The insurance company Aetna pays its workers to sleep eight hours a day. Or, more precisely, it has a bonus scheme that pays workers $25 dollars for every 20 days that they have slept for at least eight hours. It is easy to understand the motivation behind this scheme. Sleep deprivation can severely impact on health and well-being, which in turn impacts on worker productivity. The American Academy of Sleep Science has actually tried to quantify the impact, suggesting that as much as $63.8 billion is lost every year in the US due to sleep deprivation.
How should we think about schemes of this sort? Aetna’s sleep reward programme is just one of a growing number of corporate wellness programmes. These programmes encourage (sometimes compel) workers to submit to increased surveillance of their personal lives. Companies realise that a healthy worker is a productive worker and that the more they can do to improve employee health and well-being, the better their bottom lines. If this means monitoring calorie intake, sleep and physical activity, so be it — all the better if you can sell the programme to employees as being a way in which they can improve their own health and fitness. What’s more, corporate wellness programmes are just the tip of a very large iceberg. For a very long time, employees have to submit to all kinds of corporate interference, from bag searches and drug tests to e-mail and internet monitoring. All just to ensure that the company meets its shareholder’s expectations.
There is something quite odd about this level of corporate interference. As Elizabeth Anderson points out in her book Private Government, if you were to imagine the most unjust form of government, it wouldn’t be far off the kinds of government to which many people are subjected in their working lives. And yet we don’t think of this as a kind of unjust government. Why not? Anderson argues that history is partly to blame. We have inherited an idealistic view of contracts and free markets that ignores some of the distinctive features of employment contracts. Furthermore, we have a tendency to think that the state is the only form of government that is worth criticising and challenging. This has created a massive blindspot when it comes to recognising a major source of injustice in modern life.
Anderson’s argument is certainly provocative. She presents it in a distinctive style, favouring a detailed examination of the history of ideas around the free market and egalitarianism, as well as the consequences of the industrial revolution, over an analytical presentation. While this historical detail is fascinating and well-worth reading, I want to offer a more analytical reconstruction of her argument in this post. I start by identifying the major conditions of injustice that are to be found in employment contracts. This forms of the backbone of Anderson’s argument against modern forms of work. I then look at the objections to this argument.
1. The Argument for the Injustice of Workplace Governance
The idealistic view of contractual negotiation comes down to us from Adam Smith. Contrary to what you might think, Smith was an egalitarian. He lived in a society that was riven with hierarchies. His case for the free market was based, in part, on the belief that it could disrupt these inherited hierarchies (this is one of the major historical claims in Anderson’s book). When people get together to trade on a free market, they meet as equals: one person has something they want to sell; the other wants to buy. Neither has to defer to the other on the grounds of inheritance or birthright. They negotiate on a price that is mutually beneficial, otherwise they don’t make the trade. The contracts are win-win, not win-lose.
There are, of course, a number of objections and exceptions to this idealistic view. Anderson argues that employment contracts are a major exception. The tendency, since the time of Smith, is to use contracts for the sale and supply of commodities (food, cars etc) or short-term services (entertainment, medical examinations) as the norm to which all others are subsumed. Parties to these contracts negotiate, exchange and separate in short order. It thus appears that they engage with one another on largely equal terms. Furthermore, even if there is some inequality of bargaining power (and, in reality, there often is) its effects are localised in time and space. The parties will soon exit the contract and return to their prior position. Subsuming employment contracts to this norm, people tend to believe that if employment contracts have reasonable entry and exit conditions, they are just. This leads them to ignore the main feature of the employment contract: the actual period of employment during which the employee is subordinate to the employer. As Anderson puts it, most economists and politicians seem to ignore what goes on inside the ‘black box’ of employment (see image below).
And what goes on inside that black box is, according to Anderson, remarkably unjust. Employers effectively “purchase command over people” (2017, 57) and are then given much leeway in how they exercise their powers of command (though this varies from jurisdiction to jurisdiction), doing things that governments would never be allowed to do. This is justified on the grounds that it is a matter for the ‘market’ or ‘private’ individuals to determine, and that employees are adequately compensated for the intrusions visited upon them by their employers.
This is far from the truth for many workers. In practice, employment contracts are major sites of unjust treatment. Anderson’s argument, as best I can tell, runs something like this:
- (1) A system of government is unjust if it: (i) treats its subjects harshly and with a lack of dignity/respect; and/or (ii) undermines their freedom, particularly by subjecting them to arbitrary authority/domination.
- (2) The workplace, particularly for certain kinds of workers, is a system of government that (i) treats its subjects harshly and with a lack of dignity/respect; and/or (ii) undermines their freedom by subjecting them to arbitrary authority/domination.
- (3) Therefore, workplace government is unjust.
Premise (1) of this argument should be relatively uncontroversial but I should say something about it. In terms of the first condition (i), concepts like ‘dignity’ and ‘respect’, though widely used in political and moral discourse, are pretty vague and can be critiqued for this. Nevertheless there are some reasonably clearcut instances of undignified treatment, usually associated with treating people as ‘things’ or like non-human animals, and as we’ll see in a moment it is instances of this sort that animate Anderson’s argument. In terms of the second condition, ‘arbitrary authority/domination’ has a reasonably well-understood meaning in the debate about political freedom. There are two leading conceptions of political freedom in the modern era: freedom as non-interference and freedom as non-domination. The former is all about actual interferences by an authority into your life choices either through force or coercion; the latter is all about potential interferences due to living at another person’s mercy or goodwill. There has been a major revival of interest in freedom as non-domination as a condition of justice since the 1970s.
Premise (2) is the key to the argument. Anderson doesn’t provide an exhaustive defence of it in her book, but she does appeal to some choice examples. Consider the following quote with which she starts her book (footnotes excluded):
Walmart prohibits employees from exchanging casual remarks while on duty, calling this ‘time theft’. Apple inspects the personal belongings of all their retail workers, who lose up to a half-hour of unpaid time every day as they wait in line to be searched. Tyson prevents its poultry workers from using the bathroom. Some have been forced to urinate on themselves, while their supervisors mock them. About half of US employees have been subject to suspicionless drug screening by their employers. Millions are pressured by their employers to support particular political causes or candidates.
Clearly there are instances of undignified treatment here (e.g. being forced to urinate on yourself); and clearly there are interferences with autonomy and exercises of arbitrary authority (e.g. suspicionless searches and policing of speech). More generally, we can be confident that many employees are subject to harsh and undignified treatment in their workplaces. They are viewed as cogs in a corporate machine, given limited breaks, subject to debilitating shift work, given near-impossible targets, forced to work in unsafe and unhealthy conditions, and generally not treated as full human beings but as thing-like inputs into a corporate process. Add to this the increased surveillance and control to which we are subjected, particularly in the modern era of precarious and ‘gig’ employment, where there are no traditional employers to be held to account, and we have a recipe for considerable exercise of arbitrary authority. In this environment, we constantly need to ‘prove’ our employability to others, and so ‘voluntarily’ subject ourselves to ever-more intrusive forms of control (see my earlier post on quantified work and the unraveling problem for a discussion of this).
This gives some prima facie weight to Anderson’s initial argument. What are the potential objections to it?
2. Objections to the Argument
Anderson deals with several objections in the book. Indeed, one of the nice features of the book is that it contains a series of responses to her main case, penned by different academics, some of whom are deeply sceptical of the thesis. The most sceptical stance is probably taken by the economist Tyler Cowen, to whom Anderson responds with vigour. I’ll look at five major objections in what follows, including two from Cowen.
First, a simple definitional objection: you cannot have a ‘private’ government. The term is an oxymoron: government is, by necessity, public. This isn’t a very powerful objection. Even if it were true, by definitional fiat, that government is necessarily public, it wouldn’t mean that Anderson’s concerns about the unjust conditions of employment are unwarranted. But, as it happens, Anderson doesn’t think it is very persuasive as a definitional point anyway. She contends, reasonably in my view, that government exists wherever there is a relationship of authority, and this clearly exists in the private context. She also argues that the ‘public’ nature of the state is, in any case, a historically recent phenomenon. Historically, most forms of government were ‘private’ (think: feudal lords, husbands, and paterfamilias) and the state was not particularly ‘public’ (being something open to relatively few to participate in and criticise). One of the subsidiary goals of Anderson’s book is to recapture the historical view of government and challenge the public/private distinction that has arisen more recently. This is not to deny differences between public and private forms of government — employers do not (yet) have a monopoly on the use of force, for example — but these differences do not negate the reality of the governmental powers that they exercise.
Second, there is an objection from the heart of mainstream economic theory. Anderson thinks that economists should be disturbed by both the scale and scope of private government within corporations. Why are they allowed to get away with what they do? Why does the ‘market’ permit this? But economists think they have an answer: the theory of the firm. This was a theory developed in the early part of the 20th century, largely through the work of Ronald Coase. It explained why corporations exist. From the perspective of classical economics, the existence of large corporate enterprises with complex internal governance structures (i.e. ‘firms’) is something of a puzzle. Why couldn’t we just have individuals offering goods and services to one another through a system of contracts? The answer is partly to do with economies of scale (there are economic advantages to bigness when it comes to producing more for less) and partly to do with ‘transaction costs’ (it can be excessively costly to get private individuals together to negotiate the contracts needed to supply all the relevant goods and services). Firms are more economically efficient ways of doing business on a large scale and the authority they obtain is justified by the advantages of this efficiency. Anderson, as you might imagine, is sceptical. For one thing she doesn’t think that efficiency should be afforded that much normative weight. For another, she thinks that while the theory of the firm might explain a certain amount of authority, it does not explain as much authority as we currently see. What economic efficiencies are served by companies controlling bathroom breaks and getting employees to advocate on behalf of certain political parties? The theory of the firm doesn’t seem to answer these questions and yet it provides an ideological shield for not asking them.
Third, there is the claim that the injustice of the employment contract is irrelevant because there are good ‘exit’ options from employment contracts. That is to say, people can easily escape from the unjust conditions of employment by quitting their jobs and finding other forms of employment. Anderson is unpersuaded by this. The mere fact that you can exit a system of government does not entail that the system is just. If this were the case, we would think that the possibility of emigration/exile negated the injustice of public government. Furthermore, the exit options from employment are not always that good. People are often desperate for work, and may not have other jobs to which they can easily migrate. Employers can often restrict their ability to find new forms of work by adding ‘restraint of trade’ or ‘non-compete’ clauses to their employment contracts, or by refusing to provide good references to future employers. The possibility of such future interference with employment deepens the domination to which workers are subjected: they are forever trying to stay on their employer’s good side.
Fourth, there is one of the arguments made by Tyler Cowen in his contribution to the book. He doubts that the situation of workers is as bad as Anderson makes it out to be. Anderson really takes Cowen to task for this in the book, arguing that it is an easy thing for a well-paid college professor like him to say. Anderson is possibly a little bit unfair in this regard since, as I read it, Cowen’s claim that things aren’t that bad is a relatively minor part of his argument (his major objection is the one discussed below). Nevertheless, taking Cowen to task for this claim does give Anderson the opportunity to bolster her case. She provides many more examples of how bad things are for many workers. I can’t go through all the examples here, but a quick sample includes (Anderson 2016, pp 135-137):
The high levels (90%) of sexual harassment and assault experienced by restaurant workers.
The continuing struggle for the right to use the bathroom at work in the US.
The high numbers of injuries and deaths in the workplace reported each year, coupled with the fact that these figures are likely to understate the problem due to the fact that people conceal illness and injury for fear of losing their jobs or being subject to disciplinary action.
The fact that 41% of all workers in the US are subject to unpredictable work schedules, including short notice on-call work and disruptive split-shift work.
The routine abuse and mistreatment of immigrant workers, who are often completely beholden to their employers due to fears of deportation.
Examples like this could proliferate, but one source, that Anderson recommends, is Barbara Ehrenreich’s 1990s expose of working conditions in various low-paid industries in the US. Similarly one could recommend Studs Terkel's classic book Working. Both are somewhat dated though, so if anyone has more recent recommendations I would be happy to hear about them.
Fifth, and finally, there is Cowen’s main objection to Anderson’s argument, namely: that we should let the market sort out the problems that she identifies. The idea behind this objection is that even if Anderson is right to think that employment conditions are not ideal we should not assume that the state (or some other entity) is best placed to sort out these problems. He thinks there are good reasons to believe that ‘the market’ is the best tool for keeping employers honest. This is because (i) the competition for talent and reputation will make employers treat employees with respect and dignity; (ii) if there are intrusions into privacy, employers will be forced by the market to compensate workers adequately for these intrusions through higher pay and perks; and (iii) the cost of exit (again) remains low enough to redress any imbalance (although Cowen thinks it could be lower).
There is some theoretical plausibility to what Cowen says. Under certain conditions we could expect the market to sort out these problems, but those conditions aren’t met in practice. Although some labour markets are highly competitive and give potential employees a lot of bargaining power (e.g. markets for ‘superstar’ traders or academics) they are far from the norm. For many kinds of work, particularly low-paid manual work, there is a significant labour glut, partly induced by a rise in automation. There are lots of willing and able workers. This means that there is no market incentive for employers to treat employees with respect and dignity. Employers can easily find alternatives if someone starts kicking up a fuss. This labour glut creates problems when it comes to employees trying to improve their compensation and finding alternative employment. Employers hold all the good cards: they get to dictate the conditions of employment. This is reflected in the terrible working conditions in many sectors of low-paid work. It’s somewhat surprising that Cowen overlooks this dynamic, given that it is central to the argument in his 2013 book Average is Over.
3. Concluding Thoughts
That’s Anderson’s argument in a nutshell. I have to say I find it intriguing. I think much of what she says is correct. Employment conditions are often terrible for many people, and it is interesting that we don’t view them as a form of unjust private government. There is, consequently, real value in the critical-historical perspective that Anderson provides, particularly in her attempt recapture the idea of ‘private government’. Nevertheless, I have two concerns.
The first is relatively minor. It is simply that I think Anderson’s argument tends to assume a mid-20th century model of employment in much of her critique. That is to say, she seems to assume that most people are employed by large, well-integrated firms with a lot of bargaining power. But the era of such large-scale firms is past. The workplace is now increasingly ‘fissured’, with an increasing number of people working as short-term contractors or temps, within stripped-down corporate enterprises. This new reality of work doesn’t have to undermine her argument; if anything it could strengthen it. As David Weil points out in his book The Fissured Workplace, employment conditions under this new reality are often much worse than they were in the mid-20th century. I hinted at this point above when I referred to the experience of ‘gig’ workers in the new economy. That said, this new reality does put pressure on the claim that workers are subject to unjust ‘government’ by specific employers. If anything they are subject to unjust government by a more abstract and dispersed market forces. This is a much more insidious reality because there is no obvious person or organisation to blame or hold to account for the injustices visited upon workers.
The second criticism could be a little more substantive. It is that Anderson’s argument is a little bit US-centric (a point she readily concedes). Her claims about the injustice of employment contracts are likely to be much stronger in the US where at-will employment is common, and the number of legal protections for workers is relatively scant. In countries with more robust employee rights, things are a bit better. For example, Ireland has a fairly decent ‘unfair dismissal’ and ‘discriminatory dimissal’ laws that force employers to adopt fairly stringent procedural safeguards when dismissing an employee. That said, even jurisdictions with more robust employment laws are subject to the broader economic trends toward automation and gig work. In these jurisdictions, there is often a gap between what the law and what employers are actually do. The downward pressure on employment, and the labour glut, creates an environment in which employers are willing to flout the rules and in which many workers are no longer able to avail of robust protections because they are not legally classified as ‘employees’.