It starts in January.
2027 is the target.
By then, roughly 20 million Americans — non-pregnant, ages 19 to 68, in 43 states plus DC — will need to prove they’re “doing something.” The rule came out June 1. The Centers for Medicare & Medicaid Services (CMS) turned the One Big Beautiful Bill Act’s work requirement from ink into law. You have 80 hours to fill. Job training? Half-time school? Community service? It all counts. Earn $580 a month — that’s 80 hours at federal minimum wage — and you’re safe.
Not everyone has to play. Pregnant women? Out. Disabled veterans? Out. Kids under 13? Parents are exempt. It’s a long list.
But for those in the middle, it gets bureaucratic fast. States have to check compliance. Every six months. Miss a check? You get 30 days. Cover your bases. In year one, you can self-attest. Starting 2028, documentation is king.
This isn’t just paperwork.
It’s a $326 billion shift in the American economy. And nobody really knows if it will work.
Two Stories. Same Policy.
There’s the view from Capitol Hill. Katie West and Matthew VanHytte, staffers on the House Energy & Commerce Committee, see a rescue mission. They think work requirements re-engage the able-bodied. Employment goes up. Income follows. People move into employer insurance. Medicaid finally gets to do what it was “designed for” — help the disabled, kids, and the truly frail.
West calls it restoring a Clinton-era principle: benefits come with expectations. She insists the Arkansas experiment from 2018 — which dropped 18,00 people without boosting jobs — won’t happen this time. New data frameworks. Better vendor tech ($600 million worth, pledged). Automatic confirmations for workers who already punch a clock. She claims nearly eight in ten Americans support this in conservative polls.
It’s a clean narrative. Move people toward independence. Bend the spending curve.
Then there’s the clinic side. Dr. Adam Brown, emergency physician, and Dr. Steve Farmer, cardiologist. They run ABIG Health. They look at Arkansas and see the truth.
No employment gain. Just debt.
Researchers found half of those who lost coverage racked up medical debt. Six in ten skipped meds or delayed care. The problem wasn’t that people weren’t working — more than 95% of the target group were already meeting the standard. They just couldn’t prove it on time.
Brown calls it a design flaw so deep it can’t be patched. 64% of enrollees work. Another 15% can’t due to disability. The system will burn energy re-verifying the compliant 80%. Meanwhile, eligible people fall through cracks because they were short by one day in a given month. Twice a year.
West counters that data-first verification protects the workers. Attention should stay on the small slice who are truly disengaged. Defenders point to modest literature supporting employment gains. Is it worth the churn?
The Arkansas data is the only real test we have.
Brown leans into that null result. A 2025 Urban Institute Study confirmed it: zero employment boost.
Georgia adds fuel to the fire. Its “Pathways” program is the only current live wire. The General Accounting Office found it cost $54 million in administration vs. $26 million for actual care. West dismisses Georgia. They say it’s apples to oranges. Georgia required proof of work before enrollment. The new rule applies to renewal. Easier bar, theoretically.
Brown says it’s a preview. Federal dollars paid for 90% of that admin cost. Data matching — the same tool promised now — has been used since 2024. Start-up costs? Eligibility systems? They’re structural. Every state pays them.
Where Is the Money Going?
Let’s talk about the headline: $326 billion saved over ten years. That’s the Congressional Budget Office (CBO) number. Spending forecasts climb from $700 billion to $981 billion. This bill stops that. West argues people leaving Medicaid means less burden on the program. Simple.
Brown and Farmer disagree.
It’s not savings. It’s cost shifting.
When 5.2 million people lose coverage in a single year (per GW analysis), the costs don’t vanish. They migrate. To the patients. To the ERs. To the states. A recent AP review suggests implementation could cost states over $1 billion, despite only $200 million in federal help.
Jobs disappear, too. Nearly half a million, according to one model. GDP drops.
Dr. Farmer makes a sharp legal point: If you fail the work req, you’re treated as eligible for Medicaid but failed to qualify. So you can’t get marketplace premium tax credits. You’re stuck. No safety net. And most low-wage workers? Their employers don’t offer health insurance.
West sees this as steering. Away from subsidies. Toward labor. Brown sees it as a foreclosed exit trap.
Who’s right?
Owning the Outcome
The early adopters — Nebraska, Montana, Arkansas — are switching on before 2027. They’ll provide the data first.
Brown bets against them. He wants to see the metric: does the 92% compliance rate spike when reporting starts? Or does the bureaucracy eat itself?
We’ll find out fast.
But let’s be clear on the accountability. This provision was written by the Energy & Commerce Committee. Set their timeline. Defended it publicly. Chairman Brett Guthrie called it independence.
If it pulls 2.9 million out of poverty? Good for them.
If it sends sick, eligible people into debt because of a form they missed in November?
That’s on them, too.
Guthrie’s team built the test. In January 2027, 20 million people take it. The ledger won’t hide the names attached to the result.
