How Will the 2025 Reconciliation Bill Affect the Uninsured Rate in Each State?: Allocating CBO’s Partial Estimates of Coverage Loss
House Republicans are currently considering a reconciliation package that would make significant changes to Medicaid and the Affordable Care Act (ACA) Marketplaces. While the House version is evolving rapidly and there are not yet public estimates of the effect of the uninsured across all pieces of the legislation, the Congressional Budget Office (CBO) estimates that the part of the legislation passed by the Energy and Commerce Committee (E&C) could increase the number of people without health insurance by 8.6 million, due largely to changes to Medicaid and the ACA.
Additionally, these legislative changes come at a time when enhanced premium tax credits for ACA Marketplace enrollees are set to expire later this year. When combining the E&C provisions with the effect of the expected expiration of the ACA’s enhanced premium tax credits, CBO expects 13.7 million more people will be uninsured in 2034.
This analysis apportions the increase in the number of uninsured across the 50 states and DC and shows that number as a percentage of each state’s population. The number of newly uninsured as a percent of the population is equivalent to the percentage point increase in the 2034 uninsured rate. Nationally, CBO projected an uninsured rate of under 10% in 2034 under current law, which assumed the enhanced ACA premium tax credits would expire. The analysis here includes two maps: one showing the effect of the E&C provisions in the House Reconciliation package, and another showing those effects combined with expiration of the ACA enhanced premium tax credits.
Because CBO projections for the entire package are not yet available, the maps below do not include any increases in the uninsured rate that could arise from provisions not included in the E&C score. For example, Ways and Means Committee provisions that would require pre-enrollment verification of ACA eligibility – effectively ending automatic renewals of coverage – or remove repayment limits on excess premium tax credits could further increase the number of people without insurance coverage. The estimates are also likely to change as the reconciliation bill works its way through Congress, as CBO finalizes its estimates and with amendments to the bill itself.
Anticipating how states will respond to changes in Medicaid policy is a major source of uncertainty in CBO’s cost estimates. Instead of making state-by-state predictions about policy responses, CBO estimates the percentage of the affected population that lives in states with different types of policy responses. For example, in the E&C bill, Medicaid work requirements account for nearly half of the federal savings on Medicaid, suggesting they may contribute to the largest loss of insurance coverage in CBO’s estimates. However, different states might choose to implement a work requirement with reporting requirements that are easier or harder for enrollees to comply with. Reflecting the uncertainty, this analysis illustrates the potential variation by showing a range of enrollment effects in each state, varying by plus or minus 25% from a midpoint estimate.
The interactive tables at the end are sortable by state and size of coverage loss.
The biggest increase in uninsured by percentage point is in DC (5 percentage points), which also has expanded Medicaid eligibility up to 215% of poverty. 10 states and DC would have increases in their uninsured rates of 3 percentage points or more (DC, Washington, Oregon, Kentucky, New York, Louisiana, New Mexico, Connecticut, Illinois, Rhode Island, and California).
In terms of increases in the number of uninsured people, California and New York are the top two states (1.4M and 810k, respectively). Florida, Illinois, and Texas would follow at 520K, 430k, and 430k, respectively.
The combined effects of the House Reconciliation package with the expiration of the ACA enhanced tax credits, compared to a scenario where the enhanced subsidies are in place and the proposed integrity rule was not in effect, results in the greatest increases in Florida, Louisiana, Georgia, Mississippi, and Washington, where the uninsured rate is expected to increase by at least 5 percentage points. 30 states and the District of Columbia may see an increase in their uninsured rates of 3 percentage points or more.
About half (46%) of the 13.7 million more people who would be uninsured in this scenario live in Florida (1.8M), Texas (1.6M), California (1.5M), New York (800k), and Georgia (610k). Texas (2.8M Marketplace growth), Florida (2.8M) and Georgia (1.0M) experienced the most ACA Marketplace growth since 2020, the year before the enhanced premium tax credits became available.
Methods |
| This analysis first separates the number of newly uninsured people into two groups: those newly uninsured because of changes in Medicaid (7.7 million) and those newly uninsured because of changes in the Affordable Care Act exchanges (0.9 million or 6.0 million depending on the scenario). Increases in the number of uninsured by policy change were sourced from Congressional Budget Office (CBO) estimates.
Changes in Medicaid CBO estimates that changes in Medicaid from the Energy & Commerce reconciliation bill are expected to trigger two types of health insurance loss. First, an estimated 10.3 million people are expected to lose Medicaid. Second, an estimated 1.4 million people are expected to lose coverage provided to immigrants regardless of immigration status through programs financed entirely by the states. KFF uses the ratio of those numbers to first allocate the newly uninsured population (7.7 million) to Medicaid or state-funded coverage categories. This analysis allocates the newly uninsured population stemming from a loss of Medicaid across the states proportionally to each state’s estimated federal funding loss. In a prior analysis, KFF estimated how the federal Medicaid cuts would be allocated across the states using prior modeling work and state-level data. Data sources include:
The analysis allocates the newly uninsured stemming from a loss of state-funded coverage across the states proportionally to the federal spending reductions resulting from the new penalty on ACA expansion states that offer state-funded coverage. Changes in the ACA Marketplaces The effects of the E&C’s codification of the Trump administration’s proposed program integrity rule are distributed across state based on a table in the proposed rule. CBO estimates that 1.8 million more people will be uninsured if the proposed rule is codified than if these policy changes do not go into effect. However, because the policy changes have already been proposed through regulation, CBO assigns half of the effect of codifying the proposed rule (900,000 increase in uninsured) to the reconciliation legislation. Therefore, this analysis considers only half of the estimated state-level impact from the proposed rule for Figure 1 (900,000), and the whole effect in Figure 2 (1.8 million). In both figures, CBO totals of the estimated newly uninsured people from the integrity rule are distributed across states that exceed the average take-up rate of low-income people, as shown in Table 15 of the proposed rule. To estimate the effect of the expiration of the enhanced premium tax credits by state, CBO’s estimate of the increase in the uninsured population (4.2 million) is proportionally allocated to each state based on their growth in the ACA Marketplaces from 2020 to 2025, obtained from the Open Enrollment Period Public Use Files. Population Estimates Decennial state-level population projections from the Weldon Cooper Center for Public Service are used to interpolate the population in 2034 assuming compound population growth. The percentage point increase of the uninsured population per state reflects the estimated increase in the uninsured as a share of the projected population. The total impact from all changes were aggregated then rounded to two significant figures, with the percentage point increase in the uninsured population rounded to the nearest whole number. |
