Journal of Practical Ethics

The Feminist Argument Against Supporting Care

Anca Gheaus

Abstract

Care-supporting policies incentivise women’s withdrawal from the labour market, thereby reinforcing statistical discrimination and further undermining equality of opportunities between women and men for positions of advantage. This, I argue, is not sufficient reason against such policies. Supporting care also improves the overall condition of disadvantaged women who are care-givers; justice gives priority to the latter. Moreover, some of the most advantageous existing jobs entail excessive benefits; we should discount the value of allocating such jobs meritocratically. Further, women who have a real chance to occupy positions of advantage have most likely already enjoyed more than their fair share of opportunities; they lack a claim to more. Women can have a complaint grounded in the expressive disvalue of sexist discrimination. This gives them special claims against men occupying the vast majority of top positions and against their higher share of opportunities for positions of advantage. But their claim does not speak against care-supporting policies.

Introduction

Feminist scholars and activists regularly defend public policies that support care-giving and care-givers. Some of these policies support care directly, by entitling to financial compensation individuals who provide care to their family members and who thereby often forego employment or downsize their career ambitions. Examples include a mandate that employers split the paycheck between their employee and their stay at home partner (Okin, 1989), or a ‘caregiver’s allowance’ in the form of a voucher given to parents who provide hands-on care (Alstott, 2004). Other measures are meant to support the combination of care-giving and paid work, and include paid and relatively long parental leaves, flexible working hours for parents, better compensated part-time work and subsidies for non-parental childcare (Williams, 2000; Gornik and Meyers, 2003; Schouten, 2019). Finally, some policies whose aims are more general—such as a universal basic income—would also support-caregivers (Elgarte, 2008; Baker, 2008). These diverse policies are justified by different considerations. Yet most1 of them trigger a general feminist worry: that protecting care-givers unavoidably provides women with incentives to opt out of the labour market at higher rates than they already do, and at much higher rates than men. This, the criticism goes, will reinforce statistical discrimination and sexist stereotypes that portray women as less well-performing employees, and thus further undermine equality of opportunities between women and men for positions of advantage.

This criticism, I argue, doesn’t generate a powerful reason against care-supporting policies—not in general, and even less so in unjustly unequal societies. One broad argument is as follows: even if it is true that supporting care makes it more difficult for women to access positions of advantage compared to men, it is also likely to improve the overall condition of women who are most disadvantaged by their care-giving social role. Not all the above policy proposals would benefit the worse off women equally, but all would do so considerably. At the same time, it is mostly better off women who have an interest in equal opportunities to positions of advantage. It is highly plausible that the interests of the worst off ought to be given priority. Here I will not engage in any detail with this general argument but focus, instead, on the particularly strong case in favour of supporting care-giving in unjust societies.

I assume that existing societies fail to observe any plausible principle of distributing outcomes. I don’t commit here to a particular theory of justice; such a principle may be luck-egalitarian, permitting only inequalities that track individual responsibility, and therefore deeming as unjust cases where some people are worse off than others out of no fault or choice of their own. Or it can be a prioritarian principle, requiring that the worse off an individual is, the more weight we give to her interests. Or it can be a Rawlsian principle, allowing only for inequalities that are necessary to improve the prospects of the worst-off members of society.2 Finally, much of my analysis should also appeal to sufficientarians, who rule out inequalities to the extent to which they leave some individuals under a threshold of sufficiency. Societies which fail to conform to any of these principles, or alternative principles of justice, I will call “unjust societies”. In unjust societies, the better off women lack a claim of justice with respect to many of the most advantageous jobs. To this conclusion I advance two arguments. The first applies to the very best jobs, which I call, by stipulation, “top positions”. This is a placeholder term, but plausible candidates include being the CEO of a large company, the best paid jobs in the entertainment industry etc. The second argument applies to positions of advantage in general, plausible examples of which are jobs in law, academia, medicine etc. Given its focus on conflicts of interests between worse off and better off women, this paper echoes the current popular debates between a feminism interested in the fate of poor and marginalised women and so-called “boardroom feminism”. While I cannot engage directly with this debate here, my overall argument provides support to the former. I rely on a common distinction between equality of outcome—concerning where people end up with respect to, for instance, income, social status, or power—and equality of opportunity for such outcomes—that is, a level-playing field in competing for the outcomes.

Argument 1

• P1 Nobody has a right to a position that confers unjust advantages on those who hold it.
• P2 We ought to discount the value of equality of opportunity for positions to which nobody has a right.
• P3 Many of the top positions in societies with large unjustifiable inequalities of outcome confer unjust advantages on those who hold them.
• C1 Therefore, we ought to discount the value of equality of opportunity for many of the top positions in societies with large unjustifiable inequalities of outcome.

There is a legitimate question about the relationship between the magnitude of unjustifiable inequalities attached to a position and the plausibility of P2 with respect to that position. One possibility is that the more modest the deviation from justice, the less we should discount the value of distributing the position in question according to equality of opportunity. So, for instance, if the pay-offs attached to a position deviate only marginally from what justice permits, it is almost as desirable to distribute it according to equality of opportunity as it is in the case of positions that carry permissible rewards. I cannot address this problem properly here, but I assume that at least some top positions in existing societies involve benefits that are entirely out of proportion with the demands of distributive justice, and that we should attach almost no value to distributing those positions meritocratically, that is according to equality of opportunity.

Argument 2

• P4 Most people who have a shot at positions of advantage are likely to already have had more than their fair share of opportunities for such positions.
• P5 People lack a claim to more than their fair share of opportunities for positions of advantage.
• C2 Therefore, most people who have a shot at positions of advantage do not have a claim to more opportunities than they already have for such positions.

According to several feminists, the fact that a policy worsens the inequalities of opportunities between women and men to positions of advantage is a serious consideration against them (Robeyns, 2000; Robeyns, 2001; Gheaus, 2008), perhaps even a defeating one (Baber, 1990; Bergmann, 1998; Bergmann, 2008). Against these views, Arguments 1 and 2 show that all women lack a claim to top positions and most women who stand a chance for positions of advantage lack a claim to more opportunities for them than they already have. But all women do have special claims against men having more opportunities for them, including for top positions: this is a complaint grounded in the expressive disvalue of denying women the same opportunities that men have—as I explain below. The complaint provides grounds for implementing policies that combat gender stereotypes which associate women with care and present them as less desirable workers than men. In particular, the expressive complaint supports policies implementing gender parity in the distribution of positions of advantage. Such parity is desirable both for its immediate expressive value and, potentially, for its long-term consequences in breaking the association between women and domesticity on the one hand, and between men and paid employment on the other hand (Radcliffe-Richards, 2014). Yet one need not aim at the much more demanding ideal of equality of opportunities in order to address women’s expressive complaint.

I assume that, given the large unjustifiable inequalities of both outcome and of opportunities in contemporary societies, most positions where discrimination against women would increase due to care-supporting policies are likely to be like the positions in arguments 1 and 2. This is not a purely speculative point: the main feminist complaint against care-supporting policies is that they will increase a kind of discrimination against women which, as I explain below, is a lot more likely to affect access to positions of advantage. This is not to deny that other kinds of sexist discrimination apply also, or especially, to women in middle and lower grade jobs. But this is not the kind of discrimination likely to be enhanced by care-supporting policies. Therefore, care-supporting policies are significantly less likely to increase inequality of opportunities between women and men for middle-grade jobs and even less so for lower-grade jobs. Moreover, with respect to the lowest-grade positions, such as menial work, the disvalue of a great many of these positions may also mean that opportunities for them are less important to individuals likely to occupy them than receiving care-giving support. Therefore, assuming that a care-supporting policy is otherwise justified, egalitarian feminists have much better reason to welcome than to resist it, even if it sets back women’s opportunities to positions of advantage.

Equal opportunities criticism against supporting care has been raised in connection with all the above-mentioned policies (Baber, 1990; Bergmann, 2008; Miller, 2016). But the most articulated debate to date concerns the introduction, in existing developed societies, of a universal, unconditional basic income (henceforth “BI”)—that is, a regular income financed through taxation and paid to all the members of a political community regardless of any other sources of income and independent from the recipients’ productive participation in social cooperation and of their level of financial need. Here I refer to the feminist debate concerning BI merely to illustrate the tradition of appealing to equality of opportunity for positions of advantage as a weighty reason against its introduction. All care-supporting policies are open to this criticism, since they are likely to activate the same mechanisms that incentivise the perpetuation of the gendered division of labour. I do not aim, as such, to defend any of these policies, the overall evaluation of which—especially in the case of BI—depends on many normative and empirical considerations, including matters of feasibility and existing alternatives. My aim is to assess, and reject, a particular argument against them.

The next section offers some theoretical clarifications; the third presents the feminist debate about the care-supporting effects of BI. In the fourth section I elaborate on Arguments 1 and 2 and their implications. I end with brief conclusions.

Background clarifications: feminist egalitarianism

There is broad consensus that equality must govern the access to some goods, such as political and civil rights, and that avoidable inequalities in the distribution of other goods require justification. Examples of the latter include resources (like money), welfare (that is, how well one’s life goes), and capabilities (for instance, being able to move freely), and, possibly, relational goods such as social inclusion and non-stigmatisation.3 Some philosophers justify departures from equality in the distribution of such goods in cases where these departures result from individuals’ autonomous choices; others believe departures from equality are permissible if they are necessary in order to make the worst off as well off as possible. It is not obvious that any of these versions of egalitarianism—widely understood to include prioritarianism—involves, as such, any particular interest in whether more women than men are the victims of injustice, or in whether gender norms in particular (as opposed to other unjustified mechanisms) are responsible for inequalities. For instance, being a welfarist luck egalitarian—that is, endorsing equality of welfare—does not as such give you reason to find it particularly objectionable if women are disproportionately affected by unchosen inequalities in welfare due, for instance, to the gendered division of labour; on this view, what matters are inequalities in welfare between individuals. Being a prioritarian does not as such give you reason to object to the feminisation of poverty. And being a relational egalitarian—believing that justice requires that individuals relate to each other as equals—does not as such give you reason to be particularly concerned with women’s social marginalisation or stigmatisation. “As such”, because many egalitarians do in fact object to gendered inequalities; but, I assume, they do so in virtue of additional beliefs—for instance, concerning the role that a history of discrimination and oppression plays in explaining such inequalities.

By contrast, to be a feminist is to be interested in whether the group of those affected by injustice is gendered, at least to the extent to which inequalities can be attributable to gender norms. Feminism, too, comes in several varieties. A common denominator is the joint belief that (a) women and men are moral equals and therefore (b) women and men should have equal opportunities, at each stage of life, to access those goods the distribution of which is relevant to justice. Many feminists also endorse the view that (c) women should not fare worse than men, on the relevant conception of justice, due, even in part, to their sex. The last point goes beyond the request for equal opportunities: it states that women should not be worse off merely because they have different native competencies than men (if they do). So, for instance, if justice allows different compensation for different skills, and women end up being overly represented amongst the worse off because they tend to have native skills that receive worse compensation on the market, feminists may object to the fact that pay-offs are attached to marketable skills in a way that disadvantages women. Depending on their particular conception of justice, different feminists will identify different inequalities as objectionable. For instance, to qualify as feminist, a libertarian must object to women suffering worse rights violations than men; sufficientarian feminists must object to the feminisation of poverty, etc.

Egalitarian feminists’ interest in distributive justice is at least in part fueled by a sense that in many societies women constitute the majority of those denied their fair share—mainly in terms of wealth, status and power—and that this is due, in part, to the fact that they are women. This complaint goes beyond a complaint of unequal opportunities to occupy positions of advantage. If women and men had the same opportunities—in the robust sense specified at (b) above—but more women than men ended up as poor, marginalised etc., egalitarian feminists could have a double complaint. The first is that, in virtue of being egalitarians, they deem poverty unjust. But if this state of affairs is due to background economic rules and expectations that relegate care-givers to poverty, they also have grounds to complain in virtue of being feminists, even if women end up in caring positions via a distribution of work which is governed by equality of opportunity.

As specified by point (b) above, feminists believe that women and men should be able to compete unencumbered by socially created obstacles related to them being women or men. Some, such as Janet Radcliffe-Richards (2014), believe that this requirement fully captures feminist demands concerning the distribution of positions of advantage. I don’t deny that equality of opportunity between women and men at each stage of life is a worthy, and weighty, goal. But— as the arguments of this paper show—in societies that harbour large unjustified inequalities of outcome we should discount this widely-held ideal of gender justice when applied to positions of advantage. In such circumstances, feminists should, instead, strive to eliminate the feminisation of the worse off groups.

The feminist argument against care-supporting policies

Feminists have always been divided on the merits of care-supporting policies with respect to how they would affect women and gender equality. As already indicated, at least some of the disagreement is principled: such policies are likely to significantly improve, on the whole, the material situation, power and status of the worse off women, and in particular care-givers, while at the same time making it harder to achieve the goal of equal opportunities between women and men for the best jobs.

The debate about BI provides a detailed illustration of this feminist conundrum. BI would help lift from poverty women in jobs with low pay, few benefits and little security. If generous enough to provide financial independence, BI would provide exit to women who are in abusive domestic relationships and economically dependent on their partners. It would also protect from old age poverty women who have not joined the labour market (or left it prematurely) and who therefore do not qualify for welfare benefits (Whithorn, 2013; Robeyns, 2000; Elgarte, 2008). It would therefore benefit women who find themselves in these circumstances due to their caring responsibilities. At the same time, BI is likely to have a negative effect on women’s opportunities in the labour market and, possibly, on their access to positions of power and prestige (Robeyns, 2000; Gheaus, 2008).4 One of the earliest feminist concerns was that BI will diminish women’s participation in the labour market by making it affordable for more women to opt out of paid work—fully or partially—and instead do care work in the family (Orloff, 2013).5 Having withdrawn from paid work for a period and “specialised” in care work, it then becomes more difficult for women to access positions of advantage when they re-enter the labour market; more likely, such women will continue to be mainly care-providers or stuck in undesirable jobs. The status quo is already bad. In Europe, for instance, in spite of doing well, and sometimes out-competing men, in higher education, women make up the vast majority of part-time workers. They are a lot less likely than men to hold a job if they have children under 12, have an overall lower employment rate, earn less, and represent only a small fraction of executive boards’ membership or holders of top positions such as CEO (Eurostat, 2017; European Commission, 2016). Women also tend to do more voluntary work and informal, unpaid, care work, which is usually low status. As a result, they earn less and have less decision-making power than men as well as lower social status.

Some philosophers think that feminists should welcome policies that enlarge women’s opportunity sets even if as a result women will participate (even) less in the labour market. On this view, women’s lower market attachment is a morally innocent effect because it reflects individual women’s choices.6 This argument, however, is too quick—as two feminist rejoinders show.

Not always their (voluntary) choices

The first feminist rejoinder is that, at least sometimes, women’s decisions to prioritise caring for children, elderly parents or sick relatives do not carry the usual normative implications of autonomous choice because they are not made under the circumstances necessary for full autonomy. Women are (still) being socialised into assuming caring roles—with the help of gender-specific toys and bodily discipline, educational activities and generally by being pressured to comply with feminine social roles through active encouragements and/or through penalties. If so, perhaps women who choose part-time paid work or full-time home-making in order to look after family members in need for care don’t enjoy the same degree of autonomy that a person with a less gendered socialisation would enjoy in an otherwise identical choice context. It then seems unfair for such women to pay the full price of their choices in terms of loss of income, status and loss of other goods that one can only access through (good) jobs (Gheaus and Herzog, 2016). Policies are objectionable to the extent to which they sponsor imprudent choices for which the choice-makers lack full liability. The feminist worry that care-supporting policies will encourage some women to leave the job market in order to care more is indeed premised on the thought that this is can be an imprudent choice.

Recently, the gendered socialisation worry has taken one specific form, for which psychologists have coined a new term: stereotype threat. This is the fear that others will perceive you in light of negative stereotypes associated with the group with which you identify—in this case, gender—fear of which tends to inadvertently confirm the existing stereotypes (Steele and Aronson, 1995). There are reasons to think that women suffer from stereotype threat when performing within the traditional “male” territory and, as a result, perform worse than they would otherwise perform (and worse than men) precisely because they are expected to do so. This can explain, to some extent, women’s lower achievements in fields dominated by, and traditionally associated with, men (Stricker and Ward, 2004).7 It may also explain, in part, women’s higher willingness to leave the job market. Again, a policy sponsoring an imprudent choice, choice whose causal history that includes (possibly many) episodes of non-autonomous choices is, to that extent, objectionable.

Finally, a woman’s opting out of the labour market in order to care for someone in need of care can fail to qualify as a fully autonomous choice for an entirely different reason. Even someone who was free from any gendered socialisation, and unsusceptible to stereotype threat, could nevertheless have her autonomy curtailed because she finds herself unfairly saddled with a special duty of care. Individuals who become parents thereby acquire a duty of care towards their children; sometimes, foreseeably or not, one of the parents must switch to part-time work, or give up paid employment temporarily in order to care for the child, especially in societies where adequate daycare is unavailable or unaffordable. In cases when neither parent wants to do this, women, who are usually socialised as care-givers and who tend to earn less than men, face economic, and possibly psychological, pressure to be the ones who give up full employment (Okin, 1989, especially pp. 134-169; Allen, 2008). Moreover, due to path-dependency—because they breast-feed and need post-birth recovery leaves, and hence spend more time with infants—women tend to be more knowledgeable and competent with, and possibly more attached to, their young children than men. A mother’s higher level of competence and attachment does not mean that it is hers, more than a father’s, duty to opt out of (full) employment in such a case—the child’s interest may be sufficiently well served by the fathers’ hands-on care—but it contributes to the psychological pressure on mothers to be the ones who stay at home. And so, a woman’s decision to exit the labour market in order to live up to her caring duties is restricted by the lack of acceptable alternatives; when this is the case, the choice is less than fully voluntary.8 As Anne Alstott noticed (2004), because parenthood comes with a “no exit” clause—parents acquire the duty to provide continuity in care to their children—it significantly restricts parents’ autonomy once they have settled into the parental role. And it particularly restricts the autonomy of the parents who are the only hands-on care providers for their children.

The more general point here is that gender norms limit women’s autonomy by imposing higher opportunity costs on their, than on men’s, pursuit of parenthood (see also Mason, 2000): Combining parenthood with full-time employment is harder for women, who are socially expected to assume the caring role, who tend to earn less them men and who (subsequently) have more difficulties than men in finding a partner willing and capable to assume the role of main care-giver or, at least, to share this role equally. Policies that add to the incentive structure which keeps women away from the market, then, are not necessarily enhancing women’s autonomy.

Not always their choices

There is an even more important rejoinder to the claim that the negative effects of care-supporting policies on women are morally innocent: Such policies can lower the expected labour market opportunities of one group of women as a result of choices made by another group of women. Here, too, there are several complementary mechanisms to be considered. Women are typically associated with higher social skills, like being caring and competent to meet others’ needs, but not with competence in leadership and innovation skills. The retreat from the labour market of some women can reinforce these biases and thus reinforce explicit sexist beliefs against women in general. More controversially, it can also reinforce implicit biases—that is, a tendency to unconsciously associate certain characteristics with femininity and others with masculinity and to express approval when individuals’ behaviour displays the associations attached to their sex and disapproval when it does not (Brownstein, 2015). Scholars of implicit bias claim that virtually everybody harbours such biases, with the result that women are evaluated less favourably in the workplace than men who have the exact same qualifications and accomplishments (Steinpreis, Anders and Ritzke, 1999). Implicit bias may partly explain women’s lesser presence in the labour market in general and in some fields in particular, as well as women’s difficulties in breaking through various glass ceilings. Biases against women’s leadership qualities are much more likely to affect women’s opportunities to the more attractive positions. This is because, in existing societies, highly competitive individuals with leadership potential are believed to be the most qualified for the best rewarded positions.

If, as a consequence of a care-supporting policy, enough women retreat from the labour market in general, and possibly from some fields in particular, this will reinforce mechanisms that perpetuate general discrimination and marginalisation of women who apply for positions that require leadership abilities and high competitiveness. Hence, women other than those who respond to the policy will be even more likely to have their opportunities for positions of advantage curtailed by gender biases.

Further, there exists a fully conscious and rational mechanism that can impose on some women the price of other women’s choices to decrease their participation in the labour market: statistical discrimination. If women on average are more likely than men to quit their job or to reduce their labour participation when they have children, and if employee turnover is costly, then it is economically rational for employers to give preference to prospective male workers. Even nonsexist employers who are immune to implicit biases—assuming they exist—will respond with higher levels of statistical discrimination to women’s lower participation in the labour market (Bergmann, 1998). Take, for example, the case of hiring scientists who need expensive equipment tailored for the purposes of their research agenda. Each time a lab loses—permanently or for extended periods of time—such an employee they incur a great financial loss (some research projects become soon obsolete so they cannot be simply be put on hold during one’s care leave). With these kinds of jobs there are powerful economic incentives to avoid hiring individuals who are more likely to take significant leaves.

A policy that enables women to avoid the labour market therefore will also, indirectly, cause other women to pay for these choices with a more limited set of opportunities for positions in which employee turnover is costly. Thus, as in the case of implicit bias, statistical discrimination is more likely to operate against a woman the better the job to which she aspires: In existing societies such positions usually require high qualifications which, in turn, depend on a commitment to full-time careers. In contrast, the jobs where a high level of staff turnover is not particularly problematic for employers, because it is easy and cheap to train new workers, come with significantly lesser benefits.

Therefore, feminists rightly worry that policies that directly or indirectly protect care-giving will strengthen gender norms against women’s access to positions of advantage. These norms are bad for gender fairness, because they make women’s and men’s access to positions of advantage dependent, in a morally arbitrary way, on them being women or men. This would be serious criticism in circumstances characterised by background distributive justice, where the inequalities in the pay-offs associated with different jobs were just and in which the unemployed—or, at least, the involuntarily unemployed—were not at risk of undue poverty and social marginalisation. Yet, it has little, if any, weight in a world displaying unjustly large inequalities. The rest of the paper explains why.

Equality of opportunities to positions of advantage in unjust circumstances

When it comes to the distribution of jobs, there are two particularly popular versions of the principle of equality of opportunities (henceforth EO), both of which are meritocratic. One version requires that access to positions of advantage depend only on individuals’ talents and ambition. This is often called “formal EO», and prohibits sex-based discrimination. In a more egalitarian, and demanding, version, usually referred to as “fair equality of opportunity“ (henceforth FEO), the principle states that individuals‘ prospects should only be determined by native talent and ambition (Rawls, 2001, p.44). In this form, the principle also requires that girls’ and boy’s talents and ambitions be equally cultivated.

One feminist virtue of both versions is expressive: by prohibiting employers to give less consideration to an applicant merely on account of their sex, EO principles send the right message that women and men are moral equals. As I argue below, this expressive value can be realised without ensuring that all individuals have equal opportunities for positions of advantage.

The most convincing defense of the more egalitarian version of the principle, usually referred to as “fair equality of opportunity” (henceforth FEO)—is prioritarian. Rawls (2001), for instance, thought FEO regulates competition for social positions in societies that only tolerate those inequalities of outcome that are necessary to incentivise talented individuals to be socially productive.9 In such circumstances, FEO contributes to the maximisation of the social output which is then distributed according to prioritarian principles; this process maximises the prospects of the worst off. Yet, this defense of FEO depends on prioritarian redistribution actually taking place, and therefore implies nothing about the truth or weight of FEO in societies in such redistribution does not actually take place. Here I assume that existing and predicted levels of inequality in current societies fail to meet prioritarian distributive standards, or indeed any other standard likely to be supported by a purportedly egalitarian theory. They even fail to meet the more modest, sufficientarian standard, that requires that everybody has unconditional access to enough resources to lead a decent life.

We therefore have reasons to doubt that FEO has the same weight in unjustly inegalitarian societies as it has in perfectly just ones—or, indeed, that it is the correct principle regulating competition for positions of advantage. A full analysis of this matter is outside the scope of the present paper.10 To settle the question at hand, it is enough to examine the likely implications of applying EO to those kinds of positions to which women’s access is most likely to be negatively affected by strengthening the mechanisms that already obstruct gender equality. By “positions” here I mean roles with a set of rights, powers and prerogatives of office; thus, salary and political influence are not contingent add-ons to a position, but essential parts of it.

First, the very top positions come with amounts of economic reward, social power and status that nobody would enjoy in a just society. Some of the positions that lie above the glass ceiling which women find difficult to break are, plausibly, of this kind. The first argument applies to them:

Argument 1

• P1 Nobody has a right to a position that confers unjust advantages on those who hold it.
• P2 We ought to discount the value of (fair) equality of opportunity for positions to which nobody has a right.
• P3 Many of the top positions in societies with large unjustifiable inequalities of outcome confer unjust advantages on those who hold them.
• C1 Therefore, we ought to discount the value of (fair) equality of opportunity for many of the top positions in societies with large unjustifiable inequalities of outcome.

If C1 is true, improving FEO with respect to top positions cannot be an important goal of justice—and surely not one that is as important as improving the situation of women who are amongst the worse off due to their care role. I assume P1 is self-evident, and so I concentrate on defending P2. One reason why a principle of equal opportunity between women and men is so widely endorsed is that it is obviously unfair to deprive someone of opportunities to a good for morally irrelevant reasons such as one’s sex. Yet, this reason is only relevant in cases in which justice permits people to have some opportunity to achieve that good—that is, where it is morally permissible for someone to achieve that good. It is unclear that it is morally important to run meritocratically competitions for positions that nobody ought to occupy: nobody has a right to such a position. Ideally, such positions would be abolished; but, if and as long as such positions are up for distribution, a fairer way to distribute them seems to be by lotteries with a threshold of competence (to avoid hiring people that would inflict damage) and that give everyone an opportunity.

There seems to be something very prejudicial to be denied, specifically due to one’s sex, the same opportunities that others have even for goods that no one should have in the first place: As already noted, a strong reason to uphold equality of opportunity between women and men is expressive: If a person is denied, merely on grounds that she is a woman, an opportunity to a good G that nobody ought to have, she is given the message that she is worth less than a man. This gives grounds of complaint not only to her but, plausibly, to all women, since the expressive disvalue of discrimination affects women in general. This does not, however, necessarily mean that the woman in question has any claim of justice to more opportunities than she already has to G; if the very existence of G is unjust, she—like everybody else—should have zero opportunities for it. More plausibly, the woman in this case has a special claim (alongside the claim that we all have that G should be abolished) that any opportunities to G be also denied to her male competitors.

Perhaps a meritocratic organisation of competition for goods which nobody ought to enjoy has some unique value, which I have failed to consider here. (The expressive value is not unique to the meritocratic distribution, since this value could also be realised by lotteries or, as I explain presently, a quota system.) In this case, even in unjust societies, ideal feminist policies would include measures that offset the disruption of equality of opportunities between women and men for top positions. But, whether or not EO for top positions has any weight in unjust societies, the most significant progress towards justice can come from improving even slightly the situation of the worst off. This is a powerful argument in favour of care-supporting policies.

Here is a thought experiment to test intuitions about the value of improving gender parity for unjust advantages compared to improving the situation of the worst off. Any conclusion about its relevance to the issue at hand will, of course, depend on how you think about justice—especially economic justice—more generally and, therefore, on exactly how unjust you think are our present circumstances. Consider an ancient slave-holding societies that banned women from holding property and hence also from owning slaves. Assume that such a society contemplated a reform granting women the legal right to hold slaves, reform which wouldn’t have altered the number of slaves and slave-owners, but would have eliminated any gender inequality of opportunity to be a slave-owner. The reform seems to have some moral value because it eliminates one form of discrimination against women. But, because it achieves this by giving women opportunities to hold an unjust social position, 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. Therefore, the reform can possibly improve justice only if the demands of justice go beyond fully satisfying the rightful claims of all individuals involved.11 In any case, the real value of the reform seems to be expressive: it sends the message that women and men—even if only of a certain social class!—are moral equals. Yet, the reform may also have one kind negative expressive value, because it sends the message that women and men are equally responsible for injustice. The morally sound way of sending the message of women’s equality is to deny men, as well as women, the right to hold slaves—that is, to level down, not up.

Some may conjecture that the imaginary reform and, by analogy, reforms that would unblock women’s access to top positions have instrumental value in the form of trickle-down effects. In particular, they may undermine the association between women and powerlessness, and as a result improve the opportunities of women across the entire society. Yet, this consequence is uncertain (will the trickle-down really happen?) and it might be cancelled out by the other, negative, effects of associating powerful women with the wrong of slave-holding. Most relevantly, any desirable trickle-down effects would occur due to improved gender parity—that is, by an increase in equality of outcome rather than opportunity. If so, then they could be more directly realised through measures that put some women in prominent positions—such as gender quotas attached to lotteries. Any desirable consequences of having more women in top positions can be achieved through reforms—such as lotteries with quotas—that are much less demanding than achieving EO.

More generally, although it is important to undermine gender norms, not any means of fighting them is legitimate: denying care-supporting policies (assuming they are otherwise desirable in terms of justice) to disadvantaged women, in order to give other women more equal opportunities to the top positions is one of the worst ways to attempt to undermine gender norms. As well as being morally objectionable, such strategy is also likely to be inefficient: some norms concerning the gendered division of labour are likely to be strengthened by care-supporting policies (Robeyns, 2000; Gheaus, 2008). But other biases against women could be strengthened if an otherwise justified policy was opposed on grounds of equal opportunities between women and men for top positions: associating women with enjoyment of unjust benefits can easily backfire by fueling misogyny. It is safe to say, at the very least, that in unjustifiably unequal societies the first concern of justice is to make sure that existing social positions become more equal; regulating competition for top positions by EO seems much less important.

In defense of the second premise, the slavery example is meant merely as a heuristic to tease out beliefs about the intrinsic value of non-discrimination in access to unjust advantage. Its point, of course, is not to model an analogy to current societies. Slave-owning is unjust in part because it denies some people their rights to freedom from coercion, and liberals typically give priority to such rights—and their fair distribution—over the distribution of other kinds of goods. One may think that current societies in the most developed parts of the world thrive from the dire exploitation of some individuals whose lack of reasonable alternatives makes their situation, in some respects, akin to that of slaves. Examples include firms that rely on sweatshops employing individuals whose alternative is starvation, or firms paying less then subsistence wages to some workers (who, again, lack alternative) although they can afford to pay exorbitant wages to employees in top jobs. Perhaps some of the positions above the glass ceiling that is so difficult for women to crack actually depend on violations of the same rights against coerced exploitation that are violated slavery. It is far from clear that there is any value in improving equal opportunities to positions on the executive board of a company that relies on sweatshops.

But the argument does not depend on the belief that some of top positions involve rights violations on a par with slavery. It is enough to accept that current societies allow massive maldistributions of material goods that are, in some respects, akin to theft. One could run the thought experiment with other kinds of social injustices, which involve violations of lesser rights, such as the right to property. Imagine a group of individuals who trade in stolen goods and fail to give the women and the men in the group the same opportunity to enjoy the revenues. How valuable would it be to ensure that women and men have equal opportunities to share the loot? It is hard to see how individual thieves could have an entitlement to an equal opportunity to benefit from theft, or why denying such opportunities to women represents an injustice. Indeed, it would be odd if the law was to prosecute thieves not only for theft but also for failing to share the loot according to an EO principle—as it should do, if discrimination against women thieves was a violation of justice.

Even if the second analogy, too, is faulty, because the best positions in rich democratic countries are not akin to (benefitting from) theft, one can agree with the more modest, comparative claim defended here. If the distribution of economic goods, and, in particular, the sharply increasing levels of inequality and insecurity that affect lots of people are unjust, then some of the benefits attached to existing top positions would not exist in a just society. On most egalitarian accounts, in a just society there would be no wages and levels of status and power like those enjoyed by pop and sports stars, many CEOs and executive board members, or top employees in Silicon Valley. Yet, in existing societies, gender norms unfairly disadvantage women from having the same access to such benefits as men. We have much less reason to be concerned with equality of opportunity between women and men for such positions than with improvements in the situation of the worst off.

Some may think that, unlike the case of slavery or theft, it is not in itself morally wrong to enjoy pay-offs like those attached to the best positions available in current societies. Sometimes this belief comes from the intuition that it can be legitimate to hold competitions in which the highest prize is similar to a sport star salary while the lowest reward is less than a subsistence wage. I believe that the pull of this intuition depends on such competitions being optional, which makes people’s participation in them voluntary. Yet, current participation in the job market is not voluntary, since most of us depend on jobs for survival.12 Moreover, the best positions in existing societies come not only with enormous salaries, but also with levels of power and status that may render them impermissible even if the competition was entered voluntarily.

Therefore, people who, merely due to their gender, have lesser opportunities to top positions, lack a claim to more opportunities for such jobs. If so, then there is little reason to worry about how care-supporting policies set back the opportunities of the better off women relative to such positions. The wrongs that statistical discrimination and sexist biases inflict on women who aspire to top positions do not consist in depriving them of equal opportunities to the material goods, status and power of their male peers. The real wrong of, say, paying a male pop star significantly more than a female pop star is expressive, and so the female star’s special claim is that the male star’s salary be lowered, not that hers be increased.

If a consequence of care-supporting policies was the reinforcement of statistical discrimination, this would negatively impact on the opportunities of many more women than just would-be CEO, executive board members, pop and sports celebrities or top employees in Silicon Valley. Women who aspire to less extravagantly well-compensated occupations such as in law, medicine, economics, academia, research or politics—and, especially, at the higher ranks of these and many other professions—will also be negatively affected. Perhaps the benefits associated with some of these jobs would also exist in a just society. But note that it is likely that only very few positions in our society would exist in an egalitarian society. The rights, powers and prerogatives attached to most currently existing positions may be incompatible with just social relations because of the levels of power and status attached to even middle-class typical jobs.

Let me nevertheless concede that some of these positions of advantage would exist in a just society. The second argument concerns such positions and is about the distribution of opportunities for these positions across the entire population:

Argument 2

• P4 Most people who have a shot at positions of advantage are likely to already have had more than their fair share of opportunities for such position.
• P5 People lack a claim to more than their fair share of opportunities for positions of advantage.
• C2 Therefore, most people who have a shot at positions of advantage do not have a claim to more opportunities than they already have for such positions.

To defend P4: Because in existing societies most people are excluded from opportunities to positions of advantage, a majority of the women who fail to secure positions of advantage merely on account of their gender are likely to have already had more than their fair share of opportunities to such positions.13 How much more will depend on the relevant conception of equality of opportunity.14 Consider two possibilities: FEO requires that one’s opportunities should be determined by native talent and ambition alone. Yet, existing societies are far from giving the same nurturing to all children; not only gender but also class, race and many other factors contribute to unequal nurturing. Given the great inequalities in the cultivation of individuals’ talents, it is likely that many people who would compete for any particular position of advantage in a society that realised FEO cannot even enter that particular competition in existing societies. Their talents have not been cultivated enough to allow them to qualify formally. Alternatively, assume, more modestly, that equality of opportunity is satisfied by having careers open to talents; it is very plausible that factors such as friends, family, attendance of the right educational institutions and other forms of brute luck—rather than mere qualification—determine who is in the run-up for many positions of advantage.15 Given the great inequalities between individuals’ pedigrees, social connections etc., it is likely that many people who would compete for any particular position of advantage in a society that realised careers open to talents lack any chance in that particular competition in existing societies. Some will not even enter the competition, for lack of information. In other cases, individuals’ social connections, pedigrees etc. are not sufficiently good for them to be given real consideration for the job. If so, the majority of women who fail to obtain a desired job merely on account of their gender are likely to have already had more than a fair share of opportunities to that position thanks to a better than average education, social connections etc. Ensuring that there is no gender disparity of opportunity for positions of advantage would not result, in the absence of additional and more radical reforms, in the level playing field in the way required by the principle of careers open to talents—and even less so by FEO. Rather, it would mostly level the playing field between the already unfairly advantaged women and men.

Again, this is not to conclude that people who fail, merely on account of their gender, to secure jobs that confer unjust advantages on those who hold them, have no complaints of justice. But their complaint of justice is not a claim for more opportunities than they already had—which care-supporting policies may impede. Rather, they have a special claim of expressive justice that they not be given the signal that their interests are less important than men’s. This is a claim that the men who are favoured merely on account of gender have fewer opportunities than they currently have. This aim, again, can be achieved in the absence of realising EO—for instance, by adopting measures such as lotteries with thresholds of competence and gender quotas. Moreover, even if care-supporting policies entrench statistical discrimination against women, it is not obvious that, on the whole, their consequences for expressive justice will be negative. Precisely because they would sponsor some women’s decision to be care-givers, supporters of such policies believe that they have positive consequences in terms of expressive justice; they signal that care—the traditional “women’s work”—is valuable.16

The general conclusion from the two arguments is that we should discount the value of improving women’s opportunities to top positions and of improving most women’s opportunities for positions of advantage for which they stand a real chance. At the very least, in societies that seriously fail to realise any egalitarian redistributive ideal these aims cannot come close to outweighing the interests of disadvantaged women in care-supporting policies.

What about middle positions that come with decent, though not above average, pay, benefits and security? Examples may include—depending on the society—office jobs or jobs in nursing and teaching. In some places, some such positions—for instance accountancy or human resources—do not even require applicants to have university education, but, instead, an apprenticeship. German societies, for instance, still have desirable jobs that are based on (extensive) vocational education; therefore, it is more likely that individuals who did not enjoy more than average levels of opportunities throughout their lives may have a real shot at these positions. But the kind of statistical discrimination and biases against women that could be amplified by care-supporting policies are a lot less likely to further disadvantage women who compete for such jobs. First, because such jobs are already fairly feminised, a stronger association between women and care need not cause women to be perceived as less competent teachers or nurses. Second, jobs of this kind are often easier to regulate for employees’ protection, especially when the employer is the state (as it is the case, in Europe, with most schools and hospitals, and many office jobs). Third, and perhaps more importantly, when women in middle-grade jobs use care leaves or a basic income to temporarily exit the job market, this is unlikely to do the same harm to business as it is the case with jobs that require higher levels of training and specialisation. There will be little reason to discriminate against women in middle-grade jobs due to their involvement in care work.17

The last point also applies to lower-grade positions—those with benefits that are significantly below average. But even assuming that care-supporting policies would reinforce women’s association with domesticity, and thus have detrimental effects on equality of opportunity for these positions, it is not clear how much of a worry this is. The women whose interests are most likely to be directly advanced by the introduction of BI or care wages are single mothers (some of the poorest people in developed countries); elderly women who spent much of their adult years caring for children, old parents and sick relatives and who, as a result, have meagre pensions; and, in the case of BI, women migrants with little education and social capital. These are not the women whose opportunities are most likely to be negatively affected by implicit and explicit biases or statistical discrimination. Women from poor backgrounds (who are often also from immigrant backgrounds) are some of the least likely to stand a chance to be employed in positions of advantage or in the higher ranks of any profession. The glass ceiling is not a typical problem for women who are chronically unemployed, or in jobs from which they can be easily fired whether or not they are pregnant or saddled with care responsibilities. For the same reasons, these are also the women whose socialisation into caring roles is least likely to deprive them from the non-material benefits of paid work, since they have very few opportunities for jobs with such benefits in the first place.18

Further, a large proportion of the low-grade jobs are excessively repetitive and boring, demanding in terms of time, and often organised in unnecessary and oppressive hierarchies. They seem mostly undesirable and, therefore, better opportunities for them seem less beneficial than a secure, substantial and unconditional monthly income or other types of support for care-givers.

Another concern is the effect of care-supporting policies on men’s opportunities to engage in care work on a par with women. A fully spelled-out ideal of gender justice requires that men have the same access as women to objectively valuable goods such as raising a family, and the same opportunities to avoid bads like exploitative or meaningless work (Gheaus, 2012). The effect of women’s exit from the labour market may be to strengthen those gender norms which make it very hard for men to be full time, and sometimes also part-time, care-givers: There is peer pressure on men to behave less caringly than women, there are also ridiculing stereotypes of men who do “women’s work”, and men receive less training as potential care-givers.19 Implicit biases against men and statistical discrimination against their presence on the sites of childrearing and in professions that involve direct access to children (like baby-sitter or nursery worker) can make it difficult for them to gain access to some caring professions. But if, in spite of their limited opportunities in this domain, men are overall better off than women, their interest cannot be more weighty than the interest of the worse off women in care-supporting policies. This being said, there is a remaining concern those men who are among the worse off members of society and who would like to take on caring roles;20 stronger gender norms can make it even harder for such men to access paid care work. I assume they are a very small minority, and therefore this worry does not constitute a reason to oppose care-supporting policies; but it might constitute an additional reason for states to undermine the gendered division of labour by other means.

Therefore, care-supporting policies are least likely to have net negative effects on the worst off women. Advancing the interests of these categories of women carries significant weight—making the case in favour of care-supporting policies stronger than the case against them. These policies include a (suitably-designed) BI, long subsidised care leaves, or split paychecks. When it is possible to means-test care-supporting policies (obviously not in the case of BI but perhaps in the case of care leaves), doing so might be best suited to advance all the feminist goals. Means-testing would avoid incentivising women to stay away from the labour market, and thus would protect EO at the same time as helping the worse off.21

Conclusion

To the extent to which the introduction of care-supporting policies would bring systematic and enduring improvement for the worse-off women, the fact that it would also set back equality of opportunity between women and men to positions of advantage does not raise a weighty concern. If women make a majority of the disadvantaged members of society, the rebuttal of the equality of opportunity criticism to these policies can be upheld by egalitarian feminists in virtue of their feminist convictions, as well as in virtue of their egalitarian convictions.

The relevance of the present argument goes beyond its implications concerning the introduction of care-supporting policies. An influential strand in contemporary feminism is driven by the ideal of equal opportunities between women and men for positions of advantage. Yet, this ideal has very limited, if any, weight, unless and until two other goals are realised: first, ensuring that the pay-offs associated with positions of advantage are permissible, by bringing inequalities of outcome in line with the requirements of justice. The second goal is to ensure that individuals enjoy equality of opportunity at all stages of their lives. 

Acknowledgements: For discussions and feedback on previous drafts of this paper I am grateful to Gustaf Arrhenius, Erin Beeghly, Juliana Bidadanure, Simon Birnbaum, Elizabeth Brake, Amanda Cawston, Stephanie Collins, Gijs van Donselaar, Thomas Douglas, Julieta Elgarte, Mollie Gerver, Jan-Cristoph Heilinger, Lisa Herzog, Beth Kahn, Tim Meijers, Eva Maria Parisi, Tom Parr, Diana Popescu, Ingrid Robeyns, Gina Schouten, Adam Swift, Andrew Walton, Nina Wild, Andrew Williams, two anonymous referees for this journal and to audiences in Amsterdam, Barcelona, Bern, Florence, Hamburg, Leiden, Munchen and Stockholm. I am particularly thankful to John Baker for our years-long conversation about feminism and care-supporting policies.

References

Anita Allen, “Rationalising Oppression”, Journal of Power 1(1):51-65, 2008.

Anne Alstott, No Exit: What Parents Owe Their Children and What Society Owes Parents, Oxford: Oxford University Press, 2004.

Harriet Baber, “Two Models of Preferential Treatment for Working Mothers”, Public Affairs Quarterly 4(4):323-334, 1990.

John Baker, Arguing for Equality, London: Verso, 1987.

———, “All Things Considered, Should Feminists Embrace Basic Income?”, Basic Income Studies 3(3):1-7, 2008.

Barbara Bergmann, “The Only Ticket to Equality: Total Androgyny, Male Style”, Journal of Contemporary Legal Issues 9(75):76-86, 1998.

———, “Long Leaves, Child Well-Being, and Gender Equality”, Politics and Society 36(3):350-359, 2008.

Michael Brownstein, “Implicit Bias”, Stanford Encyclopedia of Philosophy, 2015. Xx partial ref?

“Gender statistics”, Eurostat, February 2017. https://ec.europa.eu/eurostat/statistics-explained/index.php?title=File:Gender_statistics_Table1.PNG&oldid=328445 [Accessed 18 June 2020].

“Gender balance on corporate boards”, European Commission, July 2016.

Julieta Elgarte, “Basic Income and the Gendered Division of Labour”, Basic Income Studies 3(3), 2008.

Anca Gheaus, “Fair equality of opportunity in unjust circumstances”, manuscript.

———, “Basic Income, Gender Justice and the Costs of Gender-symmetrical Lifestyles”, Basic Income Studies 3(3):1-7, 2008.

———, “Gender Justice”, Journal of Ethics and Social Philosophy 6(2), 2012.

———, “The feasibility constraint on the concept of justice”, Philosophical Quarterly 63(252):445-464, 2013.

———, “Three cheers for the token woman!”, Journal of Applied Philosophy 32(2):163-176, 2015.

———, “Hikers in Flip-Flops. Luck Egalitarianism, Democratic Equality and the Distribuenda of Justice”, Journal of Applied Philosophy 35(1):54-69, 2018.

Anca Gheaus and Lisa Herzog, “The Goods of Work (Other than Money!)”, Journal of Social Philosophy 47(1):70-89, 2016.

Janet Gornik and Marcia Meyers, Families That Work. Policies for Reconciling Parenthood and Employment, New York: Russell Sage Foundation, 2003.

Colin Macleod, “Equality of Opportunity and the Consequences of Choice”, (manuscript).

Andrew Mason, “Equality, Personal Responsibility, and Gender Socialisation”, Proceedings of the Aristotelian Society 100(3):227-246, 2000.

———, Levelling the Playing Field : The Idea of Equal Opportunity and its Place in Egalitarian Thought, New York: Oxford University Press, 2006.

Claire Cain Miller, “When Family-Friendly Policies Backfire”, New York Times, 26 May 2016. https://www.nytimes.com/2015/05/26/upshot/when-family-friendly-policies-backfire.html [Accessed 18 June 2020].

Susan Moller Okin, Justice, Gender and the Family, New York: Basic Books, 1989

Serena Olsaretti, Liberty, Desert and the Market, Cambridge: Cambridge University Press, 2004.

Ann Orloff, “Why Basic Income does not Promote Gender Equality”, in Karl Widerquist, Jose Noguera, Yannick Vanderborght and Juregen De Wispelaere (eds.), Basic Income: An Anthology of Contemporary Research, Chichester: Wiley-Blackwell, 2013, pp. 149-152.

Janet Radcliffe-Richards, “Only X%: The Problem of Sex Equality”, Journal of Practical Ethics 2(1):44-67, 2014.

John Rawls, Justice as Fairness: A Restatement, Cambridge, Massachusetts: Harvard University Press, 2001. XX place

Ingrid Robeyns, “Hush Money or Emancipation Fee? A Gender Analysis of Basic Income”. In R. van der Veen and L. Groot (eds.), Basic Income on the Agenda. Policy Objectives and Political Chances, Amsterdam: Amsterdam University Press, 2000, pp. 121-136.

———, “Will a BI do justice to women?”, Analyse und Kritik 23:88-105, 2001

Thomas Scanlon, Why does Inequality Matter?, Oxford: Oxford University Press, 2018. XX Place

Gina Schouten, Liberalism, Neutrality, and the Gendered Division of Labor, Oxford: Oxford University Press, 2019.

Rhea E. Steinpreis, Katie A. Anders and Dawn Ritzke, “The Impact of Gender on the Review of the Curricula Vitae of Job Applicants and Tenure Candidates: A National Empirical Study”, Sex Roles 41(7):509-528, 1999

Claude M. Steele and Joshua AronsoN, “Stereotype threat and the intellectual test performance of African-Americans”, Journal of Personality and Social Psychology, 69:797-811, 1995.

Gijsbert Stoet and David C. Geary. “Can stereotype threat explain the gender gap in mathematics performance and achievement?”, Review of General Psychology 16:93—102, 2012.

Lawrence J. Stricker and William C. Ward, “Stereotype threat, inquiring about test takers’ ethnicity and gender, and standardized test performance,” Journal of Applied Social Psychology 34:665-693, 2004.

Yannick Vanderborght and Karl Widerquist, “The Feminist Response to Basic Income”, in K. Widerquist, J. Noguera, Y. Vanderborght and J. De Wispelaere (eds.), Basic Income: An Anthology of Contemporary Research, Chichester: Wiley-Blackwell, 2013, pp. 142-144.

Ann Whithorn, “Is One Man’s Ceiling Another Woman’s Floor?”, in K. Widerquist, J. Noguera, Y. Vanderborght and J. De Wispelaere (eds.), Basic Income: An Anthology of Contemporary Research, Chichester: Wiley-Blackwell, 2013, pp. 145-148.

Joan Williams, Unbending Gender: Why Family and Work Conflict and What to Do About It, New York: Oxford University Press, 2000.

Toru Yamamori, “A Feminist Way to Unconditional Basic Income: Claimants Unions and Women’s Liberation Movements in 1970s Britain”, Basic Income Studies 9(1-2):1-24, 2014.