Chris Harman

Zombie Capitalism

Part One: Understanding the System: Marx and Beyond

* * *

State Spending and the System

An important distinction

If the enormous growth of the economic importance of the state was one feature which distinguished 20th century capitalism from the capitalism of Marx’s time, another was the growth of all sorts of expenditures that were not directly productive.

Marx had taken over from Adam Smith a distinction between “productive” and “unproductive” labour. Smith had been writing at a time when the capitalist mode of production was still in its infancy and he sought to work out what was needed for it to overcome obstacles to its further advance. He therefore distinguished between the uses of hired labour that enabled the capitalist to make profits so as to further advance production and those which simply absorbed existing resources. Employing someone to make things to sell was productive; employing someone to tend to one’s individual desires was not. Or, as it was sometimes put, employing someone in a factory created wealth; employing someone as a personal servant simply used wealth up. But it was not only servants who Smith regarded as unproductive and wasteful in this sense: he had the same attitude to the hoards of placemen and women who lived off the revenues of a state which had not been reformed fully to suit the needs of capitalist production. [1]

Marx took up this distinction as he prepared various drafts for Capital and developed his own understanding of it. He, like Smith, was interested in what made capitalism function – even if out of opposition to, not support for, the system. And so his concern was with what was “productive” in capitalist terms. [2] It was, he argued, that which was productive of surplus value. Labour which produced surplus value enabled capitalists to accumulate; labour which did not produce surplus value was of no use in this respect – it was “unproductive”.

In all this, he was careful to make it clear that the “productiveness” of labour did not depend on the physical form or how socially useful the product was. What mattered was its ability to create surplus value – nothing else. “This distinction between productive and unproductive labour”, he wrote in one of his notebooks, “has nothing to do with either the particular speciality of the labour or with the particular use value in which ... [it] ... is incorporated.” [3]

Marx’s distinction was not between material production and what today are categorised as “services”. Some “services” have a use value that is bought and sold as a commodity on the market – or make a useful addition to some other commodity. These have an exchange value which is determined by the socially necessary labour time needed to produce them and so can provide capitalists with new surplus value. They are therefore productive. Acting in a film, for instance, is productive insofar as it creates a use value (adding to people’s enjoyment and so improving their living standard) that is sold profitably as a commodity by the capitalist who employs the actor. Similarly, moving things from where they are made to where they can be consumed, as is done by some transport workers, is productive, since it is in effect part of the process of completing their production. By contrast, actors who appear on television to urge people to buy a particular good are not productive, since their labour does not create new use or exchange values. It merely aids in the selling of goods that have already been produced.

Guglielmo Carchedi has rightly argued:

The category “services” only confuses matters and should be dropped. “A service is nothing more than the useful effect of a use value, be it of a commodity, or be it of labour” [according to Marx]. [4] Therefore, “services” encompass productive labour (hotels, entertainment) and unproductive labour (advertising, market research) ... [5]

In his first discussions on the issue in the early 1860s Marx assumed, like Smith, that unproductive labour is concerned with services provided by individuals for the upper classes. [6] These included providing “entertainments”, dealing with “physical infirmities” (doctors) and “spiritual weakness” (parsons), and resolving “the conflict between private interests and national interests” (e.g. statesmen, lawyers, police and soldiers). The last sort were regarded “by the industrial capitalists themselves” as incidental expenses of production to be kept down to the most indispensable minimum and provided as cheaply as possible. [7]

Marx recognised that sometimes personal services for the ruling class were provided not by individuals working on their own account, but by capitalists employing paid labour to provide them to others. In these cases, he argued, the labour was productive because it created surplus value. The capitalists who employed it, after all, sold the produce of the labour at more than they paid for the labour power and pocketed a profit as a result. So a teacher employed personally in someone’s home to teach their children was providing a service from which no profit was made and was unproductive; by contrast, a teacher employed by a company which made a profit by running a school was productive. One did not in any way help capitalists to accumulate value; the other did. The distinction was between labour that was integral to capitalist production and accumulation, and that which was not.

But in Capital Marx also found himself having to revisit the distinction between productive and unproductive labour in a different context – a context which was integral to, not external to, capitalist production in its totality. For as capitalism developed, it became increasingly dependent on many forms of labour that produced nothing.

There was the labour involved in maintaining discipline inside the capitalist enterprise – the “work” of managers, supervisors, foremen. There was the commercial labour involved in the exchange of already produced commodities as they went through the various chains of buying and selling before reaching the final consumer. There was the financial labour involved in reckoning up profit and loss, advancing credit, and dividing up surplus value between the various sections of the capitalist class. Marx recognised that these sorts of labour would grow in quantity as capitalism expanded:

It is clear that as the scale of production is extended, commercial operations required constantly for the recirculation of industrial capital ... multiply accordingly ... The more developed the scale of production, the greater ... the commercial operations of industrial capital. [8]

Such labour could not be regarded as productive if the capitalist employed it in these ways, any more than the labour of the servant could be. Maintaining discipline, selling goods or getting the accounts done were necessary functions that had to be paid for by deductions from surplus value, not creative labour that added to surplus value. They did not produce something new, but were merely concerned with controlling the production of value by others, with transforming it from one form (commodities) into another (money), or with dividing it up between people. The activities of a supervisor, a bank clerk or a shop assistant could no more create value (and therefore surplus value) than could the valet.

But what happened if the productive capitalist used other capitalists to carry out some of these functions on his behalf? The labour employed by these other capitalists should be counted as productive according to Marx’s established definition since it enabled them to make a profit. But seeing things like this presented a problem. The profit did not arise from increasing the total amount of output any more than it did when productive capitalists directly employed people to perform the tasks. It simply amounted to the second capitalist getting a slice of the surplus value originally in the hands of the first capitalist. Marx concluded that from the point of view of capitalist production such labour was unproductive, even though this seemed to be based on a different definition of productive labour to that he used elsewhere. For this reason Jacques Bidet, for instance, has argued that Marx was inconsistent. [9] Yet it made sense in terms of the thing both Adam Smith and Karl Marx were interested in – the distinction between what advanced capitalist development and what retarded it.

So long as capitalists were operating in an economic environment in which capitalist production was not yet dominant, those who employed workers to provide personal services were providing them mainly to those whose wealth came from outside the capitalist system. The payments received, for instance, by the owners of a school constituted a transfer of resources into the capitalist sector from the pockets of pre-capitalist exploiters – resources that could then be used for productive accumulation. By contrast, the merchants or shopkeepers who handled the goods of the productive capitalists got their profit from the already created surplus value of the productive capitalist. They were not adding to total surplus value and with it the further accumulation of capital.

As Marx put it at one point:

To industrial capital the costs of circulation appear as unproductive expenses, and so they are. To the merchant they appear as a source of profit, proportional, given the general rate of profit, to their size. The outlay to be made on these circulation costs is therefore productive investment for mercantile capital ... And the commercial labour which it buys is likewise immediately productive for it. [10]

The competition between commercial capitalists with each other meant that each was subject to the same pressures as the capitalists involved in production to keep wages down to the value of labour power. For this reason their workers were exploited in the same way as workers for capital involved in production. The more a commercial capitalist held down the wages and increased the workload of his employees, the greater was the share he could keep for himself of the payment he got from the productive capitalists for providing services to them. If it took eight hours of socially necessary labour time to perform, say, a certain sales task, but only four hours to cover a sales worker’s wage, then the shopkeeping capitalist could pocket four hours worth of the surplus value supplied from elsewhere in the system.

But this did not mean that commercial labour could be equated with productive labour when it came to understanding the dynamics of the system as a whole. One created resources that could be used for further accumulation, and the other did not. That is why Marx is insistent:

Costs which enhance the price of a commodity without adding to its use value, which are therefore to be classed as unproductive expenses so far as society is concerned, may be a source of enrichment to the individual capitalist. On the other hand, as this addition to the price of the commodity merely distributes the costs of circulation equally, they do not cease to be unproductive in character. For instance, insurance companies divide the losses of individual capitalists among the capitalist class. But this does not stop these equalised losses from being losses so far as the aggregate social capital is concerned. [11]

The distinction between productive and unproductive labour is often seen as a merely scholastic question. But once seen in terms of what contributes to accumulation and what does not, it has enormous implications – including some that Marx himself never developed. What is “productive of surplus value” for the individual capitalist (the definition of productive labour Marx used in his notebooks) is not necessarily what is productive in terms of adding to the surplus value available to capital in general for accumulation. And it is this that is central for the dynamic of the system.

The scale of unproductive labour

The level of unproductive expenditures involved in sales and finance grew throughout the 20th century. Anwar Shaikh and Ertugrul Ahmet Tonak calculate that the number of workers employed in trade in the US grew from 10,690,000 in 1948 to 24,375,000 in 1989, and of those in finance and insurance from 1,251,000 to 7,123,000. Meanwhile, the number of productive workers only grew from 32,994,000 to 41,148,000. [12] Fred Moseley estimates the numbers in commerce as growing from 8.9 to 21 million between 1950 and 1980, and the number in finance from 1.9 to 5.2 million, while the productive workforce only grew from 28 to 40.3 million. [13]

The figures do not include the large number of managerial employees who Marx regarded as non-productive because they are involved in policing those who actually produce value. Simon Mohun has calculated that the growth in their numbers and remuneration caused the share of “unproductive” wages and salaries in the “material value added” in the US to rise from 35 percent in 1964 to over 50 percent in 2000. [14] These figures also understate the total growth of unproductive labour because they do not include employees involved in non-productive state functions like the military and the legal system.

Unproductive expenditures and waste production

There is another sort of labour that also has to be taken into consideration when examining 20th and 21st century capitalism. This is the labour that goes into producing commodities that are sold like other commodities but which do not then re-enter later rounds of production, whether as means of production or as wage goods. The labour producing luxury goods for the capitalist class falls into this category. So too does labour that goes into military weaponry. Although such labour has usually been regarded as “productive” by Marxists, it shares with non-productive labour the fact that it does not add to capitalist accumulation. For these reasons it was argued by Mike Kidron in the early 1970s that it should also be regarded as non-productive:

The ageing of capitalism ... opened a gulf between the two criteria of productiveness that he [Marx – CH] used interchangeably – employment by capital and augmenting capital ... Now that capital is king ... the two criteria are no longer congruous. Millions of workers are employed directly by capital to produce goods and services which it cannot use for further expansion under any conceivable circumstances. They are productive by one criterion and unproductive by the other ... Given the need to choose, productive labour today must be defined as labour whose final output is or can be an input into further production. Only such labour can work for capital’s self-expansion ... To spell it out, in late capitalism only part of the surplus can be used for the expansion of capital. The rest is waste product. [15]

More recently Alan Freeman has also suggested that the notion of unproductive labour has to be extended to involve the use of labour to produce things that are then used in an unproductive manner. “The workers who decked the European Bank for Reconstruction and Development in marble are just as unproductive as the clerks who now walk across it”. [16] Guglielmo Carchedi, by contrast, argues there is labour that is productive if it has created new value, even if this does not then contribute anything to the next round of accumulation. [17] Regardless of how it is categorised, the proportion of labour that is waste from the point of view of capital accumulation has become enormous. Kidron calculated that “three fifths of the work actually undertaken in the US in the 1970s was wasted from capital’s own point of view”. [18]

The state sector and non-productive labour

Expenditures by individual capitals that are neither going to capital investment nor to the wages of productive workers can be broken down into different categories:

  1. Those concerned with the disciplining of the workforce and ensuring it works flat out – expenditures on internal security, supervisory labour and time and motion measurement, checking on work speeds.
  2. Those concerned with keeping the allegiance of the workforce, e.g. expenditures on internal public relations, works bulletins, management-run works committees, subsidies to works sports teams.
  3. Those devoted to financial transactions, obtaining credit, bank charges, etc.
  4. Those devoted to sales, advertising, etc.
  5. Those concerned with keeping the workforce fit and able to work – company medical facilities, factory canteens, etc., in some cases the provision of housing for the workforce.
  6. Those concerned with training the workforce – what mainstream economists often call “human capital”.
  7. Expenditures on research and development.

Expenditures (a) and (b) are unambiguously unproductive. They create nothing and are only concerned with getting the maximum of already created value from the workers. Expenditures (c) and (d) are unproductive from the point of view of capital in general. They do not in any way add to the capacity of the system as whole to accumulate. But the individual firm can regard them as productive in the same way as Marx wrote that the individual merchant capitalist did – they serve to get control of surplus value which would otherwise go to rival firms. So advertising expenditure, for example, may be seen by the firm, like expenditure on new equipment, as a way of expanding its position in the market, of forestalling attempts to enter the market by other capitalists, and so on. Similar expenditure on patents and patent protection may be seen as a way of getting a stranglehold on the market (I will return to the other types of expenditures (e) to (g) below).

The growth of state expenditures in the course of the last century has involved states taking over partial responsibility for many of these outlays from the hands of the private capitals based in their national territory. So state expenditures can be broken down into categories playing the same or analogous functions to the expenditures of firms.

There are those expenditures which are clearly unproductive in terms of accumulation throughout the system as a whole. Among these are those concerned with protecting property, maintaining social discipline and ensuring the smooth reproduction of class relations; maintaining state-run or financed forms of maintaining popular allegiance to the system, such as state produced propaganda and subsidies to religious institutions; the perpetuation of the ruling ideology through sections of the educational system; maintaining the financial infrastructure of the system through the printing of national currencies and running central banks.

Alongside these there are expenditures beneficial to nationally based capitals in competition with foreign capitals, but which, like the individual capitalists’ expenditure on marketing or advertising, do not add to accumulation as a whole. This includes military expenditure, spending on export promotion schemes, negotiations with other governments over international trade and investment regulations, etc.

It was these unproductive expenditures that Marx referred to when he wrote:

Political economy in its classical period, like the bourgeoisie itself in its parvenu phase, adopted a severely critical attitude to the machinery of the state etc. At a later stage it realised and learnt from experience that the necessity for classes which were totally unproductive arose from its own organisation. [19]

Such growth in unproductive expenditures came to have a big impact on the dynamic of the system after Marx’s death.

Waste output and the system’s dynamic

Marx hinted at one important point about non-productive labour in his first attempt at a draft for Capital, the Grundrisse. He includes among the “moments” that can delay the rise in the organic composition of capital and the fall in the rate of profit:

the transformation of a great part of capital into fixed capital which does not serve as agency of direct production; unproductive waste of a great portion of capital etc. (productively employed capital is always replaced doubly, in that the posing of a productive capital presupposes a countervalue). The unproductive consumption of capital replaces it on one side, annihilates it on the other ... [20]

Marx is saying that if for some reason part of the surplus value available for investment is diverted into some other use, there is less new capital available for firms seeking innovations that will cut their costs, and the trend towards capital-intensive investment will be reduced. The same point was made much more explicitly in the 1960s by Mike Kidron – apparently without knowing that Marx had spelt the argument out. [21] He pointed out that Marx’s argument about the falling rate of profit:

rested on two assumptions, both realistic: all output flows back into the system as productive inputs through either workers’ or capitalists’ productive consumption – ideally there are no leakages in the system and no choice other than to allocate total output between what would now be called investment and working class consumption; secondly in a closed system like this the allocation would swing progressively in favour of investment.

If the first assumption, that all outputs flow back into the system, was dropped – in other words, if some of these outputs are lost to the production cycle – then there would be no need for investment to grow more rapidly than the labour employed. The law of the falling rate of profit would not operate. “Leaks” of surplus value from the closed cycle of production/investment/production would offset the tendency of the rate of profit to fall. [22]

As Kidron put it in a later work:

In Marx the model assumes a closed system in which all output flows back as inputs in the form of investment goods or wage goods. There are no leaks. Yet in principle a leak could insulate the compulsion to grow from its most important consequences ... In such a case there would be no decline in the average rate of profit, no reason to expect increasingly severe slumps and so on. [23]

The argument is impeccable, and Kidron goes on to suggest the form these leaks have taken:

Capitalism has never formed a closed system in practice. Wars and slumps have destroyed immense quantities of output, incorporating huge accumulations of value, and prevented the production of more. Capital exports have diverted and frozen other accumulations for long stretches of time. [24]

As we saw in Chapter Four, Henryk Grossman had recognised that imperialism in diverting surplus value overseas had temporarily reduced the upward pressures on the organic composition of capital in the domestic economy and the therefore tendency to crisis. He also at least partially anticipated Kidron’s point about the effect of military expenditure. He noted that, while wars were enormously destructive of use values, they had the effect of easing the purely economic contradictions of capitalism since they “pulverise values” and “slow down accumulation”. By reducing the tendency for accumulation to rise faster than the employed labour force they countered the fall in the rate of profit:

The destructions and devaluations of war are a means of warding off the immanent collapse [of capitalism], of creating a breathing space for the accumulation of capital ... War and the destruction of capital values bound up with it weaken the breakdown [of capitalism] and necessarily provide a new impetus to the accumulation of capital ... Militarism is a sphere of unproductive consumption. Instead of being saved, values are pulverised. [25]

Military expenditure is a particular form of waste that can appeal to capitalists connected to a particular state. For it enhances their capacity to struggle for control of worldwide surplus value with rival capitalists. It is functional for nationally based complexes of capital in the same way that advertising is for individual firms, even while wasting resources for the system as a whole. It was therefore a characteristic phenomenon of the classical form of imperialism that led to the First World War – and it survives today in the massive arms spending of the United States in particular.

The logic of arms-based economic expansion has escaped many Marxist economists. It is absurd, they argue, to see a deduction by the state from the total surplus value as somehow countering the tendency for surplus value to grow more slowly than total investment costs, and so overcoming the fall in the rate of profit. What they have failed to understand is that this “absurdity” is just part of the greater absurdity of the capitalist system as a whole, of its contradictory nature. They have not seen that engaging in military competition can be just as much a “legitimate” capitalist goal as engaging in economic competition for markets.

As we saw in the last chapter, one of the greatest followers of Marx, Rosa Luxemburg, could not understand how capitalism could continually expand the value embodied in means of production without producing more goods for consumption. Similarly, these Marxists have not been able to understand how capitalism could possibly benefit from continually expanding the means of destruction. They have been so bemused by the irrationality of what capitalists are doing as to try to deny that this is how the system works.

But such expenditures had enormous implications for capitalism in the latter part of the 20th century. Waste expenditures played a contradictory role. They reduced the amount of surplus value available for productive investment, so counteracting the tendency towards over-rapid accumulation and crisis. But the eventual effect in slowing down accumulation was to create a whole new series of problems for the system, as we will see in Chapter Nine.

Welfare and the supply of labour power

Not all the state expenditures listed earlier fall into the unproductive category as narrowly defined or into the wider category of waste. State-financed research and development (corresponding to category (g) in the list above) that feeds through into aiding accumulation in the wider economy clearly plays a role for those capitals that benefit from it, similar to that of dead labour embodied in means of production. But what of expenditures on health, education and welfare services (equivalent to the expenditures (e) and (f) of individual capitalists)? Here it is necessary to examine something Marx only discusses in passing – the reproduction of the working class that capitalism needs for exploitation.

The first industrial capitalists of the late 18th and early 19th centuries in Britain did not have to worry over much about the supply of labour power. It was available in abundance once “primitive accumulation” had driven sufficient peasants from the land. They assumed they could bend former peasants and their children to the discipline of unskilled machine minding [26], while relying on drawing men trained as artisans into the factories for more skilled work. For these reasons, Marx, who dealt at length with primitive accumulation and the treatment of workers in the factories, virtually ignored the problem for capitalists of getting a labour force with the right physiques and skills. Yet by the time of his death, the spread of capitalist industry to ever newer new sectors of production was making the supply and management of labour power – outside as well as inside the factory – something of increasing concern to those promoting capitalist accumulation.

The individual capitalist aimed to pay the individual worker just enough by the hour, day or week to keep him or her fit and motivated to work. But this did not cater for a number of important things if labour power of the right quantity and quality was going to be available for the capitalist class as a whole over time. It did not take into account the need for workers to learn necessary skills nor did it sustain them through periods of unemployment so as to be able to supply their labour power when the crisis ended. It did not deal with the problem of workers temporarily losing their capacity to be productively exploited through illness or injury. And it did not provide for the upbringing of working class children who would be the next generation of labour power. [27]

There were various ad hoc attempts to deal with each of these problems through the 19th century. Religious and other charitable funds provided some relief for the unemployed or the sick. Pressure was put on working class women to bear the burden of child rearing through the propagation of ideologies that treated men as the wage earner and men’s wages as a “family wage” (even though working class women always worked to some extent and a man’s wage was rarely adequate to keep a family). [28] Some firms would provide housing under their own control – and sometimes minimal health facilities as well – for their workforces. Groups of skilled workers would run funds to provide for periods of unemployment or sickness. Firms would incorporate into the factory system a version of the apprenticeship system of pre-capitalist artisanship, with youngsters learning a trade by working under skilled workers for five or seven years on minimal wages.

But over time it became clear that the ad hoc methods were inadequate and that the state had to take over many of the tasks from private capitalists and charitable concerns. In Britain it intervened as early as the 1834 Poor Law to ensure that the conditions under which the unemployed or the infirm could get poverty relief were so arduous that those who could work would, however low the pay. In 1848 it established a Board of Health to act against the spread of diseases in working class areas – which was affecting richer areas too. Over the decades it was cajoled into limiting the working hours of children and barring women from occupations that might damage their capacity to bear and bring up the next generation. In the 1870s it moved to set up a state system of elementary education and to encourage the building of homes for skilled workers. Then in the first decade of the 20th century it made the first moves to coordinate the various ad hoc measures of the previous 70 years into national structures to provide minimal social insurance benefits for unemployment, old age and sickness. [29] The impetus to do so came from the shock of discovering in the course of recruitment for the Boer War how few of the working class were healthy enough to undertake military service. Ann Rogers has summarised the reaction of the upper and middle class:

The belief that change was necessary if Britain was to compete successfully with Germany and the United States remained central. Whether the argument was formulated by the Fabians or by Liberal imperialists the concentration was on the damage that poverty was doing to society rather than the misery it caused individual workers ... The underlying reason for the desire to improve the health of the working class was the need for a healthier labour force in the factories and the army. [30]

These measures were not simply a result of capitalists getting together and deciding what was rational for their system. They came into being only after recurrent campaigns involving upper class philanthropists with a conservative disdain for the money-grabbing grubbiness of capitalism, middle class moralisers about working class behaviour, political opportunists out to get working class votes, factory inspectors and doctors with professional concerns about people’s safety and well-being – and, alongside them and often independently of them, trade union and socialist activists. But such coalitions framed the projects they pursued in terms of what they saw as rational for capitalism. And that meant what was necessary to supply it with pools of sufficiently healthy and skilled labour power. This was shown clearly by one feature that characterised the reforms of the early 20th century just as much as it had the charitable efforts of the early 19th century. Any benefits were always to be provided in such a way as to coerce into seeking work all those who were fit and able. The principle of “less eligibility” had to apply: getting the benefit must still leave the recipients worse off than the worst paid work. What is more, the benefits were not meant to come from a diversion of value from capital to labour, but by a redistribution of income within the working class through the “the insurance principle”. Weekly payments from those able to work were to sustain those unable to do so because of sickness or unemployment.

The role of the state in the supply, training and reproduction of labour power grew through the 20th century, reaching a peak in the long boom from the mid-1940s through to the mid-1970s, and continuing into the new period of crises which followed. All through this the “welfare state” continued to be tailored to the interests of nationally based capitals, even when the impetus for extending its role came from below, as during the Second World War, when the British Tory politician Quintin Hogg famously declared, “if you do not give the people social reform, they are going to give you social revolution”. [31] The British Labour minister of the 1940s Aneurin Bevan argued that public health measures had become part of the system, “but they do not flow from it. In claiming them capitalism proudly displays medals won in battles it has lost.” [32] The fact, however, is that those who formulated them – including Bevan – did so in ways that could fit in with the needs of the system.

This has important implications for the labour power that goes into such services – and for the people who supply it. There is a widespread tendency for Marxists – as well as some non-Marxists [33] – to insist that such labour cannot be productive since it does not produce commodities directly. But that also applies to much labour inside any capitalist enterprise, which is merely a precondition for other labour that produces the final products. It is productive as part of the labour of the “collective worker” [34] in the enterprise. A fully trained carpenter or bricklayer can be many times more productive than an unskilled one; a fully trained toolmaker can do jobs an unskilled labourer is incapable of. The labour of those who train them is adding to the capacity of the collective worker to produce value. And they are exploited, since they are paid the value of their labour power, not of the training they provide. There can be a debate over exactly how the skills added by their labour fit into Marx’s categories: are they to be equated with plant and equipment as a form of constant capital or as simply enhanced labour power, as variable capital? [35] There are also debates between individual firms over the merits of undertaking training programmes. They may gain in the short term, but what is to prevent other firms “poaching” their skilled labour without ever having paid for its training. [36] Finally, there are arguments about how to characterise the labour used to train other workers: is it “productive” or “indirectly productive”? But there should be no doubt about its role in increasing overall potential output and productivity: it is part of the total productive labour of the firm and of the system as a whole. [37]

A big portion of the labour that goes into the educational system plays an identical role in providing the skills capital needs, although in this case the skills are not available simply for individual capitalists, but for all the capitalists operating from within the state that provides it. The training in skills which future workers get from a teacher in an educational institution adds to the amount of socially necessary labour they can produce in an hour in exactly the same way as the training they might get inside an enterprise. And the cost of the training is part of the cost of providing labour power, just as much as the wage that goes into buying the food, clothing and shelter the workers require. Enterprises under modern capitalism require labour power with at least minimal levels of literacy and numeracy. The teachers who provide this have to be considered as part of the collective worker, ultimately working for the complex of nationally based capitals that the state services. Apologists for capitalism recognise this inadvertently when they refer to the provision of education as “adding to social capital” and demand “value added” in schools.

The same general principle applies to health services that cater for actual, potential or future workers. Spending on keeping the workforce fit and able to work is in reality a part of the wage even when it is paid in kind rather than in cash and goes to the workers collectively rather than individually. In Marx’s terminology, it is part of “variable capital”. This is absolutely clear in countries like the US where healthcare is provided for most workers through insurance schemes provided by their employers. It should be just as clear in countries like Britain where the state provides them on behalf of the nationally based capitals. The popularly used term “social wage” is an accurate description. It is just as accurate when applied to unemployment benefits available only to those who show they are able and willing to work, and to pension schemes dependent on a lifetime of labour. The capitalist wants contented workers to exploit in the same way that a farmer wants contented cows. Workers cannot be expected to labour with any commitment to their work unless there is some sort of promise that they will not starve to death once they reach retirement age. As Marx put it, there is a historically and socially determined element to the cost of reproducing labour power as well as a physiological one.

But labour power is not an object like other commodities, which are passive as they are bought and sold. It is the living expression of human beings. What from a capitalist point of view is “recuperation of labour power” is for the worker the chance for relaxation, enjoyment and creativity. There is a struggle over the social wage just as over the normal wage, even if both are, to a certain degree, necessary for capital.

The problem is compounded from capital’s point of view by the fact that not all welfare provision is in any sense productive. A good portion of it is concerned solely with maintaining the existing relations of exploitation. Studies of the schooling of working class children in the 19th century emphasise the degree to which what was involved was not education in skills so much as inculcating into them discipline and respect for authority. [38] Not until late in the 19th century did a concern with basic skills for the workforce begin to become a central preoccupation for British capitalism facing foreign competition. [39] Today disciplines like economics and sociology are about trying to reproduce bourgeois ideology, while others like accountancy are concerned with the unproductive redistribution of surplus value among members of the capitalist class.

If capital has no choice but to tolerate these unproductive “expenses of production”, there are other elements in welfare expenditure it would love to be able to do without and does its utmost to minimise. These go to those who are not needed as labour power (the long-term unemployed without needed skills) or are incapable of providing it (the chronic sick and disabled). It has a similar attitude to provision for the mass of the elderly, but is restrained to some degree by its need to give the impression to currently employed workers that their future is assured. Marx pointed out that there exists, alongside the “reserve army of labour”, able to enter the active labour force when the system undergoes periodic expansion (and in the meantime exercising a downward pressure on wages), a surplus population in whose survival it has no real interest apart from that of warding off rebellion and preventing a demoralising impact on the employed working class.

The history of welfare legislation over the last 180 years has been a history of attempts to separate that provision which is necessary for capital in the same way that wage payments are and that which is unnecessary but forced on it by its need to contain popular discontent. This finds expression in repeated debates among those who would manage national capitalisms over how welfare policy interacts with labour market policy, among mainstream economists about the “natural” or “non-inflationary” level of unemployment, and among sociologists and social work theorists about the “underclass”.

The division between social expenditures that are in some way productive for capital and those that are non-productive cuts across some of the normal ways of dividing up national budgets. So education is both training for productive labour and also training for unproductive forms of labour (e.g. in sales promotion or finance) and the inculcation of bourgeois ideological values. Health services and unemployment benefits both keep the workforce fit and ready to provide labour power and are mechanisms for maintaining social cohesion by providing at least minimal provision for the old, the infirm and the long-term unemployed. These ambiguities become important whenever capital finds the costs of state provision begin to cut into profit rates.

At such points states come under the same pressure as do big capitals when faced with sudden competition – the pressure to restructure and reorganise their operations so as to accord with the law of value. On the one side this means trying to impose work measurement and payment schemes on welfare sector employees similar to those within the most competitive industrial firms. On the other side it means cuts in welfare provision so as to restrict it as much as possible to servicing labour power that is necessary for capital accumulation – and doing so in such a way that those who provide this labour power are prepared to do so at the wages they are offered.

These pressures grow as managing labour power becomes more important for the state. In the process, employees working in the welfare, health and educational sectors who could at one stage of capitalist development regard themselves as part of the professional middle class – with salaries and conditions comparable to lawyers or accountants – find themselves subject to a traumatic process of proletarianisation. This, as we shall see, adds to the problems that beset national capitalist states as they attempt to cope with sudden crises. Public expenditures become a central focus for class struggle in a way in which they were not in Marx’s time.

* * *


1. See Marx’s comments on Smith in this regard in Karl Marx, Theories of Surplus Value, Volume One (Moscow, n.d.), pp. 170 and 291.

2. “Productive and unproductive labour is here throughout conceived from the standpoint of the possessor of money, from the standpoint of the capitalist, not from that of the workman”, as above, p. 155.

3. As above, p. 156.

4. Marx, Capital, Volume One, p. 192.

5. Guglielmo Carchedi, Frontiers of Political Economy, p. 40.

6. According to Enrique Dussel, Marx’s argument in his notebooks is that “unproductive labour” will, “with only minor exceptions, only perform personal services”. Enrique Dussel, Towards an Unknown Marx (Routledge, 2001), p. 69.

7. Marx, Theories of Surplus Value, Volume One, p. 170.

8. Marx, Capital, Volume Three, p. 293.

9. Jacques Bidet, Exploring Marx’s Capital (Leiden, Brill, 2007), pp. 104–121.

10. Marx, Capital, Volume Three, p. 296.

11. Marx, Capital, Volume Two, p. 137.

12. See Table F.1, in Anwar Shaikh and E.A. Tonak, Measuring the Wealth of Nations (Cambridge University Press, 1996), pp. 298–303.

13. Fred Moseley, The Falling Rate of Profit in the Post War United States Economy, p. 126.

14. Simon Mohun, Distributive Shares in the US Economy, 1964–2001, Cambridge Journal of Economics, 30:3 (2006), Figure 6.

15. Michael Kidron, Waste: US 1970, in Capitalism and Theory, pp. 37–39.

16. Alan Freeman, The Indeterminacy of Price-Value Correlations: A Comment on Papers by Simo Mohun and Anwar Shaikh, available at

17. Guglielmo Carchedi, Frontiers of Political Economy, pp. 83–84.

18. Michael Kidron, Waste: US 1970, in Capitalism and Theory, p. 56.

19. Marx, Theories of Surplus Value, Volume One, p. 170.

20. Karl Marx, Grundrisse, pp. 750–751.

21. The Grundrisse was not published in English until 1973.

22. Michael Kidron, International Capitalism, International Socialism 20 (first series, 1965), p. 10, available at Carchedi recognises the different effect of production of non-reproductive goods (which he calls “non-basic”, using the terminology of the Ricardians he is criticising) on the dynamic of accumulation without explicitly drawing Kidron’s conclusion about the rate of profit. “Non-basics cannot”, he writes, “be the transmission belt through which value changes in the previous production process are carried into the next one” – G. Carchedi, Frontiers of Political Economy, p. 83.

23. Michael Kidron, Capitalism and Theory, p. 16.

24. Michael Kidron, Capitalism: the Latest Stage, in Nigel Harris and John Palmer, World Crisis (London, Hutchinson, 1971), reprinted in Capitalism and Theory, pp. 16–17. For a longer exposition of this insight which takes up criticisms from Ernest Mandel, see my Explaining the Crisis, pp. 39–40 and 159–160.

25. Henryk Grossman, The Law of Accumulation, pp. 157–158.

26. T.S. Ashton, The Industrial Revolution (London, Oxford University Press, 1948), pp. 112–113. They did then have to enforce on them their own discipline of work, timed by the clock. See E.P. Thompson, Time and Work-Discipline, in Customs in Common (London, Penguin, 1993), pp. 370–400.

27. For a development of this argument with historical references, see Suzanne de Brunhoff, The State, Capital and Economic Policy (London, Pluto, 1978), pp. 10–19.

28. See Lindsey German, Sex, Class and Socialism (London, Bookmarks, 1989), pp. 33–36.

29. T.H. Marshall, Social Policy (London, Hutchinson, 1968), pp. 46–59.

30. Ann Rogers, Back to the Workhouse, International Socialism 59 (1993), p. 11.

31. Hansard, Parliamentary Debates, 17th February 1943, Col. 1818.

32. Quoted in T.H. Marshall, Social Policy, p. 17.

33. Notoriously, the attack on public sector employment, Robert Bacon and Walter Eltis, Britain’s Economic Problem: Too Few Producers (New York, St Martin’s, 1976).

34. Marx’s term in Capital, Volume One, p. 349.

35. If they are counted as part of constant capital, this is a peculiar form of constant capital that can walk away from the firm and work elsewhere, and in some interpretations this leads to the view of skilled workers as in some way possessors of their own “human capital” and a part of their wages is viewed as a “return” on this capital. It should be added, however, that disputes over this question can be in danger of turning into pure scholasticism, since in any case the costs of training add to the investment costs of a firm. At the same time, insofar as training is generalised across the system as a whole, it increases average labour productivity and so serves to reduce the value in terms of socially necessary labour of each unit of output – and in doing so reduces the gains accruing to the individual capitalist from undertaking the training.

36. The “free rider” problem: see, for instance, Mary O’Mahony, Employment, Education and Human Capital, in R. Floud and P. Johnson, The Cambridge Economic History of Modern Britain, Volume Three (Cambridge University Press, 2004).

37. Assuming the labour they train is going to end up as productive labour.

38. See Richard Johnson, Notes on the Schooling of the English Working Class 1780–1850, in Roger Dale and others (eds.), Schooling and Capitalism (London, Routledge & Kegan Paul, 1976), pp. 44–54. See also Steven Shapin and Barry Barnes, Science, Education and Control, in the same volume, pp. 55–66.

39. For a full account of developments in the major industrial countries of the time, see Chris McGuffie, Working in Metal (London, Merlin, 1985).

Last updated on 05 April 2020