Reproductive Work and Productive Fairness

Abstract

In academic and public debates, defenders of the case for sharing the costs of children often claim that by having and rearing children parents produce public goods for the rest of society, or perform socially valuable or necessary labour, and that it would be unfair to parents for others to not share the costs of children, for example through publicly funded parental leave and schools for this reason. Critics of the public goods argument have claimed that it fails because there is no defensible principle that can serve to buttress the claims of parents. Furthermore, these same critics, as well as others, have argued that a certain view of liberal equality militates against sharing the costs of children.This paper challenges both the contention that there is no defensible principle of fairness that can buttress the public goods argument and the contention that liberal equality militates against socialisation. It accomplishes these two tasks together, arguing that the ideal of equality of resources, rather than militating against socialisation, in fact grounds a principle of productive fairness that can serve as the normative premise of the public goods argument for sharing the costs of children.

1 Introduction

In public and academic debates, defenders of the view that other citizens must share with parents the costs of raising children (“socialisation”, for short) often advance the following claim: by having and rearing children parents produce public goods for the rest of society, or perform socially valuable, indeed necessary, labour. It would thus be unfair to parents who perform this type of labour if other citizens did not share the costs of children, for example through publicly funded parent al leave and schools, to mention but two of the standard family policies that many welfare states enact.¹ Whether a version of this fairness-based case for sharing the costs of children (which here I loosely refer to as the “public goods argument”) can be defended depends, among other things, on whether there is a defensible normative principle that can vindicate the charge of unfairness: why exactly is it unfair to producers of public goods or those who engage in socially valuable activities, for beneficiaries not to share the costs of production of those goods? And can this unfairness charge be made on behalf of parents?

Critics of the public goods argument have claimed that it fails because there is no defensible principle that can serve to buttress the claims of parents.² Furthermore, these same critics, as well as others, have argued that a certain view of liberal equality militates against socialisation (i.e., the sharing of costs between parents and non-parents). The view in question, equality of resources, first formulated by Ronald Dworkin, is alleged to permit or require that all of the costs of children fall on their parents, assuming that parents do not view being parents as a predicament they regret having to face, and that they are not at the short end of other injustices (e.g., they are not unjustly poor adults, or unjustly disadvantaged women).³ If these critics are right, equality of resources deems parenting to be “an expensive taste”.⁴ Because equality of resources is an attractive ideal of equality, it is important to determine what it entails about socialisation. Moreover, like almost all egalitarians, Dworkin assumes that a just society is one which distributes some of the costs of children among citizens at large, regardless of whether they are parents, through policies such as publicly funded child-care and schools, and by requiring that we all pay for all our fellow citizens’ lifetime shares. Ascertaining whether equality of resources tells against socialisation, then, is important for determining the internal consistency of this view.

This paper challenges both the contention that there is no defensible principle of fairness that can buttress the public goods argument and the contention that equality of resources militates against socialisation; it accomplishes these two tasks together, in that its central claim is that the ideal of equality of resources, rather than militating against socialisation, in fact grounds a principle of productive fairness that can serve as the normative premise of the public goods argument for sharing the costs of children. (Note that, for the purposes of this paper, I will simply assume that, with regard to a certain number of children, the crucial empirical premise of the public goods argument holds true—i.e., that a certain number of children constitute or create public goods of the relevant kind.⁵)

Productive fairness, broadly conceived, demands that, under just background conditions, if one is a producer of a benefit which others need or want, this fact is normatively significant for how institutions that distribute burdens and benefits should be set up: in short, it makes a difference to one’s claims against others. The interpretation of productive fairness I offer in what follows is an alternative to the familiar desert-based and producer-entitlement interpretations of productive fairness. On desert-based interpretations of productive fairness, producers deserve rewards as a fitting response to a certain desert basis (whether this is effort expended, contribution made, or morally praiseworthy traits). On a producer entitlement-based interpretation of productive fairness, individuals acquire ownership rights over previously unowned resources that they labour on (Locke, 1988, Ch. 5, Sects. 27–8). On the interpretation I propose, the justification for why being a producer makes a difference to one’s claims is that this is required by equality; accordingly, I will refer to this as egalitarian productive fairness.

The paper proceeds as follows. Section 1 frames the problem I am concerned with more fully. It explains why accounting for the unfairness of not sharing the costs of reproductive work is important and why it appears difficult to do so within a liberal egalitarian framework that accepts the key commitments of equality of resources. In particular, it identifies two stumbling blocks we face in showing this—Market prices for ambitions and No economic rents. This section also explains why appeal to gender inequality in the division of unpaid labour does not fully capture the unfairness under discussion. Section 2 engages with Market prices for ambitions, which holds that, when people differ only in their ambitions, egalitarian justice requires them to pay the market price of their ambitions. The section first briefly sketches the equality of resources ideal and then shows why the commitment that we hold people liable for their ambitions, which is at the heart of that ideal, tells in favour of requiring individuals to pay for the market price of their ambitions only sometimes, and sometimes—where ambitions are productive and the goods produced are non-commodifiable—it can recommend quite the opposite, i.e. that the price which producers pay for the resources they use be an adjusted price. This is what I refer to as a requirement of egalitarian productive fairness. Section 3 engages with No economic rents, which holds that, absent considerations of brute luck disadvantage, on equality of resources only incentive-based considerations tell in favour of allowing inequalities in income or wealth between persons and that to suggest otherwise, as I claim egalitarian productive fairness does, is to recommend, implausibly, that people should be paid economic rents. I argue that this challenge fails: there are no reasons of justice in favour of rewarding people differently depending on whether they are willing to hold out for incentives, and there is an important reason against permitting unequal incentives of this kind, which is that the egalitarian ideal would otherwise fail to treat neutrally different ambitions, regardless of whether they are self-interested or other-regarding.

If these arguments are successful, a version of the public goods argument can be defended which is integrated within the ideal of equality of resources.

2 The Problem of Disadvantaging Reproductive Work

During the Covid pandemic, talk of the “Crisis of Care” has made more prominent in mainstream debates claims which feminists have been making for decades, concerning the need to do justice to unpaid caring work, of which reproductive work is a sizeable part. I take “reproductive work” to refer to having and raising children; it is work insofar as it is an activity that has a purpose beyond itself (parents do not engage in their parental tasks for their own sake, because they enjoy performing them, but in order to raise their children well; it is thus different from leisure), the performance of which involves substantial costs (among other things, in terms of time and financial resources), and which, directly and indirectly, serves the needs of others. Reproductive work is thus characterisable as “a socially organized contribution to a larger group” (Gomberg, 2018: 514). Those who do most of this work (women) suffer a number of disadvantages as a result of doing it; though these vary across countries and in line with other factors such as class and race, they are wide-ranging and include an earnings gap, a pension gap and greater vulnerability to poverty.

It is no wonder, then, that feminists of various stripes—Marxists, libertarians, liberal—have called attention to the problem of reproductive work’s being unpaid and/or disadvantageous for those who carry it out. From Marxist quarters, for example, Silvia Federici has long called on the need “to reappropriate that money which is the fruit of our labor—of our mother’s and grandmothers’ labor” (Federici 2020: 5).⁶ Writing from a very different, economic liberal, perspective, Shirley Burggraf notes that “[w]hen we become conscious of the family as the socially productive institution that it is, one that has for centuries absorbed much of the human race’s energies and resources in order to perform essential social and economic functions, the question of how to reverse the current disinvestment in the family appears critical” (Burggraf, 1997: 7). Legal philosopher Martha Fineman writes that “[c]aretaker’s…labor should be treated as equally productive even if unwaged, and should be measured by its societal value…Caretakers provide a subsidy to the larger society and its institutions…Caretaking…creates a “social debt” (Fineman, 2004: xv & xvii). Nancy Folbre, a feminist economist who has argued sustainedly over the last three decades to defend the public goods argument for sharing the costs of children, writes, about contemporary liberal democracies (especially the US), that “current policies …have, in a sense, exploited parents by taxing the younger generation to help finance the spending of the older generation with little regard for those who devoted time and money to raising those taxpayers” (Folbre 2008: 7). In writings for the general public, the writers of Feminism for the 99%, highlight the need to “…[make] visible the indispensable role played by gendered, unpaid work in capitalist society… and call for “generous public support for social reproduction” (Arruzza et al., 2019: 8 & 40); The Care Collective note that “Invisible, undervalued, exploited care labor is everywhere” (The Care Collective 2020: 3 & 11). The list of writings by feminists pointing to the need to do justice to reproductive work could go on and on.

Turning to liberal egalitarian theorists of justice, they have typically formulated principles for regulating societies made up of individuals whose creation, care and numbers are taken as given (see, for example, Rawls 1999; Dworkin, 2000; the same is true for some libertarians; see Nozick 1974). When pressed by feminists, to be sure, they have acknowledged the importance of reproductive work. So, for example, engaging with Susan Moller Okin´s challenge to A Theory of Justice (Okin, 1989), Rawls acknowledges that “reproductive labor is socially necessary labor” (Rawls 2001: 162).⁷ However, Rawls does not elaborate what follows from the fact that reproductive work is socially necessary, other than by saying that women should not have fewer opportunities than men as a result of doing it (a point I come back to at the end of this section). Moreover, while, like most liberal egalitarians, Rawls assumes that the costs of children should be shared between parents and other citizens to some degree,⁸ neither he nor the other main theorists of justice have set out to make a case for this assumption (see Olsaretti, 2017 for why this view is standardly assumed).

Upon inspection, furthermore, liberal egalitarian theories might appear inhospitable to the case for sharing the costs of children with parents. On Rawls’ view, for example, unless a case is made for treating reproductive work on a par (for some purposes) with productive work rather than as leisure, parents who are economically poorly off as a result of using their resources to have and raise a child would not be regarded as unjustly badly off; they would be deemed to be in the same position as those who are economically poorly off because they chose to surf in Malibu: poor in income, but rich in leisure, which, like income, is a primary social good. Similarly, parents who work a double shift in order to not be among the economically worst off while caring for their children would also be viewed as not being unjustly disadvantaged: they would be comparable to someone who works double shifts in order to finance an expensive hobby.

These remarks make clear that, for those working within liberal egalitarianism, making a case for sharing the costs of children by appealing to the fact that reproductive work is socially valuable work faces two stumbling blocks.⁹ The first, Market prices for ambitions, is the claim that insofar as people freely choose to be parents, or identify with this lifeplan, there is no egalitarian objection against disadvantage accruing to parents as a result of that choice of lifeplan, and more specifically, that it is compatible or required by equality to hold them liable for the market price of their ambition. The second stumbling block, No economic rents, is the claim that, absent forward-looking, incentive reasons to get producers to produce, there is no egalitarian reason to reward or compensate producers.¹⁰

These two stumbling blocks are made harder to surmount by the fact that some ways of working our way around them are barred by a neutrality-respecting variant of liberal egalitarianism. On this view, we may not appeal to judgments about the good which some citizens could reasonably reject. So, we may not appeal to welfarist considerations to argue that, assuming that sharing costs with parents’ ambitions to have children is necessary to secure for them the same opportunity for welfare as that which we secure for the Malibu surfer by not charging him to access the beach, then the parent should not be held liable for the fact that his ambition is costlier to satisfy. (This line of argument meets with the well-known “expensive tastes” objection: equality does not require compensation for those who choose, or identify with, costly lifeprojects.¹¹) We may also not appeal to perfectionist considerations to argue that as a community we are justified in subsidising parenting but not surfing, and do so independently of what incentives are needed, because parenting is a worthwhile life project that merits protection in a way that surfing does not. Finally, I assume that neutrality prevents us from turning to principles of desert (these reflect judgements about value which not everyone)¹²; and that a commitment to the equal moral status of children prevents us from adopting a principle of producer entitlement, in order to justify removing disadvantage attaching to the role of parents even when this is not called for by forward-looking considerations.¹³

With our hands tied in this way, it is tempting to think we should abandon the attempt to formulate a defensible version of the public goods argument for sharing the costs of children, and instead offer an account of the injustice of disadvantaging those who do reproductive work solely through the lenses of concerns about gender inequality. On this way of thinking, instead of arguing that it is unjust for reproductive work to be costly for those who undertake it, we should focus on arguing that it is unjust that this work be distributed unequally along gender lines.

Notice that, on this gender inegalitarian reading of the problem of reproductive work, it is unjust for reproductive work to be disadvantageous only if and because women do it disproportionately. That women do the lions’ share of reproductive work is unjust insofar as it is a product of inegalitarian gender norms and insofar as it hampers women’s equal opportunity to compete for jobs and others positions of advantage. This implies that the failure to share, at society-wide level, the costs of reproductive work (e.g., the absence of publicly funded parental leave and childcare) may be unjust, but only derivatively so. That is, not sharing the costs of children, society-wide, is unjust only if and because it is important to diminish the comparative disadvantage of women relative to men.

Accordingly, the gender inegalitarian reading of the problem of reproductive work need not recommend sharing the costs of children in a society in which all man-woman households with children realise the “equal split” of reproductive work, defended by Okin (1989); or in a society in which there is what Richard Arneson calls statistical equal split (i.e. taking all households, aggregatively, women and men come out as sharing equally reproductive work and paid work; see Arneson, 1989). It need not recommend sharing the costs of children if households realise an equal split, even if these households are considerably less well off, in terms of income and leisure, than households without children. The gender inegalitarian reading would also not recommend sharing the costs of children in the event that the gender egalitarian goals mentioned above (ensuring that men and women share in domestic work and ensuring true equality of opportunity for women and men alike outside the family) could be achieved in some other way, e.g. if funded parental leave and childcare could be provided by levying a hefty family tax (or, largely equivalently, by requiring fathers to pay for these provisions). The gender inegalitarian reading of the problem of reproductive work as characterised above, then, differs from the public goods argument this paper defends, both in its diagnosis about what is unjust about the failure to share the costs of children with parents and possibly in the means it prescribes to overcome this injustice.

Before moving on, let me emphasize that it seems to me clearly true that gender inequality is an important component of the injustice in our current arrangements around reproductive work. There is, in my view, no doubt about this. Still, the gender egalitarian reading above does not fully capture that injustice.¹⁴ What exercises feminists like the ones mentioned at the beginning of this section, and rightly so, are two things, not one, and these two things, while closely tied together, are independent of each other.¹⁵ One is the fact that there is an unequal and gendered division of reproductive work, where the injustice in question is not reducible to the unfairness of reproductive work’s being costly; the other is that it is unfair that reproductive work is costly, where the unfairness in question is not reducible to the injustice of that work’s being unequally distributed along gender lines. The second claim, as much as the first one, needs vindicating. If it were not vindicated—if the injustice of the gendered division of reproductive work were fully reducible to the injustice of that work’s being unequally distributed along gendered lines—then feminists would have no more reasons to condemn the unfairness of women’s doing the lion’s share of reproductive work than they would of condemning the injustice of a social scheme in which women are confined, say, to the unpaid work of counting grains of sand, or more realistically, the injustice of women’s incurring substantial material and health costs which men do not in order to live up to gendered beauty norms.

But I believe they do have reasons beyond these: it is unjust to women that they do the bulk of unfairly costly unpaid work—work which is unduly costly, despite being, in numerous ways which feminists have identified, socially valuable, indeed necessary, work. If we accept this claim, liberal egalitarians have strong reasons to search a principle of fairness for a version of the public goods argument that surmounts the two stumbling blocks identified above, while also remaining within the confines of neutrality and that avoids appealing to producer entitlement.

3 Market Prices for Ambitions and True Opportunity Costs

In searching for a principle of productive fairness, it is helpful to offer a brief characterisation of the ideal of equality of resources within which the ideal of productive fairness I defend is embedded. As I mentioned earlier, the common reading of that ideal to date has neglected the importance of productive fairness and has taken equality of resources to be opposed to the case for sharing the costs of children.¹⁶ This is, in my view, a mistake.

According to equality of resources, equal resources must be devoted to the ambitions of each person (to their preferences, projects, or lifeplans). This has implications first, for how we should respond to unequal distributions of resources between people that are traceable to brute luck, and second, about how people who differ only in their ambitions, rather than as result of brute luck, should fare.

With regard to brute luck, equality of resources requires the mitigation of factors outside of people’s control that unequally affect the resources with which they can pursue their ambitions. It is inconsistent with equality of resources, for example, if someone has ample resources with which to pursue his lifeplans while another person does not only because brute luck has conferred on the first person, and not on the second, a marketable talent, so that the first enjoys a lucrative occupation while the second struggles with unemployment. Since the kinds of resources whose equal distribution matters include “personal” resources (and not only “impersonal” resources, such as land or other natural resources), it is also inconsistent with equality of resources if one person is blessed with good health while the second has unmet health needs.

Besides the mitigation of unequal brute luck, equality of resources also imposes a constraint on how resources should be distributed between people who differ only in their ambitions. (This point about the ideal of equality of resources is much less often noted, but it plays a crucial role in my argument below.) It requires that each person be assigned a bundle of resources according to the opportunity costs to others, judged by their lights, of his possessing that bundle—that is, according to how much those others have to give up for him to have that bundle of resources. How much land I should possess, for example, should depend on the value others place on that land, so that, if they value it greatly and I want more land than them, I must forgo, for their sake, a lot of other resources, so that our overall bundles of resources are equal. Following Dworkin, we can say that equality of resources requires that we use an “opportunity cost” metric to determine the value of impersonal resources people use to pursue their ambitions (or, as I sometimes put it in what follows, to determine the price or the payoffs of ambitions). The underlying rationale for using this “opportunity cost” metric is that this shows equal respect to all persons as valuers.

To see the implications of this view of what the price or the payoffs of different ambitions should be, in the name of equality, consider a simple example. It would be inconsistent with equality of resources to transfer resources from a person who has not gambled to a person who has gambled and lost. To do this would be to give the gambler more resources than the non-gambler in the following sense: the non-gambler costs less, so to speak, to the gambler than the gambler does to the non-gambler, judged by what each has to give up for the other’s ambition to be realised. (As I explain shortly, this commitment to price ambitions in terms of their opportunity costs to others is what accounts for productive fairness considerations, i.e. for why producers of what others want or need can acquire distinctive claims.) This conviction is often captured by referring to egalitarianism’s commitment to hold people responsible for (some) consequences of their choices.

I believe it is because of a misunderstanding about what the commitment to holding people liable for their ambitions entails that the resourcist egalitarian ideal is sometimes seen as inhospitable to the case for socialising the costs of children. The misunderstanding consists in upholding Market prices for ambitions, that is, in believing that equality requires people to pay the market price of resources they use except if they need those resources to offset brute luck disadvantage. Justine Burley, for example, writes: “Given the high costs associated with having children it is clear that the preference for offspring amounts to what Ronald Dworkin calls an expensive taste. Dworkin believes that ministering to expensive tastes is counter-intuitive from an egalitarian point of view…” (2000, 138). Matthew Clayton remarks: “it is difficult to see how individuals who choose to found a family have a justified case for resource transfers from those who prefer to remain childless. Any inequality of opportunity or income that is the product of having children compared to remaining childless is not the product of unequal brute luck, but rather the result of a difference of ambition.” (2006, 69).¹⁷ Defenders of this line of argument can remark: just as it would be unfair to non-gamblers to subsidise gambling or to non-religious people to subsidise religion, so it would be unfair to non-parents to subsidise the parental project. Equality of resources can only support sharing the costs of children if it were shown that parents are at the short end of an inequality in brute luck.

This reasoning, however, is mistaken. From the fact that parenting is an ambition with which parents identify, it does not follow that parents must pay the market price for that ambition. What follows, instead, are the following two things. First, within equality of resources, the fact that parenting is an ambition serves to disqualify parents from adducing as one possible ground of cost-sharing, that this is required as compensation for a “handicap” (i.e. for being at the short end of an inequality in brute luck).¹⁸ Second, from the fact that parenting is an ambition, it follows that those who exercise this ambition should bear the opportunity costs, to others, of resources being dedicated to that ambition rather than being put to alternative uses. This is the metric of opportunity costs I mentioned just above. Sometimes the market prices of our ambitions and the opportunity costs to others coincide. But this is not always so. Where they do not, the very principle that sometimes tells in favour of asking people to pay the market prices of their ambitions requires the opposite, i.e. adjusting the price of a resource, or, as we might put it, “subsidising” the use of a resource for a certain purpose, even if it is used in the pursuit of an ambition. The underlying principle, recall, is that, when people differ only in their ambitions (as opposed to in their brute luck), treating people equally requires that what people should pay for the resources they need to pursue their ambitions should be measured by the lights of what their appropriating those resources deprives others of, or by the lights of what others have to give up, as judged by those others. Where the ambitions that people have are productive, i.e. ambitions the exercise of which produces what others need or want, the opportunity costs metric supports the judgement that treating producers fairly requires that they be rewarded or compensated, or equivalently, that the price they pay for the resources they use to produce what others need or want be adjusted. If producers were made to internalise the full costs of production, and were not able to recoup those costs, they would be overcharged for the resources they use.

In what follows I illustrate and elaborate these points and apply them to the case of parents. I proceed by first considering the production of commodifiable goods and then turn to that of non-commodifiable goods, which include those parents produce by having and raising children. (Let me anticipate that the differences between these cases are discussed at the end of this section; because my argument rests on their similarities, it is these I focus on first.)

Consider, first, a case involving Adrian and Bruce, who differ only in their ambitions regarding how to use a previously unowned piece of land, a scarce resource which the community must allocate justly.¹⁹ While Bruce wishes to use it as a tennis court for himself, Adrian wishes to use it to produce tomatoes, which others want. How should the land be divided between Adrian and Bruce? Recall that equality requires that the price individuals pay for their ambitions reflect the true opportunity costs to others of their pursuing those ambitions. The opportunity costs are the potential benefits others miss out on—what they have to forgo—as a result of resources being assigned to one person rather than another. So, the benefits as well as the costs that result from each person’s appropriating a scarce resource—such as, in this case, a piece of land—determine what the opportunity costs are, to others, of her appropriating a resource. In the case at hand, because Adrian would produce something which others want, whereas Bruce would not, the opportunity costs to others of Bruce’s appropriating the land are higher than those of Adrian’s appropriating it. Adrian’s ambition is, in this specific sense, “cheaper” (for others) than Bruce’s, and this should be reflected in the price Adrian should pay in pursuing it. This implies that Adrian should be able to reap some rewards for selling his tomatoes, or equivalently, that he must pay less for the land than Bruce should.

Equality of resources thus captures considerations of productive fairness: assuming that people differ only in their ambitions, it is unfair to producers to treat the benefits they create as being, morally speaking, fully on a par with manna from heaven which must or may be redistributed equally, without the fact that someone produced those benefits making any difference to how producers fare relative to non-producers: to do so is to allow someone to gain at the expense of producers in a way that is contrary to equality. Note that productive fairness, I believe, underpins the often-cited claim, in discussions of responsibility-sensitive egalitarianism, that equality of resources tells against forcing the Ant to subsidise the Grasshopper. If we redistributed resources from the Ant to the Grasshopper, we would be violating equality of resources by enabling the Grasshopper, who has an expensive non-productive ambition, to gain unfairly at the expense of the Ant’s inexpensive one.

The same considerations, I submit, come into play when we focus on the ambition to have and raise children, assuming that children are public goods. To illustrate this—and without, I hope, causing offence—we can imagine a second Adrian and Bruce example, in which Adrian would like to use scarce resources (land, say, and other natural resources), not in order to grow tomatoes, but to have and raise a child. As before, in order to know what price Adrian must pay for pursuing this ambition, we need to know what the opportunity costs are, to others, of scarce resources’ being put to this use rather than devoted to their purposes instead. If we grant that Adrian’s parenting ambition is productive—Adrian will spend time and energy and some of his material resources in raising a child who will, in due time, be an asset to society in various ways—then the price Adrian must pay for his ambition should reflect that fact. It would be unfair to Adrian to price his parental ambition as if it were expensive to others, when it in fact is, in the relevant sense (i.e., in terms of its true opportunity costs to others), a cheap one; Adrian’s ambition would wrongly be priced as expensive if Adrian had to pay the same price as Bruce for the land, and use only his own resources to bid for the resources he needs to raise a child, in the same way in which Bruce has to use his own resources (fairly, in Bruce’s case) to purchase land on which he intends to play tennis.

To be sure, the two Adrian and Bruce cases are different in some important respects. Where Adrian’s productive ambition involves commodifiable goods like tomatoes, it is permissible, and, under just background conditions, there are reasons of efficiency, to introduce actual markets and entrust the pricing of tomato-growing ambitions to them. If they work reasonably well, as mentioned earlier, markets will register the opportunity costs to others of resources (e.g., land, water, atmospheric capacity) being assigned to tomato production, as opposed to alternative uses.²⁰ Relatedly, with goods of this kind, it is both permissible and efficient that those who produce them be assigned private property rights over them; they will thus reap rewards or recoup costs, as productive fairness demands they do, by exercising those property rights.²¹ Considerations of productive fairness, in the cases of commodifiable goods and given the right background conditions, can thus be realised in actual markets and through the assignment to producers of property rights over the fruits of their labour.

But note that setting up actual markets and assigning producers private property rights are not necessarily required by productive fairness. Where non-commodifiable goods are concerned, and it is impossible and/or undesirable to assign people private property rights over what they produce, productive fairness can be realised in other ways. For notice that we could avoid the unfairness to parenting Adrian in two possible ways: we could either reproduce as closely as possible, compatibly with respecting the fact that children are not ownable and tradeable, the scheme we establish with tradeable and ownable goods; or we could share with Adrian the costs of production.²² Let me explain what each of these two options consists in, and why the overall ideal of equality of resources tells in favour of the second one over the first.

The first option is that of a regime in which both the benefits and the costs of children are privatised. While children are neither ownable nor tradeable, it is perfectly possible, compatibly with respecting fully their equal moral status, to set up socio-economic institutions so as to offset the costs of parental investments in children by allowing parents to reap more rather than less of the material benefits which they contribute to producing by having and raising children. To illustrate: one of the major benefits which children will provide—their future tax contributions—could be divided so as to mostly accrue to parents instead of being divided equally among all citizens regardless of whether or not they are parents. Some social scientists have advanced proposals of this type, mostly focusing, specifically, on structuring the pension system in such a way that parents benefit more from the new generations’ contributions than non-parents (Sinn 2005; Gál, Vanhuysse and Vargha 2017; see also Burggraf, 1997). We could imagine an entire split welfare system, rather than just a split pension system, in which one’s welfare entitlements vary in line with whether or not one is a parent (for discussion of this possibility, see Olsaretti, 2013).

It is important to see that, as far as considerations of productive fairness go, either of these two schemes—which we can call privatisation and socialisation, respectively—is in line with them.²³ However, whether the overall ideal of equality of resources is indifferent between them is a further question; productive fairness is but one part of that ideal. Although I cannot develop this point at length here, let me conclude this section by saying that I think equality of resources—and indeed any liberal egalitarian view—must, by dint of its commitment to equality as a social ideal, or equivalently, to viewing society as a fair scheme of cooperation in which citizens share one another’s fate, favour socialisation over privatisation. To see why, we must note that at the extreme, in which all the costs of children, and all the benefits of children, are privatised, the society regulated by privatisation is one in which there are no primary or non-remedial obligations of egalitarian justice among citizens at large. That is, the obligations to pay for each citizen’s fair share falls, ideally, on their parents; fellow citizens are only obliged to step in as a matter of non-ideal justice, assuming parents have failed in their obligations (Olsaretti, 2017). On a view which favours privatisation, parents and children are bound to each other, but not to fellow citizens generally, by non-remedial obligations of egalitarian justice. By contrast, under socialisation, citizens are tied to one another qua fellow citizens, and not qua parties to parent–child relationships. So, while both a privatised scheme and a socialised scheme can realise the demands of productive fairness, other parts of the ideal of equality of resources arguably tell in favour of the latter.²⁴

4 Incentives, Economic Rents and the Self-Interest Bias in Pricing Ambitions

The previous section argued that Market prices for ambitions is ungrounded and that productive ambitions are not expensive ambitions, so it would be unfair to producers to ask them to pay costs for them that do not in fact reflect the true opportunity costs to others of those ambitions. If parents were charged market prices for their ambitions, they would, in effect, be overcharged. Egalitarian productive fairness thus provides a neutrality-respecting case for sharing the costs of children, where children are public goods, that surmounts the first stumbling block. But another stumbling block, No economic rents, is still in the way, and it takes the form of the following two-pronged objection:

Granted, the opportunity costs of ambitions is part of what determines the fair price of ambitions, but that price should reflect, besides the opportunity costs to others of forgoing a scarce resource, also the value to the producer of that resource’s going to him. If Adrian values tomato growing quite independently of whether he will be able to make money by selling them, that should make a difference to what he has to pay for the land: by acquiring that land, he is now acquiring something which he judges to be a very valuable resource.

Accordingly—so the second prong of the objection continues—productive fairness requires adjusting the price of resources for productive ambitions only when doing so is necessary to give people incentives to exercise those ambitions. If, by contrast, individuals do not need incentives in order to exercise their productive ambitions, equality of resources does not require that the fact that their ambitions are productive make any difference to their claims. To suggest otherwise would be to maintain, implausibly, that egalitarian justice tells in favour of paying people economic rents.²⁵ As illustration of this point, we can imagine that Adrian did not need the incentive payment to grow tomatoes, and were willing to pay enough for the land, out of his own resources, to outbid Bruce and grow tomatoes without being paid for them. (Adrian might even leave an area his cultivated land open to the public as a “pick-your-own-tomatoes-for-free” area.) Surely, the objection maintains, there would be no unfairness to Adrian in this arrangement, and to insist that the price Adrian has to pay for the land in this scenario has to be adjusted to be different from what Bruce would have to pay seems unmotivated and inimical to treating Adrian and Bruce equally.

If this objection were successful and egalitarian productive fairness considerations only kicked in where people need incentives to exercise their productive ambitions, the case for sharing the costs of children would be undermined whenever it is true that parents do not need incentives to have and raise children. This would restrict the scope of application of the public goods argument; it is an open question just how widely the argument would still apply, to be settled by reference to social scientific research on the effects of family policies on the fertility rate.

Independently of whether this would spell the failure of the public goods argument, or whether the latter may still apply under many real-world conditions, however, I think there are reasons to resist No economic rents. In particular, I believe that resisting No economic rents is required in order to offset a problematic bias which we would otherwise display towards narrowly self-interested ambitions, i.e. those lifeplans and preferences that have, as their final aims, just the narrowly defined prudential interests of the agent whose ambitions they are.

To make my point, it will help to introduce a third and last, more variegated, version of the Adrian and Bruce example, where various Adrians and Bruces are described as being different from each other in terms of both whether their preference is productive and in terms of whether they are narrowly self-interested or other-regarding. For simplicity, I will assume that an activity is productive if and only if it benefits everyone in the community; and (as we have done all along) that Adrian and Bruce differ only their ambitions. I also assume that all three Adrians’ productive ambitions, when exercised, are equally productive, and, finally, that the other-regarding ambitions are ones to benefit some others, in ways which do not involve picking up the slack of the community at large,²⁶ and are not perceived by those who have them as moral imperatives which involve a sacrifice in self-interest. In one case, i.e. with Narrowly Self-Interested Productive Ambitions, I also distinguish between two versions of Adrian, one of which (Businessman) needs incentives to cultivate tomatoes, and the other one of which (Hobbyst) does not.

	Narrowly self-interested	Other-regarding
Productive	Businessman Adrian (needs incentives) Hobbyist Adrian	Generous Adrian
Non-productive	(Original) Bruce	Generous Bruce

Consider, then, the situation of Generous Adrian, who cultivates tomatoes so as to benefit a local charity and in the process, unavoidably, produces extra tomatoes which the community takes for free. (The community would pay for those tomatoes if they needed to in order to obtain them, but let us assume that it would be costly for Adrian to fence off the extra tomatoes and that, even if he knows that a tomato levy will be charged, he will continue to produce tomatoes because he is committed to pursuing his ambition.)

On the incentive-unconstrained version of egalitarian productive fairness I favour (henceforth, “my view”), Generous Adrian will be treated differently from original Bruce: the community, I said earlier, would charge Generous Adrian, but not Bruce, an adjusted, lower, price for the land (or pay him for the tomatoes they obtain), on the grounds that the net opportunity costs, to others, of the land’s going to Generous Adrian are positive, whereas they are negative in Bruce’s case. Egalitarian productive fairness in my version would, furthermore, treat Generous Adrian equally with Businessman Adrian, on the grounds that the opportunity costs, to the community, of either’s appropriating the land are the same.

If egalitarian productive fairness were incentive-constrained, by contrast, Generous Adrian would be treated on a par with Bruce, and would be worse off, in resource terms, than Businessman Adrian. But why should fairness (as opposed to efficiency) allow for Businessman Adrian to be better off than Generous Adrian? And how can it be compatible with fairness for Generous Adrian to be treated as if his ambition were as costly, to the community, as Bruce’s is, when these two ambitions are, in fact, not equally costly? We could say, to be sure, that Generous Adrian will gain in welfare by realising his other-regarding ambition even without a price-adjustment for the land, as is evinced by the fact that he is motivated to purchase the land and cultivate it without incentives, whereas Businessman Adrian might only get a similar welfare gain if he gets paid for his tomato growing and can then use the extra money he earns to buy leisure. But even if this were the case, I have been assuming all along that our egalitarian theory is not welfarist, so—again, insofar as fairness is concerned—the fact that Generous Adrian would have more welfare than Businessman Adrian when they are treated equally in terms of resources is no reason to prefer incentive-sensitive productive fairness.

Moreover, in light of the above discussion, we could say the following. Certain ways of setting up our socio-economic institutions (i.e. in line with the demands of incentives-constrained fairness) are biased in favour of those with narrowly self-interested ambitions, that is, ambitions which involve the satisfaction of desires about oneself. The bias is unfair in that it allows those with narrowly self-interested ambitions to obtain benefits (as Businessman Adrian does) thanks to being self-interested while not allowing those whose ambitions are other-regarding (Generous Adrian) to also obtain benefits, despite the fact that the opportunity costs to others of their ambitions are the same. This bias has no justification, and institutions that exhibit it are unfair to those whose ambitions are other-regarding.

But I am not home and dry yet. At this point it might be said that once we consider Hobbyist Adrian and Generous Bruce, my contention that an incentive-constrained view is biased in favour of narrowly self-interested ambitions and against other-regarding ones, in a way in which my view is not, loses plausibility. This is because on my view, some other-regarding ambitions would also register as expensive: this is the case of Generous Bruce, who wants to buy the land to have tennis courts anyone could use, but which no one will use as no one likes playing tennis in his community. At the same time, for its part, the incentive-constrained view would register some narrowly self-interested ambition as expensive, namely, that of Hobbyist Adrian, who wants to buy the land to grow tomatoes for fun—his daily gardening is part of his mindfulness routine—and who does not fend off the land and lets the community take tomatoes from it. In light of this, the objection goes, I cannot claim that the incentive-constrained view reflects a bias in favour of narrowly self-interested ambitions, nor that my view accommodates other-regarding ambitions better.

In reply, I would like to say two things, beginning with a reply to the second part of the objection.

First, note that my point is not that having other-regarding ambitions is itself a reason for receiving favourable treatment; instead, my point is that the true opportunity-costs-to-others metric offers reasons for treating favourably some such ambitions, while the incentive-constrained view, by contrast, treats them less favourably for no good reason. Since neither being narrowly self-interested nor having other-regarding ambitions is a morally relevant reason for having to bear more or less burdens than others, we should screen out, insofar as we can, the effects of differences in whether one is narrowly self-interested or other-regarding, and in terms of whether one holds out for incentives or not, and price ambitions only in light of the true opportunity costs to others of scarce resources’ being allocated to those ambitions.²⁷

Second, even though it is true in principle that the incentive-constrained view would treat equally (i.e. equally non-favourably) the self-interested preferences of Hobbyist Adrian, which are as productive as the other-regarding preferences of Generous Adrian, this should be small consolation for the Generous Adrians of this world. I say this because I speculate that, in practice, Hobbyst Adrians who end up paying for their ambitions will be few and far between, whereas there will be many Generous Adrians who will. The reason for my saying this follows from the two-fold thought that, first, those with other-regarding ambitions are more likely than those with narrowly self-interested ambitions to be over-represented among producers of non-commodifiable goods; and second, that—as I mentioned above—we have weighty reasons to let the pricing of commodifiable goods to actual markets. Taken together, these two facts would mean that in practice, Hobbyist Adrians will be paid for their ambitions, despite the fact that they would be willing to exercise those ambitions even for little or no pay (think of all the people who get heftily rewarded for doing paid jobs they love doing); while at the same time, appealing to an incentive-constrained egalitarian ideal, we would allow the Generous Adrians to pay the costs of their productive ambitions (No need to mimic a market or share the costs of Generous Adrians’ activities! No need to incentivise Generous Adrians!). If there were truth to this speculative thought, this would strengthen my claim that adopting an incentive-constrained view of productive fairness would end up unfairly favouring self-interested ambitions.

In conclusion, let me return now to the case for sharing the costs of children when children are public goods: if the scope of productive fairness considerations is not constrained by considerations of incentives, then the case for publicly funded family policies that share the costs of children holds even when it is true, of parents, that they would have children even in the absence of those policies. It also holds irrespective of whether we think of the parental ambition as being akin to the ambition of Generous Adrian or as more like that of Hobbyist Adrian. This way, the application of the public goods argument will be both more robust and more defensible, in that it is not contingent on resolving questions whose answer can depend on controversial ethical views, i.e. whether parenting is an act of altruism or a fundamentally self-regarding project is what at issue here. This would be a suitably neutral, egalitarian vindication of why fairness requires sharing some of the costs of a society’s reproductive work.