Thirty Hours and No License: The Unregulated Authority of CASA Volunteers and Guardians ad Litem in America's Dependency Courts
Investigation by Project Milk Carton
Thirty Hours and No License: The Unregulated Authority of CASA Volunteers and Guardians ad Litem in America's Dependency Courts
A volunteer with thirty hours of training — twelve in rural Nevada — can be the court's designated voice for a child in foster care. Judges adopt what that volunteer recommends in roughly four out of five cases. No exam, no license, no credentialing body exists anywhere in the system. We spent a week verifying every claim in that sentence. This is what the record actually shows.
By PMC Intelligence Desk
In every state in America, when a child is removed from home and lands in dependency court, the judge appoints someone to speak for that child's best interests. Federal law — the Child Abuse Prevention and Treatment Act — requires it in every substantiated abuse or neglect proceeding. The appointee is called a guardian ad litem, and the statute says that person "may be an attorney or a court appointed special advocate."
In much of the country, it is the second option: a CASA volunteer. Roughly 939 CASA and volunteer guardian-ad-litem programs operate in 49 states plus the District of Columbia. In 2023, by National CASA/GAL Association's own annual report, 79,000 volunteers served 201,000 children and youth.
Here is what it takes to become one.
The Thirty-Hour Floor
The National CASA/GAL Association's Standards for Local CASA/GAL Programs — effective January 1, 2025, updated March 25, 2025, the standard in force today — sets the national baseline. Standard 8.C.8 references "the 30 hours of pre-service training." Standard 8.C.9 requires 12 hours of continuing education annually. The volunteer definition requires only that the person be at least 21 years old, pass a background screening, complete the training, and be appointed by the court.
That is the entire national gate. The same floor appears in California Rule of Court 5.655 ("at least 30 hours of formal instruction"), in Indiana's court administration standards ("minimum of 30 hours of initial training"), and in Texas CASA's standards manual ("at least 30 hours of required pre-service training").
We searched the national standards, state court rules, and program manuals for anything more. No document examined contains a degree requirement. None contains a competency examination. None contains a license, and no state licenses CASA volunteers. There is no credential-revoking body anywhere in the system — the only removal path is dismissal by the local program itself. (One partial exception on the guardian-ad-litem side: New Hampshire operates the country's only actual GAL licensing board, under RSA 490-C.)
Two qualifiers, because precision matters. First, many local programs voluntarily exceed 30 hours — the accurate formulation is a 30-hour floor, not "30 hours is what every volunteer gets." Second, the floor has not always held even at 30: legal scholars Amy Mulzer and Tara Urs, quoting the field's own national evaluation, note that actual training delivered has historically ranged as low as three hours, with training continued during service.
And in one state the statutory floor sits far below the national one. Nevada's NRS 432B.505 requires just 12 hours of initial training — and 6 hours annually after that — for special advocates in counties under 100,000 population. That is the lowest statutory figure this investigation found anywhere in the country. Larger Nevada districts must meet National CASA standards; the rural counties, by statute, need less than half of them.
The Authority: Measured Exactly Once
How much weight does this volunteer's recommendation carry? The claim you will hear — that judges adopt CASA recommendations in the overwhelming majority of cases — is true. It is also resting on essentially one quantified study, and that study's data is more than two decades old.
The Caliber Associates national evaluation, prepared for the National CASA Association with Packard Foundation support, analyzed recommendation adoption in 1,994 cases across 25 CASA programs. The findings:
In 61.0 percent of cases, 100 percent of the volunteer's recommendations were accepted by the judge.
In another 22.6 percent, between 76 and 99 percent were accepted.
Total rejection occurred in 0.5 percent of cases.
"In more than four out of five cases," the evaluation concluded, "all or almost all of volunteers' recommendations were accepted." National CASA's own effectiveness page still relays the same conclusion today. No independent replication of that measurement has been published since. The judge-satisfaction surveys sometimes cited alongside it (93 percent positive) measure how judges feel about CASA — not how often they follow it. The two should never be conflated.
Buried in the same study are two findings almost nobody quotes: African American volunteers were 2.5 times more likely than white volunteers to have all of their recommendations accepted, and male volunteers were 4 times more likely than female volunteers. Nobody has explained either number, because nobody has re-run the study.
Investigation data visualization
So the machine, as measured: a volunteer clears a 30-hour floor, files recommendations on custody, placement, and reunification, and the judge adopts all or almost all of them about 84 percent of the time. The next question is whether the record shows those recommendations make children's lives better.
The Evidence Record: Commissioned by CASA, and Unflattering
The two largest quantitative evaluations of CASA in the child-welfare context were both commissioned or funded by CASA organizations themselves. Neither found what the movement hoped.
Caliber, 2004. The same national evaluation, using National CASA's own case data merged with a national survey of child and adolescent well-being. Its most cited figure: among the closed-case subsample, 89 percent of children with a CASA volunteer were placed in out-of-home care, versus 18 percent of children without one — a covariate-adjusted comparison, significant at p<.001. The study's own next sentence is the caveat that must always travel with that number: the finding "is likely attributed—at least in part—to remaining differences in who receives a CASA volunteer and who does not, despite attempts to control for this issue." CASA cases were rated more severe going in; covariates explained less than a fifth of the variance in who got assigned a volunteer. In the open-case subsample, children with CASAs were also significantly less likely to be reunified (13 percent versus 50 percent) and less likely to be placed with kin (8 percent versus 27 percent) — comparisons that were not significant among closed cases. The authors' bottom line was that only a random-assignment study could settle the question. Twenty-two years later, none has been done in dependency court.
Osborne, 2020.The University of Texas at Austin study — Osborne, Warner-Doe, LeClear and Sexton, published in Child Maltreatment — followed 31,754 children entering Texas state custody, using intent-to-treat analysis with inverse probability weighting. Funded by Texas CASA itself (disclosed in the paper; the funder had no role in design or analysis). The results, as odds ratios for children with a CASA versus without:
Any permanency: 0.81 (p = .002)
Reunification with family: 0.84 (p < .001)
Permanent kin guardianship: 0.78 (p = .003)
Adoption, if not reunified: 1.17 (p = .001)
Children with CASA volunteers were less likely to reach permanency at all, less likely to go home, less likely to end up with relatives — and more likely to be adopted out. The authors are careful: "This study does not provide insight into the reasons for the differences… it is likely that there remain unobserved differences between the groups." They do not claim CASA causes harm. But the direction of every family-preservation outcome runs the same way, in the largest study ever conducted, on a sample thirty times the size of anything before it.
The response is its own story. Dallas CASA's chief executive published a rebuttal calling the study flawed — arguing judges, not CASAs, decide dispositions. National CASA's current research page cites the same research team's 2019 paper on who gets assigned a CASA, but not the 2020 permanency paper. The Imprint reported the lead researcher's disappointment that some Texas CASA leaders "publicly disputed the results from the study they had helped design." Texas CASA's practical answer was programmatic: a pivot toward Collaborative Family Engagement, a family-finding model, rather than a contest of the findings.
The clearinghouse verdict. The California Evidence-Based Clearinghouse — the standard registry for child-welfare interventions — carries no scientific rating for CASA. The Campbell Collaboration registered a protocol for a CASA systematic review in 2015; no completed review has ever appeared. The peer-reviewed framing as of 2024: "assertions that CASA is an effective intervention for improving child welfare case outcomes are tentative at best, due to contradictory findings… and pervasive methodological limitations." Washington State's Institute for Public Policy flagged the same comparison-group problem back in 1998.
One asterisk, in fairness, because it is new: a 2025 randomized controlled trial of CASA in juvenile justice — adolescents on probation, not children in dependency court — found reduced recidivism and improved academic self-efficacy. It is the only published CASA RCT, and it measures neither safety nor permanency. The defensible sentence, as of today: no randomized trial in the child-welfare context has ever demonstrated that CASA improves child safety or permanency.
Sit with the asymmetry. Judges follow these recommendations roughly 84 percent of the time. The evidence that the recommendations help — after four decades and hundreds of millions of federal dollars — does not exist.
Race: The Corps, the Hours, and the Data Going Dark
Who are the volunteers? The Caliber national sample: 90 percent white, 79 percent female; 8 percent of volunteers were African American. The most recent public national survey, from 2016: 84.4 percent Caucasian, 82 percent female. American Indian and Alaska Native volunteers were 0.6 percent of the corps.
Then the data went dark. National CASA's 2023 Annual Report — 79,000 volunteers, 201,000 youth served — contains no volunteer race or gender breakdown at all. Newer survey data sits behind the association's member portal. An organization whose volunteers speak for a foster-care population that is disproportionately Black and Native stopped telling the public who its volunteers are.
Investigation data visualization
The hours finding is the one that should end any debate about whether this matters. From the Caliber evaluation's primary text: volunteers spent a mean of 2.67 hours on African American children's cases, versus 4.30 hours for children of other races — significant at p<.01, confirmed by both regression and t-test. The children with the least power in the system received 62 cents of volunteer attention on the dollar.
Mulzer and Urs, in the CUNY Law Review, put the structural point plainly: "Eighty to ninety percent of CASAs are white… This exercise of white supremacy is out in the open, obvious, direct. It is a part of the case—a party to the case."
Two findings we verified that cut in different directions — both belong in print. On appointment rates: the 2019 Texas study of 32,349 children found that, in raw numbers, Black and Hispanic children were significantly less likely to be appointed a CASA at all — but after case-mix controls, the race effects wash out; the study attributes the raw disparity to case characteristics, and it contains no separate Native American category. Anyone citing that study for a racial-targeting claim is overreading it, and we won't.
The Native record comes from a different and starker source. Fletcher and Fort's study of Indian Child Welfare Act transfer appeals, in the Boston University Law Review: across 125 ICWA transfer-appeal cases, guardians ad litem took an active position in 56. In 34 of those 56 — sixty-one percent — the GAL opposed transferring the case to tribal court. Three supported it. "The position of the GAL… appears to hold sway with the court," the authors write, concluding that GALs "continue to thwart the purposes of the ICWA." Set that against a volunteer corps that is 0.6 percent American Indian or Alaska Native, and the pattern needs no embellishment.
The Money: What Actually Funds the Volunteer
Start with a correction to the common citation, because we had to make it ourselves. 34 U.S.C. § 20110 — frequently cited as "the CASA funding statute" — is nothing of the kind. It is generic Victims of Crime Act administrative machinery; CASA appears nowhere in it. The actual CASA statute is 34 U.S.C. § 20323, from the Victims of Child Abuse Act of 1990: the Justice Department's juvenile-justice office "shall make grants to initiate, sustain, and expand" CASA, with a national organization authorized to pass money down through subgrants, and funded programs required to comply with National CASA standards. Congress reauthorized it through 2027.
The direct federal line has held at 15 million dollars in fiscal 2023 and 14 million in fiscal 2024 and 2025; the administration requested 9 million for fiscal 2026 and both chambers of Congress rejected the cut, each marking 14 million (the enacted figure tracks the Senate bill; we have not read the final joint explanatory statement, so treat the FY26 line as near-certain rather than confirmed). On top of the direct line sits the far larger Victims of Crime Act stream: by National CASA's own legislative page, more than 500 state and local programs in 40 states received over 83 million dollars in VOCA victim-assistance funding in 2020 — a self-reported figure, now six years stale, and the VOCA stream has deteriorated since.
Then there are the association's own books, from IRS filings via ProPublica's Nonprofit Explorer. National CASA's total revenue: 17.0 million dollars in 2021. 16.8 million in 2022. 9.0 million in 2023. 5.4 million in 2024. A 68 percent collapse in three years, as federal money receded — with contributions making up 89 to 97 percent of revenue every year. Chief executive compensation over the same window ran from 519,966 dollars (2021) down to 305,593 dollars (2024).
For scale: Texas CASA Inc alone — the state association — runs 39 to 40 million dollars a year, roughly seven times the national association's current size, on state money flowing through the Texas Health and Human Services Commission (31.9 million dollars for the 2024–25 biennium; Texas CASA passed 33.3 million to local programs in fiscal 2025). Which is why, when Washington pulled the plug in 2025, Texas shrugged — "not a primary source of any Texas CASA funding," its release said — and other states did not.
One Hundred Days in 2025: The Defund and the Price of Reinstatement
On April 22, 2025, the Department of Justice issued more than 360 grant termination notices across the justice-grant world. National CASA's three federal grants were among them, terminated for no longer "effectuat[ing]… agency priorities." Within two days, all federally funded services and subaward disbursements were suspended. By National CASA's own accounting, 101 current subgrants to local programs were terminated and 275 anticipated grants went unawarded. (The total dollar value reported varies by source — roughly 25 million per The Imprint, versus about 48.9 million in targeted award value on a DOJ list obtained by Reuters — likely an annual-versus-multi-year measurement difference; we cite both with their sources rather than pick one.)
What happened during the appeal is the part that deserves to be remembered. Beginning May 15, 2025, as The Imprint reported, CASA leadership instructed program trainers to "Skip or Adapt as Directed" dozens of sentences across slide decks, a volunteer manual, and facilitator guides — skipping references to racial, class, and gender diversity, and changing "LGBTQ" to "LGB." On August 6, 2025, DOJ granted the appeals and restored all three grants, finding the activities "align sufficiently with the Department's priorities, including protecting American children." An OJJDP spokesperson, again via The Imprint, said National CASA then received 25.3 million dollars in new federal grants "with special conditions." (The Imprint's site blocks automated retrieval; those specifics rest on verified search records of its reporting and should be read there directly.)
Note what National CASA did not do: it is not a plaintiff in the litigation over the April terminations. It won through the administrative channel — by editing the training.
Read the two threads of this story together. The evaluation record says the volunteer corps is 84 to 90 percent white and spends significantly less time on Black children's cases. The price of restoring federal funding in 2025 was stripping bias-awareness content from the training of that same corps. As of July 2026, funding is restored and holding: grantmaking is running normally, spring volunteer-recruitment grants are flowing, California's 2026–27 budget restored 20 million dollars for 44 local programs a year after eliminating it — the state serves only about 15 percent of eligible youth — while Oregon carries backfill bills and West Virginia's governor has line-item vetoed CASA funding two years running.
Investigation data visualization
Our Own Ledger: The $97.7 Million Pass-Through Layer
Project Milk Carton maintains a database of more than 340 million records on the child-welfare money trail — federal grant streams, nonprofit tax filings, and missing-children data. For this investigation we ran CASA through all of it. Three results, and the shape they make together is the finding.
In the HHS TAGGS grant stream — 23,000 awards — CASA appears zero times. No CASA entity, no court-appointed-special-advocate entity, no guardian-ad-litem entity.
In the 1.05-million-row unaccompanied-children subgrant stream: zero rows. CASA is entirely absent.
In our 630,000-row Form 990 Schedule I data — nonprofits reporting grants they made to other nonprofits, tax years 2021 through 2024: 97.7 million dollars across 1,332 grants to 753 distinct CASA recipients.
The first two zeros make sense once you know the statute: CASA is Justice-funded, not HHS-funded. But the third number is the architecture made visible. CASA's money lives almost entirely in the nonprofit-to-nonprofit pass-through layer: state associations regranting public money down to local programs, topped up by donor-advised-fund retail giving. The largest "grantors" to CASA organizations in our data are CASA organizations — Texas CASA Inc alone regranted 26.5 million dollars across 62 grants; the California CASA Association 21.6 million; Illinois 8.0 million; National CASA itself 5.1 million across 173 grants — followed by Fidelity Charitable, Schwab, and their peers. The top single recipients are the big-city locals: CASA of Los Angeles County (4.9 million), Dallas CASA (4.3 million), CASA of Orange County (2.6 million).
Slice it by state and the imbalance is stark. Texas: 195 grants, 31.1 million dollars — 85 percent of it Texas CASA's own regranting. Nevada: 8 grants, 408,704 dollars — a 76-to-1 ratio against a state whose rural counties, remember, run on the 12-hour statutory training floor. Six of Nevada's eight grants came from the Nevada CASA Association itself; the state's association network covers 10 counties, and Washoe-area CASA materials indicate only about a fifth of Nevada's foster children have a CASA volunteer at all. (Our totals are conservative undercounts — we manually excluded roughly 25 million dollars of false-positive "casa" matches, mostly Spanish-language organization names and a housing nonprofit.)
This architecture explains something the 2025 crisis proved empirically: a network funded through a single national association's federal grants can be frozen overnight by a single agency decision. Concentration of money is concentration of fragility — and of control. The training edits of May 2025 propagated through the entire national network for the same reason the April termination did: everything runs through one pipe.
The Accountability Gap: Immune, Undisciplined, Unaudited
What happens when a volunteer or guardian ad litem gets it wrong — or does wrong?
In most of the country: nothing a family can reach. The majority rule in American courts gives GALs absolute quasi-judicial immunity for acts within their judicial duties. The Eighth Circuit reaffirmed it as recently as August 2024 in Arseneau v. Pudlowski — a civil-rights suit alleging one guardian ad litem misrepresented her qualifications and another blocked abuse reporting, prevented a therapist from testifying, and threatened a mother with loss of custody unless she settled. The court never reached whether any of that happened. Dismissal affirmed: GALs are immune "even if their conduct is wrongful or illegal."
The rule is cracking, slowly, state by state. Maryland's highest court stripped GAL malpractice immunity in 2006 (Fox v. Wills — reversing the intermediate ruling still sometimes cited as if it were law; the legislature later restored partial protection by statute). And in May 2024, the Pennsylvania Supreme Court held in N.W.M. v. Langenbach that dependency GALs get no quasi-judicial immunity from legal malpractice claims at all: "they do not operate as an arm of the court… Like adults, children deserve competent representation, as well as recourse if they do not receive it." Meanwhile, statutory shields run the other way — Washington deems dependency GALs "officers of the court for the purpose of immunity"; Texas immunizes them except for recklessness, bad faith, malice, or gross negligence.
Where discipline systems nominally exist, the record is thin to the point of vanishing. Investigate West's March 2026 investigation of Washington State found that of 31 grievances filed against GALs in the state's six largest counties across 2024 and 2025, only 4 were found to have merit — against more than 7,100 family-law GAL appointments statewide since 2020. The documented cases read like an argument for the Pennsylvania rule: a Thurston County GAL whose report a review board found lacked "independence, objectivity, or…fairness" — discipline was corrections plus two hours of training, and the flawed report stayed in the case file. A Clark County GAL who threatened a client and disclosed her confidential medical information to her employer — admonished, still working. A King County GAL who altered a medical release and fabricated a therapist quote — remained on the registry.
The state-audit shelf says the same thing in institutional language. Minnesota's Legislative Auditor, 2018: the state GAL program "has not assigned guardians ad litem to all cases for which they are required" — a statutory-mandate violation — while its board "provided limited direction and oversight." Maine's legislative watchdog, 2006: GAL reimbursements up more than 585 percent over a decade with insufficient oversight at every stage, such that "judges in many cases could not be confident they were receiving complete and accurate information"; real reforms took until 2013. Connecticut's legislature, after testimony about uncapped GAL fees bankrupting parents, passed reform unanimously in 2014.
And the federal paper trail runs through National CASA itself. A 2007 Justice Department Inspector General audit of 88.9 million dollars in National CASA grant spending found improper indirect-cost allocation and — the finding that rhymes with everything above — an absence of outcome-based performance measures. In 2022, the Office of Justice Programs designated National CASA a high-risk grantee over "significant issues with… financial management practices," after monitoring flagged roughly 2.7 million dollars in questioned costs from 61.8 million in grants — office leases, a Las Vegas public-relations firm, recruitment and trademark fees — nearly all of it since resolved or repaid, per The Imprint's reporting.
After four decades and hundreds of millions of federal dollars, we could locate no GAO audit of CASA's effectiveness. Not a critical one. None.
Investigation data visualization
The Alternative With Actual Evidence: Lawyers
The deepest structural development of the last seven years is that federal policy now pays for the thing CASA is not: attorneys.
In January 2019, the Children's Bureau changed its policy manual to let state Title IV-E agencies claim 50 percent federal matching funds for attorneys representing children and their parents in child-welfare proceedings. In May 2024, the Administration for Children and Families codified and expanded that rule at 45 CFR 1356.60 — covering independent legal representation for foster children, parents, relative caregivers, and Indian custodians. Per reporting in The Imprint, 31 states drew down roughly 300 million dollars for family-court lawyers between 2018 and 2023.
Unlike the volunteer model, counsel has randomized and quasi-experimental evidence behind it. The HHS-funded QIC-Child Rep trials in Georgia and Washington found children with specially trained attorneys were 40 percent more likely to reach permanency within six months. Washington's 2021 statewide pilot — 434 children with assigned counsel versus comparison counties — found children 45 percent more likely to be reunified with their families, with 30 percent fewer placement changes, 65 percent fewer school transitions, and over a million dollars in estimated savings; the state's official December 2025 evaluation confirms those figures (while reporting the statewide phase-in is on hold, with the remaining counties pushed to 2030–2032).
The professional bodies have chosen. The National Association of Counsel for Children's standards now state that a guardian ad litem, when appointed, should always be an attorney — and that a CASA "is not a substitute for legal representation." New scholarship goes further: a Texas Law Review article argues CASA courtroom opinions on parental fitness are, functionally, expert testimony from untrained witnesses, and should be excludable on due-process grounds.
The sharpest statute in the country may be New Hampshire's SB 463, effective January 2025: it grants a child expressed-interest counsel precisely when the child's own wishes conflict with the CASA or GAL's best-interests position. Read that twice. The legislature wrote into law the recognition that the volunteer speaking for the child and the child may be adverse parties. Right-to-counsel laws have now been established or expanded in Alaska, Montana, North Dakota, New Hampshire, and Washington; South Carolina's current session carries a cluster of GAL-accountability bills, including one placing GALs under bar discipline; West Virginia's legislature has pressed its supreme court on GAL supervision rules four sessions running; and a proposed federal bill, Senator Wyden's BRIDGES for Kids Act, would require legal representation for foster children nationwide (reported; bill text not yet independently verified).
What the Record Shows
Strip away everything contested and here is what remains, each item verified against primary sources:
A volunteer with 30 hours of training — 12 in rural Nevada — serves as the court's designated best-interests voice for a child, with no exam, no license, and no credentialing body anywhere in the system.
The only measured judicial adoption rate makes that voice close to dispositive — all or almost all recommendations adopted in roughly 84 percent of cases — and the measurement is more than twenty years old.
Both flagship evaluations commissioned or funded by CASA's own institutions, associate CASA with less reunification and less kinship care. No clearinghouse rates CASA evidence-based. No randomized trial in dependency court has ever been run.
The corps is overwhelmingly white and female as far as public data shows; volunteers logged significantly fewer hours on Black children's cases; in ICWA appeals, active GALs opposed tribal-court transfer 34 times out of 56; and National CASA stopped publishing volunteer demographics.
Federal money built and sustained the network, yet the federal oversight record on National CASA itself includes a 2007 audit finding no outcome-based performance measures and a 2022 high-risk-grantee designation — and no GAO effectiveness audit exists.
The 2025 termination-and-reinstatement showed both the network's total federal dependence and the price of readmission: cutting bias-awareness training for a corps this homogeneous, serving families this disproportionately non-white.
In our own 340-million-record database, CASA's money moves through a 97.7-million-dollar nonprofit pass-through layer, invisible to the HHS grant streams — an architecture that concentrates power, fragility, and control at the association level.
Accountability is diverging by state: absolute immunity reaffirmed federally in 2024, stripped in Pennsylvania the same year; where discipline exists, Washington's numbers — 4 merited grievances out of 31, against 7,100 appointments — show what it is worth.
The alternative with actual randomized evidence — counsel for children — is federally funded since 2019 and codified since 2024. The best state statutes now treat the volunteer's position and the child's own wishes as potentially adverse, requiring separate representation.
None of this says a CASA volunteer cannot be the one adult who shows up for a child when no one else does. Many are exactly that, and the hours they give are real. The question this record poses is not about the volunteers. It is about a system that hands near-dispositive courtroom authority to laypeople on a 30-hour floor, measures nothing, disciplines almost no one, publishes less each year, and has spent four decades of public money declining to find out whether it works. Children in dependency court get one childhood. The adults deciding where they spend it should be able to show their work.
The pending NEICE mandate is often presented as a solution. NEICE, the National Electronic Interstate Compact Enterprise, is designed to modernize ICPC transmission and reduce paper-based delays. Digital transmission matters, and full implementation is better than fax machines, mail bins, and scattered local processes. But digitizing a broken process is not the same as fixing it. If the underlying compact still lacks enforcement, penalties, staffing requirements, family remedies, and real-time accountability, then NEICE may simply move files more efficiently into the same queues. A digital bottleneck is still a bottleneck.
The ICPC crisis should force a much bigger question: why does the system make it so hard to place children with family while making it so easy to keep them in paid care? The answer is not one villain. It is design failure. It is legacy law. It is federalism without accountability. It is underfunded administration. It is risk avoidance. It is provider economics. It is the absence of a private right of action. It is decades of reform language without enforcement teeth. It is the oldest story in child welfare: the state says family matters, then builds a process that treats family as a paperwork problem.
A serious reform agenda would begin with enforceable deadlines. The 60-day home study requirement should not be a polite suggestion. If a receiving state misses the deadline without documented good cause, there should be automatic escalation, public reporting, and financial consequence. If the delay continues, relatives should have a direct administrative appeal and a fast-track court remedy. A compact that can stop a child from moving should also give the child and relative a way to force the government to act.
Second, the ICPC should include a child-centered emergency placement pathway for vetted relatives. When a relative has already passed background checks, has an existing relationship with the child, and meets basic safety criteria, the law should allow provisional placement with supervision while the full home study is completed. The system already takes risks by placing children with strangers in emergency settings. It should not pretend that relatives are uniquely dangerous simply because they live across a state line.
Third, Congress and the states must fund ICPC offices like critical infrastructure. A state with thousands of foster care cases cannot run interstate placement through one or two overburdened staff members and then act shocked when children wait. Staffing ratios should be published. Caseloads should be published. Average processing times should be published. Returned packets should be tracked. Clock restarts should be tracked. Every state should have a public ICPC dashboard showing how many children are waiting, how long they have waited, and where the file is stuck.
Fourth, the law must close the clock manipulation loophole. The 60-day period should not begin only after the receiving state accepts a completed referral while the sending state can take weeks or months to prepare it. There should be deadlines on both sides. The sending state should have a fixed period to identify relatives, prepare the referral, and transmit the packet. The receiving state should have a fixed period to accept, reject, or request specific missing information. If a packet is returned as incomplete, the reason should be documented, time-limited, and reviewable.
Fifth, ICPC delays should trigger placement cost review. If a child remains in congregate care or high-cost foster care solely because of interstate paperwork delay, the state responsible for the delay should bear financial consequence. Right now, the cost of delay is diffused across systems, while the child pays emotionally. Accountability should follow the bottleneck. If a state’s failure to process a home study causes months of unnecessary paid placement, that failure should show up in the budget.
Sixth, federal child welfare reviews should include ICPC compliance as a core performance measure. It is absurd to review state child welfare performance while allowing interstate family placement delays to remain a side issue. If a state cannot move children to safe relatives across state lines within legal timelines, that is a permanency failure. It should affect federal oversight, corrective action plans, and funding scrutiny.
Finally, families need rights, not just promises. Relatives who step forward should receive notice, status updates, reasons for delay, and access to review. Children should have counsel empowered to challenge ICPC delay when it harms permanency. Parents should be able to argue that the state cannot move toward termination while kinship placement is delayed by government inaction. No child should lose family because two state agencies could not move paperwork on time.
The ICPC began as a protective reform. That origin matters because it explains why the system has been so hard to challenge. Nobody wants to be accused of weakening safeguards for children. But after sixty-six years, the question is no longer whether children need protection across state lines. Of course they do. The question is whether the existing compact protects them from harm or creates a new kind of harm through delay, drift, and family separation. A child left in congregate care for six months while a safe relative waits is not being protected by paperwork. That child is being failed by it.
The phrase “fall through the cracks” is too soft for what is happening here. Children are not falling through cracks. They are being held in a pipeline with known failure points, known delays, known staffing shortages, known enforcement gaps, and known financial consequences. When a system knows where children get stuck and still does not build a release valve, the problem is not accidental. It is tolerated.
This is why the ICPC crisis deserves national attention. It is not as visible as a courtroom termination. It is not as emotionally packaged as an adoption ceremony. It is not as politically explosive as a removal caught on video. But it may be one of the most consequential hidden mechanisms in the child welfare system because it decides whether family can actually function as family when state lines are involved. In a mobile country where relatives often live in different states, a slow interstate placement system is a family-separation engine.
The children caught in this engine do not need another conference panel. They do not need another best-practices memo. They do not need another voluntary compact reform stalled for twenty years. They need enforceable timelines, funded staffing, transparent dashboards, provisional kinship pathways, legal remedies, and consequences for states that ignore deadlines. They need a system that treats family placement as urgent, not optional.
The ICPC was built to stop children from being sent across state lines without protection. But in 2026, the greater danger is that children are being kept from family without accountability. A compact designed to protect children has become a maze that delays them. A law built for paper files is governing digital-age families. A process created to ensure safety is too often producing separation.
And the most damning fact remains the simplest: when states miss the deadline, nothing happens to the states. Something happens only to the child.
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 or donate at
Project Milk Carton Inc. Your gift keeps our site alive, spotlighting high-risk children!
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
A Note on Method
This article is built on a Project Milk Carton OPUS investigation completed July 19, 2026 — seven parallel research tracks, including four adversarial-verification squads whose job was to break every load-bearing claim against primary sources. Five claims from our own earlier research pass failed or degraded under that verification and were corrected before writing: the Caliber 89/18 figure carries its subsample and selection caveats above; 34 U.S.C. § 20110 is not the CASA funding statute (§ 20323 is); a 2003 Maryland immunity ruling still circulating was reversed in 2006; the 2025 training edits occurred during the appeal, months before reinstatement; and "no RCT ever" now requires the dependency-context qualifier. Facts sourced to The Imprint, whose site blocks automated access, are attributed inline. Database figures come from Project Milk Carton's civicops research database and are conservative undercounts. The full sourced report, including the verification log, is available at projectmilkcarton.org.
Sources
National CASA/GAL Association, Standards for Local CASA/GAL Programs (eff. Jan. 1, 2025)
Nevada Revised Statutes 432B.500, 432B.505
California Rule of Court 5.655; Texas Family Code §§ 107.011, 107.012, 107.009
Caliber Associates, Evaluation of CASA Representation: Final Report (data ca. 2002)
Osborne, Warner-Doe, LeClear & Sexton, "The Effect of CASA on Child Welfare Permanency Outcomes," Child Maltreatment 25(3) (2020)
Osborne, Warner-Doe & Lawson, "Who Gets a CASA?", Children and Youth Services Review 98 (2019)
Mulzer & Urs, "However Kindly Intentioned: Structural Racism and Volunteer CASA Programs," 20 CUNY L. Rev. 23 (2016)
Fletcher & Fort, "Indian Children and Their Guardians ad Litem," B.U. L. Rev. Annex (2013)
34 U.S.C. §§ 20323, 20110; 45 CFR 1356.60; 89 FR 40400 (May 10, 2024)
IRS Form 990 filings, National CASA (EIN 91-1255818) and Texas CASA (EIN 75-2252358), via ProPublica Nonprofit Explorer
National CASA/GAL Association public statements, April 23 – August 6, 2025; The Imprint reporting on the 2025 terminations, training edits, reinstatement, and OJP high-risk designation
DOJ Office of the Inspector General, Audit Report 07-04 (2007)
Arseneau v. Pudlowski, No. 23-2265 (8th Cir. 2024); N.W.M. v. Langenbach, No. 25 EAP 2022 (Pa. 2024); Fox v. Wills, 390 Md. 620 (2006)
InvestigateWest, "Washington courts rarely discipline guardians ad litem accused of misconduct" (Mar. 23, 2026)
Minnesota Office of the Legislative Auditor, Guardian ad Litem Program Evaluation (2018); Maine OPEGA, Guardians ad Litem report (2006); Conn. P.A. 14-3 (2014)
QIC-ChildRep research findings, University of Michigan; Washington State Center for Court Research, Children's Representation Program reports (2021, Dec. 2025)
NACC Recommendations; N.H. SB 463 (2024); N.H. RSA 490-C
Bates, "Procedural Due Process and the Improper Expert Opinions of Court-Appointed Special Advocates," 103 Tex. L. Rev.
Project Milk Carton civicops research database: Form 990 Schedule I (630K rows), HHS TAGGS (23K rows), USAspending UAC subgrants (1.05M rows)
Project Milk Carton
501(c)(3) Nonprofit • projectmilkcarton.org • t.me/ProjectMilkCarton
When I worked at the Court of Appeal I wanted to volunteer for CASA. I learned that they probably wouldn’t take me because of my employment with the court. I never applied.
I worked in the chambers of Justice Pauline Davis Hanson for six weeks. She was a juvenile court judge in her career before becoming an appellate court justice. She was a fine judge and cared very much about children. Everyone loved her and we all took her death very hard.