More Bullshit Jobs?

Graeber’s claim that 20 to 50 percent of jobs are bullshit was based on YouGov polling that asked whether people thought their work “is making a meaningful contribution to the world.” Over the last seven years, YouGov has asked this question twice in the United States and once in the United Kingdom, and each time, the results were stark: between 20 and 40 percent of workers said their work was not making a meaningful contribution, while another 10 to 20 percent were uncertain.

It’s easy to see why some people would describe their jobs in those terms, and Graeber’s book is full of stories of pointless toil. Betsy spends her days interviewing nursing home residents and filling out forms that list their preferred recreational activities. The forms are then logged and “promptly forgotten about forever.” Ben has ten people who work for him, “but from what I can tell,” he says, “they can all do the work without my oversight. My only function is to hand them work, which I suppose the people that actually generate the work could do themselves.”

But while no one denies that these jobs exist, many people have been skeptical about how widespread they are. And, over the last few years, a small cottage industry of social scientists has emerged, all trying to prove that Graeber’s “nonacademic thesis” (as they insist on calling it) doesn’t stand up to scrutiny.

These critiques are based on two statistical sources. The first is the International Social Survey Programme (ISSP), which has, since 1989, asked workers around the world how much they agree or disagree with the claim “My job is useful to society.” The most recent data we have is from 2015, and the findings are firmly at the low end of Graeber’s estimate: in the United Kingdom, only 30 percent disagreed or were unsure whether their job was useful to society. In the United States, that figure was even lower, at just over 20 percent.

More troublingly for Graeber, the second data set used by his critics — the European Working Conditions Survey (EWCS) — gives even lower figures. There the number of European workers who say they sometimes or rarely “have the feeling of doing useful work” is less than 20 percent. If we focus on the original fifteen countries that made up the European Union, that number falls to just 14 percent.

It is always possible to quibble with the results of these kinds of surveys. In particular, it is worth noting that the EWCS says nothing about who the work is useful for, while the YouGov and ISSP questions make some reference to a wider social good. That is a higher bar, probably close to the spirit of Graeber’s critique, and might explain why the proportion of bullshit jobs is lower in that particular survey. But however you cut the data, it seems clear that the proportion of people who “find themselves laboring at tasks that they themselves consider pointless” is right at the bottom of Graeber’s estimate. That doesn’t mean that bullshit jobs are not a phenomenon worthy of study. But it does cast doubt on the claim that they capture something novel and essential about work under contemporary capitalism.

Yet Graeber’s theory doesn’t just rest on his estimate of the number of people who feel their own job to be pointless. One important element is his claim about the kinds of jobs people think are bullshit. Graeber gives many examples: lobbyists, political consultants, and PR specialists; doormen, receptionists, and bailiffs; sales, marketing, and advertising specialists; HR professionals and administrators; management consultants and corporate lawyers. He also singles out “information workers” — administrators, consultants, clerical and accounting staff, IT professionals, and the like — as being “precisely the zone where bullshit jobs proliferate.”

These examples fit neatly within Graeber’s taxonomy of different flavors of bullshit, but, unfortunately, the statistical evidence is much messier. There is some suggestion from the ISSP that information workers find it slightly harder to justify the utility of their work than everyone else. But this obscures a much stronger pattern: the workers most likely to doubt whether their jobs are useful to society are overwhelmingly found in “unskilled,” routine, and manual occupations.

This correlation between what Graeber calls “shit jobs” and a feeling of pointlessness is even more pronounced in the data from the EWCS. There the people who are most likely to describe their jobs as bullshit are cleaners, farmers, laborers, machine operators, trash collectors, sales workers, and clerks. The EWCS also shows that, in countries like the United Kingdom, information workers are less likely to feel their job is pointless than the rest of us. Finally, more granular statistical analysis suggests that some of the best predictors of doubting the value of your job are poor management, toxic culture, lack of autonomy, and low pay.

On one level, there should be nothing surprising about this. If your job pays you badly and treats you worse, then you’re more likely to complain about it to a stranger with a clipboard. But reading back through Graeber’s list of bullshit occupations with this evidence in mind, it starts to feel less like a well-developed theory and more like a list of jobs that Graeber doesn’t particularly care for.

Arguably, the most important element of Graeber’s theory is his claim that “the overall percentage of jobs considered bullshit by those who hold them has been increasing rapidly in recent years.” Graeber dedicates a whole chapter to explaining this historical trend, focusing in particular on neoliberalism. In his account, the move away from manufacturing and toward extractive and financialized industries has led to the growth of a whole series of bullshit corporate services (such as advertising, consultancy, and corporate law) and of pointless office jobs that consist mostly of moving piles of paper from one place to another. It has also defanged the organizations that might otherwise have redirected our economies toward more socially useful pursuits, crushing trade unions and left parties around the world.

Unfortunately, there is no statistical evidence to support this supposed historical trend. The ISSP data goes back to 1989 and suggests that, if anything, the proportion of people who think their own job is bullshit has fallen over the last twenty-five years. We get the same result from the EWCS, which shows the number of bullshit jobs falling between 2005 and 2015.

Graeber’s theory is premised on a long-run increase in the amount of bullshit in the economy, with a growing corporate sector awash with dejected workers. This has been a hugely influential characterization of contemporary capitalism — but on the basis of the evidence surveyed above, it stands in need of fundamental revision.

What’s the Point?

The lack of data supporting some of Graeber’s central arguments reflects a much deeper problem for his theory: his explicitly subjective definition of a “bullshit job.” Graeber begins with people who consider their own work to be pointless. This is a useful starting point for him because it aligns with the basic anarchist assumption that workers understand their own workplaces and the value of their own labor, while also allowing him to focus on the psychological impact of being forced to do work you think is meaningless. This might be an interesting line of research in its own right. But as a starting point for a theory of contemporary capitalism, the assumption that there is a connection between the feeling of uselessness and the real functionality of a job creates serious problems.

The first issue is epistemological; namely, the question of whether people know the true value of their work. This is the argument that the business press jumped on in their critique of Graeber. The complexity of the modern economy makes it impossible for anyone to see how their small, specialized role fits in. But from the high mountain of the Economist, the beauty of the system is clear to see. Graeber doesn’t have a direct response to those who would privilege this kind of top-down assessment of the value of a job. Instead, his argument relies on the wager (backed up by a few choice anecdotes) that people working the kinds of jobs Graeber thinks are pointless actually agree with him. He assumes, to put it bluntly, “that lobbyists and financial consultants are, in fact, largely aware of their uselessness.” But this argument becomes very hard to sustain once you acknowledge that the jobs people tend to describe as pointless do not match up with Graeber’s list of parasitic, neoliberal professions.

The second problem with Graeber’s subjective definition is more ideological. Because the pointlessness of a job lies entirely in the mind of the worker, Graeber is unable to differentiate between the two different forms of bullshit that emerge from his examples: the job of the office temp with nothing to do, and jobs the worker thinks have no social value. Take Jack. Jack was hired by a stockbroker to call senior traders offering

“free research material on a promising company that is about to go public,” emphasizing that I was calling on behalf of a broker. . . . The reasoning behind this was that the brokers themselves would seem, to the potential client, to be more capable and professional if they were so damn busy making money that they needed an assistant to make this call for them. There was literally no other purpose to this job than to make my neighbor the broker appear to be more successful than he actually was.

Jack is busy at work. He understands what the point of his job is. And if the broker’s scheme works, then it’s a rational move within the profit-seeking logic of the financial sector. The problem is that Jack doesn’t value it.

But this question of social value is one that Graeber is curiously reluctant to touch. In just over ten pages, he dismisses the very possibility of developing an absolute measure of the common good, arguing that all “utility” is basically subjective, that most “needs are just other people’s expectations,” that “values” are not the kind of things that can be quantified and compared, and that they are “murky,” the subject of “constant political argument,” and “more likely to be based on some kind of gut instinct than anything [we] can articulate precisely.” He then offers a brief summary of the different folk theories of value revealed in his conversations with disgruntled workers, which culminates in a rough-and-ready “I’ll know it when I see it.”

This position leaves Graeber fatally exposed to the old reactionary trope that if you don’t like what you do, you should change how you feel about it. Ultimately, without an objective theory of value, we cannot respond to Graeber’s critics in the business press, nor can we realize the agitational potential of his provocation: How did we end up with an economy that generates so little social good?

The New Shit Jobs

Graeber is not the first to criticize capitalism for producing socially useless jobs. In fact, this argument, in a more materialist and objective form, plays a central role in Paul Baran and Paul Sweezy’s landmark Monopoly Capital: An Essay on the American Economic and Social Order. Their analytic framework is made up of three interrelated trends. The first is the tendency for the economic surplus — which they define as the difference between what society produces and what that production costs — to rise. This occurs because, on the one hand, capitalism encourages innovation and competition between firms, which drives down prices; while on the other, firms’ monopoly power keeps prices high. The second tendency, drawing on John Maynard Keynes, is toward “underconsumption.” Because capitalists are few in number and workers are paid less than the value of what they produce, there is not enough consumer demand to absorb that growing surplus. Of course, that surplus could be directed toward savings and investment. But, Baran and Sweezy argue, because monopolistic firms tend to throttle production in order to keep prices high, they already have lots of extra capacity and therefore little need for largescale investment.

The challenge of what to do with this growing surplus then produces the third tendency: increasing economic “waste” in the form of pointless and unproductive work. Some of that is what Baran calls “leaf-raking,” pointless public sector jobs that keep people employed without risking pushback from business interests. Another category is military spending, which Baran and Sweezy see principally as a tool to defend American corporate interests abroad. And the final group encompasses the socially unnecessary costs of monopolistic competition: “advertising expenses + PR + legal departments + fins and chrome + faux frais [incidental operating expenditures] of product variation and model changes.”

I will return to the objective theory of value underpinning Baran and Sweezy’s analysis in a moment. But it is worth first explicitly addressing the shortcomings of their empirical predictions. Because if Graeber was wrong that neoliberalism is producing more and more pointless jobs, then, in their own way, so were Baran and Sweezy. Taking first those jobs that exist only to encourage us to spend more — advertising, marketing, brand management, and PR — there is no evidence to suggest that they have grown over time. Data going back to 1919 for the United States suggest that advertising spending has fluctuated between 1 and 3 percent of GDP, but with no clear trend. More recent global data compiled by the economists Alvin Silk and Ernst Berndt also show no evidence of a structural increase. Instead, since the 1960s, advertising expenditure seems to have followed the ups and downs of the business cycle. Baran and Sweezy’s second measurable category, military spending, has also fallen. When Monopoly Capital was published, the United States was spending 8 or 9 percent of GDP on the military. Today, the figure is just 4 percent, with similar declines across the Global North.

So, if neoliberalism hasn’t led to a wave of bullshit jobs, what kinds of employment has it produced? In the advanced economies that Graeber, Baran, and Sweezy all focused on, the answer is a surprising one. Contemporary capitalism has produced huge numbers of new jobs in education and health care.

The growth of education and health care as an employer of last resort is intimately bound up with the decline of manufacturing. As Gabriel Winant has argued, the social power that industrial workers once had allowed them to win substantial concessions from the state and from capital, concessions that often took the form of expanded welfare provision. While those manufacturing jobs have been automated and offshored, the welfare states they created have lingered on, often under attack but generally accounting for a growing share of GDP and of working-class employment. This growth is particularly noticeable in the care sector, where demographic shifts have been exacerbated by a feedback loop in which brutal industrial working conditions created demand for care, leading to newly exploitative conditions for care workers, and to further demand for care in turn.

In many ways, education and care are exactly the kind of professions that socialists should value. The problem is not that these jobs have been created in large numbers but rather that they have become what Graeber would call “shit jobs.” They are increasingly underfunded when compared with the demands placed on them. They are poorly paid, insecure, and casualized. Professional autonomy has been decimated. And the prestige attached to these vital roles has been systematically eroded.

This decline in working conditions is part of a general remolding of labor under contemporary capitalism. First, work itself continues to be the central organizing feature of most of our waking lives, with labor force participation in the OECD flatlining at around 60 percent over the past forty years. (This continuity obscures two opposing trajectories: in Europe, it has risen steadily, while in the United States, it has collapsed since the late 1990s.) Second, despite the ongoing centrality of labor, annual working hours have fallen steeply. This is fine if you can afford to live, but it’s disastrous if you’re near the breadline. Third, there has been a marked trend toward “job polarization,” by which economists mean that there are fewer medium-skill jobs on offer and many more high-skill jobs. Fourth, the middle class has been squeezed both numerically (the share of people in middle-income households across the OECD fell from 64 to 61 percent between the mid-1980s and the mid-2010s) and in terms of its economic power (the total income of all middle-income households was four times the income of upper-income households in the 1980s, while today, the ratio is less than three; this is largely because incomes for people in the middle of the distribution have stagnated in absolute and relative terms over the last forty years).

This collapse in the quality of jobs is something that even mainstream economists have at last begun to discuss. Most recently, Harvard professors Dani Rodrik and Stefanie Stantcheva described the creation of good jobs as an “existential” challenge for contemporary capitalism, placing it on par with climate change. But nowhere in this resurgence of interest in pay and working conditions do they seek to clarify Graeber’s central question: Are our jobs contributing to the social good, or are they just bullshit?

Frustration With Bureaucracy

If Graeber’s empirical predictions have turned out to be inaccurate and his theory vulnerable to conservative counterattacks, much of the problem lies with his explicitly subjective starting point. But perversely, that subjective starting point is also the most natural one for a central goal of socialist theory: informing the practice of agitation.

Although our feelings about our work must be at the heart of any attempt to organize at the point of production, this opportunity to link theory and practice is lost in Graeber’s focus on jobs that are experienced as “entirely or overwhelmingly bullshit.” In those situations, the obvious response from the worker is to quit the job entirely. And, indeed, this is what many of the people Graeber spoke to had done. But almost all of us have had the feeling that particular tasks within our job are bullshit. Often, it is this tension between the value of the job and the bullshit tasks we are asked to perform that creates the space for agitation and resistance.

As the industrial relations scholars Jacques Bélanger and Christian Thuderoz have argued, many workers who are “very proud of their craftsmanship … [feel] a deep frustration … [about the] inability of management to organise operations efficiently.” This is probably especially true in education and health care, which the labor organizer Jane McAlevey calls “mission-driven” sectors in reference to workers’ clear and preexisting moral authority and sense of purpose. The feeling that these workers could organize themselves better than their bosses do is a vital socialist impulse and a potent starting point for agitation. But it’s also something we lose sight of by focusing on bullshit jobs rather than bullshit tasks.

Shifting the level of analysis from jobs to tasks allows us to link theory more closely to socialist practice. But the next step is to provide an objective foundation for our claims that some particular task is bullshit. Graeber himself flirted with the idea of quantifying the level of bullshit tasks in an earlier book on bureaucracy, The Utopia of Rules. In the opening pages, he provides an imaginary graph tracking the ever-increasing number of hours we spend on “paperwork.” This is obviously just speculation, and, unfortunately, the only people employing time-use surveys and task monitoring to calculate how much time we spend doing what at work are our bosses. With no reliable data to quantify the levels, trends, and distribution of bullshit tasks, I will instead sketch a conceptual framework for understanding where they come from. But the suggestion that tasks of this kind are proliferating must remain a speculative hypothesis in need of firmer evidence.

The first distinction to make is between the primary task around which our job is organized (say, providing care to elders) and the secondary tasks that are arranged around it. Some of those secondary tasks are immediately necessary (for example, ordering drugs and supplies in the correct quantities), while others may indirectly improve the efficiency of that primary operation (such as taking part in trainings or keeping medical records). But often, these secondary activities slide into pointlessness.

The classical sociology of bureaucracy, drawn from the writings of Max Weber and his student Robert Michels, provides a basic framework for understanding how those secondary tasks can shift from functional to dysfunctional. For Weber, bureaucracy — a system of rule-based decision-making and information processing — emerges inevitably out of the coordination challenges faced by large and complex institutions. These rational management and organizational systems allow institutions to function effectively, giving them a competitive advantage, which then encourages their rivals to follow them down the road of rationalization and bureaucratization. But these systems also produce extra tasks for workers. Most important, workers are required to keep the records and provide the data the bureaucrats use to make decisions.

For as long as those systems enable large, complex institutions to function in a coordinated and rational way, they might have some legitimacy. But there is always a risk that bureaucracies become dysfunctional, obstructing the primary tasks they were meant to facilitate. The central danger identified by Weber, Michels, and others is that bureaucracy becomes an end in itself. The collection of data, the codification of rules, and the maintenance of the bureaucratic superstructure become the primary tasks of the institution.

Weber explains this tendency toward expansion as inherent to the logic of rationalization, which always seeks to capture within its “iron cage” any irregular, uncodified, and unregulated exceptions. For Michels, this tendency has a more human form: bureaucrats have access to information, which gives them power and privileges within the institution, and they then develop a material interest in preserving that institution and the bureaucracy that sustains their power over it. (A similar argument is made by Ernest Mandel and Rosa Luxemburg in their critiques of the reformist bureaucrats who all too often dominate trade unions and social democratic parties.) Crucially for our purposes, both mechanisms turn functional procedures into hopelessly dysfunctional bureaucracies and, in so doing, create a mountain of bullshit tasks for workers to complete.

It is important here to note that Weber’s account of bureaucracy has a certain cyclical quality. As rational bureaucratic processes turn pathological, they are left vulnerable to “charismatic” revolution. This alternative source of authority is fundamentally indifferent to the rules that govern bureaucratic thinking. Instead, under the influence of charisma, people pursue their interests by breaking rules, smashing and disrupting the bureaucratic systems in which they were trapped. Whether we reach that stage anytime soon remains to be seen.

Giving Up Control

The second category of bullshit secondary tasks is related but more pernicious: those activities that only exist to give bosses and managers control over our work. To explain the origin of these tasks, we need to go back to Karl Marx’s famous manuscript Results of the Immediate Process of Production, in which he draws a distinction between the “putting-out” system (where workers were paid a wage to, for example, weave cloth in their own homes) and the factory system (where that work was brought indoors and under the bosses’ direct supervision). This transition from “formal” to “real” subsumption is made necessary because of the strange nature of labor as a commodity. When bosses hire workers, what they want is their labor power, that productive energy that adds value to whatever assets they are working with. But what they get is a human being with its own volition and idiosyncrasies.

This “indeterminacy” of labor has two principal sources. On the one hand, workers have an irreducible power to choose how much effort to expend in their work. On the other, they can also choose whether or not to stay at that firm for the long haul. (This second source of indeterminacy is obviously historically contingent. Marx assumed that “free labor” differentiated capitalism from other systems of production. But in fact, capitalism has always combined workers in various states of freedom and unfreedom — from Atlantic slavery and England’s “master-servant” laws to migrant worker statuses and the bewildering variety of employment law around the world today.)

Capitalists have always struggled against these two sources of indeterminacy. But the tools they use have varied over time. Marx described the transition from the direct violence of feudalism to the “silent compulsion of economic relations” under nineteenth-century capitalism. Harry Braverman then chronicled the twentieth-century shift to a reliance on “Taylorism” and the various techniques of scientific management. But none of these modes of control generated additional secondary tasks for workers.

This has changed over the last forty years due to two shifts in the way bosses seek to control the process of production. The first is a refinement of Taylorism. Scientific management was traditionally achieved by outside experts, the “time-and-motion men” who came in to study the workflow and impose more efficient procedures. Control of contemporary white-collar work, on the other hand, requires us to observe ourselves, to become our own time-and-motion men. Often, this self-surveillance takes the form of an expanding bureaucratic superstructure, complete with endless forms to be completed and records to be maintained. But it has a single and more limited purpose: control.

Today, some of these tasks are at risk of being made obsolete by technological advances. The automation of supervision in what is known as “algorithmic management” has so far been concentrated in low-paid, easily standardized jobs — Amazon logistics workers, Uber drivers, and Deliveroo couriers. But wider automation of these processes for collecting data and controlling labor could reduce the burden of secondary tasks that workers have to complete, if at a cost to autonomy, privacy, and organizational capacity. However, as with much of the discourse on automation, it is worth remembering that these trends are very much in their infancy and that too often our analysis is colored by the self-aggrandizing claims of their loudest advocates. A more sober analysis of whether this kind of bullshit secondary task is growing or shrinking is simply not possible until we have far more rigorous data on what workers actually spend their time doing.

The second key shift is the move toward normative and cultural modes of control. The intuition behind these new techniques is that “managers could more effectively regulate workers by attending not only to their behavior but to their thoughts and emotions. By winning the hearts and minds of the workforce, managers could achieve the most subtle of all forms of control: moral authority.” For workers, this means taking part in a new range of secondary tasks, all designed to inculcate certain values, identities, and cultures. These tasks are often absurd — laughably bad compulsory training videos, online tests of a corporation’s “key values,” singing the company anthem — but they seem to fill our working lives like never before.

As more orthodox Marxists pointed out in the critical engagements with labor process theory, this desire for control does not arise because our bosses are vindictive or out of some purely political will to power. Instead, it is rooted in economic necessity. Capitalists are engaged in a competitive struggle for survival and so are compelled to try to extract the maximum effort and commitment from their workers. But unlike a bureaucracy, which might be genuinely necessary for the coordination and organization of a large, complex institution, the secondary tasks produced by our bosses’ desire for control serve no broader purpose. The control imperative only arises because we have to work at jobs over which we have no control or autonomy, because we are, in Marx’s terms, alienated by capitalist relations of production. With greater freedom to choose and direct our own productive labor, the indeterminacy problem would cease to exist, and so would the need for control.

One important point of nuance related to this second category of bullshit tasks is the way they play out in the public sector. Here there is no economic compulsion to extract as much labor as possible from workers. But, perhaps surprisingly, similar modes of control have arisen, most famously in Margaret Thatcher’s new public management theory and Tony Blair’s obsession with “targets.” This is a deliberate mirroring of the private sector and reflects the way a specifically capitalist mentality has spread into other areas of social life. Thatcher and Blair shared the aim of making the public sector feel more like a private enterprise, and they were explicit about that ambition. In this way, the cultural forms produced by capitalism took on a life of their own, generating bullshit tasks in workplaces far removed from the biting edge of free-market competition.

Useful vs. Useless Tasks

With the question of which primary tasks should be considered pointless, we risk stumbling over the vexed and intractable problems of moral philosophy. But, without pretending to resolve those debates, I think we can clarify some of the concepts needed to understand this third category of bullshit tasks and point toward a practical resolution of that theoretical impasse.

The starting point for thinking about which jobs make a meaningful contribution to the world is the distinction between use value and exchange value. Every object can be seen from these two perspectives: on the one hand, its manifold, qualitatively distinct uses; and, on the other, the homogeneous, quantitative price at which we sell it. This is a fundamental contradiction of capital, and it helps to explain why a capitalist economic system always has the potential to produce jobs that do not contribute to the common good. Because capital is guided at all times by exchange value, it is devoted solely to the search for profits and the ceaseless drive to increase its value. To put it another way, our economy is structured by an insatiable quest for exchange value that is not limited, directed, or controlled by what people actually need.

Mainstream economics has its own language for describing this disconnect between where investment is directed (and hence what kinds of jobs are created) and the common good. This is the language of positive and negative externalities, those spillover effects of economic activity that may be beneficial to wider society (everyone benefits from having healthier friends, family, and neighbors) or harmful (pollution from a factory may affect people living hundreds of miles away). Because these effects spread far beyond the people directly involved in making the relevant economic decisions, they tend to be ignored. The result is a misallocation of resources. We end up with too many polluting factories and not enough hospitals.

The problem with this language of positive and negative externalities is that it frames the problem as one of misallocation rather than contradiction. It therefore preserves modern economics’ basic assumption that there is a single, quantifiable measure of human utility that is translated into prices by supply and demand. Externalities appear as nothing more than creases in this otherwise smooth process — creases that can, with appropriate institutional frameworks, be ironed out. Marx’s qualitative distinction between the uses of a good or service and the price it fetches in the market allows us to do away with this assumption, revealing how deeply embedded in modern capitalism it really is.

The contradiction between use value and exchange value explains the potential for capitalism to produce jobs that don’t contribute to the common good. But it doesn’t demonstrate that they do in fact exist. Faced with this challenge, there is a tendency to fall back on jobs that are universally agreed to have social value. There is, for example, a long tradition of research in development economics demonstrating that state socialist regimes tend to invest more in health care than capitalist ones. A common example is Cuba — an underdeveloped, isolated Caribbean island of eleven million inhabitants — that has greater life expectancy, lower infant mortality, and cheaper health care than the richest country in the world, the United States.

Health care is obviously not the only category of jobs that contributes to the social good. But we are often drawn to it because it responds directly to an indisputable human need for survival. At a purely formal level, this then provides an answer to the question of which primary tasks contribute to the social good, namely those that fulfill a true human need.

But this formal answer has little content. In particular, it doesn’t explain how we are to identify those true human needs. Survival is clearly inadequate as an organizing concept. Although needs of this kind do have an important place in Marxist theory — providing an explanation “in the last instance” of why workers take on exploitative wage labor and why they also resist those relationships when capitalism endangers their basic survival — but any credible account of bullshit tasks has to push beyond it. Moreover, survival does not do justice to the nature or range of human desire. As psychoanalysts have long argued, we often choose to take risks, play dangerous sports, get high, or eat too much, all in spite of our physiological drive for self-preservation. Indeed, physiology is only the first rung of Abraham Maslow’s famous eight-tiered “hierarchy of needs,” which includes equally human desires for safety, love, esteem, and self-actualization.

The other problem in filling out the content of true human needs is that our needs are not fixed. Contemporary capitalism is engaged in a continual war of propaganda, convincing us to buy more and more commodities, constantly reshaping our desires and our sense of self. For many nineteenth-century writers (including Marx), it was possible to have an ambivalent attitude toward this, seeing in our expanding needs a certain refinement, an “education of the senses” that ultimately elevates humans above a mere animal existence. But in today’s hyperconsumerist societies, that argument feels less plausible. In the first place, contemporary capitalism seems to produce an absurd excess of needs.

Taking two authors I’ve already cited as examples, Baran draws a distinction between “sensible” commodities and those that only exist “due to the sales effort,” while Cohen emphasizes the Sisyphean insatiability of commodity production, which always and only wants to produce more. Second, advertising creates the illusion that our needs are vested in commodities themselves and so obscures their deeply social nature. We drink beer to look manly and impress our friends, thinking that the commodity will magically do the work of building relationships of love and respect. Third, by entangling our needs within capitalist relations of production, we become estranged from them. Because capitalism releases us from immediate necessity, it should be possible to realize our deep desire for collaboration and altruism. But because it simultaneously mediates those needs through private property, we remain trapped in relationships of instrumentalization and alienation.

The question of how we can judge whether a job is involved in valuable self-actualization or frivolous consumerism can be more usefully framed in terms of practice: How do we come to know where the limits of true human needs lie? Through what processes can an answer be determined? The classical neoliberal response comes out of Friedrich Hayek’s and Ludwig von Mises’s contributions to the economic calculation debates. For them, the market was an unsurpassable device for processing information, synthesizing all the knowledge and subjective preferences of its participants, and shaping our behavior in response to that wealth of data. Indeed, the “superior, if opaque, wisdom of the market” was something to be defended against meddling technocrats and prying economists. So when it comes to bullshit jobs, their response is simple: if someone is paying you to do it, then it must be valuable and answer a true human need. No mere mortal should question the outcomes of the market machine.

This is more a declaration of faith than a persuasive argument. But Hayek and Mises can point socialists toward an answer of our own. A socialist baseline is not the unfettered market but a society in which decisions about where to allocate resources and what jobs need to be done are made through open democratic debate. This rationally and democratically designed economy was the implicit “other” against which Baran and Sweezy compared the “waste” of monopoly capitalism. But it has sometimes been sidelined on the contemporary left in favor of neo-Keynesian interventions like the Green New Deal or anti-austerity politics.

Holding out a democratic economy as the yardstick against which we judge whether or not a job is pointless requires two final theoretical clarifications. First, just as advertising molds our desires and identities, so too would deliberative democratic debate change the character of our needs. In that sense, the processes through which we determine value themselves shape our value judgments. But unlike the distortions and contradictions described above, socialists assume that democracy will help to undo the spiritual and psychic damage caused by consumer capitalism, allowing us to flourish individually and collectively.

The second clarification concerns the here and now. We can obviously only approach this kind of democratic planning in small steps, and today, there are very few places where such institutions exist. The challenge is therefore to build them. But in the meantime, we must also conduct proto-democratic debates in what’s left of the public sphere. We must struggle through civil society in search of “defensible positions” from which we can vocally challenge the current allocation of resources and put forward a different vision, grounded in a different set of needs. This is the task that Graeber tried to dodge with his purely subjective definition of bullshit jobs. But it can no longer be ignored.

Where the Real Bullshit Lies

David Graeber’s Bullshit Jobs asked a vital question about contemporary capitalism. But its theoretical starting point led it down a blind alley. The central problem with work today is not that a growing proportion of people feel their own job to be pointless. It is that the economy as a whole is not oriented around satisfying true human needs. Instead, we find our working lives consumed by dysfunctional bureaucracies that constrain our capacity for play and creativity. We waste time trapped in processes that exist only to give our bosses more control over us. And we’re directed to jobs that wouldn’t exist if people had the power to choose where to invest our increasingly scarce resources. That is where the real bullshit lies.