a. Independent Realms

One possible way to look at the different allocations is to see each area as separate and independent. According to such a view, each realm is governed by independent principles. Custody should be decided only according to the best interests of the child (or the welfare principle). The allocation of the burden of child support should depend on the two parents’ relative ability to bear this burden in the future, namely on their respective income. The distribution of marital property, in contrast, should depend on the relationship between the spouses when they accumulated the property, or their relative need thereof. Accordingly, since each part of the allocation is subject to a different rationale, they should have no effect on each other.

When we focus on the burden of child support, what seems to be the most plausible solution is something in the spirit of the Australian scheme of child support,Footnote ³ which takes into account the income of both parents and only then allocates child support based on the relative time spent caring for the children by each parent.Footnote ⁴ For the fact of the matter is we should think of any child support scheme as answering two distinct questions in two different stages: the allocation of the burden (who is responsible for which portion), and the form of payment. Accordingly, in the first stage, which dictates each parent’s share in the duty of child support, the Australian scheme takes both parents to be responsible for supporting their children, according to their relative income (with no regard to residency). Only in the second stage, which dictates the form of payment, does residency enter the calculus and come into play, through the distinction between supporting the children directly (by the residential parent) and supporting them through a transfer (by the non-residential parent) to the other parent with whom they reside.

Thus, two parents with the same income should share equally the burden of support, notwithstanding the question of residency. Accordingly, after the scope of the required expenses is established (either independently or based on both parents’ income), the non-residential parent will be required to bear half of the expenses, while the residential parent is presumed to bear an equal sum in direct payments for the household’s ongoing expenses.Footnote ⁵ Where the non-residential parent has plentiful access to the child, it affects the calculus only by way of possibly deducting the direct payments she makes from the amount she has to transfer to the other parent. While the exact calculus might be complicated,Footnote ⁶ the basic principle is that the scope of care or parental time does not affect the first question—the scope of the burden; it affects only the direction of the payments, or the mechanism for making them. Similarly, it should not affect the principles of the marital property allocation. The only possible effect might be through the way in which the marital home might be attached to the children, hence to the residential parent, or the way in which the parents’ relative needs are evaluated (for the sake of alimony or such).

b. Traditional ‘Division of Labor’ View

Yet there is another way to perceive the relation between parental care and the other distributions. I call this the “traditional scheme.” In times past, the traditional household undertook a division of labor in which the two duties of rearing the child, in terms of care and finance respectively, were divided between the parents: the mother’s duty to care for the children was in exchange for the father’s responsibility to provide for the family. When the possibility of breadwinning was opened to women, the father’s duty of support became also a kind of a payment for the mother’s care, or compensation for the opportunity costs the homemaker bears in the absence of paid work. This reflects an acknowledgment that caring for the day-to-day needs of the children is a substantial effort and contribution. Thus, while fairness requires equal sharing in the parental responsibility, the contribution might be via either direct care or a financial payment, rather than through splitting the financial burden.Footnote ⁷

What seems to follow from such a view is a scheme of allocation that imposes the main burden of financial support on the parent who refrains from taking the burden of day-to-day care,Footnote ⁸ because requiring the residential parent to both give care and share the duty of financial support would violate the dictates of fairness. Thus, the basis for calculating the burden of support is the non-residential parent’s income only, rather than the ratio between the parents’ incomes. The income of the residential parent is not considered, since this parent has a lesser duty to participate in the financial support. Surely, the residential parent is expected to share her own income with the child, and had she had an equal income that would lead to equal sharing of the burden of support. Yet her income does not affect the non-residential parent’s burden: the support payments are presumed to suffice for caring for the child even if the residential parent has no income, and are considered to be fair even if this residential parent has a very high income. Likewise, while the obligor’s payments are compulsory, the way the residential parent divides the income between herself and the children is subject to her own discretion. Her main share is fulfilled through providing care, and so—unlike the former model—her own income has no bearing on the extent of support duties of the parent who do not provide care.

Such a scheme of allocation is reflected, even if not purposely or unequivocally, in the Canadian child support law, which regularly takes into account only the income of the non-residential parent for the sake of the “table amount”, considering both parents’ income only where they both share the care duties through shared custody.Footnote ⁹ The same idea is reflected also in the legal regime of those U.S. states that apply the “Percentage of Income” model.Footnote ¹⁰ If the residential parent’s income is taken into account, this is primarily due to the need to evaluate the scope of the child’s needs, yet once this scope is determined it is only the non-residential parent who bears the responsibility to provide the support, while the residential parent is not required to do so.Footnote ¹¹ Without committing to an exact interpretation of any concrete legal regime, which might be disputable (due to the complexities of child support policy), for the sake of my discussion here it is sufficient to recognize the possibility of this line of reasoning, which imposes the main responsibility of financial support on the parent who does not contribute through providing care.

A similar line of reasoning might apply also to spousal support or to the allocation of marital property. Many theories of marital property hinge on the value of housework, which consists, in modern times, mostly of childrearing-related labor. According to these views, while one spouse acquires property through the marketplace during the marriage, the housekeeping spouse should be recognized as having a share in this property. This is due to the way the housework contributed to and enabled the other’s breadwinning, and due to the career-loss it inflicted on the housekeeper.Footnote ¹² An advanced version of this claim holds that when upon divorce the couple still have young children, the residential parent continues to bear those responsibilities, hence should continue to share in the income even beyond the separation.Footnote ¹³ Similarly, a duty of alimony or spousal support might be due, as long as the breadwinning-housekeeping divide survives the separation, since it stems mainly from the division of parental labor rather than the couple’s romantic bond.Footnote ¹⁴

a. Custody as a Burden or a Benefit

The last view, which I have termed the “traditional scheme,” is rooted in a claim regarding fairness, according to which both parents should share the burdens of parental responsibility via either direct care or financial maintenance. Note, however, that such a model assumes residency with one’s child to be an overall burden, part of the general burden of parenthood, deserving of relief and reimbursement for enduring. However, seeing residency as an overall burden does not seem to fit the modern realities. At least since custody began to be an issue for contestation and competition between the parents, giving care might be considered, overall, as a benefit rather than a burden.

This may be due to a change in the social construction of parenthood. Traditionally, children were conceived as potential adults—immature specimens of the “human” organism. Parents had to provide a supportive environment and conditions that would enable the child to develop into its final and complete form: an adult. The child was therefore apprehended mainly as a deficit, configured by the list of missing adult capacities. And once children could no longer provide an economic benefit (due to the prohibition of child labor), having them in one’s home became mainly a burden. Today, however, childhood is perceived much more positively, being associated with innocence, joyful play, freedom and a special creativity.Footnote ¹⁵ Bringing up children is therefore not just the burden of fostering future adults. Children are valuable, so caring for children has become a desirable thing. This change, together with economic pressures relating to maternal employment, led also to a new model of proper fatherhood, the “distant breadwinner” model being rejected in favor of a more “involved co-parent” model.Footnote ¹⁶ Accordingly, both parents sometimes wish to win custody, and courts have begun to grant fathers broader custodial rights, turning custody into a valuable entitlement that both parents should have equal opportunity to claim.Footnote ¹⁷

According to this view, acknowledging the benefits of parenthood precludes the view of caregiving as only a burden in need of relief. While parenthood admittedly involves both burdens and benefits, the positive aspects might outweigh the burdensome aspects. After all, other areas of life are burdensome as well, yet that doesn’t override their positive aspects. Indeed, it was in the realm of family law that the benefits of work in the market were acknowledged, despite its burdensome aspects as compared to leisure, since it affords the worker both resources and, sometimes, a meaningful life.Footnote ¹⁸ In the same spirit, the fact that parenthood and its attendant caregiving have burdensome aspects cannot lead to its being seen as an overall burden, at least if both parents genuinely desire to provide caregiving, since the affective benefits of childrearing might counterbalance the cost of care work. Under such conditions, a demand both to be granted custody and to be exempted from the burdens of caregiving (or to be compensated for them) seems at first glance not only unjust, but maybe even incoherent.

b. Why not Independence

If viewing childrearing only as a burden will not do, one might be tempted to prefer the allocation scheme that I have called “independent.” As we have seen, however, while it is not easy to determine whether custody should be regarded as a burden or a benefit, everyone seems to agree that it is not neutral. Why, then, should we adopt the independent scheme? Here, three possibilities come to mind. One possible claim is that the burdens and benefits of caregiving offset each other, rendering their effect negligible as regards the general fairness of the allocation. A second possibility is to focus on the difficulty of measuring and quantifying the impact of caregiving, which makes it impossible to take it into account. The third possibility is that the children’s interests and welfare, as well as their status as autonomous persons, preclude the possibility of taking them into account as part of the allocation, as if they were a mere commodity. Let us consider these claims, one by one.

Various scholars believe that the burdens and benefits of caregiving offset each other. Such a view is reflected, for instance, in the comments of the American Law Institute’s Principles of the Law of Family Dissolution.Footnote ¹⁹ Similarly, Merle Weiner writes that “it makes a lot of sense to assume that caregivers experience an equal amount of nonmonetary costs and benefits and that these nonmonetary components balance each other out. This reasonable assumption means that noneconomic considerations can be ignored in an assessment of fairness.”Footnote ²⁰ However, while such a presumption might be practically plausible, given the difficulties in assessing the magnitude of the burdens and benefits, such precise offsetting is in fact very unlikely, given the plurality in preferences and life scenarios. It seems, then, that the claim that the benefits and burdens of caregiving offset each other should not be accepted at face value, but rather seen as actually reflecting the second claim—the need for a pragmatic strategy to cope with the difficulties of measurement.

Indeed, measuring—and more importantly, quantifying—the exact value of caregiving is a complicated task, and it is hard to imagine how courts or official authorities could undertake such a mission.Footnote ²¹ However, we should not conflate evaluation difficulties with valuing something as zero, and consequentially focus only on those aspects of familial life that can easily be quantified, such as the financial assets and burdens. Indeed, as we have learned from feminist critiques of tort law, failing to adequately compensate for nonpecuniary loss is not fair.Footnote ²² Moreover, to distribute only the financial items would do injustice to the very nature of the familial endeavour, creating systematic and fundamental errors. After all, the familial endeavour has both financial and relational aspects, as it strives in many cases to create a financially functioning household that will enable the flourishing of the relational aspects (in both the spousal and parental dimensions). Creating an environment for stable parenthood and establishing parental relations are often the most important goals in the spouses’ decision to establish a family.Footnote ²³ These goals dictate the couple’s decision-making from questions of division of labor to the very question of coupling and sharing a household. Financial and relational trade-offs therefore substantially factor into the parties’ planning during their relationship. To completely ignore them at the time of dissolution would not do justice to the parties.Footnote ²⁴

Surely, acknowledging the pressing need to evaluate does not solve the practical problem of how to actually conduct this evaluation. Below I will devote a separate discussion to the possible ways of overcoming this difficulty. At this stage, however, evaluation does not have to be easy, suffice it to be possible. Recall further that I focus at this stage on a couple that are negotiating a divorce settlement agreement seeking a fair solution, with full mutual knowledge and good faith. They may therefore be able to overcome evaluation problems that would stand in the way of an external evaluator. Indeed, while it might be tempting to think of the benefits and burdens of caregiving as incommensurate, exactly such a trade-off actually seems to be fairly common among divorcing couples.Footnote ²⁵ So, at least such sincerely intentioned couples should not assume that the burdens and benefits of caregiving completely offset each other, when they can set a negotiated price to them.

But even if money can buy custody, maybe it should not? In other words, maybe two sincerely intentioned partners should not see custody as a benefit in their allocation scheme, since it violates their duty towards their child. Doesn’t it amount to treating the child as a commodity, or violating their duty to promote the child’s welfare? Below I will elaborate on how to reconcile the child’s interests and dignity with the idea of fair allocation between the parents. For now, suffice it to recognize that even if such a trade-off does violate duties toward the child, it does not entail the irrelevance of fairness considerations. Fairness does not cease to be relevant in the presence of other normative defects. After all, even those who oppose transactions such as paid surrogacy, for instance, are able to distinguish between fair and unfair surrogacy agreements, and might concede that adequately paid surrogates are treated more fairly than poorly paid ones.

Thus, while the prevailing schemes for considering caregiving often regard it either as a burden or inconsequential, the opposite scenario, namely that custody is a benefit allocated to the custodian, is at least possible. Presumably, instead of imposing the burden of support on the non-custodian, such a scheme might impose it (or the greater part of it) on the residential parent, or even—in an extreme version—allow the allocation of residency to affect the distribution of marital property in favour of the non-residential parent. After all, as long as the familial “partnership” is aimed not only at gaining financial profits but also at promoting other interests like childrearing,Footnote ²⁶ the dissolution of the partnership calls for allocating all the familial assets, and taking all of them into account when considering the fairness of the allocation.Footnote ²⁷ One might therefore be tempted to conclude that the party that was deprived of a “share” in the children’s life should get a bigger share in the other assets, as a counterbalance. Yet this would be too hasty a conclusion. The next section is devoted to a more careful examination of this possibility, its limits and ramifications.

a. Simplifying I: Abstracting away from Gender

In deciding matters of family law in general, matters of caregiving in particular, gender justice claims loom large. Admittedly, the way society constructs familial roles affects the wishes and expectations of all the involved parties. Men and women might differ in the way they undertake caregiving responsibilities, as well as in the surrounding circumstances and alternatives they encounter in the workplace.Footnote ²⁸ While the role of women as primary caregivers has a well-documented effect on the subordination and deprivation of women,Footnote ²⁹ contesting custody by men might have a similar effect, since it might prompt mothers for whom child custody is important to concede other financial entitlements in order to secure custody.Footnote ³⁰ One cannot discuss fairness in this regard without taking such considerations into account, and indeed an extensive literature has been devoted to this task, in discussing the proper design of custody rules.Footnote ³¹

However, while being salient, considerations of gender justice do not exhaust the problem at hand. Since they might mask other components of the normative picture, it is helpful to first conduct the discussion before such considerations come into play. Therefore, in order to abstract away from gender justice challenges, I will take as my primary example the marital dissolution of two married women, who are similarly affected by social conventions regarding parenthood and motherhood and have to divide their assets and parental responsibilities towards their joint children.Footnote ³² The complex task of addressing gender-related considerations in this already complicated normative picture, and applying my discussion to hetero-normative families, will have to wait for another occasion, though I will provide preliminary thoughts towards the end of this paper.

b. Beyond Fairness: The Children as Items for Distribution

A major challenge to this discussion is that it seems to amount to treating children as property. Surely, the determination of custody should be responsive mainly to the best interests of the child.Footnote ³³ Sometimes those interests are even regarded as the sole criterion for custodial decision-making,^{Footnote 34} though many jurisdictions take a slightly more nuanced approach, seeing the best interests of the child as a “primary consideration” in all actions concerning children,^{Footnote 35} yet leaving a secondary place for considerations such as fairness between the parents.Footnote ³⁶ Similarly, the allocation of child support obviously should be guided by the goal of securing the child’s needs,Footnote ³⁷ while considerations of fairness between the parents play only a secondary role.Footnote ³⁸

However, this is not to say that we can ignore parental interests and considerations of fairness between them. First, parental interests often align with those of the child, since often it would be best for the child to be raised by the parent who wishes or deserves to be custodian.Footnote ³⁹ Second, there are cases in which there is no real difference between the relative fitness of the parents, as either caregivers or supporters. Even if there is such a difference, sometimes there is no reliable way to identify or evaluate it (hence the two options are epistemically equal). In such cases, other considerations will apply even under a strict priority rule.^{Footnote 40} Finally, accepting the priority of the child’s interests does not necessarily mean that this priority is of a ‘lexical order’.^{Footnote 41} In other words, the priority of the child’s interests does not require that those interests be fully satisfied or maximized before any other interest or moral consideration is taken into account.^{Footnote 42} Therefore, fairness might come into play even if there is a slight difference between the parents in terms of the child’s welfare.

Nevertheless, it is hard to deny that the interplay between the child’s interests and considerations of parental fairness is complicated. A useful way to proceed will therefore be to abstract away from the child’s interest. I thus stipulate at the first stage a scenario in which both mothers are equally fit to take care of the child. Moreover, each of them is wealthy enough to support the child by herself if necessary, namely the child’s financial needs—however defined—could be met by either of them alone.Footnote ⁴³ In other words, for the sake of the discussion I assume that any kind of allocation of responsibilities between the two parents will secure the child’s financial and emotional interests, making the child indifferent to the actual allocation. The remaining question is, therefore, a question of distribution only.

c. A Few More Simplifying Stipulations

So far, I have made room for the possibility of thinking about the allocation of custody in terms of distribution, thus allowing it to be taken into account within a general scheme of fair distribution of the child support burden and the marital property. Surely, however, the allocation of parental time cannot be easily translated into the allocation of either the burden or the benefit of parenthood: it is not as if any marginal unit of time represents the same magnitude of this burden or benefit. A parent might, for example, enjoy the benefits of parenthood with half of the parental time, without gaining twice as much benefit from being the sole custodian (since that would involve an extra burden, or due to the diminishing marginal value of such benefit). Similarly, moving from two weekly nights to three does not increase one’s benefit (or burden) by 50%.Footnote ⁴⁴ There are so many possible contingencies, giving rise to a complicated picture that stands in the way of appreciating the possible role of this allocation in terms of fairness. For that reason, let me stipulate a few more simplifying assumptions in portraying the basic case I will discuss.

First, I stipulate that the two (female) partners have fully conflicting interests: both have a symmetrical wish to take care of the child (fully and alone). I assume that each of them prefers to maximize her parental time, i.e., prefers to be the child’s fulltime caregiver to any other possible allocation. Second, I stipulate further that the above preference of the parents is based on their own self-interest, namely that notwithstanding any altruistic preference they might have, their self-interest suffices to establish their wish to gain full physical custody and residence. Third, in order to simplify the translation of parental time into parental interests, I stipulate also that any kind of joint residency or shared custody, or even substantial visitation, is not viable (due to, let us say, both spouses’ impending relocation, or such), so the benefit of a substantial relationship with the child should be allocated exclusively to one of them only. Finally, as mentioned above, for the first two rounds of the discussion, I shall assume that both parents wish to settle their dispute in good faith and in a fair manner through an adequately negotiated agreement, and that they are acting with full knowledge and mutual disclosure. Under all these simplifying assumptions, and supposing that the parental plan is already settled, let us now consider our main question: how the allocation of custody should affect the other allocations.

a. Basic Principles of Distributive Justice

Burdens and benefits should be distributed. Choosing the appropriate distributive frameworks and the resulting distributions is thus a matter of distributive justice. Principles of distributive justice should provide normative guidance not only as to how political or legal institutions are structured, but also the way private people act within their capacity in dividing assets or responsibilities among themselves. Following Susan Moller Okin, I assume here that both in terms of private morality and in terms of governmental interference, principles of distributional justice should apply also within the nuclear family, in the division of the marital assets and responsibilities upon dissolution of a marriage.Footnote ⁴⁵ Indeed, such a commitment usually accompanies the norms for distributing marital property.

This is not the place to develop or defend a full theory of distributional justice or fairness in allocation. I will therefore conduct the discussion according to the following seemingly plausible assumptions, which reflect what I believe to be common intuition and practice. Those who do not share these views might try to amend the argument accordingly, or just see it as a conditional claim regarding the ramifications of these common views. For the sake of the current discussion, I will assume without argument that in principle, fair allocation means that each of the parents is entitled to an equal share in the benefits and an equal share in the burdens, as long as considerations of responsibility or desert do not dictate otherwise.Footnote ⁴⁶ Moreover, in the distribution of the marital assets and parental responsibilities upon dissolution, I assume that the relevant considerations are only those relating to the familial unit and the responsibilities of its members, leaving the task of promoting general justice to the general political body (e.g., through the tax system). Accordingly, if an asset that is worth 100 stands for distribution, the fair way to divide it between the parties is to let each have 50; likewise, if a burden of 100 need be allocated, both parties should share it equally.Footnote ⁴⁷ Differences in the parties’ ability to bear the burden, or in their responsibility for creating an asset or a burden, might affect the allocation. Similarly, the allocation might be affected also if one of them deserves a bigger share than the other (according to independently specified criteria). These basic intuitions indeed reflect roughly the common schemes for allocation of both marital property and child support.

Generally, it is reasonable to limit questions of distribution to financial resources only, other assets being dealt with through either actual division or a buyout that enables the party who got the money to find a substitute in the market. As regards contact with one’s child, however, that will not do. Under the simplifying assumptions, contact with children is not divisible in itself, and there is no possible market in which its value could be translated into other goods, or a meaningful substitute purchased. This item should therefore be regarded as part of the distribution, a view which is actually already common within the scheme that takes caregiving into account as a burden. The only remaining question is that of aggregation and commensuration, an issue I will discuss below.

b. Interim Conclusion for the Simplified World

How should the above-discussed separating couple, who wish to dissolve their marriage in a fair manner, allocate the burden of child support and their marital property, given their inability (ex hypothesi) to share the custody? More importantly, how should the allocation of custody affect the other allocations? In order to explore this question, let us begin with the allocation of custody itself.

a. Dropping the Symmetrical Claims Assumption

So far, the discussion has assumed that each parent has a symmetrical claim to caring fulltime for the child. What would be the consequences of dropping this assumption? This section will consider the case where one mother wishes to be a fulltime caregiver, while the other wishes to have no access at all. Section (e) below will consider the case where Mother Two wishes to have some access, but not full time.

If one mother wishes to have custody and the other equally fit mother does not, it seems clear that the first mother should gain custody. But how should that affect the other allocations? In order to find out, recall first the moral verdict that applies when both parents have a symmetrical claim not to have the child, namely both see custody as a burden and wish the other to have it. Under such an assumption, it seems clear that when deciding who should be the custodian (maybe by tossing a coin, as discussed above), both parents have to share the burden of their parental responsibility. Therefore, the one who undertakes the burden of caregiving is entitled to be compensated for it by the non-residential parent assuming a bigger share of the financial responsibility. If the burden of caregiving is high enough, it can lead to a duty of the non-residential parent to bear all the costs of childrearing, or even to compensate the custodian further for her extra burden. In other words, if the burden of caregiving is higher than the burden of paying even the entire sum of child support, extra payment may be required (maybe in the form of a transfer of marital property, or post-divorce income sharing). Moreover, the duty of child support will be directed towards the other parent, as a duty of fairness, rather than only towards the child.Footnote ⁶¹

But what if one parent wishes to be the custodian and the other does not? Here there are several possibilities, which seem to depend on one’s general theory of distributive justice. Following principles of distribution that focus on equal welfare, the custodian—who ex hypothesi gains extra welfare through gaining custody—would still be required to compensate the other, notwithstanding the other’s preferences. Otherwise, one parent would find herself with a bigger share than the other (in terms of wellbeing or welfare), violating the principle of equal distribution. Thus, the party who wishes to be the custodian still has to compensate the other, even when the other saw custody as a burden.

In terms of equality of resources, however, if custody is not desired by both parents, it ceases to be a positive resource, and hence should be neglected in further distributions. As a parallel, think about dividing a cake and a carrot between two children, where one of them does not like vegetables. Welfare-based distribution would give the carrot to the one who likes it, and compensate the other by a bigger share of the cake (maybe the whole slice). Resource-based distribution (assuming inalienability of the carrot) would give the carrot to the one who likes it and divide the cake equally, since the other child makes no claim for the carrot. The intuitive thought is that a child who would not have the carrot for himself should not profit from the fact that someone else can, an intuition that might underlie the Dworkinian “envy test”.Footnote ⁶² Such a conclusion seems to be even clearer given the fact that the burden of rearing the child should be allocated between the ex-spouses, so the fact that one of them is willing to take the child exempts the other from this burden, enriching her in this regard. According to this view, then, the non-residential mother has no claim to compensation. It is unclear, however, why she should be required to share the burden of care with the residential mother who sees the care as overall a net-benefit. The allocation of the entire financial burden to this mother seems to be even less warranted.

The cake-carrot parallel might conceal another factor. Preferring the carrot may seem to us, in a way, as a better attitude (due to the health advantages of carrots over cakes, or the politeness and discipline it reflects given that children are commonly encouraged to have vegetables over cakes). But isn’t that also the case with a wish to be the custodian? In other words, couldn’t there be considerations of desert at play here, holding that one is entitled to a bigger share of the distributed goods due to her actions or even her character?Footnote ⁶³ If so, it seems clear that there is no call for compensating the non-residential parent, since the residential mother deserves a bigger share in the allocation. I will return to this issue when discussing desert-based claims below.Footnote ⁶⁴

One last reservation is necessary. The above discussion has assumed that by being allocated custody, the party who wished to gain it also gains extra welfare relative to the other. Yet it is important to note that the non-residential parent is not untouched by this allocation: she too might gain extra welfare through leisure, work or mere freedom, which would parallel the benefit to the custodian. Moreover, the good in parenthood is at least partly non-rivalrous: it can be consumed simultaneously by both separating spouses. This is because parenthood might be considered a benefit even if one doesn’t spend a substantial amount of time with the child. In the absence of the child’s spontaneous trust, affection, and intimacy, which are associated with significant shared time and residency, a parent might still enjoy the fact that she has a child that will carry her genetics, or derive joy from the possibility of establishing a future relationship with the child, or such. Where the non-custodian preserves a more robust relationship (such as regular phone calls or other sorts of ongoing relations), or where she maintains ongoing authority to direct the child’s life (through educational choices and such), this option is more likely. In these ways, the non-residential parent may gain a substantial share in the good of parenthood and find herself well off in terms of the internal distribution, by being granted both an exemption from the burden and a share of the benefit. In such cases, there is again a call for sharing the burdens of parenthood, especially given the fact that the benefit of the non-residential mother free rides on the ongoing nurturing and care by the custodial parent.Footnote ⁶⁵

To conclude, then, both normative and factual factors dictate whether or not abandoning the symmetrical claims assumption changes the normative picture. The normative question here concerns how an item should be counted for the sake of distribution, when only one party wishes to have it. The factual question relates to the direct parental benefits enjoyed by the non-residential parent, as well as the magnitude of the leisure-related benefits (including career options) that accompany the non-residency. The answer to the factual questions might vary from case to case.

b. Dropping the Similar Fitness Assumption

So far, the discussion has assumed that both parents are similarly fit to be the custodians, in terms of meeting the best interests of the child.Footnote ⁶⁶ This assumption should not be too strict: since the standard of the best interest of the child is infamously vague,Footnote ⁶⁷ when both parents are above a certain threshold of fitness it is hard to say which one, if any, is better than the other. Even if such a difference exists, it will often be epistemically impossible to determine in advance, due to ignorance regarding the future needs of the child and the future life circumstances of the parent. Beyond that, some aspects of good parenting seem to be incommensurate. Here again, determining which parent is more fit than the other might be difficult.

Sometimes, however, it is not so hard. There are cases where upon reflection it is clear (to both parents) that one parent is better than the other. This may have to do with personality or character (of either the parent or the child) in terms of patience or attentiveness, with the flexibility of one’s career and the extent to which it enables a reasonable work-life balance, and so on. For the sake of the current discussion, I will assume without argument that in any such case the better parent should be awarded custody. Still, it is necessary to inquire how that should affect the allocation of the other assets and burdens, namely property and the duty of child support.

c. Dropping the Self-Interest Assumption

So far, the discussion has assumed that the parent’s (symmetrical or not) claim to parental time with the child stems from self-interest. This is not to say that this parent does not care about the child’s welfare, but only that her own interest is for her a sufficient reason to prefer having custody over not having it. However, surely self-interest is not the only possible reason. A parent might wish to gain custody for the sake of the child, even though out of her own self-interest she would prefer to be a non-residential parent. Even in ordinary cases (not to mention cases of children who suffer from severe disabilities), a parent might prefer to be free of the burden of the kids, yet not trust the other’s capacity to take care of children properly, be it physically, emotionally, or even morally.^{Footnote 79} Accordingly, she might believe that taking care of the child by herself is her duty, even if she would rather be exempted from that duty, if only the other parent were better suited.Footnote ⁸⁰ In such cases, asking for full residency does not coincide with seeing custody as a benefit. How should such a scenario affect the distributional scheme?

Consider first a case in which both parents wish to gain custody out of altruistic commitment. After settling the allocation through whatever relevant mechanism for handling the disagreement between them (since apparently each of them believes the other to be the inferior custodian), they have to allocate the financial burden between them. Since one of them has taken on the burden of caregiving, it seems reasonable that the other will assume a bigger share of the financial burden, including—in proper cases—a component of compensatory payment in terms of marital property. For the reasons discussed above, the same seems to hold where one parent wishes to gain custody altruistically, while the other waives her claim to custody.

Consider now cases where one mother claims custody out of self-interest (let us call her SI), while the other claims it only due to altruism (let us call her A). Surely, if SI gains custody she has no claim to burden sharing. Actually, maybe she even has to compensate A, subject to the above discussion on asymmetrical claims. In contrast, if A got the child, she indeed has a genuine claim to burden sharing, since there is no reason to believe that gaining custody is a net-benefit for her. On the other hand, in terms of marital property, unlike in the basic scenario here there is a situation of mutual loss: the loss inflicted on SI, the non-custodian, due to her having lost custody, and the burden assumed by A, the custodian. These losses might offset each other and hence call for no further redistribution.

Similarly, that one parent refrains from contesting custody might not reflect a ‘custody as a burden’ view. On the contrary, a parent might understand that the other parent is a better custodian. Under such conditions, despite a genuine self-interest in having the child, a parent may want the other parent to be the custodian. This parent’s loss, in terms of losing contact with her child, is not inferior to the loss suffered by the parent who cares about the child to the same extent but also believes she is the better custodian and therefore claims custody. In other words, it is the loss in terms of self-interest, rather than in terms of the frustrated claim, that is of genuine importance. The result should resemble the verdict on the above-discussed dissimilar fitness case, this time even without the need to claim custody in the first place. Actually, relinquishing custody against one’s self-interest in order to promote the child’s welfare might also offset the way desert-based considerations work in favour of the custodian in asymmetrical fitness cases!

To conclude, what matters for the fairness in allocation is not whether or not the parties have claimed custody, but rather the extent to which their wish to be custodians is based on self-interest. Surely, this complication adds another challenge to the application of these considerations in the real world, in the absence of full knowledge regarding the parties’ real interests. However, for our all-knowing, well-intentioned couple, it is important to inquire what exactly stood behind the claim to custody (or refraining from making such a claim): self-interest or altruistic regard for the child’s welfare. Given the diverse options and nuances, it is hardly possible to come up with one general verdict on the fair allocation in such cases, but the mere possibility of altruistic motives does not preclude the claims of the non-residential parent.

d. Dropping the “Ability to Meet the Child’s Needs Alone” Assumption

So far, I have assumed that each mother could bear all the costs of rearing the child by herself. Such cases may occur if the parties are rich enough, or if we set the scope of duty of support low enough. If the custodian remarries, becoming part of a double-income household again, it might enable her to solely provide the level of support enjoyed by the child prior to the separation, even without being especially rich. However, in many cases one parent cannot provide for the child by herself. In other words, allocating the full burden of support to the parent who has gained custody would fail to meet the needs of the child.

In such cases, it seems reasonable to require the other parent to share the burden and pay, in order to secure the child’s needs. Note, however, that nothing has changed in the norms that apply horizontally between the parents themselves. If the custodian suddenly becomes rich, the non-custodian should be able to argue for shifting the burden back to the custodian’s shoulders. After all, although the non-custodian has duties towards the child, this provides no basis for the custodian to enrich herself. Sharing the burden is thus limited to the need to provide for the child.

Yet if this is true, then the duty of the non-custodian is only to help the custodian in meeting the child’s needs, not to share the burden equally. What seems to follow is that the custodian should bear the burden to the limit of her ability (taking her own needs into account, of course), letting the non-residential parent bear only the remaining residual part.Footnote ⁸¹ Thus, the basic normative picture is not sensitive to the parent’s level of wealth; only its practical application is such. Moreover, while the non-custodian will be required to contribute in order to meet the child’s needs, and forego her possible entitlement to compensation, that solution will not be fair in the relations between the parents. In terms of their internal distribution, the custodian will be unfairly better off, relative to the non-residential parent.

e. Dropping the “Winner Takes All” Assumption

So far, I have assumed that any kind of joint residency is unavailable, including any scheme that leaves the non-residential parent with substantial access rights or such. However, in many real-life cases, this is surely not the case. Even where the parents cease to live nearby each other, there are all sorts of access plans that enable each parent to secure substantial and significant time with her child, or to otherwise preserve a meaningful relationship. Given that possibility, a case of asymmetrical claims again raises the possibility of a positive sum game in which both parents benefit from a certain allocation of custody: one parent enjoys the custody as a benefit, while the other is satisfied with access rights, given the fact that someone else is raising the child and ensuring his health and education. It is not easy to translate the allocation of time into an allocation of the benefit of rearing the child, especially given the large amount of housework time, which falls mostly on the shoulders of the custodian.Footnote ⁸² A parent may well wish to reap the benefits without committing herself to tedious and grimy physical household labor such as “cleaning toilets, scrubbing stains off of floors and out of shirts, changing diapers and bedpans.”Footnote ⁸³ In such cases, while the benefit from the child is non-rivalrous, hence enjoyed by both parents, the burden falls mainly on one of them.Footnote ⁸⁴ So the demand for burden-sharing arises again. After all, if both parents enjoy the benefit, fairness and equality require that the burden be shared too. Therefore, it is reasonable to assume that the non-residential mother will be required to assume a bigger share of the financial burden due to her relatively smaller share in the burden of direct care, from which she benefits.

However, as long as both parents sincerely and symmetrically wish to gain full custody, such a preference reflects a view of full custody as superior to mere access or even a shared custody scheme. Thus, the mother who is actually granted custody still gets a bigger share relative to the other. The conclusions of the former sections still stand, then, though their effect might be mitigated by the smaller gap, in terms of benefit, between being the custodian and being the non-residential yet involved parent. Thus, the claim either to have the burden be shared or to be compensated for the other’s bigger benefit will depend on the details and contingencies of each scheme. Generally, however, we can still conclude that in the case of genuine symmetrical claims, the burden of financial support should be slanted against the residential mother, rather than the other way around.

To conclude, then, even if there is a possibility of access or visitation, the interim conclusion is mitigated but not abrogated. It is only the combination of discarding both the symmetrical claims assumption and the “winner takes all” assumption that wholly undermines the interim conclusion.

a. Dropping the ‘Good Faith and Full Knowledge’ Assumption: Why Claiming Custody is not Informative

So far, I have proceeded on the assumption that we know the parties’ genuine wishes and preferences. In that spirit, I have stipulated whether or not each mother preferred to have custody, and if she did, whether she was motivated by self-interest or altruism. This standpoint was also regarded as transparent to the other negotiating mother. However, in real-life scenarios the possibility of relying on the parent’s revealed claims is surely very limited. First of all, even a sincere claim to exclusive custody of the child might not reflect the value of the child but rather other interests that go beyond the distribution of parental interests. For example, one mother might prefer a certain custodial arrangement due to a wish to avoid the requirement of cooperation with her ex-partner in case of a conflict,^{Footnote 85} in an attempt to enable a “clean break”. Hence, a parent who wants some custody might demand full custody, although that does not reflect the nature of custody as a benefit in her eyes but only the wish to avoid contact with the other spouse. Hence, it is hard to translate the parties’ standpoints when contesting custody into the value they assign to contact with the child or the loss they will suffer from losing it.

The problem becomes more acute if, along the lines proposed above, one’s claim to custody is expected to yield financial gains (in terms of support or property). Under current background rules, real-world parents are expected to bluff in their claims, pretending strategically to have a strong interest in custody in order to reduce their duty of support or financial obligation.Footnote ⁸⁶ In other words, the parties’ representations are not credible or trustworthy, since they are difficult to verify and incur little cost or risk. Therefore, constructing the legal doctrine around the parents’ representations regarding their residency wishes is not feasible in a functioning legal system. The judicial system simply lacks a reliable fact-finding tool regarding the parties’ genuine wishes. Moreover, that there is such an incentive to bluff might even disrupt the process of determining the best residency plan, hurting the child while failing to secure parental fairness. It seems, then, that there is no way to construct a reasonable legal system that takes into account the loss of the non-residential parent for the sake of distribution.

However, while that may be a decisive reason against implementing such a legal regime, there is still room for considering the ramifications of the argument. If my conclusions in Part II are sound, then for many of the discussed distributions there is a substantial gap between what the fair distribution is and what the foreseen (justified) legal distribution would be. Parents who wish to be custodians (and so should pay less according to the principles of fairness depicted above) pay most of the price for the deficiencies in legal fact-finding to the benefit of parents who prefer to neglect their child (and so should pay more). This is unfair.

b. Why the Fairness-Based Rule Might Create Bad Incentives

Earlier (at section III(b)) I concluded that sometimes a compensatory payment between the parties is appropriate in the spousal dimension even in cases where the parents differ in their parental fitness. In such cases, the placement of the children will be determined according to their best interest, while a financial transfer between the parents will ensure fairness. However, in moving from the moral duties of a person seeking the good to the realm of legal regulation, it is necessary to consider the possible ramifications of such a regime in a non-ideal world, where people sometimes violate their moral duties. So a legal norm that requires the custodian to bear alone the costs of rearing her children (or even compensate the other) might lead the most fitting parent to waive her claim to custody, in order to avoid the expected financial consequences.Footnote ⁸⁷ This would be against the best interest of the child.

Once more, that seems to be a valid consideration for rejecting such a legal regime. However, it is important to note the distributional ramifications of this argument. Suppose that Rachel and Leah are the divorcing parents of a child; both wish to have custody, but Rachel is the better custodian in terms of the child’s interests. According to the dictates of fairness (as outlined above), Rachel should gain custody and assume the major burden of support (and sometimes even be required to compensate Leah). Such a legal regime is flawed, however, since it might incentivize Rachel to waive her claim to custody, as a result of which the child will reside with Leah, the less fit parent. Therefore, the legal regime should not require Rachel to bear the major burden of support, let alone pay compensation. This solution deprives Leah of her fair share in the distribution for fear of Rachel violating her own duties towards her child. In other words, it punishes Leah for Rachel’s (expected) sin. This is unfair!

c. Measurement and Quantification

Under the suggested scheme of Part II (the “interim conclusion”), the extent of the duty of child support and sometimes even the allocation of marital property or post-divorce income sharing should hinge on the relative worth of custody in terms of either a benefit or a burden. However, nonmonetary benefits and burdens are hard to measure and quantify. How much property should be allocated in return for the loss of custody? How exactly should we estimate, in monetary terms, the pleasure a parent derives from being with her child, or the pain of having to confront him or worrying for his safety?

Note that the problem here is not that such pleasure or pain and monetary payments are incommensurable. Trade-offs between monetary items and levels of parental responsibility are common among divorcing couples. While such trade-offs might raise a challenge in terms of the limits of permissible commodification (a question I have tackled elsewhereFootnote ⁸⁸), here I would like to address a more practical problem of implementation. Even a judge or lawmaker that has been totally convinced by the arguments in section II would find it hard to deny that courts are unable to implement such a legal regime in the absence of reliable external evidence. It seems, then, that the discussion has produced no more than a theoretical exercise, with no possible ramifications for law and policy.

Nonetheless, as stated earlier, difficulties in evaluating something should not entail that its value is zero. Zero is as arbitrary as any other estimation, and it is especially groundless where it is quite clear that unlike any other guess, ‘zero’ would probably be the wrong answer in almost all cases. Ignoring the value of custody is notably problematic due to the systematic nature of the evaluation problem within the familial arena: it will lead to injustice in any divorce case where custody is truly contested. On the personal level, therefore, it seems that given the evaluation difficulties we should simply do the best we can, in an attempt to mitigate the gap between the just solution and the reality, if not preclude it altogether. On the institutional level, while it might be advisable for courts not to consider such issues in the basic norms of allocation, the insight and sensitivity that follow from section II might inform the way the law handles private ordering, an issue to which I will return. Finally, even in the absence of any practical effect, it is important to note again that the cost of the institutional deficiency (i.e., the inability to evaluate properly the relevant factors) is borne mostly by parents who have a genuine wish to gain custody (that goes unfulfilled), while neglectful parents benefit from the fact that relieving their burden is not counted. Again, this is unfair.

d. Gender is Back

So far, the discussion has assumed two women negotiating their separation. In reality, however, divorce bargaining—especially when it involves children—usually takes place along gender lines, between fathers and mothers. Moreover, both the preferences and the options of the parties, in terms of their wish to be the primary caretaker as well as their earning capacity in the market, are shaped by structural conditions beyond the parties’ immediate control. On the personal-moral level, the parties should take into account their concrete characters and circumstances (for which gender serves only as a proxy), but the law should also take generalizations into account.Footnote ⁸⁹ The legal regime should consider, for example, the general differences in negotiation style between men and women,Footnote ⁹⁰ as well as the fact that women are subjected to social pressure to assert that their child is a benefit (pressure to which men are much less subject). Similarly, given the structure of gender inequality, the law should resist incentivizing bargaining tactics designed either to regain men’s patriarchal privileges lost through the divorce, or to attempt to avoid financial responsibilities.Footnote ⁹¹ More importantly, the role of historic injustice in the allocation of caregiving-related missions, as well as the effect of gender construction on the parties’ preferences, should be taken into account. While the discussion above hinted that a parent who sees custody as a benefit might have a duty to compensate the other, it seems especially objectionable to let men benefit from the fact that women tend to seek custody: having first shifted the burden of caregiving to women, and then constructed the identity of mothers accordingly, patriarchy is now seeking compensation? Talk about adding insult to injury!

Within the limits of the current framework I cannot address all the possible ramifications these claims may have for the proper regulation of familial distributions in the case of hetero-normative couples. However, without attempting to spell out the entire normative picture in this regard, note again that if the conclusions of section II must be withdrawn due to considerations of gender justice, the price would once more fall mainly on the wrong shoulders. After all, the deepest effect of the current legal position, in terms of deviation from section II’s “interim conclusion” (according to which the allocation of support duties and maybe even property should be slanted against the residential parent), would be on those who most closely fit the simplifying assumptions: fathers who genuinely wish to assume the role of caregivers (hence see custody as a benefit, yet are still required to pay as if it was only a burden), and mothers who earn enough to support the child by themselves (hence presumably are not the primary victims of patriarchy, within the family and beyond). “Traditional” fathers who see caregiving as a burden and serve as sole breadwinners are treated fairly by the current legal scheme, while fathers who share caregiving and breadwinning with their partners (and, as a result, might see custody as a benefit) and have high-income wives are treated less fairly, even though they are less liable for the wrongs of patriarchy. Such an allocation not only raises concerns of fairness, but also creates incentives that might hinder the adoption of more egalitarian lifestyles and a more just division of household labor between fathers and mothers, since it incentivizes fathers to avoid caregiving while incentivizing high-earning mothers to seek custody (by not requiring them to assume the burden of support). In other words, this structure of the law has a role in confining couples to a gendered division of childcare responsibilities and reinforcing traditional gender roles, since those who adopt a more egalitarian course of living are “punished” upon divorce by a legal result that allocates their burdens and benefits according to a more traditional scheme. This is unfair (and maybe also unwise).

a. The Law of Private Ordering

The last section provided compelling reasons against designing the law of family dissolution in a way that relates to caregiving as a benefit and compensates the non-residential parent for the loss of this benefit. The law should not adopt my “interim conclusion” as the default rule, then. However, many of those compelling reasons do not apply to couples who wish their privately negotiated agreement to be approved by a judge. Private parties do not need a judge to determine factual findings about their preferences, and are better equipped to evaluate the other’s true preferences, based on their own close relationship and experience.Footnote ⁹² They also have an advantage in evaluating the good faith of their ex-spouses, and hence are able to call their bluff in cases of strategic behavior. It seems, at first glance, that where the parties have negotiated a divorce settlement agreement that lets the residential parent bear most of the burden of child support or even requires the transfer of marital property to the non-residential parent [hereinafter: “Residential-Pays-Agreements”, or “RPA”], the law should be permissive towards such agreements.Footnote ⁹³

However, at second glance there are limits also to the permissiveness of the law of settlement agreements. After all, these are very sensitive agreements.Footnote ⁹⁴ The law should therefore seek to regulate these agreements in order to secure them against exploitation and unconscionability. An RPA that shifts all the burden of child support to the custodian, leaving the non-residential parent with more of the marital property, might indeed be the result of a fair division that takes caregiving into account, but it might also be the result of exploitative bargaining that unjustly impoverishes the weaker side. If the law cannot properly distinguish between these scenarios, then again enabling such agreements would not be advisable in a functioning legal system.

What should the law do, then, with RPAs? The conclusion seems to depend on one’s general position on procedural review of contracts. Those who believe that courts should limit themselves to procedural review only, and in their ability to conduct such review reliably and efficiently (based on the parties’ declarations and access to legal counsel, for instance), will be inclined to approve such agreements. As long as courts can tell which RPA represents a well-considered fair distribution and which is built on one-sided oppression, or distinguish voluntary agreements from those entered into under duress, they should be able to endorse the former and reject the latter. The consequence will be a more flexible standard that applies to settlement agreements (including those which allocate child support duties), focusing on ensuring sufficiency of resources for children, with more room for the parents to decide the origin of those resources and the division of the assets and burdens between them.

On the other hand, those who believe that courts should review the terms of contracts, either generally or in the specific context of divorce settlement agreements, are likely to conclude that if courts are unable to conduct such review the law should limit parties in forming RPAs, even if some of them might actually be fair. The same is true also for those who see the procedural review as hinging on the contract’s content (i.e., evaluation of the consideration), inasmuch as a review cannot be conducted in the absence of an objective standard to evaluate the fairness of the bargain. Thus, the same limits that apply to the law in general apply also to the norms of private ordering: the law should bar parties from forming RPAs since it is required to bar unfair agreements and it cannot tell fair RPAs from exploitative RPAs.

However, despite this conclusion the current discussion is not entirely immaterial. Understanding the reasons behind the regulation of such agreements helps us to understand and appreciate the price we pay for our commitments. This price might be worth paying. Yet acknowledging the fact that it is paid, and identifying those who pay it, helps in fully recognizing the effect of the law on familial life.

b. The Morality of Private Ordering

So much for the legal ramifications. Yet there is also another important conclusion that bears on the way parties should look at the law from a moral point of view. We can now conclude that couples who wish to dissolve their marriage in a fair manner should not take the law as a reference point for their private agreements. Indeed, parties commonly bargain in the shadow of the law, and often this is also due to the perception that law determines the norms of fairness.Footnote ⁹⁵ From a behavioral point of view also, the legal resolution will often anchor the parties’ perceptions regarding the point of departure for their bargain, as well as shape their notions of entitlement. While privately negotiated agreements often raise a concern regarding their fairness,Footnote ⁹⁶ the current discussion has demonstrated that for some couples and under specified circumstances, the legal solution might be totally detached from fairness. Therefore, such couples should not treat their legal rights as determinative of a fair agreement.Footnote ⁹⁷

Putting aside the question of the ability to form such legally binding agreements, it is clear that those separating parents who are wise enough to solve their dispute without resorting to the law, should avoid the law not only as a mechanism of enforcement but also as an inspiration for their desired settlement. This lesson should be noted by well-intentioned couples as well as by mediators that help parties to become such couples. Insofar as the parties wish to be fair and decent, they should not look to the law as a guide to the proper solution, as neither the starting line nor the finish line. While the law should take into account generalizability, enforceability, ignorance, bad faith and incompliance with one’s duties, the well-intentioned couple are free to consider the dictates of fairness, as long as the child’s interests are not compromised.