Is it a concern of justice if millionaire women don’t get paid as much as millionaire men solely because of their gender? In a recent article in this journal, James Christensen, Tom Parr and David Axelsen argue that millionaire salaries are unjust and women have no grounds of fairness to unjust salaries (Reference Christensen, Parr and Axelsen2022). They do accept that paying women less than men purely on account of their gender is wrong for expressive reasons, namely, that such gender disparities express that women are less skilled, talented or valued than men. We argue that their argument threatens women’s claims to fairness in many contexts beyond millionaire salaries, and as such should have been more cautious. In addition, by drawing on the distinction between ideal and nonideal theory, we also argue Christensen et al. hold women to ideal standards of justice even in contexts where injustice is widespread, to the benefit of men.

Christensen et al. consider two arguments for rectifying gender inequality with respect to millionaire holdings: a fairness argument and an expressive argument. We deal with each argument in turn.

The fairness argument for rectifying gender inequality between millionaires says that women and men should be paid equally because women should not be disadvantaged because of a morally irrelevant factor such as gender. Christensen et al. reject the fairness argument for essentially the same reasons as Anca Gheaus (Reference Gheaus2020) did so earlier: fairness has no value with respect to the distribution of goods that are unjustly held.Footnote ¹ We discuss two concerns we have with their view: firstly, a complaint that one might wish for a higher standard of caution before dismissing claims that have been at the heart of gender equality movements, and, secondly, an objection that they hold women to ideal standards of justice in nonideal conditions.

Christensen et al. defend the claim that millionaire salaries are unjust on the grounds that millionaires are morally required to give up their excess wealth, either voluntarily or through taxation, because it would make a significant contribution to ending extreme poverty at very little cost to themselves. As such, when millionaires hoard such wealth, it represents ‘a failed opportunity to rectify severe disadvantage’ (Christensen et al. Reference Christensen, Parr and Axelsen2022: 337).

Christensen et al. acknowledge they are leaning on moral claims developed by, among others, Peter Singer (Reference Singer1972). We know of the argument that Singer develops in ‘Famine, Affluence and Morality’ that it can entail consequences that are anything but modest: certainly Singer’s own preferred version of the principle entails quite radical consequences. Given this, we raise similar questions about whether the argument developed by Christensen et al. poses possible ‘nihilistic’ consequences for women’s fairness claims more generally. Specifically, we question what grounds they have for supposing only millionaire salaries are unjust. After all, Singer (Reference Singer1972, Reference Singer2009) insists that even modest earners could contribute in highly effective ways to help eliminate extreme deprivation without sacrificing anything of moral importance. If this is true, we should also deem (at least some portion of) the holdings of those modest earners as unjust. Of course, if we were to do so the implications for women’s pay parity claims would be stark, even ‘nihilistic’ (Christensen et al. Reference Christensen, Parr and Axelsen2022: 341): pervasive unfairness against even middle-class women would have no intrinsic moral disvalue. Women would have no complaint of fairness against gender pay gaps across a vast range of industries in wealthy, modern economies.Footnote ²

At different points in their article, Christensen et al. suggest a number of reasons as to why their argument does not have such nihilistic implications. Firstly, toward the end of their article they explicitly claim their argument will not target middle class earnings as unjust. They suggest, Singer’s arguments notwithstanding, that ‘the wealth that is held by any moderately well-off individual need not represent an excessively large forgone opportunity to assist the world’s poor’ (Reference Christensen, Parr and Axelsen2022: 350). This suggestion overlooks that the middle-class remain very large in number, and much larger than the number of multi-millionaires; despite dwindling somewhat in some developed countries, their numbers have been rapidly rising in many developing countries. Collectively their wealth could plausibly eliminate much global deprivation. Christensen et al. also suggest that taxing or requiring donations from the middle-class to tackle world poverty would require them to forgo benefits that are ‘not trivial’ (Reference Christensen, Parr and Axelsen2022: 350), whereas such impositions on millionaires would come at ‘very little cost to them’. Again, given that Singer and others have argued extensively against these claims, a substantial argument of some kind would be needed to defend them, incorporating both empirical data about what it would actually take to address global poverty and some philosophical analysis as to what counts as no, a small, or a considerable ‘cost’.

A second option for staving off the ‘nihilistic’ threat is one which Christensen et al. discuss at length (Reference Christensen, Parr and Axelsen2022: 338–339): namely, that rectifying the unfairness suffered by millionaire women has some value, but because wealthy women already have so much, that value is outweighed by the desperate unmet needs of the poor. On this approach it might follow that rectifying middle-class gender unfairness also has some value, and because middle class wealth is not so massively unjust, the value of fairness is not necessarily overridden. Christensen et al. ultimately reject this option. Their own preferred view is that the value of fairness is not merely outweighed by severe injustice (such as millionaire holdings in the context of severe deprivation); they argue that the intrinsic value of fairness is extinguished by injustice. They defend the Just Holdings Condition according to which ‘improving the fairness of a distribution is intrinsically valuable only when the distribuenda are justly held’ (Reference Christensen, Parr and Axelsen2022: 339).

Christensen et al. defend the Just Holdings Principle by borrowing and modifying an example from Gheaus. Gheaus’ original example better illustrates the issue of gender fairness, so we stick to her version. She asks us to consider a slave-owning society in which women are not permitted to own slaves. This society proposes a reform to allow women to own slaves and since this would eliminate gender inequality of opportunity (or unfairness) to be a slave owner, the reform appears to have some moral value. However, nobody should have an opportunity to own slaves because owning slaves is morally impermissible. Even though it would eliminate a morally arbitrary form of gender-based unfairness if women were also allowed to hold slaves, fairness has no intrinsic value here because slaves are not justly held. ‘Nobody should hold slaves, hence nobody has a claim [of fairness] to hold slaves’ (Gheaus Reference Gheaus2020: 16).

Christensen et al. agree with Gheaus’ view that no one has a claim of fairness to hold slaves. They deny that the plausibility of this conclusion is based on the fact that the injustice of slavery is so massive that it outweighs the intrinsic value of a fairer distribution. They argue that if two people – a man and a woman – steal a pizza meant for their housemate, there is no intrinsic value owing to fairness in ensuring that the woman receives an equal share (Christensen et al. Reference Christensen, Parr and Axelsen2022: 340). It follows from their commitment to the Just Holdings Principle that any unjust level of wealth extinguishes women’s fairness claim to be treated equally to men. Christensen et al. acknowledge that some readers might ‘demur’ from this analysis (perhaps because of the potential for the nihilistic implications we have discussed): they might insist that in cases where the injustice is not massive (such as stealing a pizza) then there is at least some value to fairness. Nevertheless, they argue that even if this position is correct, their overall argument with respect to millionaire fairness remains untouched, because the unfairness of millionaire inequality has more in common with an unfair distribution of slaves than with an unfair distribution of stolen pizza. The hoarding of excess wealth by the super-rich which could otherwise be used to address extreme deprivation is massively unjust, although they do state ‘we do not suggest that being rich is as bad as being a slave owner’ (Reference Christensen, Parr and Axelsen2022: 341, their emphasis).

Let us now state our complaint directly. Given the potential ‘nihilistic’ threat to a huge range of women’s fairness-based claims in wealthy economies, we might have hoped for a little more caution than we see in this article. Gender fairness with respect to pay and prize money has been an important aim of feminist and gender equality movements. We believe it is particularly reasonable to expect caution, if not downright modesty, whenever a philosopher evaluates the common justice claims made by socially salient disadvantaged groups.Footnote ³ Yet in this particular case, the authors’ argument, which runs a clear risk of nihilism with respect to a vast range of women’s fairness claims, is based on a series of claims and arguments that are quite far from compelling: (a) the claim that only millionaire holdings but not wealthy middle-class holdings are unjust would probably be rejected by most theories of distributive justice and is rejected on Singer’s account of beneficence; (b) and the claim that millionaire holdings are almost (but not quite) as unjust as slavery obviously needs more discussion, discussion that, in turn, would lead us back to complex debates about distributive justice.

None of this is to state that Christensen et al. should have avoided critically discussing women’s fairness-based claims to very high pay and prize money. Indeed, much of their discussion could have proceeded to a large extent in its current form. What they should have done is shown more modesty in their conclusions. Given their conclusion ultimately rests on highly contestable foundations about beneficence and distributive justice, a more cautious approach would have been to instead highlight some of the complex distributive justice considerations that have to be resolved before women can convincingly substantiate fairness-based claims to gender parity with respect to very high levels of wealth. Such a discussion we would have no complaints about.

We readily acknowledge that not all readers will share our views about how philosophers ought to exercise caution in discussing common justice claims of disadvantaged groups. So we now turn to our objection, which is based on the distinction between ideal and nonideal political theory (Rawls Reference Rawls1971; Sen Reference Sen2006).

Having dismissed women’s fairness complaints on the grounds that no one should earn unjust salaries they are left with the obvious question: what should be done about men’s unjust holdings? One suggestion would be to level men’s salaries down. This is Gheaus’ suggestion (Reference Gheaus2020: 16, 17). Christensen et al. caution against the argument that men’s millionaire salaries should be levelled down. Given that on their view a fairer distribution of unjust holdings has no intrinsic value, there can be no fairness-based argument in favour of either increasing women’s salaries or lowering men’s salaries to the level of millionaire women. It is worth again emphasizing that this argument would also hold against levelling men down to a middle-class level of wealth if that too is deemed unjust (as it would be on Singer’s view). Indeed, Christensen et al. comment that levelling men down on grounds of fairness might be a perilous strategy insofar as it could be interpreted as endorsing injustice, namely the injustice of the holdings of wealthy women. It is true that levelling men down might run the risk of endorsing injustice, if the level to which they are lowered is still unjust. But that risk would obviously have to be balanced against the risk of endorsing injustice if men’s salaries are kept at their current higher level. This is not a risk which Christensen et al. mention, nor suggest how to mitigate.

Christensen et al. do, of course, endorse lowering men’s holdings for reasons other than fairness namely, simply for the reason that millionaire salaries are unjust. They say that ‘[i]n a just society, the excess wealth of the super-rich would be taxed away … 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)’ (Reference Christensen, Parr and Axelsen2022: 337). This brings us directly to our main objection to their argument: millionaires are not taxed so that their holdings do not exceed their fair share and they do not give away all of their unjust holdings to poverty relief. In our actual world, many people, overwhelmingly men, do, as a matter of fact, continue to enjoy millionaire levels of wealth, and will continue to do so in our foreseeable future. This fact compels us to consider women’s fairness claims in nonideal contexts.

Briefly, we here understand ideal theory to assume full compliance with principles of justice, where nonideal theory acknowledges lack of full compliance and asks what we ought to do given other people are not doing what they ought to do (Valentini Reference Valentini2012). We question why ideal standards of distributive justice are being invoked to reject women’s fairness claims to millionaire (or even middle-class) pay parity in the context of widespread non-compliance with those very standards: many people do earn millionaire salaries, predominantly (a small number of) men. At the very least, women can reasonably expect an additional argument about what nonideal theorizing might say about gender parity in contexts of widespread noncompliance with the demands of justice, which is not offered, or mentioned, in this article. As such, we suggest that Christensen et al. can be plausibly accused of holding women to higher standards of justice than men.

Here is one nonideal suggestion that Christensen could have made, one taken directly from feminist and gender-equality advocates: that women should receive the same remuneration as men when they hold the same positions, including when they hold positions that attract millionaire salaries or prize money. In other words, they could have recommended levelling women up as a nonideal response in contexts of noncompliance. Ensuring that women earn millionaire salaries for certain positions for as long as men do would at least offer us a feasible way to rectify the moral wrong of being treated less favourably purely on account of gender, even though full justice is out of our reach. Of course, nonideal theory does not require that we accept whatever solution is most feasible (although we do think it requires us to reject completely infeasible solutions). Feasibility considerations are balanced against, among other things, whether an option brings us closer to justice or makes existing injustice even worse. Nonideal theorizing is rather messy in this way: it does not divide our options into either fully just or completely unjust (Sen Reference Sen2006), but requires the complex assessment of a number of empirical and ethical issues to decide the comparative acceptability of the range of options on the table, none of which can secure full conformity with ideal principles of justice. It would be a complex matter beyond the scope of this short article to determine whether gender parity would make things better or worse: on the one hand, it would eliminate some gender unfairness and expressive disrespect (see below), but on the other hand do very little or nothing at all to address broader inequality of opportunity and distributive injustice; it may even make them worse. Whatever the case, our objection stands: it is unreasonable to reject women’s claims to gender fairness using ideal standards of justice in a context where noncompliance with such ideal standards is rife. Women are owed some argument as to how their fairness-based demands would fare if we took seriously the need for nonideal reasoning.

It is perhaps worth mentioning that unlike Christensen et al., Gheaus concedes the need to propose nonideal solutions to existing gender unfairness given noncompliance: she acknowledges that men will probably not be levelled down and that unjust holdings will continue to exist in the foreseeable future. She suggests that ‘if and as long as such positions are up for distribution, a fairer way to distribute them seem to be by lotteries with a threshold of competence’ (2020: 15). She suggests that gender quotas could be attached to such lotteries to ensure they deliver an equal gender distribution. On the one hand, we are pleased that Gheaus accepts that women should not necessarily be held to ideal standards of justice in contexts of widespread noncompliance with those standards. On the other hand, it is very unlikely that lotteries offer a practical solution in many contexts. For example, ‘Pop and sports stars’ are included in Gheaus’ target of people who earn unjust salaries. It is difficult to imagine how ‘lotteries’, even with a ‘threshold of competence’ could be implemented in these domains. In many areas implementing a lottery system to allocate jobs would require a massive overhaul of current labour market systems which would be entirely politically infeasible, and we suspect probably very undesirable for a number of market-driven, fairness-driven and practicality-driven reasons. Our aim with these brief comments is not to show that lotteries could never be a nonideal solution to gender unfairness. To the contrary, we believe that they might work in some cases. Nevertheless, if we are doing nonideal theory, exploring what justice for women might mean in the context of noncompliance with ideal standards of distributive justice, it is not unreasonable to hold out hope for feasible and practical suggestions. Why would Geaus suggest lotteries when gender pay parity seems much more feasible? We can only speculate here, but our suspicion is that despite explicitly acknowledging the shift to nonideal theory, Gheaus continues to labour under the assumption that solutions to gender inequality of opportunity must not violate ideal standards of justice, which gender pay parity would certainly do. Gender pay parity with men does nothing to address broader inequality of opportunity or unjust holdings.Footnote ⁴

Christensen et al. are much more sympathetic to the second argument against millionaire men being paid higher salaries than women, namely, that such inequality expresses something deeply disrespectful about women. Following Gheaus, Christensen et al. suggest that the real wrong done to millionaire women when they are paid less than millionaire men is expressive: 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’ (Reference Christensen, Parr and Axelsen2022: 348).

Although they endorse the argument that women have a respect-based claim to rectification of millionaire inequality, Christensen et al. do not offer any suggestions as to what such rectification might look like. They can of course emphasize that their aim was to assess the various arguments in favour of millionaire equality, not to suggest how to achieve such equality when good arguments support it. Yet expressive disrespect of women is a serious concern, as Christensen et al. rightly acknowledge. Such disrespect is tied to widespread and pervasive gender inequality across most spheres, including persistent disadvantageous treatment purely on account of gender. Expressions of disrespect for women leads to (sometimes savage) inequality, neglect of girls’ and women’s interests, and even violence. If we take such injustice seriously then we believe women are owed minimally feasible and practical suggestions as to how we might respond to it.

Feminists and others concerned with gender inequality have offered such suggestions: that such injustice is (partially) addressed by paying women the same salaries and prize money as men. As we’ve conceded, this is a thoroughly nonideal solution in that it would leave unjust salaries intact. The mere fact that it is a nonideal solution to expressive disrespect however cannot itself be a reason against it given widespread noncompliance with such ideals. Or at least, not without considerable argument. We live in a world rife with serious distributive injustice that is currently beyond our capacity to eliminate and which largely benefits (a small number of) men. Women are owed serious proposals for tackling gender unfairness and expressive disrespect, especially in contexts where widespread lack of conformity with ideal standards of justice is conceded as marring our foreseeable future. It cannot be acceptable to hold women’s claims to gender equality and their claims to equal respect hostage to ideally just standards that do not exist.

Someone could object to pay parity as a feasible nonideal solution to rectify disrespect on the grounds that it would lead to even more injustice. Quite a lot of argument would be needed for us to find this objection even prima facie plausible. It is true that if women were to also receive millionaire salaries in positions where men do, then it is likely that more people overall will receive unjust salaries.Footnote ⁵ But gender pay parity would also decrease at least some highly salient expressive disrespect toward many more. No simple conclusions follow about the impact of gender pay parity on ‘overall’ levels of justice.

Nonideal theory takes place in the bog of injustice and involves much muddier thinking about how to respond to injustice when none of our options will realize a fully just state of affairs, and many of them will themselves fail to conform to ideal standards. We accept that we may be far too sanguine about the unique or special badness of millionaire salaries (almost as bad as slavery?). Just as much more argument is needed to bolster many of the claims about justice and fairness made by Christensen et al., more argument is obviously needed to substantiate our assumption that millionaire salaries probably do not represent a form of vicious injustice which women must never enjoy, even when men do, and even when they will continue to do so. There should be no injustice in the world. But there is: a lot, everywhere. Our main aim has been to argue that there is one approach to gender injustice that should be treated with a hefty dose of scepticism: to either dismiss women’s claims on the back of highly contestable suggestions about distributive justice and the duties of beneficence, or to hold their claims hostage to ideal standards of justice even in radically nonideal contexts that persist to the benefit of (a small number of) men.