Foster care as policy — the data, the harm, the alternatives

The pathway in begins with a report. Mandatory reporters — teachers, doctors, nurses, social workers, mental health professionals, daycare staff, sometimes clergy — are required by state law to report suspected abuse or neglect. The threshold for reporting is low ("reasonable suspicion") and the penalties for failure to report are real. The result is a reporting culture that produces approximately four million CPS reports per year, involving roughly seven million children. Roughly half are screened out at intake. Of those screened in, roughly fifteen to twenty percent are substantiated. Of substantiated cases, a minority result in out-of-home placement. The funnel is wide at the top and narrows dramatically; the question is whether the funnel's wide top is itself a harm, given that investigation alone — even without removal — has documented adverse effects on family functioning.

Neglect accounts for roughly sixty percent of all substantiated CPS cases. The legal definition of neglect varies by state but typically includes failure to provide adequate food, clothing, shelter, medical care, supervision, or education. Each of these failures has an economic correlate. Bridges's work on the moral construction of poor motherhood documents how the same parenting decisions are read as resourcefulness in middle-class contexts and neglect in poor contexts. The hot car case, the unsupervised eight-year-old, the empty refrigerator — none of these is, in isolation, a neutral fact. They are interpreted through frames that import class and racial assumptions into the substantiation decision.

Hospitals are the largest single source of CPS reports for infants. Maternal drug screening at delivery, hospital social work assessments, mandatory reporting of positive infant toxicology — these mechanisms route a substantial fraction of poor mothers' postpartum encounters into CPS investigations within days of birth. The implementation is profoundly racially disparate. Studies of urban hospital systems have found that Black mothers are tested at higher rates than white mothers with similar risk profiles, and that positive results in Black mothers are more likely to be reported. Roberts's work documents how the Plans of Safe Care required under the federal CARA legislation have, in practice, become surveillance mechanisms targeting Black and Brown families.

The decision to remove a child is usually made by a CPS worker in the first hours after a report, often without judicial review until forty-eight to seventy-two hours later. The standard for emergency removal is typically "imminent danger" or "substantial risk," but the practical standard is the worker's judgment under organizational pressure to avoid the worst-case scenario. The worker who fails to remove a child who is then harmed faces career and legal consequences; the worker who removes a child who would not have been harmed faces no comparable consequences. The asymmetry of professional risk produces a structural bias toward removal. The first hearing, which technically reviews the removal decision, is short, often conducted without effective representation for the parent, and predominantly confirms the agency's decision.

Quality legal representation for parents at the first hearing and throughout the case is one of the highest-leverage interventions known to the system. The 2009 New York City Office of Family Court Operations parent representation reforms, evaluated by Sankaran and others, produced substantial reductions in time to permanency and reductions in time spent in care, without compromising child safety. The federal recognition that parent legal representation is reimbursable under Title IV-E (effective 2019) was a significant policy change. Implementation depends on state-level appropriations and on the existence of high-quality parent defender offices. Most jurisdictions still rely on overworked appointed counsel whose preparation time per case is measured in minutes.

Placement with relatives — kinship care — accounts for roughly thirty-five percent of all foster placements nationally, with substantial variation by state. Children in kinship care experience fewer placement changes, more sibling co-placement, better behavioral and mental health outcomes, and higher rates of permanency than children in non-relative foster care. The system's historical reluctance to use kinship placement was rooted in concerns about "the apple not falling far from the tree" and in lower licensing standards for relatives that translated into lower financial support. The Family First Act and a series of state reforms have pushed toward kinship preference. The implementation gap — relatives who could care for a child but are not identified, are not financially supported, or are screened out by licensing barriers — remains substantial.

Roughly ten percent of children in foster care live in group homes, residential treatment centers, or other congregate settings. Congregate care is more expensive than family-based foster care and produces worse outcomes on every measured dimension. The Family First Act restricted federal reimbursement for placement in non-family settings beyond two weeks except in narrow circumstances. Implementation has been uneven; states have requested and received extensions, and congregate care populations have declined more slowly than the statute envisioned. The political economy of congregate care — providers with state contracts, employees, capital investments — produces strong institutional resistance to its reduction.

The Adoption and Safe Families Act of 1997 imposed a fifteen-of-twenty-two-months timeline: when a child has been in foster care for fifteen of the previous twenty-two months, the state must file a petition to terminate parental rights unless an exception applies. ASFA's intended effect was to reduce "foster care drift" — children growing up in care without permanency. Its actual effect has been to dramatically increase terminations of parental rights, often against parents whose primary obstacle to reunification was poverty rather than safety. Guggenheim and others have called for ASFA's repeal or substantial reform. The fifteen-month clock, in operation, frequently runs out while a parent is waiting for an appointment with a court-ordered service that was not actually available in their community.

A termination of parental rights is, in legal terms, a civil death — the complete severance of the legal relationship between parent and child. It is, in most states, irrevocable. The child's adoption may or may not follow; in a substantial fraction of cases, the child is freed for adoption but never adopted and ages out of foster care with no legal family of any kind. The "legal orphan" population — children whose parental rights have been terminated but who have no adoptive placement — is large and demographically concentrated in older youth, youth of color, and youth with disabilities. The system has produced a category of children with no parents that it itself created.

Each year, roughly twenty thousand young people age out of foster care without permanency. The Fostering Connections to Success Act allowed states to extend foster care to age twenty-one (and a small number to age twenty-three), and most states have taken up the option. The outcomes for aged-out youth remain poor by every measure. The Midwest Evaluation has tracked outcomes at age seventeen, nineteen, twenty-one, twenty-three, twenty-six, and beyond. The accumulated disadvantage — interrupted education, limited employment history, mental health untreated, no family to fall back on — produces patterns of homelessness, incarceration, early parenthood (often with their own children entering foster care), and chronic instability that persist into mid-adulthood.

The Family First Prevention Services Act, signed in 2018 and implemented progressively from 2019, was the first major federal child welfare statute to allow Title IV-E funds — historically restricted to out-of-home care — to be used for evidence-based prevention services delivered to families before removal. The statute permits funding of mental health services, substance use treatment, and in-home parenting programs for families at imminent risk of placement. Implementation has been state-by-state and uneven. The shift is meaningful and incomplete; the federal financial incentives still tilt substantially toward placement.

The Indian Child Welfare Act of 1978 established placement preferences for Native children: first with extended family, then with members of the child's tribe, then with other Native families. ICWA was passed in response to documented patterns of removal of Native children from their families and tribes at rates that constituted, in the assessment of congressional findings, cultural destruction. ICWA has been under constant litigation since enactment. The 2023 Haaland v. Brackeen decision rejected facial challenges to the statute on equal protection and anti-commandeering grounds, preserving the statute but leaving room for as-applied challenges. The political and legal coalition seeking to weaken ICWA — driven in part by white prospective adoptive parents and adoption agencies — has not gone away.

Roberts's Torn Apart (2022) makes the case that the family-policing system cannot be reformed and should be abolished, replaced by a system of direct material support to families and a much narrower, voluntary, community-based response to actual abuse. The argument has gained traction in the academic literature and in some advocacy circles. Its operational implementation has been pursued through specific reforms: narrowing mandatory reporting, eliminating anonymous reporting, raising the threshold for substantiation, mandating provision of concrete material support before removal, ending the use of CPS investigation as the response to poverty-driven conditions. Whether these reforms constitute abolition by another name or an alternative to abolition is a definitional dispute. The underlying claim — that the existing system causes more harm than it prevents — is supported by an increasing body of evidence and is no longer a minority view in the research literature.

1. Roberts, Dorothy. Shattered Bonds: The Color of Child Welfare. New York: Basic Civitas Books, 2002.

2. Roberts, Dorothy. Torn Apart: How the Child Welfare System Destroys Black Families—and How Abolition Can Build a Safer World. New York: Basic Books, 2022.

3. Guggenheim, Martin. What's Wrong with Children's Rights. Cambridge, MA: Harvard University Press, 2005.

4. Bridges, Khiara M. The Poverty of Privacy Rights. Stanford: Stanford University Press, 2017.

5. Sankaran, Vivek S., and Christopher Church. "Easy Come, Easy Go: The Plight of Children Who Spend Less Than Thirty Days in Foster Care." University of Pennsylvania Journal of Law and Social Change 19 (2016): 207-231.

6. Wildeman, Christopher, and Natalia Emanuel. "Cumulative Risks of Foster Care Placement by Age 18 for U.S. Children, 2000-2011." PLoS ONE 9, no. 3 (2014): e92785.

7. Wildeman, Christopher, et al. "The Cumulative Prevalence of Confirmed Maltreatment Among US Children." American Journal of Public Health 104, no. 2 (2014): 274-280.

8. Courtney, Mark E., et al. Midwest Evaluation of the Adult Functioning of Former Foster Youth: Outcomes at Age 26. Chicago: Chapin Hall at the University of Chicago, 2011.

9. Pecora, Peter J., et al. Improving Family Foster Care: Findings from the Northwest Foster Care Alumni Study. Seattle: Casey Family Programs, 2005.

10. Krinsky, Miriam A., and Jamie Yoon. "Aging Out of Foster Care: The Challenges of Transitioning to Adulthood." Family Court Review 47, no. 2 (2009): 297-309.

11. Pelton, Leroy H. "The Continuing Role of Material Factors in Child Maltreatment and Placement." Child Abuse and Neglect 41 (2015): 30-39.

12. Edwards, Frank, et al. "Family Surveillance: Mandated Reporting, Foster Care, and the Policing of Poor Mothers." Annual Review of Law and Social Science 19 (2023): 1-22.