For months, Joan Alker and the Georgetown University Center for Children and Families have sounded the alarm over a purported crisis in children’s health coverage. Based on monthly administrative data from CMS and state Medicaid agencies, Alker wrote in May that roughly two million fewer children were enrolled in Medicaid and the Children’s Health Insurance Program (CHIP) in April 2026 than in January 2025. Her conclusion was absolute: “when child enrollment in Medicaid and CHIP goes down, the child uninsured rate goes up.”
Alker laid this drop squarely at the feet of the Trump administration and the One Big Beautiful Bill (OBBB). Because the OBBB made no changes to child eligibility rules — and its major Medicaid financing and coverage provisions have not yet been implemented — Alker offered a number of speculative mechanisms to explain the recent drop in Medicaid/CHIP enrollment: a “chilling effect” driven by enhanced immigration enforcement, “confusion” among families assuming prospective policy changes are already in effect, and a purported lack of federal enforcement of Medicaid/CHIP policies resulting in improper disenrollments.
As Brian Blase and I documented this past June, this narrative rests on a chain of unverified assumptions and disregards basic facts, such as:
- Alker’s analysis measures Medicaid/CHIP enrollment trends against a January 2025 baseline and ignores the stickiness of the COVID-era continuous coverage requirement that allowed the Medicaid rolls to massively expand as states were blocked from dropping ineligible enrollees. Although regular eligibility redeterminations were re-instituted in early 2023, even after well over a year of unwinding activity Medicaid/CHIP enrollment in January 2025 (79.4 million) remained roughly 8.0 million beneficiaries above pre-COVID levels (71.4 million). This suggests that part of the decline in Medicaid/CHIP enrollment after January 2025 reflects the continuing effect of removing accumulated improper and duplicate enrollees from the program.
- Shrinking Medicaid/CHIP rolls do not necessarily imply a loss of health insurance. On the contrary, leaving Medicaid/CHIP frequently indicates upward economic mobility, as improving household earnings and higher parental employment allow children to transition into employer-sponsored insurance (ESI). Moreover, these shifts from Medicaid/CHIP to private coverage often result in broader provider networks, lower wait times for services, and better access to care.
- Alker’s narrative ignores demographic trends that affect Medicaid/CHIP enrollment, specifically the sustained decline in U.S. birth rates that has reduced the national child population by roughly one million since 2020. As older youth age out of the child eligibility group and are replaced by smaller birth cohorts, total child Medicaid/CHIP enrollment naturally contracts.
- In January 2026, child enrollment in Medicaid/CHIP remained above its pre-pandemic level as a share of the total population aged 0-18. In January 2020, 45.3 percent of children were enrolled in Medicaid or CHIP, nearly identical to the average level of 45.4 percent from 2018 to 2019. In January 2026, that share was 47.3 percent, two percentage-points above the pre-COVID level.
- According to Alker, heightened immigration enforcement has deterred undocumented parents from enrolling their eligible children in Medicaid. If this hypothesis were true, states with larger undocumented populations should have experienced the steepest drops in child enrollment. In reality, there is virtually no correlation between the two variables.
Assessing Alker’s central claim — that the drop in child enrollment in Medicaid/CHIP was accompanied by an overall increase in the child uninsured rate across all sources of coverage — using hard data has been challenging because complete administrative data on private insurance enrollment is not available publicly. So while tracking the number of children on Medicaid/CHIP is easy with government administrative data, monitoring the other side of the ledger is more challenging.
National trends in private insurance coverage are mainly measured using household surveys, which are published with significant time lags due to the complexity of data collection and processing. Last week, the Census Bureau released 2025 estimates from the Current Population Survey Annual Social and Economic Supplement (CPS ASEC), a nationally representative survey of approximately 75,000 households. The CPS ASEC serves as the federal government’s official benchmark for measuring annual poverty, household income, and health insurance coverage — and it’s one of the first concrete tests of Alker’s arguments.
The 2025 CPS Data Shows No Deterioration in Children’s Coverage
Contrary to Alker’s theory, the 2025 CPS ASEC data did not show a spike in child uninsurance from 2024 to 2025.
Instead, the survey found small (but mostly statistically insignificant) declines in uninsurance across multiple groups:
- The estimated uninsurance rate for children nationwide fell from 6.1 percent in 2024 to 5.8 percent in 2025. The share of children on public health insurance (mainly Medicaid/CHIP) declined slightly year-over-year, but this was more than offset by an increase in private coverage. In absolute terms, the number of uninsured American children dropped by an estimated 226,000.
- Among children living below the federal poverty line — the cohort presumed to be most vulnerable to administrative barriers and supposed “chilling effects”—the uninsurance rate dropped from 10.3 percent in 2024 to 9.0 percent in 2025.
- Nationwide, the total number of uninsured Americans dropped by 270,000 from 2024 to 2025.
In short: the best available national data show no evidence of the deterioration in health insurance coverage that Alker predicted. Rather, coverage, including for children, likely expanded.
It is important to note that the definition of coverage in the CPS ASEC is fairly narrow. An individual is classified as uninsured in the CPS ASEC only if they lacked health insurance of any kind for the entire year; anyone covered for even a single month or day is counted as insured. This inevitably dampens the CPS ASEC’s sensitivity to within-year coverage churn and makes it more difficult to capture the point-in-time impact of policy changes.
But the basic takeaway from the new CPS ASEC data is corroborated by other federal surveys that measure current insurance coverage. Newly available 2025 data from the Behavioral Health Risk Factor Surveillance System (BRFSS), for example, reveals that 7.4 percent of adults were uninsured at the time of the survey in 2025, virtually unchanged from the 7.5 percent who reported being uninsured in 2024. The National Health Interview Survey (NHIS) echoes the same pattern: The proportion of adults aged 18-64 who reported being uninsured at the time of the interview dropped from 12.0 percent in the second half (July-December) of 2024 to 11.6 percent in the first half (January-June) of 2025 and 11.5 percent in the second half of 2025. And while these estimates do not capture children, the absence of a decline in the adult uninsurance rate from 2024 to 2025 cuts against Alker’s arguments because several of the mechanisms she proposes – including confusion over whether forthcoming OBBB policy changes are already in force and the federal government’s alleged attempts to increase administrative barriers to enrollment in the exchanges – would apply to adults as well as to children.
Looking Ahead: The Real Impact of the OBBB
The long-term impact of the OBBB on children’s health coverage cannot be fully evaluated today, as its primary financing and coverage provisions have yet to take effect. Yet when these reforms are eventually phased in, their net effect on children is likely to be modest — and quite possibly positive. By improving program integrity and instituting stronger mechanisms to prevent ineligible able-bodied, working-age adults from enrolling in Medicaid, the OBBB will ease severe provider shortages that have long plagued the program and denied adequate care to those who genuinely require a public safety net.
Constructive policy debate requires patience and empirical rigor, not extreme claims that conflict with the best available data.
