2.1. Introducing the Fairness Argument

The Fairness Argument holds that millionaire inequalities are morally objectionable because they are unfair, and that we should rectify such inequalities because of this unfairness. This argument draws on the idea that it’s unfair for some individuals to receive less favourable remuneration than others simply in virtue of factors that are morally arbitrary. Female millionaires who are paid less generously than their male counterparts simply because they are women are disadvantaged by a morally arbitrary factor, their gender, and are therefore treated unfairly.

We can characterize morally arbitrary factors in one of two related but distinct ways. We can take a factor to be morally arbitrary (i) if it fails to track individual choice and responsibility, or (ii) if it fails to track any consideration that we could plausibly regard as a determinant of one’s entitlements. ^{Footnote 5} When it comes to explicating the unfairness of granting less favourable treatment to members of a particular gender, the second notion of moral arbitrariness seems more appropriate than the first. To see why, imagine a science-fiction scenario in which we are able to choose our gender freely before entering society. There may be good reasons why one might choose to be a woman even if one knows that society is structured in a manner that burdens women with socioeconomic disadvantages. But one’s choice to be a woman does not deprive one of justice-based resources with which to criticize the inegalitarian nature of one’s society. It remains the case that there is no good reason why socio-economic disadvantages should attach to that choice. One’s gender remains morally arbitrary in the second sense distinguished above.

As a property that is morally arbitrary in this sense, one’s gender contrasts with a number of other considerations that might plausibly be regarded as relevant to a moral assessment of distributive outcomes. It’s prima facie plausible that someone who performs a job that is especially socially valuable or arduous should enjoy higher pay than someone who occupies a less valuable or burdensome position. But it’s not at all plausible that someone should enjoy higher or lower pay than others simply in virtue of their gender.

The Fairness Argument is intuitively attractive. However, in the next subsection we demonstrate that it’s defeated by what we shall call the Excess Objection. But before proceeding, let’s note that there are two ways in which millionaire inequalities could be rectified. These can be described using Jennifer Lawrence as our example of a female victim of millionaire inequality and Christian Bale as our example of a male beneficiary of millionaire inequality. The first rectificatory strategy is to raise Lawrence’s pay to the level of Bale’s; the second rectificatory strategy is to lower Bale’s pay to the level of Lawrence’s.Footnote ⁶ We develop the Excess Objection with the first rectificatory strategy in mind, but, at the end of the section, we show how it also impugns the second.

2.2. The Excess Objection

According to the Excess Objection, female millionaires, like their male counterparts, already have more than their just share of resources, and we cannot plausibly maintain that they have a fairness-based claim to even more. The details of this objection, which we endorse, are as follows.Footnote ⁷ In a just society, the excess wealth of the super-rich would be taxed away. The higher rewards for female millionaires recommended by the first rectificatory strategy would, if paid, be immediately expropriated by the state. Moreover, in the absence of just tax-and-transfer arrangements, the super-rich have a moral obligation to relinquish their fortunes voluntarily (e.g. by donating them to charity). The most compelling theories of distributive justice yield the conclusion that the excess wealth of the super-rich should be regarded as morally analogous to stolen loot. The fact that the loot in question is distributed unequally does not provide those who receive less with a fairness-based claim to more.

A variety of familiar reasons support the claim that inordinately large accumulations of wealth among individuals have no place in a just society. ^{Footnote 8} Some egalitarians argue that it’s simply unfair for some people to have so much more than others, given that brute luck inevitably plays a prominent role in determining who gets what. Others explain the injustice of large inequalities by pointing to the pernicious effects that they engender. These egalitarians argue that large economic inequalities often translate into inequalities of power, and that the better-off are often able to dominate or control the lives of the less advantaged in objectionable ways (Schemmel Reference Schemmel2011). Furthermore, large disparities of wealth can undermine the fairness of certain political procedures, and impede the creation of genuinely equal opportunities in education and the labour market (Christiano Reference Christiano and Estlund2012).

But perhaps the most powerful and straightforward objection to large inequalities is that they represent a failed opportunity to rectify severe disadvantage. This objection, which can be effectively mobilized to serve our ends, focuses on absolute deprivation, and makes no reference to the idea that it’s necessarily bad for some to fare worse than others in merely comparative terms.Footnote ⁹ As T.M. Scanlon writes:

It’s no secret that in our world there are not merely ‘some’, but a great many people living under terrible conditions. More than 700 million people survive each day on less than what could be purchased in the USA for $1.90 (United Nations Development Programme 2016: 29). Each year, almost 6 million children die before reaching the age of five, and nearly half of these children lose their lives because they do not have enough food (UNDP 2016: 30; Hug et al. Reference Hug, Sharrow and You2017: 3). Moreover, those involved in movements such as ‘Effective Altruism’ have drawn our attention to the fact that these kinds of deprivations can be ameliorated at little cost. For example, Peter Singer points out that a donation of $250 could provide rehydration therapy to a child who would otherwise die from diarrhoea (Singer Reference Singer2009: 88–89).

It’s these well-known facts about the extreme poverty that disfigures our world, and about the ease with which it can be mitigated, that make the extravagant fortunes of the super-rich so instinctively repellent. It’s also such facts that provide the strongest support for our claim that these fortunes should be seen as analogous to stolen loot. Given the crushing burdens endured by the world’s poorest people, the super-wealthy can have no just claim to the enormous wealth that they have amassed. Like thieves in possession of stolen goods, the rich must relinquish their holdings.

But how, exactly, does the Excess Objection defeat the Fairness Argument? This depends on how we conceive of fairness as a value and on how considerations of fairness interact with the needs of the world’s poor. The Excess Objection might defeat the Fairness Argument in three possible ways, which we can again describe using Lawrence and Bale as our respective examples of victims and beneficiaries of millionaire inequality.

First, we might concede that Lawrence does suffer some morally problematic unfairness. After all, Lawrence is paid less than Bale, and, assuming her lower pay is attributable to her gender, which is morally arbitrary, this inequality in remuneration is certainly unfair. However, because Lawrence (like Bale) already has so much more than most, and can contribute to ameliorating the desperate plight of the world’s poor at such little cost to herself, she is morally required to tolerate this unfairness all-things-considered. Indeed, she is morally required to donate much of the wealth that she already possesses to charity. Raising Lawrence’s pay would enlarge that portion of her holdings that she is morally required to relinquish. On this account of how the Excess Objection defeats the Fairness Argument, each raises distinct moral concerns (of fairness and, say, basic need fulfilment), and Lawrence’s interest in fairness is outweighed by the basic needs of the world’s poor. On this view, fairness is one value to be weighed against others.

The second account is closely related. Again, we begin by conceding that Lawrence suffers some morally problematic unfairness. But we then point out that the severe burdens borne by the world’s poorest individuals, and the radical inequality between these individuals and people like Lawrence, are themselves a product of unfairness. Lawrence has been spared various misfortunes – such as being born into an impoverished country – that have devastated the lives of millions, while being blessed with various advantages, arbitrary from the point of view of morality, that enable her to lead an extraordinarily privileged life. On this account, Lawrence’s interest in fairness conflicts with the interest that the world’s poor have in fairness. And, because the unfairness suffered by the latter is so much greater than the unfairness suffered by Lawrence, and exposes its victims to such catastrophic disadvantages, it’s the claims of the world’s poor that win out. Here, the Excess Objection and the Fairness Argument raise similar (fairness-related) moral concerns of different gravity. And, once again, Lawrence’s interest in fairness is outweighed.

The third account of how the Excess Objection defeats the Fairness Argument is more distinctive. The general idea at the heart of the Fairness Argument is that unfairness obtains when some individuals receive fewer goods than others simply in virtue of factors that are morally arbitrary. This is a plausible claim. But it’s also plausible to hold that the value of fairness is conditional. Sometimes, the fact that a particular distribution of goods is fairer than others provides no reason at all to favour that distribution.Footnote ¹⁰ More specifically, the fact that a particular distribution of goods is fairer than others provides no reason to favour that distribution when the goods in question are not justly held. The Excess Objection reveals that the wealth of millionaires like Lawrence and Bale is not justly held; these millionaires are morally required to relinquish their holdings. Unlike the two previously stated accounts, the present account identifies no morally problematic unfairness in the unequal pay between the two actors. Therefore, while increasing Lawrence’s pay to the level of Bale’s may succeed in rectifying some unfairness, this fact provides no reason to increase Lawrence’s pay. Because of this, there is no morally relevant unfairness to weigh against the claims of the world’s poor. We can characterize this point by saying that the fairness achieved by rectifying millionaire inequality has no intrinsic value.Footnote ¹¹ We can call this the No Intrinsic Value Claim.

While the first two accounts of how the Excess Objection defeats the Fairness Argument hold that the unfairness suffered by Lawrence is outweighed – either by a concern for the basic needs of the world’s poor or by their interest in fairness – the third account that we have just sketched denies that the unfairness suffered by Lawrence constitutes any value that we should place on the scales. It premises this denial on what we can call the Just Holdings Condition, according to which improving the fairness of a distribution is intrinsically valuable only when the distribuenda are justly held. To see whether this condition is plausible, let’s consider two scenarios:Footnote ¹²

We imagine that most readers will share our intuitive judgement that, even if the government reallocation in Slaves is conducive to fairness, this fact provides no reason to favour the reallocation. Fairness has no intrinsic value here and, as Gheaus plausibly notes, one natural explanation of this result is that the distribuenda in question (slaves) are not justly held: ‘the reform cannot cater to any individual’s claim of justice: Nobody should hold slaves, hence nobody has a claim to an opportunity to hold slaves’ (Gheaus Reference Gheaus2020: 16).

Our judgement regarding Pizza is analogous. The fairness of the equal division has no intrinsic value given that the pizza is not rightfully theirs. However, we recognize that some readers may demur from our analysis of Pizza, believing that a fair distribution of the stolen pizza has some intrinsic value. Some readers might hold that, if an equal division really is fairer than Brenda’s original proposal, then that fact provides a reason to favour the former in this case. Those who are willing to countenance this view must abandon the Just Holdings Condition. Moreover, those readers who do not want to jettison the belief that we have no fairness-based reason to favour the reallocation in Slaves must provide an explanation of our conflicting judgements in the two cases.

One candidate-explanation is as follows. While it’s possible to own a pizza permissibly, it’s not possible to own a slave permissibly: owning slaves is always unjust, whereas owning a pizza is not. Because owning slaves is never permissible, distributing slaves fairly cannot be a virtue. By contrast, because it’s possible to own pizza permissibly, improving the fairness of a distribution of pizza can be a virtue, even when the pizza is unjustly held.

If this is an adequate explanation of why a fair distribution of unjustly held goods can be an intrinsic virtue under some circumstances but not under others, then we should concede that there are reasons of fairness to rectify millionaire inequalities, since money (like pizza but unlike slaves) can be held justly. However, we do not find this explanation appealing. While we can certainly distinguish between things that can and cannot be permissibly held, employing this distinction to justify the rectification of millionaire inequalities is puzzling. The fact that a particular good can be justly held might help to establish why ensuring a fair distribution of that good can be intrinsically valuable. But it does not help to establish that a fair distribution of that good can be intrinsically valuable even when the good is not justly held. In the absence of a compelling reason to believe that the fairness of a distribution of unjustly held goods is a virtue, we should reject the current explanation.

An alternative explanation of why a fair distribution of unjustly held goods can be intrinsically valuable under some circumstances but not under others refers to the degree of injustice that individuals suffer in being deprived of their just holdings (Gheaus Reference Gheaus2020: 4). When some people own slaves, others suffer massive deprivations. By contrast, the retention of stolen pizza translates into a trifling burden for the rightful owner. Because slaves are deprived of their most basic entitlements, a fair distribution of slaves cannot be intrinsically valuable. The fairness is robbed of its value by the gravity of the injustice involved. By contrast, we might think that, because the wrongness of retaining stolen pizza is so minor, the intrinsic value of a fair distribution is not extinguished.

Supposing that this is a plausible explanation, the relevant question for present purposes is as follows: does the unfairness of millionaire inequality have more in common with an unfair distribution of slaves or with an unfair distribution of stolen pizza? For reasons that should by now be apparent, our view is that it has more in common with the former (though we do not suggest that being rich is as bad as being a slave owner). The hoarding of excess wealth by the super-rich constitutes the withholding of resources that could otherwise be used to ameliorate the desperate plight of the world’s poor, and the super-rich are morally obligated to donate this wealth to effective poverty-relief organizations.Footnote ¹³ It’s because the hoarding of excess wealth translates into massive deprivations for others, and because the costs of avoiding this outcome are so small, that millionaire inequality cases have more in common with Slaves than Pizza. In virtue of these features, measures to rectify millionaire inequality are strong candidates for fairness-promoting actions that lack intrinsic value.

It’s worth emphasizing at this point that, in our view, the Just Holdings Condition is correct: improving the fairness of a distribution is intrinsically valuable only when the distribuenda are justly held, and the degree of injustice involved when the distribuenda are not justly held is immaterial. We went in search of an alternative defence of the No Intrinsic Value Claim only because we acknowledge that some readers may be reluctant to accept some implications of the Just Holdings Condition. By demonstrating that we can vindicate the No Intrinsic Value Claim even if we reject the Just Holdings Condition, we hope to have strengthened our position.

The implications of our view may be radical, but they are not as nihilistic as one might initially suspect. This is because, though the number of people who have more than their just share may be high, so too is the number of people who have less than their just share, and our view does not imply that fairness cannot have intrinsic value among this latter group of people. Suppose that a development agency is distributing medicine among the inhabitants of an impoverished village but, due to under-funding, it does not have enough medicine to go around. Assuming that everyone in the village has a just claim to the medicine, it follows that injustice will inevitably obtain. But it does not follow that, by seeking to distribute the medicine in a fair manner, the agency does nothing of intrinsic value. If distributive fairness requires that all those who have a just claim to the medicine have an equal chance of receiving it, say, then we should acknowledge that satisfying this standard is intrinsically valuable.Footnote ¹⁴ Thus, even when justice is unachievable, ensuring a fair distribution can still be a virtue.

In developing the Excess Objection, we have had the Fairness Argument’s first rectificatory strategy in mind. That is, we have focused on the claim that we should increase Lawrence’s pay to the level of Bale’s. But we are now in a position to see how our arguments also impugn the second rectificatory strategy, according to which we should decrease Bale’s pay to the level of Lawrence’s. We can concede that this strategy conduces to the achievement of fairness by eliminating a morally arbitrary inequality. But if the No Intrinsic Value Claim is correct – and we have offered two independent arguments for this verdict – then this fairness is not of any intrinsic value.

Of course, if the money withheld from Bale were donated to a poverty relief organization, then decreasing Bale’s pay would be commendable. But this does not vindicate the Fairness Argument, for the Fairness Argument recommends decreasing Bale’s pay in order to address Lawrence’s fairness complaint, not in order to benefit the world’s poor. Similarly, we might commend decreasing Bale’s pay on the grounds that it reduces the degree to which his holdings exceed his just share. But, again, this observation does not salvage the Fairness Argument.

In case some readers are sceptical of the No Intrinsic Value Claim, it’s worth noting that the considerations adduced in our presentation of the Excess Objection ground an additional argument against the second rectificatory strategy. As we have accepted throughout, Lawrence suffers some unfairness. She is paid less than Bale, and her lower pay appears to be attributable to a morally arbitrary factor. But as we have also emphasized, Lawrence’s holdings are unjust. These holdings comprise resources that Lawrence is morally required to relinquish. Now, the second rectificatory strategy recommends eliminating the unfairness between Lawrence and Bale by reducing Bale’s pay to the level of Lawrence’s. One problem here is that, in doing so, it also appears to offer a normative endorsement of Lawrence’s pay. It states that we should maintain Lawrence’s pay, and that other millionaires such as Bale ought to be remunerated at the same level. But Lawrence’s pay is unjust. These observations reveal that the second rectificatory strategy is vitiated by a narrowness of vision. It focuses myopically on the morally irrelevant micro-unfairness between Lawrence and Bale, while ignoring larger problems. As a result of this tunnel vision, it ends up sanctioning injustice.

Our conclusion, then, is that the Fairness Argument for rectifying millionaire inequality is unsuccessful. In the next section, we examine an alternative case for rectifying millionaire inequality.

3.1. Discrimination and Market Demand

What makes discrimination wrong? According to one view, discrimination can be wrong in virtue of the discriminator’s bad intentions (Alexander Reference Alexander1992; Slavny and Parr Reference Slavny and Parr2015). The strongest intuitive support for this judgement is provided by reflection on cases like the one described above, where an unqualified applicant is denied a job on explicitly prejudiced grounds.Footnote ¹⁵ Here, the character of the employer’s intention is sufficient to render her decision wrongfully discriminatory, and thus we need not appeal to any other considerations in order to deliver the conclusion that the applicant is a victim of a serious injustice. We can call this the Intention-Based Discrimination Argument.

We suspect that some cases of millionaire inequality involve discrimination of this kind, which is performed by employers acting on explicitly sexist motives (and recent charges brought against certain Hollywood producers has strengthened this suspicion). Under these circumstances, it’s straightforward to establish wrongful discrimination and we should condemn the treatment on this basis. However, in other cases, those who determine the salaries of female millionaires might not be motivated by explicit prejudice, such as sexist beliefs, and may instead be responding to variations in the level of market demand for the talents of male and female stars. One possibility is that, when differences in earnings between victims and beneficiaries of millionaire inequality are generated in this way, there is no wrongful discrimination. We can call this the Market Demand Reply to the Intention-Based Discrimination Argument.

Two considerations support this reply.Footnote ¹⁶ First, even if the contributions of male and female millionaires seem relevantly identical at first blush, closer examination may reveal subtle differences between them. For example, male tennis players typically exhibit higher levels of strength, stamina, and speed than their female counterparts. Thus, there can be variations in the goods that the relevant employees produce. Second, even when this is not the case, differences may still emerge simply in virtue of variations in others’ reactions to apparently identical goods. For example, it may be that audiences tend to prefer watching male news presenters to equally talented female ones. The point is not that male news presenters are typically more talented than female news presenters, in any technical sense of that claim. It’s possible that both sets of presenters perform identically well, but that audiences still happen to respond more positively to male news presenters than female ones.

In practice, it’s often difficult to discern which of these factors affects market demand in a given case, since this task is an epistemically complex one and the line between the two factors is blurred, even at the conceptual level. As Iris Marion Young notes, market evaluations of qualifications and contributions of individuals are greatly influenced by societal values, such as whether they incorporate norm-following behaviour, social competencies, and conform to role expectations (e.g. expectations toward women that differ from those toward men). Because of this, it’s practically difficult to separate variations in goods from variations in reactions (Young Reference Young2011: Ch. 7).Footnote ¹⁷

The difficulty of determining the exact basis of market demand is important for two reasons. First, it casts doubt on the claim that those who determine millionaires’ salaries must be motivated by explicit prejudice, including sexist beliefs (which is not to deny that they sometimes are). According to the Market Demand Reply, some millionaire inequalities are better understood as tracking differences in the goods produced by different celebrities and/or variations in consumer reactions to those goods. Second, it may be possible to appeal to these considerations in order to defend some millionaire inequalities. According to this defence, it’s justifiable to pay female celebrities less than their male counterparts when and because this reflects the lower level of market demand for their services. This is the view of the tennis player, Novak Djokovic, who has claimed that tennis players’ salaries should reflect ‘who attracts more attention, spectators and who sells more tickets’ (BBC News 2016).

The Market Demand Reply appeals to the suggestion that, when determining Lawrence’s pay, some employers may be responding to the level of market demand for her talents, rather than acting on bad intentions. What should we make of the Market Demand Reply? A first response invokes the fact that political philosophers disagree about whether and why differences in the level of market demand for employees’ services can justify inequalities in pay. For example, proponents of some forms of radical egalitarianism believe that the fact that an individual’s services are in high demand never justifies paying her more than someone whose services are in low demand (Cohen Reference Cohen2008). Those drawn to this view will be unpersuaded by the Market Demand Reply, believing that the misfortune of lacking talents that are in greater demand is as morally arbitrary as gender and race, and so allowing salaries to depend on this factor is as wrongfully discriminatory as allowing salaries to depend on these other attributes. However, since this rejection of market mechanisms is controversial, and because the relevant considerations have been explored at length elsewhere, we shall set it aside.

A second response holds that the Market Demand Reply merely passes the buck. More specifically, it suggests that even if employers are not wrongfully discriminating against female victims of millionaire inequality, perhaps consumers are doing so. According to this line of argument, consumers act on wrongfully discriminatory preferences that generate an incentive for employers to offer lower pay to female actors than to their male co-stars. By focusing on consumers’ preferences, we can uphold the verdict that Lawrence is a victim of intention-based wrongful discrimination without laying the blame at the door of an employer, who may simply be responding to market demand.

In order to sustain this response, it’s necessary to do more than show that consumers’ preferences generate an incentive for producers to offer lower salaries to Lawrence and other female celebrities. In particular, proponents of this reply must show that consumers are motivated by explicit prejudice, or have a similarly morally defective kind of intention. (From the perspective of the Intention-Based Discrimination Argument, if consumers’ intentions are entirely innocent, then there can be nothing more objectionable about the inequality in income between Lawrence and Bale than the inequality in income between a highly talented actor and a weak actor.) However, this task is not straightforward, since the judgements that inform consumers’ preferences regarding female actors can be very different from the obviously morally defective intentions of an employer who rejects an applicant on explicitly racist grounds. Certainly, consumers with a stronger preference for seeing Bale than Lawrence need not exhibit the kind of animus and explicit prejudice that motivate the employer in the case with which we began and that are necessary to classify the employer’s act as intentionally malicious.

In the light of these facts, we conclude that the Intention-Based Discrimination Argument is likely to fail. But we cannot infer from this that we must reject the Discrimination Argument tout court. This is because employers and/or consumers may be engaged in forms of wrongful discrimination against female millionaires that do not rely on the nature of their intentions. In the remainder of this section, we explore these possibilities in greater detail.

3.2. Reconsidering the Market Demand Reply

One problem with the Market Demand Reply is that it adopts an uncritical attitude towards consumer preferences and the inequalities that these preferences generate. In particular, it fails to recognize that, irrespective of whether these preferences are explicitly prejudiced in themselves, they may have been generated under unjust or otherwise problematic circumstances, and this may make a difference to our moral evaluation of the inequalities that emerge from them.

For example, unjust gender-related social norms play a significant role in shaping our preferences and expectations about goods and services (Haslanger Reference Haslanger2015). Social norms determine which roles men and women are expected to fulfil, often to the disadvantage of the latter (Haslanger Reference Haslanger, Anthony and Witt1993; Young Reference Young2011: 193–197). This is important because, irrespective of whether consumers’ preferences are themselves explicitly sexist, they are often shaped by an unjust culture of sexism.

Do these considerations show that Lawrence and other female millionaires are victims of wrongful discrimination? One way in which those who answer ‘yes’ might support this judgement is by appealing to the fact that the sexist injustice that characterizes our world harms Lawrence, in the sense that it renders her worse off economically than she would otherwise be. In a world free from sexist injustice, Lawrence’s earnings would be greater than at present, more closely equalling the income of Bale. This is because, in the absence of sexist injustice, social norms would not pervasively shape consumers’ preferences in a way that creates an incentive for producers to offer lower pay to Lawrence than to her male co-stars. We can call this the Harm-Based Discrimination Argument.Footnote ¹⁸

Three features of this argument warrant emphasis. First, despite some similarities – for example, both are concerned with the distribution of income – the Harm-Based Discrimination Argument and the Fairness Argument make rather different claims. The Harm-Based Discrimination Argument relies on a counterfactual claim about how Lawrence would have fared economically under alternative conditions, and, in particular, about how she would have fared in a world without sexist injustice (Goff Reference Goff2016). By contrast, the Fairness Argument invokes an interpersonal claim about how Lawrence fares economically in comparison with other individuals, and, in particular, about how she fares in comparison with her male co-stars.

Second, the success of the Harm-Based Discrimination Argument does not depend on the controversial premise that consumers act wrongly when expressing a stronger preference for Bale over Lawrence, or for Djokovic over Williams. The argument appeals to the more modest claim that consumers’ preferences may be shaped by sexist background injustice and that we should be sensitive to the harmful effects that this may have on people’s earnings. It’s consistent with this latter claim that consumers’ preferences, though shaped by unjust social norms, are not themselves unjust.

Third, the Harm-Based Discrimination Argument’s core claim is that female millionaires are wronged because they fare less well economically than they would if consumers’ preferences were not shaped by sexist social norms. In order to make sense of this idea, we do not need to commit to any particular account of what makes a social norm sexist. It could be because of these norms’ propensity to undermine equality of opportunity; it could be because of what they express; or it could be something else. We do not need to take a stand on this tricky issue. We take this to be an attractive feature of the Harm-Based Discrimination Argument since it means that we can combine it with a range of different answers to that question.

However, the Harm-Based Discrimination Argument is vulnerable to an objection, which stems from the fact that, though Lawrence is economically worse off than she would have been in a world without sexist injustice, it’s not obvious that this is the relevant counterfactual baseline against which to make the comparison. An alternative option is to accept what we call the Comprehensive Baseline Claim. This claim holds that the relevant comparison for the purposes of the Harm-Based Discrimination Argument is with how an individual would have fared in a world without any economic injustice, sexist or otherwise. This claim takes seriously the fact that, while Lawrence can complain that sexist norms shape consumers’ preferences in a way that renders her less well off economically than she would have been in the absence of these unjust norms, there are other economic injustices from which Lawrence benefits. In determining whether Lawrence is harmed in the relevant sense, we should compare Lawrence’s current economic situation with how she would have fared in a world that lacks not only the sexist component of economic injustice, but economic injustice more generally.Footnote ¹⁹ Since Lawrence is not worse off economically than she would have been in the absence of all economic injustice, those who endorse the Comprehensive Baseline Claim must deny that she is a victim of wrongful discrimination, according to the Harm-Based Discrimination Argument.

In support of this objection, let’s return to the earlier example of Slaves, in which the government of a slave state distributes slaves on a random basis, with the result that some citizens end up owning more slaves than others. Now, let’s also suppose that the ruling government came to power through an unjust coup d’état, and that the pre-coup government distributed slaves more equally. Can a citizen who receives no slaves appeal to the Harm-Based Discrimination Argument, citing the fact that she is worse off now than she would have been had the unjust coup not occurred? Intuitively, the answer is ‘no’, and our analysis helps to make sense of why this is the case. The slaveless citizen cannot complain in this way because the pre-coup state of affairs is the wrong counterfactual baseline against which to make the comparison. As the Comprehensive Baseline Claim states, the slaveless citizen must compare her situation to how she would have fared had the unjust coup not occurred and had the pre-coup government not been a slave state. Along similar lines, Lawrence and other female millionaires cannot appeal to the fact that they would have earned more in a world without distributive gender injustice to support the judgement that they are victims of wrongful discrimination. They cannot appeal to this fact because they are much better off economically than they would be in a world that satisfies not only the requirements of distributive gender justice, but also those of distributive justice more generally.

At this point, one might press an objection against the Comprehensive Baseline Claim. We can illustrate this objection by considering a case in which a female millionaire is a victim of sexual harassment. It might be said that such harassment need not render this millionaire worse off, all-things-considered, than she would be in a fully just world in which she is not sexually harassed, but in which she also earns much less than she does at present, and that the Comprehensive Baseline Claim therefore prevents us from saying that she is harmed. Given that this conclusion is unacceptable, the Comprehensive Baseline Claim must be rejected.

In reply, this objection misunderstands our claim. Our claim is that in order to identify the specific kinds of economic harm that concern us, the relevant counterfactual world with which to draw comparisons is one that lacks any kind of specifically economic injustice. If in that counterfactual world one fares less well economically than one does in our world, then one is not harmed economically. This claim has no implications for the kind of counterfactual scenario that is appropriate when attempting to identify other, non-economic, forms of harm. Thus, we can readily accept that the envisaged millionaire is harmed by the sexual harassment she suffers, even if she would be less well off economically in a fully just world.

The upshot of these considerations is that we should reject the above attempt to salvage the Harm-Based Discrimination Argument, and thus that we are not entitled to appeal to this argument in order to sustain the conclusion that Lawrence and other female millionaires are victims of wrongful discrimination.

3.3. Discrimination and Disrespect

So far, we have focused exclusively on individuals’ economic entitlements. But this approach is too narrow, since it overlooks the way in which inequalities can express an insulting message about the comparative moral worth of the victims and beneficiaries, and how the expression of such a message can be disrespectful. This idea is a powerfully intuitive one. It helps to explain what is so morally abhorrent about laws that require black passengers to sit at the back of the bus, or policies that offer payments to minorities to emigrate (Anderson and Pildes Reference Anderson and Pildes2000; Hellman Reference Hellman2011; Gerver Reference Gerver2018). Failing to recognize the symbolic significance of these policies is failing to recognize the nature of the injustice that the victims suffer. It might be argued that we can sustain similar judgements in the cases with which we are concerned. This is the Disrespect-Based Discrimination Argument.

This argument rests on the claim that paying female celebrities less than their male counterparts expresses an insulting message about their comparative moral worth, and perhaps also about the comparative moral worth of men and women more generally, and that we should object to millionaire inequalities on this basis.Footnote ²⁰ Importantly, this argument is not concerned with what an agent intends his acts to express. Rather, it’s concerned with how his acts can reasonably be interpreted. If an act can reasonably be interpreted as insulting, performing that act can be disrespectful, regardless of whether the insult is intended. These considerations allow us to condemn conduct as wrongfully discriminatory on the basis of the message it sends, without having to inspect the intentions of those acting.

One advantage of the Disrespect-Based Discrimination Argument is that it furnishes us with a response to the Market Demand Reply. To see how it does this, let’s consider a US Supreme Court case from 1964, in which an Alabaman restaurant, Ollie’s Barbeque, refused to open its doors to black customers following the enactment of the Civil Rights Act on the grounds that, if it were required to do so, the restaurant ‘would lose a substantial amount of business’ (Katzenbach v. McClung 1964: 297).Footnote ²¹ The injustice of the owner’s conduct can be fully explicated only if reference is made to the insulting message that it conveyed. Intentionally or otherwise, the owner’s act sends a clear symbolic message, namely that black and white customers are not moral equals. Given the social context in which it’s performed, and how it could reasonably be expected to be perceived in light of this context, the owner’s act is deeply disrespectful. In 1964 Alabama, black citizens were routinely treated as moral inferiors and had been subject to unequal and unjust treatment for centuries. Part of this unjust treatment consisted precisely in segregation and in refusing black citizens entry to various establishments. Refusing to admit black customers is disrespectful, regardless of the owner’s intentions, in virtue of this unjust social background, and the role it plays in determining how the action is reasonably interpreted. In other words, refusing admission is disrespectful because it’s expressively in line with the enforcement of racial hierarchy and segregation (Anderson and Pildes Reference Anderson and Pildes2000: 1528).

Moreover, the fact that the owner was (allegedly) acting only to maximize profits does little to mitigate the disrespectfulness of their act. The judgement that they have acted disrespectfully is grounded in their failure to be adequately responsive to the social context in which they were acting, and to the social meaning that this context would inevitably lend to their conduct. A crucial implication of these observations is that the Market Demand Reply is insufficient to stave off assertions of wrongful discrimination when the conduct in question expresses an insulting message, as it does in this case.

Importantly, it seems plausible to hold that these considerations transfer over to the case of millionaire inequality. Even if unequal remuneration really does reflect variations in the level of market demand for celebrities’ talents, such remuneration plausibly remains wrongfully discriminatory given the insulting message that it’s likely to convey regarding the comparative moral worth of men and women (Gheaus Reference Gheaus2020: 13–14). Contemporary liberal societies, including Britain and the USA, do not treat women in a manner that is comparable to the manner in which 1960s Alabama treated African Americans. Nevertheless, these societies have deeply sexist histories. In the recent past, these societies have upheld practices that deprived women of vitally important goods (such as the right to vote and to own property) and that failed to protect women from serious harms (such as marital rape). Moreover, these societies continue to be marked by deep gender inequalities. Women suffer from a variety of serious inequities, including a disproportionate burden of household labour, high levels of domestic violence, negative stereotyping, and socio-political marginalization (Satz Reference Satz2010: 144–146). Given this social context, it’s reasonable to regard unequal remuneration as insulting, whatever its actual source. Employers who pay women less than men often fail to be adequately responsive to the social context in which they are acting, and to the meaning that their acts can reasonably be expected to convey in light of that context. This is a failure of respect.

In drawing the parallels that we have drawn, we do not mean to claim that the symbolic content in cases of millionaire inequality is as unambiguously disrespectful as it is in paradigmatic examples of racial injustice. In the example of Ollie’s Barbeque, the restaurant claims to be acting on market demand shaped by explicit prejudice on the part of its patrons. However, in cases of millionaire inequality, we grant that employers may be providing unequal remuneration on the basis of consumer demand that is not motivated by explicit prejudice. Some critics of our analysis might argue that this difference renders cases of millionaire inequality invulnerable to the Disrespect-Based Discrimination Argument. We disagree. What matters is that, in both cases, the acts with which we are concerned are widely and reasonably understood (particularly amongst the victims) to express a disrespectful message about the comparative moral worth of different individuals.

These issues are part of a larger picture regarding the systematic oppression of members of a socially salient group (Young Reference Young2011: Ch. 7). Our observations emphasize that cases of millionaire inequality occur within the context of a world in which women of all backgrounds are systematically disadvantaged relative to men, and perhaps it’s these facts that make Lawrence’s complaints so serious.