Sweatshops, Exploitation, and the Nonworseness Claim

Abstract

According to the nonworseness claim, it cannot be morally worse to exploit someone than not to interact with them at all when the interaction 1) is mutually beneficial, 2) is voluntary, and 3) has no negative effects on third parties. My aim in this article is to defend the moral significance of exploitation from this challenge. To that end, I develop a novel account of why sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage despite the fact that their relationship is entirely optional. More precisely, I defend two main claims. First, I show that sweatshop owners are morally obligated to pay sweatshop workers a nonexploitative wage even though they have a right not to hire them and even though that will require them to pay sweatshop workers a wage that is higher than the one they voluntarily accepted. Second, I explain why this obligation on the part of sweatshop owners is not defeated by the fact that other individuals not party to the transaction would benefit even more than sweatshop workers from receiving this additional level of pay.

It seems wrong to exploit people. But is that really the case? In recent years, a number of philosophers have challenged this intuition by advancing a puzzle that seeks to cast doubt on the moral significance of exploitation. Following Alan Wertheimer, call this puzzle the nonworseness claim, or NWC (Wertheimer 1996, 2011; Wertheimer and Zwolinski 2015). The puzzle, roughly speaking, is that it is difficult to see how it can be worse to exploit someone than not to interact with them at all when the interaction 1) is mutually beneficial, 2) is voluntary, and 3) has no negative effects on third parties. The NWC seeks to show, in other words, that, despite initial appearances to the contrary, there cannot be something morally wrong about a mutually beneficial and voluntary transaction between two parties if neither one of them has an obligation to enter into such a transaction in the first place.

By way of example, consider the following case: ¹

Apple and Mapple. Two American firms—Apple and Mapple—sell consumer electronics. But whereas Apple manufactures its products under sweatshop ² conditions in China, Mapple manufactures its products in highly automated factories in California. For that reason, Apple is widely condemned for paying workers only the market-clearing wage, and Mapple is widely praised for keeping production in the United States.

But as defenders of the NWC have forcefully argued, there is something deeply puzzling about these claims. If it is morally permissible for firms not to benefit people in the developing world by keeping production at home, ³ then how can it be morally wrong of them to pay sweatshop workers the market-clearing wage simply because that does not seem to benefit them enough? ⁴ Or, to put the point another way, how can it be morally worse to exploit the global poor than not to transact with them at all when exploitation is voluntary and makes them better off? For whatever else one might be tempted to say about firms that outsource production, they are at the very least doing more to help workers in developing countries than firms that do not. As Joan Robinson (1962, 46) once famously quipped, “the misery of being exploited by capitalists is nothing compared to the misery of not being exploited at all.”

My aim in this article is to defend the moral significance of exploitation from this challenge. To that end, I develop a novel account of why sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage despite the fact that their relationship is entirely optional. More precisely, I defend two main claims. First, I show that sweatshop owners are morally obligated to pay sweatshop workers a nonexploitative wage even though they have a right not to hire them and even though that will require them to pay sweatshop workers a wage that is higher than the one they voluntarily accepted. Second, I explain why this obligation on the part of sweatshop owners is not defeated by the fact that other individuals not party to the transaction would benefit even more than sweatshop workers from receiving this additional level of pay. Wertheimer calls these claims the interaction principle and the greater obligation claim, respectively. And because he believes that rejecting the NWC entails accepting both the interaction principle and the greater obligation claim (Wertheimer 2011, 256–57), my account is (to the best of my knowledge) the most comprehensive response to this puzzle in the literature to date.

More generally, this article advances the debate on the NWC in three distinct ways. First, I will not simply be arguing that, intuitively speaking, the wage that sweatshop workers are paid is too low (Faraci 2019) or that it violates our considered judgments about what it means to respect sweatshop workers (Arnold and Bowie 2003; Sample 2003; Snyder 2008, 2013). Although I largely agree with those intuitions, they are precisely what is in dispute in this debate. Appealing to them thus begs the question. By contrast, my account is based on a widely held moral principle that I will articulate below. Now, that does not mean that everyone will in fact accept this principle. But it does nevertheless shift the argumentative burden onto defenders of the NWC. If they want to reject this principle, then they must explain why that is the case.

Second, and related to the preceding point, I will not simply be claiming that the NWC is vulnerable to counterexamples (Meyers 2004; Bailey 2011) or that it ignores the existence of “special obligations,” that is, obligations that individuals have to specific people in virtue of their intimate ties or special relationships (Wertheimer 2011). For example, some have argued that the NWC should be rejected because husbands have an obligation to treat their wives fairly even though they have a right not to marry them in the first place (Wertheimer 1996, 290). And others similarly believe that parents have special obligations to their children despite the fact that they are morally permitted not to procreate (Snyder 2009a, 306). But the problem with these arguments is that it is perfectly consistent for defenders of the NWC to grant that special obligations exist in those cases and yet deny that the same holds true for the employer–employee relationship (Powell and Zwolinski 2012, 469). My account avoids this problem because it relies only on considerations internal to the employment context.

Finally, unlike other critics of the NWC, I will not be arguing that it does not apply in the present circumstances because mutually beneficial and voluntary transactions often have negative effects on third parties (Malmqvist 2017). To the contrary, I will grant the underlying premises of the NWC for the sake of argument and yet show why it should be rejected. I will assume, in other words, that the relationship between sweatshop owners and sweatshop workers 1) is mutually beneficial, 2) is voluntary, and 3) has no negative effects on third parties. But I will show that the truth of the NWC does not follow from these assumptions.

My discussion proceeds in three stages. First, I draw an analogy between the employer–employee relationship and two variations on the familiar problem of how to divide a cake. Second, I show how this analogy helps both to expose a mistake in the NWC and to explain why it is typically morally impermissible for sweatshop owners to pay sweatshop workers only the market-clearing wage. Finally, I conclude by responding to three natural objections to my argument. These objections force me to clarify 1) the role that consent plays in employment relationships, 2) whether the obligation not to exploit people can be waived, and 3) the relation between nonexploitation and other values.

Before proceeding further, one final comment is in order. For the purpose of this article, I will assume that to exploit someone is to take unfair advantage of them. This definition both accords with common usage and is ubiquitous in the philosophical literature (see, e.g., Wertheimer 1996, 6; Vrousalis 2013, 148–49; Ferguson 2016a, 7; 2016b, 951). It does, however, raise a problem. It seems to imply that defenders of the NWC believe that it is morally permissible for sweatshop owners to treat sweatshop workers unfairly or that the value of fairness is simply not important in employer–employee relationships. For example, David Faraci (2019, 177) has recently argued that one of the reasons why we should reject the NWC is that it seems to ignore the “independent moral significance [that] fairness has… within their interaction.” But that is not entirely accurate. Defenders of the NWC do not deny that fairness matters in employer–employee relationships. Rather, they believe that the market-clearing wage is fair because it is better for sweatshop workers than the available alternatives and the wage they voluntarily accepted. Indeed, given that sweatshop owners are already doing more than anyone else to help sweatshop workers escape poverty, it would be unfair to require them to pay sweatshop workers a wage that is even higher. It thus seems more charitable to assume that the disagreement between defenders and critics of the NWC is not over whether fairness matters in employer–employee relationships but over what fairness consists in. So the question is not, Are sweatshop owners morally obligated to pay sweatshop workers a fair wage? but rather, Is the value of fairness exhausted by the contractual obligations that sweatshop owners have to sweatshop workers? ⁵ My answer to the latter question is no. For reasons that I explain below, sweatshop owners typically have a moral obligation to pay sweatshop workers a wage that is higher than the one they voluntarily accepted. ⁶ Anything less would be exploitative.

1. AN ANALOGY

To motivate the argument that follows, let us begin with an analogy. Consider two variations on the familiar problem of how to divide a cake. In particular, let us suppose that, unlike in the original example, where the cake is created ex nihilo, the issue is not simply how it should be distributed but how it gets produced in the first place as well. This will enable us to focus on two distinct scenarios that are directly relevant to the problem at hand.

In the first scenario, A has the necessary equipment (oven, measuring cups, mixer, etc.) and ingredients (flour, milk, eggs, etc.) and bakes the cakes alone. ⁷ After doing so, he then asks B (out of a group of neighbors that includes C and D) if she would like a certain number of pieces of it as a gift. Here it seems that B has no legitimate grounds for complaint if A offers her a smaller share of cake than she would have hoped for, for example, less than one-half of it. Why? The reason, quite simply, is that it is his cake (in the morally relevant sense of that term at issue in this article), and so it seems to follow quite straightforwardly that A has a right to decide how to distribute it, that is, a right to decide how many pieces, if any, he is willing to give her. Indeed, the latter claim follows from the former one. I do not have a right to decide how to distribute something unless it is mine in the first place. I cannot, for example, rightfully give you something that belongs to someone else. Now, that is not to say that B is required to accept whatever A wants to give her. She can, of course, reject A’s offer and ask for a larger share of cake. The point, however, is that B has no moral entitlement to demand that she be given more of it. The matter is up to A’s discretion instead. If he wants to give her a larger share of cake, then he may do so. But because there is no restriction on his choice in the matter, it is ultimately up to him to decide how much or how little cake he is willing to offer B or anyone else.

Now contrast the first scenario with a second one in which A and B bake the cake together. Suppose, for example, that although A has the necessary equipment and ingredients, he lacks the necessary skills to bake it. And so he decides to enlist B (out of a pool of qualified candidates that includes C and D) to do so instead. Here it seems that even though A has the right to decide with whom to bake a cake, he may not decide how much cake B ultimately gets. For there are no legitimate grounds for assuming that it is A’s cake to begin with. To the contrary, because it is a joint product of their cooperative efforts, it seems hard to deny that, morally speaking, the cake is theirs.

By the same token, it would be a mistake to inquire how many pieces of cake A should offer B in exchange for her labor. That question falsely presupposes, once again, that the cake is his. Of course, that is not to deny that, as in the first scenario, both A and B are morally permitted to transfer part of the cake to someone else. Each has a right, for example, to give C or D a piece of cake as a gift. But they may only do so with that part of the cake that is rightfully their own to begin with.

Notice, then, that despite the fact that, in both scenarios, A has a right to decide with whom to interact, it is only in the first one that he has a right to decide how much the other party to the interaction may get. And this can be explained by how the good to be distributed was produced in the first place. In the first scenario, A produced the cake alone. Accordingly, it is up to him to decide how it should be distributed, that is, how much or how little he is willing to offer anyone else. In the second scenario, by contrast, A and B produced the cake together, and so each only has a right to decide how to distribute his or her own share.

Now, it might be tempting to object that because A has the necessary equipment and ingredients, he should have the right to decide how many pieces of cake B should get. After all, B would not have been able to use her skills to bake the cake if she had no equipment or ingredients. But that is incorrect. With equal force, we could say the same about A: without B’s baking skills, A could never have baked a cake with only his equipment and ingredients. Because their positions in this regard are entirely symmetrical, there is no reason to assume that either one has a right to decide how much the other should get. The upshot is that because the cake was a joint product of their cooperative efforts and would not have existed if either one of them had failed to do his or her part in baking it, each party has a claim to a certain share of it. ⁸

By “claim,” I mean something quite specific. ⁹ At the most general level of abstraction, to say that a person has a claim to something is to say that there is a reason why she, and not anyone else, should have it. For if I have a claim to a good, then I would be wronged if I did not get it. More importantly for our present purposes, if I have a claim to a good, then I have a pro tanto moral entitlement to it, independently of whether you have a desire to give it to me or whether I can successfully bargain for it instead. For I need not bargain for something that is, morally speaking, mine to begin with. To the contrary, I can only bargain for something that is not rightfully my own in the first place.

Now, it is important to emphasize that there are a number of potential explanations for why someone has a claim to a particular good as opposed to someone else. ¹⁰ Because exploring them is not an aim of this article, it should suffice to note that, intuitively speaking, one such explanation is that I produced it. That is to say, I have a pro tanto moral entitlement to a particular good if I was the one that made or created it. By the same token, if you and I produced a particular good together, then, absent special reasons to the contrary, it seems to follow quite straightforwardly that we each have a claim to it as well.

One virtue of this explanation is that it is compatible with the views of philosophers on both the Right and the Left. For example, it is entailed by the idea of self-ownership, and, as G. A. Cohen (1995, 151) explains, not only is self-ownership “the foundation of libertarianism” but “the Marxist contention that the capitalist exploits the worker depends on the proposition… of self-ownership” as well. As he puts it, “the underlying idea is that a person should be sovereign with respect to what he will do with his own energies. He should not deploy them under another person’s orders in the manner of a slave and have part or all of his product taken from him for nothing in return” (146–47). Or, in the words of A. John Simmons (1992, 223), “there has been no more widespread or enduring intuition about property rights than that labor in creating or improving a thing gives one special claim to it.” ¹¹

Turn now to the relationship between sweatshop owners and sweatshop workers and ask which of the preceding two scenarios is more directly analogous to it. In answering this question, I want to defend two minor claims. First, the relationship between sweatshop owners and sweatshop workers is more directly analogous to the second scenario, that is, the scenario in which A and B baked the cake together. Second, for reasons I will explain below, defenders of the NWC (mistakenly) assume that the more direct analogy is to the first scenario, that is, the one in which A bakes the cake alone and gives part of it to B as a gift. Let us consider these claims in reverse order.

Begin by asking, What are the grounds for believing that defenders of the NWC assume that the relationship between sweatshop owners and sweatshop workers is more directly analogous to the first scenario? Although the evidence for this claim is primarily textual in nature, it is also, I believe, necessary to make sense of the NWC’s underlying rationale or motivation. Consider, by way of example, the following quotation by Benjamin Powell and Matt Zwolinski (2012, 468–69, emphasis added): “MNEs [i.e., sweatshop owners] confer considerable benefit on their workers. But why should the very act of providing such a benefit impose on MNEs a moral obligation to confer an even greater benefit? Why does providing some help to workers in the developing world confer an obligation to help more, especially when those who provide no help are… guilty of no moral wrongdoing?”

There are two things to note about this quotation. First, it is highly representative in nature. By that I mean that it is indicative of the language that defenders of the NWC typically use to describe the relationship between sweatshop owners and sweatshop workers. Thus, elsewhere, Zwolinski (2007, 707) writes that these “firms [i.e., sweatshop owners] are doing something to help. The wages they pay to workers make those workers better off than they used to be—even if it is not as well off as we think they ought to be made. Do they have an obligation to do more? Consider the fact that most individuals do nothing to make Third-World workers better off.” And in a slightly different context, Zwolinski (2008, 357) also emphasizes how “it is puzzling how it could be worse… for [a putative exploiter] to provide some help than it is for him to provide none.” Indeed, the same can be said about other defenders of the NWC (e.g., Wertheimer 2011, 256).

Second, and more important, it is wrong to suggest that sweatshop owners are “providing” sweatshop workers with “help” or that the former are somehow “conferring” benefits on the latter. ¹² For that seems to imply that the wage a sweatshop worker receives from a sweatshop owner should be treated as if it were a gift. Sweatshop owners have a right, after all, not to start a business in the developing world, and because they are already providing sweatshop workers with some benefits, it would be a mistake to require them to provide even more. And this, of course, is structurally similar to the assumption that B has no legitimate grounds for complaint in the first scenario if A offers her a smaller share of cake than she was hoping to get.

But that assumption is incorrect. To see why, we must ask, How were the benefits of social cooperation produced in the first place? The answer, quite clearly, is that they were the joint product of the cooperative efforts of both sweatshop owners and sweatshop workers. And so, in a manner directly analogous to the second scenario, it seems to follow that each has a claim to a specific share of those benefits. Indeed, we can put this point another way by saying that if defenders of the NWC wish to deny that sweatshop workers have an independent moral entitlement to the benefits they helped create with sweatshop owners, then the burden falls on them to explain why the relationship between A and B in the first and second scenarios is fundamentally the same rather than different. They need to show, in other words, that in both cases, B is entitled only to that for which she can successfully bargain, despite the fact that she helped bake the cake in one of them and received it in the other as a gift. But why should the fact that she helped bake the cake make no difference to her moral entitlements? Why is it morally irrelevant how something was produced to begin with?

Notice, for example, that a sweatshop owner cannot respond to a demand for a higher wage ¹³ from a sweatshop worker by, in effect, saying the following: “Because you would not have been able to produce a social surplus if I had not hired you, then I should be able to make a deal with you on any terms that I am willing to accept.” This response is unavailable to the sweatshop owner because there is no logical connection between the premise that those benefits exist only because he hired the sweatshop worker and the conclusion that he has a right to decide how much or how little she should get. ¹⁴ It is equally true that had she chosen not to work for him, there would be no benefits to distribute as well. The positions of sweatshop owners and sweatshop workers are, once again, entirely symmetrical, and so neither has a greater initial claim to the joint product of their cooperative efforts. It thus seems hard to deny that the relationship between sweatshop owners and sweatshop workers is exactly the same as the relationship between A and B in the second scenario.

2. THE CASE FOR A NONEXPLOITATIVE WAGE

With these points as background, we are now in a position to explain why sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage even if doing so requires them to pay sweatshop workers a wage that is higher than the one they voluntarily accepted and individuals not party to the transaction would benefit even more than sweatshop workers from receiving this additional level of pay. ¹⁵ I begin with some general remarks about the concept of fairness. ¹⁶

Fairness is a comparative standard. It concerns, that is to say, how much A has relative to B rather than whether B has enough in an absolute sense. For A to defend himself against a charge of unfairness, therefore, it is not sufficient for him simply to point out that B is better off than she was prior to their interaction, or even that she is able to satisfy her basic needs. To the contrary, A must instead show that the terms of their interaction are not illegitimately biased in his favor. The value of fairness is thus comparative in the sense that its realization depends on what a particular person has relative to another and not just what a person has in and of herself.

But if fairness is a matter of how A fares relative to B, then how do we know whether one of them has too much or too little by way of comparison? To answer this question, we must determine why someone should have a particular good in the first place. And although there are many different types of reasons to consider, fairness is concerned exclusively with what I have called a claim. ¹⁷ Only if I have a claim to a particular good am I eligible, from the point of view of fairness, to receive it (Broome 1990). By contrast, if I have no claim to a particular good, then I have no legitimate complaint of unfairness if I do not get it. Claims thus precede fairness in terms of the order of explanation. Once a person’s claim to a particular good is established, then whether or not they are treated unfairly depends on the extent to which their claim to that good is satisfied in comparison to others.

More precisely, fairness requires that “claims should be satisfied in proportion to their strength” (Broome 1990, 95). The greater the strength of my claim is, the more am I entitled to, and vice versa. For example, if A makes a greater contribution to a cooperative endeavor than B, then he should receive a greater share of the benefits. By contrast, if A and B make the same contribution to a cooperative endeavor, then their shares should be equal. As Broome emphasizes, however, proportion should not be taken too precisely. Rather, the basic idea is that, roughly speaking, equal claims require equal satisfaction, stronger claims require more satisfaction than weaker ones, and so on. If these claims are denied, then those who get less than what they are entitled to are being treated unfairly.

Notice, however, that I have already shown that both sweatshop owners and sweatshop workers have a claim to the benefits of social cooperation. It thus seems to follow that fairness requires that those claims be satisfied in direct proportion to their strength. Now, it is an open question whether, in virtue of their respective contributions, sweatshop owners and sweatshop workers have an equal claim to the social surplus or whether one of them is morally entitled to a greater claim than the other instead. For the sake of argument, I will remain agnostic on this issue. The important point is that whatever the correct answer to this question, each party’s claim to the social surplus should be satisfied in direct proportion to its strength. And so that means that sweatshop workers are entitled to a certain division of the social surplus even if that is more than what they voluntarily accepted. Accordingly, it will be unfair if they get a smaller share of benefits than is mandated by their claim.

We can put this point another way by saying that the reason why sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage is not that paying them the market-clearing wage does not benefit sweatshops workers enough. Rather, it is that if sweatshop owners failed to pay sweatshop workers a wage that is proportionate to the strength of their claim, then they would be extracting a share of the jointly produced social surplus that is, comparatively speaking, too much. By the same token, it is morally irrelevant that sweatshop owners are already doing more than anyone else to help sweatshop workers escape poverty. Because they benefited from the latter’s cooperative efforts, they are not entitled to a disproportionate share of the gains.

One important implication of this account, therefore, is that no fundamental moral significance attaches to the wage for which sweatshop workers bargained. Because they have a claim to a certain division of the social surplus, they must be provided with a level of compensation that meets that standard, even if it is more than sweatshop owners are willing to pay them. For, as I already explained, if I have a claim to something, then that just means I have a pro tanto moral entitlement to it, independently of whether someone else has a desire to give it to me. And so, even if sweatshop workers agreed to a lower level of compensation, their initial claim is not negated.

Indeed, that is also why it is a mistake to defend the NWC by appealing to the fact that sweatshop workers are paid more than workers in, say, the agricultural or informal sector (Powell 2014, chapter 4). The relevant issue is not whether sweatshop workers gained relative to noninteraction or whether they are better off than those who work elsewhere; rather, it is whether the distribution of the social surplus between sweatshop owners and sweatshop workers is itself fair. That is the appropriate baseline of comparison because it is only by virtue of the cooperative efforts of both sweatshop owners and sweatshop workers that a social surplus exists to begin with. The question thus arises as to how the social surplus should be distributed. Seen in that light, it seems difficult to deny that sweatshop workers have a claim to it, because it would not have been produced without their labor.

In a similar manner, it makes no difference that workers in the agricultural or informal sector would benefit even more than sweatshop workers from receiving this additional level of pay. Because the former did not help to produce the social surplus, they have no claim to it. And because they have no claim to it, they have no legitimate grounds for complaint if they do not receive a share of it. It is thus beside the point that giving it to them would increase overall welfare. If sweatshop owners and sweatshop workers already have a claim to the social surplus, then that preempts any other reason why someone else should get it. Now, that is not to say that if one fails to produce something, then one thereby lacks any claims at all. Not only have I already explained that claims have multiple grounds or bases but workers in the agricultural or informal sector may also have claims to various goods in their capacity as citizens. Indeed, the view I am defending is perfectly consistent with the idea that all people are morally entitled to certain goods simply by virtue of their status as human beings. The point is simply that only sweatshop owners and sweatshop workers have a claim to the particular goods they produced together.

This provides us with a test or standard by which to evaluate the truth of the NWC. For each accusation of unfair treatment, we can ask, Does the complainant have a claim to the good that is being distributed? If the answer to this question is yes, then there are grounds for believing that she is morally entitled to a fair distribution of it. By contrast, if the answer to this question is no, then we may conclude that her complaint is without merit.

By way of example, consider price gouging. If my analysis is correct, then an increase in the price of an essential good, such as food or water, in the wake of an emergency counts as an instance of wrongful exploitation only if the consumer of such a good has a claim to it. It is not clear, however, why that would be the case. After all, consumers do not generally have a claim to the goods they wish to purchase, and unlike sweatshop workers, they cannot legitimately argue that they helped to produce them. Now, that is not to say that if someone lacks a claim to a particular good, and thus has no basis on which to level the charge that she is being exploited, then that entails that her treatment is beyond moral criticism. For not only can one person wrong another without exploiting him but there may be other grounds on which to condemn an individual’s lack of access to basic necessities. For example, one can argue that governments have a duty to ensure that such goods are not “out of reach for the poorest members of the affected community” (Snyder 2009b, 281). The upshot is that the truth of the NWC depends on the facts of the case and, in particular, whether the person issuing the charge of exploitation has a claim to the good being distributed. Sweatshop workers do indeed have such a claim because they helped to produce the good in the first place. ¹⁸

3. OBJECTIONS

This completes my provisional argument for the thesis that sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage. As it stands, however, my account is potentially vulnerable to three natural objections. To fully vindicate my argument, therefore, I must explain my response to them.

3.1 Consent?

Thus far, I have been proceeding on the assumption that sweatshop owners and sweatshop workers have a claim to the social surplus, that is, an independent moral entitlement to the benefits they helped to create. Defenders of the NWC might argue, however, that this assumption ignores the crucial fact that both parties will have reached an agreement prior to their interaction as to how the social surplus should be distributed. They will, in other words, have already consented to the market-clearing wage. But if that is the case, then what grounds are there for paying sweatshop workers more than the market dictates?

This is an important objection. My response to it will help clarify the relation between the market-clearing wage and consent as well as the conditions under which the latter is morally decisive. I will begin by sketching two different ways in which consent matters in market transactions and then proceed to show that neither one of them undermines my analysis. Although defenders of the NWC accept only the first such account, I explore them both for the sake of completeness.

According to what we may call an “actual consent account,” consent matters because it is a way of showing equal concern and respect for both parties to a transaction. ¹⁹ On the one hand, consent is a way of showing equal concern to both parties because, other things being equal, individuals are the best judges of their own interests. Voluntarily agreeing to the terms of a transaction is thus an effective way of promoting their welfare. On the other hand, consent is a way of showing equal respect for both parties because, other things being equal, exercising autonomous choice over the conditions of one’s employment helps to realize one’s own values and commitments.

The problem with the actual consent account, however, is that “other things” are not always “equal.” In particular, sweatshop owners typically have decisive bargaining power over sweatshop workers given the income they receive from their wealth and capital. Unlike sweatshop owners, that is to say, sweatshop workers cannot hold out for a better wage. If they do, they will starve. For that reason, the distribution of the gains of social cooperation between them will be determined, for the most part, by the relative adequacy of their exit options. And because the greater the strength of one’s bargaining power is, the greater the distribution of the social surplus one will receive, there are strong grounds for believing that the terms of the exchange between sweatshop owners and sweatshop workers will be systematically biased against the latter. ²⁰ To be sure, that does not mean that the consent of sweatshop workers is invalid, nor that it is morally permissible for third parties to interfere with their transaction. ²¹ The point, however, is that sweatshop workers typically lack the bargaining power either to reliably promote their welfare or to realize their own values and commitments. We should thus be skeptical of the claim that the actual consent of sweatshop workers is the final word on the matter. ²²

Turn now to what we may call a “hypothetical consent account.” On this view, consent matters not because it is a way of showing equal concern and respect for both parties to a transaction but because it provides us with a test or standard for determining what a nonexploitative wage is. According to Alan Wertheimer (1996, 230–36), for example, sweatshop workers are paid a fair price for their labor if their actual wages are sufficiently close to the wages to which they would have consented in a perfectly competitive market. ²³ More precisely, the “hypothetical” market price is fair because, in a perfectly competitive market, sweatshop owners cannot take unfair advantage of defects in the decision-making capacities of sweatshop workers or unfair advantage of vulnerabilities in their circumstances. Sweatshop owners cannot, for example, exploit asymmetries of information or monopoly power to dictate unilaterally the terms of their transaction. Rather, both parties must accept the terms that the market dictates. Why? The reason is that, in a perfectly competitive market, no party to a transaction can influence the price at which it occurs. Both parties are “price takers,” and so market competition will force them to interact at a unique equilibrium point. Hence there is no way, in a perfectly competitive market, for sweatshop owners to offer sweatshop workers better terms on which to transact.

Given that Wertheimer’s hypothetical consent account is designed to eliminate the possibility that sweatshop owners can take unfair advantage of their superior bargaining power, it strikes me as a better explanation for why the market-clearing wage is morally permissible. ²⁴ But his view does not deny that sweatshop workers have an independent moral entitlement to a certain division of the social surplus; to the contrary, it explicitly affirms that. For if the wage to which sweatshop workers voluntarily agreed is lower than the wage that sweatshop owners would have paid them in a perfectly competitive market, then we have grounds for concluding that their actual wage is wrongfully exploitative. Sweatshop owners would thus have an obligation to pay sweatshop workers a wage that is higher than the one to which they actually consented.

The same thing can, of course, be said about my account. Like Wertheimer’s, it denies that (actual) consent is always morally decisive. That is why I said earlier that no “fundamental” moral significance attaches to the wage for which sweatshop workers bargained. Because consent is not the only thing that matters on my view, sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage even it is higher than the wage they voluntarily accepted.

Nevertheless, it is important to emphasize that although both Wertheimer and I are in agreement on this point, our accounts differ in how to explain it. Whereas his account focuses on the gap between the actual market-clearing wage and the hypothetical one, my account focuses on whether the wage to which sweatshop workers voluntarily agreed is lower than the wage to which they have a claim. The justification and content of our views are thus quite different. And so that means that not only will Wertheimer’s account generate a number of “false positives,” that is, cases, such as price gouging, in which it is mistakenly assumed that an unfair transaction is necessarily exploitative, but the hypothetical market price for sweatshop labor will also likely be much lower than the wage to which sweatshop workers have a claim. For even in a perfectly competitive market, the supply of sweatshop labor will far exceed its demand. Market competition will thus force sweatshop owners and sweatshop workers to transact at a relatively low equilibrium point—one that will almost certainly be below the value of what they each contributed.

What is more, my account of a nonexploitative wage (NEW) does not imply that it is never morally permissible to pay sweatshop workers the (actual) market-clearing wage (MCW). Rather, the relation between the two is far more complicated. To see why, note that, logically speaking, there are three possibilities. Either 1) MCW > NEW, 2) MCW < NEW, or 3) MCW = NEW. Consider possibility 1 first. If MCW > NEW, then sweatshop owners would indeed be morally permitted to pay sweatshop workers the wage they voluntarily accepted. But that is because having a claim to a certain division of the social surplus only establishes a “floor” below which sweatshop workers cannot be permissibly paid, not a “ceiling.” ²⁵ Accordingly, it is morally permissible for sweatshop owners to pay sweatshop workers a wage that is higher than their initial claim if the latter are able to bid up the price of their labor. ²⁶ Similar reasoning applies to possibility 3: if MCW = NEW, then it is morally permissible for sweatshop owners to pay sweatshop workers the wage to which they consented. But that is only because MCW just so happens to equal the wage to which sweatshop workers have a claim.

And so that leaves us with possibility 2, that is, when MCW < NEW. In light of what I have said thus far, this is the only scenario in which sweatshop workers’ consent is not morally decisive. Two things immediately follow from this fact. First, it is an empirical question whether this is the state of the world in which most sweatshop workers actually find themselves. However, given their lack of acceptable alternatives and weak bargaining power vis-à-vis sweatshop owners, I think there is good reason to believe that this is the case. Second, consent is, on my view, a necessary but not a sufficient condition of the moral permissibility of market transactions. But that is not particularly surprising. Because very few philosophers believe that consent is the only thing that matters in the employment context, my view is far less counterintuitive than it may appear at first glance.

3.2 Waive?

At this point, however, a defender of the NWC might object that even if sweatshop owners have a moral obligation to pay sweatshop workers a wage that is higher than the one they voluntarily accepted, that still does not fully settle the issue. Sweatshop workers can always waive that obligation as a condition of their employment. In the words of Powell and Zwolinski (2012, 469), “we do not deny that entering into a relationship can create new obligations. We simply hold that it is implausible to hold that those new obligations are not waivable, even when one party regards the other’s waiving of the obligation to be a necessary precondition for entering into the relationship, and the other party strongly prefers the relationship without the obligation to no relationship at all.”

What does it mean to say that sweatshop workers can waive this obligation? As a general matter, to say that an individual can waive an obligation is to say that they can authorize another person not to discharge or fulfill an obligation simply by offering their permission. Thus, suppose that A has an obligation to B. He must do X or not do Y. If B can waive this obligation, then she can permit A not to do X or to do Y simply by giving her consent in the appropriate situation. Of course, B’s consent must be informed and voluntary. But if that is the case, then A is permitted not to perform an obligation that would otherwise be morally binding. Consider the obligation to fulfill a promise, for example. If A promises to paint B’s house and B can waive this obligation, then she can authorize A not to paint it, even though that contradicts his original promise. A would, therefore, not be acting wrongly if he failed to paint B’s house under the circumstances. By the same token, if sweatshop workers can waive the obligation sweatshop owners have to pay them a nonexploitative wage, then they can permit the latter to pay them the market-clearing wage despite the fact that doing so would otherwise be morally impermissible. Sweatshop owners would thus be released from the obligation to pay sweatshop workers a wage that is directly proportional to the strength of their claim.

Before responding to this objection in detail, two initial points are worth mentioning. First, although both this objection and the previous one focus on the role of consent in employer–employer relationships, they are nevertheless different. Whereas the previous objection denied that sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage, the present one grants that obligation and yet holds that it can be waived. Second, and related to the earlier point, if this objection presupposes that sweatshop owners have such an obligation, then this seems to represent a retreat from the NWC’s original aim. Recall that the NWC was meant to show that sweatshop owners do not actually have a moral obligation to pay sweatshop workers more than the market-clearing wage. But if defenders of the NWC simply wish to argue that this obligation can be waived, then that seems to assume they already had such an obligation in the first place. Thus, even if this objection is true, it is not clear that it actually vindicates the NWC.

Let us, however, set this point aside and grant the force of this objection for the sake of argument. The question remains as to whether A’s obligation to pay B a nonexploitative wage is really waivable. Is A morally permitted to pay B the market-clearing wage simply because B gave him the permission to do so? I will argue that, even if we assume that obligations in general can be waived, there are good reasons to believe that B cannot waive this obligation in particular. ²⁷

To see why, notice that if B were to waive A’s obligation to pay her a nonexploitative wage, then this would likely have a negative impact on other potential workers. That is, if B chooses to waive this obligation as a condition of her employment, then this would predictably weaken the bargaining power of those who would otherwise hold out for the wage to which they have a claim. For why would A pay anyone a nonexploitative wage if he is able to find people who are willing to work for him for a wage that is significantly lower? The problem with this objection, therefore, is that it ignores the negative effect that B’s action would have on other workers. ²⁸

Now there are several points to emphasize about this argument. First, it is not paternalistic in nature. I am not arguing, in other words, that the reason why B cannot waive A’s obligation to pay her a nonexploitative wage is that doing so would be in her best interest. ²⁹ Rather, I am arguing that even though B may want to waive this obligation, and her consent is informed and voluntary, she must nevertheless refrain from doing so because that will protect the interests of nonconsenting third parties. This is important because it shows that one can reject paternalism and yet still accept the conclusion that this obligation cannot be waived. Second, it is an empirical question whether B’s waiving of this obligation always harms other people. For example, if the labor market is “tight,” that is, there is a high number of job openings relative to the number of unemployed people, then waiving this obligation might not have a significant impact on the bargaining power of other workers. By contrast, if the labor market is “loose,” that is, there is a low number of job openings relative to the number of unemployed people, then matters will be quite different. But because the market for sweatshop labor more closely resembles the latter situation, this should not be a problem for my argument. Finally, it is important to note that I am not providing a justification for why developing countries should legally prohibit workers from waiving this obligation. To the contrary, I am merely saying that there are good moral reasons for them not to do so. It is an entirely separate question whether the law should be also be changed.

3.3 Other Values?

The final objection concerns the relation between nonexploitation and other values. Brian Berkey (2020) has recently argued that even though it is pro tanto wrong to exploit sweatshop workers, there may nevertheless be circumstances in which doing so is morally permissible, all things considered. In particular, Berkey claims that focusing on the fairness of sweatshop worker wages potentially obscures the moral significance of justice and need in hiring decisions. ³⁰

For example, suppose that A can hire either B or C to work in his sweatshop. Although they are both equally qualified for the position, B is significantly worse off than C. Indeed, suppose not only that B is very badly off in absolute terms but that C is quite well off by comparison. Faced with this situation, A has the three options shown in Table 1. ³¹

Table 1: A’s Options

In Berkey’s view, there are two things to notice about this situation. First, if fairness is the only value at issue, then A should be indifferent between options 1 and 3. Because neither option is exploitative, there are no grounds to discriminate between them. But, according to Berkey, that seems hard to believe. Surely there are reasons for A to choose option 1 over option 3, given that B is in need. Berkey believes, however, that critics of the NWC cannot reach that conclusion because both B and C are treated fairly. They thus have no basis for drawing the correct verdict in this case.

Second, there might also be a reason for A to choose option 2 over option 3: even though option 2 is exploitative, it would nevertheless better enable B to meet her basic needs. It would thus be a mistake for critics of the NWC to insist that it would necessarily be wrong for A to exploit B. If the alternative is simply that A would hire C, then perhaps there are good reasons to prioritize needs satisfaction over fairness in cases when the two conflict. Indeed, this point can also be made in terms of the value of justice: if it is unjust for people to lack a sufficient number of resources to meet their basic needs, then A would have an additional reason for choosing option 2 over option 3. The upshot is that “there are further values that might compete with the value of fairness and thereby potentially make [A] hiring [B] at an intuitively exploitative wage morally better, all things considered, than her hiring [C] at a fair wage, even if doing so is worse in terms of the value of fairness alone” (Berkey 2020, 422).

Let us assume for the sake of discussion that Berkey’s substantive judgments in these cases are correct. Does that undermine my argument? I believe it does not. There are at least two reasons why this objection is fully compatible with my account of a nonexploitative wage. First, it is false to assume that the reason why critics of the NWC focus on the fairness of sweatshop wages is that they believe there are no ethical constraints on hiring decisions, or, as Berkey puts it, “there are no other potentially obligation-generating reasons for employers to hire anyone in particular” (420). Rather, the reason why they do so is that they are granting one of the central premises of the NWC. ³² Recall that one of the underlying assumptions of the NWC is that firms are morally permitted not to benefit people in the developing world by keeping production at home. Indeed, that is what makes the NWC so puzzling to begin with: if it is morally permissible for firms not to benefit sweatshop workers at all, then how can it be morally wrong to pay them the market-clearing wage? Viewed from this perspective, therefore, the reason why critics of the NWC disregard the ethics of hiring decisions is not that they think it unimportant but that they are trying to accommodate defenders of the NWC. ³³

Second, I never claimed that the obligation not to exploit people is absolute. To the contrary, I have consistently maintained that the obligation not to exploit people is a limited or defeasible one. More precisely, I argued that sweatshop workers have a pro tanto moral entitlement to a fair division of the social surplus irrespective of whether sweatshop owners have a desire to give it to them. But if that is the case, then one can accept the claim that there are strong moral reasons not to exploit someone without denying the further claim that it may be morally permissible, all things considered, to pay the worker an unfair wage. Indeed, that should not be surprising. Exploitation is not the only wrong, and so it is unclear why the obligation to treat people fairly cannot sometimes be outweighed by other moral values.

A potentially more challenging objection, however, turns on the value of fairness itself. According to Berkey, many people believe it is unfair if one person is worse off than another through no fault of their own. Thus suppose that B is significantly worse off than C not because of the choices she made but because of circumstances beyond her control. If that is the case, then not only might there be a fairness-based reason for A to hire B over C but there might even be a fairness-based reason to do so if A pays B an exploitative wage. Because doing so would reduce the level of inequality between B and C, it would make the overall outcome fairer. And so the upshot is that considerations of fairness alone do not necessarily condemn exploitative wage relationships. As Berkey puts it, “the value of fairness itself might favor hiring [B] rather than [C] and, somewhat more controversially, might even favor hiring [B] at an intuitively exploitative wage over hiring [C] at a fair wage” (422). Paradoxically, therefore, the value of fairness might itself require A to treat B unfairly!

But although this seems like a more challenging objection, its force depends on an equivocation between two different senses of the word fairness. ³⁴ On one hand is what we might call “noncooperative fairness.” This is the sense of fairness that Berkey has in mind when he says that it is unfair if one person is worse off than another through no fault of their own. It is a form of “noncooperative” fairness because it completely abstracts from the social, political, and economic relationships among people. On this view, it is unfair if one person is worse off than another through no fault of their own and thus even if there is no interaction between them. On the other hand is what we might call “cooperative fairness.” This is the sense of fairness that exploitation theorists have in mind when they are concerned with how the benefits of social cooperation between sweatshop owners and sweatshop workers are distributed. On this view, what matters is not whether one person has more or less than another in an abstract sense but whether the terms of their interaction are unfair.

Despite the fact that both these views are concerned with the value of “fairness,” their understanding of that term is clearly different. An example will help to illustrate the point. Suppose two candidates apply for the same job. Although the first is better qualified than the second, she is not hired because the employer is sexist. Now, there is no doubt that the employer’s decision is unfair because it makes someone worse off than another through no fault of her own. But it is clearly not exploitative. One person cannot exploit another if they have no relationship. Yet if these judgments can come apart, then we must be dealing with two separate issues. Indeed, they represent two entirely distinct ways in which one person can wrong another. If I exploit someone—that is, I act in a way that is (cooperatively) unfair—then I use the person’s vulnerability or desperation to my advantage. I gain at the person’s expense and thus treat the person as a means to my own end. In the words of Nicholas Vrousalis (2013, 131–32), exploitation is the “self-enriching instrumentalization of another’s vulnerability.” By contrast, if I treat someone in a way that is (noncooperatively) unfair—if I engage in, say, an act of wrongful discrimination—then I wrong the person, but I do not necessarily benefit from doing so. Now, that does not mean that the latter form of wrongdoing is less morally significant than the former; that would depend on a broad range of other factors. The point is simply that because one cannot be reduced to the other, it would be a mistake to confuse them in our analysis.

Pace Berkey, therefore, there is nothing paradoxical about the claim that “fairness” requires employers to hire badly off rather than well-off people or that it might even permit them to pay the former an exploitative wage. Because we are actually concerned with two different values here—noncooperative fairness, on the one hand, and cooperative fairness, on the other—this is simply another illustration of the fact that the moral obligation not to exploit people can be outweighed by other factors. But given that critics of the NWC need not deny this point, this objection poses no problem for my argument. ³⁵

4. CONCLUSION

My aim in this article has been to explain why sweatshop owners have a moral obligation to pay sweatshop workers a nonexploitative wage. To that end, I have defended two main claims. First, sweatshop owners are morally obligated to pay sweatshop workers a nonexploitative wage even though they have a right not to hire them and even though doing so will require them to pay sweatshop workers a wage that is higher than the one they voluntarily accepted. Second, this obligation on the part of sweatshop owners is not defeated by the fact that other individuals not party to the transaction would benefit even more than sweatshop workers from receiving this additional level of pay. But even if there are strong grounds for believing that it is morally wrong to exploit people, several important questions nevertheless remain. If there are no good reasons to deny the moral significance of exploitation, then what exactly follows from that claim? How, in particular, should we respond to the situation that victims of exploitation face? Because one way not to exploit people is to refrain from interacting with them, what is the strength of our moral obligation to enter into mutually advantageous and fair relationships? Although these questions are beyond the scope of this article, the first step in answering them is correctly explaining the wrongness of exploitation.