Book Summary: Bullshit Jobs – A Theory by David Graeber

Book Cover for Bullshit Jobs by David Graeber

This is a summary for Bullshit Jobs: A Theory, a book following the viral success of David Graeber’s article, On the Phenomenon of Bullshit Jobs: A Work Rant. In the book, Graeber attempts to flesh out some of the ideas in his 2013 article.

Buy Bullshit Jobs at: Amazon | Kobo (affiliate links)

You may also want to check out these blog posts with my thoughts on the ideas raised in Bullshit Jobs:

Key Takeaways from Bullshit Jobs

  • A bullshit job is one that the employee considers is pointless, could be easily automated and/or has no (or negative) social value. An element of fraud or pretence is also required.
  • Many people are in bullshit jobs. One survey found as many as 37% of the population feel that their jobs are not meaningful.
  • Business owners often refuse to accept the idea that bullshit jobs could exist, at least in the private sector.
    • Their main argument was that no one would spend money on an employee who isn’t needed. So employees who believe they are in bullshit jobs just don’t understand their real function.
    • Alternative, bullshit jobs in the private sector are only created by government interference (e.g. box-ticking).
  • Bullshit jobs are demoralising and depressing for the employee involved. Despite conventional economic wisdom, people don’t want to do nothing. People want to have some effect on the world.
  • Graeber suggests multiple reasons for why bullshit jobs exist:
    • The rise of finance has contributed to the rise of bullshit jobs. When there is a large pot of money, people will find ways to get a part of that pot. They do so by moving the money around in inefficient (bullshit) ways. Bullshit jobs seem to defy the logic of capitalism because the existing system isn’t capitalism. It is a system of rent extraction and feudalism.
    • Graeber looks back at the medieval feudal system, the fall of that, the industrial revolution and workers’ movements. The upshot of all those changes is that work became valued as an end in itself. There is a deeply held idea that employment makes you a full, moral person.
    • This is why there is an inverse relationship between social value and compensation. Our society has come to believe that, as a matter of morality, you should not be paid for things you enjoy. You should only be paid for work that is difficult.
    • Politically, the Left and the Right both agree that more jobs is better. Whether those jobs are useful is irrelevant.
    • Having people in jobs keeps them occupied. They won’t have time or energy to plot uprisings. As George Orwell has suggested, keeping people busy helps to keep order.
  • Automation actually has led to mass unemployment. We’ve just plugged the gap by adding bullshit jobs.
  • A universal basic income is a possible solution to bullshit jobs. It detaches livelihood from work. A UBI allows people to leave their jobs without economic consequences. This will allow people to do other things – things less likely to be pointless or bullshit.

Detailed Summary of Bullshit Jobs

Definition of a bullshit job

  • Graeber offers up the following definition in his book.
A bullshit job is a form of paid employment that is so completely pointless, unnecessary, or pernicious that even the employee cannot justify its existence even though, as part of the conditions of employment, the employee feels obliged to pretend that this is not the case.
—David Graeber in Bullshit Jobs
  • The three key points seem to be:
    • Pointless, unnecessary or pernicious.
    • Fraud or pretence that the job does some good
    • Paid for it (both the job and the pretence)
  • A YouGov poll in the UK found that only 50% of people thought that their made a meaningful contribution to the world. 37% of people thought that it did not. [I looked up the actual poll. The wording of the question is: “Is your job making a meaningful contribution to the world?” Seems to me that many more people would answer “no” to such a question compared to “is your job completely pointless, etc”. Graeber refers to this 37% figure at several points in the book.]
  • In that poll, men were far more likely to feel their jobs were pointless (42%) than women (32%). [This is not just because more men answered the YouGov poll than women: the full poll results suggest responses were about evenly divided along gender lines.]
  • Bullshit jobs are not the same as shit jobs. Bullshit jobs are pointless but often pay quite well and have good working conditions. Shit jobs are bad, in that the work is pointless, the pay is crap and workers are treated poorly – but they are not pointless at all. Graeber notes it is possible to have a job that is both bullshit and shit.

Fraud or pretence

  • A mafia hitman is not a good example of a bullshit job. There has to be some degree of pretence and fraud for why the job exists for it to be a bullshit job. A mafia hit man presumably is quite upfront about what they do and is unlikely to pretend that their job benefits society. [Not sure I buy this. I can easily imagine a mafia hit man justifying why them doing their job is good. For example, “I pick and choose the jobs I do so I’ll only kill bad people” or “If I didn’t do it, someone else will, and I’m reducing suffering by killing people in a clean and less brutal way”. In fact, Graeber seems to contradict himself by saying “Most mafiosi believe they are part of an ancient and honorable tradition that is a value in its own right, whether or not it contributes to the larger social good.” He suggests that unless the hit man is personally convinced that his job should not exist, it’s not a bullshit job.]

The subjective element

  • The definition depends on whether the worker considers the job to be pointless, unnecessary or pernicious, but it also suggests that the worker is correct.
  • Graeber defends the subjective element of this definition as follows:
    • on whether a job does anything at all, we should defer to the individual workers who is likely best placed to know this.
    • on whether a job does anything of value, we should defer to the overall opinion of those who work in the industry.
  • In this way, Graeber avoids the situation where some people will have a bullshit job while others in the exact same job won’t, because they don’t think they do.
  • Sometimes people working for a large corporation won’t fully understand their contributions in the context of the “big picture”. Graeber accepts this may be the case (particularly if the corporation is doing anything illegal). But he quickly dismisses it by saying that it’s been “his experience” that any underling who works for the same outfit for any length of time will normally be taken aside and let in on the company secrets. [I don’t find this answer very satisfactory. A better response might be that, it is demoralising to be in a job that you believe is pointless. If, contrary to that belief, the job actually is (socially) valuable in some way, you’d expect the higher-ups to tell that to you, to boost morale. I note, however, that this only applies to jobs that are socially valuable, rather than jobs that have a point but that point is to seek rents.]
  • Graeber points out that higher-ups in the organisation won’t necessarily have an accurate picture of whether a job is pointless either. People have an incentive to hide things from higher-ups. For example, many underlings wonder, “Does my supervisor actually know that I spent eighty percent of my times designing cat memes?”
  • To determine if certain kinds of work (e.g. telemarketing, market research) are bullshit, Graeber suggests deferring to the judgment of the majority of people doing that work. He says that if corporate lobbyists and financial consultants generally really did believe in the social value of their work, they’d be more like hit men than bullshit jobs.
  • In April 2013, at a conference on “Fixing the Banking System for Good”, Jeffrey Sachs gave an unusually candid assessment of Wall Street. Many people in the financial system were quite upfront with him because they assumed he was on their “side”. Sachs described the moral environment of Wall Street as “pathological”. They genuinely believe they have no responsibility to anyone, and that it is their right to take as much money as they possible can in any way possible (legal or otherwise).
  • Graeber acknowledges that people can be wrong about what they do. But, for the purposes of this book, he doesn’t think it matters so much as he is mainly interested in the subjective element. His primary aim is not to lay out a theory of social utility or value. Rather, it is to understand the psychological, social and political effects that result from so many people believing that their jobs have no social value.

Public and private sectors

  • Pointless work is equally likely in the public and private sectors. The main difference is that pointless work in the private sector is likely to be more closely supervised.
  • Some people think bullshit jobs only exist in the public sector. Part of this is because they remember countries like the Soviet Union, which made up jobs for people to pursue “full employment”.
  • While pressures to downsize and increase efficiency do exist in the private sector, they are directed almost exclusively at the people on the bottom of the pyramid who actually make things. Middle management, white-collar roles rarely face those same pressures.

Partly bullshit jobs

  • Most jobs involve at least some pointless elements (e.g. teachers having to do administrative paperwork).
  • The bullshit quotient of jobs has increased over time. The process of “bullshitization” (as Graeber calls it) is highly inconsistent. It has affected the middle class more than the working class; and within the working class the main target has been traditionally female, caregiving work.
  • According to the US edition of the 2016-2017 State of Enterprise Work Report, the amount of time US office workers say they spend doing their actual duties has declined from 46% in 2015 to 39% in 2016. That time has been redirected to dealing with emails, “wasteful” meetings, and administrative tasks. Graeber acknowledges that figures so dramatic must be partly the result of statistical noise. But nonetheless he concludes that more than half of US working hours are spent on bullshit and that the problem is getting worse.
  • Graeber says that his book is about jobs that entirely or overwhelmingly bullshit – not just “mostly”.

Sex work

  • A former exotic dancer made the case that most sex work should be considered a bullshit. Sex work clearly does answer a genuine consumer demand. But something was terribly wrong with a society that effectively tells its female population that they are worth more dancing on boxes between the ages of 18 and 25 than they will be at any subsequent point in their lives.
  • Graeber says that it’s hard to deny the power of her argument. His only objection is that the argument may not go far enough. It’s not so much that stripper is a bullshit job. It’s that this situation shows that we are living in a bullshit society.

Five major types of bullshit jobs

  • Graeber’s research is drawn from online discussions in response to his 2013 essay, and from people responding to a tweet soliciting testimonies. The research is qualitative in nature.
  • The book suggests 5 main categories of bullshit jobs. These are not exhaustive. It is certainly possible that a bullshit job may fall into more than one category.


  • Flunky jobs exist to make someone else look or feel important.
  • Flunkies may be given a minor task to justify their existence, but this is a pretext. In some cases, however, flunkies may end up doing the job of other people in the organisation. For example, secretaries working for male executives.
  • Examples include doormen and receptionists at places that don’t need them.
  • A company has to have at least 3 levels of command to be considered a “real” company. A receptionist can act as one of the layers, as a collective underling to others.
  • When an organisation grows, higher-ups’ importance is measured by the number of employees they have working under them. So there is an incentive for leaders to hire people first and only then decide what to do with them.


  • Goon jobs exist only because other people have goons. These jobs have an aggressive element and, like actual goons, largely a negative impact on society.
  • Examples include armed forces, lobbyists, PR specialists, telemarketers, corporate lawyers, advertising.
  • Graeber argues that these are bullshit jobs, rather than mafia hit men, because of the subjective element. Because so many people in those jobs feel that the jobs have no social value and should not exist. [Seems hard to understand why this is difficult from hit men. I would wager that mafia hit men are just rarer in general and also less likely to post about their jobs online.]
  • Goons find their jobs objectionable not just because they don’t have positive value, but also because they see the jobs as manipulative and aggressive. These jobs may involve deception by tricking people into doing something that wasn’t in their best interest. [But it may well be in the best interests of their employer – in which case these jobs aren’t “pointless”.]

Duct tapers

  • Duct tapers are there to solve a problem that should not exist. Graeber adopts this term from the software industry, but generalises it.
  • For example, cleaning is a necessary function as ordinary life will mean that some things need to get tidied up. But sometimes people will make an unnecessary and gratuitous mess. Duct tapers are mostly underlings whose jobs are to undo the damage done by sloppy or incompetent superiors. [So a copyeditor will presumably be a duct taper under this definition. But while a writer can proof their own work, it may be more efficient to get someone else to do it. It can be cheaper and more effective to hire a fresh pair of eyes. The work – the proofreading and editing – needs to be done either way. Why is it a bullshit job if it’s just one person doing it?]
  • Many duct-taper jobs are tasks that could easily be automated but haven’t. A reason why the job hasn’t been automated may be because no one has gotten around to it or some structural confusion. Or the manager may want to maintain as many subordinates as possible. Or some combination of the above. [Doesn’t suggest the job is pointless as it still fixes a problem that requires fixing. Hiring someone to duct tape it may be cheaper or otherwise preferable to fixing the underlying problem.]

Box tickers

  • Box tickers are employees who exist to allow an organisation to claim it is doing something that it is not in fact doing.
  • For example, if a government’s employees are caught doing something bad, its first reaction is to create a fact-finding commission. This has two functions. First, it’s a way of insisting that, apart from the small group of employees involved, no one had any idea it was happening. Second, it implies that once all the facts are in, something will get done (which is not usually true, either). [Very cynical. Only somewhat true. I do agree it’s an easy way for the government to be seen to be doing something. But sometimes they do end up doing things.] Large corporations will do the same if they are founded to be using child labour or something.
  • Once you introduce formal measures of success, “reality” becomes that which exists on paper.
  • But some indicators may be a poor measure of success. For example, whether someone meets monthly “target figures” will sometimes be due to random statistical noise, rather than their actions.
  • Many box-tickers are involved in making reports and presentations look physically attractive. An employee’s sole purpose may be to prepare PowerPoint presentations or create visually attractive graphs or charts. The reports produced are often just props, too – no one reads them all the way through.
  • Some large corporations maintain their own in-house magazines or even TV channels. Ostensibly, the reason for these is to keep employees up-to-date on interesting news. The real reason is to give executives warm fuzzies when they see a favourable story about them in that media. [Seems more like a flunky job than a box-ticker.]


  • There are two subcategories of taskmasters:
    • Those who just assign work to others. This job may be bullshit if the underlings are perfectly capable of doing the tasks without the taskmaster. In this case, the taskmaster is useless but not harmful.
    • Those who create bullshit tasks for others to do and supervise it. This type of taskmaster does actual harm. For example, a taskmaster may set up quantifiable methods for assessing performance. This forces people to spend more time assessing and justifying what they do and less time actually doing it.

Bullshit jobs cause unhappiness

  • Bullshit jobs regularly induce feelings of hopelessness, depression, and self-loathing. Graeber claims they are forms of “spiritual violence directed at the essence of what it means to be a human being”.
  • It isn’t obvious that a bullshit job should cause unhappiness. Employees are being paid – often well – to essentially do nothing.
  • Graeber gives various examples of people who had bullshit jobs and were unhappy. He also provides two examples of people that are happy in bullshit jobs:
    • The first example is a substitute teacher in the US. Graeber doubts whether this is even a bullshit job. It is unsupervised, nonmonotonous, involves social interaction, and allows the person to spend a lot of time doing whatever he likes. Moreover, the person does not expect to be doing it for the rest of their life. So it’s about as good as a bullshit job will get.
    • The second example is a tax official in France. The woman gets along with her coworkers and has a lot of autonomy. She also gets the respect and security of government employment and is aware that the whole thing is ultimately quite silly.
  • Graeber also suggests that everyone knows that substitute teachers in the US and tax officials in France are mostly bullshit, so there’s little room for disillusionment or confusion. People applying for those jobs know what they’re getting into
  • In many large cities, most middle-class people now spend so much time at work that they have few social ties outside it. If that’s true, then the fact that most people in bullshit jobs are so miserable is even more striking.
  • Bullshit jobs can even lead to both physical and mental health problems. Graeber has read many anecdotal reports of depression and anxiety overlapping with physical symptoms like carpal tunnel syndrome and autoimmune breakdown.
  • The only accounts Graeber received from workers who didn’t feel anguish at their bullshit jobs were those who had found a way to keep them down to one or two days a week. He quickly dismisses this as being “logistically extremely difficult, and usually impossible, for either financial or career reasons”. [It seems like he hasn’t heard of the FIRE movement then.]

The classical economics assumption that humans hate work is wrong

  • In classical economic theory, the model of homo economicus (economic man) is motivated by a comparison of cost vs benefits. If that were correct, people should be delighted to have a bullshit job. They receive money for virtually zero expenditure. [Time cost is the largest cost of a bullshit job, same as with any job. Most of the examples Graeber gives still require the employee to be physically present, and often they can’t do other things they’d want to while at the bullshit job. On top of that you also have costs of boredom, and having to tell people your job is something you think is bullshit.]
  • Much of our public discourse about work is based on this underlying economic model. If people are offered the option not to work, they will take it. That’s why, if we give poor people relief, we have to deliver it in humiliating and onerous ways. Otherwise they will have no incentive to find a proper job.
  • But almost every bit of available evidence indicates that this is not in fact the case. For example, working-class people who win the lottery rarely quit their jobs. If they do, they often regret it. People in prisons who are provided free food and shelter and are not required to work will still choose to do so – even if it’s unpaid.

Bullshit jobs make us unhappy because they show us how we don’t affect the world

  • Bullshit jobs cause unhappiness because they are a direct attack on the very foundations of our sense of self. A human being unable to have a meaningful impact on the world ceases to exist.
    • In 1901, German psychologist Karl Groos found that infants expressed extraordinary happiness when they find out they can cause predictable effects in the world. Groos described this as “the pleasure at being the cause”. He suggested it was the basis of play – exercising powers simply for the sake of exercising them.
    • But if you then deny a child the ability to cause an effect, they get very upset. First they will rage, then refuse to engage, then they’ll kind of fold in on themselves in a kind of catatonic way and withdraw from the world entirely. Francis Broucek called this the “trauma of failed influence”.
  • Bullshit work can create a disincentive for workers to be efficient. If they get all their work done early, they will not be rewarded but instead punished with pointless busywork.
  • Most people in the world today, particularly in wealthy countries, are taught to see their work as the main way they can have an impact on the world. One woman testified that she gets most of the meaning in her life from her job. So the fact that her job was bullshit was very demoralising and depressing.

The falseness of bullshit jobs adds to the unhappiness

  • It’s not just the purposelessness of a bullshit job that causes unhappiness, it’s the falseness. A person who scams someone might enjoy it – we see grifters and con-artists as romantic figures, living by their own wits. But being forced to scam someone is different. You’re in the same position as the person you’re scamming because you’re similarly being pressured and manipulated (by your employer), with the added indignity of having to betray someone’s trust.
  • Being forced to pretend to work just for the sake of working is an indignity, since the demand is perceived (rightly, in Graeber’s view) as the pure exercise of power for its own sake. [I think one reason why managers or supervisors force others to pretend to work is to cover their own asses – so that they cannot be blamed for their underlings not working. So it’s not really exercise of power for its own sake.]

Ambiguity adds to the misery

  • A bullshit job may not have a demonstrable bully. It’s not always clear who is making you pretend to work – your manager? The company? Society? This ambiguity can be infuriating.
  • The emotional labour required of those in bullshit jobs is usually a lot less than that required of service workers, but is complicated by ambiguity. People in bullshit jobs are never sure exactly what is required of them or how they’re meant to feel about their jobs. The overwhelming majority of people Graeber talked to didn’t know if their supervisor knew that they weren’t doing anything.
  • Graeber never heard of a single case of a supervisor just sitting down with an employee and spelling out the rules regarding when they had to work and when they didn’t.
  • Sometimes managers will communicate this indirectly, by their own behaviour. Particularly in large organisations [and probably government] where managers don’t have much of a proprietary feeling and don’t have reason to believe they’ll get in trouble with their own superiors. But there is a taboo on being too explicit about this. [Probably because of the Clueless, and because of any other employees whose jobs are not bullshit.]
  • Sometimes, in longer-term positions, there is enough camaraderie among employees that they can discuss the situation openly and share strategies. For example, a legal aide in a law firm said that he always made sure to have at least two projects run by different bosses going, so that he could tell both bosses that he was spending a lot of time on the other’s project. [I’ve been in this situation before. There is always the risk that the two bosses talk to each other, so your claims still have to be somewhat plausible.] But it can be hard to know exactly who you can and can’t confide in.

Our normal human work patterns are irregular

  • Historically, most people would have assumed that normal human work patterns take the form of periodic bursts of energy, followed by relaxation, followed by slowly picking up again toward another intense bout. In other words, there are ebbs and flows.
  • Examples include farming, being a feudal lord, hunting.
  • Peasants and servants may work more steadily – but even then, nothing as regular or disciplined as the current 9-5 work week. Medieval serfs probably worked from dawn to dusk for 20-30 days a year, and just a few hours a day otherwise.
  • The main reason work could be so irregular was because it was unsupervised. If people produced what was required of them, their superiors wouldn’t really care how much they worked to produce it. [Nowadays output is less measurable.]
  • The demand to work continually at a steady pace for 8 hours a day regardless of what there is to do defies common sense.
  • Much pointless work comes about because employers can’t accept that work will ebb and flow, and that they are sometimes really just paying someone to be on call.

The time spent at a bullshit job can rarely be used for anything else

  • The most common complaint among those trapped in offices doing nothing is how difficult it is to repurpose the time for anything worthwhile.
  • Specific conditions vary considerably from one job to another. Some workers are supervised relentlessly; others are expected to do some token task but otherwise left mostly alone. Most are somewhere in between.
  • Even in the best of cases, the need to be on call, constantly look over one’s shoulder, and maintain a false front lends itself much more to a culture of computer games, YouTube rants, memes and Twitter controversies than to, say, rock ‘n’ roll bands, drug poetry and experimental theatre. [In my experience it’s certainly true that the need to pretend to work creates stress. But I found that reading and learning programming were much easier to do “undercover” than playing games or going on YouTube or Twitter. A programming website looks a lot more like “real work” – even in a law office – than the familiar YouTube site]
  • Some people may feel too mentally numb from their bullshit jobs to do anything but consume meaningless media. [I find this is true for most full-time jobs, not just bullshit ones.]
  • Utilising a bullshit job to pursue other projects isn’t easy. It requires ingenuity and determination.

Rebuttal to market enthusiasts

  • Graeber says that market enthusiasts are committed to the assumption that a market economy could not create jobs that serve no purpose (but will happily accept that bullshit jobs can exist in the public sector). There are broadly two market enthusiast arguments.

First argument – some employees just don’t understand the point of their job

  • The first argument is that competing firms would never pay workers to do nothing. Their jobs must therefore be useful in some way they simply don’t understand.
  • Graeber points to The Economist for an example of this argument. The Economist article says that the world economy has gotten more complex, and that many tasks have been broken down to become more efficient. Disaggregation may make the task look meaningless since workers will be doing things far removed from the end points of the process.

Second argument – pointless jobs only exist because of government interference

  • The second argument is that pointless jobs only exist in the private sector as a result of government interference.
  • Graeber points out that the problem with this argument is that it’s circular and can’t be disproved. All existing market systems are to some degree state regulated, so it’s easy to insist that any results one likes are the result of the workings of the market and the results one doesn’t like on government interference – and then insist the burden of proof is on anyone who suggests otherwise.
  • That being said, Graeber does accept that government regulation can play a role in the creation of bullshit jobs, particularly the box-ticker type.

Graeber’s rebuttal

  • Graeber uses the example of private universities to rebut both of the above arguments.
  • Benjamin Ginsberg’s book, The Fall of the Faculty, found that over the 30 years from 1985 to 2005, the number of administrative staff at universities ballooned while the number of teachers per student remained largely constant.
  • The “production” side of universities hasn’t increased that much. For the most part, teachers just continue to give lectures, hold office hours, and grade papers as they’ve always done.
  • The number of administrators and mangers actually increased at more than twice the rate at private institutions than at public ones. It seems unlikely that government regulation was the reason that so many more private sector admin jobs were created in this period.
  • Back in the 1950s or 1960s, universities still ran on the old medieval principle that those involved in a certain form of production had the right to organise their own affairs. Universities were basically craft guilds run for and by scholars. Their most important business was producing scholarship; second-most was training new scholars.
  • In the 1980s, Ginsberg argues that university administrators effectively staged a coup. They wrested control of universities from the faculties and directed them toward new purposes.

Why bullshit jobs exist

  • Graeber believes that bullshit jobs, and the bullshit quotient of jobs that are not entirely or overwhelmingly bullshit, have been increasing rapidly in recent years. [As far as I can tell, he doesn’t cite any basis for this belief.]

Different levels of causality

  • A classic problem in social theory is the different levels of causality. For example, if you look at the causes for poverty, the reasons for why any individual person is poor may put the “blame” on the individual. But this reason is not the whole picture because you can may also find that poverty rates vary over time, and across countries. So there are higher level causes that should be considered also. [This reminds of a child persistently asking “Why? Why? Why?” to everything.]
  • Graeber thinks that the rise of bullshit jobs is a broad pattern of social change. He proposes that we need to look at 3 levels of explanation:
    1. Individual level – why do people agree to do bullshit jobs? [And presumably also why the employer hires someone in a bullshit job.]
    2. Social and economic levels – what are the larger forces that have led to the proliferation of bullshit jobs?
    3. Cultural and political levels – why is the bullshitisation of the economy not seen as a social problem? Why has no one done anything about it?
  • The different explanations given by the answers to these questions are not alternatives. Rather, they can all operate at the same time. Much of the confusion in the debate around social issues is caused by people misunderstanding this.

Individual level causes of bullshit jobs

  • Some immediate causes for bullshit jobs include:
    • managers whose prestige is caught up in the number of subordinates they have;
    • weird corporate bureaucratic dynamics – e.g. it may be too difficult to remove someone for incompetence if they have seniority and a long history of good performance, so you neutralise them by giving them a bullshit job;
    • bad management;
    • poor information flow;
    • artificial contests that require box-ticking – e.g. best company to work for, best council.
  • Graeber says that these immediate causes are important but don’t really explain the bullshit job phenomenon. Why were bad organisational dynamics more likely to occur in 2015 than, say, 1915?

Social and economic level causes of bullshit jobs

Rise of the information economy
  • Graeber thinks that it is deceptive to refer to the rise of the “service economy”. We aren’t serving each other more iced coffees or doing more housework for each other. Robert Taylor, a library scientist, suggested in 1992 that we should use the term “information work” to describe that part of the service economy that primarily deals with information.
  • When you break down the economy into four sectors (farming, manufacturing, service and information), you see that service jobs have remained at roughly 20% for over a century. Whereas information jobs have increased drastically, as a proportion of the total economy, since 1860.
  • Graeber says that information work is the zone where bullshit jobs proliferate. He makes clear that not all information workers feel they are engaged in bullshit, and not all those who are engaged in bullshit are information workers. For example, teachers, scientists and librarians are all classed as information workers. But he does think that a lot of information workers feel that, if their jobs disappeared, it would make very little difference to the world.
  • Graeber seems to lay the blame for bullshit jobs largely at the feet of the finance sector, which he seems to think is mostly a scam. The finance sector represents itself as largely about directing investments towards profitable opportunities when it actually does very little of that. Graeber suggests that the overwhelming bulk of its profits come from colluding with governments to create, and then to trade and manipulate, various forms of debt.
  • Whenever a large sum of money – hundreds of millions – is set aside to compensate an entire class of people, a bureaucracy is set up to portion out that money. The money that pays the bureaucracy comes from the same pot, so there is no incentive to distribute the money efficiently – in fact, quite the opposite.
    • For example, in the payment protection insurance (PPI) scandal in the UK in 2006, banks had to pay out a large number of PPI claimants. One of the case studies in the book worked on paying out claimants. But they were paid by the hour so were purposefully mis-trained and disorganised so that they could earn more money.
  • This is basically what the FIRE (finance, insurance and real estate) sector does. It creates money (by making loans) and moving it around in often extremely complicated ways, extracting a small cut with each transaction.
  • Surprising number of bank employees can’t figure out exactly what their bank does. For example, one fund accountant at a custodian bank didn’t really understand how custodian banks safeguard stocks and bonds. In banks, employees are under enormous pressure not to ask too many questions. Graeber says that almost all bank workers he talked to insisted on elaborate secrecy.
  • The closest Graeber could find to an employee with an overall picture of what banks do was “Simon”, who had been employed by a series of large international banks in risk management.
    • Basically Simon’s job was to analyse and find problems in their internal processes.
    • Simon estimated that 80% of the bank’s 60,000 staff were not needed. Their jobs could either be automated or were not needed at all because the process was bullshit to begin with.
    • In one case, Simon came up with a program that would remove the need for 25 staff, and would cost 5% of those staff’s salaries. But the executive would not go for it because it would mean losing his 25 staff.
    • Simon said he found many similar problems and solutions that were never actioned, because they would have resulted in people losing jobs.
Increase in bullshit jobs tied to increase in finance
  • From around 1945 to 1975, there was a “Keynesian bargain” between workers, employers and government. Part of the tacit understanding was that increases in worker productivity would be matched by increases in worker compensation.
  • In the 1970s, compensation remained largely flat as productivity increased. Graeber refers to a graph cited as “EPI analysis of Bureau of Economic Analysis and Bureau of Labor Statistics Data”.
  • Some of the profits from this increased productivity ended up going to the wealthy business owners. But another considerable chunk when to creating new and pointless managerial positions.
  • Graeber says that during most of the 20th century, large industrial corporation were very independent of finance. Executives saw themselves as having more in common with production-line workers in their firms than with speculators and investors. But in the 1970s the financial sector and executive classes effectively fused. CEOs began paying themselves in stock options and moved back and forth between unrelated companies. This set off a cycle whereby workers, who no longer felt any loyalty to corporations, had to be increasingly monitored, managed and surveilled.
Managerial feudalism and classical feudalism
  • Feudalism is essentially a redistributive system. Peasants and craftsmen produce things, and lords siphon off a share of what they produce. The lords then redistribute shares to their own staff, flunkies, warriors, retainers. They also give some of it back to the peasants and craftsmen by sponsoring feasts and festivals and occasional gifts.
  • In a feudal arrangement, it “politics” and “the economy” are one and the same. The goods are extracted through political means and redistributed for political purposes.
  • Under a classical capitalist economic system, in a very competitive market, those who hire unnecessary workers are unlikely to survive. But in a feudal system, where economic and political considerations overlap, the same behaviour makes sense. The whole point in such a system is to grab a pot of loot and then redistribute it.
  • Large corporations are less and less about making, building, fixing or maintaining things and more and more about political processes of appropriating, distributing and allocating money and resources. Even companies that started out making things now derive most of their profits from finance – e.g. General Motors making money from interest on car loans rather than on making cars.
  • One crucial difference between medieval feudalism and the modern version.
    • Medieval feudalism was based on a principle of self-governance in the domain of production, with minimal supervision from anybody else. For example, wheelwrights would collective regulate their own affairs and decide who to allow to be wheelwrights.
    • With modern feudalism, more and more power is vested in managers and supervisors. The actual producers have almost no autonomy.
  • Graeber argues that managerialism has become the pretext for creating a new covert form of feudalism, where wealth and position are allocated not on economic but political grounds (but every day it’s getting harder to tell the difference between the economic and the political). This has happened in corporations as well as in institutions such as universities and charities.
  • One reason why bullshit jobs seem to defy the logic of capitalism is that the existing system isn’t capitalism. It is increasingly a system of rent extraction and, in many ways, resembles classic medieval feudalism.

Cultural and political level causes of bullshit jobs

  • I found most of Chapter 6 “Why Do We as a Society Not Object to the Growth of Pointless Employment?” really long, meandering and hard to follow. At times it was borderline unreadable.
  • I also disagreed with his underlying assumption that we, as a society, think it is morally right to reward those with crappy bullshit jobs and punish those with socially valuable jobs. As a result, I have not bothered to summarise most of the chapter here.
  • One thing I did agree with though, is when Graeber says that much of the bullshitisation is driven by the desire to quantify the unquantifiable. It requires a lot of labour to standardise and quantify certain things.
Politicians love “more jobs”
  • It’s not that anyone at the political level has every directed that bullshit jobs be created. They don’t have to. It’s just that they’ve directed, and create policies that incentivise, more jobs. That is something the Left and Right both agree on. It’s just never specified that the jobs should also be useful.
  • A politician calling for job creation may not be entirely aware of the likely effects of their action. But once a situation is created, politicians can be expected to weigh up the larger political implications of that situation when deciding whether to do anything about it.
  • In 2006, Obama gave the following quote in an interview:
Everybody who supports single-payer health care says, ‘Look at all this money we would be saving from insurance and paperwork.’ That represents one million, two million, three million jobs [filled by] people who are working at Blue Cross Blue Shield or Kaiser or other places. What are we doing with them? Where are we employing them?
—Barack Obama as quoted in Bullshit Jobs
  • Graeber suggests that this quote might be considered a “smoking gun”. Obama acknowledges that millions of jobs in companies like Kaiser or Blue Cross are unnecessary. He even acknowledges that a single-payer health care system would be more efficient. But he’s also saying that the reason (or at least, one reason) for maintaining the current system is to maintain millions of useless jobs.
Difficulty of ascertaining value
  • Graeber notes the difficulty of assigning value to things. Beyond a human’s minimal nutritional requirements and a few other physical things, value must always be subjective.
  • Most people’s perceptions of value are based on some sort of gut instinct. Graeber attempts to write this out explicitly by saying that “when a good or service answers a demand or otherwise improves people’s lives, then it can be considered genuinely valuable, but when it merely serves to create demand … it is not”.
  • Most economists don’t spend time wondering what people should want. Rather, they take people’s preferences as given and assume that if there’s a market for a good or service, it’s clearly valuable to someone.
  • Graeber points out that there is a market in labour as well. If the market were always right, then someone paid $40,000 to do nothing all day would have to accept that the service he provides for the company is worth $40,000 even though it clearly is not. [No, but that’s because of information asymmetry, principal-agent problem and transaction costs. ] Therefore Graeber suggests that the market is not an infallible arbiter of value – but the problem is, nothing else is, either.
  • Money allows us to make precise quantitative comparisons between two things that are otherwise hard to compare.
  • Commodities have economic “value” because they can be compared with other things, but “values” are valuable because they cannot be compared with anything. Values are considered priceless.
  • Life is not really divided between an “economy” where everyone thinks only in material self-interest and other spheres where people behave entirely differently. Real motives are always mixed.
Inverse relationship between social value and economic value
  • Our society has reached the point that the social value of work is usually inversely related to its economic value. That is, the more your work benefits others, the less you are likely to be paid for it.
  • Many people now believe that this is morally right – that we should reward useless or even destructive behaviour and effectively punish those that make the world a better place. [I disagree with this, and in fact almost this entire chapter. I don’t think most people think this way, and I don’t think that this is the reason for the inverse relationship.]
  • If suddenly all the nurses, teachers, firefighters, garbage collectors, bus drivers, mechanics, etc all disappeared, the results would be catastrophic. Whereas there’s a reason why those in the financial sector almost never go on strike. [I mean sure, but the effects of some jobs are longer term than others.]
  • Some apparently crucial positions at the top of an organisation can go unfilled for long periods of time without any noticeable effect, even on the organisation itself. [It’s reasonable to expect more redundancy in a large organisation.]
  • In a 2017 paper, Lockwood, Nathanson and Weyl looked at externalities (both positive and negative) associated with a variety of highly paid professions. They found that medical researchers added $9 value to society for every $1 they are paid. The least valuable was the financial sector, which subtract $1.50 from society for every $1 they are paid. The breakdown was:
    • researchers +9
    • schoolteachers +1
    • engineers +0.2
    • consultants and IT professionals 0
    • lawyers -0.2
    • advertisers and marketing professionals -0.3
    • managers -0.8
    • financial sector -1.5
  • In 2009, Lawlor, Kersley and Steed as part of the New Economic Foundation in the UK conducted a similar study. They looked at 6 occupations – 3 high-income, 3 low. They found that city bankers, advertising executives and tax accountants destroyed between $7 and $11.50 of social value per $1 paid, while hospital cleaners, recycling workers and nursery workers generated between $7 and $12 in social value per $1 paid.
  • Graeber acknowledges that doctors may be an exception to the inverse relationship observed. But even then, he doubts how much doctors contribute to human health as it is often cited that the overwhelming majority of improvement in longevity since 1900 is due to hygiene, nutrition and other public health improvements rather than to improvements in medical treatment.
Reasons for the inverse relationship
  • Graeber finds it difficult to explain this inverse relationship. He thinks that the obvious reasons do not work:
    • It’s not about education and training because there are thousands of PhDs living at or near the poverty line.
    • It’s not about supply and demand because the US has a shortage of trained nurses and an oversupply of law school graduates. [But the positive and negative externalities mentioned above exactly do explain why there is a shortage of socially valuable workers and an oversupply of socially negative ones.]
    • In a footnote, he refers to Alex Tabarrok’s point about marginal utility and the diamonds-water paradox, but doesn’t answer it.
  • There is a sense, Graeber argues, that those who choose to benefit society, especially those who get the benefit of knowing they benefit society, shouldn’t be rewarded for it. By the same token, those who have to suffer through pointless or even harmful work just for the sake of money should be rewarded for more money. [I find it really hard to take this argument seriously. It’s like Graeber is suggesting that “society” is a singular entity that doles out jobs and pay according to what the society thinks is moral. That’s just not how things actually work.]
  • Many businesses now feel that if there’s worth that’s gratifying in any way at all, they shouldn’t have to pay for it. There is an emerging “voluntariat” with capitalist firms using the work of unpaid interns, internet enthusiasts and other volunteers. The free software industry is a paradigm in this respect.
  • There seems to be a broad consensus not so much that work is good, but that not working is very bad. [I agree, but I think there are some good reasons for this. If someone became rich through their own merits and retired – even if retired early – I don’t think they face much hostility. As compared to someone who can work, but chooses not to, supporting themselves instead through the work of others.]
Bullshit jobs keep people from changing the system
I believe that this instinct to perpetuate useless work is, at bottom, simply fear of the mob. The mob (the thought runs) are such low animals that they would be dangerous if they had leisure; it is safer to keep them too busy to think.
—George Orwell, Down and Out in Paris and London
  • The more the economy becomes a matter of distribution of loot (rather than producing it), the more inefficiency and unnecessary chains of command actually make sense. Inefficiency helps soak up as much of the loot as possible.
  • Some people say that the reason we are not working a 15 hour week, as Keynes predicted, is because we have chosen to increase consumption rather than to increase leisure. Graeber says this is partly true. But it’s not because we’re spending more time manufacturing PlayStations and serving each other sushi. It’s because we’ve invented a bizarre sadomasochistic dialectic whereby we feel that pain in the workplace is the only possible justification for our furtive consumer pleasures.
  • Bullshit jobs foster a political landscape filled with hatred and resentment. The unemployed resent the employed. The employed are encouraged to resent the poor and unemployed. Those with bullshit jobs resent those who do real productive or beneficial labour. Those who do real productive or beneficial labour increasingly resent the “liberal elite”, who have jobs that allow them to live well while doing something useful, high-minded or glamorous.

History of bullshit jobs

People came to view time like money

  • Nowadays we have this idea that an employee who is not working is stealing from their employer. “You’re on my time – I’m not paying you to lounge around.” Idleness is theft.
  • The idea that one person’s time can belong to someone else is quite peculiar and modern. To an ancient Roman or Greek, the closest they could come was the idea of renting someone as a slave for a limited period – e.g. a day. But no free man would have agreed to enter into this arrangement. To be a slave, even temporarily, was considered the most degrading thing possible.
  • The vast majority of examples of wage labour in the ancient world are with people who are already slaves. A slave master may send his slave to work in a factory, with the wages earned being shared between the slave and master.
  • When clocks and pocket watches were disseminated more widely, beginning in the late 1700s, attitudes to time changed. People were encouraged to see time as a finite property to be carefully spent, much like money. It also allowed a person’s time to be divided into uniform units that could be bought and sold for money. This was both a moral and technological change.
  • Over the course of the 18th and 19th century, the middle classes came to see the poor as being poor because they lacked time discipline. They were poor because they spent their time recklessly, just like they spent their money.

Life cycle service in feudal systems

  • In theory, feudal society was a vast system of service. In addition to serfs, lower-ranking feudal lords served higher ones, and higher-ranking feudal lords served the king.
  • Feudal systems also had what is called “life-cycle” service. Almost everyone was expected to spend roughly the first 7-15 years of their working lives as a servant in someone else’s household. For example, pages were apprentice knights, noblewomen would serve as ladies-in-waiting. Servants were paid.
  • Almost everyone in the Middle Ages assumed that service was something respectable people did in the first part of their working lives. The reason for the service was to educate the young people. They would learn self-discipline, manners, and how to behave like a respectable adult.
  • Because of these life-cycle arrangements, attitudes towards work in medieval Northern Europe were quite different from those in other parts of the world. For example, Aristotle thought work made you a worse person. Because it took up so much time, it made it difficult to meet your social and political obligations.
  • With the advent of capitalism, relations of service transformed into permanent relations of wage labour – relations between those who owned capital and those who did not. Guild structures broke down. This meant that millions of young people found themselves trapped in a sort of permanent social adolescence. For example, apprentices could become journeyman but journeyman could not become masters. In traditional terms, this meant they could not marry and start families of their own.
  • Many young people started to rebel and just started marrying early. This was the birth of the proletariat as a class. “Proletariat” is derived from a Latin word meaning “those who produce offspring”.
  • This set off a moral panic among the employing classes of English Calvinists/Puritans. A long history of attempts to reform the manners of the lower classes followed, from Victorian workhouses to workfare and similar government programs today.
  • The poor were viewed as frustrated adolescents. Work had traditionally been the means by which they became proper, self-contained adults. When this avenue became unavailable, work was taught as having value in itself. After the industrial revolution, work was celebrated even more.

The Universal Basic Income (UBI) as an alternative to bullshit jobs

  • The current allocation of labour has nothing to do with economics or even human nature. It is ultimately political. Instead of having tons of bullshit jobs, we could instead distribute the tasks that needed to be done more evenly, and reduce everyone’s working hours.
  • People only become individuals through the care and support of their fellow human beings. “The economy” is just the way we provide ourselves with the material provisions to do this.
  • Graeber suggests a universal basic income (UBI) as a program that might detach work from compensation. He emphasises that Bullshit Jobs is about a problem, not a particular solution. But there are no anti-bullshit job movements and the UBI is the only solution he’s been able to identify that would reduce the size and intrusiveness of government.
  • Graeber notes that there are radically different visions of what a UBI would look like. Some conservative versions would replace most or all existing state support (e.g. hospitals). More radical versions would leave most existing supports in place [but presumably not unemployment benefits].

Reasons for favouring a UBI

  • Graeber explains that he is an anarchist and generally does not prefer solutions that give more power to governments or corporations. Rather, he prefers solutions that give people the means to manage their own affairs. The reason he likes a UBI is because it would allow huge parts of government to be closed down. In particular, it would close down the most intrusive parts that engage in surveillance of ordinary citizens.
  • There are apparently studies that show that any system of means testing, no matter how it’s framed, will mean at least 20% of those who legitimately qualify for benefits give up and don’t apply. In the UK, currently 60% of those eligible for unemployment benefits don’t get them. [Unfortunately, Graeber does not have references for either of these figures.]
  • Feminist critics also point out that the wage-labour system defines what is “real” labour (i.e. that which was bought and sold) and what isn’t. Most women’s labour was not considered “real” labour under this definition.
  • A pilot study of Basic Income in India found that domestic violence goes down significantly, and that social inequalities lessen. Girls started being given the same amount of food as boys, disabled people were more accepted in village activities, and young women stopped having to act shy and modest. Giving everyone an equal amount of money has symbolic power.
  • The UBI should be given even to people who don’t need it. It’s worth it, to establish the principle that, when it comes to what’s required to live, everyone deserves that, equally, without qualification.
  • Graeber seems to view the main advantage of a UBI as detaching livelihood from work. – A UBI would remove all the gratuitous sadism of workplace politics as people will be able to quit if they want, without any economic consequences.

Objections to a UBI

  • A common objection to the UBI is that people will just stop working.
    • Graeber disagrees. He thinks that many humans will continue to “work” in the sense of doing something useful or beneficial to others. Some workaholics – or just people who want extra money – will more than compensate for occasional slackers. The work required to maintain people in comfort and security is not really that formidable, if much of our current work is bullshit anyway.
  • Another objection is that people will work, but they’ll just do things that are interesting to them – not to anyone else. So there may be tons of bad poets and street mimes, but nothing will get done.
    • Graeber considers this a more serious objection than the one that says people just won’t work. However, he doesn’t take this one too seriously either. He says it’s hard to imagine more than 10 or 20% of the population choosing to do things most others consider pointless. But already 37 to 40% of workers in rich countries already feel their jobs are pointless. So he thinks it’s unlikely that a UBI would result in a distribution of labour more inefficient than what we currently have.
  • In a footnote, Graeber suggests that if a UBI were implemented, rent controls could need to be imposed else landlords could just double rents to grab the additional income. [This seems to betray a very poor – or at least superficial – understanding of economics.]

Other Interesting Points

Points made by Graeber

  • In places without clocks, time is measured by actions rather than actions measured by time. For example, in Madagascar, rural people would say that a walk to another village would take two cookings of a pot of rice.
  • The public school systems, which sound a bell every hour to get students to move to a different class, was consciously designed like this to train children for future factory work.
  • Many early factories didn’t let workers bring their own timepieces since the owner would rig the factory clock.
  • Emotional labour:
    • The work required to create an maintain a happy, good-natured persona is a type of “emotional labour”.
    • Sociologist Arlie Russell Hoschschild first introduced the idea of emotional labour in a study of flight attendants. He found that the amount of emotional labour required caused them to feel haunted by emptiness, depression or confusion, unsure of who or what they really were.
  • Baumeister, Wotman and Stilwell (1993) found that men or women who experienced unrequited love during adolescence could usually get over it eventually. In contrast, people who had been the objects of unrequited love still struggled with guilt and confusion long after. The researchers thought that one reason for this was the lack of cultural models. There is much literature on people who feel unrequited love but very little featuring the objects of unrequited love.

Points made in testimonials

  • Internet banner ads are largely a scam. The agencies selling those ads have studies that show Internet users didn’t even notice and almost never clicked on them. But they would still hold junkets with clients and present elaborate “proof” of the ads’ effectiveness. [If the “proof” is misleading or false then, yeah, that’s a scam. But the fact that people almost never click on them doesn’t make it a scam since it implies that some people still do. This is built in to the price of banner ads. You can get thousands of impressions for mere cents.]
  • Many government or non-governmental organisations (NGOs) may be actively doing harm. NGOs often profit from inequality – it is the reason they exist.
  • A lot of charity work in large companies is driven by awards given for “Best company to work for in X” or similar. Charity work may be part of the criteria for those awards.
  • One young woman who worked as a personal assistant to a NATO official claimed to end up writing many strategic plans for operations in a war zone. This was despite her having no military experience.
  • One guy who writes term papers for lazy (and dishonest) students found himself largely writing business and marketing papers. It seems people study business and marketing are rarely doing it because they are passionate about it. Instead, they just do it get a high paying job. If they see the degree as merely a business investment, it makes sense that they’d be willing to invest a little more to reduce their workload and guarantee a good grade.

My Thoughts

I don’t think Bullshit Jobs is very well-written. The first parts are all right, and the anecdotes of people’s various bullshit jobs are entertaining. Chapter 6 and 7 are borderline unreadable in parts. They come off like long, unstructured rants. Graeber also uses plenty of unnecessarily long sentences that are difficult to parse.

Bullshit Jobs seems a bit schizophrenic. At times it seems to lean into the ridiculousness of the underlying idea that so many jobs are just bullshit. For example, when it uses the terms flunky, duct-taper, etc to describe the different types of bullshit jobs. At other times, Graeber seems to treat the bullshit jobs phenomenon as a serious problem that merits thorough and thoughtful consideration.

The thing is, I do think the bullshit jobs phenomenon merits more thoughtful consideration. It is therefore a shame that Graeber doesn’t really do this in his own book. There isn’t much more to be gained from reading the book than in reading his original 2013 article.

For example, Graeber initially sets out 5 types of bullshit jobs, but goes on to discuss them all in the same way. Flunkies and goons clearly provide value to someone, even if not to society as a whole. The reasons why these jobs exist are therefore likely to differ from the reasons why completely pointless jobs (i.e. that provide no value to anyone) exist.

Overall, I think the book raises some interesting ideas and points but does not go on to develop them properly (and fairly). It is important to note I suppose that the full title is Bullshit Jobs: A Theory. So Graeber is not necessarily trying to prove that his theory is correct, just raising the question.

Buy Bullshit Jobs at: Amazon | Kobo <– These are affiliate links, which means I may earn a small commission if you make a purchase. I’d be grateful if you considered supporting the site in this way! 🙂

Do you think I’m being too harsh on Graeber? Have you had a bullshit job yourself? Share your thoughts in the comments below!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.