On Bureaucracy and the Left

I’VE WANTED TO read a book like David Graeber’s The Utopia of Rules: On Technology, Stupidity, and the Secret Joys of Bureaucracy for a long time. Like me, Graeber believes bureaucratic solutions are ultimately insufficient to deal with the economic and ecological problems facing society. Lamenting that “when those who identify with the Left do have anything negative to say about bureaucracy, they are usually forced to adopt a watered-down version of the right-wing critique” — a situation he calls a “political catastrophe” — Graeber attempts to sketch out a genuinely left-wing critique of bureaucracy. It’s long overdue. But it’s also a very difficult proposition.

“Bureaucracy” can be hard to pin down. The word conjures up images of paperwork, red tape, and fussy clerks arbitrarily enforcing rules that are barely explained, much less understood. But it also suggests the idea of the regulatory state — the bureaucracy that both protects and restrains commerce and governs the use of public spaces. One of the reasons that “deregulation” has become so politically fashionable in the West over the last few decades is precisely because “regulation” in the abstract reminds people of frustrating situations. After all, no one likes the DMV — and no one wants to belong to the political party of the DMV either. But regulation also plays an important role in holding powerful interests in check. When I come across a book like Cass Sunstein’s Valuing Life: Humanizing the Regulatory State, and it talks about the role of the bureaucrats in correcting “market failures” in areas like the environment or public health, I can’t help but nod along. To be honest, this has tripped me up for a long time. Finding bureaucracy unsatisfying isn’t enough. A real critique needs to offer an alternative — anything else is just whining.

Certainly many complaints about bureaucracy turn out to be facile up close. In a review of yet another critique of red tape, Ben Kafka’s The Demon of Writing: Powers and Failures of Paperwork, Rob Horning writes:

The inefficiencies and inadequacies of paperwork — and the clerks whose thankless task is to manage it — lend themselves to being used to construct the fantasy of a natural, uncorrupted, hierarchy- and bureaucracy-free direct democracy […] In this sense, paperwork encourages a misunderstanding of power as a medium, divested of actual people responsible for abusing it. Pursuing the dream of domesticating power becomes a matter of endless formal restructuring, while the mundane organizational work of politics is demonized as the sort of bureaucratic troublemaking “everyone” wants to dispense with.[1]

Horning isn’t defending bureaucracy here — he’s just pointing out that railing against paperwork isn’t automatically productive. Discussions about political forms have their place, but they are no substitute for actually challenging those who hold power in the here and now.

Bureaucracy, like almost all human inventions, is a tool, and it reflects political priorities. It will never, on its own, be able to counteract power imbalances, because those imbalances will influence the laws that set the perimeters bureaucracy acts within, the priorities executives set for regulators, and the market signals that regulators look to for information. But it is not the source of those imbalances, and any critique of bureaucracy worth keeping needs to acknowledge this. To Graeber’s credit, he readily acknowledges that there are situations where bureaucracy is not only useful but necessary, offering organ donation programs as an example where waiting lists, lotteries, and paperwork are the safest ways to tame an inherently arbitrary process. More importantly, he embeds his criticisms of bureaucracy in a much larger intellectual framework. For Graeber, the bureaucratization of society is a part of something even more troubling: “The gradual fusion of public and private power into a single entity, rife with rules and regulations whose ultimate purpose is to extract wealth in the form of profits.”

Before launching into his critique of total bureaucratization, though, Graeber turns his attention to the more prevalent, conservative critique of bureaucracy, specifically as articulated by Austrian School economist Ludwig von Mises in his 1944 book Bureaucracy. This is important, because the Austrian critique still influences thinkers and regulators from across the political spectrum. For example, Sunstein’s Valuing Life owes no small debt to von Mises’s fellow Austrian School economist Friedrich Hayek — Sunstein even explicitly credits Hayek with forging many of his ideas. Since Sunstein served as administrator of the White House Office of Information and Regulatory Affairs (OIRA) during most of Barack Obama’s first term, he shows the impact of the Austrian School critique in the mainstream American politics, even among liberals. In Graeber’s summary, von Mises argues “that by definition, systems of government administration could never organize information with anything like the efficiency of impersonal market pricing mechanisms.” Hayek famously developed this line of thought even further, and this is what appeals to Sunstein, who advocates “government by discussion,” and details how the Obama-era OIRA “aggregates” information from a variety of public and private sources, as well as incorporating market cues, in an effort to alleviate this inefficiency. So the laissez-faire Austrian School not only influences how we talk about bureaucracy, but how the regulatory state itself operates.

Graber however, finds that critique lacking. First he notes that von Mises’s ideas about bureaucracy are rooted in the belief that the kinds of welfare states then being constructed in nations like France and the United Kingdom would inevitably lead to “fascism” — a prediction that has quite obviously not come true in the 70-plus years since Bureaucracy’s publication. He also objects to the way von Mises fails to account for the way bureaucracies and markets actually come into being, noting that, “Historically, markets are generally either a side effect of government operations […] or were directly created by government policy.” Indeed, much of Graeber’s most famous book, Debt: The First 5000 Years, concerns itself with explicating that history. The result, in Graeber’s view, is that markets and bureaucracy cannot be separated. He points at the explosion of legal clerks, registrars, and police during the British Industrial Revolution as one example of this process. He argues that “even right-wing critics like von Mises were willing to admit — at least in their academic writing — that markets don’t really regulate themselves, and that an army of administrators was indeed required to keep any market system going.”

It’s worth noting that the deregulation of the post-Reagan/Thatcher years — which was often executed in the name of Austrian School principles — hasn’t reduced the role of bureaucracy in everyday life. Graeber himself labels deregulation a “scam,” and half-jokingly asserts the “Iron Law” that:

[A]ny market reform, any government initiative intended to reduce red tape and promote market forces will have the ultimate effect of increasing the total number of regulations, the total amount of paperwork, and the total number of bureaucrats the government employs.

Financial markets in particular are symbiotic to government. After all, governments maintain currencies, enforce debts (especially through the court system), and underwrite financial institutions. Graeber puts it succinctly: “There’s no such thing as an ‘unregulated’ bank. Nor could there be.” Certainly, it’s fair to wonder how a nation where most bank deposits are insured by the federal government (and where economic crises are answered with massive public bailouts for financial institutions) could ever really “deregulate.” So Graeber is correct that any real left-wing critique of bureaucratization needs to reckon with the financialization of the American economy.

But Graeber takes this reckoning and uses it to introduce an even more potent conversation about structural violence and its effects. Graeber traces the arbitrariness of bureaucracy to the violence underlying society’s institutions: “All of these are institutions involved in the allocation of resources within a system of property rights regulated and guaranteed by governments in a system that ultimately rests on the threat of force.” Throughout history, property rights have only existed when backed by some form of government. And that backing has traditionally meant that those rights will be enforced through physical (i.e., violent) means. For example, if a stranger sets up camp in your backyard, you can call the police to have him taken away. To Graeber, this explains why bureaucracies tend toward arbitrariness — force is inherently arbitrary, and any system that uses force will ultimately draw arbitrary lines itself. In this way he echoes the 20th century novelist/revolutionary Victor Serge’s eloquent formulation, “All violence is dictatorial.”[2]

Drawing on the work of thinkers like Jo Freeman, Graeber advocates direct democracy, where decisions are made by consensus, and structures are built to minimize the power of small factions and emphasize consensus building. This means he rejects the legitimacy of violence, even in the name of defending property rights. Graeber’s society would be ad hoc, with no permanent formal rules or institutions. The problem, though, is that it’s not immediately apparent how this vision can help us in the here and now. It needs to be tested. That means taking the ideas contained in The Utopia of Rules and letting them rub against a more conventional view of bureaucracy, to see what sparks emerge.

Sunstein’s Valuing Life offers us just such a work. It begins with an “insider” account of Sunstein’s time in the White House. Rather than a memoiristic tale about the personalities of the Obama administration, however, Sunstein focuses almost entirely on the OIRA’s day-to-day operations, laying out exactly how the office comes to approve regulations. He clearly invests government regulation with a great deal of importance, and while I wouldn’t necessarily call him passionate, he is certainly convinced that bureaucracy can do real good, specifically in the context of the regulatory state. Sunstein never discusses the kind of structural violence that alarms Graeber, but he does address the apparently arbitrary nature of government regulations — and by implication the red tape they generate.

As noted above, Sunstein sees bureaucracy as the necessary remedy for behavioral market failures, “a set of problems that make markets work imperfectly, including unrealistic optimism, myopia, and self-control problems.” Sunstein is deeply impressed with Hayek’s ideas about how markets process information, but he worries that certain heuristics (mental shortcuts) can sometimes influence behavior in negative ways. While he admits this leads him in directions Hayek would never go, Sunstein ultimately sees his conclusions as an outgrowth of his faith in markets, not a negation of them. To illustrate why behavior market failures are worth confronting, he looks at many different examples, but the most convincing ones involve public heath issues, such as vaccinations:

In deciding whether to vaccinate their children from risks for serious diseases, people show a form of “omission bias.” Many people are more sensitive to the risk of the vaccination than to the risk of the diseases against which the vaccination would provide protection — so much so that they will expose their children to a greater risk from “nature” than from the vaccine.

It isn’t hard to see this reflected in the real world. In early 2015, there was a nationwide outbreak of measles, which was tied in part to unvaccinated children coming into contact with the disease at an amusement park. (The CDC estimated 113 cases could be tied directly to this outbreak.[3]) Sunstein’s point is that these kinds of behavioral market failures are fairly common, and that a strong regulatory state can offset them.

While Sunstein never puts it quite so explicitly, his book strongly implies that the “arbitrariness” people attribute to regulations stems from the way they contradict our (deeply flawed) intellectual shortcuts. I certainly see the appeal of this line of thinking. For one thing, it acknowledges one of my biggest concerns about bureaucracy — that regulatory solutions always risk a certain brittleness, because their benefits are diffuse (it’s not like there’s a way of knowing who definitely would contract measles absent a vaccine), and most people will only be made aware of a particular regulation when it contradicts what they want to do. So a political program built on regulation and top-down, technocratic solutions will be at a disadvantage, forever associated with policies that look arbitrary, or ineffective, even when they’re not.

But Graeber’s concerns go far beyond that, insisting that bureaucracy, in all its forms, contains arbitrary distinctions by its nature. And that this arbitrariness betrays the inherent conservatism of all bureaucratic solutions. Valuing Life provides an interesting test for this idea because Sunstein himself is so comfortable with markets, and actively seeks market inputs when formulating regulatory decisions. He even uses markets in places a lay reader might not expect, such as when he proposes changing the way the US government calculates the value of a statistical life, or VSL. A VSL is the dollar value assigned to an individual life saved (or expected to be saved) by a new regulation. In the United States, it’s approximately $9 million.[4] The idea of a VSL draws a lot of criticism from certain quarters — in Debt, Graeber himself explicitly scoffs at the practice of assigning a dollar value to a human life — in part because it seems to be an obvious case of a bureaucracy’s need for quantification leading to the dehumanization of humans, turning lives into numbers. But Sunstein defends the process vigorously, and variable VSLs form the backbone of his plan to “humanize” the regulatory state.

Currently, the United States uses a static VSL, using the same figure to evaluate each proposed regulation. Sunstein believes that this is a mistake, and that regulators should use different VSLs for different situations, and different populations. This takes him down a path diametrically opposed to Graeber. After all, variable VSL might mean that — on paper — the life of a poor person would be “valued” lower than that of a rich person. But Sunstein is emphatic that this practical benefit outweighs the abstract idea that some people’s lives are being “valued” less: “The real question is the real effects of different numbers for actual human beings.” In many ways, he is the ultimate technocratic reformist, eager to use statistical formulas and logical (or logical sounding) arguments to advance an idea that on paper might seem contrary to humanist values. As Sunstein explains, “If an agency uses a VSL of $15 million, many more regulations will be justified than if it uses a VSL of $2 million.” A high VSL for certain public health risks, like cancer (people often express a special aversion to cancer in surveys), would strengthen regulator’s hands against some businesses that pollute. And a low VSL for poor people would mean that regulators would impose fewer costs on them, which (if it worked) could alleviate some financial strains on those who can’t afford it.

The question, though, is does this suggestion seem likely to work — or would it only introduce more arbitrariness into the system? Whatever their statistical uses, VSLs themselves are pretty abstract. Sunstein describes a few inputs, but in Valuing Life, the population at large mostly seems to express itself through two main paths: surveys and markets. Neither of these strikes me as perfect. Sunstein never really explains why surveys are exempt from the heuristics that lead people astray during other occasions. For example, people might express a special aversion to cancer because it’s more common than other serious diseases, some of which they might be willing to pay more to avoid if they knew more about. Also, markets themselves can be subject to all manner of distortions. After all, that’s the entire point of regulation in Sunstein’s view — that heuristics sometimes lead to cognitive market failures. It is genuinely surprising that Sunstein never adequately explains how regulators distinguish a genuine preference as expressed through market signals from a preference expressed due to a cognitive market failure. Finally, reading Sunstein alongside Graeber brings another objection into focus. Namely, that Sunstein’s reliance on markets inevitably means that regulations will end up serving the needs of the market itself, not society as a whole.

It’s worth seeing how this might play out. Variable VSLs already exist internationally, in the sense that regulators in developing nations like China and India often use a lower VSL than the United States. This makes sense in Sunstein’s view, because the loss of economic growth might be more serious in a nation that is still struggling to bring up its average standard of living. The problem, however, is that in a nation like China, where there are strict limits of both speech and movement, it’s hard for workers and citizens to really articulate their preferences. So when he writes,

Why should a worker in Beijing be subject to significantly higher death risks then a worker in Los Angeles? The answer is that so long as the distribution of global income has the form that it does a system that gives [Chinese] workers the same protections as American workers is not in the interest of [Chinese] workers[5]

he is relying on abstract economic principles and his own intuitions about the interests of Chinese workers, not on their actual preferences (which they cannot adequately express in the present political environment). So it’s fair to ask if the VSL invoked here reflects the actual preferences of Chinese workers, or the needs of the Chinese government and international commerce. This applies domestically as well. When economic activity becomes the primary way for people to express their preferences, it’s inevitable that markets — and therefore major market players — will become incredibly powerful.

None of this invalidates the regulatory state as a whole. But it shows that the regulatory state has its limits, especially when it comes to offsetting the considerable social and economic power of finance and big business.[6] Where paperwork, bureaucratic red tape, and the regulatory state all come together is that they are all products of society as it exists now — and therefore are poorly suited to changing it. None of this means that leftists should reject “the mundane organizational work of politics.” In fact, it indicates the opposite — that the left needs to emphasize organizing and activism, and work to facilitate change from the bottom up.

Certainly, there are times when the regulatory state is the only viable way to address an immediate political problem (for example, in the wake of the 2008 financial crisis, it certainly seems prudent to try and increase capital requirements at banks). But it’s important to remember these kinds of solutions are inadequate over the long haul, because they do not change the underlying balance of power in society. And that balance is the key not only to a true left-wing politics, but to democracy itself. This is why unions are so important to a democratic society: They are one of the few means available for workers to express their own preferences without the distorting filter of markets or market-oriented bureaucracies.

I admit that I remain unable to articulate the kind of all-purpose critique of red tape that I want. But Graeber certainly demonstrates that the best response to the Austrian School is to reject its premises: bureaucracy cannot be discussed separately from markets, because the two are entwined. Perhaps what’s important is not to denounce bureaucracy in broad terms but to simply acknowledge its limitations, and the way its most frustrating aspects stem from imbalances of power that exist in society at large (imbalances the Austrian School refuses to confront). The correct way to address the tedium and arbitrariness of bureaucracy — both in the public sphere and the corporate one — is to address those imbalances.

This brings to mind my own personal favorite lament of bureaucracy, from Chinua Achebe’s essay “What Has Literature Got to Do with It?” There, Achebe acknowledges the importance of bureaucratic experts, but also notes, “having an expert among us does not absolve the rest of us from thinking.” As he explains:

He can build a bridge for us perhaps, and tell us what weight of traffic it can support. But he can’t stop us from hiring an attendant who will take a bribe and look the other way while the prescribed weight is exceeded. He can set up the finest machinery for us, but he can’t create the technician who will stay at his post and watch the controls instead of going for a chat and some groundnuts under a mango tree outside.

Achebe is writing to defend the place of cultural work, specifically writing fiction, in the wider political conversation. As someone who is most at home in books, I certainly agree with that sentiment — while poorly suited to articulating a specific political program, the arts can and do play a part in the wider cultural discussions that democracy requires. But the conversation itself is what really matters. And that conversation requires independent voices, ones not beholden to big market actors or other powerful entities.

Bureaucracy seems central because it is one of the most visible parts of political life — from tax forms, to lines at the DMV, to the regulations well-funded lobbyists decry on television. It’s easy for a conversation about making bureaucracy work better to take time away from the more important work about making society work better, specifically by empowering citizens to express the preferences not through markets but through organizing and democratic participation. Achebe reminds us that we all have an obligation to think. Even if you find Graeber’s direct democracy too radical, he is correct that citizens need to communicate their desires themselves, directly. The Black Lives Matter movement, which has largely kept itself separate from bureaucratic political parties, while still pressuring elected officials, offers one example of an independent left. More are needed.


[1] Horning’s essay can be found here.

[2] A more detailed discussion of Serge’s political writing can be found here.

[3] The CDC data on this outbreak can be found here.

[4] Sunstein himself specifies that the figure is “around $9 million in 2013 dollars.”

[5] In the book itself, Sunstein talks about China and India simultaneously, and uses the names of the countries interchangeably in this passage. Since his statements are intended to apply to both, I have edited the quote for clarity. In the original, “Indian” appears in some parts of this passage instead of “Chinese.” My “edits” appear in brackets.

[6] It’s worth acknowledging that Sunstein himself readily concedes that “it is important to see that the best response to unjustified inequality is a redistributive income tax,” but he also clearly sees the regulatory state as the primary means of protecting citizens from risks to the environment and public health, among other problems, that might stem from imbalances of social or political power.


Guy Patrick Cunningham is a writer and editor living in Brooklyn, New York.