ANALYSISApril 13, 2026
Should Medicaid be expanded in every state?
As of 2026, 40 states and the District of Columbia have expanded Medicaid under the Affordable Care Act, leaving 10 states — mostly in the South — without expansion. Approximately 1.4 million uninsured individuals fall into a 'coverage gap' in those non-expansion states, ineligible for both Medicaid and Marketplace subsidies. The debate has intensified following the 2025 passage of H.R. 1, the 'One Big Beautiful Bill,' which imposed new work requirements, cut expansion incentives, and is projected to significantly reduce Medicaid enrollment.
Fourteen years after the ACA made it optional, twelve states still haven't expanded Medicaid — leaving millions in a coverage gap. Is that a policy failure that Washington should fix, or proof that states should keep the power to decide how far their safety nets stretch?
Not familiar with this story? Get context →
Coverage gap: real harm or pretextual
C
The 1.4 million people in the coverage gap are real, and the geographic concentration in Texas, Florida, and Georgia is verifiable — pretending otherwise is intellectually dishonest. But acknowledging the harm doesn't settle the mechanism question. The issue isn't whether the gap exists; it's whether federal Medicaid expansion is the right tool to close it, or whether it creates a dependency structure that compounds state fiscal problems over time.
L
You're conceding the harm is real, which matters — but then the mechanism question you're raising requires an answer, not a deferral. Non-expansion states aren't protecting their budgets by refusing expansion; they're leaving nine federal dollars on the table for every one they'd spend, while their hospitals absorb uncompensated care costs that expansion would reduce. That's not fiscal caution — that's ideology at their poorest residents' expense.
C
The 9-to-1 match rate is precisely the mechanism by which exit becomes impossible — states that expand must absorb 10% of an open-ended, Washington-controlled entitlement, and face overwhelming political pressure never to withdraw even when budgets tighten. The generosity is the trap, not the gift.
L
If the match rate is a trap, Arkansas, Montana, and Louisiana walked into it voluntarily and haven't collapsed — the predicted fiscal catastrophe keeps not arriving, which at some point becomes evidence rather than coincidence.
Medicaid's original categorical design
C
Medicaid was designed in 1965 to serve categorical populations — children, pregnant women, the disabled, and the elderly poor. The ACA's expansion to able-bodied childless adults earning up to 138% FPL wasn't a marginal adjustment; it was a categorical transformation of what the program fundamentally is. Conservatives who resisted weren't indifferent to poverty — they were insisting that a program purpose-built for dependency populations shouldn't be universalized without a serious structural rethink.
L
The 1965 design argument proves too much — the same logic would have blocked Medicare expansion to younger disabled people in 1972, or CHIP in 1997. Programs evolve when the coverage gaps they leave become undeniable. The relevant question isn't what Congress intended in 1965 but whether 1.4 million working-age adults earning under $22,000 a year have a legitimate claim on a functioning safety net.
C
CHIP and the 1972 disability expansion targeted populations with categorical barriers to private insurance — children and the severely disabled. Extending Medicaid to able-bodied working-age adults without dependents is a different kind of move, one that creates a permanent alternative to labor-market participation rather than a bridge for people who genuinely can't access coverage otherwise.
L
The 'able-bodied' framing assumes the coverage gap is populated by people who could get employer coverage but chose not to — the actual population is disproportionately in low-wage service jobs with no employer offer, which makes the categorical distinction do less work than you need it to.
Work requirements: fiscal reform or coverage loss
C
The CBO's 5.3 million figure deserves real scrutiny — the briefing itself acknowledges that some of those losing coverage are ineligible enrollees removed through proper eligibility verification, not people being denied necessary care. Attaching work requirements to Medicaid for able-bodied adults is a legitimate exercise of federal spending conditions, and it advances both fiscal sustainability and the principle that a safety net should support labor-market engagement, not substitute for it.
L
The Arkansas experiment you're describing as legitimate is the one courts halted after finding that most people who lost coverage were already working, caregiving, or in school — they just couldn't navigate the paperwork. The 'ineligible enrollees' framing in the CBO brief doesn't account for that: bureaucratic churn isn't the same as catching fraud, and 5.3 million more uninsured is 5.3 million more uninsured regardless of the mechanism.
C
Arkansas is a single implementation of one design — the argument for work requirements doesn't stand or fall on one state's rollout. A well-designed system with adequate exemptions for caregivers and students looks different from Arkansas's, and dismissing the principle because of one flawed execution is not the same as disproving it.
L
Every state that has implemented work requirements has produced the same result: coverage loss concentrated among people already engaged in work or caregiving, not able-bodied non-participants. At some point the pattern is the evidence, and 'better implementation next time' stops being a response.
Red-state ballot initiatives undercut federalism case
C
The ballot initiative results in Oklahoma, Missouri, and South Dakota are genuinely difficult to dismiss — these are conservative voters, in their own states, deciding the coverage gap was real and the federal offer acceptable. A principled federalism argument has to take this seriously. Subsidiarity isn't only about restraining Washington; it's about respecting genuine local democratic choice, and those votes are exactly that.
L
You're acknowledging the force of this, so let's be precise about what it means: voters in those states didn't just prefer expansion — they enshrined it in their state constitutions, which is the highest form of local democratic expression available. The ten remaining holdout states aren't defending some coherent conservative popular will; their political class is overriding it. The federalism argument becomes a shield for legislative majorities to ignore their own electorate.
C
The ballot initiative argument actually supports state-by-state determination, not federal mandate — Oklahoma and Missouri voters chose expansion for Oklahoma and Missouri, which is exactly how cooperative federalism should work. Using those outcomes to justify imposing expansion on states whose voters haven't made that choice inverts the logic.
L
That would be persuasive if the ten holdout states had put expansion on the ballot and it failed — but they haven't, because their legislatures won't allow the vote. Preventing a referendum isn't the same as reflecting popular will, and federalism shouldn't protect that maneuver.
Mortality evidence and safety net adequacy
C
The mortality evidence on Medicaid expansion is genuinely mixed — rigorous work by Finkelstein and others finds modest or statistically uncertain mortality effects, which complicates the life-or-death urgency framing. If the primary benefit is financial protection and access rather than dramatic mortality reduction, the argument that safety-net hospitals and community health centers partially fill the gap becomes harder to fully dismiss.
L
The mixed mortality signal narrows the argument, it doesn't close it — financial catastrophe from medical bills and delayed diagnosis until conditions become emergencies are serious harms even if they're statistically harder to count as deaths. And uncompensated care at safety-net hospitals is a more expensive, less dignified, and less effective mechanism than insurance — you're not actually defending the status quo when you cite it, you're just declining to improve it.
C
The honest conservative response to this isn't that uncompensated care is adequate — it's that block grants, catastrophic coverage alternatives, or state innovation waivers could close the gap without permanent federal administrative control. That argument is harder to make politically, but it's the one that holds if you actually believe the gap is real.
L
Block grants and waivers have been on the table for fifteen years and zero non-expansion states have used them to close the coverage gap — at some point the alternative mechanisms have to actually produce coverage, not just exist as theoretical rebuttals to the mechanism that works.
Conservative's hardest question
The Republican-state ballot initiative data is genuinely difficult to dismiss: if Medicaid expansion were primarily a federal imposition opposed by ordinary conservatives, it would not keep winning at the ballot box in red states. A principled federalism argument that ignores this democratic signal risks looking like elite policy preference dressed up as constitutional principle.
Liberal's hardest question
The evidence on whether Medicaid expansion meaningfully reduces mortality is genuinely mixed — some rigorous studies, including work by Finkelstein and others, find modest or statistically uncertain mortality effects, which is the strongest empirical challenge to the life-or-death urgency framing. If expansion's primary benefit is financial protection and access rather than dramatic mortality reduction, the conservative counter that safety-net hospitals and community health centers partially fill the gap becomes harder to dismiss entirely, even if it remains an inadequate substitute for actual insurance.
Both sides agree: Both sides accept that the 1.4 million people in the coverage gap represent a genuine, verifiable harm — neither argues the problem is fictional or adequately solved by existing safety-net alternatives.
The real conflict: They disagree on a factual-causal question: whether the 90% federal match rate is genuinely generous federal partnership or a structural mechanism designed to eliminate states' practical ability to exit the program once enrolled.
What nobody has answered: If federalism is the conservative principle at stake, why does it protect state legislatures overriding their own voters' ballot-mandated expansion decisions — and at what point does 'state flexibility' become a veto by officeholders against the populations they represent?
Sources
- Search results: Current Medicaid expansion status by state, 2025–2026 landscape
- Search results: Coverage gap statistics — 1.4 million uninsured in non-expansion states, state breakdown (Texas, Florida, Georgia)
- Search results: H.R. 1 'One Big Beautiful Bill' provisions — work requirements, CBO estimates, incentive cuts
- Search results: Constitutional amendments in Missouri, Oklahoma, and South Dakota requiring Medicaid expansion
- Search results: House budget resolution, February 2025 — $880 billion Medicaid cut proposal
- Search results: Idaho legislature signals potential Medicaid expansion repeal consideration
- Search results: Kelly Hall quote, Fairness Project, on state constitutional constraints