Products are selected by our editors, we may earn commission from links on this page.

A policy meant to encourage employment is projected to remove millions of people from Medicaid rolls regardless of whether they’re actually working. New Urban Institute projections estimate that average monthly Medicaid expansion enrollment nationwide could fall by almost 8 million people by 2028 under new federal work requirements and more frequent eligibility checks. California and New York are expected to see the largest numerical declines of any state.
The new rules stem from the One Big Beautiful Bill Act, the sweeping budget law President Trump signed in July 2025. Adults covered through Medicaid expansion in 41 participating states will generally need to complete at least 80 hours of approved activity each month, including paid work, community service, an approved employment program, or part-time education, with states required to implement the requirement by January 1, 2027.
The law includes specific exemptions meant to protect certain groups from the new requirement entirely. Pregnant people, former foster youth, American Indians and Alaska Natives, totally disabled veterans, and people in drug or alcohol treatment programs are excluded, along with parents of children 13 or younger and caregivers supporting someone with a disability, among several other protected categories written into the statute.
A Second Rule Change Compounds the Work Requirement’s Effect

A separate change compounds the work requirement’s overall effect on enrollment. States must now reassess Medicaid expansion recipients’ eligibility every six months instead of annually, checking whether income and other rules are still met. The Urban Institute estimates that this shift to twice-yearly reviews alone could reduce average monthly enrollment by as much as 3.1 million people, even before counting the work requirement’s separate impact.
Administration officials frame these changes as a path toward employment and long-term independence rather than a simple coverage reduction. “This rule helps Americans build skills and independence through work, education, job training, or community service,” said Dr. Mehmet Oz, head of the Centers for Medicare and Medicaid Services, describing the policy as a way to create new opportunities for recipients and their families.
California is projected to record the single largest numerical reduction of any state under the medium scenario, with almost 2 million fewer people enrolled during an average month in 2028. New York follows with an estimated decline of 955,000 enrollees, while Illinois could lose 324,000, reflecting how sheer population size shapes which states see the biggest raw numbers under the new rules.
Massachusetts Could Lose More Than Half Its Expansion Enrollment

Several other populous states also appear high on the list of largest projected numerical declines. Michigan and Ohio are expected to lose 287,000 and 285,000 enrollees respectively, while Pennsylvania, Virginia, Washington, North Carolina, and New Jersey each face projected declines ranging from roughly 242,000 to 276,000 people, rounding out the ten states facing the steepest raw enrollment losses nationwide by 2028.
Measuring the impact as a percentage of each state’s affected enrollment tells a noticeably different story than raw numbers alone. Massachusetts faces the largest proportional reduction under the medium scenario, projected to lose 54% of its expansion enrollment. Connecticut and Maryland could each see declines around 51%, while Vermont and Minnesota face projected losses near 50% and 49% respectively by 2028.
Some states are projected to weather these changes far better than others in percentage terms. North Dakota has the smallest projected decline nationwide, at just 30%, followed closely by Oregon and South Dakota at 34%. Indiana and Nebraska are each expected to see enrollment fall by about 35%, meaning even the least-affected states still face a significant loss of coverage overall.
Some People Could Lose Coverage Even While Following Every Rule

These projected numbers describe average monthly enrollment, not the total number of individual people who might lose coverage at some point across the entire year. That distinction matters because someone could briefly lose coverage during one reporting period and later regain it, meaning the true number of people affected at some point could differ meaningfully from these specific monthly snapshot estimates.
Perhaps the most striking finding involves people who are actually complying with the new rules entirely. The Urban Institute cautioned that some eligible people could still lose coverage even while working enough hours or qualifying for an exemption, largely due to administrative friction in reporting irregular hours, self-employment income, or documenting a qualifying medical or caregiving situation accurately and on time.
The deepest risk in this policy isn’t really about people who refuse to work. It’s about eligible people losing coverage anyway because a form arrived late, a paystub didn’t match a database, or a caregiving situation proved hard to document on paper. Losing insurance, even temporarily, can delay treatment and create new barriers to the very employment this policy was designed to encourage.
