The recent commentary on poverty and neglect by Will Schneider — cosigned by other academics — calls for the field to try “something new in child welfare policy”: Expand the social safety net and cash assistance.
Given the widespread claim that child neglect is nothing more than poverty, such a proposal should be easy for Schneider to argue. And yet he makes a wholly unconvincing case — eliding specifics and ignoring tradeoffs. In this brief rejoinder, we address the most obvious issues.
First, Schneider’s commentary contains no specifics as to the scope of cash assistance and other support that would be appropriate to mitigate poverty and reduce child neglect. Still, he claims these as-yet-undefined investments are “likely to be cost effective” because “the federal government spends nearly $8 billion per year on out of home care for children (e.g., foster care, adoption and guardianship).”
Note the carefully qualified statement that additional programs to address poverty are merely “likely to be cost effective.” Schneider knows it is impossible to evaluate whether expanded services would be cost-effective without offering a price tag or quantifying the expected benefits. Although $8 billion — or even the $31 billion in total federal, state, and local expenditures by child welfare agencies — may seem like a lot, it pales in comparison to cumulative federal expenditures through the current social safety net.
Consider the budgetary scope of some major social safety net programs that serve low-income children and parents:
Supplemental Nutrition Assistance Program ($112 billion)
Earned Income Tax Credit ($64 billion)
Special Supplemental Nutrition Program for Women, Infants, and Children ($7 billion)
School meal programs ($28 billion)
Supplemental Security Income ($65 billion)
Housing assistance ($62 billion).
Portions of Temporary Assistance for Needy Families, Medicaid, and numerous other programs also support low-income families with cash or in-kind support.
Still, Schneider argues that additional cash assistance is needed to enable parents to “change their circumstances” and seems to believe that “caseworkers don’t have anti-poverty services to offer.” But how much financial support is enough before parents are considered to have agency over their choices and responsibility for their children’s safety? Schneider and his signatories do not say.
Absent such details, it’s illustrative to examine one of the many recent proposals to expand cash assistance: making the child tax credit fully refundable (i.e., available to families with no or limited earnings). The estimated price tag for modifying the existing credit to be fully refundable up to $2,000 per year, per child, is $21 billion annually. To what extent would those costs be offset by decreased involvement in the child protection system? The most generous estimate of what such an expansion would achieve is a 15% reduction in investigations. And we know the small subset of investigated children who enter foster care (especially those who ultimately exit to adoption or guardianship) come from families with much more severe and complex issues. For example, an estimated 75% of children entering foster care come from families with substance use problems, who also tend to have very low rates of voluntary engagement in treatment.
To be clear — an expansion of the child tax credit, or any other part of the safety net, may be cost-effective or good policy on other grounds. We take no position on that question. But the costs of these policies would not be meaningfully offset by reductions in out-of-home care or other child protection activities. The reason is obvious: Child protection and out-of-home care are essential safeguards regardless of the size or scope of the safety net. Just look at countries like Sweden, Denmark or Norway — which have both expansive safety nets and sizable out-of-home care populations.
Second, even modest effects on CPS involvement or foster care entry are hardly certain. Schneider asserts that research “tells us that social programs help to protect against child maltreatment” and that it is “increasingly clear that poverty is causally linked to child neglect” because “this relation cannot be explained by any other factors.” Such confident dismissal of alternative explanations is striking given the current evidence base. Schneider invokes smoking and lung cancer as a parallel to poverty and neglect. But surely he knows that smoking’s causal role in cancer is supported by decades of mechanistic research, epidemiological consistency, dose–response evidence, and randomized trials of cessation programs.
In reality, recent randomized controlled trials raise serious questions about the benefits of unconditional cash transfers for low-income families. For example, one such trial found that guaranteed income did not increase total income for low-income individuals, because recipients simply worked less or accrued more debt. Another recent randomized controlled trial found — even after excluding high-risk families typically observed in the CPS-involved population — that unconditional cash transfers had no positive effects on maternal well-being or children’s development.
And importantly, many randomized controlled trials of unconditional cash payments have excluded individuals with serious substance use disorders or related conditions associated with substance use, like child custody loss or infant neonatal intensive care unit stays. Perhaps — contrary to Schneider’s assertion that there is “little reason” to think CPS-involved families will spend cash assistance differently than other disadvantaged families — this is because they know that cash transfers to individuals experiencing drug addiction can be dangerous.
Rather than acknowledge these nuances, Schneider contends that any concerns about unconditional cash transfers in the context of child maltreatment is founded in an “old stereotype” and “myth of the welfare queen”— a not-so-subtle way of silencing concerns among those who don’t want to be called bigots.
Schneider’s argument rests largely on quasi-experimental studies that collectively feature notable limitations — most prominently, their reliance on aggregate state-level rather than child- or family-level data. Moreover, many studies have asserted positive effects of employment-conditioned transfers (e.g., the Earned Income Tax Credit) — even if true, those findings may not generalize to the 42% of custodial CPS-reported parents who don’t work. As evidence that cash assistance works in the CPS context, Schneider references an easily debunked claim that Indiana’s Family Preservation Program — which provided a small amount of funds to approximately 9% of participants — drastically reduced racial disparities in foster care placement.
Last, Schneider and the researchers endorsing his commentary acknowledge the prevalence of substance use, untreated mental illness, and domestic violence among families involved with CPS, but they wave it away by noting that poverty is also present. No one disputes that CPS-involved families are predominantly low-income.
But they further acknowledge that the factors contributing to poverty are complex and that cash assistance is no panacea for child neglect. Given these acknowledgements, it is unclear why they have such confidence their proposed solution is good policy.
Although we have focused our response on Schneider’s argument for cash assistance as a cost-effective means to stabilize families and reduce out-of-home care, we would be remiss not to re-center children. The consequences of neglect, especially when chronic, can be devastating, inflicting lifelong harm and in some cases, resulting in death. We should absolutely experiment with different strategies to prevent or mitigate parenting conditions that give rise to all forms of child maltreatment. But advancing sound policy requires rigor and humility. Overstating what we know — and obscuring what we don’t — serves no one.
