Reforming Policy for Single-Parent Families to Reduce Child Poverty =================================================================== * Maria Cancian * Daniel R. Meyer ## Abstract We argue that child support, the central program specifically targeting single-parent families, should increase financial resources for children living with a single parent, with a secondary goal of holding parents responsible for supporting their children. Current child support policy is substantially successful for divorcing families in which the noncustodial parent has at least moderate formal earnings. However, the system does not work well for lower-income families, especially unmarried couples: far too few children regularly receive substantial support and the system is sometimes counterproductive to encouraging parental responsibility. We propose: a public guarantee of a minimum amount of support per child, assurances that no noncustodial parent will be charged beyond their current means, and a broadening of child support services. * child support * divorce * guaranteed income * nonmarital births * single-parent families Recognition is widespread that single-parent families with children are economically vulnerable but less so on the policies and programs to address these vulnerabilities (see, for example, Maldonado and Nieuwenhuis 2015). Policies addressing custodial parent families (those who have children who are living with only one of their parents) confront the fundamental challenge of balancing the role of public benefits and private support from the noncustodial parent. Efforts to hold noncustodial parents responsible for their children encounter issues related to the relative importance of encouraging financial support and encouraging noncustodial parents’ active engagement in their children’s lives. The response to these challenges has varied over time, and in some cases, for divorced and never-married families, and for families who do or do not receive means-tested public benefits. Current policy includes general programs for low-income families and specific ones for custodial parent families. It prioritizes private support over public, economic support over other engagement, and generally makes no distinction in the financial responsibilities of noncustodial parents who have had different types of relationships with the other parent. Policy goals for custodial parent families include encouraging private support from both low-income and other noncustodial parents by trying to set an appropriate amount of economic support to be transferred, monitoring whether it is transferred, and then enforcing the transfer through a variety of threats and penalties. Public and private support schemes function as substitutes, rather than complements, so that when private support is paid on behalf of a single-parent family receiving public support, all or a part of those resources are typically retained by the government or public support is reduced, making custodial parent families no better off if private support is or is not paid (Cancian, Meyer, and Caspar 2008; Skinner et al. 2017). In this article we highlight central policy challenges in meeting the needs of children in single-parent families and consider the role of the current U.S. child support system in responding to those challenges. We argue that the traditional approach to child support, though functional for many middle-income families facing divorce, fails to address key challenges for lower-income families facing divorce and for individuals who, regardless of income, did not have a stable romantic relationship. Unrealistic child support expectations can harm noncustodial parents, create additional barriers for noncustodial parents to be involved with their children, and may even yield less support to vulnerable families than an alternative scheme (for example, Waller and Plotnick 2001). The traditional child support system also fails these families because it does not address the risk children face when their noncustodial parents do not pay support, despite nonpayment, partial payment, and irregular payment being common. By enforcing financial support while ignoring never-married noncustodial parents’ access to their children, the current system is particularly flawed and unsustainable for the growing number of children of lower-income never-married parents. The traditional child support enforcement strategy is premised on an often inaccurate view of noncustodial fathers’ economic resources and employment stability and of parents’ relationships. We highlight the costs of ignoring the disjuncture between ideals and current reality, and some of the key challenges that must be confronted in developing an appropriate policy response. In the next sections, we describe and then evaluate the current system. We then recommend a set of changes to private and public child supports that aim to address the identified challenges. ## THE LOGIC AND FUNCTIONING OF THE CURRENT SYSTEM A number of programs and policies, many covered in other papers in this volume, address the resources available to poor individuals, and especially families with children. Custodial parent families are disproportionately poor and therefore disproportionately affected by these general poverty policies. However, our focus here is on policies designed to address the challenges of families with children in which parents live apart. Although many policies are means-tested and account for the resources provided and required by members of these separated-parent households, policy governing child custody and child support are the primary policies specifically addressing the additional challenges arising when parents live apart.1 In particular, we argue that a child support guarantee is needed, even given a general children’s allowance, as proposed elsewhere in this volume, which reduce the poverty rate for those in married-couple families more significantly than for those in single-parent families (Wimer, Collyer, and Kimberlin 2018). Children who live apart from a parent are at substantially greater risk given economic and other vulnerabilities that emerge when parents live apart. In addition, custodial parents, who typically must serve as both breadwinners and caretakers, face economic and other challenges beyond those faced by “intact” (two-parent) families. Child support policy comes into play when parents with children divorce. Divorce is a legal process within the judicial system and each state has its own rules (or guidelines). In general, a divorce where children are present involves formalized decisions on who will make important decisions for the child (legal custody), with whom the child will live (physical custody), whether there will be financial transfers and at what level (child support), and how joint assets will be divided. Legal and physical custody are typically set based on the best interests of the child, though many states have stated preferences for both parents sharing responsibility and children spending substantial amounts of time with each parent, unless these arrangements are not feasible or determined to be not in the child’s best interests (Cancian et al. 2014). If a child is to live with one parent most of the time, the other parent may have specified visitation privileges, even including a detailed parenting plan specifying which parent has responsibility at each time and how the transitions between parents are handled. Child support obligations are set based on each state’s guideline. Nearly all states have a guideline in which the central principle is *continuity of expenditures*, the idea that noncustodial parents should provide the level of support that they would have had the parents lived together (Garrison 1999). When a child support order is in place and the noncustodial parent is employed, policy requires that the employer automatically withhold the amount of support due and transfer it to a central processing agency that then records the amount paid and distributes it (Pirog and Ziol-Guest 2006). In addition to these services, which should be available to all parents, custodial parents can request the services of the child support agency in their state. This agency can help parents locate the other parent, establish an order for child support, actively monitor whether the order is being paid, and take enforcement actions if it is not, through such steps as taking away a driver’s license, intercepting a tax return, or even bringing civil or criminal charges that may result in imprisonment. The child support agency can take aggressive steps to enforce child support orders, but no comparable enforcement of parenting time is practicable; a parent who does not follow the agreed plan can eventually be brought to court, but no public agency monitors this and enforcement of a parenting plan is quite difficult. Divorce law and procedures are not available to unmarried parents regardless of whether they were living together. The same child support policy does apply, but an extra step is required before child support can be ordered: paternity needs to be formally established or voluntarily acknowledged. If paternity is formally established in a court proceeding, or if a child support order is established in a court proceeding, then an opportunity to formally establish custody and visitation, and to set the rights and responsibilities of each parent, is possible. But in some states, child support orders need not be established by a court, and can instead be done within the child support agency; if so, then an opportunity to formalize custody and visitation is not possible because these are not part of the child support agency’s purview. The federal Office of Child Support Enforcement recently acknowledged that “there is currently no systematic, efficient mechanism for families to establish parenting time agreements for children whose parents were not married at the time of their birth” (2013, 1). This lack of a systematic opportunity for unmarried parents to define roles and set rights and responsibilities is a key way that unmarried parents are disadvantaged relative to divorcing parents. This disadvantage is even greater for lower-income unmarried couples, who are particularly likely to be served by the child support agency. The overrepresentation of low-income families in child support enforcement efforts occurs because the agency serves those having difficulty with child support issues who apply for services (who are more likely to have low incomes) and because lower-income custodial parents are required to cooperate with the agency as a condition of receiving some public benefits and even to sign over their right to child support to the state during periods when they receive Temporary Assistance to Needy Family (TANF) benefits. These same low-income families are also less likely to have the resources to pursue separate legal hearings related to parenting time. The one-sided focus of the child support system thereby leaves low-income families facing potentially punitive enforcement of orders for financial support, without effective access to agreements for parenting time. Lower-income families and individuals are more likely to encounter challenges in the child support system both because of their income and employment status, and because they are more likely to have children outside of marriage. One key difference is that lower-income noncustodial parents are often ordered to pay a higher proportion of their income in child support than middle-income noncustodial parents (for example, Meyer 1998). In part this regressivity is intentional: the guidelines used in most states are called *income shares* and require a smaller percentage of income as the couples’ income increases, consistent with the lower proportion of total income typically spent on children as family income rises. But this is also the result of other factors: when there is no income information available for a noncustodial parent (or they have very low incomes), some states have set orders based on imputed income—though this practice is restricted by recently finalized federal regulations.2 Imputed income often reflects expectations that the noncustodial parent can work full time all year, which many do not, resulting in orders that are a high percentage of actual income. Finally, lower-income fathers are less likely to be in stable marriages and more likely to have had children with multiple partners, so their resources are being stretched across multiple families (Cancian and Meyer 2011; Sinkewicz and Garfinkel 2009). In these ways the child support enforcement system may exacerbate, rather than manage, the inevitable tension between setting orders high enough to provide enough income for children, and low enough to impose a manageable burden on noncustodial parents. This tension is inevitable because noncustodial parents with very low incomes do not have sufficient resources to support their children, even less so if they live apart. On the one hand, even setting orders at a relatively high proportion of income, or assuming income based on full-time low-wage work, may fall short of providing enough resources to meet half of children’s needs. On the other hand, some states have tried a variety of efforts to lower the burden on lower-income noncustodial parents, including allowing the noncustodial parent a certain amount of income for their own purposes before child support is assessed (a self-support reserve) or having a lower percentage requirement for lower-income noncustodial parents. But this comes at the cost of support to the parent caring for the child. And, although orders are often insufficient, and a minority of low-income custodial parents receive all the support due, an entitlement to alternative support is no longer in place.3 Based on this review of the child support system, what are its explicit and implicit goals and what should they be? The federal legislation governing child support services (Title IV-D of the Social Security Act) states a multifaceted purpose: > [E]nforcing the support obligations owed by noncustodial parents to their children and the spouse (or former spouse) with whom such children are living, locating noncustodial parents, establishing paternity, obtaining child and spousal support, and assuring that assistance in obtaining support will be available under this part to all children. (SSA 2000) But these are more on the order of activities than goals. What is the problem that policy for single-parent families, and the child support program in particular, is trying to solve? There are a number of potential answers, and a research literature and legal and regulatory documents that seek to clarify the current state of affairs. Here we clarify a set of priorities that we then use to evaluate policy options. We argue that the primary goal of the child support system is to increase the financial resources available to children living with a single parent. This goal is mostly consistent with an anti-poverty strategy, given that children living with single parents are more likely to be economically vulnerable, but this is also consistent with a recognition of the rights of the child. The secondary goal is to hold parents responsible for the financial support of their children. There are at least two motivations for this goal, which justify different policy preferences. First, private support from parents is often preferred to public support because holding noncustodial parents responsible reduces the burden on taxpayers, and maintains the U.S. policy preference for private support of children (Meyer 2012). Many U.S. benefits to families with children are means-tested, and therefore available only to families who are judged unable to meet the need themselves. In qualifying for means-tested benefits the incomes of both parents are considered for two-parent families. Requiring noncustodial parents to pay child support, and considering child support received as an income source for custodial parents, may be seen as equivalent treatment for single-parent families. A second motivation for requiring financial support from noncustodial parents is that it reduces the economic incentive to stop living with children that would exist if noncustodial parents were free of the obligation to support their children. If the couple made a shared decision to raise children together, but later change their minds, holding noncustodial parents financially responsible reduces the negative consequences for children, custodial parents, and taxpayers. In other words, it retains the right of adults to end their relationship with one another, but requires that they accept responsibility to support children to adulthood. It renders separation (what was called *abandonment* in an earlier era) an ineffective strategy for escaping the financial responsibilities of parenting. On the other hand, it makes it more attractive for a parent who would prefer to care for their child, but not live with the child’s other parent. Both these arguments—providing for continuity of contributions after relationship dissolution, and avoiding a financial incentive to become a noncustodial parent—are more difficult to apply to couples or sexual partners who do not have an affirmative interest in becoming parents, and noncustodial parents who have never lived with their child.4 With respect to continuity of contributions, child support cannot restore what never existed; assessing contributions based on hypothetical living situations requires confronting a number of challenges. (For example, if a father has children with multiple partners, do we imagine him living and sharing resources with each child, neglecting the others, or living with all the children simultaneously, or sequentially?) In addition, avoiding financial incentives to dissolve a prior commitment is arguably quite different from creating incentives to form a partnership that was not otherwise intended. Holding noncustodial parents financially responsible increases the incentive for men to avoid a pregnancy or birth; without a child support requirement, a father with limited connections to the mother would otherwise potentially face few consequences. By the same logic, child support mitigates the financial burden of unintended motherhood, though given the significant consequences of a birth for custodial mothers the incentive effects of financial support might be expected to be relatively small.5 Providing institutional support to regularize never-married noncustodial parents’ contact with their children is a potential policy goal; it is less directly connected to our focus on economic resources, but it is not unrelated. In recent years there has been a growing focus on the potential importance of father involvement, and concern about how fathers’ involvement with other aspects of their children’s lives is related to financial support (for example, Garasky et al. 2010). With respect to noncustodial parents’ involvement, the formal child support system may increase noncustodial parents’ nonfinancial support and involvement with their children when noncustodial parents comply (Garasky et al. 2010; Huang 2009; Koball and Principe 2002; Nepomnyaschy 2007; Peters et al. 2004). But, for noncustodial parents who do not pay, or whose payments do not benefit their children because they are used to offset public welfare costs, child support may be a barrier to involvement (Edin 1995; Gunter 2016; Nepomnyaschy and Garfinkel 2010; Waller and Plotnick 2001). Further, clarifying the rights and responsibilities of noncustodial fathers, for example, with respect to parenting time, is an important challenge largely unmet by the current system for never-married parents. As the dominance of marital childbearing declines, and the traditional pattern of caregiving mother and breadwinning father become less prevalent, the future relevance and success of the child support program may depend on addressing this challenge. Although we do not include support in navigating nonmarital parents’ relationships as a primary goal of the current child support system, we come back to this issue in our recommendations. A number of other goals have been articulated for the child support system. These include recovering public expenditures made in other systems (such as Medicaid), improving the equality of outcomes between custodial and noncustodial parents, and discouraging nonmarital births. These may be worthy goals, at least in some instances, but we do not prioritize these goals in evaluating alternative approaches to child support. We do not believe recovering public expenditures is an appropriate goal for the child support system. A set of policies focused on cost recovery has been found to be ineffective and is regressive, transferring resources from economically vulnerable noncustodial parents to taxpayers (Cancian, Meyer, and Caspar 2008); meanwhile, those most vulnerable—children in low-income custodial parent families—receive the least. Moreover, for the potential goals of equality of outcomes and discouraging nonmarital births, the implications are not straightforward. For example, with respect to improving equality between custodial and noncustodial parents, this is consistent with the other goals of child support when the noncustodial parent is relatively better off, because requiring support from the noncustodial parent is equalizing, but also increases resources for the child and enforces parental responsibility. However, these goals are at odds when noncustodial parents are relatively disadvantaged (Ha, Cancian, and Meyer 2016). Finally, enforcing child support shifts the burden of nonmarital births somewhat from custodial to noncustodial parents, rather than simply discouraging nonmarital births. In sum, we argue that designing policies that provide adequate resources to children and appropriately encourage parental responsibility across a range of situations is a challenge that requires and deserves a focused response. ## EVALUATING THE CURRENT SYSTEM How well does the current system meet the twin goals of supporting children and enforcing parental responsibility? The answer is mixed, and substantially different for families eligible for child support due to divorce or nonmarital birth. Recent federal estimates suggest that more than one in four children under the age of twenty-one in the United States lived with one parent but not the other (Grall 2016). In most (more than 80 percent) cases, census data suggest that the child or children lived with their mother, but not their father, and that pattern has remained fairly stable over time. Other analyses, using more detailed data for select samples, such as court cases in Wisconsin, suggest that shared custody is becoming more common in divorce cases, only about half of cases living with only their mother (Cancian et al. 2014). However, shared custody is less common for low-income divorcing families, and, especially, for never-married couples (Cancian et al. 2012). Thus, although most children in single-parent families live only with their mother, differences by class and union status are substantial, children of lower-income and never-married parents being less likely to share time with each parent. Children living with single parents are much more likely to be poor, and tend to have worse long-term social and economic outcomes, than children raised by married parents do (see, for example, Amato 2005; McLanahan and Sandefur 1994). In 2015, 11 percent of children living with both their married parents, relative to 43 percent of children living with just their mother, were in poor families (U.S. Census Bureau 2015). To the extent that these discrepancies are related to the reduced economic resources available from a single parent, requiring financial support from the noncustodial parent may seem an obvious solution. About half of the thirteen million custodial parents have a formal child support order—53 percent of all custodial mothers but only 31 percent of custodial fathers (Grall 2016). The proportion of custodial parents with orders increased through 2003, but has declined since then (Grall 2016). The change could be due to declines in the number of TANF families who are required to cooperate with child support, the economic prospects of noncustodial parents, or increases in shared custody (Meyer, Cancian, and Chen 2015; Schroeder 2016). Orders are more common among white Non-Hispanic (56 percent) than Hispanic (44 percent) or African American (37 percent) parents, among college graduates (54 percent) than those with only a high school degree (46 percent) or less (38 percent), and among divorced (58 percent) than among never-married (42 percent) parents. For unmarried parents, paternity establishment is a prerequisite for a child support order. Many parents voluntarily establish paternity, often immediately following birth. However, establishing paternity creates legal and financial obligations for the father but extends relatively few rights; as discussed, child support agencies are tasked with enforcing the same financial expectations for divorced and never-married parents, but only divorcing parents typically have formal custody and visitation agreements. Moreover, for low-income families receiving means-tested assistance, formal child support payments may be used to reimburse government costs rather than to directly benefit children—undercutting the incentive for parents to engage with the system. Even for those who have orders, payment is not ensured. Fewer than half of all custodial parents who were supposed to receive support received all the child support due, and a quarter received nothing in a given year (Grall 2016). Moreover, although national data do not provide detailed accounting on the timing of payments, an analysis of administrative records in Wisconsin showed that only about half of those who received some child support in a year received it regularly, that is, in at least ten months (Ha, Cancian, and Meyer 2011). This irregularity is another important limitation of the current system, and can cause uncertainty and stress and make it difficult for custodial parents to plan for the future. Nonetheless, national estimates suggest that average orders are substantial for those who have them ($6,772 in 2013 dollars for divorced parents, and $4,486 for never-married parents), as are average amounts received for those supposed to receive support, especially divorced parents ($5,209, twice the $2,538 average for never-married parents) (Grall 2016). This brief review of the most current data and related research suggests that in many cases the current child support system can be judged a success in meeting our two primary goals, because it transfers significant support from some noncustodial parents to their children, thereby supporting children and enforcing parental responsibility. The system works best for families in which the noncustodial parent has stable formal employment—which gives the noncustodial parent the means to pay, and generally results in automatic wage withholding of the child support due. However, custodial parents whose children are most in need of assistance are less likely to be owed support, and when they are, they are disproportionately more likely to be owed support from noncustodial parents who also have limited resources (Sinkewicz and Garfinkel 2009). As a result, they are less likely to receive support and to receive less when they do receive it; for example, even when custodial parents below poverty received support, they received on average about $1,000 less than nonpoor custodial parents. Only one-quarter of custodial parents below poverty received any child support and only 13 percent received all the child support due to them in 2013 (Grall 2016). Further, although the data are limited, it appears likely that low-income custodial parents are also least likely to receive *regular* child support because the noncustodial parents owing this support are more likely to have irregular employment. Thus, with disadvantages at every point in the process, the imperfect system falls far short of what is required for those who need it most. The first overall problem with the current child support system, then, is straightforward: far too few children receive child support, receive substantial amounts, or receive this regularly. What are some of the reasons that child support provides so little support, on average, to low-income custodial parents? First, under- or unemployment means many noncustodial fathers of low-income children do not have enough income to provide substantial or consistent support. To the extent that these fathers would have provided relatively little support even if they lived with their children, this is less a problem specific to the child support system than a limitation of general support for low-income families. (Other papers in this volume address related policy options—including a child allowance and employment programs.) Still, a policy regime that relies on the support of noncustodial parents will often fail if these parents do not have the resources to provide support. Second, mass incarceration leaves many noncustodial parents of economically vulnerable children unable to pay support while incarcerated, and with reduced earnings potential after they are released (Chung 2012; Geller, Garfinkel, and Western 2011). Moreover, child support enforcement is a contributing factor to high levels of incarceration: sometimes we incarcerate those who are behind in their child support payments, and unmanageable child support obligations can discourage formal employment (Cancian, Heinrich, and Chung 2013). Another contributing factor is the instability of marital and nonmarital relationships. A substantial proportion of noncustodial parents have had children with more than one partner, a more common phenomenon among those with the fewest resources (for example, Carlson and Furstenberg 2006; Cancian, Meyer, and Cook 2011). We highlight the problems created by multiple-partner fertility here because they are common, cause payment difficulties, and challenge basic notions of fairness. Some research suggests that more than half of children who were their mother’s first child born outside of marriage will have a half-sibling by their tenth birthday (Cancian, Meyer, and Cook 2011). Failing to account for noncustodial fathers’ potential obligations to multiple families creates significant overestimates of how much child support could be collected (Sinkewicz and Garfinkel 2009). Moreover, multiple-partner fertility is most prevalent among lower-income couples, making it even more difficult for noncustodial parents to provide support across more than one custodial parent family. Finally, multiple-partner fertility creates classic tradeoff problems for child support policy in that basic notions of fairness cannot all be simultaneously met (Meyer, Cancian, and Cook 2005; Meyer, Skinner, and Davidson 2011). In summary, a policy scheme that relies on private support cannot meet our first goal of providing financial support to children when a noncustodial parent’s few resources are spread across multiple families. Moreover, some of the cause of no, low, or irregular child support receipts is social policy itself. We noted that the policy scheme in the United States makes public benefits for low-income custodial parents and child support from noncustodial parents function as substitutes. The TANF program requires recipients to sign away their right to child support payments during their period of recipiency, and many states retain all child support paid on behalf of children receiving benefits and use these receipts to offset public expenditures rather than sending them to the children. As a result, children receiving TANF do not benefit from a noncustodial parent’s payment. But this problem also occurs for non-TANF families (and TANF families in states that do pass through a portion of the child support) because child support counts as income in some means-tested programs (food stamps, housing vouchers).6 Thus, even if child support goes to the custodial parent family, their other benefits may then be reduced, making child support function as a cost recovery mechanism rather than supporting children, our first goal. Child support policy is limited in encouraging parental responsibility, especially for never-married families. The U.S. child support scheme tends to focus exclusively on private financial transfers, in which any benefit to one parent is taken from the other. Current policy also has a large divide between child support payments and other aspects of parenting for never-married parents, even though these other aspects could generate more parental agreement. We noted that the child support system includes no natural place in which unmarried parents decide custody and visitation, let alone to gain skills in co-parenting. In most jurisdictions, if parents are able to come to some agreement, few if any institutional supports and few ways to enforce them are available for maintaining the agreement, for both unmarried and divorcing parents. Thus, many noncustodial parents feel that the child support system does not support their concerns, does not help them with developing a relationship with their children, and sees them only as a financial resource (Waller and Plotnick 2001). This may undercut noncustodial parents’ willingness to provide financially. In our assessment, current child support policy is often ineffective because it is based on unrealistic assumptions. The enforcement system was designed to enforce notions of paternal financial responsibility that, even if somewhat idealized, were grounded in broadly held views of appropriate family structure (for example, parents should marry), and gender roles (for example, fathers, more than mothers, should work for pay). Increasing nonmarital births, declines in postconception marriage (and in marriage in general), and union instability, signal a disjuncture between these views and contemporary realities.7 More than 40 percent of all children are now born to unmarried parents. Imposing paternal responsibility based on biology alone is now more contested. Moreover, although mothers (who remain the more likely custodial parent) still work and earn less than fathers, the gap has declined, and reversed for some subgroups, making fathers’ expected contributions more contested (Cancian, Meyer, and Han 2011). Finally, changing patterns of contraception and abortion that give women more control over fertility, also make biological fathers’ rights and responsibilities more contested. As a result, the current child support system, which still prioritizes biological responsibility, assumes that a noncustodial parent had a relationship with a child before separation (and thus should continue to support that child financially), was designed for simple families with one noncustodial parent and one custodial parent who have had children with no one else, and assumes that fathers have advantages in the labor market, will increasingly confront challenges to its effectiveness. At the same time, changes in the structure of the U.S. safety net, especially the lack of entitlement to cash assistance and program rules that require that custodial parents provide for their children financially as well as providing caregiving, makes it essential that children with only one parent’s financial support have additional resources. To effectively address these challenges, we need to reorient child support policy for low-income single-parent families. ## A NEW APPROACH TO SUPPORTING CHILDREN AND ENCOURAGING PARENTAL RESPONSIBILITY Even while maintaining a focus on providing economic support and encouraging parental responsibility, we argue for transforming the child support program and broadening its mission beyond the assessment, collection, and enforcement of financial transfers from noncustodial parents to custodial parent families. The more comprehensive child support program would provide guaranteed payments to custodial parents, essentially extending a combination of grants (that is, with no expectation of repayment) and loans (with repayment required) to noncustodial parents who are unable to meet their obligations. Although we focus on the agency’s role in financial support (the core of the agency’s mission), we also argue for a complementary set of agency activities supporting stronger family relationships. ### Financial Support At the core, our model of family support would achieve the two central goals of the current child support program—supporting children and encouraging parental responsibility—by balancing private responsibility and support, with public responsibility for both enforcement and support of parents’ ability to meet their responsibilities. The essential elements of our proposal are a minimum monthly support amount per child, a maximum child support obligation for noncustodial parents, and a public guarantee to bridge the gap when the minimum support for children exceeds what the noncustodial parent can reasonably pay. We include a specific proposal, to illustrate and support estimates of costs and impact. In particular, we propose a guaranteed minimum child support of $150 per month per child, and a child support order standard of 12.5 percent of noncustodial parent income per child. Each noncustodial parent’s total current contributions are capped at 33 percent of income. In our basic proposal, we also assume that child support income up to the amount of the guarantee (whether this comes from the noncustodial parent or the government) would not count in determining eligibility and benefits for means-tested programs; we also assume that noncustodial parents would accrue debt when they failed to pay the current support due, or when the 33 percent maximum cap reduced their current payments below 12.5 percent per child. We next detail the proposal and outline a set of potential variations, each of which have implications for the costs and benefits of the new system. The minimum guarantee of $150 per month per child ensures a reliable income source for all children living apart from a parent, regardless of the noncustodial parent’s income or payment status. We propose that the government guarantee support up to the minimum benefit, so that every child would receive the minimum benefit every month.8 This minimum benefit would be available to all children covered by a child support order, whether they were receiving other benefits or not. Because it provides a reliable income stream, it would decrease the insecurity and stress currently associated with irregular child support payments. Although $150 per child is modest relative to continuity of expenditures for middle-income families, it is higher than that standard would provide in very low-income cases. The per child minimum, the 12.5 percent per child order, and disregarding the minimum amount of child support in means-tested benefits all reflect a child’s rights, which are not diminished when there are siblings nor when the custodial parent has a low income and is therefore receiving means-tested benefits. For noncustodial parents with very low incomes, the minimum benefit of $150 would be higher than their expected contribution of 12.5 percent, and the government would provide the difference.9 If the noncustodial parent failed to pay their expected contribution the government would pay up to the minimum benefit and hold noncustodial parents responsible. By failing to account for economies of scale, a per child approach requires less of a contribution from noncustodial parents with only one child, and relatively more from noncustodial parents with multiple children with the same partner, relative to a continuity-of-expenditures approach.10 While it does not account for differences in family size, it reduces inequities and complications that arise with complex families (Meyer, Cancian, and Cook 2005), which are estimated to be a majority of families of low-income never-married parents (Cancian, Meyer, and Cook 2011). The maximum per noncustodial parent contribution of 33 percent would avoid clearly unmanageable child support burdens. At the values we propose—12.5 percent contribution per child, and 33 percent maximum total contribution—the expected contribution for noncustodial parents with more than two children would exceed the 33 percent maximum. This gap would be paid by the government up to the $150 per child minimum benefit level. In our base proposal, amounts above the maximum noncustodial parent contribution rate would be considered arrears to be paid (with minimal interest) after the children reach majority and the order for current support ends. Figures 1 and 5 illustrate the implications of our base proposal for noncustodial parents with one child who pay all current support due (1) or who pay nothing (5). For each level of noncustodial parent income, the figure shows support owed by the noncustodial parent and received by the custodial parent. In the case of custodial parent receipt, we differentiate between support provided from the noncustodial parent, from a government subsidy (public support provided when 12.5 percent of the noncustodial parent’s income is less than the $150 per month minimum), or from a government loan to the noncustodial parent (public support advanced by the government, but to be repaid). For example, on figure 1, for one child, the noncustodial parent owes 12.5 percent of income, and the government guarantees up to $150 per month. Thus, the custodial parent receives $150 a month, from a combination of public subsidy and noncustodial parent payments, when the noncustodial parent income is below $1200 per month. When noncustodial income is higher, the custodial parent receives more than the minimum, all from the noncustodial parent (and there is a government guarantee, but no government payment). In contrast, as shown in figure 5, if the noncustodial parent fails to pay child support, the custodial parent will receive the $150 per month minimum at all income levels. For cases with a low-income noncustodial parent, the payment will include a subsidy and an advance (a loan from the government to the noncustodial parent), and no additional support will be owed to the custodial parent. For cases with higher-income noncustodial parents, the $150 minimum will all be an advance, and additional support will remain due from the noncustodial parent to the custodial parent. ![Figure 1.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F1.medium.gif) [Figure 1.](http://www.rsfjournal.org/content/4/2/91/F1) **Figure 1.** One Child, Full Payment The remaining figures (2 through 4 and 6 through 8) show the outcomes for noncustodial parents with additional children. For noncustodial parents with three or more children we distinguish *total support* owed, and *current* support owed—since noncustodial parents with more than two children will have some amount deferred, given the 33 percent maximum current support level. For example, for four children (figure 4) the noncustodial parent will owe a total of 50 percent of income (12.5 percent for each of four children), but current support will be capped at 33 percent. Thus, a custodial parent will receive $150 per month per child, or a total (across all children) of 33 percent of income, whichever is greater.11 Noncustodial parents will pay 33 percent of income, and will owe the remaining 17 percent of income when current child support is no longer due, to the government (for the amount due toward the $150 minimum) or to the custodial parent (for the amount due over the $150 minimum). In the case of noncustodial parents who are not paying support (6), the custodial parent will receive $150 per month per child, and the noncustodial parent will owe 33 percent of income immediately, and the additional 17 percent of income when current support is no longer due. ![Figure 2.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F2.medium.gif) [Figure 2.](http://www.rsfjournal.org/content/4/2/91/F2) **Figure 2.** Two Children, Full Payment ![Figure 3.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F3.medium.gif) [Figure 3.](http://www.rsfjournal.org/content/4/2/91/F3) **Figure 3.** Three Children, Full Payment ![Figure 4.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F4.medium.gif) [Figure 4.](http://www.rsfjournal.org/content/4/2/91/F4) **Figure 4.** Four Children, Full Payment ![Figure 5.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F5.medium.gif) [Figure 5.](http://www.rsfjournal.org/content/4/2/91/F5) **Figure 5.** One Child, No Payment ![Figure 6.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F6.medium.gif) [Figure 6.](http://www.rsfjournal.org/content/4/2/91/F6) **Figure 6.** Two Children, No Payment ![Figure 7.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F7.medium.gif) [Figure 7.](http://www.rsfjournal.org/content/4/2/91/F7) **Figure 7.** Three Children, No Payment ![Figure 8.](http://www.rsfjournal.org/https://www.rsfjournal.org/content/rsfjss/4/2/91/F8.medium.gif) [Figure 8.](http://www.rsfjournal.org/content/4/2/91/F8) **Figure 8.** Four Children, No Payment Some of the potential costs and benefits of the minimum guarantee are included elsewhere in this double issue (see Wimer, Collyer, and Kimberlin 2018). Those estimates show a modest reduction in the overall poverty rate (1.3 percent), and a larger effect among those who receive it, decreasing their poverty rate by 5.2 percentage points (or by 22.7 percent), for an estimated cost of $8.2 billion. These estimates, the best available in the current context, are limited in several ways. First, they are based on only one of our proposals, the guarantee; they do not consider the proposals for changing the child support formula and for setting a maximum amount of support required. These other parts of our proposal might be consequential. For example, a cap on the percentage of income required from noncustodial parents would lower the amount that some noncustodial parents pay, which would then increase their disposable incomes and lower their poverty according to the Supplemental Poverty Measure. The $150 minimum monthly support per child would exceed the support received by many low-income families, but those currently receiving both more than $150 per month and more than 12.5 percent, per child, would receive less support under this proposal, decreasing their disposable incomes. More generally, our proposal to switch to a fixed proportion of income per child would have implications for the disposable incomes of custodial and noncustodial parents. These effects could not be estimated with available data. Second, the estimates do not incorporate any second-round effects (behavioral changes) induced by a minimum child support guarantee (or by the other changes we propose); similarly, they do not consider changes that may take longer to emerge.12 At the end of this section, we discuss what we view as the major concerns: changes in the incentives to have an order and to make payments. More minor limitations include not counting changes in TANF or SNAP, and the lack of a labor supply effect. For example, some custodial parents may be receiving TANF because of the regularity of income it provides; they might find the package of a guaranteed benefit and earnings better than continued TANF receipt. Some dated research has examined the sensitivity of cost estimates to incorporation of labor supply effects of a guarantee (Meyer and Kim 1998); this research shows that typical estimates of a labor supply response do not change overall cost and benefit estimates much because some welfare recipients are predicted to work more and some nonrecipients to work less.13 The guarantee may also affect fertility, family dissolution, or custody, but none of these effects have been estimated. It is difficult to compare these results with previous estimates, in part because the parameters differ and in part because the child support system, labor market, and policy regime has changed substantially from the 1980s to which previous estimates refer. Elaine Sorensen and Sandra Clark estimate a 1.2 percent poverty reduction among all children, at a cost of about $4.3 billion in 2016 dollars (1994). Daniel Meyer and Rebecca Kim, using a different model, different data, and a higher guarantee, estimate a decline in poverty among custodial parent families of 2.1 percent and net costs of $6.6 billion, also in 2016 dollars (1998).14 These estimates are generally comparable to previous estimates: there would be a small reduction in poverty overall, a larger one among custodial parents, and particularly among those custodial parent families that were receiving the benefit. The current cost estimates are broadly consistent with the previous work, though they suggest somewhat higher costs because some savings are not considered and because the per child guarantee is more expensive than a guarantee that has marginal declines with the number of children. We have outlined one approach to balancing private responsibility and public support with a guaranteed per child minimum support amount, a per child percent of noncustodial parent income due, and a maximum current per noncustodial parent contribution. The basic framework can be modified along a number of dimensions, altering the costs and benefits: > *The minimum assured support could be set higher or lower than $150 per month per child*. Our proposed guarantee of $150 per child per month is modest relative to some prior proposals.15 A lower minimum would be less expensive for taxpayers, but would reduce support to children of the lowest income noncustodial parents, for which the minimum benefit exceeds the percentage-of-income contribution expected from the noncustodial parent. A higher minimum benefit would increase costs but provide more to children of low-income noncustodial parents, and be a larger reliable income source for all children living apart from one of their parents. > > *The percentage of noncustodial parent income could be set higher or lower than 12*.5 *percent per child*. A lower amount would yield orders below current guidelines in most cases with one or two children, but orders more consistent with current guidelines for larger families. A higher percentage standard would not affect current support due from noncustodial parents with more than two children, unless the maximum contribution were increased. > > *The maximum noncustodial parent contribution could be set higher or lower than 33 percent of noncustodial parent income, for all noncustodial parents, or for those noncustodial parents with higher incomes*. In our base scheme, where support above the maximum contribution is to be paid by the noncustodial parent at a later date, and where support up to the $150 per child minimum is assured, a change in the maximum contribution would not affect the support received by children with low-income noncustodial parents, but would delay a greater share of the payments and allow the noncustodial parent additional income for current expenses. For higher-income noncustodial parents, a lower maximum would delay payments to the children. An alternative would be to provide for a noncustodial parent self-support reserve in calculating maximum current contributions. > > *Part of the minimum support payment could be contingent on the noncustodial parent’s contribution*. For example, given a minimum of $150 per month per child, guaranteed support could be reduced to $100 per month, with the remaining $50 per month per child provided only if the noncustodial parent paid at least $50 per month.16 This would reduce the amount of reliable support for families of noncustodial parents who did not pay, but would increase the incentive for noncustodial parents to pay; children would receive more income if noncustodial parents paid support than if they did not pay—not otherwise the case for low-income noncustodial parents with orders below the $150 minimum. > > *Child support income up to the per child minimum could be considered in determining eligibility and benefits for means-tested programs*. Although this is not consistent with child support being the right of the child and would result in a lowered value of the child support for single-parent families, who are arguably more vulnerable than two-parent families with comparable needs-adjusted incomes, it would be less expensive for taxpayers. Moreover, ignoring the benefit in these other programs generates inequities between families with parents living together and those living apart. We suggest minimizing the disregard to the per child minimum (proposed to be $150 per month) because the horizontal inequity generated by the disregard would be more problematic at higher levels. > > *The guarantee could be provided to all children living with a single parent, even without a child support order.* This would be more consistent with guaranteed support as the right of the child, and also recognizes the vulnerability of many children who do not have a child support order on their behalf (and may not have legal paternity established). However, this would undercut the incentive for low-income families to establish an order, and would significantly raise costs. A more modest expansion would require a child support order to qualify for the guarantee, except when there is good cause (for example, due to domestic violence) for not pursuing an order. In addition to concerns regarding public costs, guaranteed minimum child support raises important concerns regarding the incentive for noncustodial parents to pay child support and to work. Because children would receive the minimum $150 per month per child support regardless of the noncustodial parents’ payments, a child support guarantee would reduce the incentive for noncustodial parents to pay support. Some evidence indicates that noncustodial parents are more likely to pay support, or to pay more, when their children benefit (Cancian, Meyer, and Caspar 2008), but the estimated effects are relatively small and, we would argue, not enough to justify risking the minimum economic well-being of children. Moreover, although noncustodial parents working outside the formal economy may largely avoid child support enforcement efforts, their counterparts in the formal economy are subject to automatic withholding, largely making child support payments nondiscretionary (Bartfeld and Meyer 2003). To the extent that noncustodial parents are working so that they can support their children, a guaranteed payment means they have less incentive to work. However, previous analyses suggest that this potential effect is likely to be small (Freeman and Waldfogel 1998). Moreover, the minimum guarantee, available only to children with a child support order, would create a substantial positive incentive for custodial parents (and many noncustodial parents) to cooperate in establishing an order, which may reverse the decline in custodial parents participating in the child support system (Schroeder 2016). ### Broader Supports for Parents A new approach to assessing and ensuring financial support is at the core of our proposal. But policies enforcing noncustodial parents’ financial responsibilities to their children will be most productive in a context that also supports parental responsibility more broadly, rather than focusing only on financial transfers. Addressing an array of issues for separating parents could improve co-parenting relationships, noncustodial parent-child relationships, and would rebalance the system from its overemphasis on finances. Although not directly focused on financial resources, improving these relationships might be expected to also lead to additional financial support. Moreover, putting these services in an agency that serves not only separating parents but also parents who are together would encourage a focus on child well-being and helping all parents. This model is similar to models being implemented in Australia (Moloney et al. 2013) and the United Kingdom (Skinner 2012). Although married, cohabiting, and separating parents confront many related issues, effective policy must acknowledge and provide an institutional context for managing differences in biological parents’ relationships with each other and with their children. Many children are born to parents who are not in a sustained or stable relationship, to parents who may not have intended to conceive a child, and even in circumstances in which the father may have had no part in the mother’s decisions with respect to the pregnancy. Difficult policy challenges arise in a context where biological and social or emotional ties are inconsistent. Current policy largely ignores the inconsistency and imposes equivalent financial responsibilities on all noncustodial parents, but does not offer clear guidance or institutional support for managing noncustodial parents’ rights or responsibilities with respect, for example, to time with their children. One step in resolving this tension would be to expand institutional supports for unmarried parents to resolve issues related to their relationship with each other and their children. Many unmarried mothers express an interest in their child’s father’s playing a role. The lack of formal institutional supports, including access to mediation and agency and court agreements related to visitation, may contribute to the challenges unmarried parents face in establishing and maintaining positive relationships. When both parents want to share responsibilities for their child, child support services should include efforts to establish both orders for financial support and parenting plans.17 When parents disagree about their roles and responsibilities, the appropriate policy response is considerably less clear. At present, pregnant women have a legal right to terminate a pregnancy, or, if they carry a child to term, to initiate termination of parental rights so that the child may be adopted. Unmarried mothers who do not rely on public benefits are largely free to decide whether to identify the biological father, establish paternity, and pursue a child support order. In contrast, low-income mothers are routinely required to cooperate with the child support agency as a condition of receiving Medicaid (including for childbirth), or other means-tested benefits. Once paternity is established, fathers are subject to the same child support responsibilities, regardless of their relationship with the mother, or their intentions or interests with respect to the birth. The current approach to child support, holding biological fathers responsible irrespective of their relationship to the mother or child, was once consistent with widely held attitudes and behaviors. But, as we have argued, declines in postconception marriage, increases in nonmarital childbearing, as well as increases in women’s socioeconomic independence have undercut much of the logic for this approach. The contested nature of parental rights and responsibilities highlights the importance of institutional support for resolving issues related to parents’ relationships with each other and their children. ## SUMMARY The United States has a variety of programs for low-income families with children, but current policy for single-parent families is primarily located in the child support system. We argue that the system’s primary goal should be increasing the financial resources available to children living with a single parent, with a secondary goal of holding parents responsible for the support of their children. When assessed by this standard, current policy has been substantially successful for divorcing families in which the noncustodial parent has at least moderate earnings in the formal economy. But the current system clearly does not work well for lower-income families, especially those couples who were not married. The problems are clear: far too few children regularly receive substantial amounts of child support and, by being overly focused on financial transfers, the system is sometimes counterproductive to the broader mission of encouraging responsibility. We argued that a key part of the problem is unrealistic, outdated assumptions about separated parents. We propose a reformed system that includes a guaranteed minimum support per child and assurances that no noncustodial parent will be charged beyond their means. We also propose a new agency that focuses not only on these financial transfers, but also on supporting parents’ relationships with their children. These reforms, aimed especially at addressing the challenges facing families with parents living apart, combined with other reforms supporting low-income families in general, would transform the resources available to economically vulnerable children and families. ## FOOTNOTES * 1. Although social policy discussions often presume that single parents are entitled to programs that they would not receive if they were to marry, we find little evidence of programs that are available only to (or provide extra benefits to) those who are single parents, per se, outside of the child support system. Single parents are entitled to (or eligible for) some programs because they are parents who have low incomes, but in most cases are not differentially eligible based on single-parenthood in and of itself. In fact, single parents who marry someone without income would be eligible for *more* of some benefits because their family size is larger. The federal income tax system does have a special filing status for those who are head of household (that is, single parents). However, the amount of tax assessed for those filing head of household is the same or more than those who are married filing jointly, and those married have more exemptions and higher standard deductions, all else equal. One exception relates to work requirements, which may be greater for married couples—for example, in the case of TANF. * 2. The “Flexibility, Efficiency, and Modernization in Child Support (CMS-2343-P) Enforcement Programs” Rule (CMS-2343-P) clarifies policies designed to ensure that orders are consistent with a noncustodial parent’s ability to pay. * 3. Another difference is that middle-class families are more likely to opt for shared physical custody, and this custody arrangement typically requires fewer financial transfers (lower child support orders, if any). Because the difference in support due is intended to reflect differences in expenses associated with physical custody, the implications for resources available to each parents are not clear. * 4. We also note that these arguments, and our proposal, do not address the case of children with a deceased (rather than nonresident) parent. * 5. We use gendered language and assume that children born to parents without a significant relationship will remain with the birth mother. Father custody has increased, but remains unusual for children born to never-married parents (Grall 2016). An additional complication arises outside a stable relationship when mothers have a legal option to terminate a pregnancy, and fathers cannot be expected to have had an influence on the decision. We address this issue in the final section. * 6. Although child support generally counts as income in other means-tested programs, child support payments are not consistently subtracted from the income of noncustodial parents. * 7. For example, the percent of premaritally conceived children born to married parents fell from more than half in the 1960s to just over a quarter in the late 1980s (DHHS 1995), and has continued to fall, even among those cohabiting at the time of conception. Daniel Lichter, Sharon Sassler, and Richard Turner, using data from the 2006–2010 National Survey of Family Growth, estimate that for couples in which the mother is at least twenty-five years old, 16 percent of those cohabiting, and 5 percent of those noncohabiting at the time of conception, marry before a birth (2014). * 8. A guaranteed child support amount is not a new idea. We owe much to the work of Irwin Garfinkel, who has written extensively about how it might work (see especially Garfinkel 1992; Garfinkel et al. 1992). Note that an assured benefit would now be even easier to implement than when it was proposed by Garfinkel, because nearly all child support payments go through a central registry. The mechanism for collecting child support and distributing it (or the guarantee, if the amount collected is less than the guarantee) is basically in place. * 9. A flat percentage of income has the advantage of simplicity. Current policy in many states calls for a lower percentage for high-income cases (reflecting continuity of expenditures). For lower-income noncustodial parents, policies in many states suggest a lower percentage (to reduce burden). This concern remains relevant, though less urgent given the 33 percent cap. Other policies call for a higher percentage for lower-income parents in order to reach a minimum level of support, but that concern is mitigated by the minimum guarantee. * 10. The implications of family size for child support owed by noncustodial parents and due to custodial parents is complicated. Following the continuity-of-expenditures logic, child support guidelines for simple families—in which the mother and father have children together, and with no other partners—generally call for higher orders for larger families, with smaller increments for each additional child. For example, the Wisconsin guidelines call for 17 percent of the noncustodial parent’s income for one child, 25 percent for two children (that is, 8 percent more than for one), and 29 percent for three (4 percent more than for two). However, whereas a father paying support for two children born to the same mother would owe 25 percent of his income, a father paying support owed to two children born to different mothers would owe 17 percent to the first born, and 17 percent of his remaining income (14 percent of total income) to the second, for a total of 31 percent. Similarly, a mother owed support for two children would be due 25 percent of the father’s income if both children had the same father, and 17 percent of each father’s income (net of any prior child support owed by those fathers) if there were two fathers. * 11. We have not specified the distribution of support across multiple families, when support is owed to more than one custodial parent. As we discuss elsewhere, the appropriate distribution is complex (Meyer, Cancian, and Cook 2005; Cancian and Meyer 2011). However, the minimum monthly guarantee per child substantially improves the outcomes and options for the most complex cases. * 12. For example, the current proposal calls for arrears to accumulate if the noncustodial parent’s expected contribution of 12.5 percent of income is greater than the cap. The estimates do not consider eventual payments on these arrears. * 13. The conclusion about limited effects of allowing labor supply responses come from their estimates of a guarantee of $1,500 per year (in 1988 dollars) for a one-child family; this is equivalent to $3,043 in 2016 dollars using the CPI-U (relative to the guarantee we propose here of $1,800 per year). Daniel Meyer and Rebecca Kim did not use a per-child model, so their benefit for a three-child family of $2,500 translates into $5,072 in 2106 dollars, similar to the $5,400 proposed here. * 14. The estimates from Sorensen and Clark are for a guarantee of $2,491 for one child, $4,152 for two (both 2016 dollars), no reduction in AFDC benefits, and only families with child support orders being eligible. The estimates from Meyer and Kim are for a guarantee of $3,043 for one child and $4,058 for two (both 2016 dollars), a dollar-for-dollar reduction in AFDC benefits, and only families with child support orders being eligible. Meyer and Kim do show components of costs and savings, and the difference that including labor supply makes. In their model costs include the direct cost of the guarantee, and small increases in EITC benefits and other tax decreases; savings come from AFDC and food stamps. The labor supply module adds about $500 million to costs (2016 dollars). * 15. A guarantee of $150 per child per month yields an annual minimum of $1,800 per year. In 1991 the National Commission on Children recommended a minimum annual benefit of $1,500 for one child; others proposed a guarantee of $3,000 (Sorensen and Clark 1994)—equivalent to about $2,650 to $5,300 today. * 16. If the noncustodial parent paid less than $50 per month, the child would receive less than the minimum of $150, if the noncustodial parent paid between $50 and $150 per month the child would receive $150, and if the noncustodial parent owed and paid more than $150 per month the child would receive the higher amount. * 17. For example, see the discussion of the Co-Parent Court program in Minnesota (Marczak et al. 2015). * © 2018 Russell Sage Foundation. Cancian, Maria, and Daniel R. Meyer. 2018. “Reforming Policy for Single-Parent Families to Reduce Child Poverty.” *RSF: The Russell Sage Foundation Journal of the Social Sciences* 4(2): 91–112. DOI: 10.7758/RSF.2018.4.2.05. Paper presented at the Russell Sage Foundation Conference “Anti-poverty Policy Initiatives for the United States” and at the related session at the 2016 Association for Public Policy Analysis and Management Research Conference. We thank Chris Wimer and Sophie Collyer for the cost and poverty estimates, Maria Serakos and Veronique Yeo for research assistance, and the reviewers, the editors, and conference participants, especially Lawrence Berger, Robert Doar, Irwin Garfinkel, and Elaine Sorensen, for their helpful comments. Any opinions expressed here are our own. Direct correspondence to: Maria Cancian at maria.cancian{at}wisc.edu, 3436 Social Sciences Building, 1180 Observatory Dr., Madison, WI 53706. Open Access Policy: *RSF: The Russell Sage Foundation Journal of the Social Sciences* is an open access journal. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. ## REFERENCES 1. Amato, Paul. R. 2005. “The Impact of Family Formation Change on the Cognitive, Social, and Emotional Well-Being of the Next Generation.” The Future of Children 15(2): 75–96. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1353/foc.2005.0012&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=16158731&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000232149200005&link_type=ISI) 2. Bartfeld, Judi, and Daniel R. Meyer. 2003. “Child Support Compliance among Discretionary and Nondiscretionary Obligors.” Social Service Review 77(3): 347–72. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1086/375793&link_type=DOI) 3. Cancian, Maria, Yiyu Chen, Eunhee Han, and Daniel R. Meyer. 2012. “Exploring Reasons for the Decline in Child Support Orders Among Paternity Cases.” Report to the Wisconsin Department of Children and Families. Madison, Wisc.: Institute for Research on Poverty. 4. Cancian, Maria, Carolyn J. Heinrich, , and Yiyoon Chung. 2013. “Discouraging Disadvantaged Fathers’ Employment: An Unintended Consequence of Policies Designed to Support Families.” Journal of Policy Analysis and Management. 32(4): 758–84. 5. Cancian, Maria, and Daniel R. Meyer. 2011. “Who Owes What to Whom? Child Support Policy Given Multiple-Partner Fertility.” Social Service Review 85(4): 587–617. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1086/664199&link_type=DOI) 6. Cancian, Maria, Daniel R. Meyer, Patricia R. Brown, , and Steven T. Cook. 2014. “Who Gets Custody Now? Dramatic Changes in Children’s Living Arrangements After Divorce.” Demography 51(4): 1381–96. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1007/s13524-014-0307-8&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=24811135&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) 7. Cancian, Maria, Daniel R. Meyer, , and Emma Caspar. 2008. “Welfare and Child Support: Complements, Not Substitutes.” Journal of Policy Analysis and Management 27(2): 354–75. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1002/pam.20328&link_type=DOI) 8. Cancian, Maria, Daniel R. Meyer, , and Steven T. Cook. 2011. “The Evolution of Family Complexity from the Perspective of Children.” Demography 48(3): 957–82. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1007/s13524-011-0041-4&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=21671198&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000292930900007&link_type=ISI) 9. Cancian, Maria, Daniel R. Meyer, , and Eunhee Han. 2011. “Child Support: Responsible Fatherhood and the Quid Pro Quo.” Annals of the American Academy of Political and Social Science 635(1): 140–62. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1177/0002716210393640&link_type=DOI) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000289886900009&link_type=ISI) 10. Carlson, Marcia J., and Frank F. Furstenberg Jr.. 2006. “The Prevalence and Correlates of Multipartnered Fertility Among Urban U.S. Parents.” Journal of Marriage and Family 68(3): 718–32. 11. Chung, Yiyoon. 2012. “The Effects of Paternal Imprisonment on Children’s Economic Well-Being.” Social Service Review 86(3): 455–86. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1086/667779&link_type=DOI) 12. Edin, Kathryn. 1995. “Single Mothers and Child Support: The Possibilities and Limits of Child Support Policy.” Children and Youth Services Review 17(1/2): 203–30. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1016/0190-7409(95)00009-2&link_type=DOI) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=A1995QE54100009&link_type=ISI) 13. Freeman, Richard B., and Jane Waldfogel. 1998. “Does Child Support Enforcement Policy Affect Male Labor Supply?” In *Fathers Under Fire: The Revolution in Child Support Enforcement*, edited by Irwin Garfinkel, Sara S. McLanahan, Daniel R. Meyer, and Judith A. Seltzer. New York: Russell Sage Foundation. 14. Garasky, Steven, Susan D. Stewart, Craig Gundersen, , and Brenda J. Lohman. 2010. “Toward a Fuller Understanding of Nonresident Father Involvement: An Examination of Child Support, In-Kind Support, and Visitation.” Population Research and Policy Review 29(3): 363–93. 15. Garfinkel, Irwin. 1992. *Assuring Child Support: An Extension of Social Security.* New York: Russell Sage Foundation. 16. Garfinkel, Irwin, Sara S. McLanahan, and Philip K. Robins, eds. 1992. *Child Support Assurance: Design Issues, Expected Benefits, and Political Barriers as Seen from Wisconsin*. Washington, D.C.: Urban Institute Press. 17. Garrison, Marsha. 1999. “Child Support Policy: Guidelines and Goals.” Family Law Quarterly 33(1): 157–89. [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000081793000006&link_type=ISI) 18. Geller, Amanda, Irwin Garfinkel, , and Bruce Western. 2011. “Paternal Incarceration and Support for Children in Fragile Families.” Demography 48(1): 25–47. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1007/s13524-010-0009-9&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=21318455&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000290961300002&link_type=ISI) 19. Grall, Timothy. 2016. “Custodial Mothers and Fathers and Their Child Support: 2013.” *Current Population Reports* series P60, no. 255. Washington: U.S. Census Bureau. 20. Gunter, Samara R. 2016. “Child Support Wage Withholding and Father-Child Contact: Parental Bargaining and Salience Effects.” *Review of Economics of the Household*. DOI: 10.1007/s11150-016-9330-4. 21. Ha, Yoonsook, Maria Cancian, , and Daniel R. Meyer. 2011. “The Regularity of Child Support and Its Contribution to the Regularity of Income.” Social Service Review 85(3): 401–19. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1086/661923&link_type=DOI) 22. Ha, Yoonsook, Maria Cancian, and Daniel R. Meyer. 2016. “Child Support and Income Inequality.” Working Paper. University of Wisconsin–Madison. 23. Huang, Chien-Chung. 2009. “Mothers’ Reports of Nonresident Fathers’ Involvement with Their Children: Revisiting the Relationship Between Child Support Payment and Visitation.” Family Relations 58(1): 54–64. 24. Koball, Heather, and Desiree Principe. 2002. “Do Nonresident Fathers Who Pay Child Support Visit Their Children More?” *New Federalism: National Survey of America’s Families*, series B, no. B-44. Washington, D.C.: The Urban Institute. 25. Lichter, Daniel T., Sharon Sassler, , and Richard N. Turner. 2014. “Cohabitation, Post-Conception Unions, and the Rise in Nonmarital Fertility.” Social Science Research 47 (September): 134–47. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1016/j.ssresearch.2014.04.002&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=24913950&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) 26. Maldonado, Laurie C., and Rense Nieuwenhuis. 2015. “Family Policies and Single Parent Poverty in 18 OECD Countries, 1978–2008.” Community, Work & Family 18(4): 395–415. 27. Marczak, Mary S., , Emily H. Becher, Alisha M. Hardman, Dylan L. Galos, , and Ebony Ruhland. 2015. “Strengthening the Role of Unmarried Fathers: Findings from the Co-Parent Court Project.” Family Process 54(4): 630–38. 28. McLanahan, Sara, and Gary Sandefur. 1994. *Growing Up with a Single Parent: What Hurts, What Helps.* Cambridge, Mass.: Harvard University Press. 29. Meyer, Daniel R. 1998. “The Effect of Child Support on the Economic Status of Nonresident Fathers.” In *Fathers Under Fire: The Revolution in Child Support Enforcement*, edited by Irwin Garfinkel, Sara S. McLanahan, Daniel R. Meyer, and Judith A. Seltzer. New York: Russell Sage Foundation. 30. Meyer, Daniel R. 2012. “Child Maintenance Policies in the United States.” European Journal of Social Security 14(4): 252–66. 31. Meyer, Daniel R., , Maria Cancian, , and Yiyu Chen. 2015. “Why Are Child Support Orders Becoming Less Likely after Divorce?” Social Service Review 89(2): 301–34. 32. Meyer, Daniel R., , Maria Cancian, , and Steven T. Cook. 2005. “Multiple Partner Fertility: Incidence and Implications for Child Support Policy.” Social Service Review 79 (4): 577–601. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1086/454386&link_type=DOI) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000233577600001&link_type=ISI) 33. Meyer, Daniel R., and Rebecca Y. Kim. 1998. “Estimating a Proposed Family Policy’s Effects: Incorporating Labor Supply Responses to an Assured Child Support Benefit.” Journal of Family Issues 19(5): 534–55. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1177/019251398019005003&link_type=DOI) 34. Meyer, Daniel, R., Christine Skinner, , and Jacqueline Davidson. 2011. “Complex Families and Equality in Child Support Obligations: A Comparative Policy Analysis.” Children and Youth Services Review 33(10): 1804–12. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1016/j.childyouth.2011.05.011&link_type=DOI) 35. Moloney, Lawrie, Lixia Qu, Ruth Weston, , and Kelly Hand. 2013. “Evaluating the Work of Australia’s Family Relationship Centres: Evidence from the First 5 Years.” Family Court Review 51(2): 234–49. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1111/fcre.12023&link_type=DOI) 36. Nepomnyaschy, Lenna. 2007. “Child Support and Father-Child Contact: Testing Reciprocal Pathways.” Demography 44(1): 93–112. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1353/dem.2007.0008&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=17461338&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000245291900005&link_type=ISI) 37. Nepomnyaschy, Lenna, and Irwin Garfinkel. 2010. “Child Support Enforcement and Fathers’ Contributions to Their Nonmarital Children.” Social Service Review 84(3): 341–80. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1086/655392&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=20873018&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) 38. Office of Child Support Enforcement. 2013. *Promoting Child Well-being & Family Self-Sufficiency.* Child Support Fact Sheet Series No. 13. Washington: U.S. Department of Health and Human Services, Administration for Children and Families, Office for Child Support Enforcement. 39. Organization for Economic Co-operation and Development (OECD). 2014. “Social Expenditure Update: Social spending is falling in some countries, but in many others it remains at historically high levels.” Directorate on Employment Labour and Social Affairs. Accessed October 3, 2017. [https://www.oecd.org/els/soc/OECD2014-SocialExpenditure\_Update19Nov_Rev.pdf](https://www.oecd.org/els/soc/OECD2014-SocialExpenditure_Update19Nov_Rev.pdf). 40. Peters, H. Elizabeth, Laura M. Argys, Heather Wynder Howard, , and J. S. Butler. 2004. “Legislating Love: The Effect of Child Support and Welfare Policies on Father-Child Contact.” Review of Economics of the Household 2(3): 255–74. 41. Pirog, Maureen A., and Kathleen M. Ziol-Guest. 2006. “Child Support Enforcement: Programs and Policies, Impacts and Questions.” Journal of Policy Analysis and Management 25(4): 943–90. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1002/pam.20215&link_type=DOI) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000240575600008&link_type=ISI) 42. Schroeder, Daniel. 2016. “The Limited Reach of the Child Support Enforcement System.” Washington, D.C.: American Enterprise Institute. 43. Sinkewicz, Marilyn, and Irwin Garfinkel. 2009. “Unwed Fathers’ Ability to Pay Child Support: New Estimates Accounting for Multiple-Partner Fertility.” Demography 46(2): 247–63. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1353/dem.0.0051&link_type=DOI) [PubMed](http://www.rsfjournal.org/lookup/external-ref?access_num=21305392&link_type=MED&atom=%2Frsfjss%2F4%2F2%2F91.atom) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000268057900002&link_type=ISI) 44. Skinner, Christine. 2012. “Child Maintenance in the United Kingdom.” European Journal of Social Security 14(4): 231–51. 45. Skinner, Christine, Daniel R. Meyer, Kay Cook, , and Michael Fletcher. 2017. “Child Maintenance and Social Security Interactions: The Poverty Reduction Effects in Model Lone Parent Families Across Four Countries.” Journal of Social Policy 46(3): 495–516. 46. Social Security Administration (SSA). 2000. “Title IV: Grants to States for Aid and Services to Needy Families with Children and for Child–Welfare Services.” Washington: U.S. Department of Health and Human Services. Accessed October 4, 2017. [https://www.ssa.gov/OP_Home/ssact/title04/0400.htm](https://www.ssa.gov/OP_Home/ssact/title04/0400.htm). 47. Sorensen, Elaine, and Sandra Clark. 1994. “A Child-Support Assurance Program: How Much Will It Reduce Child Poverty, and at What Cost?” American Economic Review 84(2): 114–19. 48. U.S. Census Bureau. 2015. “America’s Families and Living Arrangements: 2015: Children.” Table 8. Washington: U.S. Department of Commerce. 49. U.S. Department of Health and Human Services (DHHS). 1995. *Report to Congress on Out-of-Wedlock Childbearing*. DHHS Pub. No. (PHS) 95-1257. Hyattsville, Md.: National Center for Health Statistics. 50. Waller, Maureen R., and Robert Plotnick. 2001. “Effective Child Support Policy for Low-Income Families: Evidence from Street Level Research.” Journal of Policy Analysis and Management 20(1): 89–110. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.1002/1520-6688(200124)20:1<89::AID-PAM1005>3.0.CO;2-H&link_type=DOI) [Web of Science](http://www.rsfjournal.org/lookup/external-ref?access_num=000166115600005&link_type=ISI) 51. Wimer, Christopher, Sophie Collyer, , and Sara Kimberlin. 2018. “Assessing the Potential Impacts of Innovative New Policy Proposals on Poverty in the United States.” RSF: The Russell Sage Foundation Journal of the Social Sciences 4(3): 167–83. DOI: 10.7758/RSF.2018.4.3.09. [CrossRef](http://www.rsfjournal.org/lookup/external-ref?access_num=10.7758/RSF.2018.4.3.09.&link_type=DOI)