Theories which seek to explain the inadequacy of historical leftist and socialist traditions are a most cherished tradition of leftism and socialism. A common point of criticism is the fact that many wage-labourers are not engaged in “classic” productive work of the kind many Marxists are familiar with, which is to say industrial production in which the labor force, using the means of production, transforms inputs, resulting in a mass of commodities of a greater value than the prior input.
Industrial productive workers have consistently constituted a minority within any capitalist society. Such workers are the locus of value production in that they multiply usable values. There are also other forms of labour in the world which are or have been essential to the functioning of production, but which do not produce surplus values themselves, such as subsistence agriculture, or reproductive labour: the caring labours performed primarily by subjugated women, deeply necessary to reproduce both workers and the working day.
The utility of such forms of labour, and the basis of their exploitation by capitalism, can be explained by their connection to the locus of value creation. In this sense, they cohere to known laws of production. However, what has become increasingly identifiable over the course of the 20th century are forms of labour which are seemingly disconnected from the locus of value creation entirely. Labour that is, as far as one can tell, entirely unproductive.
I will be tracking the development and expansion of these forms of unproductive labour in the work of three authors. The Frankfurt school philosopher Jürgen Habermas calls this Reflexive Labour, the anarchist anthropologist David Graeber calls it Bullshit Jobs, and imperialism theorist Zak Cope might call it a symptom of Labour Aristocracy.
These three authors, in examining similar phenomena, draw wildly different conclusions. Habermas, ever the optimist, believes that the existence of new forms of labour which negate crisis mean late capitalist societies can continue to improve as the state is transformed through engagement in the public sphere by a revived civil society. Graeber is reluctant to draw conclusions, and tentatively points towards Universal Basic Income as a means of separating work from wages. Cope argues that we must throw away the dominant tendencies of Leftism in the Imperial Core, and locates his hopes within the most marginalised sections of the working class, as well as in the workers and farmers of the periphery. I will attempt to draw my own conclusions around how marxists might engage with the theories described, as well as what the implications of reflexive labour might be for the workers’ movement.
Habermas and Reflexive Labour
Central to much of Jürgen Habermas’ thought in Legitimation Crisis is the idea that Capitalism has transformed since the time of Marx to such a degree that it constitutes a new mode of produc- tion, and, with it, new relations of production and class hierarchies. Habermas borrows from the historical materialism of Marx and Engels, positing that instead of a trajectory through modes of production from Feudalism, to Capitalism, to Lower Communism, there has instead been a trajectory from feudal “Traditional Society,” to “Liberal Capitalism,” to a “Late” or “Advanced”Capitalism.
This distinct mode of production came into place in four stages. The state, as the most conscious and forward-thinking sector of capitalist societies, secures the conditions of its continued existence in property and contract law, rescuing the market from self-defeating side-effects, promoting market competitiveness, and securing national sovereignty. The state then undergoes market-complementing actions, increasing the size and competitiveness of markets. These actions give way to market-replacing actions, such as regulations, price adjustments, stimulus, and other subversions of free-market principles which affect the organising principles of society, creating new “classes” such as the public service. Finally the state compensates for the dysfunctions of the earlier Liberal Capitalist period, responding to any organised aggrieved section of society such as unions, minorities, or failing industries, taking charge of the consequences of unregulated industry and propping up unprofitable sectors through subsidies.
A society which has undergone this change should possess three characteristics. The market-replacing actions of the state lead to new forms of surplus value accumulation, such as new forms of labour, new forms of reinvestment, and new social classes. A “politicised” wage structure develops in which capitalists, unions, and the state collaborate to determine wage levels, a process which further changes class dynamics by promoting collaborationism. Finally the contradictions between the new society and the old “liberal capitalism” produce a legitimation crisis which creates demands around “use-values.”
But why would a capitalist class allow these radical state reforms to take place? Habermas calls the constellation of class relations described by Marx “improbable,” which is to say that the market allocation of wages was inefficient, illogical, and unstable. A “political” (class-collaborationist) system of wage allocation was a natural solution more likely to avoid the classical overproduction crisis which Marx described in Capital’s third volume.
Proving that late-capitalism had evolved mechanisms to avoid an overproduction crisis is essential to Habermas. To do so he attempts to prove that the Tendency of the Rate of Profit to Fall (TRPF) has been offset by mechanisms which not only delay crises (something Marx accounted for as countervailing tendencies), but prevent it entirely. A basic knowledge of Marx’s crisis theory is necessary before understanding Habermas’ supposed negation: technology leads to more and more efficient means of production, which leads to greater outputs of use-values per unit of capital invested. This technological progress also means that labour is automated in order for businesses to stay competitive, leading to an increase in the Organic Composition of Capital (OCC). Since living labour-time is the locus of value creation, the lack thereof leads to a decline in the rate of profit. Marx considered this development of crisis theory to be his greatest innovation in part because it proved the long-term instability of capitalist productivity, imposing a hard limit on any optimistic projections of capitalist futures. Any theorist who wishes to propose non-revolutionary solutions to crises, without refuting each of Marx’s theories in-detail, would therefore have to disprove the TRPF and provide evidence of ongoing countervailing tendencies.
In Legitimation Crisis, Habermas proposes several countervailing tendencies, but the two most relevant to us are the creation of masses of public service workers, and the creation of reflexive labour jobs.
Habermas believed that the increasingly massive state bureaucracies of advanced capitalism would themselves delay or negate the TRPF by steering markets away from crisis; buying up commodities at above-market rates, employing workers in an ever-expanding bureaucracy, and effectively compensating for any market failure through social spending.
This might appear to be one of the more dated components of Habermas’ thought as it is clear to any modern reader that the meteoric rise in public spending and public sector growth he describes was short-lived. Figures from most western countries show a consistent rise in social spending around the time Habermas was writing, with most countries plateauing in the late 1970s, and some (particularly late-adopters of neoliberal policy, such as Sweden) declining. However, in dollar terms the change is less dramatic than might be imagined, and many states, including the UK and US, soon resumed the general trend towards increased spending.
Spending generally increased under Neoliberalism, but privatisation actually leads to reduced efficiency and there has been a corresponding decrease in social outcomes per dollar spent. The missing value has instead gone to an increasingly thick layer of parasitic companies which exist to milk government contracts – job-searching or temping agencies, call centres, pension providers and the like – all fulfilling roles once performed by the state, and each with their own attendant armies of managers and administrators.
While Habermas did not account for these changes, this does not in itself invalidate his argument that public spending would constitute a countervailing tendency against the TRPF. However when we consider a groundbreaking study by Esteban Ezequiel Maito, which measured the rate of profit across fourteen countries, correctly adjusted for growth in turnover speed, we can see that the fall in the rate of profit accelerated precisely when social spending was being expanded in the west, and this acceleration halted precisely when social spending plateaued.
Habermas’ other key countervailing tendency is reflexive labour, defined as “labour applied to itself with the goal of increasing the productivity of labour.” This begins as a collective property of mankind – naturally occurring human ingenuity, or inventions brought about by necessity and chance. Habermas argues that Marx considered ingenuity to be a background phenomenon, like air or land, and not a fundamental part of production. This is arguable, but for Habermas it is clear that invention is a process that can be forced to occur in a predictable and controllable fashion, but only by an “advanced” capitalist state, as in liberal capitalism there was no systematically utilizable, productivity-increasing innovations. In advanced capitalism, the state can incorporate teaching, science, engineering, and management innovations into the production process by funding research institutes, schooling, and think-tanks. In neoliberal capitalism of the last few decades, we would call much of this reflexive labour force the tech or information industry, and the degree to which this form of labour has become uncoupled from the state will become important later.
In terms of the specific way in which these reflexive labourers affect productivity, Habermas is somewhat contradictory. It is important to understand that increases in productivity typically lead to lower profitability in the long term. Capitalists must continuously invest into fixed capital in order to remain competitive. This increases the Organic Composition of Capital (OCC) and thus reduces the rate of profit. These functions can be expressed as an equation to find an annual rate of profit:
In this equation we can see that a reduction in the value of wages relative to the means of production will bring about a reduction in the rate of profit. In Habermas’ model the combined efforts of reflexive labourers act simultaneously to add to the output of direct producers (increasing s), as well as additional labourers themselves (increasingv), effectively reducing the value composition of capital (which correlates with the aforementioned organic composition) and increasing productivity. While it would certainly be fortuitous for capitalists if this were the case, as productivity could be increased endlessly with no downside, it makes little sense while profitability decline continues to persist. Instead, one or the other must be true. Since reflexive labourers are still living wage-labourers, they add to the overall wages invested in production, increasing the value composition, and reducing the surplus value produced per unit of labour time in the short term. This has the effect of increasing long-term profitability, at the cost of short-term profitability.
The capitalist alone has little incentive to engage in this process as it increases the number of labourers involved without any short-term increases to profitability. It’s for this reason that Habermas is correct when he says that there are no productivity-increasing innovations that are intrinsically part of capitalism as Marx described it. It’s strange then that Habermas believes that the advanced capitalist state is so much more forward thinking than its captains of industry, as if government bureaucrats read enough Marx to understand the value of counter-acting the TRPF in anything but the most crude and superficial terms, such as a decline in GDP.
It’s difficult to think of a critique of Habermas’ work more encompassing than to point out the anti-marxist orientation of his world-view. This is not to say that Habermas is an anticommunist – he is no cold warrior – but his work is typified by a dogged determination to come to conclusions contrary to Marx no matter how improbable. This was noticed as early as 1971 by Habermas’ contemporary, Göran Therborn, who notes:
In Habermas’s academic oration there is no room for any workers, still less for a revolutionary proletariat. The ‘emancipatory interest’ (to use a term of Habermas’s) which guided Horkheimer and which he regarded as the ‘only concern’ of the critical theorist, was ‘to accelerate a development which should lead to a society without exploitation’. It was clear from the context that this development was the proletarian socialist revolution. Habermas’s interest in emancipation produces merely 'self-reflection'
Habermas’ distinctly anti-revolutionary development of Frankfurt school critical theory contrasts poorly with the useful contributions of other Frankfurt school writers like Walter Benjamin, in that it is “pure negation”; his objective being to insubstantiate a collection of sociological orthodoxies (mostly Marxist ones) without any real regard for developing an alternative theory of comparable coherence. He achieves this by criticising an imagined, dogmatic, ultra-orthodox Marx. For example when Habermas explains that social relations have lost a supposedly Marxian pre-determinative quality through the “politicisation” of the labour market, he neglects to mention that Marx described states suspending the laws of the “apolitical” labour market for economic gain several times. Most of the characteristics of Habermas’ “advanced” capitalism either existed long before Habermas said they should, or have ceased to exist since Legitimation Crisis was published.
All this said, the concept of a class of reflexive labourers who exist as a countervailing tendency to the TRPF is an interesting one, albeit a revelation undercut and distorted by Habermas’ anti-Marxian analytic framework which necessarily ignores the colonial underpinnings of his “advanced” capitalism. That such higher-order intellectual labourers can exist is predicated on imperialist superprofits, as we will see. So too is the impact of anticommunism itself in the formation of this class under-emphasised by Habermas, as the rise of such workers was inexorably connected to the west’s herculean crusade against Soviet communism. This is easier to see in the context of a modern information economy built on the infrastructure (internet, satellite and wireless communications) of the cold war.
The greatest mistake of Habermas’ Legitimation Crisis was that he failed to situate his new epoch on a fluid historical continuum: his “advanced” capitalism would be superseded by later forms eerily similar to the supposedly obsolete “liberal” capitalism of the 19th century, and his reflexive labourers, once the heralds of a new crisis-free capitalism, would be subsumed into the logic of the neoliberal decades.
Graeber and Bullshit Jobs
While Habermas is single-mindedly concerned with negating the theories of others, in Bullshit Jobs anarchist anthropologist David Graeber is equally single-minded in his determination to prove, without doubt, the existence of the titular phenomenon, with less emphasis on why it exists or its implications.
In addition to jobs which are entirely pointless, Graeber suggests that most jobs are undergoing a process of bullshitisation, or a decline in the time spent performing socially-necessary labour. He uses the example of British nursing, wherein nurses reported that up to 80% of their time is spent doing paperwork at the expense of their primary care duties. Likewise, in many office jobs the time spent doing primary duties is declining, with one study showing a decline in hours spent performing primary duties from 49% of total working hours in 2015, to 39% in 2016. Even in workplaces where workers self-report that their work produces real use values, there is an overwhelming trend towards an “inverted pyramid” structure wherein the workers who perform the most clearly necessary labours (direct production, cleaners, maintenance workers, logistical staff etc.) are increasingly understaffed, underpaid, and subject to “scientifically” precise productivity quotas, while the top of the pyramid (management, administrators, those designing the aforementioned quotas) are multiplying exponentially. To have such a high ratio of management and clerical staff to labourers makes little sense, as the multitudes involved are above and beyond that which would be necessary to reproduce class relations within the workplace, or to administer scientific management techniques.
It is easy to challenge Graeber’s claims by pointing out that alienation of the worker from the act of production, the segregation of the processes of production into a series of discrete, repetitive motions, could account for workers under-reporting their output of use values, as this would prevent workers from seeing the end result of any action.
However, while I believe that while this is a factor to consider, I do not think it accounts for all of Graeber’s revelations. Alienation of the worker from production, such as it extends outside the bounds of an individual workplace, typically applies more so when looking “up” the chain of production rather than down. To give one example, Congolese cobalt miners have at least a loose understanding that the resource they produce will end up in batteries for consumer electronics. By contrast it is doubtful that workers engaged in final assembly of those batteries would know where the cobalt hydroxide comes from, in part because that information is deliberately obscured. For this reason I believe self-reportage is an appropriate means of gauging the degree to which workers are actually producing values.
From this perspective, the revelations in the first half of Bullshit Jobs, which is to say the parts that are wholly based in worker testimony, are relatively incontrovertible. What follows in the latter half of the book is much less clear-cut, as Graeber avoids making any definitive arguments as to why bullshit jobs exist, or at least, avoids making any sweeping political-economic claim as to their origin. To be fair, Graeber warns his readers many times that his work is light on conclusions.
As such it is more productive to talk about the dominant recurring themes in Graeber’s broad discussion of tendencies which correlate with bullshitisation, rather than misrepresenting the book as a series of discrete arguments around causation. These themes include privatisation, financialisation, infeudation, and informationalisation.
This first tendency, in which social-democratic self-conscious make-work programs give way to much more inefficient neoliberal job-creation programs, is a perfect illustration of something I referred to earlier while discussing Habermas’ optimistic treatment of the public sector: neoliberalism is an extension of the goals of social democracy with added inefficiency. As Graeber says:
For much of the twentieth century state socialist regimes dedicated to full employment created bogus jobs as a matter of public policy, and their social democratic rivals in Europe and elsewhere at least colluded in featherbedding and overstaffing in the public sector or with government contractors, when they weren’t establishing self-conscious make-work programs...the new neoliberal age was supposed to be all about efficiency. But if patterns of employment are anything to go by, this seems to be exactly the opposite of what actually happened.
The mass of public sector jobs that Habermas described as something which could forestall or even negate the worst excesses of capitalism, would likely have been bullshit jobs themselves, and indeed these inflated state bureaucracies are one key source of bullshitisation today. However, such public sector bureaucracies are a lesser factor than their privatised counterparts. Most aspects of the public sector, if not privatised outright, at least outsource much of their function to private contractors, or else are subject to budgetary responsibility rules that mean that many state functions are subject to the same profit motive as private industry. These privatised replacements for what once functioned as public services run at higher costs, and also tend to employ far more managerial and administrative staff. From 1975 to 2005 the number of administrators and managers at privatised institutions in Britain increased at more than twice the rate it did in public ones, an expansion of 135% vs. 66%.
Once again, neoliberalism proves to be less of a process by which states shrink and more of a gradual process of states expanding so as to support a thick layer of parasites: companies which milk government contracts by finding the least efficient means of dealing with public issues. Graeber’s examples of various FIRE sector (Finance, Insurance, Real-Estate) worker testimonies illustrate the link between privatisation, bullshitisation and financialisation well.
As states privatised their huge public sectors largely for ideological reasons, privatisation emerged as an alternative means of fulfilling some of the state’s functions in terms of stimulus and job creation.
By fulfilling government functions through contracts, the state gained a means of ongoing, direct stimulation of the economy by allowing companies to milk tax dollars directly. In return, corporations fulfilled the state’s political goals of reducing unemployment to acceptable levels. Many of the financial functions of the state were also given over in the great sell-off, and now the finance sector is primarily concerned with “colluding with government to create, and then trade and manipulate, various forms of debt.” Having been given the means to create debt, “it creates money (by making loans) and then moves it around in often extremely complicated ways, extracting another small cut with every transaction.”
The necessity of accruing political capital to facilitate collusion between the state and finance capital is one factor leading to infeudation. Graeber believes that the entanglement of the economic and political since the 1970s has led to a neo-feudal dynamic in many corporate structures. The purpose here is to loot, steal, or extract rent from what little productive labour is being done, in order to create an entourage of followers which can create political favour. Graeber here goes so far as to offer a general theory of infeudation:
In any political-economic system based on appropriation and distribution of goods, rather than on actually making, moving or maintaining them, and therefore, where a substantial portion of the population is engaged in funneling resources up and down the system, that portion of the population will tend to organise itself into an elaborately ranked hierarchy of multiple tiers (at least three, and sometimes ten, twelve, or even more). As a corollary, I would add that within these hierarchies, the line between retainers and subordinates will often become blurred, since obeisance to superiors is often a key part of the job description. Most of the important players are lords and vassals at the same time.
This hierarchical structure is not only the result of financial-political elites but also a century of managerial ideologies starting with the “scientific” management of the early 1900s. Under feudalism, workers with specialised knowledge had basic rights to collectively regulate their own affairs. The alienation of production brought about by Industrial Capitalism changed this, and managerialism, later known as taylorism, pushed this further. Under financialised capitalism the map has completely become the territory to the point that efficiency is not measured by outputs, but rather, by how much effort is put into efficiency. Thus an efficiency industry, or information sector can grow, which is by its nature hugely inefficient, but which can effectively monopolise the measuring of efficiency. As Graeber says:
Efficiency has come to mean vesting more and more power to managers, supervisors, and other presumed ‘efficiency experts’ so that actual producers have almost zero autonomy.
But who are the “actual producers” in Graeber’s view? Much of the remainder of Bullshit Jobs is spent on a series of comparatively unconvincing, disjointed theses: a latent cultural memory of a protestant/puritan work ethic is the cause of our collective fetishisation of employment, the real locus of value creation is in caring work, and the only way Graeber can imagine uncoupling those in Bullshit Jobs from their dependency upon bosses is the uninspired phantasm of the Universal Basic Income. It is his claims caring work which I find most condescending as a former care worker: it was clear to me even then that washing the elderly was no locus of value creation, rather I existed to assist a multinational in milking pensions in return for labour better provided by an adequate healthcare system.
Much of the problem for anyone trying to apply lessons from Bullshit Jobs to political economy is the fact that bullshit jobs themselves are a series of several discrete categories of work that have been collated together by reason of social similarities in their subjective experience of work. Some examples only fit Graeber’s definition of bullshit jobs because the worker is morally opposed to their own labour (such as the bank worker opposed to usury). Once again, this isn’t really a criticism of Graeber as he did not set out to write a work of political economy, and in fact he divides bullshit jobs into several sub-categories based on some social similarities, or roles which make sense in the context of his feudal analogy. Nonetheless in order to draw any political economic conclusions from Bullshit Jobs we must devise our own system of sub-categories:
Once we prune away some of the socially similar but economically different aspects of bullshitisation, we can see that Bullshit Jobs is in part the synthesis of Habermas’ earlier theories on reflexive labour with its antithesis: the redundancy and economic impossibility of reflexive labour as it actually came to exist in the world. Bullshitisation, therefore, is the oversized socio-political impact of the much more specific economic factors of reflexive labour and informationalisation generally.
To see the shadow of reflexive labour within Graeber’s theory shows that Habermas’ theory of how capitalism might avert economic and political crisis was correct in a way, but not at all in the way he imagined: far from the armies of enlightened, state-employed scientists, teachers, and engineers who would force indefinite productivity, we instead got armies of privately-employed overskilled peons, paid to shift money and paperwork hither and fro. Graeber seems to understand the irony of how the hopes and fears of the contradictions of the social-democratic era led us here: in discussing Keynes he says:
Automation did in fact, lead to mass unemployment. We have simply stopped the gap by adding dummy jobs that are effectively made up.
The victory of this reflexive-unproductive labour, hollow though it may be, was to address key contradictions of the social democratic and neoliberal eras. Both have their origin in a classic over-production crisis caused by the Tendency of the Rate of Profit to Fall, and Graeber acknowledges this briefly through discussion of a short story by Stanislew Lem. I have already discussed how a decline in the Organic Composition of Capital is the key factor in initiating an overproduction crisis, but this can be interpreted in several ways.
Social-Democratic reformers have typically concluded that the key symptom of an overproduction crisis is unemployment and wage reductions, which in turn lead to reduced consumer spending and shrinking markets for goods, and so this is the cause for the correlation between an increase in OCC and the exacerbation of the TRPF. This analysis would suggest that governments stepping in to subsidise wages and stimulate consumer spending is enough to end a crisis, and indeed this is what governments of all stripes have done in the current crisis. However this does not strike at the root of the issue: the cause of an overproduction crisis is the over-productivity of labour as a result of the full sum of investments into productive forces, and short of actively underdeveloping those forces, the problem can only be relocated through asset sell-offs and growing monopolies. In the long term, the output of increasingly productive forms of labour will always outstrip demand, no matter how much consumer spending is stimulated. A much more direct means of retarding the development of crises is to both reduce overall productivity and artificially deflate the OCC, which is exactly what reflexive-unproductive labour achieves. This was able to effectively solve several economic and political anxieties of the social democratic era: it reduced the likelihood of overproduction crisis, it reduced unemployment, and it secured a large consumer base in the developed countries.
To situate the role of reflexive-unproductive labour within the context of an over-production crisis does leave one question remaining: if there is a considerable decline in the production of use values in the developed world, does this indicate a shift in the locus of value creation, or even an invalidation of Marx’s theory of value? Both Habermas and Graeber suggest that their theories invalidate materialist economics in whole or in part, for curiously similar reasons. To Habermas, the incorporation of reflexive labour from an aspect of system environment (a factor external to the mode of production) into a peculiar aspect of production disproves Marx’s theory of value because it represents an institutionalisation of indirect production. Graeber makes a very similar claim around a form of indirect labour, in this case reproductive labour, saying that the locus of value creation has shifted into caring labour precisely because it is impossible to quantify, automate, or incorporate into materialist economic theory. Habermas claims to invalidate Marx’s theory of value, but he doesn’t challenge Marx’s definitions of value, whereas Graeber proposes an entirely new definition of value, saying that it is that which we feel is valuable, or which we care about – really just a phenomenological definition.
Both authors argue against an imagined, dogmatic, inflexible Marx, and since neither works through a Marxist lens it does not feel particularly productive to be too invested in these debates around the theory of value. Suffice to say that it is a mistake to think that we are forced to look beyond Marxism, or even Marx’s own writings, to uncover terms flexible enough to explain the phenomena described. While it is true that Marx and many orthodox Marxists did not dedicate significant time to analysing reproductive and indirect labour, it is a mistake to say that in saying these forms of labour are not a locus of value creation we diminish their importance.
Reproductive labour can simultaneously be the most widespread and fundamental form of labour in the world, and still not be a locus of value creation for the capitalist. Likewise intellectual or emotional forms of indirect labour can be socially valued and can affect use value creation, or can be loaned out by capitalists for a profit, but this doesn’t mean they are a locus of value creation. In addition, Marx thought it entirely possible for more abstract forms of value, sometimes called locum tenens or placeholders, to be used in place of the production of use values.
Habemas and Graeber, taken together, have traced the development of a growing strata in developed nations which are neither the owners of capital nor engaged in any form of production or socially-necessary service. But in displacing a great deal of our understanding of where use-values come from they have not