. It is expressed above all in the intensification of the class struggle at the point of production. By challenging conditions of work bound up with the fragmentation of tasks and intensification of effort, these struggles showed the limits to the increase in the rate of surplus value that were inherent in the relations of production organised in this type of labour process. This was the root of the crisis ...

It can be seen in the halt in the fall of real wage costs that occurred simultaneously with the outbreak of sporadic conflicts and endemic contradictions challenging work disciplines of the kind that Fordism had established. [61]

From this initial source, the crisis spreads throughout the economy. The growth of Department I of the economy is retarded because it no longer produces new techniques capable of increasing productivity and so counteracting the tendency for the organic composition of capital to rise. There is a decline in investment, growing unemployment and increased job insecurity. At the same time, the failure of productivity to grow in industries producing consumer goods leads management to attack living standards. This attack has to be accompanied by an attack upon ‘so-called collective consumption’ (what is often called the ‘social wage’) – since productivity in the sectors producing ‘collective consumption’ goods and services rises much more slowly than in the other sectors of the economy.

Either these services are produced by capitalists with under-developed methods, and their costs grow astronomically, as social demand for them rises [, or] these services are produced by public bodies. They then absorb labour which is unproductive from the point of view of surplus value ... Far from being complementary to labour that does produce surplus value, this unproductive labour is from the capitalist standpoint antagonistic to it when it absorbs a share of social value that grows more quickly than the sum total of surplus value.

A point is reached where the ‘Fordist’ conditions which allowed an expansion of the capitalist system begin instead to throw it into crisis.

As long as major transformations in the production of standardised commodities and a corresponding upsurge in the mode of consumption were predominant, the collective costs of the reproduction of wage labour could be held steady and the rising rate of surplus value could still be imposed. But these forces themselves generate a more and more rapid increase in the collective costs, at the same time as they exhaust the potentialities contained in the mechanisation of wage labour. It is not surprising, therefore, that the crisis of Fordist work organisation should at the same time have been the occasion for a general drive of the capitalist class to curtail social expenditures ... [62]

Aglietta clearly makes a number of powerful points. But there are weaknesses in his explanation of the onset of deepened crisis, very similar to those of a number of the earlier theorists. He does not really explain why, at a certain point in time, the expansive elements in Fordism should suddenly cease to operate. Take, for instance, the possibilities of raising labour productivity. Aglietta seems to imply that a point is reached where the worker can (or will) not work any harder. But labour productivity is increased not only by the worker working harder – indeed, it has to be proven that workers today work that much harder than workers a hundred years ago. What increases labour productivity is a rise in the technical level of production. And Aglietta provides neither proof that the technical level of production stopped rising in the mid-1960s, nor any explanation as to why it should have done.

Even if ‘Fordist’ forms of raising labour productivity have been exhausted in the old industrial centres of Western Europe and North America, what about parts of the globe where ‘Fordism’ has barely been introduced? Should not the system gain a new lease of life, on Aglietta’s argument, by shifting growing sectors of production to parts of the globe where ‘Fordism’ can still displace previous forms of organisation of the production process? This is the conclusion of some thinkers who share Aglietta’s terminology. Why does Aglietta himself argue otherwise? Why does he insist that global productivity can no longer be raised on a ‘Fordist’ basis?

It only needs to be added that key points in Aglietta’s argument depend upon dubious generalisations about the growth of ‘unproductive’ government expenditure, like those of Bacon and Eltis, and an international increase in the level of class struggle, taken from ‘wage push’ theorists, which, as we have seen earlier, can by no means be taken for granted.

The gaps in Aglietta’s argument could be filled in if it was first shown that investment had to fall before the beginning of his periods of crisis. Then much else of what he has to say would follow. But you cannot explain a decline in investment on Aglietta’s own premises. He cannot really say why, in the heyday of the first forward rush of Fordism, in 1929, the bottom should suddenly have dropped out of capitalist expansion, or why a new wave of ‘Fordist’ expansion should be grinding to a halt today. So other explanations have to be looked for. Taken by itself, his account is as unconvincing as those which merely talk of ‘waves of innovation’ or ‘long waves’ or wage push or rising public expenditure.

* * *

Appendix seven:
Theories of monopoly and stagnation

THE AMERICAN MARXISTS Baran and Sweezy [63], on the one hand, and the neo-Keynesian Steindl [64], on the other, have developed theories that ascribe the onset of general crisis to the monopolisation of production. These were originally intended to explain, not the crisis of the 1970s, but the great depression of the 1930s. But they have been used to explain the new lurch into world crisis since 1973.

Both Baran and Sweezy and Steindl see a tendency towards stagnation as a necessary result of the effect of monopolisation on profit margins.

For Steindl, the growth of monopolistic trends from the 1890s onwards led to an increase in the level of profit margins, combined with a deliberate policy of developing excess capacity in order to protect profits. This in turn led to a lower rate of accumulation, so that in the decade before 1899 the rate of growth of the US economy was 5 per cent, in the 1920s it was 3 per cent, in the 1930s nil. [65] ‘An increased fear of excess capacity, due to the transition to monopoly, will always reduce the limiting rate of growth [by which he means the maximum rate of growth that is possible] ... I believe this has in fact been the main explanation of the decline in the rate of growth which has been going on in the US from the end of the last century ...’ [66]

But reducing investment and cutting the rate of growth does not restore full capacity utilisation and protect profit margins. Instead, it reduces effective demand and leads to still greater excess capacity: ‘A given degree of capacity utilisation can be restored adequately only by the method of eliminating capacity by price-cutting, but never by the method of reducing investment, because this leads only to an even greater excess capacity ...’ [67]

The result is a general crisis, unless other factors mask it:

Stagnation did not come overnight, preceding it there had been a long process of secular change which passed almost unnoticed ... Hardly anyone during the ‘New Era’ was aware of the fact that the annual rate of growth of business capital was only half what it had been thirty years earlier. [68]

For Steindl the ‘masking’ factor since the Second World War has been a much higher level of government expenditure than previously. ‘The post-war economy has been transformed by the unprecedented role which government public policy and politics have played.’ [69] The decline in cold war tension on the one hand, and the preoccupation of governments with inflation and public debt on the other, reduced government willingness to spend. But a fall in government spending in no way leads to a rise in the willingness of industry to invest. So even the new lower level of government spending can, according to Steindl, be financed only by budget deficits.

Baran and Sweezy’s argument is not fundamentally different, although they express it in Marxist rather than Keynesian language. They argue that under monopoly conditions, the determination of prices no longer depends on the free play of market forces. The trend is for corporations to increase their profits, and for the total ‘surplus’ of society, what remains after workers’ consumption and depreciation have been accounted for, to grow. The monopolies are disinclined to transfer this surplus to shareholders for private consumption, and so, ‘not only the surplus, but also the investment-seeking part of the surplus tends to rise as a proportion of total income’. [70]

But it is not possible for investment to rise fast enough to absorb this proportion of the surplus that is not privately consumed.

If total income grows at an accelerating rate, then a larger and larger share has to be devoted to investment and conversely, if a larger and larger share is devoted to investment, total income must grow at an accelerating rate. What this implies, however, is nonsensical from an economic standpoint. It means that a larger and larger volume of producer goods would have to be turned out for the sole purpose of producing a still larger and larger volume of producer goods in the future ...

One is left with the inescapable conclusion that the actual investment of an amount of surplus which rises relative to income must mean that the economy’s capacity to produce rises faster than its income ... Sooner or later excess capacity grows so large that it discourages further investment. When investment declines, so do income and employment and hence the surplus itself.

In other words, this investment pattern is self-limiting and ends in an economic downturn – the beginning of the recession or depression. [71]

Again they argue

These mechanisms tend to generate a steadily rising supply of investmentseeking surplus, but ... in the nature of the case they cannot generate a corresponding rise in the magnitude of investment outlets. Hence if [these] investment outlets were the, only ones available, monopoly capitalism would bog down in a permanent state of depression.

Fluctuations of the kind associated with the expansion and contraction of inventories would occur, but they would take place within a relatively narrow range, the upper limit of which would be far below the economy’s potential. [72]

This, for them, is the explanation of the 1930s. But they go on to argue that exceptional forms of investment were able, for a time, to counter these pressures towards stagnation in the 1940s, 1950s and 1960s. Such, they argue, was the case when technological innovation necessitated massive new investment – as did the spread of the railways in the last century, and the spread of the automobile in the 1920s and the 1940s and 1950s. They also see expenditure on the sales effort itself, through advertising and promotion of goods as disposing of some of the ‘surplus’. So too does foreign investment.

They conclude, however, that the most important recent form of ‘surplus absorption’ has in fact been the activity of government, especially its military activity. They show the huge shifts in the pattern of expenditure before 1929 and after the Second World War.

In other words:

Some six or seven million workers, more than 9 per cent of the workforce, are now dependent for their jobs on the arms budget. If military spending were reduced once again to pre-Second World War proportions the nation’s economy would return to a state of profound depression, characterised by unemployment rates of 15 per cent and up, such as prevailed during the 1930s. [74]

Baran and Sweezy show, in passing, that attempts to end the crisis of the 1930s on bases other than military expenditure, did not succeed.

Measured in current dollars, government spending increased from 1929 to 1939 more than 70 per cent. At the same time, GNP declines ... 12.7 per cent, and unemployment rose from 3.2 per cent to 17.2 per cent ... Regarded as a salvage operation for the US economy as a whole, the New Deal was a clear failure ...

War spending accomplished what welfare spending had failed to accomplish. From 17.2 per cent of the labour force, unemployment declined to a minimum of 1.2 per cent in 1944. [75]

Baran and Sweezy’s account of the fluctuations of the world economy in the past 80 years has the merit that it seems, at first sight, to fit reality. In this respect it has a great advantage over most of the theorists we have looked at so far – and, for that matter, over various writers (from Yaffe to Bleaney) who simply dismiss Baran and Sweezy’s theories out of hand as ‘underconsumptionist’. It is true that their theory does depart from Marx’s account of capitalist crisis in a ‘Keynesian’ or ‘underconsumptionist’ direction, as I will attempt to show later. But they do capture empirically some of the major shifts in the dynamic of capitalism as most of their critics do not.

Baran and Sweezy conclude from their analysis that the trend has been towards the stagnation of the world economy ever since the first development of monopoly capital towards the end of the last century.

If the depressive effects of growing monopoly had operated unchecked, the United States economy would have entered a period of stagnation long before the end of the 19th century, and it is unlikely that capitalism would have survived into the second half of the 20th century. [76]

But, they argue, this was avoided in the US of the 1880s and 1890s by continuing huge expenditure on railways. ‘Census data suggest that from 1850 to 1900 investment in railways exceeded investment in all manufacturing industries combined’ until ‘the crisis of 1907 precipitated a sharp drop in railway investment’ which ‘remained permanently at a lower level.’ [77]

This, they argue, has an immediate effect on the general dynamism of the economy. After 1908, the tendency is for depressions to last longer and for booms to be shorter than previously. Unemployment rose so that even in the ‘boom’ years of 1909–10 and 1912–13 it was no higher than in the ‘slump’ years of 1900 and 1904.

The First World War lifted the economy out of this stagnation, and the restructuring of industry associated with the first wave of ‘automobilisation’ continued to keep stagnation at bay throughout the 1920s. But this first wave soon exhausted itself, and from 1923 onwards, excess capacity accumulated rapidly until in 1920 it hit production and the great slump of the 1930s began.

This, as we have seen, was ended for Baran and Sweezy by the Second World War. After the war stagnation was again kept at bay as in the 1920s – this time partly by a second wave of automobilisation, but more importantly by a level of arms spending much higher than at any previous peacetime period.

With the aftermath [post war] boom triggering a great upheaval in the living patterns of tens of millions of people, and with arms spending growing nearly five fold, it is probably safe to say that never since the height of the railway epoch has the American economy been subject in peacetime to such powerful stimuli ...

Yet it was already running out by the mid-1950s.

What is remarkable is that despite the strength and the persistence of these stimuli, the familiar symptoms of inadequate surplus absorption – unemployment and under-utilisation of capacity – began to appear at an early stage, and, apart from cyclical fluctuations, have been gradually growing more severe ... [78]

Despite its descriptive power, there are overwhelming objections of both an empirical and a theoretical kind to be made to Baran and Sweezy’s account.

First of all, it is by no means certain that the ‘surplus’ has increased in the long-term way suggested by Baran and Sweezy. Their collaborator, Phillips, purports to give factual evidence for its growth. But this depends upon the assumption that all government spending is part of the surplus. If some of it is not (for instance, if some is indirectly part of wages or is directly productive as nationalised industry investment) then the figures can show a fall, not a rise. [79]

Baran and Sweezy try to counter such objections in advance by insisting they are talking about ‘potential surplus’ – the surplus which would exist if industry worked at full capacity. But clearly this counter argument cannot apply to periods like 1942–5 or the early 1950s when industry actually did work very close to full capacity.

Secondly, the timing of the transition from periods of growth to periods of stagnation seems very arbitrary. Why, for instance, should automobilisation have led to massive new investments in the early 1920s and not in the late 1920s?

Baran and Sweezy provide no real explanation – unless it is the old Keynesian explanation that government did not then understand the needs of the system, and that the war taught them otherwise. The same gap in explanation characterises their account of the move from boom to stagnation over the past 25 years. They seem to suggest it resulted from accidents. [80]

If coincidental factors have produced stagnation, it would seem that other coincidental factors (or changes in government policy) could reverse the trend again. Indeed, it has been suggested by O’Connor that the move from the ‘Warfare State’ to the ‘Warfare-Welfare State’ could open up just such a new period of capitalist expansion.

Steindl’s account of the 1950s and 1960s discussed above can be faulted on many of the same grounds. The ‘theory’ of ‘maturity and stagnation’ becomes so overlaid with ‘accidental’ countervailing influences as to provide no guide at all to understanding what is likely to happen in future.

Thirdly, the whole basis of Baran and Sweezy’s (and Steindl’s) argument, the notion of ‘monopoly profits’ as the cause of the ‘rising surplus’, is open to factual criticism – as the Argentinian Marxist Alejandro Dabat has shown. He gives figures that indicate that US monopoly concerns do not have higher than average profits:

The spheres of business in which the average rate of profit is more than a third above the general corporate average are not, in general, those with a high monopolistic concentration, except for the tobacco industry, the brewing industry and petrol distribution and sales.

At the same time, the spheres which obtain a corporate rate of profit a third less than the average are highly concentrated spheres.

The most important industries of the North American economy, almost all totally controlled by a very few monopoly enterprises (automobiles, aerospace, the electrical industry, the food industry, telephone and electrical utilities, petroleum extraction and coal mining) are found in an intermediate position, close to the average rate of profit. [81]

These factual criticisms of Baran and Sweezy are reinforced by criticisms of a theoretical nature.

Fourthly, on the basis of Marx’s theory of profits, there are clear limits to the growth of the ‘surplus’ accruing to the monopolies. As Dabat has argued, once monopolisation has proceeded beyond a certain point, it is very difficult for monopoly profits to maintain themselves above the average. For monopoly profits do not come out of thin air – but are a result of the ability of some firms, through monopoly prices, to force smaller firms to give them an unduly large share of the total surplus value.

But as the proportion of industry that is monopolised grows, and as the non-monopoly sector correspondingly shrinks, the creation of surplus value comes to take place predominantly in the monopoly sphere itself. That means there is less and less non-monopoly surplus value for the monopolies to gain control of through their pricing policy. A point will eventually be reached at which factors other than monopolisation will determine where any super-profits go – especially the extent to which different firms are in the most dynamic, rapidly growing sectors of the economy. But such is the scale of their fixed investment in established industries that many monopolies find it difficult to switch to the newer, more dynamic industries. Non-monopoly firms are as likely to be found there as monopoly firms. [82]

In this way Dabat destroys Baran and Sweezy’s theoretical edifice. In doing so he also destroys certain conclusions drawn by their followers, for instance O’Connor, who argues that the workers in monopolies are privileged because the monopolies’ ability to protect their profits leads them to grant wage increases more or less automatically.

Fifthly, underlying Baran and Sweezy’s and Steindl’s argument is the assumption that declining price competition between monopolies also means a decline in the pressure to use the surplus at their disposal for accumulation. Instead, they can try to protect their profits by not investing and by maintaining surplus capacity. Yet there is much evidence that declining price competition is accompanied by an increase in other forms of competition: pressures for the innovation of products, pressures for the expansion of the scale of production so as to reduce costs and raise profits at existing prices, pressures on the state to expand its investment in arms industries so as to provide the military wherewithal to back up the monopolies in their international struggle for markets. At the same time the greater internationalisation of production has often made the various national markets become the meeting point for competition between the monopolies of different nations.

Taking the international aerospace, car, or chemicals industry over the past decade – a decade which has seen a growing trend towards stagnation – it is difficult to claim that there has been a reduction in competitive pressures for component firms to invest, even though they are near-monopolies within national markets. If they have not invested all the ‘surplus’, it has not been because there was reduced (international) competitive pressure on them – some other factor has made them frightened to expend the huge sums needed to finance their response to such pressures. [83]

A final theoretical point against Baran and Sweezy. They make the mistake of all ‘under-consumptionists’ of assuming that capitalism has to have a rational goal. What else can be meant by their argument that capitalism cannot simply produce means of production in order to produce further means of production?

It certainly can, providing it finds it profitable to do so. Since, for Baran and Sweezy, the ‘surplus’ goes on rising indefinitely then production of means of production should be able to go on rising indefinitely. The fact that no humans beings benefit from this in terms of improved consumption does not prove that capitalism must break down. It only proves that capitalism is a dehumanised system, that, as Marx put it in the Communist Manifesto, ‘in bourgeois society, living labour is but a means to increase accumulated labour.’

The mistake made by Baran and Sweezy is not new. It is the same made by Rosa Luxemburg [84] to justify her view that capitalism must break down eventually. But it is surprising that Sweezy makes this mistake, since he explicitly denounces the notion, in one of his earlier works, that ‘all economic behaviour’ under capitalism, ‘is directed towards the satisfaction of human need.’ [85]

In making this final criticism of Baran and Sweezy we are also, however, indicating the route towards the discovery of the rational core of their mistaken analyses. If it could be found that over time the dynamic of capitalist growth, the creation of profit, was somehow slowed down, then regardless of whether ‘surplus’ grew or contracted, you would expect the conditions for its investment to become more and more unfavourable. Then you would get precisely the stagnation described by Baran and Sweezy.

Thus our criticism here, as with other theories of the crisis, leads back to the main arguments of the book.

* * *

Notes

1. Glyn and Sutcliffe, British Capitalism, Workers and the Profits Squeeze (Harmondsworth 1972).

2. Glyn and Sutcliffe, p. 10.

3. New Left Review 98, p. 67.

4. Ernest Mandel, Late Capitalism (London 1975), p. 179.

5. See Report of National Institute for Social and Economic Research, 1978, No. 4.

6. See his articles in Brookings Papers on Economic Activity, 1977.

7. The figures given are for ‘industrial and commercial corporations in the us’ before tax and after deducting stock appreciation and capital consumption.

But:

1. The after tax figure would be likely to show a quite different trend, since ‘the level of taxation on US profits has fallen continually throughout the post-war period, and was 48.5 per cent in 1961 and only 26.9 per cent in 1970’ (figures given by Victor Perlo in the Review of Radical Economics, Fall 1976. I use Perlo’s critique of the Nordhaus figures here, although I think he can be faulted on certain points).
    
2. Corporations have an incentive to overstate their capital consumption costs, since these are tax deductible. ‘During the first five years after World War II corporate profits after taxes were about three times as large as corporate capital consumption allowances. But during the 1970s corporate capital consumption allowances were considerably larger than profits after tax. This did not reflect itself in the real rate of wear and tear on equipment and structures.’ (Perlo)
    
3. The figures exclude interest and rents, which are transferred from industrial capital to financial capital.
    
4. The figures exclude from profits all company write-offs for stock appreciation, although in a period of rapid inflation a good proportion of the rising cost of financing stocks will be paid for by borrowing at interest rates which may well be negative in real terms (ie less than the rate of inflation). The result is ‘double accounting’ that under estimates the real level of profits (this, incidentally, is the criticism made by the Cambridge Economic Policy Review, 1978, p. 65, of the Sandilands procedure for inflation accounting in Britain).

8. He includes all the cost of stock appreciation and capital consumption in profits.

9. Perlo.

10. Source: Brookings Papers, 1977, No. 1, p. 216. The figures are for total returns (profits plus interest payments), before tax, to non-financial capital, after deducting stock appreciation, and based on net capital stock.

11. Conference of Socialist Economists, Bulletin, February 1975, p. 8.

12. Lloyds Bank Review, 1974 No. 112, p. 11. It is true, however, that recent issues of the Bank of England Quarterly Bulletin have provided figures showing that the share of profits added (for manufacture, service and distribution), after deducting stock appreciation and capital consumption, has been falling for many years. ‘They show a significant downward trend in profitability throughout the 1960s and a sharp decline since 1973’ (Bank of England Bulletin, 1978, p. 517). But these figures must be subject to the same charge of ‘double accounting’ made by the Cambridge Economic Policy Review as the American figures. In any case, the Bank of England Quarterly itself points out that this alleged fall in the ‘share of profits’ in company value added is insufficient in itself to account for the scale of the fall in the rate of profit in recent years. Something else besides the allegedly ‘rising share’ of wages must be responsible for that (p. 517).

13. Review of Radical Political Economy, Spring 1975.

14. This point is elaborated in an article by Weeks in Science and Society, Fall 1979.

15. US Department of Commerce, 1978 Statistical Abstract of the United States, p. 429.

16. Perlo, p. 62. This is hardly consistent with the view that the ‘increased strength’ of the workers provided the mechanism for the ‘squeeze on profits’. Further more, in a review of research into US labour productivity, C. Bourdon of the Harvard Business School ‘sets out the evidence to show that the power of organised labour has actually diminished during the past decade’. (Financial Times, 21 November 1979)

17. They presented their view first of all in a series of sensational articles in the Sunday Times which was later expanded into a book: R. Bacon and W. Eltis, Britain’s Economic Problem: Too few producers, London 1976.

18. P. Mattick, Marx and Keynes, London 1971; D. Yaffe and R. Schmiede, State expenditure and the Marxist theory of crisis (IS internal publications, London 1972).

19. Mattick, pp. 161–3.

20. Sources: British Labour Statistics Year Book and Economic Trends.

21. F.T. Blackaby in F.T. Blackaby (ed.), British Economic Policy 1960–74, p. 649.

22. Blackaby, p. 649.

23. Blackaby, p. 650.

24. R. Matthews in The Economic Journal, September 1968.

25. Source: D. Glynn in Lloyds Bank Review, October 1976, p. 23.

26. Source: The Economist, 31 July 1976.

27. New Left Review 92, pp. 57 and 83.

28. This, effectively, is the argument produced by Glyn in 1975 to justify his claim that it is a ‘declining share’ of profit that has produced crisis. He argued that the share after tax of profit has declined, even though the share after tax of wages has also declined – because of a rise in the ‘social wage’ of workers.

He assumes that the ‘workers’ share’ of public expenditure amounts to ‘92 per cent of current expenditure on housing, health, education and other social services; half expenditure on fire services, a quarter of expenditure on roads; plus current grants to persons (net of tax) and consumption of social service means of production ...’

Using these proportions he finds that ‘far from there being a fall in the share of the net social product going to labour, there has been a rise from 69.3 per cent in 1955 to 73.7 per cent in 1972.’

Even on his own terms, Glyn’s figures are open to objection. As I have argued elsewhere (Socialist Review, No. 10, London 1979) a much larger proportion of the services he lists benefit the ruling class and the petty bourgeoisie (old and new) than he claims. On a rough calculation, revised figures could not produce a total increase in the ‘share’ of workers in the National Income of more than 2 per cent over 20 years – quite insufficient to explain the scale of the crisis.

29. J. O’Connor, The Fiscal Crisis of the State, New York 1973, p. 6.

30. O’Connor, p. 99.

31. J. Cypher, Review of Radical Political Economy, Fall 1974.

32. Source: Steindl, Cambridge Journal of Economics, March 1979.

33. New Left Review 92, p. 64.

34. New Left Review 92, p. 80.

35. The deviations were averaged, running 19-year deviations from the trend. For one version of Kondratieff’s theory, see the translation of his Long waves in economic life reprinted in Lloyds Bank Review, July 1978.

36. Mandel, pp. 141–2.

37. See for instance L. Trotsky, The First Five Years of the Communist International, New York 1945, p. 174, and the article by R.B. Day in New Left Review 92. In his essay The Curve of Capitalist Development (Fourth International, May 1941), he argued that:

‘It is already possible to refute in advance Professor Kondratieff’s attempt to invest epochs labelled by him “major cycles” with the self same rigid lawful rhythm that is observable in minor cycles. The character and duration (of large sections of the capitalist curve of development) is determined not by the cyclical interplay of capitalist forces, but by those external conditions through whose channel capitalist development flows.’

Trotsky sees these ‘external conditions’ as ‘the absorption by capitalism of new countries and continents, the discovery of new natural resources, and, in addition, significant factors of a “superstructural” order, such as wars and revolutions ...’

The point at issue between Kondratieff on the one hand and Trotsky on the other was not whether there were periods in which the system expanded more rapidly than others. It was whether the periodisation could be explained in terms of waves, of patterns which not only had occurred empirically, but which were bound to recur, with each downturn preparing the way for a new upturn.

38. There is an excellent account of these criticisms in George Garvey, Review of Economic Statistics, Vol. XXV No. 4 (November 1943). This article is a devastating rejoinder to all theories of long waves.

39. Quoted in Garvey.

40. Quoted in Garvey, p. 210.

41. W. Rostow, Trend periods revisited, in Journal of Economic History, 1975, p. 749.

42. Freeman, Clark and Soete, Unemployment and Technical Innovation (London 1982), p. 22.

43. Quoted in R.B. Day, New Left Review 99, p. 77.

44. Day, p. 77.

45. Garvey.

46. For a fuller critique of Mandel’s version of the theory, see my review of Late Capitalism in International Socialism (new series) 1.

47. Quoted in Day, pp. 78–9.

48. Kindelberger, The World in Depression 1929–39 (London 1973), p. 292.

49. Kindelberger, p. 293.

50. Kindelberger, pp. 307–8.

51. Arthur MacEwan, US Capitalism in Crisis, in URPE 1977, p. 46.

52. MacEwan, p. 46.

53. MacEwan, p. 51.

54. An example is Rowthorn.

55. D. Gordon, US Capitalism in Crisis, p. 31.

56. Gordon, p. 30.

57. Gordon, p. 29.

58. P. Aglietta, Theory of Capitalist Regulation (London 1979).

59. Aglietta, p. 154.

60. Aglietta, p. 165.

61. Aglietta, p. 162.

62. Aglietta, p. 167.

63. Baran and Sweezy, Monopoly Capital.

64. J. Steindl, Maturity and Stagnation in American Capitalism (London 1953).

65. Steindl, p. 155 and following.

66. Steindl, p. 255.

67. Steindl, p. 135.

68. Steindl, p. 166.

69. Steindl, Cambridge Journal of Economics, March 1973, p. 8.

70. Baran and Sweezy, p. 89.

71. Baran and Sweezy, p. 70.

72. Baran and Sweezy, p. 95.

73. Baran and Sweezy, p. 155.

74. Baran and Sweezy, pp. 155–6.

75. Baran and Sweezy, p. 162.

76. Baran and Sweezy, p. 216.

77. Baran and Sweezy, pp. 218 and 223.

78. Baran and Sweezy, p. 240.

79. This is the point made by Bleaney in his Underconsumptionist Theories (London 1977), p. 230.

80. See Baran and Sweezy, p. 243.

81. A. Dabat, Debate, May/June 1979, p. 17 (my translation).

82. Dabat.

83. It is because Baran and Sweezy claim such external pressures are of diminished importance in firms’ handling of the product of exploiting labour that they call it a ‘surplus’ – we follow Marx in calling it surplus value because it is not a self-contained entity, but something continually related, by national and international competitive forces, to the surplus value in the hands of other firms.

84. See R. Luxemburg, The Accumulation of Capital, and her Anti-Critique.

85. P. Sweezy, Theory of Capitalist Development (London 1946), p. 82.

 
Top of the page

Last updated on 28 June 2019