The Paper-Ready Pipeline
How Federal Incentives, Fast-Track Terminations, and Failed Adoptions Created America’s Hidden Class of Children Without Families
How America Creates Legal Orphans, Pays the System Anyway, and Calls It Permanency
There is a phrase buried deep inside America’s child welfare bureaucracy that sounds compassionate, procedural, and even hopeful until you stop and examine what it actually authorizes the state to do to a child. A child can be described as being “freed for adoption,” a phrase so sanitized and carefully engineered that most Americans never pause long enough to ask what happened before that point or what happens afterward. The wording suggests liberation from danger, movement toward stability, and the clearing away of legal obstacles so a child can finally reach safety and permanence. But hidden underneath that language is one of the most consequential and least publicly understood realities inside the American foster care system: a child can be permanently stripped of every legal tie to their parents before any new permanent family has been identified, approved, or guaranteed. In many cases, the state does not actually know where the child will end up when the judge signs the termination order. The child may eventually be adopted, may spend years moving through foster placements, may cycle through group homes, may experience failed adoptive placements, or may age out entirely without permanency. Yet the legal severance can happen first, long before the future the system promises has materialized.
To understand how extraordinary this is, imagine any other area of American life where the government was allowed to destroy an existing legal relationship based primarily on the possibility that a replacement relationship might eventually appear later. Imagine a hospital legally dissolving a patient’s access to current medical care before securing a new provider because it intended to “concurrently recruit” one. Imagine a court terminating housing rights before confirming alternative housing existed. The public would regard that as reckless. Yet in child welfare, the destruction of the child’s original legal family is routinely justified by the anticipated possibility of a future adoptive family that may not yet exist. The child becomes what advocates and researchers have increasingly described as a “legal orphan,” meaning a child whose parental rights have been terminated without a finalized adoptive placement replacing them. This category is not symbolic. It is a real legal condition created by state action.
The tragedy is that the public rarely sees this reality because the system’s language has evolved to obscure it. The bureaucracy speaks in terms of “permanency,” “timelines,” “best interests,” and “concurrent planning,” terms that sound reassuring unless you examine how they function operationally. Once translated from bureaucratic language into plain English, the machinery becomes much more disturbing. The state can remove a child from a family, place the child into foster care, begin a federally monitored clock, file a petition to terminate parental rights once enough time has passed, and legally erase the child’s relationship to their parents before an adoptive family has been secured. At that point the child becomes administratively available for adoption, but availability is not the same thing as permanency. The courtroom can generate a legal orphan in a single afternoon. Building an actual family for that child can take years or may never happen at all.
This is not a fringe occurrence caused by isolated mistakes or unusually negligent jurisdictions. The data compiled in the Project Milk Carton research brief paints a picture of a national structural pattern. In fiscal year 2022 alone, according to the AFCARS-based data cited in the report, states filed 64,561 terminations of parental rights nationally while finalizing only 53,665 adoptions. That created an annual national gap of 10,896 children, meaning nearly eleven thousand more children entered legal severance than reached finalized adoption during the same reporting period. Texas recorded a gap of 2,837. California recorded a gap of 2,316. Ohio’s ratio was especially alarming, with more than two terminations filed for every adoption finalized. These are not small discrepancies buried in administrative noise. They are evidence that the system is legally severing children from families faster than it is creating new permanent families for them.
What makes the issue even more unsettling is that the federal government has effectively normalized this imbalance rather than treating it as a warning sign. The central architecture comes from the Adoption and Safe Families Act of 1997, legislation passed during a period when foster care “drift” had become a major political concern. Policymakers argued that children were remaining in foster care too long while agencies endlessly attempted reunification with parents who were unlikely to regain custody. ASFA’s solution was to accelerate movement toward permanency by imposing stronger timelines for termination of parental rights. The political framing sounded humane: children should not spend their childhoods trapped in temporary care. But while ASFA successfully accelerated legal severance, the system never built an equally effective mechanism for guaranteeing actual adoptive permanency on the back end. The result was a system that became dramatically better at legally dissolving families than at consistently replacing them.
The consequence is a pipeline that can convert family instability into permanent legal severance with remarkable efficiency while leaving the child’s long-term future uncertain. The public still imagines foster care as a temporary bridge toward safety and adoption as the predictable final destination. The numbers suggest something more troubling. Thousands of children each year are entering a legal state where their original family has been erased while the promised replacement family remains hypothetical. That is the reality hidden underneath the phrase “freed for adoption.” A child can be “freed for adoption.” On paper, that phrase suggests rescue, movement, progress, and the clearing of legal obstacles so a child can finally belong somewhere. In practice, it can mean something far darker. It can mean the state has permanently severed a child from every legal tie to their birth parents before any adoptive family has been found, before any new permanent home exists, and before anyone can honestly say the child is moving into anything resembling stability. The child is not adopted. The child is not reunified. The child is not permanently placed. The child has simply been made available, converted by court order from someone’s son or daughter into a paper-ready adoption candidate inside a system that often mistakes legal severance for actual safety.
That distinction matters because the entire public story around child welfare depends on the public not seeing it. The average person hears “termination of parental rights” and assumes the state must have reached the end of a long road, exhausted every possible service, ruled out every safe relative, identified a better permanent family, and then acted only because no other option remained. That is the moral framing people are given. It is the version that makes the machine tolerable. But the research record tells a more disturbing story: federal law allows states to terminate parental rights without having an adoptive family already identified, and the federal government has treated judicial reluctance to create legal orphans not as a warning sign, but as a barrier to compliance. In other words, when some judges hesitate to destroy a child’s only legal family because no new family is waiting, the bureaucracy does not necessarily call that prudence. It can call it delay.
This is the central scandal of the paper-ready pipeline. The system is not merely failing after termination of parental rights; in many cases, the failure is baked into the design. The Adoption and Safe Families Act, or ASFA, pressures states to move toward termination when a child has been in foster care for 15 of the most recent 22 months, subject to exceptions that are available in theory but often underused in practice. The statute tells states to file a petition to terminate parental rights and, concurrently, to identify, recruit, process, and approve a qualified adoptive family. That word “concurrently” carries enormous consequences. It does not require a family to already exist. It does not require a child to have a specific adoptive placement ready. It does not require the court to pause and ask whether severing the child from all legal family will produce an actual permanent home. It allows the state to begin the search for a new family at the same time it seeks to legally destroy the old one.
The implementing regulation makes the problem even sharper because it says agencies must concurrently begin to identify, recruit, process, and approve an adoptive family. The protection is not that the child must have a family. The protection is that the agency must begin looking for one. That is the legal hinge on which thousands of children can be swung from family crisis into legal orphanhood. “Begin” is a bureaucratic word, but inside this system it functions like a trapdoor. Once the state can show it started recruitment, it can pursue the final legal severance even if the child’s future is uncertain, even if the child is older, disabled, traumatized, part of a sibling group, or otherwise statistically less likely to be adopted quickly. The law does not require the state to prove that adoption is realistically imminent before it annihilates the parent-child legal bond. It requires process, not certainty; motion, not outcome.
The numbers in the research brief make the abstract legal architecture impossible to dismiss. In fiscal year 2022, the CivicOps AFCARS-based data cited in the brief shows 64,561 terminations of parental rights filed nationally and 53,665 adoptions finalized, producing an annual TPR-to-adoption gap of 10,896 children. That gap is not an academic artifact. It represents the difference between the number of children being pushed through legal severance and the number actually reaching finalized adoption in the same reporting year. Nationally, 108,877 children were waiting for adoption in FY2022. Texas alone had 10,232 children waiting, 7,198 TPRs filed, and only 4,361 adoptions finalized, leaving a gap of 2,837. California had 8,180 TPRs filed and 5,864 adoptions finalized, leaving a gap of 2,316. Ohio had 3,461 TPRs filed and only 1,543 adoptions finalized, a ratio of 2.24 terminations for every adoption finalized. Florida, Michigan, and Kentucky also showed large annual gaps. The pattern is not isolated. It is structural.
A system designed around actual permanency would treat those gaps as emergency indicators. It would ask why legal severance is outpacing family creation. It would ask whether states are filing TPR petitions faster than they can realistically secure adoptive homes. It would ask whether some children are being converted into legal orphans because the law rewards movement through the procedural pipeline more than it rewards durable family outcomes. But the official vocabulary works to anesthetize the public. Children do not become legal orphans in the language of the system; they become “waiting children.” They are not abandoned by a failed permanency apparatus; they are “freed for adoption.” Their legal connections are not destroyed without replacement; the agency is engaged in “concurrent planning.” This is how a moral catastrophe is made administratively manageable.
The term legal orphan needs to be understood with precision. A legal orphan is not simply a child in foster care. A child in foster care may still have parents whose rights remain intact, relatives who can petition for custody, and a reunification plan that remains legally alive. A legal orphan is different. This is a child whose parents’ rights have been terminated but who has not been adopted. The state has erased the legal parent-child relationship, but no new parent-child relationship has replaced it. The child may remain in foster care, move through placements, live in group settings, age out, or eventually be adopted, but during that interval the child exists in a legal void. The state has removed the old family from the law and has not yet created a new one. In human terms, the child has lost the only legal family they had and has gained a case plan.
The reason this matters is that termination of parental rights is often described as if it is synonymous with permanency. It is not. Termination is only a legal precondition that can make adoption possible. It does not itself produce an adoptive family, does not stabilize a placement, does not guarantee attachment, does not prevent disruption, and does not make a child less traumatized. A court order can erase a parent from the legal record, but it cannot manufacture a bonded adult willing and able to raise the child to adulthood. When agencies or policymakers treat TPR as progress toward permanency without honestly measuring whether permanency is achieved, they create a false success metric. The file has moved. The child may not have.
The federal government has known for years that some judges see the danger. The research brief cites a 2021 report from the HHS Office of the Assistant Secretary for Planning and Evaluation, “Freeing Children for Adoption within the ASFA Timeline,” which documented that in seven states, judges were reluctant to grant TPR until an adoptive home was identified specifically because they wanted to avoid creating legal orphans. That finding should have triggered a national policy reckoning. Judges on the ground were effectively saying that the legal machine was asking them to sever children too early, before the promised adoptive future was real. Yet HHS framed this judicial reluctance as a barrier to compliance with the ASFA timeline. That framing tells us something essential about the system’s priorities. When judicial caution protects children from legal orphanhood, the bureaucracy can interpret that caution as noncompliance.
This is not a small technical disagreement. It is a conflict between child-centered judgment and timeline-centered administration. A judge looking at a child may ask whether severing all legal family ties is truly in that child’s best interest if no adoptive home has been found. A federal compliance monitor may ask whether the petition was filed within the required time frame. Those are not the same question. One is moral and factual. The other is procedural. The danger is that the procedural question has been elevated until it overwhelms the moral one.
The ASFA exceptions are supposed to provide safety valves, but the research brief makes clear that they are limited and often underused. States may avoid filing TPR when a child is being cared for by a relative, when the case plan documents a compelling reason that TPR would not be in the child’s best interest, or when the state has failed to provide required reunification services. On paper, the “compelling reason” exception could be used when no adoptive family exists and termination would create a legal orphan. In practice, the system’s compliance culture can push the other way. If the federal government views reluctance to terminate without an adoptive placement as a barrier, agencies have a built-in incentive to avoid using the exception too broadly. The exception exists, but the pressure is toward filing.
That pressure is intensified by the 15-of-22-month trigger. Under ASFA, when a child has been in foster care for 15 of the most recent 22 months, the state is generally required to file a TPR petition unless an exception applies. The months are cumulative rather than necessarily consecutive, and the clock can become the central fact in the case. Time in foster care becomes evidence of urgency, even when the reasons for delay may include agency failures, lack of services, housing instability, disability accommodations, delayed treatment access, or court scheduling. The law was meant to prevent children from drifting indefinitely, but in practice it can punish families for delays the state helped create. The clock does not care whether the parent had a meaningful chance to reunify. The clock matures anyway.
Some states go further than the federal baseline. Ohio uses a 12-consecutive-month threshold rather than simply following the federal 15-of-22-month framework, making its standard more aggressive than the federal floor. The brief notes that Ohio’s FY2022 numbers were among the nation’s worst, with 3,461 TPRs filed and only 1,543 adoptions finalized. That means Ohio was terminating parental rights at more than double the pace of finalized adoption in that reporting year. If the state’s stated objective is permanency, such a ratio should be treated as a flashing red warning. Instead, it sits inside the national pattern of legal severance outpacing actual family formation.
Texas reveals a different but equally troubling mechanism. The research brief cites the Texas DFPS Handbook, stating that in most counties the agency requests termination of parental rights as an alternative in the original petition filed at the time of removal. That means the legal death penalty of family law can be placed into the case from day one, before the federal 15-of-22-month trigger has even had time to develop and before the family has had a realistic opportunity to demonstrate whether reunification is possible. This is not merely procedural housekeeping. When termination is pleaded at removal, it changes the posture of the case. The family enters the system under immediate threat of permanent severance. The state can say reunification remains the goal, but the petition already contains the machinery to end the family.
This matters because legal threats shape behavior. Parents under the pressure of possible TPR may be navigating poverty, unstable housing, addiction recovery, domestic violence, mental illness, disability, transportation barriers, court appearances, service plans, supervised visitation, and caseworker turnover all at once. The system often describes these parents through the language of compliance, as if each missed appointment or incomplete class reflects moral indifference rather than systemic overload. Meanwhile, the state may already be building an alternative permanency track. Once a child is in foster care long enough, the agency can point to time itself as justification for escalation. In this way, the system can transform its own delays into evidence against the parent.
The constitutional stakes are enormous. The Supreme Court has recognized that parental rights are fundamental liberty interests protected by the Fourteenth Amendment, and in Santosky v. Kramer the Court held that termination of parental rights requires clear and convincing evidence. That standard exists because TPR is one of the most severe civil actions the state can take. It is often described as the family-law equivalent of the death penalty because it permanently destroys the legal relationship between parent and child. Yet no appellate court has squarely held that TPR without an identified adoptive placement is unconstitutional. That absence of precedent should not be confused with moral clarity. It may simply mean that the right case has not been litigated in the right way with the right record.
The constitutional argument is not that the state can never terminate rights without an adoptive family. There are cases involving severe abuse, chronic danger, abandonment, or aggravated circumstances where termination may be necessary even if adoption is uncertain. The real question is whether the state may routinely sever a child’s legal family in ordinary dependency cases without proving that the severance is the least restrictive means of achieving actual child safety and permanency. If the state’s justification is permanency, but the child remains in foster care after TPR, then the state’s logic deserves scrutiny. A court should ask whether guardianship, kinship custody, subsidized relative placement, permanent legal custody, or continued reunification services could protect the child without destroying every legal tie to the birth family.
This is where the least-restrictive-alternative framework becomes critical. If parental rights are fundamental, then the state should not be allowed to use the most destructive legal tool merely because it is administratively cleaner. The government should have to prove why lesser interventions are inadequate. It should have to prove that termination is necessary not in the abstract, but for this child, in this case, at this time. It should also have to confront the consequences of legal orphanhood. If no adoptive family has been identified, and if the child is likely to remain in foster care, the state should not be permitted to wave the word “adoption” like a talisman while producing only uncertainty.
The existence of parental rights reinstatement statutes in several states quietly confirms the problem. California, Illinois, Maine, Nevada, Oklahoma, Oregon, and Washington have created pathways, under limited circumstances, for parental rights to be restored when a child has not been adopted after a set period, often around three years after TPR. These laws are not radical family-preservation manifestos. They are state-level admissions that termination sometimes fails to produce permanency. They recognize that a child can be legally severed, left unadopted, and harmed by the permanency system’s own failure. But these statutes are narrow, often difficult to use, and frequently arrive too late. By the time reinstatement becomes possible, the parent may have been destabilized by years of separation, the child may have suffered multiple placements, and the relationship may have been starved by court order.
A system that needs reinstatement laws should also need stronger pre-termination safeguards. It is not enough to say that, years later, a child might be able to return to a parent if adoption never happened. The better question is why the state was allowed to terminate without a viable permanent alternative in the first place. Reinstatement is a repair mechanism. It is not a substitute for preventing unnecessary legal orphanhood.
The financial structure deepens the concern because the child welfare system does not operate in a vacuum of pure benevolence. Federal money flows through foster care, adoption assistance, and adoption incentive programs, and every funding stream creates behavior. The federal adoption incentive program under 42 U.S.C. § 473A pays states based on adoption performance above baseline measures. Since FY1998, according to the research brief, states have received more than $1.04 billion in cumulative adoption incentive bonuses nationally. The stated purpose is to encourage states to move children into adoptive homes. But the incentive is tied to finalized adoptions, not to lifelong permanency, not to the absence of later dissolution, and not to whether the original TPR should have happened.
That distinction is not a technicality. It is the heart of the financial ratchet. If an adoption finalizes, the state can receive credit. If the adoption later fails, the original finalization does not disappear from the incentive history. The payment is not clawed back simply because the child re-enters foster care years later. If the child is later adopted again, that re-adoption can count as a new adoption in the later year and may generate another bonus if the state exceeds its baseline. A fresh Title IV-E adoption assistance agreement may also be executed with the new family. If the child is older by the time of re-adoption, the state may qualify for a higher older-child bonus. The result is a structure where the system can be paid for an adoption that later fails and may be paid again if the same child is re-adopted.
No serious safety industry would tolerate that design without aggressive tracking and accountability. If a bridge contractor were paid for completing a bridge, the bridge collapsed, and the contractor was paid again to rebuild it, the public would demand investigations, clawbacks, engineering audits, and criminal referrals if fraud were involved. In child welfare, the equivalent collapse is often hidden behind terms like disruption, dissolution, re-entry, and re-adoption. The child’s life breaks apart, but the payment history can remain intact. The machine records the finalization. The failure becomes another case episode.
The research brief distinguishes between disruption and dissolution, and the distinction matters. Disruption occurs when an adoptive placement fails before the adoption is legally finalized, causing the child to re-enter foster care or move to another placement. Dissolution occurs after finalization, when the legal adoptive relationship breaks down and the child re-enters the system. National disruption estimates range from 10 to 25 percent. Post-finalization dissolution estimates are usually cited at 1 to 5 percent nationally, but state-level investigations suggest the number can be substantially higher in particular contexts. The Colorado Sun investigation cited in the brief found that 1,094 children re-entered Colorado’s system after failed adoptions over a decade, amounting to roughly 13 percent, with an average of 8.5 years before re-entry.
The Colorado data is especially important because it reveals a tracking problem as much as a failure problem. According to the brief, Colorado updated its tracking system in 2020 using persistent child ID numbers, implying that prior systems may have undercounted dissolution because they could not reliably follow the same child across episodes. That is a devastating admission if applied nationally. If states cannot consistently track children from original removal through TPR, adoption, dissolution, re-entry, and possible re-adoption, then policymakers do not actually know how durable their claimed permanency outcomes are. They know how many adoptions finalized. They may not know how many stayed intact.
The public is sold a clean adoption number. The child lives the messy longitudinal reality. A finalization ceremony can become a press release, a court photo, a success story, and a federal incentive event. But if the adoption dissolves eight years later, the child may return to state custody as a teenager with deeper trauma, fewer placement options, and a greatly reduced chance of adoption. The original success story remains on the books while the child becomes evidence of a failure few agencies are eager to publicize.
This is why the incentive program needs to be judged not by how many adoptions it produces, but by how many safe, durable, supported families it sustains. Adoption is not a transaction. It is a lifelong legal and emotional relationship. A system that rewards the transaction while weakly tracking the durability has created a market signal that is too shallow for the stakes involved. The state should not be rewarded merely for moving a child across the legal finish line if it does not also remain accountable when the placement collapses.
The research brief identifies another hidden pipeline feeding children into state custody: therapeutic relinquishment. This is one of the most morally damning parts of the entire system because it exposes how child welfare can become the dumping ground for failures in health care, disability services, and mental health treatment. Therapeutic relinquishment occurs when parents surrender custody or place children into child welfare or juvenile justice systems not because they abused or abandoned them, but because the family cannot access necessary mental health services while retaining custody. In plain English, some parents have had to give up their children to get them treatment.
The Government Accountability Office documented this problem more than two decades ago. In GAO-03-397, issued in 2003, child welfare directors in 19 states estimated that in fiscal year 2001, parents placed more than 12,700 children into child welfare or juvenile justice systems solely to obtain mental health services. That number was not a comprehensive national count. It was a floor. Fewer than half of states could provide data, and no federal system tracked the placements comprehensively. The children were disproportionately male and adolescent, and they entered state systems because families needed residential behavioral health services they could not otherwise access.
The cruelty of this arrangement is difficult to overstate. A parent may be doing everything society claims to want: seeking treatment, trying to keep a child safe, acknowledging that the child needs intensive care, and asking for help before crisis turns catastrophic. But if the service system says the only path to treatment runs through custody transfer, the parent faces an impossible choice. Keep custody and be denied the level of care the child needs, or surrender custody and risk losing the child to the child welfare machine permanently. That is not child protection. That is policy extortion dressed as service delivery.
Once a child enters state custody through this kind of pathway, the same permanency timelines and incentives can begin to operate. The case may not have started as abuse. It may have started as a desperate search for psychiatric care. But the system can still place the child in foster care, start clocks, pressure permanency decisions, and eventually move toward TPR if the family cannot satisfy the plan or if returning home is deemed impractical. In this way, failures in mental health infrastructure can be converted into child welfare cases, and child welfare cases can be converted into adoption-eligible files. The original sin is not parental abandonment. It is the state’s refusal to make treatment available without custody destruction.
Nebraska proved how widespread the pressure was in 2008 when it passed a safe haven law with no age limit. The law allowed children up to age 18 to be surrendered at hospitals. In just 127 days, 36 children were surrendered. They were not newborns. They were older children, including children from other states. One father from Iowa surrendered nine of his children at once. The political system reacted with alarm and quickly amended the law to apply only to newborns. But the brief Nebraska episode revealed something the country should not have ignored: when a legal door opened, desperate families walked through it. They were not exploiting a loophole for convenience. They were revealing a buried crisis.
The Nebraska safe haven crisis showed that therapeutic relinquishment was not an exotic anomaly limited to a few irresponsible parents. It was a rational response to an irrational system. Families with older children in severe behavioral health crisis needed help. They could not get it. When a state accidentally created an exit ramp, they drove to it. The emergency was not the surrender of children in Nebraska. The emergency was the national service failure that made surrender appear to be the only option.
As of 2025, according to the research brief, two-thirds of states still did not systematically collect data on voluntary custody relinquishments. A 2024 HHS rule may classify custody relinquishment for mental health access as disability discrimination, but implementation remains pending. A 2023 study documented children admitted to psychiatric hospitals and then transferred directly into state custody as a pathway to services. No updated national prevalence figure exists, leaving the 2001 GAO estimate of 12,700 children in 19 states as one of the most cited benchmarks more than twenty years later. That is not just a data gap. It is a public accountability failure.
When a system does not count something, it can avoid explaining it. If the federal government does not maintain current national data on custody relinquishment for mental health services, then Congress cannot easily hold hearings on the scale of the problem, journalists cannot easily map the hotspots, and families cannot prove they are part of a systemic pattern rather than isolated tragedies. The absence of data protects the institution, not the child.
The ultimate human consequences of legal orphanhood appear in the aging-out data. The research brief relies heavily on the Chapin Hall Midwest Evaluation, a landmark longitudinal study following former foster youth from ages 17 and 18 into young adulthood. By age 26, 36 percent of former foster youth in the study had experienced homelessness. Among males, 74 percent had been incarcerated by age 26, compared with 23 percent in the general population. Among females, 43 percent had been incarcerated, compared with roughly 6 percent in the general population. Pregnancy rates among young women by age 21 were dramatically higher than the general population. College completion was sharply lower. Employment was unstable. These numbers describe the afterlife of failed permanency.
It is tempting for policymakers to talk about aging out as if it is a separate issue from TPR policy, adoption incentives, and legal orphanhood. It is not separate. A child who is legally severed from parents but never adopted is at elevated risk of remaining in care until adulthood. A child whose adoption dissolves may re-enter care older, more traumatized, and harder to place. A child surrendered for mental health treatment may become entangled in systems never designed to preserve family bonds. These pathways converge at the same cliff: young adulthood without a permanent family.
The cost estimates are staggering, but they are still inadequate because they cannot capture the full moral damage. The brief cites estimates of $300,000 to $343,000 in social cost per youth who ages out, including costs associated with homelessness, incarceration, public benefits, lost earnings, and related consequences. With roughly 23,000 youth aging out nationally each year, the estimated annual social cost reaches $6.9 billion to $7.9 billion. But even that large number understates the issue because the deepest costs are not merely fiscal. They include the absence of a parent at a hospital bedside, the lack of a co-signer for an apartment, the missing emergency contact, the unpaid emotional debt of birthdays and holidays spent without permanent belonging, and the lifelong instability that follows a child who was processed but not parented.
This is where the system’s moral language becomes almost unbearable. Agencies speak of permanency while children age out. They speak of best interests while severing legal ties without replacement families. They speak of adoption success while failed adoptions send children back into custody years later. They speak of family preservation while parents surrender children for mental health services, they should have been able to access without losing custody. The words are warm. The machinery is cold.
The inverse relationship described in the research brief should haunt anyone who still believes the incentive system is functioning as designed. Adoption incentive payments have grown over time, totaling more than $1.04 billion since 1998, yet national adoptions fell to 50,193 in FY2023, described in the brief as a 20-year low. Meanwhile, the TPR-to-adoption gap remained substantial, with 10,896 more TPRs than adoptions in FY2022 and an estimated larger gap in FY2023. If incentives were truly aligned with the creation of permanent families, rising incentive investment should not coincide with declining adoption completions and persistent legal orphan production. Something is wrong in the design.
One former Los Angeles County child welfare official quoted in the research brief captured the structural problem with unusual bluntness: “What you have now is an incentive to initially remove the child, and an incentive to adopt them out. I think when you put those two together, there is a problem.” That sentence deserves to be read slowly. The problem is not that individual caseworkers wake up plotting to harm children. The problem is that systems follow incentives. If funding streams reward foster care entry, procedural movement, and adoption finalization more clearly than they reward preventing removal, stabilizing families, supporting relatives, or sustaining adoptive placements over time, then the system will bend toward what is funded and measured.
This does not mean every removal is illegitimate or every adoption is suspect. Many children are in real danger. Many birth parents cannot safely care for children. Many adoptive families are heroic. Many caseworkers are overburdened people trying to do impossible work inside broken institutions. But acknowledging those truths does not excuse a financial architecture that can reward throughput while failing to guarantee durable permanency. The issue is not whether child welfare sometimes saves lives. It does. The issue is whether the system also creates avoidable legal orphans, hides failed permanency, and converts service failures into family destruction.
The public debate often collapses into false binaries: child safety versus parental rights, adoption versus foster care, permanency versus drift. Those binaries protect the system because they make any criticism sound like indifference to child abuse. But the real question is not whether children should be protected. The real question is whether protection requires the current machinery. A child can be protected without unnecessary legal severance. A parent can be held accountable while still receiving meaningful services. A relative can be supported instead of bypassed. A guardianship can provide stability without erasing identity. A mental health crisis can be treated without forcing custody surrender. A state can pursue adoption without being rewarded for failed adoptions that later dissolve.
The child welfare system should be forced to answer a set of questions it has avoided for too long. How many children each year become legal orphans after TPR with no adoptive placement identified? How long do they remain in that status? How many age out after TPR without adoption? How many adoptions used for incentive calculations later disrupt or dissolve? How many children generate more than one adoption-related incentive event after re-entry and re-adoption? How many custody relinquishments occur solely because families cannot access mental health or disability services? How often do agencies invoke the compelling-reason exception to avoid TPR when no adoptive family exists? Which states file TPR petitions most aggressively, and which courts resist creating legal orphans?
These are not radical questions. They are basic accountability questions. If a state can permanently sever a child’s legal family, the public has a right to know whether that act actually produced permanency. If federal dollars reward adoption finalizations, taxpayers have a right to know how many of those adoptions later failed. If families are surrendering children to access treatment, Congress has a duty to know how often it happens and why. If judges are resisting TPR without placement, policymakers should ask whether those judges are identifying a real child-protection problem rather than treating them as obstacles.
The reform agenda should begin with a simple principle: no child should be made a legal orphan for administrative convenience. Before filing or granting TPR in a case without an identified adoptive family, courts should require a heightened showing that termination is necessary, that less restrictive alternatives have been considered and rejected for specific reasons, that relatives have been meaningfully searched and supported, that required services were actually provided, and that the child’s risk of legal orphanhood has been weighed on the record. The state should not be allowed to hide behind generic claims that adoption is the goal. It should have to show the path.
Second, federal reporting should require states to publish TPR-to-adoption gap data annually in plain language. The public should be able to see, state by state, how many TPRs were filed, how many adoptions finalized, how many children were waiting, how many children remained legally orphaned, and how many aged out after TPR. This should not require specialized databases or live queries. It should be a mandatory public dashboard. If a state terminates parental rights faster than it finalizes adoptions, that gap should be treated as a performance failure, not hidden inside separate tables.
Third, adoption incentives should be redesigned around durability, not merely finalization. States should not receive full credit for an adoption unless the placement remains intact for a meaningful period, and federal law should require tracking of re-entry after adoption using persistent child identifiers. If an adoption dissolves, the event should be connected back to the original incentive record. If the same child is re-adopted, the system should flag the child’s history rather than treating the re-adoption as an entirely fresh success. Incentives should reward stable permanency, not repeated transactions.
Fourth, therapeutic relinquishment should be treated as a civil rights and health care failure, not a child welfare inevitability. Families should not have to surrender custody to obtain residential treatment, psychiatric stabilization, or disability-related services. Federal law should require states to track every voluntary custody transfer linked to service access and should prohibit agencies from using custody relinquishment as the gateway to medically necessary care. If a child needs treatment, treatment should follow the child into the family whenever safely possible. Custody destruction should not be the price of admission.
Fifth, courts should expand the use of alternatives that preserve family connections when full reunification is not immediately safe. Subsidized guardianship, kinship care, permanent legal custody, open adoption agreements with enforceable contact provisions, and post-reunification supports should not be treated as lesser options simply because they do not produce the clean legal finality of adoption. For some children, especially older youth and children with strong family identity, legal severance may be more harmful than a carefully structured guardianship or kinship arrangement. Permanency should be measured by the child’s durable relationships, not by the system’s preference for closed files.
Finally, the language itself must change. “Freed for adoption” should not be used when a child has no adoptive family. A more honest term would be “legally severed and awaiting adoption.” That phrase is uglier because the reality is uglier. Bureaucratic euphemism allows the system to transform violent legal acts into comforting abstractions. The public cannot fix what it cannot name. If a child has no legal parents and no adoptive parents, the system should say so plainly.
The paper-ready pipeline survives because it is wrapped in the language of rescue. It tells the public that children are being moved out of limbo, that adoption is being promoted, that timelines are being enforced, and that permanency is the goal. But when you follow the data, the legal structure, the incentive payments, the dissolution problem, the therapeutic relinquishment pathway, and the aging-out outcomes, a different picture emerges. The machine is very good at producing legal finality. It is far less reliable at producing lifelong family.
That distinction is the story. America has built a child welfare system that can permanently erase a family before it has built another one. It can count the severed child as waiting, count the adoption as success, keep the bonus if the adoption later fails, potentially count the child again if re-adopted, and leave thousands of young people to age out carrying the lifelong consequences of institutional promises that were never kept. This is not permanency. It is a paper-ready pipeline.
And the children trapped inside it are not statistics, not compliance units, not incentive events, and not entries in a federal database. They are children who were promised safety and belonging. Too many received a court order instead.
If you enjoyed this work and feel encouraged by independent investigations like this, please consider becoming a paid subscriber to The Constitutional Republic on Substack.
Your support helps us continue:
investigative reporting
public records research
educational transparency projects
community outreach
and future investigations like this series
Most importantly, please share this article.
Independent journalism survives when ordinary people help spread information powerful institutions would prefer remain ignored, misunderstood, or buried beneath complexity.
Because silence protects systems.
Informed communities protect children.
ALL paid subscriptions are tax deductible Project Milk Carton | 501(c)(3) | EIN: 33-1323547
PRIMARY SOURCES AND CITATIONS
AFCARS/CivicOps (TPR, adoption, waiting data)
CivicOps AFCARS database, live query May 18, 2026 — FY2021 and FY2022 data. CivicOps is sourced from HHS Children's Bureau AFCARS annual reports.
42 U.S.C. § 675(5)(E) — ASFA TPR mandate https://www.law.cornell.edu/uscode/text/42/675 — Full statute with concurrent family identification language
45 CFR § 1356.21(i)(1)-(3) — Implementing regulation https://www.law.cornell.edu/cfr/text/45/1356.21 — Regulatory framework for 15/22-month trigger and concurrent search
ASPE/HHS (2021) — Freeing Children for Adoption within ASFA Timeline https://aspe.hhs.gov/reports/freeing-children-adoption-within-adoption-safe-families-act-timeline — Documents 7 states where judges resist creating legal orphans; classified as compliance failure
GAO-03-397 (August 2003) — Therapeutic Relinquishment https://www.gao.gov/products/gao-03-397 — The 12,700/19-states finding for FY2001; methodology and state breakdown
Santosky v. Kramer, 455 U.S. 745 (1982) https://supreme.justia.com/cases/federal/us/455/745/ — Constitutional floor for TPR; clear and convincing evidence standard
Troxel v. Granville, 530 U.S. 57 (2000) Reaffirmed parental rights as fundamental liberty interests under 14th Amendment — strengthens constitutional challenge space
Texas Family Code Chapter 161 and 263 https://statutes.capitol.texas.gov/Docs/FA/htm/FA.161.htm | Texas DFPS Handbook §5560: TPR pleaded at original removal petition
Ohio Revised Code §§ 2151.413–2151.414 https://codes.ohio.gov/ohio-revised-code/section-2151.413 | Ohio's 12-consecutive-month threshold (more aggressive than federal floor)
Chapin Hall Midwest Evaluation — Age 26 Outcomes https://www.chapinhall.org/wp-content/uploads/Midwest-Eval-Outcomes-at-Age-26.pdf — Gold standard longitudinal study: 36% homeless, 74% males incarcerated
Colorado Sun Failed Adoptions Investigation (Nov. 2022) https://coloradosun.com/2022/11/14/colorado-failed-adoptions-foster-kids-welfare/ — 1,094 children/13%/8.5 years — based on CDHSS CORA data
Colorado Sun — Cost of Aging Out (Nov. 2023) https://coloradosun.com/2023/11/15/it-costs-colorado-343453-per-kid-who-ages-out-of-foster-care/ — Common Sense Institute: $343,453 per youth
Wildeman, Edwards, Wakefield (2020) — Cumulative TPR Prevalence https://journals.sagepub.com/doi/10.1177/1077559519848499 | Child Maltreatment journal — 1 in 100 children experience TPR by 18; 60% increase 2000–2016
NCCPR Issue Paper 14 — Legal Orphans Methodology https://nccpr.org/nccpr-issue-paper-14-family-preservation-and-adoption/ — 124,000+ additional aging-out figure; NCCPR analysis of HHS trend data
NAMI — Keeping Families Together Act https://www.nami.org/press-releases/keeping-families-together-act-re-introduced — Congressional history of failed legislation to prohibit therapeutic relinquishment
CRS Report R43025 — Adoption Incentive Program https://www.everycrsreport.com/reports/R43025.html — $1.04B cumulative bonuses; double-counting concern; proposed H.R. 4980 tracking requirement
Applying Strict Scrutiny in TPR Proceedings (2024) https://law.ua.edu/wp-content/uploads/2024/12/Applying-Strict-Scrutiny-in-Termination-of-Parental-Rights-Proceedings.pdf — U of Alabama Law Review
CWLA Testimony on ASFA Implementation (April 8, 2003) https://www.encyclopedia.com/social-sciences/.../cwla-testimony-hearing-implementation-of-adoption-and-safe-families-act — Time from removal to TPR decreased; time from TPR to adoption increased; net unchanged at 44 months
Nebraska Safe Haven 2008 https://history.nebraska.gov/safe-haven-law-2008/ — 36 children surrendered in 127 days; emergency amendment to LB1 (newborns only, Nov. 22, 2008)
Annie E. Casey Foundation — Aging Out https://www.aecf.org/blog/what-happens-to-youth-aging-out-of-foster-care — 1 in 5 homeless after 18; $300,000 lifetime taxpayer cost
Project Milk Carton — ARIA Investigative Research | projectmilkcarton.org | 501(c)(3) EIN: 33-1323547