a. Impacts on Medicaid Enrollment and Benefit Expenditures

This IFC implements the statutory definition of applicable individuals at §§ 435.119 and 435.551 to describe Medicaid applicants and beneficiaries who must demonstrate community engagement as a condition of their Medicaid eligibility. With certain exclusions specified at § 435.554, applicable individuals include those who are eligible for, or enrolled under, the State plan adult group described in section 1902(a)(10)(A)(i)(VIII) of the Act and § 435.119, and individuals eligible for or enrolled in coverage under section 1115(a)(2) expenditure authority providing MEC who meet the other criteria in statute.

This IFC specifies the steps States must take to assess and verify compliance, established at §§ 435.556 and 435.557, with the community engagement requirement at application and renewal; renewals for most beneficiaries who are subject to the community engagement requirement occur once every 6 months. At § 435.556, the IFC describes the statutory requirement that, as a condition of eligibility at renewal, States must require applicable individuals to demonstrate or be deemed to demonstrate community engagement for at least 1 month since the individual's most recent eligibility determination or redetermination, though States may elect to conduct more frequent verifications and/or require more than 1 month of compliance or deemed compliance.

To assess the impact of the new community engagement requirement on Medicaid enrollment and benefit expenditures, we estimate these impacts based on the President's Fiscal Year 2027 Budget (PB 2027) Medicaid enrollment and expenditure projections, with FY 2023 as the base year for actual observed data and exclude the impact of any Medicaid provisions of the WFTC legislation. Several provisions of the ( printed page 33456) WFTC legislation are expected to have effects on Medicaid enrollment and expenditures. We have excluded those other effects from this analysis to present clearly the anticipated impacts of the community engagement requirement on Medicaid. Projected national total Medicaid enrollment and expenditures for the adult group, which includes both newly eligible and not-newly eligible subgroups, are as follows in Table 36. We project enrollment will increase at an average rate of 0.7 percent per year and expenditures will increase at an average rate of 6.6 percent per year.

The community engagement requirement established by section 71119 of the WFTC legislation has not previously been implemented as a condition of receiving coverage under the Medicaid State plan. Prior to this IFC, States could only impose such requirements through section 1115 demonstrations. The limited section 1115 demonstration experience that exists involved different implementation patterns, including reinstatements following terminations of eligibility and self-selected enrollment populations, that are not directly applicable to estimating the impact of mandatory requirements applied to an existing State plan enrollment. We have not relied on these previous demonstrations for data or assumptions used in this analysis. Accordingly, there is no direct historical experience from which to derive empirical estimates of how many enrollees will not meet the requirement, the community engagement implementation policies States will adopt, or how verification systems will perform in practice.

The primary data challenge in developing these estimates is that the characteristics most relevant to this analysis—work status, educational enrollment, disability status, caretaker responsibilities, and incarceration history—are not currently captured in Medicaid administrative data. As a result, we relied on a combination of external survey data, Medicaid administrative data, and published research to develop the key assumptions underlying these estimates. Each of these sources has limitations that are discussed further in the limitations and caveats section below.

A second source of uncertainty is that the estimates depend heavily on State implementation choices that are not yet known. For example, the statute establishes minimum requirements for compliance checks at application and renewal, with at least 1 month of demonstrated compliance between redeterminations, but it gives States discretion to require more frequent verifications and more months for beneficiaries to demonstrate compliance. These choices will affect both the noncompliance and procedural disenrollment rates. Similarly, States have discretion over whether to adopt the option to consider short-term hardship events, described at § 435.555, and the extent of adoption will affect the share of enrollees subject to the requirement at a given time. We model a range of scenarios to reflect this uncertainty, but the actual distribution of State choices is unknown.

Finally, the estimates in this section do not capture potential behavioral responses, such as increased workforce participation or educational enrollment in response to the requirements, or the extent to which individuals who lose Medicaid coverage may obtain alternative coverage. Both factors could partially offset the projected coverage losses, but neither is quantifiable with available data.

Enrollment and per-enrollee expenditure projections are drawn from PB 2027, as described previously. Total computable and Federal per-enrollee expenditures are trended separately. Federal per-enrollee figures reflect current law FMAP rates throughout the ( printed page 33457) projection period. For purposes of the underlying cost model, the adult group is separated into newly eligible and non-newly eligible subgroups. The FMAP differs for these subgroups; projecting expenditures separately allows the projections to apply the appropriate average FMAP to each subgroup. The distinction between newly eligible and non-newly eligible adult group drives the Federal and State cost split. All other assumptions in the analysis are identical for newly eligible and non-newly eligible adult group.

We do not assume any change to the average per-enrollee costs for the remaining enrolled population relative to current projections. There are some reasons per-enrollee costs for the remaining population may be lower than for those that lose coverage. Individuals meeting the community engagement requirement may be healthier on average than those that lose coverage, as individuals not working or active in community engagement may have poorer health, making it harder to participate. In addition, younger individuals are more likely to meet the requirement based on already established activities, such as being enrolled in a full-time education program. If a higher proportion of younger individuals meet the community engagement requirement and, as a result, remain enrolled than the proportion of older individuals that maintain coverage, that may contribute to lower average costs per enrollee. However, there are also some reasons that the costs of those remaining could be higher. Most notably, individuals experiencing a short-term medical hardship (including inpatient hospital or nursing facility care) and those determined medically frail likely have substantially higher health care costs than others in this eligibility group, and they would remain covered in Medicaid under this new requirement. Given the factors that could lead to lower or higher per-enrollee costs, and the uncertainty of their relative magnitude, we made no adjustment to the average per enrollee cost for those losing coverage or retaining coverage under section 71119 of the WFTC legislation.

To develop this analysis, we started by determining how many enrollees may be subject to the community engagement requirement under this rule. We started with the projection of the number of enrollees who would be: (1) a newly eligible adult made eligible under 1902(a)(10)(A)(i)(VIII); (2) a non-newly eligible adult otherwise not eligible under 1902(a)(10)(A)(i)(I) through (VII); and (3) an adult eligible under a section 1115 demonstration who would otherwise be eligible under 1902(a)(10)(A)(VIII). Those projections are shown in Table 34.

Several groups of enrollees would have mandatory exceptions from the community engagement requirement, which include individuals under age 19, individuals also entitled to or enrolled in Medicare Part A and/or Part B, and those described in another mandatory categorically needy eligibility group in sections 1902(a)(10)(A)(i)(I) through (VII). Those individuals are not included in our analysis.

For the purposes of these estimates, we estimated how many people would be excluded from or meet the community engagement requirement in two steps. In the first step, we developed assumptions for the percentage of individuals that would not be subject to the community engagement requirement because they qualify for certain specified exclusions. In the second step, we made assumptions for the percentage of remaining individuals who would meet the community engagement requirement. This group would also include individuals who would be eligible for other specific exclusions, including those currently enrolled in TANF and/or SNAP and meeting the work requirements for those programs. Generally, we assumed that if an individual would meet the Medicaid community engagement requirement (for example, by meeting the employment or education requirements described in this IFC), then they were also likely to meet those requirements for TANF and/or SNAP (if they were enrolled in those programs). Therefore, we did not separately estimate how many individuals would qualify as specified excluded individuals on the basis of meeting or not being exempt from the work requirements of other programs.

There are also several categories under which an individual may qualify as a specified excluded individual and as such would not be subject to the community engagement requirement. As described in the preceding paragraph, in the first step we developed assumptions about how many people would qualify as a specified excluded individual under a subset of the exclusions in the statute. This subset includes: American Indians and Alaska Natives; pregnant women or individuals entitled to postpartum medical assistance; a parent, guardian, caretaker relative, or family caregiver of a dependent child 13 years of age and under or a disabled individual; inmates of a public institution; and individuals who are medically frail or otherwise have special medical needs. Individuals who have been an inmate at any point in the previous 3 months are mandatorily excepted from the community engagement requirement and are deemed to have demonstrated community engagement in each of the 3 months following their release. We have combined the impact of the mandatory exception and the specific exclusion for current inmates together in this analysis. We estimate that about 24 percent of applicable individuals would be specifically excluded under one or more of these categories and therefore not subject to the community engagement requirement. This estimate is derived by applying these exclusions and reflects the combined effect of removing American Indian and Alaska Native specified excluded individuals (2.5 percent) and applying the other 4 listed exclusion categories additively (21.7 percent combined). (There may be some overlap between these categories—for example, someone could be both pregnant and a parent of a child under age 13. Calculating the impact of these exclusions multiplicatively instead of additively to account for potential overlap could lead to a lower percentage of those estimated to receive specific exclusions; however, we believe the differences would be small (less than 1 percent) and that the 21.7 percent assumption is a reasonable estimate of the percentage of individuals who would be specifically excluded under one of these criteria.)

The combined impact of these 5 specific exclusions is 24 percent; that is, we estimate 24 percent of applicable individuals would meet one or more of these 5 specific exclusions. The assumptions for each specific exclusion and sources for those assumptions are shown in the following Table 37.

We also estimated the effect of short-term hardship exceptions, which we estimate would be equal to 3.45 percent of applicable individuals. States electing the optional exception must adopt all qualifying circumstances listed as short-term hardship events. We assume that 75 percent of States will adopt the optional exception for short-term hardship events. The assumptions for the short-term hardship exceptions are described in Table 38.

Combining the impacts of the subgroup of specified exclusions and the short-term exceptions, we estimate that 26 percent of applicable individuals would either have a specific exclusion (as listed in Table 37) or a short-term hardship exception (as listed in Table 38), and the other 74 percent of applicable individuals would either need to demonstrate compliance or be deemed to demonstrate compliance (via a mandatory exception) with the community engagement requirement or receive one of the other specified exclusions.

States are required to implement the community engagement requirement on or before January 1, 2027. States may elect to implement early, and the statute provides for delayed implementation for States demonstrating good-faith compliance efforts. We assume that the majority of States will implement the requirements effective January 1, 2027. Nebraska began implementing the community engagement requirement on May 1, 2026, and other States may also start later in 2026.

Because enrollment impacts depend heavily on State implementation choices that are not yet known, we modeled four scenarios representing a range of possible State policies. The scenarios vary by verification frequency (semi-annual or quarterly) and the number of months within each review period during which an applicable individual must demonstrate compliance.

We assigned scenario weights based on our actuarial judgment about the distribution of likely State implementation approaches, with 50 percent of enrollees assumed to be in States adopting the minimum statutory requirement of semi-annual verification with 1 month of compliance required between verifications, and the remaining 50 percent in States adopting more frequent verifications and/or longer compliance periods.

From these scenarios, we derive two distinct disenrollment rates. First, we estimate that 12 percent of applicable individuals subject to the requirement will not meet them and lose coverage. This estimate reflects a scenario-weighted average. Scenario-specific noncompliance rates are derived from KFF work status data for Medicaid adults, [ 149 ] reflecting the share of enrollees who are not working, not in school, and do not otherwise meet the community engagement threshold under each scenario's compliance rules. We assume that all individuals who report working full time would meet the community engagement requirement in each of the four scenarios. The portion of individuals that report working or attending school part-time who are assumed to meet the community engagement requirement varies from 100 percent to 75 percent depending on the scenario. This is because part-time work or school attendance may be subject to seasonal variation or other variations in either the number of work (or school) hours that are offered or that the individual is able to work (or attend school), and these variations may mean that the individual is not able to meet the community engagement requirement in every month of the year. Individuals who report not working for reasons other than caretaking, illness, or disability are assumed not to meet the community engagement requirement in any scenario. Taken together, these assumptions result in 81 percent of enrollees meeting the community engagement requirement in the lowest impact scenario and 75 percent of enrollees meeting the community engagement requirement in the highest impact scenario.

In addition, we estimate that 7 percent of applicable individuals who may be working, enrolled in school, or otherwise performing activities in line with community engagement requirement, or qualify for a mandatory exception or short-term hardship exception that deems them as demonstrating community engagement, would lose coverage due to administrative or procedural reasons (or in the case of a new applicant, may have their application denied and thus not enroll). These potential reasons for loss of coverage include, for example, not responding to verification requests or submitting insufficient documentation. The reasons individuals may not respond or submit insufficient documentation are manifold. This could include scenarios in which the required documentation was not received by the individual in a timely manner due to mail delivery delays or the individual had difficulty understanding or completing the required paperwork. Those fraudulently or improperly enrolled in the program or attempting to enroll in the program are unlikely to furnish documentation or respond to verification requests. Additionally, individuals with access to employer coverage or other forms of coverage may not complete documentation or respond to verification requests. Coverage losses may also be attributable, in part, to procedural processing errors at the State level. The gross procedural disenrollment rate is a scenario-weighted average anchored to post-unwinding (of the continuous enrollment condition related to the COVID-19 Public Health Emergency) Medicaid renewal data, which shows an average procedural disenrollment rate of about 12 percent for annual redeterminations over the most recent 12 months. [ 150 ] The procedural disenrollment rate may include non-responses by individuals who no longer meet requirements to qualify for Medicaid, documentation errors or non-responses by individuals who do meet requirements to qualify for Medicaid, or other factors. We assume that about half of those with procedural disenrollments (about 5.5 percent) are due to documentation errors or non-responses by individuals who do meet program requirements, a population which is more closely related to applicable individuals who either demonstrate community engagement or are deemed to demonstrate community engagement via an exception. Where redeterminations take place more than once a year, we assume that this procedural disenrollment rate applies at each redetermination, resulting in a 12 percent procedural disenrollment rate for scenarios with semi-annual renewals and a 20 percent rate for scenarios with quarterly verifications. The gross rate is then reduced by 50 percent to exclude baseline renewal attrition that would occur regardless of the community engagement requirement. The reduction accounts for the share of procedural disenrollments that would occur at routine renewal regardless of the community engagement requirement, which should not be attributed to this policy.

The 12 percent noncompliance rate is applied to the 74 percent of adult group enrollees estimated to be applicable individuals who are not receiving a specific exclusion nor excepted under a short-term hardship. The 7 percent net procedural disenrollment rate is applied to both the applicable individuals deemed compliant due to a mandatory exception or short-term hardship exception and to the applicable individuals subject to and meeting the requirements. Together, they yield a combined estimated disenrollment rate of approximately 15 percent of total adult group enrollment.

These rates are applied as full annual rates beginning in FY 2027, the first year of implementation. Because we assume implementation is effective January 1, 2027, the impacts in FY 2027 reflect 75 percent of the full-year impact.

The estimated enrollment impact is the estimated number of enrollees (1) who would be subject to the community engagement requirement and not meet the requirement, or (2) would meet the requirement or qualify for an exception, and would not successfully demonstrate their compliance or exception, developed based on the assumptions described above. The estimated impact on expenditures is the enrollment impact multiplied by the average per-enrollee expenditures.

We project that enrollment would be reduced by 2.3 million individuals in FY 2027 (accounting for implementation occurring in the second quarter of the fiscal year) and by between 3.1 to 3.3 million individuals in subsequent years. We project Federal Medicaid spending would be reduced by $350.3 billion over the next 10 years and State Medicaid spending would be reduced by $41.6 billion over the same time period. The impacts are shown in Table 41.

These estimates are subject to uncertainty. The disenrollment estimates reflect a weighted average of four implementation scenarios, and the actual distribution of State choices regarding verification frequency, compliance period length, adoption of short-term hardship exceptions, and implementation timing is unknown and could produce outcomes materially different from the central estimate. Additionally, assumptions are based on annual or average annual data; the timing of an enrollee's work or school hours relative to the timing of their redetermination may cause individual-level variation not captured in the aggregate estimates. The estimate for the incarceration mandatory exception is based on November 2025 Eligibility Operations and Enrollment Snapshot data [ 151 ] and does not capture individuals recently released from incarceration who remain covered by the 3-month review provision.

This analysis does none of the following: (1) estimate the extent to which individuals who lose Medicaid coverage may obtain alternative coverage, and (2) model interactions with other provisions of Public Law 119-21 . We assume both of these factors could partially offset or modify the projected coverage losses. Per-enrollee cost projections extend 10 years into the future and are subject to uncertainty inherent in long-range medical cost forecasting.