Capitalism, Class Conflict, and Domination

Abstract: Capitalists have a great deal of power over workers. However, because they are forced by market pressures to maximize profits, capitalists frequently must use their power to undermine rather than track the interests of workers. It follows that capitalists do not merely have power over workers; they have arbitrary power over them. Capitalists therefore dominate workers. This result holds even across ‘friendlier’ versions of capitalism, including those that feature robust labor market competition and unconditional income support.

1. Introduction

Capitalism’s defenders have long praised it for its freedom. It is, they say, a system of free markets, free enterprise, and free labor—a system that extends maximal “freedom to choose” (in Milton Friedman’s famous phrase) to all members of society, from the towering corporate raider to the humblest worker.1

 Nobody, after all, forces workers to work, or employers to hire, or consumers to buy. Rather, under capitalism people freely choose to do these things. So whatever else one might say about it, surely it must be admitted that capitalism scores very well in terms of freedom: it promises to replace the coercion and domination characteristic of alternative modes of production (e.g., feudalism, bureaucratic socialism) with voluntary, mutually beneficial exchange.

Socialists standardly reply that this paean to capitalist freedom obscures more than it illuminates, especially the bit about “free labor”. True, under capitalism workers are self-owners, and therefore cannot be forced to work for any capitalist in particular (or indeed, for any capitalist at all). They have, that is, the formal freedom to refuse employment. But—the socialist will insist—workers’ propertylessness renders this formal freedom empty. Lacking access to the means of production, workers have little choice but to work for some capitalist or other merely to survive. As Marx puts this point, workers “belong not to this or that capitalist but to the capitalist class”.2 But what is it to belong to another—even if that other is a group rather than an individual—if not slavery? Capitalism, socialists concede, represents an advance on feudalism and chattel slavery by eradicating personal dependence. But because it places workers at the mercy of the capitalist class, capitalism does not overthrow the basic logic of enslavement. It simply exchanges one form of enslavement for another: chattel slavery for “wage slavery”, as socialists call it. So (socialists conclude) it is profoundly misleading to describe capitalism as a paradigmatically free society. In truth, capitalism is something close to the opposite: a realm of deep of unfreedom in which the “haves” use their economic advantages to systematically dominate the “have nots”.

I believe that the standard socialist reply just rehearsed is fundamentally correct: surface appearances to the contrary, capitalists do dominate workers. My aim in this article is to convey this old socialist truth in a fresh, rigorous, and convincing way. It might be wondered why such a revamping is needed. The answer is that the socialist argument sketched above suffers from two important weaknesses. First, its central concept—domination—is often left rather obscure and under-analyzed. What is domination? Intuitively, it has something to do with subjection to power, but this can’t be the whole story, since there are many examples of subjection to power and/or dependence—e.g., the relationship of a student to her teacher, or a citizen to her democratic state—that we would not want to classify as inherently dominating.

Second, even if the domination complaint (however we cash it out) rings true with respect to the rapacious, unregulated capitalism of, say, Marx’s day, surely (defenders of capitalism will say) it applies with greatly reduced force to contemporary capitalism (at least as it presents itself in the developed world), what with its safety nets, competitive labor markets, worker-friendly regulations, and so on. In short: is it even remotely plausible to describe a person with multiple job options, a decent wage, and reasonable working conditions as a “wage slave”?

In what follows, I repair these weaknesses, thus putting the old socialist truth that capitalists dominate workers on a sturdier foundation. I repair the first weakness by drawing on recent work in political philosophy to clarify the concept of domination. I repair the second weakness by explaining the sense in which capitalists dominate workers even when workers have ample ability to exit particular firms, and indeed even when they have ample ability—thanks to unconditional income support (such as a “basic income”)—to exit employment altogether.

My plan is as follows. Section 2 introduces the concept of power and draws on recent work in political philosophy to clarify the conditions under which power turns into domination. Power becomes dominating, on the view I shall endorse, to the extent that the power-holder is not forced to track the interests of those in his grasp. Sections 3 through 5 put this account of domination to work, using it to examine the capitalist/worker relationship. In particular, these sections establish, first, that capitalists have power over workers, both within the site of production and at the level of politics, and second, that this power constitutes domination, for it is not forced to track the interests of workers. Indeed, quite to the contrary: drawing on the Marxist theory of class conflict, I show that capitalists, far from being constrained to track workers’ interests, are actually forced (by the logic of capitalist competition) to use their power to subvert workers’ interests. It follows that capitalists dominate workers. Sections 6 and 7 show that this result holds across ‘friendlier’ versions of capitalism—even those that include robust labor markets and unconditional income support.

What, then, is to be done? If capitalism is the problem, is socialism the solution? I believe so, but I cannot make that case here. My present aim is to resurrect (and improve) a traditional criticism of capitalism rather than to explore socialist alternatives. But I hope that the critical work done here might facilitate constructive efforts down the road, since we must first understand capitalism’s flaws before we can build something better.

2. Power and Domination

To introduce my argument’s central normative concept—domination—I must first discuss several varieties of power.

Social power is the ability to get one’s way in a relationship despite resistance. Or, more precisely, let’s say that A has social power over B if A can get B to do something that B would otherwise prefer not to do. Frequently, one person acquires social power over another in virtue of having some form of interference power over that other. A thug can get you to hand over your wallet (i.e., can exercise social power over you) by threatening to hit you with a club (i.e., by making you aware of his interference power, and of his intention to use it unless you toe the line). Interference can be subtler than this, however, as when A threatens to withhold something that B needs. Suppose that your health depends on a certain medicine, and that there’s only one supplier of this medicine for miles and miles around. The supplier can interfere with you, can cause you damage, by withholding the medicine. This is a kind of omission-based (rather than act-based) interference power. Yet just like the thug’s more active approach, it can ground social power: should the supplier command you to hand over your wallet, you will almost certainly comply.

Social power, then, is the ability to get people to do things they would not otherwise prefer to do, and it frequently rests on what we might call interference power—broadly, the ability to produce consequences that matter to another, whether by acting (as in the case of the thug) or by refraining from action (as in the case of the medicine supplier).3

Now, is it wrong for one party to have power—whether interference power or social power, or both—over another? Sometimes, but not always. As recent developments in political philosophy have taught us, what is morally problematic is not power per se, but arbitrary power, subjection to which constitutes a serious form of unfreedom, domination.4 And what is arbitrary power? Some philosophers say that power is arbitrary when it is not forced to track the interests of the person subject to it.5   This is the conception of arbitrary power—and thus of domination—that I shall use here.

To illustrate this conception, consider the form of power possessed by a master over a slave. The master’s interference power being nearly unconstrained, so too is his social power practically unlimited. Nothing forces the master to use his power over the slave in ways that take the slave’s interests into account. He can beat him, whip him, etc., just as he has a mind. But precisely because he can do to the slave anything he likes, he can get the slave to do virtually anything he wants. The slave, knowing this, cannot look the master in the eye; he must grovel, and act servilely, and in general try to anticipate and satisfy the master’s desires, all in order to keep him sweet and thereby hopefully stay his hand. Indeed, notice that these ill effects obtain even if the master never actually interferes with the slave. The master’s mere capacity to interfere suffices to render the slave unfree.

Yet power does not always domination make. My university’s library can fine me: this is a form of interference power. So too, it can deny me access to materials: this is another form of interference power. And because it can interfere with me in these ways, it can get me to do things I would not otherwise prefer to do, such as return library materials on time. Yet there’s nothing morally amiss with this relationship; in having these forms of power over me, the library does not dominate me.6 But why not? The basic answer is that the library’s power is constrained to track my interests, and is therefore not arbitrary. It can’t fine me without reason; nor can it deny me access to its collections on a whim. It must operate in accordance with public rules. And indeed, not only is the library’s power constrained; it is constrained in ways that force it to track the interests of its patrons. We patrons benefit from the library’s having the powers that it does. Weaken these powers, and library materials wouldn’t be returned on time, etc., thus undermining the good that the library can do for us. Moreover, if a patron feels that the library has used its powers in a way inconsistent with its brief, she can protest to some administrator or official somewhere. Anticipating this sequence of events, library staff will tend not to abuse their power in the first place. In all of these ways, then, the library’s power is kept in check, is forced to track my interests, and therefore—by the present analysis’s lights—is neither arbitrary nor (therefore) a source of domination. That my relationship with the library is freedom-preserving is reflected in the fact that I can stand tall when I enter its premises: I can look librarians in the eye, I needn’t flatter them, I needn’t conform my actions to their arbitrary wills, and so on.

To sum up, then: social power is the ability to get others to do things they would otherwise prefer not to do; this form of power frequently rests on interference power; and both types of power constitute domination insofar as they are not forced to track the interests of the person (or group) subject to them.

3. Capitalists Have Power Over Workers

Do capitalists dominate workers? They do if and only if (and to the extent that) they have arbitrary power over workers. But of course they do have such power over workers; and indeed, they have this power in great supply. Or so I will now argue. I first defend the claim that capitalists have power over workers at work. I then show that capitalists’ workplace power, being largely unconstrained to track worker’s interests, is arbitrary, hence dominating (section 4). I then show (in section 5) that capitalists also have power over us politically, and that this political power—being no more constrained to track our interests than is capitalist’s workplace power—is arbitrary, hence dominating.

I start by clarifying terms. By capitalists, I mean those people who own sufficient means of production such that they do not need to sell their labor power to live, or to live well, and who hire others to work using these means of production. Workers, by contrast, are those people who, lacking sufficient ownership of the means of production in the above sense, must sell their labor power to buyers of labor power in order to live or to live well. More sophisticated, finer-grained frameworks for thinking about class are, of course, available, but for our purposes this simple, two-class model will suffice.7

Can it be doubted that capitalists have an awesome degree of social power over workers? Simply consider the many things that capitalists are able to get those human beings in their employ to do:

  1. show up to their jobs;
  2. toil, often long and hard, under hierarchical command;
  3. follow labor process that—being designed for profit rather than producers’ interests—are frequently routine, simple, and stultifying, and sometimes hazardous;
  4. absorb abuses and indignities (some petty, others consequential) from customers, from managers, from co-workers;
  5. pour their time and energy into creating products that usually mean nothing to them, products that they may neither understand nor value, products whose immediate aim is not the satisfaction of human needs but rather the creation of profits for the capitalist;
  6. alienate these products, handing them over to the capitalist, thereby forfeiting any control over the use to which these products (or the profits they enable) are put;
  7. accept, in return for their labors, less than the value of the products they create (otherwise capitalists would not hire them).

In all of these ways, capitalists wield social power over workers, directing their actions, subordinating them, getting them to do things they would not otherwise prefer to do.

And all this without the threat of force! For the capitalist is no thug; he (typically) doesn’t resort to violence to achieve his aims.8 How, then, do capitalists get us to conform, in all the ways just canvassed, to their wills? It would appear that capital acquires its social power over labor through a straight deal, a fair exchange among equals. As David Harvey observes,

The remarkable thing is that capitalism does not appear to rely on cheating, theft, robbery, or dispossession…This ‘fairness’ rests on the conceit that laborers have an individualized private property right over the labor power they are capable of furnishing to capital as a commodity and that they are ‘free’ to dispose of that labor power to whomever they like.9

But if capitalists and workers are free equals—if capitalists do not have any legally-enforceable right to the labor power of workers—how does it happen that workers are invariably subordinated to capitalists, rather than the other way around? The answer, in a word, is property (or the lack thereof). As G.A. Cohen explains, the worker’s “subordination ensues because, lacking means of production, he can ensure his survival only by contracting with a capitalist whose bargaining position enables him to impose terms which effect the worker’s subordination.”10 Capital’s asymmetric social power is founded, in short, on its unequal control of the resources without which workers cannot live or live well. Or, to put this more plainly: we do what capitalists say because they own the things we need.

4. Class Conflict and Capitalist Domination in the Workplace

But do capitalists dominate workers? Yes, for capitalists not only have power over workers, they have arbitrary power over workers—power that is, in many respects, not forced to track workers’ interests. Indeed, if anything capitalists are forced by capitalism’s “laws of motion” to wield their power over workers in ways deleterious to workers’ interests. Or so I will argue in this section, drawing on the Marxist idea of class conflict.

That the interests of workers and capitalists conflict in myriad ways is a central Marxist thesis, but one needn’t be a Marxist to appreciate its truth. Within the capitalist firm—that “hidden abode of production”—workers and owners (and their surrogates, such as managers and supervisors) are at loggerheads over various issues swirling around three central conflicts: first, operational conflicts over how to run the firm on a day-to-day basis; second, strategic conflicts over what to do with the surplus created by the firm’s activity; and third, interpersonal, largely non-economic conflicts over how to organize human relations between owners and their surrogates, on the one hand, and workers on the other.

Consider, first, operational issues. Capitalist firms aim, of course, to make money, but to make money they must produce goods or services that people will buy. And to do that, they must combine their workers’ labor power with other means of production—tools, equipment, and so on. However, this general recipe—‘add workers to means of production, yielding commodities’—leaves almost everything up for grabs. In particular, the firm needs to determine what workers will produce, how they will produce it (that is: using which labor processes, at what levels of intensity and surveillance, under what sorts of working conditions, etc.), and in return for what sorts of compensation, pecuniary or otherwise. I call these “operational” issues because they concern the day-to-day activity of the firm.

In addition to these operational issues, firms face larger “strategic” questions as well. After selling its wares and paying its costs of production, the firm may find itself with a surplus. What should it do with this surplus? Again, the general recipe is clear—it must use this surplus in profit-maximizing ways—but the concrete implications of this recipe are not. The firm has to decide, for instance, how much of the surplus to consume (and by which of its members), and how much to reinvest. And of that portion that it reinvests, it must decide the specifics of this reinvestment. For instance, should it expand existing lines of production, or branch out into new ones? Insofar as it does expand, where should it expand—locally, or should it seek greener pastures elsewhere? Moreover, what kinds of jobs, if any, will it create as it expands? Will it hire skilled, well-paid workers, or cheaper workers with lower skills? Or perhaps it will strive to dispense with workers altogether—swapping expensive humans with cheaper technology?11 And so on.

Now, there are three important things to appreciate about all of these various operational and strategic issues.

First, capitalists decide. Generally speaking, it is capitalists who have the legal and effective authority to resolve the operational and strategic conflicts described above. This is not to say that capitalists always get their way; workers are not wholly without influence over capitalists, especially those workers whose skills are in high demand. The balance of power between, say, Google and its software engineers is much different than that between Google and its janitors. And the bargaining position of all workers, even those with low skills, is improved when unions are strong and labor markets are tight. The class struggle can be somewhat fluid, with capital’s dominance ebbing and flowing as broader social, political, and economic conditions change. Furthermore, even when workers are at their weakest relative to capital, bosses may still voluntarily solicit worker input, or at least make a show of doing so.

But none of this threatens my argument. All I am claiming is that when it comes to the “what” and “how” of production, capitalists generally decide; it is generally they (or their agents) who determine what to produce, under what conditions, using what mix of inputs, at what rates of pay, what to do with the surplus, etc. This claim, which seems to me uncontroversial, is fully compatible with all of the qualifications just noted.

Second, capitalists and workers have conflicting interests across these issues. The set of solutions or policies that advance the interests of workers will generally not overlap with the set of solutions or policies that advance the interests of capitalists. Thus, capitalists want lower wages; workers want higher wages. Capitalists want to maximize work intensity and effort; workers do not. Capitalists want to choose the profit-maximizing labor process, even if (as is frequently the case) it’s stultifying and alienating for the worker; workers want labor processes that incorporate a variety of engaging and autonomous tasks.12 Capitalists have an interest in choosing that level of monitoring and surveillance—at work and away from it—that maximizes work intensity, and thus profits; workers have an interest in being free from surveillance (whether extreme or mild). Capitalists, but not their workers, have an interest in spending part of the surplus on mansions, yachts, private jets, and so on. Capitalists, but not their workers, have an interest in producing using the lowest-cost techniques, even if this means moving production to India, slashing staffing levels, dumping waste in workers’ communities, etc. So too, capitalists, but not their workers, have an interest in producing whatever will fetch the largest profits, even if this means ignoring urgent, unmet human needs. These examples could be multiplied, but the basic point should be clear: class conflict is real, it is deep, and it is broad—and one needn’t be a Marxist to agree.

Third, this conflict between workers and capitalists is structural, rooted in the nature of capitalism itself. Capitalists do not minimize wages or maximize work intensity, relocate to distant shores or replace workers with robots (etc.) because they are greedy or mean. They do these things—and indeed, must do these things—because to fail to do them is, in the long run, to go out of business. Capitalists who do not resolve key operational and strategic conflicts in profit-maximizing ways do not long remain capitalists. In an important sense, then, individual capitalists lack real discretion over these issues: they must economize on their wage bill, increase work intensity, and so on and so forth, on pain of being driven from the field—only to be replaced by a different capitalist more willing and able to do what the ‘laws of capital’ command.13

Consider, next, a third area of conflict within the firm: largely non-economic—hence, largely optional—conflicts over interpersonal relations. Capitalists can use their power over workers for ends unrelated to that power’s underlying economic rationale. Thus bosses can harass employees—sexually, emotionally, physically; they can bully workers into disclosing their political preferences, into attending political rallies of the bosses’ choosing, into contributing to favored political causes; they can force employees to dress a certain way, to speak a certain way, to use certain modes of address but not others, to pick up their dry-cleaning, and so on. These examples are familiar and by no means exhaustive. What unites them is that in each case, we see bosses interfering with and controlling workers in ways totally disconnected from the immediate economic functions of the firm. The capitalist who harasses his employees, or forces them to contribute to his favored super-PAC, etc., cannot point to the laws of capital and say: ‘capitalism made me do it!’ In this, he is quite unlike the capitalist who, say, drives down wages or ramps up work intensity. This opens an important possibility. If this form of abuse is not integral to capitalism’s workings, perhaps capitalism can be cleansed of it. We shall return to this possibility below, in section 7.

5. Class Conflict and Capitalist Domination in Politics

Thus far I have identified three conflicts—operational, strategic, and non-economic—that mark intra-firm relations between specific capitalists and their workers. There is another site of worker/capitalist conflict, however, that we must register: this is the political conflict between the two classes. This conflict is global rather than local, and large scale rather than small scale; it does not pit this particular capitalist against this particular group of workers, but rather capitalists as a class versus workers as a class.

The central issue at stake in this political conflict concerns the country’s laws, policies, and regulations: and in particular, to what extent will these ‘rules of the game’ favor one class over the other? For example, will there be regulations concerning wages, working conditions, working hours, etc., and if so, how “worker-friendly” will these be? Will workers have the right to form unions, or to strike? Will capitalists and their businesses (and bequests, and mansions, and yachts, etc.) face heavy rates of taxation? Will there be any controls on capital flows across borders, or will capitalists be able to move their money wherever they want, whenever they want? What sorts of environmental policies and regulations will constrain economic activity? What degree of economic security, if any, should workers be granted? Will there be a safety net, and if so, will access to it be conditional on working or trying to find work—or will there be some form of unconditional income support? And so on and so forth: across these and many other political issues fateful for the fortunes of capitalists and workers alike, the two warring classes face off, revealing the depth and breadth of the antagonism between them.

Regarding workplace conflicts between capitalists and workers—conflicts that arise within the site of production itself—I stressed three points. First, capitalists usually have sufficient social power (rooted ultimately in their considerable omission-power) to get their way. Second, this is bad for workers, since the interests of capitalists and workers are at odds. Third, the ‘laws of capital’ themselves are largely (although not completely) responsible for producing the antagonism between capitalists and their workers: capitalists do not fire workers, drive down wages, etc., out of spite, but rather because these things are required by the logic of capitalist competition. To what extent do these same three points apply to the political conflicts just canvassed?

Clearly, the second point remains applicable. The set of policies favorable to capitalists will be largely disjoint from the set of policies favorable to workers: capitalists, but not their workers, will want lower taxes on the rich, fewer regulations, no controls on the movement of capital, weaker unions, and so on. The third point also remains valid. Capitalists press for, say, laxer workplace regulations, more porous safety nets, lower taxes, etc., not (solely) from personal motives, but out of considerations of global competitiveness; when it comes to political preferences, their hands are—to some extent anyway—tied by the objective requirements of profitability. But does the first point apply? Do capitalists have the power to get their way, not just in the realm of production, but in the realm of politics as well?

At first glance, you might not think so. Capitalists are few, workers are many. Surely it follows that workers are in the driver’s seat, politically speaking: they have the numbers, hence, the votes, hence, the influence to secure policies favorable to their interests.

Were this logic sound, democracies the world over would be awash in social-democratic or perhaps even socialist policies and institutions. The reality, of course, is much different. Despite enjoying numerical superiority, workers cannot match the political sway held by their less-numerous capitalist antagonists. This is largely because capitalists control something more important than votes: namely, the bulk of society’s wealth. This matters for a variety of familiar reasons. When campaigns are privately funded, as they are in the United States, wealth equals influence, and the Sheldon Adelsons of the world get to decide which candidates are viable contenders. This makes it very difficult for radical parties to gain power via electoral means. And of course wealth also buys influence in other ways—through lobbying, through control of the “consciousness industry” (education, mass media, think tanks, etc.), and so on. But suppose that somehow these familiar obstacles were surmounted, and socialists or social democrats—riding a swell of popular support, a torrent of $50 donations—grasped the reins of government. Would they be able to implement their progressive agenda?

Formally, yes, but in reality, their ability to pursue leftist ends within a capitalist framework would be sharply curtailed by the threat of capital flight. Precisely because capitalists control the bulk of the social surplus—precisely because finance and investment are largely privatized under capitalism—capital can say to government: “Sure, you can pass your radical agenda. But if you do, we’ll take our investment funds elsewhere.” In which case the economy will crash, people will sour on leftist policies, and the leftist government will find itself booted from power at the next opportunity—only to be replaced by politicians eager to restore conditions more favorable to the interests of business. Anticipating this chain of events, leftist governments will tack to the center from the start. As David Schweickart concludes, “Clearly, as long as investment decisions remain in private hands, governments that want to survive—which is to say, all governments—have little choice but to cater to the sensibilities of the capitalist class.”14 Not for nothing did Charles Lindblom describe the market system as a kind of “prison,” an anti-democratic institution that confines political possibilities to a narrow range of options agreeable to capitalists.15

6. Taking Stock

Allow me to sum up our results thus far. Workers, I have pointed out, rely on capitalists for the wages they need to live, or to live well, while the political community as a whole relies on capitalists for the finance and investment it needs to prosper. But as we saw in sections 2 and 3, dependence breeds subordination: just as you must bend to the will of the pharmacist who has unilateral control over the medicine you need to survive, so too must workers—and indeed, even governments—cater to the whims of the class that has unilateral control over the means of production without which life, let alone prosperity, is possible. Across a wide range of issues, then, within both the realm of production and the realm of politics, capitalists have the power to get their way. As discussed in detail in the previous two sections, it is (generally) they who decide how to run the firm; it is (generally) they who determine what to do with the social surplus; it is even (generally) they who control the broad parameters of political possibility, despite their numerical inferiority vis a vis workers.

But this is highly unfortunate for workers. For the interests of capitalists and workers are opposed. We thus have a situation in which:

  1. Capitalists have a great deal of power over workers—both within the workplace and at the level of politics.
  1. Capitalism’s ‘laws of motion’ force capitalists to use this economic and political power to further their own class interests (e.g., in profitability and in reproducing the political, social, and economic preconditions thereof).


  1. The class interests of capitalists conflict with the class interests of workers (in all the ways described in sections 4 and 5).

It therefore follows that

  1. Capitalists, in using their power over workers to pursue their own interests, are at the same time subverting the interests of workers.

Notice that if A can use his power to subvert B’s interests, it follows (a fortiori) that A’s power is not forced to track B’s interests. Hence, 4 implies the weaker conclusion that

  1. Capitalists’ power over workers is not forced to track worker’s interests.

However, as we saw in section 2, the best available account of domination holds that

  1. If A has power over B, and this power is not forced to track B’s interests, then A has arbitrary power over B; which is to say that A dominates

Put 5 and 6 together, and we arrive at the following conclusion:

  1. Capitalists dominate workers.

And indeed, we have seen that capitalists dominate workers in two senses and at two sites. First, within the firm, individual capitalists dominate workers in their employ; second, at the level of society as a whole, capitalists as a class dominate workers as a class, insofar as they are able to use their economic power (especially their control over finance, investment, and the “consciousness industry”) to secure political results favorable to their interests (hence, opposed to those of workers).

7. Competition?

Might labor market competition come to capitalism’s rescue here? Philip Pettit suggests that “in a well functioning labor market, no one would depend on any particular master, and so no one would be at the mercy of a master…[Any given worker] could move on to employment elsewhere in the event of suffering arbitrary interference.”16 Walmart, the thought goes, won’t abuse its workers if it knows that the Target down the street is hiring. Or, to put the point in more general terms, when workers have viable exit options they are less dependent on their current employer, greatly reducing—and perhaps even eliminating—that employer’s power over them.

There is some truth to this line of argument, but we must be careful not to overstate that truth—as Robert Taylor does, I think, when he writes that “under perfect competition, economic power is not so much dispersed as extinguished”.17 Labor market competition, whether perfect or imperfect, does not extinguish the power of capitalists over workers; rather, it defines the limits within which that power operates, as I will now explain.

To begin, we must drill a bit deeper into the concept of dependence. According to Frank Lovett’s illuminating account, a person B’s degree of dependence on another person A varies inversely with the attractiveness of B’s alternatives to dealing with A.18  That is to say: as B’s alternatives to dealing with A improve, his dependence wanes; as they worsen, his dependence grows. And this matches intuition. Say you are the only employer in town, and I am a worker who must sell his labor power to live, or to live well. My alternative to dealing with you is therefore starvation or perhaps life on the dole—either way, my alternatives are bleak. Lovett’s analysis therefore suggests (correctly, I submit) that my level of dependence on you is going to be very high. So too, then, will your degree of social power over me: precisely because I depend on you to such a great degree, you will be able to get me to do a wide range of things I would not otherwise prefer to do.

Now suppose that another buyer of labor power arrives on the scene. Suddenly my best alternative to dealing with you looks more tenable. And so my overall level of dependence on you won’t be very high. Indeed, if the new employer’s terms are similar to yours, I might not depend on you at all: maybe I am indifferent between working for you and working for your competitor. Having a viable alternative to dealing with you greatly reduces and perhaps even dissolves my dependence on you. And if I don’t depend on you, what happens to your social power over me? Does it vanish?

No, not necessarily, and it is absolutely crucial to see why not. Competition for my labor has merely shifted the locus of my dependence, not eliminated it. Before, I was dependent on you individually; now, I am dependent on a group that includes you and one other capitalist. So I remain highly dependent; it’s just that now I’m dependent on a group rather than on any individual in particular. (As Marx would put it, I “belong not to this or that capitalist but to the capitalist class.”19)

Now, what kind of power over me does my dependence on the group give the individuals comprising that group? Considerable power indeed. In a principle:

Group dependence: If B depends on the group of As, in the sense that B must contract with some A or other in order to get what he/she needs to live, or to live well, then what each A can get B to do (or can do to B) depends on what (B thinks) other As would try to get B to do and/or do to B.

To illustrate, suppose I ask you for your terms. “In exchange for wages, I expect you to work for 8 hours a day, obey my instructions while at the worksite, and hand over the fruits of your toil,” you say. Not wanting to perform any of these actions, I check with the new employer—and find identical demands. “Well, I suppose it can’t be helped,” I sigh, holding my nose as I report for work. Despite the fact that I depend on neither capitalist individually, there I am, toiling for 8 hours, obeying orders, alienating the product of my labor—in short, bending to a capitalist’s will. Exit has, in this case, utterly failed to inoculate me against the power of capital.

Which is not to say that exit is always so impotent. Suppose that, prior to the new capitalist’s arrival, you were able to get me to do something x in exchange for wages—something truly beyond the pale. (Perhaps you could get me to submit to harassment, or to work for inhuman hours or at an inhuman pace, etc.) But then the new capitalist comes on the scene, and I find (much to my relief) that (for whatever reason) she will not require me to do x in exchange for wages; instead, ysomething I hardly relish, but which I deem preferable to x—will do. “Do x,” you say. “Not a chance,” I reply, heading over to the new capitalist. Exit has, in this case, constrained your power over me: before, you could get me to do x, but now you can’t.

Two lessons emerge. First, labor market competition does indeed constrain capitalists, forcing them to offer terms that are no worse than those offered by any of their competitors. Individual capitalists cannot impose demands outside the group norm. And so we should expect to see certain egregious and/or idiosyncratic abuses—especially those unconnected with the economic purpose of the firm, such as sexual harassment or unnecessarily oppressive working conditions—dissipate in the face of robust labor market competition. This is no small thing, and it underscores the (limited) moral importance of competition for labor. But, secondly and crucially, while individual capitalists cannot impose demands outside the group norm, they can impose demands inside that norm. When capitalists speak in one voice, workers must obey.

But do capitalists speak in one voice? Across a wide range of issues, yes, of course they do. For they are, recall, themselves beholden to the laws of capital—forced, by the logic of capitalist competition, to maximize profits. In pursuit of this goal, certain demands are simply unavoidable. Let labor market conditions be as competitive as you please, capitalists will still require workers to show up for work. They will still require workers to toil at a profit-maximizing pace, under profit-maximizing conditions, for a profit-maximizing wage. They will still force workers to lay down their tools at the end of the day and hand over the surplus created by their labors. And they will still reserve for themselves the right to decide what to do with this surplus. These demands are non-negotiable; they are the sine qua non of capitalism. Capitalists cannot relinquish them without ceasing to be capitalists.

In sum, earlier sections showed that capitalists have considerable power over workers, and that this power, being arbitrary, amounts to domination. In reply, a defender of capitalism might draw on remarks by Pettit, Taylor, and others, which purport to show that employment relationships do not foster domination provided that workers are able to exit these relationships. In a slogan, where exit is possible, domination is not—or so this reply would suggest. But as we can now appreciate, exit’s domination-fighting properties have been overblown. Just as a wave can wash away trash on the beach, and perhaps even some sand, but not the bedrock on which the trash and sand rest, so too can labor market competition cleanse capitalism of some of its more inessential abuses—but not those forms of worker-subordination, of capitalist-power-overing, that constitute its basic bedrock. Competition, in short, can mitigate capitalist domination, but it cannot eliminate it.

One final point before moving on. Recall from our earlier argument that capitalists dominate workers in two ways, at two levels: first, at the workplace, second, at the level of politics. Notice that the “exit” argument considered in this section addresses only the first of these forms of domination. Labor market competition by itself does absolutely nothing to mitigate the political power of capitalists over workers. And indeed, we should expect to see—and in fact do see—capitalists using their political power to install and preserve anti-competitive labor market conditions wherever possible (unemployment, collusion, and so on): which is to say, we see capitalists using their global dominating power to entrench their local dominating power.

8. Basic Income?

So long as people need to sell their labor power to live or live well, they are at the mercy of capitalists, even if there are many capitalists to choose from. Perhaps, then, the thing to do is eliminate the need to sell one’s labor power. This is the basic rationale behind an unconditional basic income (UBI), a proposal that has become quite popular amongst philosophers concerned with domination.20 Like traditional welfare programs, a UBI partly decouples income from market earnings, but unlike traditional welfare, it does this in a no-strings-attached fashion: everyone in the target population gets it, regardless of income level or employment status. Depending on the level at which it is set, a UBI’s likely effects—on the economy; on individual independence; on social relations, etc.—vary dramatically. For our purposes, suppose that we set the UBI at the highest sustainable level (as philosophers like Philip Van Parijs propose),21 and that this level is sufficient to enable each person to achieve a “culturally defined respectable standard of living, say 125 percent of the poverty line”22 —which in the United States would amount to around $15,000 per year per person. With such a UBI in place, would capitalists still dominate workers?

Yes, they would. True, no longer would workers be forced to sell their labor power in order to live, or even to live decently. But to live decently is not necessarily to live well. To do the latter, workers will still generally need to sell their labor power—and so people will still generally be dependent on the capitalist class, and thus dominated by their particular employers (as per our earlier argument).

But why will people need wage labor in order to live well, despite receiving a UBI? There are three main reasons. First, most people are at least partly consumerist in orientation—hardly a surprise, given the enormous sums spent by capitalists and governments to encourage materialism and consumerist values.23 For people in the grip of such values—which is to say, for nearly all of us—living well requires a level of consumption totally unattainable on any feasible UBI. (As a quick test, ask yourself if you could live well, as you define that target, on $15,000 per year.) The problem is exacerbated by the “positionality” of most people’s consumption desires: many of us want not merely to have lots of stuff, but to have more stuff than others in our “consumption reference group,” as sociologists call it. This is obviously impossible on a UBI, which provides equal purchasing power to all.

Second, most people wish not merely to consume, but also to contribute in a socially-recognized way. No one wants to be regarded as a free-rider or a parasite. To be sure, there are many ways to contribute besides having a job: one needn’t produce “exchange value” for a capitalist in order to produce “use value” for other human beings. Yet in our culture, shaped as it is by capitalist norms, the dominant socially-recognized form that contribution takes is paid employment. “What do you do?” is a standard opening move in conversation, and—as norms are at present—“I live on the UBI and pick up trash in the local park” is unlikely to be met with much respect. Insofar as people wish to be, and to be seen as, social contributors in good standing, then, they will face tremendous pressure to get a job. A UBI does nothing to alleviate this pressure.

Finally, many people want to develop and deploy their productive capacities, whether for the sake of contributing to the good of others or simply for personal satisfaction. Yet in many cases, production of the desired kind requires access to resources monopolized by capitalists. Say you have some carpentry skill, and you wish to build things. Does your UBI enable you to satisfy your productive aspirations without entanglement in capitalist employment? Unlikely. You’ll need tools, materials, a worksite, a community of skilled workers, a large-scale project: all things that—given the present distribution of property—you can access only within the framework of wage labor. As one 19th-century critic of wage slavery puts it:

The land, the tools and materials of labor are still the exclusive property of the privileged few, and the worker cannot produce without giving himself a boss or master. It must not be supposed that the proclamation of emancipation liberated mankind from slavery. The most odious, because the most subtle form of slavery—wages slavery—remains to be abolished.24

And my present point is that wage slavery would still remain to be abolished even with a UBI, for even with such a policy in place, many people will find themselves utterly dependent on capitalists for the income, opportunities for social contribution, and opportunities for productive activity that they need to flourish.

The emancipatory potential of a UBI, it would seem, has been widely exaggerated. A UBI loosens capital’s grip on workers but it does not eliminate that grip. UBI or no, capital retains its monopoly on societal wealth, and thus remains in control. Indeed, not just at the worksite, as I have been discussing thus far in this section, but also in the halls of government: a UBI does nothing whatsoever to free citizens from the dominating political power of capitalists, for their political domination rests (as discussed in section 5) on their vast reserves of wealth and their stranglehold on the means of productions—conditions untouched by a UBI.

9. Conclusion

In closing, allow me to summarize my core argument from a slightly different angle. Edward Bellamy, an influential 19th century American socialist and novelist, says something true and important when he writes that “if you own the things men must have, you own the men who must have them”. But capitalists, of course, own the things workers must have. Therefore capitalists own workers. Not, admittedly, in any legal sense; capitalism is not chattel slavery, and workers are legally entitled to withhold their labor power. But there are other senses of ownership besides the legal one, and other forms of slavery besides chattel slavery. If I can consistently control your actions, getting you to serve my interests in ways that undermine your interests, do I not have a kind of ownership over you? Do I not dominate you in a morally objectionable way (section 2)? Yet—as I have argued here—a clear-eyed assessment of capitalism suggests that capitalists enjoy precisely these forms of objectionable power over workers. Because capitalists have what workers need, they control workers’ actions, getting workers to serve their interests (section 3). But because the interests of workers and capitalists are opposed, capitalists are not merely using their economic advantages to get workers to do their bidding; more than this, they are getting workers to undermine their own interests in the process of serving the interests of capital (sections 4 and 5). It follows that capitalists have a morally problematic kind of ownership over workers (section 6). Despite surface appearances to the contrary, capitalism produces and maintains a kind of slavery and is therefore a system of profound unfreedom. Capitalists, in short, dominate workers—or so I have argued here.

Indeed, this domination persists even across the ‘friendliest’ forms of capitalism (sections 7 and 8). Let demand for workers be high, and exit costs for workers low; let a basic income free workers from the need to sell their labor power to live; even so, workers will remain subject to a hostile class of owners whose basic economic and political interest clash with their own; dependent, both as individuals and as citizens of a political community, on these owners for the wages, investment funds, avenues for social contribution, and productive facilities without which they cannot prosper, flourish, or fully exercise their human powers.

What, then, is to be done? Is another world possible? Can we envision a viable alternative to reformist capitalism—one that would truly eliminate economic domination rather than merely nibble at its edges? If the analysis of this paper is correct, any such alternative would have to break capital’s monopoly over the means of production, and so would have to be broadly “socialist” in orientation—but what, today, does that label even mean? These are issues well worth exploring.


1. Milton Friedman and Rose Friedman, Free to Choose (New York: Harcourt, 1990).

2. Karl Marx, “Wage Labor and Capital,” in The Marx-Engels Reader, ed. R. Tucker, 2nd ed. (New York: W.W. Norton, 1978), 203-217, at 205.

3. When A can produce consequences that matter to B, B is vulnerable to A. For further analysis of vulnerability, see Robert Goodin, Protecting the Vulnerable (Chicago: University of Chicago Press, 1986).

4. Key works in this literature include Philip Pettit, Republicanism (Oxford: Oxford University Press, 1997); Philip Pettit, On The People’s Terms (Cambridge: Cambridge University Press, 2012); Frank Lovett, A General Theory of Domination and Justice (Oxford: Oxford University Press, 2010); Lovett, “What Counts as Arbitrary Power?” Journal of Political Power 5, 1 (2012), 137-152.

5. See, e.g, Pettit, Republicanism, Chapter 2: Liberty as Non-Domination.

6. For this example, see Lovett, A General Theory of Domination and Justice.

7. See, e.g., Erik Olin Wright, Classes (London: Verso, 1985).

8. Which is not to say that capitalists have no club at their disposal. They do; it’s just held at one or two removes, and by a different party—the state. As radicals have long noted, capitalists will not hesitate to use the coercive power of the state to further their class interests when necessary (e.g., to facilitate “primitive accumulation,” to put down strikes, to suppress radical movements, to protect their fortunes, and so on).

9. David Harvey, Seventeen Contradictions and the End of Capitalism (New York: Oxford University Press, 2014), 63.

10. G.A. Cohen, Karl Marx’s Theory of History: A Defence, expanded edition (Princeton: Princeton University Press, 2000), 70.

11. It may seem like mere science fiction, but many researchers have concluded that the robots are coming for our jobs. One particularly influential study suggests that “nearly half of US jobs could be at risk of computerization over the next two decades”. Carl Benedikt Frey and Michael Osborne, “The Future of Employment: How Susceptible Are Jobs to Computerisation?” Available at

12. For the classic Marxist analysis of class conflict over the labor process, see Harry Braverman, Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century, 25th Anniversary Edition (New York: Monthly Review Press, 1998).

13. We must take care, however, not to exaggerate this point. Although capitalists must maximize profits, sometimes there are multiple plausible paths to this goal, in which case capitalists do enjoy real discretion. For instance, economic considerations might dictate layoffs, but leave open which particular workers get the sack. Or again, perhaps costs must be cut, but there are many ways to do this: salaries could be reduced, waste eliminated, executive perks rolled back, etc. Insofar as economic rationality underdetermines this choice, the capitalist has real discretion.

14. David Schweickart, After Capitalism, 2nd ed. (Lanham: Rowman & Littlefield, 2011), 157.

15. Charles Lindblom, “The Market as Prison,” The Journal of Politics, 44, 2 (May, 1982), 324-336.

16. Philip Pettit, “Freedom in the Market,” Politics, Philosophy, and Economics (June, 2006) 131-149, at 142.

17. Robert Taylor, “Market Freedom as Antipower,” American Political Science Review (August, 2013), 594-602, at 598.

18. Lovett, A General Theory of Domination & Justice, 50.

19. Karl Marx, “Wage Labor and Capital,” in The Marx-Engels Reader, ed. R. Tucker, 2nd ed. (New York: W.W. Norton, 1978), 203-217, at 205.

20. See, e.g., Philip Pettit, “A Republican Right to Basic Income?” Basic Income Studies 2, 2 (December 2007) 1-7; Lovett, A General Theory of Domination & Justice, 196-203.

21. Philip Van Parijs, Real Freedom For All (New York: Oxford University Press, 1995).

22. Erik Olin Wright, Envisioning Real Utopias (London: Verso, 2010), 217.

23. NYU professor Bertell Ollman reports that American businesses spend approximately one trillion dollars a year on advertising and public relations, and that the advertising and public relations industry produces or influences 40% of what we see, hear, and read (How 2 Take an Exam…And Change the World [Montreal: Black Rose Books, 2001], 63). Nor are children exempt from this onslaught. Corporations are increasingly infiltrating schools through, e.g., biased “classroom kits” (climate change education by Chevron), advertisements for junk food in school cafeterias, ad-drenched “BusRadio” on school buses—even report cards emblazoned with Ronald McDonald’s smiling visage. Exposure to pro-consumerist messaging on this scale cannot help but profoundly shape our conceptions of the good. See Commercial Free Childhood’s fact sheet “Marketing in Schools,” available at .

24. Cited in Alex Gourevitch, From Slavery to the Cooperative Commonwealth: Labor and Republican Liberty in the Nineteenth Century (Princeton: Princeton University Press, 2014), 108.

This entry was posted in 73, Volume 31, No. 1. Bookmark the permalink.

Leave a Reply

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