p.490
Benjamin Ferguson
Consumers confront many ethical choices. The purchase of consumer goods in a globalized economy often has far-reaching consequences. Sometimes we are unaware of the ethical implications of our consumer behavior, or choose not to think very hard about such choices, but even when we are ethically motivated to “do the right thing,” it often seems that figuring out just what the “right thing” involves is a rather complex matter. One part of this complexity involves the ethical evaluation of the businesses with which we interact.
Suppose you would like to buy a new smartphone and, faced with the large number of options on the market, you ask a friend to help you choose one. As it happens, you think the latest model from brand X is a good option, but your friend disagrees. “Sure,” she says, “it has great features and a low price, but brand X exploits the workers who produce the phone. You should not buy it.” When you ask what she means by exploitation, she cites the low pay of the employees. But, you ask, aren’t the wages in the factories where the phones are produced often higher than those available elsewhere? She acknowledges that this may be true in some cases, but, she counters, “the problem is not with the amount they are paid, but with the proportion of the profits the workers receive—they are paid pennies for producing phones that sell for hundreds of dollars.” “Besides,” she continues, “exploitation is not only about money, but also about the quality of work.” “The factories,” she claims, “hire underage laborers, who they harass and force to work long hours in physically and psychologically unsafe conditions.”
You agree that her list of objections concerns you, but something seems odd. If the conditions are horrible, then why don’t the workers simply quit? Well, your friend responds, “sometimes the workers are forced to labor, but in many other cases, these jobs, though horrible, represent the best options for the workers.” While you do not dispute the facts your friend presents, you are unsure whether she has reached the right normative conclusion: if these jobs—despite the low pay and harmful conditions—really are the best thing going for poor workers, it seems odd to conclude that we should refuse to buy brand X phones. After all, if we asked, wouldn’t the workers themselves claim they would rather work in factories producing phones for brand X than not? This line of reasoning is neatly summarized by Joan Robinson’s claim that “the misery of being exploited by capitalists is nothing compared to the misery of not being exploited at all” (Robinson 1962: 45).
p.491
This imagined conversation references many—but not all—of the confusing features that complicate our ethical judgements about fair wages and working conditions experienced by the poorest of laborers. It also highlights an important tension in our thinking about these kinds of cases: on the one hand, because there appears to be something wrong with these kinds of labor environments, it seems like they should be abolished; on the other hand, it seems that doing so would be even worse for the very people we are trying to help. An important part of understanding this tension involves determining whether and how firms can wrong their employees. A related issue concerns how consumers should respond to firms when firms do wrong their employees. Both issues are interesting and important, but in the chapter that follows I will focus my attention on the behavior of the firm. In the first two sections I address questions related to what we might broadly call the exploitation of labor: “How are laborers wronged?” and “What do firms owe their employees?” Having addressed these questions, I focus on responsibility in the third section, asking “Who is responsible when workers are wronged?”
It is common to say that firms can wrong their employees by exploiting them. But what do we mean by “exploitation”? Philosophers and economists have long quibbled about the details, but most agree that exploitation broadly involves “taking unfair advantage.” While this definition seems correct, it doesn’t tell us very much. We still need to know what it means to “take advantage of” someone and what it means to treat them “fairly.” Rather than focusing on philosophical definitions of exploitation, we might make a better start by simply considering the particular ways in which workers can be wronged before revisiting the exploitation question at the end of the chapter.1
There are many unique ways that people can be wronged, but the ways in which they can be wronged as a laborer fall into two broad categories. Laborers may be inappropriately compensated for their work, or they may work in unacceptable conditions. So, while workers might be cheated by their employers, paid unfairly low wages, denied benefits or wages to which they are entitled, and even enslaved, each of these wrongs involves an extraction of labor from the employee that is not appropriately compensated. Workers may also be subject to physical and psychological abuse, unreasonably dangerous or uncomfortable production processes, forced to work long hours, and exposed to dangerous chemicals. Each of these represents a potentially morally unacceptable working condition. Let’s consider each source of wrong in turn.
Unjust wages
There is significant disagreement about what counts as appropriate compensation, but various theories about fair wages often take one of two broad approaches. According to “distributive” theories, laborers are paid enough when they receive a certain proportion of the mutual gains that are created by the combination of their labor and the firm’s capital. So, for example, a distributive theory might claim that if Alice employs Bob to work in her factory and pays him 10 percent of the gains while keeping 90 percent for herself, Alice wrongs Bob by unfairly distributing the gains. Various distributive approaches offer different criteria for what counts as a fair division of gains. Some claim a fair division is an equal division. Others that a fair division is proportional to the individuals’ efforts or contributions. Still others that any division is fair, provided the ways in which the parties bargain for their shares do not violate certain procedural requirements. For example, proponents of this last approach may claim that distributions compromised by fraud and coercion are unfair.2
p.492
A second approach to just wages can be called an “absolute” approach. According to these theories, Alice ought to ensure that Bob is paid a certain absolute amount, regardless of the total profit amount. Often these theories describe this amount as a living wage, or an amount compatible with human flourishing. As with distributive accounts, absolute accounts provide many different criteria for what counts as a living wage.3
Finally, these two broad approaches may also be combined in a “hybrid” account. For example, a hybrid account might claim that Alice is normally required to pay Bob a living wage, but if the firm’s profits are too low to support this wage without it going out of business, then Alice must at least provide Bob with a fair distribution of the firm’s profits. This kind of hybrid account prioritizes an absolute level of pay and uses a distributive criterion only when the absolute requirement cannot be satisfied. Alternatively, a hybrid account may prioritize distributive criteria by requiring Alice to pay Bob a certain proportion of the profits but never less than a living wage.
Unjust conditions
Laborers may also be wronged if the labor environment is unacceptably hazardous. Workplace hazards may be necessary features of the work itself, or they may be caused by the firms and managers (though, as we will see, this distinction is not always sharp). Certain jobs—oil rigging, crab fishing, and mining, for example—are hazardous because of the activities they require and the risks to which they expose laborers. Although firms can implement risk management procedures to mitigate these hazards, given current technological constraints, the tasks required for some jobs mean that their hazards are “necessary hazards” and cannot be completely removed.
Other jobs may be harmful to workers not because of the nature of the work, but because of the actions of firms or managers. For example, workers may be exposed to physical and psychological abuse, or unsafe working conditions may be introduced through managerial negligence. These are “unnecessary” hazards because they are not caused by the nature of the work.
Although distinguishing these two sources of hazard is a helpful heuristic, many of the real-world risks to which laborers are exposed involve a combination of both. Consider leather tanning, which involves the use of many dangerous substances. The most popular method of tanning uses chromium (CrO2 and Cr2O3), which causes chronic health problems and leads to a significantly elevated risk of cancer (EPA 1998). Overall, tanners exposed to chromium are twice as likely to suffer some form of morbidity than persons with similar demographic traits that do not work in leather tanning (Rastogi et al. 2008). While chromium is necessary for tanning many kinds of leather, contact with chromium is avoidable (and so unnecessary). In high-income countries industry regulation significantly limits exposure to chromium, reducing the risks associated with leather tanning. However, these controls are expensive and are rarely utilized by tanners in low-income markets. So, although some of the hazards to which workers are exposed are simply necessary hazards of the job; others are unnecessarily introduced. In many—and perhaps most—cases, the nature of the work involves hazards that, once necessary, are now removable, albeit for a cost. Such hazards, while not needlessly introduced by managers, are still avoidable and so fall somewhere between purely necessary and unnecessary hazards.
So, laborers may be wronged while working either because they are not paid appropriately, or because the conditions in which they labor are unacceptably harmful. Furthermore, wages may be too low in either an absolute or a distributive sense. Hazardous working conditions may be necessary because of the nature of the work, or unnecessary because they are needlessly introduced by the firm. However, these latter two distinctions are not without exception. Some hybrid accounts of fair wages combine both absolute and distributive criteria and some workplace hazards are necessary aspects of the job, but the level of risk can be mitigated at a cost.
p.493
Some philosophers and economists also challenge the very first distinction between wages and working conditions. While accepting that pay can be distinguished from working conditions, they argue that there is no meaningful moral difference between the two because workers can make trade-offs between both kinds of goods. Standard economic theory endorses the following:
Fungibility thesis. Working conditions, monetary wages, and other employment benefits are fungible goods that can always be traded off against one another: increases or decreases in the amount of one of these goods can be compensated for by increases or decreases in the amount of another.
According to standard economic theory, rational laborers select a compensation bundle—a mix of these goods—that provides them with a combination of benefits, working conditions, and wages that maximizes their overall utility. For example, offshore oil rig engineers have higher average salaries than similarly skilled laborers in other industries. This differential is explained by the high risks associated with the work: the offshore workers are compensated for their hazardous work environment with higher pay. Those who choose this work prefer lower safety and higher pay to higher safety and lower pay.
If these goods really are fungible, then the proper subject of moral evaluation is the bundle that workers consume. Rather than looking at pay, benefits, and working conditions separately, we should evaluate the total package that workers receive from employers. This does not mean that we no longer need to consider both working conditions and wages; after all, we need to know the level of each to calculate workers’ total compensation bundles. But it does mean that we cannot conclude employees are mistreated simply from the fact that their compensation in one of these dimensions is low, for they may be highly compensated in another dimension and, consequently, sufficiently compensated overall. Although this taxonomy answers the question of how workers can be wronged, it does not answer the more pressing question of what levels of these goods firms owe employees.4
What do firms owe their employees?
The discussion so far has supposed that firms owe their employees something; that certain working conditions and wage rates are morally unacceptable. However, before considering what acceptable conditions and wages might be, it is worth considering the radical—yet rather widespread—claim that firms owe their employees nothing.
The laissez faire approach
More carefully stated, the claim is that if a laborer’s association with a firm is voluntary, then the firm is not required to ensure that the employee works under any specific working conditions or receives any particular distribution (or absolute level) of pay.5 This laissez faire approach claims that voluntary association is both necessary and sufficient for permissible labor contracts. The approach comes with two important caveats. The first is that while there may be no restrictions on the content of voluntary contracts that firms and laborers can enter into, once a contract has been established, then firms are morally and legally obliged to respect the terms of the contract. The second is contained in the conditional statement above: the laborer’s association must really be voluntary. While the first condition is relatively straightforward to understand, the second requires some unpacking.6
p.494
There is disagreement about what features must be present in order for a contract to be voluntary. On the “minimal understanding,” a laborer’s association is voluntary when he or she “tokens consent,” that is, when the person ostensibly expresses consent to the contract. The minimal understanding of voluntariness is not very attractive since it would allow, inter alia, fraudulent contracts and contracts with mentally incapacitated persons. Not only do many philosophers reject the minimal account, but there is a strong legal precedent that contracts made with those who lack sufficient mental capacities, or which are predicated on deceit, are unenforceable.
A broader notion of voluntariness requires “informed consent”: contractees must not only token consent, but this consent itself must also be “informed.” The conditions of informed consent vary, but it is generally taken to require that the contractee is mentally competent, informed of the relevant facts, risks, and consequences of the contract, and that she understands these elements.
Although the informed consent understanding prohibits many intuitively undesirable contracts, it still permits contracts made under coercion. Suppose Bob threatens to kill Alice unless she agrees to pay him 70 percent of her future earnings. Even if Alice tokens consent to Bob’s offer and fully understands its terms and implications—that is, even if she tokens her informed consent—few would call her agreement voluntary. So perhaps voluntariness is “informed consent without coercion.” What qualifies as coercion varies under different accounts, but almost all accounts accept that coercion includes imminent threats to wrongfully harm the contractee by the person offering the contract.7 In order for the laissez faire approach to be plausible, it seems it should adopt an understanding of voluntariness that equates it to informed consent without coercion, as outlined above. Narrower understandings would allow, for example, contracts for slavery signed under a threat of death, introduced by the other party. Such contracts are clearly impermissible. If we take this approach, the laissez faire approach makes the following claim:
Laissez faire duties. A firm’s only duty to employees is to honor voluntary contracts made with these persons. Contracts are voluntary if and only if they are made with the uncoerced informed consent of all parties.
Whether this account is true depends on whether there are plausible accounts of additional duties that firms have towards their employees. This brings us to accounts that argue firms have more substantive duties towards their employees than simply honoring voluntary agreements.
Substantive duties
While those who endorse the claim that firms have more substantive duties toward their employees agree that voluntariness is necessary for contracts to be morally permissible, they do not think it is sufficient for permissible contracts. Instead they argue that firms must provide minimally acceptable working conditions, or minimal levels of remuneration, or both. If the fungibility thesis outlined in the previous section is true, then provided firms do have substantive duties, they must provide workers with an overall bundle that satisfies a certain criterion of fair compensation.
The details of firms’ specific substantive obligations will, in theory, depend on the truth of the fungibility thesis. However, in practice, the thesis applies primarily to those who either earn very high salaries, but work in very hazardous conditions, or who work in excellent conditions, but are paid very little. Sweatshop workers, whose situations motivate charges of exploitation, are paid little and work in poor conditions. The form of firms’ substantive duties towards these workers is likely to be little affected by the truth of the fungibility thesis, since both components of their compensation package are low. Since the primary focus of the chapter is on severely disadvantaged workers, I shall consider firms’ possible substantive duties under the simplifying assumption that the fungibility thesis is true.
p.495
If firms’ duties to their employees extend beyond honoring voluntary contracts and include a duty to provide a minimally acceptable compensation bundle, then what exactly is a minimally acceptable bundle? The assumption that the fungibility thesis is true means that we can focus directly on monetary compensation, since any workplace hazards can be traded off for higher pay. As we saw in the previous section, there are two general approaches to fair pay. According to the absolute approach, workers are paid fairly when their compensation is equal to or exceeds an absolute level, usually defined as a “living wage”; on the distributive approach, they are paid fairly when the benefits produced by their work are fairly distributed between the firm and the employee. Also, of course, these two approaches may be combined in a hybrid account.
Absolute criteria for fairness
Ruth Sample argues that firms are obliged to honor a person’s value when they engage in mutually beneficial cooperation. She claims firms can fail to honor a person’s value by, inter alia, “neglecting what is necessary for that person’s well-being or flourishing” (Sample 2003; see also Snyder 2008 and Arnold and Bowie 2003). Sample’s appeal to basic needs makes hers an absolute account of fair compensation. The basic needs approach makes the following claim:
Basic needs duties. In addition to respecting voluntary contracts, firms must also ensure that the contracts they make with their employees provide wages that are sufficient to meet employees’ basic needs.
Although this account is intuitively attractive—we all want to flourish and have our basic needs met—an appeal to basic needs appears neither necessary nor sufficient for fair compensation.
It is not sufficient because it seems that transactions between two persons who have their basic needs met may still be unfair. If meeting basic needs were sufficient for fair compensation, then any wage that met basic needs (and satisfied the laissez faire duties) would be fair. Yet, it seems possible that middle-class managers may still be paid unfairly despite the fact that their basic needs are met. Meeting basic needs is also not necessary for a transaction to be fair. In some cases firms cannot afford to pay a wage that meets their employees’ basic needs. Of course, what exactly firms can and cannot “afford,” is a contentious issue. But Sample herself includes this caveat: “if the only mutually beneficial transaction possible is one in which [basic] needs cannot be satisfied, then such interaction is not exploitative” (Sample 2003: 75).
A final (relatively minor) problem for such accounts is that they are not easily applied to specific transactions. Suppose that in Alice and Bob’s community basic needs can be met by $10,000 per year. If Alice buys an apple from Bob, surely she does not need to pay him $10,000 in that single transaction. But how much should she pay him? Perhaps she should pay him $10,000 divided by the number of apples Bob can sell in a year. But what if Bob sells few apples? What if he sells many? These issues do not directly undermine the flourishing account. It can still be applied to employees’ monthly or yearly salaries, since there we may be more interested in the aggregate amounts. But it does mean it may not be useful as a general theory of fair transaction.
p.496
Should we abandon basic needs or flourishing accounts in favor of some other absolute criterion of fair compensation? It is not clear this will help. Not only will alternative absolute accounts inherit many of the above problems, they must also justify any new threshold levels of compensation they claim are essential to fair compensation. However, unlike the claim that we should meet persons’ basic needs, other absolute levels of compensation do not have intuitive appeal. In place of absolute accounts, let us consider the distributive alternative.
Distributive accounts of fairness
Distributive accounts of fair compensation do not claim that firms must provide their workers with any particular level of compensation, but rather that they must distribute the mutual benefits of the worker’s labor. Perhaps the most (in)famous distributive account was championed by Karl Marx, but many other criteria of fair distribution also exist. Here I will outline four: Marx’s account, based on the labor theory of value; an egalitarian distribution; Alan Wertheimer’s (Wertheimer 1996) hypothetical market approach; and Hillel Steiner’s liberal account (Steiner 1984; Steiner 1987; Steiner 2010).
Marx (1887) claims that workers are exploited (and thus unfairly compensated) when the pay that they receive embodies less labor than is embodied in the goods they can consume with their wages. Or, in other words, firms must pay workers an amount of labor equal to what workers provide firms. Marx claimed that the value of a good is determined by the amount of labor required to produce it. So, if it takes Alice five hours to produce a chair, the chair “embodies” five hours of labor. A fair price for her chair is one in which Alice receives either material goods that also embody five hours of labor, or enough money to buy such goods.
The labor theory of value has prima facie appeal: goods that take more time to produce do seem to have greater value—the labor theory of value can explain why a handmade Rolls Royce costs more than a mass produced Honda. Furthermore, Marx’s criterion captures two common intuitions about fairness. First, it is egalitarian, because it requires an equal exchange of labor. Equality is tightly linked with fairness: in a straightforward sense a fair division of cake would suggest an equal division of cake. Second, it incorporates a notion of desert because it claims that persons should be compensated for the work they do. So, some may object to the equal cake example, claiming that it is not compelling in cases where Bob contributes only half as much time to the cake’s production as Alice. Yet, Marx agrees that in such cases the producers’ compensation should reflect their differential inputs. He would claim the cake’s fair price is three units of labor, two are owed to Alice and one to Bob. According to Marx, then, firms have the following duties towards their employees:
Marxist duties. In addition to respecting voluntary contracts, firms must also ensure that the contracts they make with their employees provide wages that can be used to buy goods that embody an amount of labor equal to that expended by the employee.
One interesting aspect of Marx’s account is that it applies to both laborers and firms: while firms pay employees unfairly when they fail to discharge their Marxist duties, employees receive an unfairly high wage if they provide less labor than is embodied in their wages. In contrast, the basic needs approach does not imply that workers act inappropriately when they receive wages that are greater than those required to meet their basic needs.
Although the labor theory of value captures intuitions about equality and desert, it faces some significant challenges. Robert Nozick famously listed many exceptions to the labor theory of value:
p.497
found natural objects (valued above the labor necessary to get them); rare goods . . . that cannot be reproduced in unlimited quantities; differences in value between identical objects at different places; differences skilled labor makes; changes caused by fluctuation in supply and demand; aged objects whose producing requires much time to pass . . . and so on.
(Nozick 1974: 258)
The labor theory of value faces both normative and descriptive problems. Normatively there seem to be examples of goods, like those in Nozick’s list, that we think are of greater value than others, but which embody less labor. The causes of this divergence are varied, but the general problem is that the labor theory does not always map onto our intuitions about what really does (or ought to) have value.
Descriptively, the labor theory of value has a hard time explaining why the prices that goods command on the market diverge from their labor values. Although numerous attempts have been made to solve this “transformation problem,” none has been successful (see Samuelson 1971). These and many other problems have led most economists to reject the labor theory of value and, with it, Marx’s criterion of fair exchange.8
In place of Marx’s labor theory of value, contemporary neoclassical economics uses a subjective theory of value, according to which the values of goods and services are determined by individuals’ preferences. In short, the theory claims that goods are simply worth whatever people are willing to pay for them. While the subjective value theory faces some difficulties, it generally does a better job not only of capturing our normative intuitions about what has value, but also of accurately describing the relationship between prices and values. A second distributive criterion combines Marx’s egalitarian insight with the subjective theory of value:
Subjective egalitarian duties. In addition to respecting voluntary contracts, firms must also ensure that the contracts they make with their employees equally divide the benefits of the employment between the firm and the worker.
In order to understand this duty, we must take a step back to distinguish two concepts central to distributive fairness: the “distribuendum,” and the “fairness criterion.” The distribuendum is the object of distribution, the thing being distributed. On the Marxist approach, the distribuendum is labor time; on the subjective egalitarian approach, it is “benefits.”
The fairness criterion tells us how the distribuendum should be allocated. Although the Marxist and the subjective egalitarian approaches offer different accounts of what is being distributed, they both claim that what is being distributed should be distributed equally. So, the Marxist and subjective egalitarian approaches use the same fairness criterion, but different distribuenda. The Marxist approach fails because its distribuendum, labor time, is problematic. But how, exactly, are we to think about the subjective egalitarian distribuendum of benefits?
The best way to think about how someone benefits from a transaction is to consider the difference between the terms of the actual transaction and the person’s “reservation price.” For example, suppose that Alice wants to buy an apple from Bob. The most Alice is willing to pay for an apple is $2—at any higher price she’d rather keep her money. This means $2 is Alice’s reservation price. Alice is indifferent between buying the apple and not buying the apple at $2; she benefits when she buys the apple for less than $2, and the cheaper the price, the greater her benefit. Bob, on the other hand, would like to sell the apple for as much as possible. Suppose that the least Bob is willing to accept for his apple is $1. This is Bob’s reservation price. Anything less and he’d rather keep the apple. Bob benefits when the apple sells for more than $1 and the greater the price of the apple, the more he benefits. If Bob’s reservation price is $1 and Alice’s is $2, then the only mutually beneficial transactions that are possible are at prices between $1 and $2. At any price outside this range one or the other of the traders would prefer to walk away without transacting. There are many distributions of benefits between $1 and $2. If the price is $1.10 Alice receives 90 percent of the benefits; if it is $1.55 Bob receives 55 percent of the benefits.
p.498
The subjective egalitarian fairness approach claims that the distribuendum—what is being distributed—are the benefits each receives over her or his reservation price. Its fairness criterion—how things should be distributed—is egalitarian. In the apple case, the subjective egalitarian criterion claims that the fair price is $1.50. This appeal to equality makes the approach attractive. Unfortunately, it also faces a serious problem.
Consider a case where Alice’s car breaks down on her way to an important interview for a lucrative job, one she is likely to be offered. The high probability of years of highly paid work is quite desirable for Alice. Suppose that in order to make it to the interview she would be willing to pay up to $20,000 to have her car repaired. Suppose that Bob, the auto mechanic, discovers that the repair is simple to perform—the problem is a broken fan belt. Bob normally charges $150 to repair the fan belt and his reservation price is $50. In this case, both Alice and Bob benefit at any price between $50 and $20,000. According to the subjective egalitarian criterion, the fair price for the repair is $9,975.9 This seems unfair. The problem generalizes to a wide number of cases. Whenever one transactor attaches extreme importance to the transaction’s occurrence, and consequently has an atypically high (or low) reservation price, the subjective egalitarian criterion delivers counterintuitive results.
There is good reason to believe that any plausible fairness criterion must at least select a fair transaction point (or range) that is mutually beneficial. To see why this is so, consider a fairness criterion that places the fair division point outside the interval of mutually beneficial transactions. Suppose that the “fair” price for the apple in the prior example was $0.50, which is below Bob’s reservation price of $1. Suppose Bob is both rational and morally motivated in the sense that he wants to avoid acting wrongly. Bob has two options that involve no wrongdoing. He can either transact fairly, or he can refrain from transacting at all. There is, after all, no reason to think he is obliged to transact. Given a choice between transacting at $0.50 and not transacting, Bob will not transact. The $0.50 price is below his reservation price and will make him worse off. A fairness criterion that requires either harm to one party or no transaction at all seems absurd; after all, there are many possible transactions between the reservation prices that would be better for both Alice and Bob. Yet, these would be prohibited if the fair division point lies outside the mutually beneficial interval. In the case of sweatshop workers, part of our concern is that they are treated unfairly, but a large portion of our worry also stems from the fact that they are very poor. Surely we should avoid fairness criteria that may incentivise non-transactions and, in so doing, leave these vulnerable persons even worse off than they would have been if they had traded. Thus, any plausible fair division point must lie between the transactors’ reservation prices (see also Ferguson and Steiner 2018).
If a fair transaction is one in which the joint benefits from transacting are fairly distributed and the most obvious fair division point—an equal division of benefits—is not always plausible, then what is a fair division of benefits? And why does the egalitarian division give odd results in certain cases? One problem for the egalitarian division is that although it takes into account individuals’ demands for a particular good or service, it is insensitive to how rare these goods are, or how many other persons would also like to have them. The approach is sensitive to the demand for goods, but not the supply.
Alan Wertheimer’s fair market value (FMV) fairness criterion addresses this problem by claiming that fair transactions and, by extension, fair employment contracts, are those that fulfill the following duties:
p.499
FMV duties. In addition to respecting voluntary contracts, firms must also ensure that the contracts they make with their employees provide wages that are equivalent to those the worker would receive in a hypothetical fair market environment.
Markets incorporate information about both supply and demand. Wertheimer claims that fair transactions are those that take place at the “fair market value . . . the price that an informed and unpressured seller would receive from an informed and unpressured buyer if the [good] were sold on the market” (Wertheimer 1996: 230). At such prices, he claims, neither party “takes special unfair advantage of particular defects in the other party’s decision-making capacity or special vulnerabilities in the other party’s situation” (Wertheimer 1996: 232).
Wertheimer’s appeal to the market is compelling. The fair price for goods and services depends both on how rare they are and how much others desire them: if everyone really wants something that is relatively rare, then it is only fair that the person who receives that thing should give up many other goods in order to get it. Furthermore, FMV avoids the problems of the subjective egalitarian approach because it abstracts from individual desires and circumstances to appeal to aggregate demand (and supply). So, in the interview example, Alice’s circumstance-driven preferences mean she has a very high reservation price and, consequently, that the fair price for repairing her fan belt is $9,975. The story also mentions that Bob “usually” charges $150 to replace a fan belt. Assuming that this is the price Alice would have to pay in a hypothetical market environment, then, according to FMV, the fair price for repairing the fan belt is $150. Not only is this a more intuitively plausible fair price, it is a price at which Bob does not take “special advantage” of Alice’s vulnerability. Despite its attractiveness, FMV also faces problems.
It is unclear how the fair market value is to be determined. The most obvious hypothetical market environment is a perfectly competitive market, characterized, inter alia, by an infinite number of buyers and sellers, perfect information, and an absence of externalities and barriers to entry. While these idealized features make perfectly competitive markets useful for economic theory, they also mean that idealized markets differ significantly from real world markets. This difference is so stark that many situations we would like to evaluate with the FMV criterion simply would not exist in perfect competition. Indeed, the very idea of a fair division of benefits cannot arise in perfectly competitive markets, because in perfect competition all agents are “price takers” and there is no room for bargaining. In other words, all transactions that occur are those in which the most Alice is willing to pay is exactly equal to the least Bob is willing to accept and thus there are no benefits to divide. So it seems Wertheimer’s approach cannot appeal to hypothetical markets. Nor can it appeal to actual markets. Using actual markets as the fairness baseline would preclude criticism of existing market prices because the fair price would also be the actual market price. If Wertheimer’s hypothetical markets are neither actual markets, nor perfectly competitive markets, then what are they?
Wertheimer’s claim that the fair market value is a price that informed and unpressured agents would receive suggests his hypothetical markets would be similar to actual markets, but would exclude intuitively unfair circumstances, such as personal duress and price gouging during disasters. It is the influence of these kinds of features more than the influence of perfect information or barriers to entry that Wertheimer wants to prohibit from effecting the market price. But what circumstances that influence actual market prices should be excluded when we calculate the fair hypothetical market prices? Unfortunately, Wertheimer does not tell us.
One plausible answer to this question is provided by Hillel Steiner’s liberal account of distributive fairness. According to Steiner, prices are unfair when one party benefits less and the other party benefits more “than they would each have gained had a prior injustice not occurred” (Ferguson and Steiner 2018). This claim implies that employers have the following liberal duties towards their employees:
p.500
Liberal duties. In addition to respecting voluntary contracts, firms must also ensure the distribution of benefits in the contracts they make with their employees do not reflect prior injustices.
The liberal duties account is similar to Wertheimer’s claim that the fair price should exclude certain kinds of influence. Also, like both the FMV and subjective egalitarian accounts, the liberal duties account retains benefits as the distribuendum. But it also adds a crucial reference to justice, claiming the features that should be excluded from influencing the price in the hypothetical market are just those that a given theory of justice claims are unjust. Thus, what counts as a fair price depends crucially on the theory of justice that we adopt.
Different theories of justice make different claims about what counts as an injustice. This, in turn, means that different theories will also deliver different criteria of distributive fairness in transactions. The dependence of a theory of fair distribution on a prior, and more general, theory of justice means that Steiner’s account can explain why there are so many different intuitions about what counts as a fair price: these disagreements are often caused by prior disagreements about justice. Nevertheless, there is widespread agreement about the injustice of certain acts, such as murder or theft. Also, in many cases of sweatshop labor it does seem that those whom we would describe as exploited are, in fact, clear victims of prior distributive injustice. However, two important questions remain for the liberal duties account.
The first concerns the scope of unfairness. As Steven Walt points out, nearly everyone’s position has been affected in some way by prior injustice. If unfair transactions result from these unfair starting points, then nearly all transactions are unfair: “But this is implausible. The presence of exploitation is unquestionable. Its omnipresence is questionable” (Walt 1984). Moreover, liberal duties cannot explain what is particularly upsetting about sweatshop labor, compared with, say, my purchase of an apple in the market. Both the sweatshop worker’s wage and my purchase of the apple may be affected by prior injustice but, unlike the sweatshop case, the apple purchase does not seem to be unfair.
The second question concerns how the fair price is to be calculated. Although the liberal duties approach tells us that a fair wage should exclude the influence of prior injustices, it does not provide specific instructions for how to determine what that wage would have been. This question is further complicated by the fact that absent prior injustices mean that many transactions simply would not take place. For example, suppose Alice’s car is stolen and she needs to rent a car from Bob to use while her old one is replaced. The liberal duties account tells us that Bob should not take advantage of this injustice to gain more than he would have from Alice had her car not been stolen—yet, had Alice’s car not been stolen she would not transact with Bob in the first place. This problem cannot be solved by simply claiming Bob should not charge Alice anything for the car. Not only is this implausible, but this price is ruled out by the liberal account. If Bob were to let Alice have the car for free, then Bob, not Alice, would get less than he would have received from a transaction because of a prior injustice, for presumably Bob could have sold the car to someone else for a price greater than $0.10
It is time to take stock. I began this section by considering whether firms had any obligations beyond the minimal requirement that they respect voluntary contracts with their employees. The laissez faire duties approach denies that firms have any stronger substantive duties to their employees. Since the truth of this negative claim depends on the plausibility of any more substantive duties, I then considered substantive accounts to see whether they could offer convincing rebuttals. Significant problems surfaced with each account. Although these problems do not necessarily mean the laissez faire approach is correct, they do reveal that more work must be done if one wants to claim that it is false.
p.501
Who is responsible when workers are wronged?
The idea that firms do owe their employees more than merely the respect of voluntary contracts has a great deal of intuitive traction. Surely, we might think, there must be some way to ground more substantive duties. Unfortunately, for advocates of substantive accounts, even if we suppose that a plausible substantive account can be developed, another potential problem looms.
Are firms tied by invisible hands?
When workers are wronged because their compensation bundles are too low, who is responsible for this wrong? At first the answer seems obvious. Surely firms are responsible. In a causal sense, this answer is correct. Yet, when we expand our view to consider the broader regulatory and market environments the claim that firms are morally responsible is no longer as compelling. Firms may argue that they are merely middlemen in an economic chain that connects their investors to consumers. Thus, even if there are substantial obligations to provide workers with a certain compensation bundle, firms may not bear any responsibility for failing to discharge this duty. The argument is similar to the laissez faire duties claim, but it differs in an important way. Rather than attempting to establish that there are no substantive duties to workers, this argument claims that even if such duties do exist, firms are not responsible for fulfilling them.
The fiduciary argument. Firms have a fiduciary duty to their investors which dictates that firms maximize profits, subject to legal constraints. This duty trumps all other duties. Firms’ profits are primarily dictated by a) the regulatory environment in which they operate, b) the price of inputs used to produce their product, and c) consumers’ demand for their product. Thus, firms exercise control only over the efficient conversion of inputs to finished products: both the price of finished products and the price of inputs are dictated by the market. Any substantive duty firms might have to pay their workers more is either superfluous, or trumped by fiduciary duties.
The argument claims that because firms cannot control input prices or output prices, they exercise free choice only over the conversion of inputs to outputs. Yet, their fiduciary obligations dictate that this conversion must be done in a profit maximizing (and legal) way. Thus, firms have no room to set wage rates. If we assume that what one ought to do is constrained by what one can do, then firms cannot be held responsible for failing to discharge substantive duties because they can do only one thing: maximize profits to fulfill their fiduciary obligation. If workers ought to be paid more, as defenders of substantive duties claim, then wage increases must come from a change in one of the other variables: input costs, the regulatory environment, or consumer preferences. This means that if substantive duties do exist, the responsibility for fulfilling them lies with consumers, shareholders, and politicians (on this point, see also Powell and Zwolinski 2011).
The argument is compelling and forceful. However, it makes a number of questionable assumptions. First, although firms have limited choices in perfectly competitive markets, existing markets are not perfectly competitive. Firms may have greater latitude for choice in the conversion of inputs to outputs than the argument admits. Second, since real-world firms operate outside perfectly competitive markets, they can, at times, influence input costs and sales price through supply chain integration and advertising. Nevertheless, even when firms can influence these factors, the claim that firms have an overriding fiduciary obligation to maximize profits dictates that any gains from the choices that are available to them should be passed on to the investor. Thus, the premise that does the real work in the argument is the claim that firms’ fiduciary duties are overriding.
p.502
Perhaps the most famous defense of this crucial premise is Milton Friedman’s claim that “the social responsibility of business is to increase its profits”—to do otherwise, he argues, is tantamount to “spending someone else’s money for a general social interest” (1970: 33). Friedman allows few constraints on this responsibility, claiming a firm must maximize profits “so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud” (Friedman 1962: 133). Elsewhere he makes the more substantial concession that profit maximization should be constrained not only by law but also by rules “embodied in ethical custom” (Friedman 1970: 33).11 Though intended as mere qualifying caveats, these comments are important, for they show that fiduciary obligations are secondary to at least some moral considerations.
As noted, the core of the question is whether fiduciary duties outweigh (potential) substantive duties. If fiduciary duties are overriding, then firms’ first obligation is to shareholders. If they are not overriding, then there is room for substantive duties towards employees. There is a tension in Friedman’s claims. On one hand, he asserts “the manager is the agent of the individuals who own the corporation . . . and his primary responsibility is to them” (Friedman 1970: 33). On the other hand, he allows that this responsibility is subject to the aforementioned constraints. If fiduciary obligations are subject to other moral constraints, then fiduciary obligations cannot be primary, for other moral constraints will “kick in” first. But if fiduciary obligations are truly primary, then they will outweigh any potential moral side constraints, making the latter superfluous.
One way to make sense of Friedman is to see him as distinguishing two kinds of moral considerations: those that are prior to fiduciary obligations (laws and ethical customs) and those that are not (social responsibilities). Consider two possible acts: murder and a donation to environmental causes. It seems Friedman would claim that managers’ duty to maximize profits falls “between” these acts. That is, profit maximization is subject to a constraint not to murder, but outweighs donation to environmental causes.
The idea that certain moral considerations outweigh fiduciary obligations and certain moral considerations do not is plausible. Consider a similar case: Alice promises to return Bob’s library book for him by the end of the day. Surely Alice should not return the book if somehow by doing so she were to cause serious injury to another person. This injury would outweigh her promise to Bob. And it is just as reasonable to say that she is not permitted to break her promise to Bob in order to have a coffee with a friend. Although socializing over coffee is a good thing, unlike serious injury it does not override her obligation to return the book. Similarly, then, Friedman may claim fiduciary duties are outweighed by some moral considerations and not others.
However, this means that the fiduciary argument cannot simply assume that potential substantive duties are indeed trumped by fiduciary obligations. Rather, this claim must be defended. Without additional premises, the fiduciary argument cannot show that fiduciary obligations excuse firms from fulfilling substantive duties to their employees.
Of course, as we have seen, the existence of substantive duties has yet to be fully established. This criticism of the fiduciary argument shows that if substantive duties do exist, then firms are not necessarily excused for failing to fulfill these duties by their fiduciary obligations. While the fiduciary argument fails to establish its strong conclusion, it does draw our attention to the fact that firms are not the only parties with some control over remuneration. Consumer demand and market prices for inputs also shape employee compensation.
p.503
Concluding remarks
So, should you refrain from buying a brand X phone? Can we make sense of the idea that a corporation’s production of brand X wrongs its workers? In an attempt to answer these questions I considered first the ways in which workers might be wronged; second what firms owe their employees; and, finally, whether firms are responsible when employees are wronged.
In the first section I noted employees may be wronged when their working conditions are wrongfully harmful or when their wages are unfairly low. Furthermore, if the fungibility thesis is true, we can combine these concerns to say that employees are wronged when their compensation bundles are too low.
In the second section I considered how a compensation bundle could be “too low.” There I canvassed a number of possibilities. The laissez faire approach maintains that there should be no restrictions on the contracts firms offer their employees, provided they are made voluntarily; firms wrong their employees only when they fail to honor their contracts. I also considered various substantive approaches including basic needs, Marxist, and subjective egalitarian duties. Unfortunately, each of these faces serious challenges. Though two nuanced approaches—Wertheimer’s hypothetical market account and Steiner’s liberal account—fare better, they both have trouble specifying the fairness baseline. So, while there are no strong arguments against the theoretical possibility of substantive duties, none of the accounts we surveyed escaped unscathed.
In the final section I considered a question about responsibilities: if substantive duties do exist, are firms responsible for fulfilling them? The fiduciary argument claims they are not. However, in order for the fiduciary argument to work it must show that fiduciary obligations specifically trump potential substantive duties. Thus, it is not clear that firms can escape responsibility for fulfilling these duties, should they be shown to exist.
So, what about exploitation? Exploitation involves taking unfair advantage. Yet without a plausible account of substantive duties, it is difficult to show that firms actually treat their employees unfairly. And if we cannot show firms treat their employees unfairly, we cannot show that they exploit them. Nevertheless, many people, myself included, have the intuition that some firms do exploit their workers. The task still at hand is to offer a plausible account of substantive duties that underlie this intuition in order to show convincingly that workers’ wages or working conditions are exploitative.
Essential readings
The most sophisticated defenses of labor conditions commonly described as sweatshops have been written by Matt Zwolinski. In two articles “Sweatshops, Choice, and Exploitation” (2007) and “The Ethical and Economic Case Against Sweatshop Labor: A Critical Assessment” (2011), written with Benjamin Powell, one encounters a strong case for the laissez faire approach. For discussion of how we could wrong persons even as we benefit them see my article “The Paradox of Exploitation” (Ferguson 2016). Helpful discussions of Kantian approaches to fair wages and exploitation can be found in Denis Arnold and Norman Bowie’s article “Sweatshops and Respect for Persons” (2003), as well as Ruth Sample’s book Exploitation: What it is and Why it is Wrong (2003). The core of Karl Marx’s theory of exploitation can be found in his Capital, volume I, Chapter 9, Section 1 (1887). Important discussions of the Marxist theory are provided in John Roemer’s book, A General Theory of Exploitation and Class (1982), and in G.A. Cohen’s article “The Labour Theory of Value and the Concept of Exploitation” (1979). An interesting but often overlooked article that relates the Marxist and Kantian approaches is Jonathan Wolff’s “Marx and Exploitation” (1999). Perhaps the most prominent contemporary discussion of exploitation is Alan Wertheimer’s book Exploitation (1996). An excellent anthology comprising contemporary and historical accounts of this subject is Modern Theories of Exploitation, edited by Andrew Reeve (1987).
p.504
For further reading in this volume on the obligations of owner and managers to employees, see Chapter 28, Employee ethics and rights. On property relations in general, see Chapter 18, Property and business. For an account of business ethics that prioritizes open and free communication as a basis for just practices, see Chapter 12, Integrative Economic Ethics: concept and critique. On the fiduciary duties of managers to shareholders, see Chapter 11, Stakeholder thinking. On the practical ethics of management, see Chapter 27, The ethics of managers and employees.
Notes
1 Two anthologies with many articles on theories of exploitation are Reeve (1987) and Neilson and Ware (1997).
2 Distributive accounts based on equal division are endorsed by Marx (1887), Roemer (1982), Moore (1973) and Levine (1988). Purely procedural distributive accounts are defended by Nozick (1974) and Steiner (1984, 1987). A bargaining-based distributive account is defended in Ferguson (2013). A discussion of distributive accounts can be found in Wertheimer (1996).
3 Many absolute accounts exist, but three that explicitly reference labor and exploitation issues are Goodin 1987, Sample 2003 and Arnold and Bowie 2003.
4 Both Maitland (1996) and Arnold (2010) provide helpful overviews of some of the empirical aspects of sweatshop working conditions, as well as discussions of their normative importance.
5 See Zwolinski (2007) for an elegant defense of the laissez faire approach.
6 For an extended discussion of voluntariness see Olsaretti (2004). Wertheimer (2003) contains a helpful discussion of consent.
7 Detailed discussions of coercion can be found in Nozick (1969), Wertheimer (1987), and van der Rijt (2011).
8 Many “equal exchange” accounts are Marxist in spirit, yet independent from the labor theory of value, as in Roemer (1982) and Veneziani (2013). Cohen (1979) argues the labor theory of value is unnecessary for Marxist accounts. The relationship between Kantian and Marxist accounts is discussed in Wolff (1999). Reiff (2013) defends a cost of production account.
9 Any transaction in the interval $50 to $20,000 is mutually beneficial; the midpoint of this interval is ($20,000-$50) / 2 = $9,975.
10 Ferguson (2013) suggests Steiner’s problems can be mitigated by focusing on awareness of prior injustice and counterfactual bargain power.
11 This remark about ethical custom is odd since constraints based on ethical custom seem to open the door to the social interests he wishes to exclude.
References
Arnold, D. (2010). “Working Conditions: Safety and Sweatshops,” in G. Brenkert and T. Beauchamp (eds), The Oxford Handbook of Business Ethics. New York, NY: Oxford University Press, 628–653.
Arnold, D. and N. Bowie (2003). “Sweatshops and Respect for Persons,” Business Ethics Quarterly 13:2, 221–242.
Cohen, G. (1979). “The Labour Theory of Value and the Concept of Exploitation,” Philosophy & Public Affairs 8:4, 338–60.
Environmental Protection Agency (EPA) (1998). “Toxicological Review of Hexavalent Chromium,” Summary Information on the Integrated Risk Information System, CAS No. (18540-29-9). Washington, DC: US Environmental Protection Agency.
p.505
Ferguson, B. (2013). “The Paradox of Exploitation: A New Solution.” PhD thesis, London School of Economics and Political Science.
Ferguson, B. (2016). “The Paradox of Exploitation,” Erkenntnis 81:5, 951–972.
Ferguson, B. and H. Steiner (2018). “Exploitation,” in S. Olsaretti (ed.),The Oxford Handbook of Distributive Justice. Oxford: Oxford University Press.
Friedman, M. (1962). Capitalism and Freedom. Chicago, IL: University of Chicago Press.
Friedman, M. (1970). “The Social Responsibility of Business is to Increase its Profits,” The New York Times Magazine Sept. 13, 1970, 32–33, 122–126.
Goodin, R. (1987). “Exploiting a Situation and Exploiting a Person,” in A. Reeve (ed.), Modern Theories of Exploitation. London: Sage, 166–97.
Levine, A. (1988). Arguing for Socialism. London: Verso.
Maitland, I. (1996). “The Great Non-Debate over International Sweatshops,” in T. Beauchamp and N. Bowie (eds), Ethical Theory and Business. British Academy of Management Annual Conference Proceedings. Englewood Cliffs: Prentice Hall, 593–605.
Marx, K. (1887). Capital: A Critique of Political Economy. Moscow: Progress Publishers.
Moore, B. (1973). Reflections on the Causes of Human Misery. Boston, MA: Beacon Press.
Neilson, K. and R. Ware (eds) (1997). Exploitation. New York, NY: Humanities Press.
Nozick, R. (1969). “Coercion,” in S. Morgenbesser, P. Suppes and M. White (eds), Philosophy, Science, and Method: Essays in Honor of Ernest Nagel. New York, NY: St. Martin’s Press, 440–472.
Nozick, R. (1974). Anarchy, State, and Utopia. New York, NY: Basic Books.
Olsaretti, S. (2004). Liberty, Desert and the Market. Cambridge: Cambridge University Press.
Otsuka, M. (2003). Libertarianism Without Inequality. Oxford: Oxford University Press.
Powell, B. and M. Zwolinski (2011). “The Ethical and Economic Case Against Sweatshop Labor: A Critical Assessment,” Journal of Business Ethics 107:4,449–472.
Rastogi, S., A. Pandey and S. Tripathi (2008). “Occupational Health Risks Among the Workers Employed in Leather Tanneries at Kanpur,” Indian Journal of Occupational & Environmental Medicine 12:3, 132–135.
Reeve, A. (ed.) (1987). Modern Theories of Exploitation. London: Sage.
Reiff, M. (2013). Exploitation and Economic Justice in the Liberal Capitalist State. Oxford: Oxford University Press.
Robinson, J. (1962). Economic Philosophy. London: Transaction.
Roemer, J. (1982). A General Theory of Exploitation and Class. Cambridge, MA: Harvard University Press.
Sample, R. (2003). Exploitation: What it is and Why it’s Wrong. Lanham, MD: Rowman & Littlefield.
Samuelson, P. (1971). “Understanding the Marxian Notion of Exploitation,” Journal of Economic Literature 9:2, 399–431.
Sandel, M. (1998). “What Money Can’t Buy: The Moral Limits of Markets,” in The Tanner Lectures on Human Values. Oxford University: May 11–12.
Satz, D. (2010). Why Some Things Should Not Be For Sale: The Moral Limits of Markets. Oxford: Oxford University Press.
Snyder, J. (2008). “Needs Exploitation,” Ethical Theory and Moral Practice 11:4, 389–405.
Steiner, H. (1984). “A Liberal Theory of Exploitation,” Ethics 94:2, 225–41.
Steiner, H. (1987). “Exploitation: A Liberal Theory Amended, Defended and Extended,” in A. Reeve (ed.), Modern Theories of Exploitation. London: Sage, 132–48.
Steiner, H. (2010). “Exploitation Takes Time,” in J. Vint, J.S. Metcalfe, H.D. Kurz, N. Salvadori and P.A. Samuelson (eds). Economic Theory and Economic Thought: Essays in Honour of Ian Steedman. London: Routledge, 20–29.
van der Rijt, J. (2011). The Importance of Assent: A Theory of Coercion and Dignity. Dordrecht: Springer.
Veneziani, R. (2013). “Exploitation, Inequality, and Power,” The Journal of Theoretical Politics 25:4, 526–545.
Walt, S. (1984). “Comment on Steiner’s ‘A Liberal Theory of Exploitation,’” Ethics 94:2, 242–47.
Wertheimer, A. (1987). Coercion. Princeton, NJ: Princeton University Press.
Wertheimer, A. (1996). Exploitation. Princeton, NJ: Princeton University Press.
Wertheimer, A. (2003). Consent to Sexual Relations. Cambridge: Cambridge University Press.
Wolff, J. (1999). “Marx and Exploitation,” The Journal of Ethics 3:2, 105–20.
Zwolinski, M. (2007). “Sweatshops, Choice, and Exploitation,” Business Ethics Quarterly 17:4, 689–727.