I. Introduction
One of the most common criticisms made by activists and scholars opposed to sweatshop labor is that it is wrongfully exploitative.Footnote 1 The economist's standard response to this charge is to point out that sweatshop labor provides tremendous benefits to workers—a fact demonstrated both by quantitative data on sweatshop wages and by the preferences of workers as revealed in their voluntary, indeed often eager, acceptance of sweatshop jobs.Footnote 2 Such a response does not exactly defeat the charge of exploitation—an exchange can be mutually beneficial and nevertheless wrongfully exploitative.Footnote 3 But it does seem to deflate the charge. Sweatshops, even if acting exploitatively, are at least doing something that provides substantial benefit to people in desperate need. By contrast, the vast majority of us—including companies that do not outsource their production overseas—do nothing to provide any comparable benefit. If this is exploitation, then how bad can exploitation be?Footnote 4
To some critics of sweatshops, however, this analysis appears to vindicate the moral status of sweatshop labor only by ignoring the very factors that make it morally abhorrent. It is a mistake, such critics contend, to focus so narrowly on the nature of the exchange between a particular sweatshop and a particular worker, and to note that the exchange is beneficial to the worker relative to the position in which she stood prior to the exchange. For it is precisely in the conditions that lie in the background of the exchange that the real injustice of sweatshop labor is to be found. People choose to work in sweatshops only because their situation is desperate, and the desperation of their situation is often—or, depending on one's theory of justice, almost always—a result of serious injustice in the background political and economic institutions against which their decision is made. The choice to accept sweatshop labor might be a smart move for workers, but it is the game itself, and not the particular moves that players make within it, in which the true nature of workers' exploitation is to be found.
My goal in this essay is to assess the adequacy of this response. What role, if any, should consideration of background injustices play in the correct understanding of exploitation?Footnote 5 My answer, in brief, is that it should play fairly little. Structural injustice matters, of course, but it does not typically matter for determining whether a sweatshop is acting exploitatively, and it does not typically matter in a way that grounds any kind of special moral responsibility or fault on the part of sweatshops or the MNEs (multinational enterprises) with which they contract. My argument shall proceed as follows. In Section II I will present a brief account of the concept of exploitation. In Section III I will discuss the way in which the dominant contemporary analysis of exploitation has seen it as a feature of discrete transactions, and will survey three ways in which some theorists have argued that a greater focus on structural issues is needed. I will provide an overview of the moral and empirical issues concerning sweatshop labor in Section IV. And in Section V, I assess the arguments introduced in Section III in order to determine the ways in which structural injustice does, and does not, matter for our thinking about exploitation. I conclude in Section VI with a discussion of the implications of structural injustice for the debate over sweatshop labor.
II. What Exploitation Is
In the most general sense, to wrongfully exploit someone is to take advantage of him in a way that is unfair or degrading.Footnote 6 Characteristically, exploiters interact with their victims in such a way that they derive some benefit from their victim, either by advancing their self-interest narrowly construed or by advancing their non-self-interested goals or projects.Footnote 7 The victim, in turn, might be harmed in the sense that she suffers a setback to her interests relative to how she would have fared in the absence of any interaction. But this is not necessary. For, even if the victim benefits relative to this baseline, the interaction might still be degrading or unfair to her. Imagine someone in danger of drowning who is offered rescue by the only nearby boat only if she agrees to sign away 75 percent of her future earnings. Intuitively, the rescuer's offer will strike most of us as deeply unfair. But just as clearly, the offer is one that benefits both parties relative to the status quo ex ante. After all, the victim almost certainly values her continued life more highly than the income she trades for it. If she did not, she presumably would not have agreed to the offer.Footnote 8
This last point suggests another feature of wrongful exploitation—namely, that exploitative transactions can be fully voluntary, at least in the sense of being free from outright coercion, force, or deception. To the extent that exploitative interactions are mutually beneficial, even if unfair, exploited parties will often have good reason to enter into them. Indeed, the more desperate a person's situation, the greater the reason she will have to accept any offer that promises to improve it. To the extent that it is unfair or degrading to take advantage of another person's vulnerability, then, voluntariness and unfairness are not only logically compatible, they often go hand in hand.
So far, my discussion of exploitation has remained deliberately neutral among competing theories. To whatever extent those theories might differ in their finer details, they are all (as far as I am aware) compatible with the broad analysis I have given so far. The most contentious aspect of a theory of exploitation, however, is its specification of precisely which features of a voluntary and mutually beneficial transaction make it wrongfully exploitative. At the beginning of this section, I characterized exploitative transactions as involving taking advantage of another in an “unfair” or “degrading” way. This disjunctive signals two approaches that have been used recently by philosophers to understand the wrongfulness of exploitation—the former focuses on the fairness or unfairness of the transaction, while the latter bases its assessment on a broadly Kantian idea of respect for persons.Footnote 9 Both of these approaches, in turn, admit of specification in a broad variety of ways. Unfairness, for instance, can be understood in terms of one party's receiving a disproportionate share of the cooperative surplus, by comparing the gains and losses of each party to that which would emerge in a hypothetical competitive market, or by any number of other measures. Similarly, what constitutes degrading treatment, or a failure of respect, will depend on one's answers to a host of other moral questions, such as whether it is inherently disrespectful to purchase from another—even at a fair market price—services such as their sexual or reproductive labor.Footnote 10
For the most part, I hope to avoid these controversies in this essay. I certainly will not be attempting to put forward any comprehensive theory of what makes exploitation wrong, nor will I assess the theories that others have put forth qua comprehensive theories. My aim in this essay is more limited. It is to explore the relationship between exploitation and background injustice. To the extent that it is possible, I aim to conduct this investigation in a way that is neutral between competing accounts of exploitation's wrongfulness. My hope is that it will be possible to shed light on this particular issue in a way that makes as few controversial assumptions as possible about the overall nature of a theory of exploitation, or the broader moral debates that underlie competing theories.
III. From Transactional to Structural Exploitation
Much contemporary philosophic thought has focused on discrete transactions as the locus of exploitation. For instance, Alan Wertheimer's 1996 book, Exploitation, which has structured and inspired much of the ensuing literature on the topic, contains chapters on unconscionable contracts, the exploitation of student athletes, commercial surrogacy, unconstitutional conditions, and sexual exploitation in psychotherapy.Footnote 11 And each of these topics is analyzed almost exclusively by focusing on some offer that A makes to B, and asking whether the terms of the offer are sufficiently unfair, or the process by which the offer was made and accepted sufficiently defective, to render the interaction wrongfully exploitative. There is little exploration of the sense in which relationships, rather than discrete interactions, can be exploitative. And there is little discussion of the way in which exploitation might be a feature of broader, structural features of the economy or the social context in which the interactions he analyzes take place.
Things were not always this way. Prior to Wertheimer's paradigm-shifting book, almost all philosophical discussion of exploitation focused on a Marxist or quasi-Marxist understanding of that concept. And for Marx, exploitation was more a feature of capitalism as such than it was a feature of any particular interaction between a capitalist and a worker. By putting exclusive control of the means of production in private hands, capitalism forces workers to choose between working for the owners of capital and starving to death. Capitalists, in turn, are able to take advantage of the existence of a large reserve army of the unemployed to pay workers just as much as is necessary to keep them alive to work another day, but not as much as the value of what their labor produces. In a way, then, capitalists are guilty of exploitation defined as the extraction of surplus value from the workers. But in a more important way, it is capitalism, and not particular capitalists, that deserves our blame. For the ruthless competitiveness of a capitalist system forces the hand of capitalists just as much as it does that of workers. Capitalists have no real choice but to attempt to extract as much value from workers as possible, lest their businesses be undercut by a competitor with fewer moral scruples.
Tied as it was, however, to a deeply implausible labor theory of value, to the idea that exploitation must entail forced (and not voluntary) transfers, and to a host of other problematic commitments, the Marxist account of exploitation fell almost entirely out of favor among philosophers.Footnote 12 This is not to say that philosophers rejected Marx's claim that capitalism and wage labor were inherently exploitative. But his analysis of what precisely that exploitation consisted in was found to be untenable, and so sympathetic philosophers set about reconstructing the Marxist critique of capitalism in a way that would stand up to more rigorous philosophic scrutiny.Footnote 13 By and large, this reconceptualization has been fruitful. Understanding exploitation in terms of unfair or degrading use rather than in terms of the forced extraction of surplus labor allows us to apply the idea of exploitation in contexts other than the economic relationships on which Marx focused, such as relationships between intimates.Footnote 14 And it has allowed philosophers to examine the possible injustice involved in capitalist relationships without attempting to fit their inquiry into the awkward Marxist analytic framework. Still, these advances did not come without a cost. For along with those elements of the Marxist analysis that almost certainly deserved to be discarded, much contemporary analysis of exploitation has also neglected Marx's insights into the role of social and economic structures in understanding that concept. And this, many have argued, has been a tremendous mistake.
There has thus arisen a growing consensus that consideration of the ways in which discrete transactions can be exploitative must be supplemented with an analysis of the way in which exploitation is related to structural injustice. Or, as one commentator has recently put it, that we require an account of exploitation that takes into consideration issues of fairness at the “macro” level as well as at the “micro” level.Footnote 15 Beyond the agreement that structural issues matter, however, there has been little agreement regarding how it matters. At least three distinct claims seem to have emerged in the literature.
The first is that the identification of structural injustice is necessary in order to determine which particular interactions are to count as instances of exploitation. This view has been defended most recently by Ruth Sample, who claims as an advantage of her view its recognition that “exploitation is not simply a feature of a particular transaction but derives its badness from preexisting social institutions that underwrite and encourage the transaction.”Footnote 16 Background injustices, especially those of a structural or institutional sort, establish inequalities between persons that can facilitate the exploitation of some by others. When we gain advantage by interacting with a victim of such injustice, she holds, “and that advantage is due in part to an injustice he has suffered, we have failed to give him appropriate respect.”Footnote 17 On Sample's account, failures of respect of this sort are a form of exploitation. Taking advantage of structural injustice is thus, for Sample, a sufficient condition of exploitation, and hence an identification of macro level or structural injustice is necessary to correctly identify at least certain kinds of micro level or transactional exploitation.
The second claim is that the identification of structural injustice is necessary, not so much in order to classify particular interactions as exploitative or not, but in order to think correctly about blameworthiness and responsibility for actions and situations of injustice. Iris Marion Young's recent essay on global justice, though not specifically focused on the concept of exploitation, makes essentially this point. In attempting to locate responsibility for global injustice, she argues, we too often rely exclusively on a liability model of responsibility that determines faulty conduct as that which deviates from a baseline of a normal, morally acceptable, background situation. A gunman's proposal of “your money or your life,” for instance, is judged coercive against what we deem to be the morally acceptable background situation in which I am entitled to both my money and my life. But in cases of global injustice, we cannot make this assumption. “When we judge that structural injustice exists, we mean that at least some of the normal and accepted background conditions of action are not morally acceptable. Most of us contribute to a greater or lesser degree to the production and reproduction of structural injustice precisely because we follow the accepted and expected rules and conventions of the communities and institutions in which we act.”Footnote 18 Young's account of the relevance of background injustices thus differs from Sample's. Unlike Sample, Young does not believe that reflection on structural injustice should lead us to identify as coercive or exploitative actions that, in the absence of that injustice, would have been regarded as morally innocent. Young does not recommend that we reject the liability model as a way of assigning responsibility to individuals. Instead, she argues that the liability model of responsibility must be supplemented by a different kind of responsibility—one that is political, rather than individualistic, in nature, and that draws our attention to ways in which collective action can be mobilized to correct injustice in the future, rather than to whom should be blamed for actions that have occurred in the past.Footnote 19
Finally, some have argued that a focus on structural injustice is necessary in order to identify political and economic institutions themselves as exploitative, and not merely as a means to identifying micro level transactions as exploitative. While it is not focused specifically on the concept of exploitation, Onora O'Neill's discussion of coercion and consent makes essentially this point. According to O'Neill, a focus on discrete transactions can lead us to lose sight of the real source of moral concern in situations of systematic injustice. Take, for instance, a capitalist's offer of low-wage labor to a worker. Such an offer appears, on its face, to be paradigmatically noncoercive, and the maxim on which the employer acts in making it seems perfectly universalizable and innocent. But, she notes, if we turn from the principles on which the employer herself acts and instead focus on “the principles which guide the institution of employment in a capitalist system,” we may find that “the underlying principle of capitalist employment, whatever that may be, may perhaps use some as means or fail to treat them as persons, even where individuals' intentions fail in neither way.”Footnote 20 On this line of reasoning, the critics of capitalism might be right in finding exploitation in the capitalist system, but it is the system as a whole, and not the discrete actions of individuals within it, that is wrongfully exploitative.
IV. The Sweatshop Debate
Before proceeding to evaluate these three claims, a brief detour is warranted in order to review the debate over the moral status of international sweatshops. This debate, which has received a tremendous amount of attention both from philosophers interested in the subject of exploitation and from social scientists interested in the empirical issues surrounding sweatshop wages and working conditions, can provide a useful case study through which the theoretical claims above can be better understood and assessed. This section will provide a brief background on the sweatshop debate, discuss some of the relevant empirical issues, and present the state of the moral debate regarding sweatshops and exploitation.
A. Background
There is no generally agreed-upon definition of the term “sweatshop,” but its connotation is clearly pejorative. The U.S. General Accounting Office defines a sweatshop as “an employer that violates more than one federal or state law governing minimum wage and overtime, child labor, industrial homework, occupational safety and health, workers compensation, or industry regulation.”Footnote 21 For the legal purpose of identifying and regulating industries within the United States, such a definition might be good enough. For the purpose of general moral analysis, however, it is clearly too narrow. Presumably, the kinds of activities in which sweatshops engage are wrong, if they are wrong at all, because they are unfair or disrespectful of persons, and not merely because they are prohibited by law. To avoid this problem, then, we might define a sweatshop as a place of employment in which worker compensation or safety is compromised, child labor is employed, and/or local labor regulations are routinely disregarded in a way that is prima facie morally objectionable. This definition captures the role that the term “sweatshop” plays as a signal of moral disapprobation, while leaving open as a conceptual matter the possibility that sweatshop practices might, on closer examination, be morally justifiable.
Sweatshops, so defined, exist throughout the world, including within the United States. For the most part, however, the moral debate over sweatshops has focused exclusively on sweatshops in the developing world. These sweatshops are typically independent firms that produce goods on a contractual basis for export to MNEs. The outsourcing of production to such sweatshops began in the 1970s when the technological and legal changes associated with globalization began to accelerate, and has been especially common in the apparel industry where the demand for low-skilled labor is high and the ability to avoid the risks associated with developing and stocking products is financially advantageous.Footnote 22
B. Empirical issues
Three kinds of empirical issues regularly come up in debates regarding the moral status of sweatshops. The first involves sweatshop wages; the second has to do with working conditions and safety in sweatshop labor; and the third has to do with the extent of labor law violations by sweatshops.
Perhaps the most widely-repeated claim about sweatshops, at least among nonacademic critics, is that the wages they pay are objectionably low. Ironically, this is also the claim that is the least supported by empirical evidence, of which there exists an abundance. A 1996 study, for instance, found that wages paid by MNEs in the developing world were generally significantly higher than wages paid by domestic firms in the same countries.Footnote 23 Because most sweatshops are not owned by MNEs, however, but are instead independent firms that work with MNEs on a contractual basis, it was initially thought that this study might not have much relevance for the debate over sweatshops. However, a 2006 study by Benjamin Powell and David Skarbek found that sweatshop wages are higher than wages elsewhere in the economy regardless of whether they are paid directly by MNEs or by domestic subcontractors—three to seven times as high, for example, in the Dominican Republic, Haiti, Honduras, and Nicaragua.Footnote 24 This is compatible, of course, with the (obviously true) claim that sweatshop wages are significantly lower than wages paid for comparable work in the United States. And it is also compatible with the claim that sweatshop wages might still be insufficient to allow workers to achieve a certain standard of living. This means that sweatshops might still be failing to meet Denis Arnold's often-repeated standard that they “ensure that employees do not live in conditions of overall poverty by providing adequate wages for a 48 hour work week to satisfy both basic food needs and basic non-food needs.”Footnote 25 But it does demonstrate that sweatshop wages, even if they are not high enough to lift workers entirely out of poverty, are generally significantly higher than wages available to workers elsewhere in their economy.
Worker safety is another main complaint of anti-sweatshop activists. And while comprehensive data on this topic is more difficult to collect, the literature is replete with shocking anecdotes.Footnote 26 Workers are pressured by employers to remain on the job in spite of serious and pressing injury, are exposed to dangerous chemicals and airborne pollutants, and often work on poorly maintained machinery with inadequate training or safety measures.Footnote 27 In some ways, concern over safe working conditions is distinct from concern over wages. The steps needed to correct issues of workplace safety will be different—perhaps easier, perhaps more difficult—than those needed to increase wages. And morally, one might hold that unsafe working conditions violate workers' rights in a way that low wages do not. In many ways, however, the issues are inexorably linked. Both increases in wages and improvements in working conditions, for instance, cost employers money and resources.Footnote 28 From the perspective of the employer, both kinds of cost can be considered as part of workers' overall package of compensation. And while profit-maximizing employers are indifferent to the mix of goods that goes into an employee's compensation package, they are extremely sensitive to the overall cost of that package. Increases in worker safety, therefore, will very often come at the expense of other forms of compensation.Footnote 29 Moreover, there is some reason to believe that workers themselves would be unwilling to make this tradeoff. A 2010 survey of Guatemalan sweatshop workers found that few of them are willing to sacrifice any wages in order to receive more health and safety benefits.Footnote 30 Employees were asked about ten improvements in working conditions and if they would be willing to accept lower wages to improve any of these conditions. On eight of the questions more than 90 percent of the workers answered “No.” Paid vacation was the most popular improvement but even here more than 81 percent of the workers answered that they would not sacrifice any amount of wages for it. Nearly 65 percent of workers surveyed answered that they were unwilling to sacrifice any wages for each of the ten improvements.
A final empirical issue relevant to the debate over sweatshops and exploitation involves sweatshops' violation of labor laws. According to Denis Arnold and Norman Bowie, such violations are widespread and often perpetrated with the implicit or explicit cooperation of state institutions.Footnote 31 Employers force workers to work overtime, suppress workers' right to unionize, engage in the illegal pregnancy screening of female potential employees, and fail to pay workers wages on time and in the correct amount. Internal investigations by the clothing company Gap revealed that “between 25 and 50 percent of its contract factories lacked full compliance with local labor laws in North Asia, Southeast Asia, the Indian subcontinent, sub-Saharan Africa, Mexico, Central America, and the Caribbean, and South America. In China, more than 50 percent of its contract factories lacked full compliance with local labor laws.”Footnote 32 And this was the situation in a company that was making an active effort to ensure compliance.
There are probably a variety of relatively benign reasons for such noncompliance including ignorance of the law, lack of clarity in the law, and ignorance of implementation mechanisms.Footnote 33 But a significant part of the reason for both the lack of enforcement on the part of the state and the lack of compliance on the part of the sweatshop is almost certainly financial. Sweatshops do not want to comply with laws that raise their costs and decrease profitability. Governments, in turn, face cost-pressures of two different sorts. First, there is the expense associated with enforcement itself. Given the strained budgets within which many developing countries operate, this can be a significant constraint. Second, there is the possible cost associated with lost tax revenue should overly-vigorous enforcement lead the sweatshop to shut down, or its affiliated MNE to search elsewhere for a less burdensome regulatory environment. Sweatshops' willingness to violate labor law in the face of weak enforcement is thus another possible source of worker exploitation.Footnote 34
C. Moral issues
How, then, do these empirical issues bear on the question of the moral status of sweatshops, especially the question of whether sweatshops are guilty of exploiting their workers? On an account of exploitation that is focused on transactional fairness alone, the case against sweatshops appears thin. Alan Wertheimer's account, arguably the most influential fairness-based account of exploitation, is founded on the idea that for at least a broad range of cases, the fairness of a transaction between A and B can be analyzed in terms of the agreement that would be reached between them in a hypothetical competitive market.Footnote 35 A motorist who comes across a driver stranded in a snowstorm and takes advantage of his monopoly position by charging the driver an exorbitant sum to be rescued acts exploitatively on this account, because the price he charges is higher than the price that would be charged were multiple rescuers to compete for the driver's “business.”Footnote 36 But there is little reason to suppose that sweatshops and the multinational companies are operating in a noncompetitive environment.Footnote 37 The compensation sweatshop workers receive for their labor is generally no higher than the amount they contribute to the firm (their marginal revenue product), and no lower than the value they place on their next best alternative employment.Footnote 38 Their compensation is low because the values of these upper and lower constraints are low, and not because of any failure of competitiveness in the labor market. This is why even Denis Arnold, himself one of the most persistent philosophical critics of sweatshops, thinks that on a transactional account of exploitation of the sort Wertheimer defends, the claim that sweatshop workers are exploited is implausible.Footnote 39
Arnold seems to believe that a Kantian account of transactional exploitation of the sort defended by Allen Wood will be more successful in substantiating the charge of sweatshop exploitation. But it is not clear why this should be so. Demonstrating that the compensation sweatshop workers receive is disrespectful is hardly a simple matter, especially if one has already conceded that the compensation is not unfair. Arnold insists that any wage below a “living wage” is inherently disrespectful and fails to treat workers as beings with dignity and autonomy who are ends in themselves.Footnote 40 But it is not at all clear how he would respond to the claim, raised for instance by Sollars and Englander, that the requirement to treat others as ends in themselves generates a duty of beneficence that is imperfect in form, and cannot ground a perfect obligation on the part of employers to pay a specified wage rate.Footnote 41
Jeremy Snyder puts forward a more promising argument, holding that while Kantian ethics grounds only an imperfect duty of beneficence in the abstract, this duty can come to be “specified” by the particular relationships we enter into with others, thereby taking on a “perfect, strict form.”Footnote 42 In the abstract, we must use our own practical reason to determine who to help, on what occasions, with what kind of and how much assistance, on what conditions, and so on. But when we enter into “relationships of use” with certain, particular others, our wide and imperfect duties become more narrowly specified. In this way “employers do not simply have an imperfect duty to help some of their employees to achieve a decent minimum some of the time; rather, employers are required to cede as much of their benefit from the interaction to their employees as is reasonably possible toward the end of the employees achieving a decent minimum standard of living.”Footnote 43
Snyder's account thus has the potential to do what Arnold and Bowie's account could not—ground a perfect obligation on the part of sweatshops to pay a living wage.Footnote 44 But Snyder's argument faces difficulties of its own. The most significant of these, I think, is not unique to Snyder's account but is characteristic of almost all accounts that condemn transactions that are mutually beneficial but in some way unfair.Footnote 45 There is a certain incoherence, I think, between our intuitions regarding exploitation and our intuitions regarding duties of aid. This incoherence manifests itself when we condemn sweatshops and the MNEs with which they contract for exploiting workers in the developing world, but do not condemn businesses that refrain from outsourcing production at all. If the empirical data we have surveyed about sweatshop wages are correct, sweatshops are providing significant benefits to their workers, while firms that do not outsource are (as far as we know) doing nothing to benefit them. How, then, can it be permissible to neglect workers in the developing world, but impermissible to exploit them, when exploitation is better for both parties (including workers who are in desperate need of betterment)?Footnote 46 Consistency seems to require that we revise our moral beliefs: either by holding that neglect is worse than we previously thought, or by holding that exploitation is less bad than we previously thought. Either way of resolving the inconsistency, however, leads us to abandon the belief that sweatshops who exploit their workers are doing anything especially wrong.
Snyder is aware of this objection to his position, and has responded to it by arguing that the idea that special relationships create special moral obligations is one that is supported by both moral theory and intuition.Footnote 47 Persons can choose not to enter into relationships with individuals who stand in positions of vulnerability but, Snyder contends, “should they choose to enter these relationships, this choice helps to specify a general duty of beneficence” in a way that prohibits exploitatively low wages.Footnote 48 But this is still puzzling. For, suppose we agreed with Snyder that a company's entering into an employment relationship with a needy individual was sufficient to generate a strict, perfect duty of beneficence on the part of the company toward that employee.Footnote 49 Compliance with such a duty, let us stipulate, would require the company to provide a benefit to its employee of amount Y. But let us suppose that the employer is only willing to provide its employees with a benefit of amount X, where X < Y. Would it be permissible, on Snyder's account, for the employer to make its offer of employment contingent upon its workers' willingness to waive their right to benefits of amount Y? Prior to entering into a relationship with employees, the employer has only an imperfect duty of beneficence. No prospective employee has any valid moral claim upon its assistance. Thus, the company would not be acting wrongly if it refused to hire or assist the prospective employee at all. But if it is permissible for the employer not to hire prospective workers, and if hiring prospective workers at benefit level X is better for both the employer and the worker than not hiring the prospective workers at all, then how could doing so be wrong? If employees' claim to benefits at level Y is one that they may waive when it is in their interest to do so, then employers are not necessarily acting wrongly in providing their employees with benefits at level X. If, on the other hand, employees' claim to benefits at level Y is not waivable, then Snyder's account is committed to holding that employers are forbidden from entering into agreements with employees to hire them at wage level X, but permitted to refrain from hiring them at all, even if hiring them at level X is better for both workers and employers than not hiring them at all, and even if both workers and employers would prefer to enter into an employment relationship at wage level X than to not enter into any wage relationship at all. This is deeply counterintuitive.
To be clear, I am not making the strong claim that any agreement that is voluntary and mutually beneficial is thereby nonexploitative and morally permissible. I agree, for instance, with Chris Meyers that someone who offers to rescue another person from the desert only if the other consents to an act of sodomy, or to handing over their entire net worth, is acting in a way that is repugnant, exploitative, and morally wrong.Footnote 50 My claim is that the kinds of considerations we have surveyed in the section tend to undermine the claim that MNEs and the sweatshops with which they contract are treating workers unfairly and hence exploitatively. These considerations include the empirical facts about the benefits sweatshops provide to their workers, as well as the fact that MNEs are under no moral or legal obligation to provide any benefits at all to workers in the developing world. It is also worth bearing in mind that the only motivation MNEs have for providing any benefit is the increased profit they hope to gain from the exchange, and that their capacity to provide benefit to workers is a function of this motivation. In this way, MNEs are less like the kind of adventitious rescuer described by Meyers and more like a “professional salvor whose rescue capacity results from a planned investment.”Footnote 51 Admiralty law has long recognized that the awards that may legitimately be claimed by the latter are, while not unlimited, nevertheless significantly higher than those to which the former has good claim.
V. Does Structural Injustice Matter?
Still, even if sweatshops cannot be said to be acting exploitatively on a transactional account of exploitation, perhaps a structural account could do better? After all, most sweatshops operate within a social context that is plagued by longstanding, deep, and massively destructive injustice. For example:
• Suppression of Unions—Governments in the countries in which sweatshops operate routinely suppress workers' ability to organize and bargain collectively. The suppression is sometimes violent, and is sometimes executed at the request of MNEs and/or the sweatshops with which they contract.Footnote 52
• Seizure of Land and Natural Resources—Land in developing countries, like land elsewhere in the world, generally did not come into the hands of its current owners by a process of peaceful Lockean labor mixing and voluntary exchange. Violence and theft are as much responsible for current holdings as investment and labor. And Western governments' deference to rulers' alleged property rights in trade and lending reinforces this injustice.Footnote 53
• Protectionism—Powerful Western governments, through their influence in international organizations such as the World Trade Organization, compel developing countries to open their markets while at the same time protecting their own markets with “tariffs, quotas, anti-dumping duties, export credits, and huge subsidies to domestic producers,” thus crippling developing economies' ability to compete in those areas where they may have a genuine comparative advantage.Footnote 54
This list could be extended indefinitely to include not just Western countries' massive imposition of externalities such as CO2 on the developing world but, depending on one's theory of justice, their ruthless enforcement of intellectual property rights or perhaps even the system of global capitalism itself. MNEs and the sweatshops with which they contract might provide workers with significant benefits, and the jobs they offer might be better than any of the other options available to workers, but if a large part of the reason that workers' options are so limited is the corporatist-capitalist structure of which MNEs are a significant part, then it is not at all obvious that the benefit they provide should be sufficient to get sweatshops off the moral hook.Footnote 55
Earlier, I discussed three ways in which considerations of structural injustice might be relevant to a correct understanding of exploitation.Footnote 56 In the remainder of this section, I will critically assess each of those suggestions with specific reference to the moral debate over sweatshop exploitation. I will consider, first, whether identifying structural injustice might be necessary or sufficient in order to determine which particular interactions are to count as instances of exploitation (I argue that it is not); second, whether identifying structural injustice is necessary to think correctly about blameworthiness and responsibility for actions and situations of injustice (I argue that it is, but not in a way that is helpful for thinking about the special obligations of MNEs and sweatshops); and finally, whether identifying structural injustice is necessary in order to identify political and economic institutions themselves as exploitative, and not merely as a means to identifying microlevel transactions as exploitative (I argue that it is).
A. Structural injustice and transactional exploitation
Injustice at the structural level is not necessary for particular transactions to be exploitative. Nor is it sufficient. In this, I agree with Alan Wertheimer and Mikhail Vladman, and disagree with Ruth Sample and John Roemer.Footnote 57
Injustice at the structural level is not necessary because, as Wertheimer notes, “even in a reasonably just society, people will find themselves in situations in which they can strike an agreement that will produce mutual gain, and some of these cases will give rise to allegations of exploitation.”Footnote 58 Indeed, the canonical examples of exploitation of the sort offered by Meyers involve no obvious background or structural injustice.Footnote 59 Offering to rescue an individual who became stranded in the desert because of her own poor planning in exchange for her entire net worth is wrongfully exploitative because of the unfairness of the terms of the transaction, not because of the history or institutional background to that transaction. Thus it is not the case, as Roemer claims, that “we view exploitation as a bad thing only when it is the consequence of an unjust unequal distribution of the means of production.”Footnote 60
Nor is it the case that, as Sample claims, “if we gain advantage from an interaction with another, and that advantage is due in part to an injustice he has suffered, we have failed to give him appropriate respect” and hence exploited him.Footnote 61 Someone who takes pleasure in helping victims of injustice, and performs helpful acts solely in order to receive this pleasure, probably does not act wrongly at all and certainly does not wrong the person whom she helps.Footnote 62 Likewise, someone who performs a useful service at a fair price for victims of injustice does not wrongfully exploit them. A contractor who rebuilds a home that was destroyed by arson, but charges only the normal market price, is not taking unfair advantage of the owner. Perhaps a more morally virtuous contractor would rebuild the home at a discount, or for free. But a failure to exhibit supererogatory kindness does not render one's actions exploitative, and at any rate our judgment that it would be nice to give a break to someone whose home was destroyed seems independent of the cause of the destruction—cutting a break to someone whose home was destroyed by lightning-induced wildfire seems no more and no less virtuous. The identification of structural or background injustice, then, is not sufficient for identifying particular transactions as exploitative.
That background injustices are neither necessary nor sufficient to the identification of particular transactions as wrongfully exploitative should not be too surprising. To say this, after all, is not to say that background injustices do not matter. It is simply to say that not everything that matters, morally speaking, falls under the category of exploitation. We need not view every issue of moral concern as a nail to be hammered with the concept of exploitation. Political philosophy has other tools.
B. Structural injustice and moral responsibility
If considerations of structural injustice are not necessary or sufficient in order to determine the status of a particular transaction as wrongfully exploitative, perhaps they are nevertheless important in some way for thinking about the moral responsibility and/or blameworthiness of agents. For Iris Marion Young, as we have seen, sweatshops are a prime example of structural injustice. Structural injustice “exists when social processes put large categories of persons under a systematic threat of domination or deprivation of the means to develop and exercise their capacities, at the same time as these processes enable others to dominate or have a wide range of opportunities for developing and exercising their capacities.”Footnote 63 Individuals who contribute to these unjust social processes bear a kind of moral responsibility for them and for the results they produce,Footnote 64 though Young notes that it is a different kind of responsibility than that with which moral and legal philosophers are most familiar. It is a form of political responsibility, which, among other things, is a kind of responsibility that is forward-looking and discharged through collective action, rather than backward-looking and individualistic.Footnote 65 For Jeremy Snyder, this model of responsibility has direct implications for the moral responsibility of sweatshop managers and MNEs, since “even if unjust structures limit the degree of choice that employers have over the wage levels that they may offer while remaining competitive, they may yet have a [political responsibility] to change these structures. If so, the employer's [political responsibility] must be discharged if the employer is to avoid exploiting her workers.”Footnote 66
But is it fair to hold MNEs and sweatshops responsible for the unjust background conditions in which they operate?Footnote 67 Certainly in some cases it is. Consider those cases in which the MNE is causally responsible for the unjust background conditions in a fairly direct way. Some such cases will involve a form of collusion between the MNE and the host country government. The MNE agrees to contract with firms in the government's jurisdiction and thereby provide the government with tax revenues and other economic benefits, but only in exchange for the government's promise to forcibly suppress labor unions. In cases like this, where the MNE is part of the cause of the background injustice, it is certainly appropriate to say that they hold partial moral responsibility for it.Footnote 68 Other cases will be more difficult to analyze. What are we to say about a case in which the structural injustice is perpetrated by the government only because it anticipates that some MNE will show up to take advantage of it? The government suppresses workers' unions, for instance, just because it believes that in doing so it will create a more attractive environment for MNEs, which will then be more likely to do business in the government's jurisdiction and generate various economic benefits of which the government can take advantage. These cases are like those in which a pimp coerces a woman into serving as a prostitute in anticipation of the revenue that will be generated by the customers who will take advantage of her services.Footnote 69 To what extent does the person who hires the prostitute bear moral responsibility for the injustice of her situation? To what extent is he blameworthy?
The analysis of such cases is difficult.Footnote 70 They are not, however, the kind of cases that Young's account is designed to deal with.Footnote 71 For the entire point of Young's account was to provide a supplement to the standard liability model of responsibility, one that is applicable even to agents who are not “guilty or at fault for having caused a harm without valid excuses.”Footnote 72 It is an account designed to apply to those who are “connected to but removed from” the harms produced by structural injustice, but not to establish the blameworthiness of such agents.Footnote 73 Rather, its point is to show that such agents have a forward-looking political responsibility to take part in remedying injustice.
But it is not clear that Young's account can accomplish even this much. Or, rather, it is not clear that it can show that sweatshops or MNEs have any special responsibility to remedy injustice, merely by virtue of their “social connection.” Young argues, seemingly innocently, that special responsibilities to correct for injustice arise from social connection. And, rather less innocently, that simply pointing out that the persons suffering from injustice are human is “not enough” to ground the claim that we have obligations of justice toward them.Footnote 74 But the special status of social connections is never defended in an entirely clear way. Young states that the kind of social connection that generates political responsibility is “participation in the diverse institutional processes that produce structural injustice.”Footnote 75 And this might make sense if we thought that agents who participated in such structures were wrongly benefiting from them at the expense of victims of injustice. But this cannot be what Young is saying. For, if what grounds responsibility is not participation in unjust structures as such but wrongful participation in unjust structures, then Young's account would seem to collapse into a standard liability model of responsibility. One might think instead that participation matters because it provides one with opportunities to fight the injustice. But there is no reason to think that all who participate in unjust structures will have such opportunities, nor that all nonparticipants will lack them.Footnote 76 It is hard to see, then, how “social connection” could be specified in any way that would nonarbitrarily assign any kind of special responsibility to sweatshops or MNEs for remedying structural injustice.Footnote 77 Those who own or operate sweatshops or MNEs might have the power or opportunity to work to remedy certain kinds of structural injustice, but if this power and opportunity is neither universally nor uniquely present among them, it will not generate any universal or unique obligation on their part.
C. Structural injustice and structural exploitation
Perhaps we have been looking in the wrong place. Instead of taking structural injustice to be evidence that particular interactions within the context of those structures are exploitative, or that parties to those transactions are morally responsible or blameworthy for the results of the injustice, perhaps structural injustice should lead us to conclude that the legal and economic institutions that gave rise to them are themselves exploitive.
This, I think, is the most promising approach for linking structural injustice to a theory of exploitation. And its absence from the contemporary philosophical literature on exploitation should be nothing sort of shocking. The modern centralized state bears significant responsibility in structuring almost every interaction that persons enter into, commercial, professional, legal, social, and otherwise. And yet, when contemporary theorists discuss exploitation, it is almost always exploitation that occurs within that structure, between various participants who are subject to the state's rules. To the extent that the state itself is brought up, it is as a kind of deux ex machina the powers of which can be brought to bear to correct exploitation in the market, the family, and elsewhere. The notion that the state itself can be a significant source and tool of exploitation is almost never mentioned.Footnote 78
This is not the place to develop a full theory of state exploitation. But we might perhaps make some preliminary points toward the end of developing such a theory. We can note, first, that any theory of state exploitation will need to develop or rely on some theory of collective actions and/or intentions, and to be careful to distinguish those actions that are properly attributable to the collective entity we call “the state” and those that ought to be ascribed to specific individuals. Onora O'Neill's discussion of state coercion illustrates the difficulties inherent in such an approach. For determining whether capitalism is coercive, she notes, requires determining “the principles that guide the institution of employment in a capitalist system.”Footnote 79 But whatever might those principles be? How are we to ascribe maxims to institutions which may have evolved as “the product of human action but not of human design”? Similar problems will no doubt plague a theory of state exploitation.
Second, it will be important to distinguish between some of the different forms that state exploitation can take. At least two distinct types are readily apparent. In one, the powers of the state can be used to unfairly benefit the state itself, or the persons who constitute it in their capacity as agents of the state.Footnote 80 Suppose, for instance, that an agent of the state, acting in her official capacity, threatened unfairly to block a company's access to a particular market unless that company paid a special tax, the proceeds of which would be used to fund the activities of the state. To the extent that such a demand is illegitimate, we could say that it constitutes a kind of state exploitation—an unfair use of state power to take advantage of the (relatively) vulnerable business. A second form state exploitation might take, however, involves the use of state power to benefit persons outside the institution of the state, or agents of the state in their capacity as private persons. This phenomenon has been exhaustively catalogued and analyzed by public choice economists under the label of “rent seeking,” but to the extent that we find the activity unfair or otherwise morally illegitimate, it could very well be described as a kind of wrongful exploitation.Footnote 81 A MNE that makes its entrance into a developing country (and payment of taxes to that country's government) contingent upon the state's illegitimate restriction of potential competition can be said to be using the state as a tool of exploitation in this sense. Perhaps we might also distinguish between two more specific forms this scenario might take: one in which the state agent is neither morally nor legally authorized to restrict competition, and one in which she is legally but not morally authorized to restrict it. In the first, the agent of the state is acting as a kind of rogue—abusing the powers of her office in order to benefit privately. In this case, both the MNE and the state agent are using the state as a tool of exploitation. In the second case, only the MNE can be said to use the state in an exploitative way. The agent's action is probably best regarded as an action of the state and so, if anything, is an instance of the first type of state exploitation described in this paragraph.
Finally, a full theory of state exploitation will need some way to distinguish between the kinds of legitimate exchanges that occur between private persons, and the exploitative deals that occur between, say, state agents and rent seekers. Nineteenth-century classical liberals tended to make this distinction in a rhetorically powerful but philosophically clumsy way, distinguishing, as in the case of Jean-Baptiste Say, between “productive” exchanges on the market and the “unproductive” activities of government.Footnote 82 But the activity of rent seeking can be productive both in the sense of creating actual goods or services, and in the sense of producing an overall increase in utility (obviously for the parties to the exchange but also, perhaps in cases where rent seeking is employed to spur the provision of a genuine public good, for society as a whole). What makes rent seeking exploitative is not its being unproductive, but its being unfair. But what exactly this means in the context of political institutions is something that requires much further exploration.
Before bringing this section to a close, it is worth reminding ourselves that not all forms of structural injustice will be instances of state exploitation. Injustice can take many forms, and exploitation is only one of them. States can oppress their populations as well, in which case the injustice they do may neither yield nor be intended to yield any benefits of the sort at which exploitation characteristically aims.Footnote 83 And coercion can be unjust without necessarily being exploitative.Footnote 84 State exploitation, then, although it is an unduly neglected form of state injustice, is only one form among many.
VI. Conclusion
This essay began with a debate over the moral status of sweatshops. On one side of that debate are economists and others who hold that the benefits sweatshops provide to their workers renders their conduct and the conduct of the MNEs with which they contract not wrongfully exploitative, or at least not seriously so. On the other side are those who hold that the true exploitative nature of sweatshops only becomes apparent once we shift our focus from the particular details of the exchange between sweatshop and worker, and look instead at the deep injustices in the background of that exchange.
This essay has sought to explore the moral debates over sweatshops, and to examine the way in which considerations of structural injustice might or might not be relevant to the charge of sweatshop exploitation. And the result of our inquiry has been a modest vindication of the economists' defense of sweatshops. We have seen that the existence of background injustices does not render an otherwise fair exchange wrongfully exploitative. And we have seen that while sweatshops and the MNEs with which they contract have a moral responsibility to fight unjust institutions, this is an obligation they share with everyone who is able to do so, and so cannot ground any special fault on their part.
Nevertheless, it is important not to overestimate the reach of my argument here. I have not attempted to offer an impossibility proof. It is not only possible, but it is virtually certain that some sweatshops do wrongfully exploit their workers. Those that actively cooperate with government to suppress their workers or their competitors, for instance, are guilty of not only benefiting from structural injustice but also perpetuating it for their own gain. This is wrongful exploitation of the clearest sort. And even without government intervention, it is possible for sweatshops to wrongfully exploit their workers. The fact that sweatshop workers are willing to accept a certain package of wages and benefits from sweatshops is not enough to show that package to be fair.
Drawing the line between fair and unfair offers—like many cases of conceptual line-drawing—is of course a difficult business. The unfairness that constitutes exploitation, on my view, is not the sort of thing that can be assessed by means of any precise formula. Determining the fairness or unfairness of conduct is largely a matter of balancing the reasons weighing for and against that conduct, and thus often reduces to an exercise of judgment about which reasonable people can disagree. I have argued that canonical conditions of sweatshop labor—low wages, long hours, and unsafe conditions, for instance—do not suffice to render sweatshop labor wrongfully exploitative, once the broader range of empirical and moral features of sweatshop labor are taken into account. But when, for instance, some sweatshops use their employee's fear of termination to perpetrate physical and sexual abuse, then to my thinking at least the balance of reasons tips decidedly toward a judgment of unfairness.
Structural injustice, then, does not render sweatshop labor wrongfully exploitative. But I sincerely hope that readers will not take away from this paper the message that structural injustice does not matter, or is of small importance. The unjust seizure of land and other natural resources by governments and plutocratic interests, the suppression of alternative opportunities for workers through restrictive intellectual property laws and trade barriers, the failure to uphold the right of workers to freely organize into labor unions, and so on are all significant injustices that affect the lives of workers in the developing world for the worse. And they may, indeed, be injustices without which sweatshops as we know them would cease to exist. To the extent that it is in our power to do so, we should work to stop and rectify them.
My concern in this essay has been to explore the moral responsibility of sweatshops, not of humanity in general. And to explore in particular their responsibility for one specific kind of wrongdoing—exploitation. There is a danger, however, in the analytic philosopher's tendency to brush aside as irrelevant those issues that do not fit neatly within the concept he is exploring. Structural injustice may not have much to do with transactional exploitation, but it has a tremendous amount to do with people's ability to live peaceful and fulfilling lives. It is thus an issue of the utmost concern for us not only as philosophers (in other papers, at other times), but as human beings.
