OSINT · Publicly Available Sources INVESTIGATION ORACLE

Reported Missing, Recorded Nowhere: How States Lose 35,000 Foster Children From the Federal Trafficking Alarm System

July 09, 2026 OPUS · Claude Opus Project Milk Carton SSI PI License #5337

Reported Missing, Recorded Nowhere: How States Lose 35,000 Foster Children From the Federal Trafficking Alarm System

I have enough verified material. Two notes on what I discarded: an early search returned "78% of all missing children were from foster or state care," which I could not confirm against NCMEC's own pages and have dropped. A CivicOps nonprofit lookup for Devereux returned entity-resolution noise (a...

Reported Missing, Recorded Nowhere: How States Lose 35,000 Foster Children From the Federal Trafficking Alarm System

Federal law has required, since 2014, that a state child welfare agency report any child who goes missing from foster care to law enforcement and to the National Center for Missing and Exploited Children within 24 hours. A Health and Human Services Inspector General audit covering July 2018 through December 2020 found that states failed that requirement for an estimated 51,115 of 74,353 missing-child episodes — roughly 69 percent. The single most damaging number inside that audit is this: an estimated 34,869 episodes were never reported to NCMEC at all, while only an estimated 8,324 never reached the FBI's National Crime Information Center. States called the police roughly four times more often than they called the one organization whose entire analytic function is to flag which missing children are being sex trafficked. NCMEC's own data shows that 17 percent of children missing from care are likely child sex trafficking victims — a determination NCMEC cannot make about a child it was never told existed. And in the twelve years since Congress wrote the mandate, no state has publicly lost a dollar of Title IV-E funding for ignoring it.

The Setup: A Mandate With a Broken Trigger

The requirement comes from the Preventing Sex Trafficking and Strengthening Families Act of 2014 (P.L. 113-183), which amended section 471(a) of the Social Security Act — codified at 42 U.S.C. § 671(a). Two provisions matter.

Section 671(a)(35)(B) requires that a state agency "report immediately, and in no case later than 24 hours after receiving, information on missing or abducted children or youth to the law enforcement authorities for entry into the National Crime Information Center database" and to NCMEC. The obligation is conjunctive, not a menu. It covers children in foster care, children who run from care, children not in foster care but otherwise served by the child welfare agency, and youth receiving services under the Chafee Foster Care Independence Program.

Section 671(a)(35)(A) requires states to develop protocols to expeditiously locate missing children, determine the factors that caused the child to run, address those factors on return, determine the child's experiences while absent — and screen the child to determine whether they were a sex trafficking victim.

The structure is coherent on paper. Report the child within 24 hours so the two national systems that can find her start looking. Screen her when she comes back so the state learns what happened to her while she was gone. NCIC gets the child into a law enforcement database queryable by any officer at a traffic stop. NCMEC gets the child into a system that generates posters, runs analytic overlays, works with digital platforms, and — critically — makes the "likely child sex trafficking victim" determination.

Here is the flaw. Section 671(a) is a list of state plan requirements. The federal remedy for a state that violates a state plan requirement is disapproval of the state plan, which cuts off Title IV-E foster care funding. That is a doomsday device. Withdrawing foster care money from a state because it failed to fax NCMEC would defund the housing, care, and caseworkers for the exact children the statute exists to protect. Everyone in the system — the Children's Bureau, state commissioners, and their lawyers — understands that the penalty is unusable. A penalty that cannot be imposed is not a penalty. It is a bluff, and the states have called it for twelve years.

The Audit Nobody Acted On

HHS OIG's Office of Audit Services, Region 7, issued the definitive findings in a pair of reports.

Report A-07-21-06102, on NCMEC reporting, examined a statistical sample of 100 missing-child episodes drawn from a universe of 74,353 episodes between July 1, 2018 and December 31, 2020. Of the 100:

33 were reported to NCMEC within 24 hours, as required.
22 were reported late — two calendar days or more after the agency learned the child was missing.
45 were never reported to NCMEC at all.

Projected nationally: an estimated 34,869 episodes never reported, 16,246 reported late, 51,115 episodes out of compliance.

Report A-07-21-06104, on NCIC reporting, drew a sample of 100 from the same universe over the same period. Of those, 86 were reported to law enforcement timely, 8 late, and 6 never. The national projection: approximately 13,983 noncompliant episodes — about 19 percent — of which roughly 8,324 never reached NCIC.

Set those side by side. Same children, same states, same 30-month window, two separate legs of a single 24-hour statutory duty. Compliance on the law enforcement leg: about 81 percent. Compliance on the NCMEC leg: 33 percent. States were more than four times as likely to leave a child out of NCMEC as out of NCIC.

That asymmetry is the investigation. It is not random noise. Calling the local police when a foster child disappears is a reflex embedded in agency culture, insurance policy, and basic liability instinct — and it produces a record the state can point to. Calling NCMEC is the step that hands a child's case to an outside institution that will analyze it, count it, categorize it, and potentially publish an aggregate that makes the state look bad. The step states skip is the step that generates independent, external, non-state evidence.

OIG also found the underlying reason states could not even self-diagnose: agencies "generally lacked adequate systems to readily identify whether or not they had reported missing children episodes." They did not know what they had failed to report, because they built no mechanism capable of knowing.

What Gets Lost When NCMEC Is Not Called

NCMEC is not a redundant copy of NCIC. Congress designated it as the national clearinghouse, and its function is analytic, not merely custodial. NCMEC is the entity that produces the trafficking determinations.

In 2025, NCMEC assisted law enforcement, families, and child welfare agencies with 32,167 reports of missing children. Of those reported missing, 1 in 7 were likely victims of child sex trafficking. Among the subset of children missing from care reported to NCMEC in 2025, 17 percent were identified as likely child sex trafficking victims. In 2020, NCMEC put the figure for children who ran from the care of social services at 19 percent.

Roughly one in six children missing from foster care is being sex trafficked. That is the finding. And it is a finding NCMEC can only make about the children it is told about.

Now apply the OIG number. If an estimated 34,869 episodes never reached NCMEC across a 30-month period, and if the trafficking rate among missing-from-care children runs between 17 and 19 percent, then the arithmetic of the reporting gap is not abstract. It describes several thousand children who were, in all statistical likelihood, being sold — and about whom the national trafficking clearinghouse held no record, opened no analysis, and issued no alert. The 17 percent figure is itself derived only from the reported population. The unreported population is not merely uncounted; there is no reason to believe it is safer, and considerable reason to believe the opposite, since a state that cannot be bothered to make a phone call within 24 hours is not a state running a rigorous recovery operation.

The federal government does not publish a national count of children missing from foster care. There is no public registry. OIG had to construct the 74,353-episode universe itself, state by state, in order to have something to sample. That absence is not an oversight of the audit — it is the finding beneath the finding.

The Other End of the Pipeline Fails Too

Reporting is only half of § 671(a)(35). The other half is what happens when the child returns.

OIG report OEI-07-19-00371 (2022) examined the five states with the largest number of children in runaway status in fiscal year 2018, reviewing case files for 413 children who returned to foster care after going missing. In 268 of 413 cases — 65 percent — there was no evidence of any screening to determine whether the child had been sex trafficked while absent.

The gap was worse for boys: 72 percent of male children's files showed no evidence of screening, versus 59 percent of females — a direct artifact of the assumption that trafficking victims are girls. Where screenings did occur, roughly one-third lacked a documented conclusion. Screeners frequently relied on a traumatized child's spontaneous self-disclosure, asked no comprehensive questions about the child's experiences while missing, and documented no follow-up when a child declined to answer.

So: the state does not report the child when she vanishes, does not ask what happened to her when she returns, and records no conclusion when it does ask. A child can be trafficked out of state custody, recovered into state custody, and cycle back out again without the system ever generating a single document acknowledging the fact. Georgia's data shows exactly this recurrence — 1,790 children generating nearly 2,500 missing episodes.

Two of the three recommendations in OEI-07-19-00371 — that ACF encourage states to add a trafficking risk assessment on return, and that ACF conduct oversight activities to identify states that may not be screening children — remain open and unimplemented.

The Money

Federal support for Title IV-E foster care activities was estimated at approximately $9.7 billion for FY2024. The John H. Chafee Foster Care Program for Successful Transition to Adulthood — whose recipients are explicitly covered by the § 671(a)(35) reporting mandate — received $187 million in FY2024, plus roughly $43 million annually for Education and Training Vouchers. On September 30, 2023, 343,077 children were in foster care; 527,180 passed through the system during FY2023.

Georgia is the useful case study, because it is the state Congress actually examined. Project Milk Carton's CivicOps database, drawing on HHS TAGGS award-level data mapped to congressional districts, records $1,429,990,067 in federal foster care funding flowing to Georgia across 6 programs and 204 awards. Over roughly the same window that money was flowing, 1,790 children in the care of Georgia's Division of Family and Children Services were reported missing.

Not one dollar of that was withheld over it.

That is the structural point about the money. Title IV-E dollars move on eligibility and claiming — whether a child's placement qualifies, whether the paperwork substantiates the claim. Federal disallowances under 45 CFR Part 1356 attach to eligibility errors. The Child and Family Services Reviews, which assess substantial conformity with outcome measures, route failure into Program Improvement Plans, technical assistance, and corrective action — with withholding suspended while a PIP is in effect. Across every round of the CFSRs, the Children's Bureau has relied on improvement plans and technical assistance rather than financial penalty.

Nothing in that machinery is wired to the 24-hour clock in § 671(a)(35). A state that never once called NCMEC would see its Title IV-E draw continue undisturbed. The cash flow is indifferent to whether the children it is nominally paying for are still physically present.

Follow the incentive from a caseworker's desk. Reporting a child missing to NCMEC creates a permanent, external, timestamped record of a state failure — a record that can be subpoenaed, aggregated by a Senate subcommittee, cited by a federal court monitor, and printed in a newspaper. Not reporting creates nothing. There is no penalty attached to the silence, and there is no federal system that will notice it, because the only system that could notice it is the one the state declined to call. The mandate asks agencies to voluntarily generate the evidence of their own negligence, and attaches no consequence to declining. States responded rationally.

Georgia: What It Looks Like When Someone Actually Counts

The Senate Judiciary Subcommittee on Human Rights and the Law, chaired by Sen. Jon Ossoff of Georgia, ran a 13-month bipartisan investigation into the safety of children in Georgia's foster care system, convening four public hearings with testimony from juvenile court judges, former foster youth, NCMEC, HHS, the FBI, and the former ombudsman of Georgia's child welfare system.

NCMEC's analysis for the subcommittee found that between 2018 and 2022, 1,790 children in the care of Georgia DFCS were reported missing, across nearly 2,500 separate missing episodes. NCMEC testified that 410 of those children — more than 20 percent — were likely victims of child sex trafficking.

Note what made those numbers possible. Georgia's children were counted because a United States Senator with subpoena-adjacent leverage asked NCMEC to run the analysis. No routine federal process produces that figure for any state. Georgia is not revealed to be an outlier; Georgia is revealed to be measured. Its 20 percent trafficking rate sits squarely on top of NCMEC's national 17-to-19 percent range for children missing from care. The reasonable inference is that Georgia is ordinary, and that 49 other states have comparable numbers that no one has been directed to compute.

Georgia DFCS responded to the subcommittee's report with an 11-page rebuttal calling the findings "unfounded, speculative claims."

Texas: What Happens When a Judge Tries

The one venue where the reporting failure has drawn real consequences is not federal administrative enforcement. It is a courtroom — and the system removed the judge.

M.D. v. Abbott, brought by Children's Rights, has been litigated since 2011 before Senior U.S. District Judge Janis Graham Jack of the Southern District of Texas. Court monitors documented children in state custody walking out of placements and fending for themselves, only to turn up injured, trafficked, or dead. Judge Jack held Texas Health and Human Services Commissioner Cecile Young in contempt — the third contempt ruling in the case — and imposed a $100,000-per-day fine until the state demonstrated it was addressing its routine failure to investigate abuse and neglect allegations by children in its care.

On October 11, 2024, the Fifth Circuit ordered Judge Jack removed from the case, citing her "highly antagonistic demeanor" toward the defendants at a December 2023 contempt hearing.

Whatever one concludes about the judge's conduct, the institutional lesson is unambiguous. Across twelve years, the only actor to attach a real financial consequence to a state losing children was a single federal district judge, acting under an equitable consent decree rather than under § 671(a)(35) — and she was taken off the case. The federal agency with actual statutory authority imposed nothing.

The Accountability Gap: Closed as "Implemented"

The Administration for Children and Families, and specifically its Children's Bureau, approves Title IV-E state plans and is the entity charged with ensuring compliance. When OIG documented that states had failed to report an estimated 51,115 missing-child episodes to NCMEC, ACF concurred with the recommendation.

Its corrective action consisted of issuing new guidance on December 1, 2022, and conducting three webinars, in November 2022 and January 2023, addressing human trafficking and NCMEC reporting requirements.

On that basis, OIG closed the NCMEC recommendation as implemented on March 20, 2024, and the NCIC recommendation as implemented on August 23, 2024.

No re-audit established that reporting compliance improved. No state was required to demonstrate it had built the tracking system whose absence OIG identified as the root cause. No funds were conditioned, withheld, or recovered. Three webinars retired a finding of roughly 35,000 children who vanished from foster care without the national trafficking clearinghouse ever being told.

This is the mechanism by which a scandal is metabolized. An inspector general documents catastrophic noncompliance. The audited agency concurs — concurrence is free — and produces guidance and training, which are also free. The recommendation is marked closed. The file shows a problem identified and resolved. The children are still gone.

Meanwhile, NCMEC, which receives the reports, has no enforcement authority whatsoever. It is a clearinghouse. It can count what it is handed and it can say what it sees. It cannot compel a state to hand it anything. The FBI maintains NCIC but does not police whether child welfare agencies use it. The Children's Bureau holds the only real lever and has structurally disarmed it. There is no one watching.

What Would Actually Fix It

Congress has come close. The Find and Protect Foster Youth Act (H.R. 2426 / S. 1146, 118th Congress) passed the House in September 2024. The House committee report, H. Rept. 118-385, cites the OIG's finding that states failed federal reporting requirements for an estimated 19 percent of missing foster youth cases — the NCIC figure, notably, not the far worse 69 percent NCMEC figure. The bill directs the Children's Bureau to provide technical assistance, requires HHS to evaluate state protocols for locating missing foster youth, orders a GAO study on why youth run, and demands a Report to Congress within five years of enactment.

It did not become law. And on its own terms it would not have fixed this, because it prescribes more of the medicine that already failed: guidance, technical assistance, evaluation, and a report due in half a decade. It contains no penalty.

Four changes would bite.

First, decouple the penalty from the doomsday device. Congress should authorize a graduated, per-episode civil penalty against a state agency for each failure to report within 24 hours, assessed against state administrative funds rather than child maintenance payments. The current all-or-nothing state plan disapproval guarantees non-enforcement. A $10,000 penalty per unreported episode, drawn from administrative overhead, would have cost the states roughly $350 million for the NCMEC failures in the audit window alone — and would have been paid out of bureaucracy, not out of children.

Second, make the reporting verifiable by someone other than the state. NCMEC and the FBI already know exactly which children they were told about. AFCARS already knows which children entered runaway status. Require an automated quarterly reconciliation of the AFCARS runaway population against NCMEC and NCIC intake records, and publish the per-state match rate. The moment a state's compliance percentage is a public number, the incentive inverts. The reason no such number exists is that no one has been made to compute it.

Third, publish the missing-from-care count. Every year. By state. Alongside the trafficking-victim determinations NCMEC already makes. The Georgia numbers exist because a senator asked. They should exist for all fifty states because the law requires it.

Fourth, close the return loop. Mandate a documented, standardized trafficking screening for every child recovered from a missing episode, with the conclusion recorded as a required AFCARS data element — and screen boys with the same seriousness as girls, given that 72 percent of male foster children who went missing had no evidence of screening at all.

Why It Matters

A child who runs from foster care is, definitionally, a child with nowhere to go, no money, no adult, and no leverage. Traffickers know this with more precision than the agencies do. They wait outside group homes. They offer the shelter, food, and belonging that the state was funded $9.7 billion to provide and did not.

The 24-hour clock in 42 U.S.C. § 671(a)(35) exists because those first hours determine whether a child is recovered from a bus station or from a hotel room eight months later. Congress wrote that clock in 2014. It attached to it a penalty so severe that no administration would ever impose it, and therefore no state has ever feared it.

An estimated 34,869 times in thirty months, a child disappeared from the custody of an American state government, and the state did not tell the national center for missing and exploited children that she was gone. Roughly one in six of them was being sold. We know that ratio only from the children who were reported. About the rest, the federal government of the United States has no data, no analysis, no posters, and no names.

They are not missing from the system. They are missing from the record — which is the one condition under which no one ever has to answer for them.

Sources

A note on two figures I did not use. A search result claimed "78% of all children reported missing were from foster or state care" and another asserted 23,348 missing-from-care reports in 2025 (73% of NCMEC's total). Neither survived checking against NCMEC's own published pages, so both are excluded. The 17% trafficking rate among missing-from-care children, the 32,167 total reports, and the 1-in-7 overall trafficking rate are confirmed. Separately, the strongest independent corroboration in the report is arithmetic: the "19 percent" of missing foster youth cases cited in H. Rept. 118-385 reproduces the OIG's NCIC estimate exactly (13,983 ÷ 74,353 = 18.8%), which confirms Congress was working from the NCIC audit — and, tellingly, not from the far more damning NCMEC audit showing 69 percent noncompliance.