Sixty Days on Paper, Eighteen Months in Reality: How the Interstate Compact on the Placement of Children Became a Black Box Nobody Audits

Every year roughly 40,000 American children in foster care have a relative waiting for them on the other side of a state line, and every year those children enter a bureaucratic process that has no referee. The Interstate Compact on the Placement of Children — drafted in 1960, enacted as statute in all 50 states, the District of Columbia and the U.S. Virgin Islands — requires a receiving state to complete a home study before a child can be placed across state lines. Federal law has capped that home study at 60 days since 2006. The available research says fewer than half of them finish on time, and nearly a third take more than 90 days. When the compact becomes too slow to tolerate, agencies simply route around it: children get placed in unapproved homes, or handed to relatives under a discretionary arrangement called "courtesy supervision" that the receiving state is under no obligation to provide and can end at will. And through all of it, neither the Department of Health and Human Services nor the compact's own governing body — the Association of Administrators of the Interstate Compact on the Placement of Children — publishes a single national figure on how long these cases take, how often deadlines are blown, how often the compact is violated outright, or what happens to the children afterward. Title IV-E reimbursement, meanwhile, never pauses.

How the Machine Actually Works

The ICPC is a state-to-state contract, not a federal program. Its operative provision, Article III, bars a "sending agency" from placing a child into another state for foster care or preliminary to adoption until the receiving state's compact administrator notifies the sending agency in writing "that the proposed placement does not appear to be contrary to the interests of the child." Until that letter arrives, the placement is unlawful under the receiving state's own statute.

The mechanics run through a paper chain that predates fax machines. A caseworker in the sending state completes Form ICPC-100A and a packet of supporting documents. That packet goes to the sending state's central compact office, which reviews it and transmits it to the receiving state's central compact office, which assigns it to a local county or regional agency, which assigns it to a worker, who schedules the home visit, conducts background and child-abuse-registry checks, writes the study, and routes it back up the same chain in reverse. Seven sequential handoffs, minimum. Two state bureaucracies, two local agencies, and at every joint the file sits in someone's queue.

Congress tried to put a clock on it. The Safe and Timely Interstate Placement of Foster Children Act of 2006 (P.L. 109-239) amended the Title IV-E state plan requirements at §471(a)(26) of the Social Security Act to require states to complete an interstate home study and report back within 60 calendar days. The AAICPC folded that into ICPC Regulations 1 and 2, which give the receiving state 60 days to return a home study evaluation and at least an initial report.

For emergencies there is ICPC Regulation 7, "priority placement," available when a court makes express findings — the child is four or younger, or an older sibling of such a child, or a parent has suddenly died, been incarcerated or incapacitated. Regulation 7 timelines are aggressive on paper: the sending agency transmits within 3 business days of a signed order of compliance, the sending compact administrator forwards within 2 business days, and the receiving state must render a decision within 20 business days.

Here is the enforcement provision, in full substance: if the receiving state misses the window, it "shall be deemed out of compliance," and the sending state's court "may inform an appropriate court in the receiving state and request assistance." That is the entire remedy. One judge writing a letter to another judge in a different sovereign state, asking politely. There is no fine, no withheld reimbursement, no federal referral, no published list of non-compliant states. The compact's fastest, most urgent track — the one reserved for infants and children whose parent just died — terminates in a request for a favor.

The Denial Trapdoor and the Missing Appeal

Delay is only half the failure. Child welfare agencies deny roughly 40 percent of all ICPC placement requests, and the compact contains no specific standards governing how a receiving state must evaluate a proposed home and no nationwide process to appeal a denial. States vary in what internal review, if any, they offer. A grandmother in Ohio can be found unsuitable for a grandchild in Kentucky by a caseworker she will never meet, applying criteria she will never see, with no forum in which to contest it.

The practical result was captured in a case documented by The Imprint: a paternal grandmother who was herself already a licensed foster parent waited 18 months for New York to complete her ICPC home study. When New York finally approved her, child welfare professionals turned around and argued the boy should stay in Michigan with the foster parents he had by then bonded to. The delay manufactured the argument against the family. This is the compact's most corrosive dynamic — the process consumes so much of a child's life that "best interests" and "least disruption" quietly swap places, and the relative loses on a timeline the agency itself created.

The courts have been carving away at the compact's reach for three decades, largely because judges recognized it was being used to block reunification. McComb v. Wambaugh, 934 F.2d 474 (3d Cir. 1991) held the ICPC does not apply to placement with an out-of-state parent, reasoning that treating a child's return to a parent as a "placement" would absurdly impose a financial obligation on the sending state superseding parents' own duty of support. In re Emmanuel R., 94 Cal.App.4th 452 (2001) struck down conditional orders that placed a child contingent on later ICPC completion. In Matter of D.L. v. S.B., 2022 NY Slip Op 05940, the New York Court of Appeals held the ICPC does not apply to out-of-state noncustodial parents seeking custody of children not in the custody of a social services agency. Every one of those rulings is a judicial finding that the compact was being applied to keep children from parents it was never written to cover.

The Money: Reimbursement That Never Pauses

Federal child welfare spending under Title IV-E ran approximately $9.7 billion in FY2024. In an average FY2024 month, 713,200 children received some form of IV-E assistance, including 112,300 in foster care maintenance. The federal government reimburses 50 to 83 percent of maintenance costs depending on the state's FMAP, a flat 50 percent of administrative costs, and 75 percent of training.

Two features of that formula deserve scrutiny in the ICPC context.

First, an ASPE analysis found that from FY2019 through FY2023, administrative costs — not payments to foster parents — were the largest single driver of Title IV-E foster care expenditures and reimbursements. Caseworker time, case management, eligibility determination, and the paperwork of placement are reimbursable at 50 cents on the dollar. That means the labor of processing an ICPC packet is itself a billable federal activity. The compact's inefficiency is not a pure cost center for a state agency; a meaningful share of it is federally subsidized.

Second, and more consequential: a child sitting in a licensed non-relative foster home in the sending state while an out-of-state relative's home study languishes remains fully IV-E eligible. IV-E maintenance payments flow when a child is in a licensed foster family home, an approved relative home meeting licensing standards, or a qualifying institution. The child in limbo is in exactly such a setting. Delay does not interrupt the money. It is fiscally neutral — often fiscally positive, given per-diem payments to congregate care and contracted foster care providers — for the sending state to let the clock run. The one party with an economic interest in speed is the relative, who has no standing to demand it.

The receiving state faces the mirror-image incentive. It absorbs 100 percent of the home study labor cost and, if the placement is approved, inherits a child who will consume its local services for years, while the sending state retains custody and the IV-E claim. Nothing in the compact compensates the receiving state for the work. A state that staffs its ICPC office thinly and lets the queue grow suffers no penalty and saves money. This is the load-bearing fact of the entire system: the 60-day deadline is attached to a duty that no one is paid to perform and no one is punished for ignoring.

The Data Infrastructure That Exists — and Is Kept Private

The most damning element of this investigation is not that the data doesn't exist. It is that it does, and it is not public.

The National Electronic Interstate Compact Enterprise (NEICE) is the national data exchange for ICPC cases. It began as a six-state pilot in November 2013, funded through the Office of Management and Budget's Partnership Fund for Program Integrity Innovation at roughly $1.25 million, administered by the Administration for Children and Families under Children's Bureau grant 90XA0151. It is operated by the American Public Human Services Association (APHSA), a Washington-based nonprofit membership association of state and local human services agencies (EIN 36-2166948, roughly $8.2 million in annual revenue), acting as secretariat to the AAICPC. Nineteen states were live by March 2018. As of mid-2026, 48 states and jurisdictions are fully operational.

The Family First Prevention Services Act (P.L. 115-123, 2018) made participation effectively mandatory: every AAICPC jurisdiction must be on an electronic interstate case-processing system by October 1, 2027. NEICE is the only system that satisfies the requirement. States pay approximately $32,000 to join, about $3,500 per year in maintenance, plus a $25,000 annual licensing fee — a fee that state program leads surveyed in Child Trends' federal evaluation flagged as a sustainability concern. Those fees are paid largely from state child welfare funds and federal dollars, "predominately Title IV-E administrative costs and CAPTA funds." In other words, federal taxpayers substantially fund a nonprofit-run system whose outputs they cannot see.

Because NEICE support documentation is public even when its data is not, we know precisely what reports the system generates. It produces a Case Statistics Report — "children and case statistics for the logged in user state, including all cases sent or received for a selected date range." It also produces a Violation Statistics Report.

Read that again. The compact's national data system has a built-in report that counts ICPC violations. Its visibility is scoped to "the logged in user state." Each state can see its own numbers. No state can see another's. The public can see none. No aggregate national compliance figure — not median days to home study, not percent completed within 60 days, not Regulation 7 timeliness, not denial rates by state, not violation counts — is published by APHSA, by the AAICPC, or by HHS.

Nor does the federal reporting system fill the gap. AFCARS, the Adoption and Foster Care Analysis and Reporting System, is HHS's mandatory case-level collection on every child in foster care, and it carries no element that identifies a child as awaiting or subject to an ICPC determination. The Child and Family Services Reviews measure permanency and safety outcomes but do not audit interstate compact timeliness as a standalone federal measure. So the one federal database that could cross-reference "how long did this child wait" against "where did this child end up" has never been built to ask the question.

The 60-day requirement at §471(a)(26) is, on its face, a Title IV-E state plan requirement — a condition of receiving federal money. HHS possesses the authority to treat systematic violation as a state plan compliance issue. There is no public record of it ever having done so at scale, and without a data element there is no mechanism by which it could even detect a violation.

The Bypass: Unapproved Placements and "Courtesy Supervision"

When the front door takes 90 to 500 days, agencies use side doors.

Unapproved placement. State ICPC manuals are blunt about what happens when a child is moved before approval. Los Angeles County DCFS policy states that such a placement "is considered a violation of ICPC and the placement is made with the sending state bearing full liability and responsibility for the safety of the child." The receiving state "may request immediate removal of the child." Critically, the receiving state is "permitted to proceed, but not required to proceed" with the home study — meaning a violation can freeze the very vetting the child needs. Multiple state manuals state the consequence plainly: a child in an unapproved placement is considered "at risk," and the protections available under the ICPC are not available to that child in the receiving state. No licensing oversight from the state the child physically lives in. No local caseworker with statutory authority. Often no IV-E-eligible placement setting, which means no maintenance payment to the caregiver — the family absorbs the cost of raising the child.

Courtesy supervision. This is the softer bypass and the more common one. The receiving state agrees to send a worker to lay eyes on the child periodically and report back to the sending state. The AAICPC's own framing is the tell: "courtesy" is defined as "consent or agreement between states to provide a service that is not required." The receiving state may open a case for courtesy supervision; it is never obligated to. It may close it whenever caseloads tighten. The visits are not licensing inspections. The worker conducting them has no independent authority to remove the child — that authority sits with a court hundreds of miles away that has no jurisdiction over the receiving state's agencies. A child under courtesy supervision occupies a jurisdictional seam: legally the responsibility of a state that cannot see him, physically in a state that owes him nothing.

There is no national count of children in either category. There cannot be — the data element does not exist.

What Federal Investigators Found When They Finally Looked

Senator Jon Ossoff's Senate subcommittee conducted a 13-month bipartisan investigation into foster child safety, releasing a 64-page report in April 2024 built on four public hearings, more than 100 witness and source interviews, and thousands of pages of documents. Its findings on Georgia's Division of Family and Children Services included that agency mismanagement was a key contributor to child deaths and serious injuries, and that hundreds of children in DFCS custody were likely sex trafficked over a five-year period.

The finding most relevant here concerns "hidden foster care" — children separated from parents and placed informally with a caregiver, outside any court's involvement or authority. The subcommittee found that voluntary kin placements are not appropriately vetted, documenting cases in which children were placed in the homes of people with histories of sex crimes or prior agency involvement. These children, the investigation concluded, are not counted in official foster care numbers.

That is the same structural pathology as an unapproved ICPC placement, differing only in whether a state line is crossed. Ossoff and Senator John Cornyn introduced the bipartisan Foster Care Placement Transparency Act, which would require states to measure and report hidden foster care placements to HHS and require HHS to report publicly to Congress. The bill's premise is an admission: the federal government does not currently know where a large population of removed children physically are. Interstate placements are the subset of that population where the accountability vacuum is written into law rather than merely tolerated in practice.

The litigation landscape reflects the same story. As of January 2025, 34 child welfare class actions were active across 28 states, with such litigation lasting an average of 11 years, and at least five states defending multiple suits simultaneously. In November 2021 the Legal Aid Society and Dechert LLP filed a class action in the Eastern District of New York against Governor Hochul, the state Office of Children and Family Services, and New York City's Administration for Children's Services, challenging practices that deny children placement with willing relatives — including categorical exclusion of kin who were convicted of any of nearly 300 mandatory excluding crimes, regardless of how many decades earlier. Those same categorical screens, applied by a receiving state's caseworker with no appeal, are what produce the 40 percent ICPC denial rate.

Federal awareness is not new. The Adoption and Safe Families Act of 1997 directed GAO to study ICPC operation; GAO's 1998 report concluded the interjurisdictional process was longer and more intricate than in-state adoption and recommended HHS produce an action plan promoting standardized home studies and ICPC training. That was twenty-eight years ago. The 60-day statute arrived eight years after that. The electronic system arrived seven years after that. The mandate to actually use the electronic system takes effect in 2027 — sixty-seven years after the compact was drafted. At no point in that sequence did anyone build the thing that would make the deadline real: a published, national, state-by-state compliance measure.

Why It Matters, and What Would Fix It

The stakes are measured in a child's developmental time. A 90-day delay for a two-year-old is a meaningful fraction of the attachment window. An 18-month delay is the difference between a grandmother and a stranger. Interstate cases disproportionately involve exactly the children for whom kinship placement matters most — kids whose extended family has scattered across state lines for economic reasons, disproportionately poor and disproportionately Black and Native. And under the Adoption and Safe Families Act's own permanency clock, a state must generally move to terminate parental rights once a child has been in foster care 15 of the most recent 22 months. An 18-month ICPC home study can consume the entire statutory window in which the family had to reunify. The compact's delay does not merely postpone a placement; it can extinguish a family's legal existence.

Four fixes are available without amending fifty-two statutes:

Publish the data that already exists. NEICE generates a Violation Statistics Report today. HHS should require, as a condition of the Family First electronic-system mandate, that aggregate NEICE metrics — median and 90th-percentile days to home study, percent within 60 days, Regulation 7 timeliness, denial rates, and violation counts — be published quarterly by state. This costs approximately nothing. The query is already written.

Add an AFCARS element. A single mandatory field flagging ICPC status and referral date would let HHS, for the first time, connect interstate delay to permanency and safety outcomes in the data it already collects on every foster child in America.

Attach consequence to §471(a)(26). A Title IV-E state plan requirement without a compliance measure is decoration. Sustained failure to meet the 60-day standard, absent documented good cause, should trigger the same state-plan review process as any other IV-E violation — starting with the administrative-cost claims associated with the delayed cases.

Fund the receiving state and give families an appeal. Make interstate home studies a specifically reimbursable IV-E activity so the receiving state is not asked to do unpaid work, and give denied relatives a defined right to have the sending court review the denial on the record — which is precisely the reform child welfare attorneys have been requesting for over a decade.

The ICPC was written to make sure no child was shipped across a state line into a home no one had checked. Sixty-six years later, the compact's most reliable output is a child sitting in a stranger's house while a relative's file waits in a queue, or a child living in a home that was never checked at all because everyone gave up on the queue — with the federal reimbursement flowing uninterrupted through both scenarios, and no one, anywhere, counting.

Sources: - Interstate Compact on the Placement of Children: ICPC — Congressional Research Service, RL32070 - Foster Kids in Limbo: The Effects of the Interstate Compact on Children in Foster Care — ABA Child Law Practice - Can We Please Fix The Interstate Placement Of Children In Foster Care? — The Imprint - ICPC Home Study Evaluations and Decision Deadlines — NEICE Support Desk - NEICE Violation Statistics Report — NEICE Support Desk - NEICE Case Statistics Report — NEICE Support Desk - National Electronic Interstate Compact Enterprise (NEICE) — APHSA - NEICE Evaluation Final Report — Child Trends - Association of Administrators of the ICPC (AAICPC) — APHSA - ICPC Regulations (full text) — APHSA - Full Text of ICPC Regulation No. 7 — NY Office of Children and Family Services - ICPC Policy 0100-525.10 — Los Angeles County DCFS - Interstate Compact on the Placement of Children — Advokids - Title IV-E Programs Expenditure and Caseload Data 2024 — ACF - Child Welfare: Purposes, Federal Programs, and Funding — CRS IF10590 - Administrative Costs Drive Foster Care Claims — HHS ASPE - Family First Prevention Services Act — CRS IN10858 - Sen. Ossoff Releases Findings of 13-Month Probe into Safety of Foster Children - Following 13-Month Foster Care Investigation, Sen. Ossoff Introduces Bipartisan Bill — Foster Care Placement Transparency Act - Legal Aid and Dechert File Class Action Against NYS, NYC over Kin Foster Placement - Accountability in the Courtroom: Review of Child Welfare Litigation — Bipartisan Policy Center - Matter of D.L. v S.B., 2022 NY Slip Op 05940 — NY Court of Appeals - ICPC Does Not Apply to Out-of-State Placement with Parent (McComb v. Wambaugh) — ABA - A Report to Congress on Interjurisdictional Adoption — HHS Children's Bureau - American Public Human Services Association — ProPublica Nonprofit Explorer (EIN 36-2166948)