Who Is Responsible for the ACA Premium Increases?

Information current as of August 22, 2026. Open enrollment for the 2027 plan year begins November 1.

Overview

This post examines the argument over what happened to Affordable Care Act premiums after Congress allowed the enhanced premium tax credits to expire on December 31, 2025, and over what should happen next.

A note on the numbers first, because this is where most of the public argument goes wrong.

Three figures circulate constantly: a 26 percent average premium increase, a 114 percent increase in what enrollees pay out of pocket, and between 2.2 million and 7.3 million people expected to drop coverage. All three are real, all three come from credible sources, and all three were projections made before the expiration took effect.

Actual 2026 data is now available, and it differs. KFF reports that the average monthly premium payment net of tax credits rose 58 percent, from $113 to $178, not 114 percent.

The reason is not that the projection was wrong on its own terms. The 114 percent figure estimated what enrollees would pay to keep the same plan. Many did not keep the same plan. They bought down to bronze plans with lower premiums and much higher deductibles, and the people facing the steepest increases, those above 400 percent of poverty who lost eligibility entirely, left the market at disproportionately high rates.

That distinction matters for every claim below. A smaller measured increase in premium payments is partly the result of people accepting worse coverage or no coverage. Anyone citing 58 percent as evidence the warnings were overblown, and anyone citing 114 percent as a description of what happened, is misusing the number.

What Critics of the Expiration Say

Claim: Premium payments more than doubled. This was KFF’s projection, an increase from $888 to $1,904 annually for subsidized enrollees keeping the same plan. As a projection it was sound and widely reported. As a description of realized outcomes it is not accurate, for the reasons above.

Claim: The subsidy cliff has returned and hits older middle-income people hardest. Accurate and well documented. Above 400 percent of the federal poverty level, roughly $62,600 for a single person in 2026, the credit drops to zero rather than phasing out. Because premiums are age-rated, a person in their early sixties just over that line can face an increase of several thousand dollars a year. This group is too young for Medicare and often self-employed. The mechanism is not disputed by either side.

Claim: Enrollment has fallen sharply. Supported. Effectuated enrollment has declined toward roughly 17.5 million from about 24 million in 2025. KFF notes this aligns closely with the Congressional Budget Office’s earlier projection of a roughly 25 percent marketplace contraction. The share of enrollees receiving any premium tax credit fell from 92 percent to 87 percent, the first decline in subsidy uptake since 2020.

Claim: Deductibles rose by about $1,000 per person. KFF reports this alongside the premium figures. It is the other half of the trade-off, and it is the part a premium-only comparison hides.

Claim: Millions will become uninsured. The estimates vary and the range should be given rather than a single number. CBO projected the uninsured population rising by 2.2 million in 2026, by 3.7 million in 2027, and by 3.8 million on average annually through 2034. The Urban Institute projected 7.3 million losing marketplace coverage in 2026, of whom 4.8 million would become uninsured. The Commonwealth Fund and the Economic Security Project produced figures near 4 million. These measure different things over different periods, which is why they differ.

Claim: Republicans chose this outcome. Partly supported and partly a framing. The enhanced credits were temporary by statute and expired on schedule, so no vote was required to end them. What is checkable is that competing bills failed and that Senate Majority Leader John Thune said there was no appetite for an extension.

What Defenders of the Expiration Say

Claim: The enhanced credits were always temporary. Accurate and important. They were enacted in the American Rescue Plan Act of 2021 as a pandemic measure and extended by the Inflation Reduction Act. The underlying ACA premium tax credit is permanent and did not expire. Describing the change as the end of Obamacare subsidies, as some coverage has, is wrong.

Claim: Extending them is expensive. Supported. CBO and the Joint Committee on Taxation estimated a permanent extension would add roughly $350 billion to the deficit over ten years, and scored the three-year House bill at about $80.6 billion. Cost is a legitimate consideration rather than a pretext.

Claim: The program had fraud problems. Thune said the enhanced credits were especially susceptible to fraud. There is a documented issue with unauthorized enrollments, which is why the bipartisan Senate proposal included a $100,000 fine on insurers for signing someone up without consent, along with a $5 monthly minimum premium. That both parties negotiated over these provisions indicates the underlying problem was taken seriously by both. The scale of it is not established by the existence of the reforms.

Claim: Premiums were rising anyway. Supported and often overlooked. KFF attributes the 2026 increases substantially to underlying medical costs: hospitalizations, physician visits, and prescription drugs including GLP-1 medications, plus labor costs. For 2027, the underlying cost of medical care and prescription drugs rose 10 percent, above the 8 percent average of recent years. Attributing the entire premium picture to the subsidy decision is inaccurate.

Claim: Most enrollees kept coverage. Partly supported. Roughly 17.5 million did enroll. But 70 percent of KFF survey respondents said that if their payments doubled they would look for a plan with a lower premium and higher out-of-pocket costs, and about 40 percent of enrollees moved to bronze or other low-cost high-deductible plans. Keeping coverage and keeping comparable coverage are different claims.

Claim: An extension would mostly benefit people who do not need help. The bipartisan Senate framework would have capped eligibility at 700 percent of the federal poverty level, which concedes that unlimited eligibility was a legitimate target for reform while rejecting a hard return to 400 percent. That both sides negotiated in that range suggests the disagreement was about where the line goes, not whether one should exist.

Where Both Sides Are Making the Same Mistake

Treating projections as outcomes. Discussed above, and it runs in both directions.

Ignoring that healthy people leaving raises costs for everyone left. CBO expected healthier-than-average enrollees to exit and insurers to raise premiums for those remaining. Insurer rate filings for 2027 cite exactly this, with one Indiana carrier explicitly projecting increased statewide morbidity. This is a mechanism both sides acknowledge in principle and neither emphasizes when inconvenient.

Ignoring the states. At least ten states funded their own support in 2026. New Mexico fully replaced the federal enhanced credits and reported a 17.1 percent increase in plan selections, while California allocated $190 million against roughly $2.5 billion its residents had been receiving federally. National averages conceal enormous variation.

A Note on How Unusual This Is

Among wealthy countries, the United States is alone in delivering coverage to this population through a subsidized private marketplace whose affordability depends on a tax credit Congress must periodically renew. In Germany, Japan, the Netherlands, and Switzerland, private or quasi-private insurance is universal and the subsidy structure is set in permanent law rather than reauthorized every few years.

That difference is the reason this argument exists in the form it does. The policy question of how generous a subsidy should be is live everywhere. The question of whether coverage disappears when a temporary provision lapses is largely particular to the American design, and it is worth naming when readers outside the United States ask why an expiration date could move millions of people.

Where This Stands

The enhanced credits remain expired. The House passed a three-year extension in January 2026 by 230 to 196 with 17 Republicans, and the Senate did not take it up successfully. A bipartisan working group led by Sens. Bernie Moreno and Susan Collins negotiated a framework involving a two- or three-year extension, an income cap, a minimum premium, fraud penalties, and a health savings account option. Those talks stalled with each side blaming the other. Four Republicans, Collins, Josh Hawley, Lisa Murkowski, and Dan Sullivan, crossed over on an earlier Senate vote.

Insurers have filed largely double-digit proposed increases for 2027. Open enrollment for that year begins November 1, 2026, days before the midterm elections.

Discussion Questions

Should a temporary program be extended once people have come to rely on it?
Is a lower premium with a much higher deductible better coverage or worse?
How would you decide where the income cutoff for help should be?
Does the cost of a program to the government change whether it is worth doing?
Why do you think people cite the projection rather than the actual figure?
Should states have to fill a gap left by federal policy?
Who is responsible when a temporary law expires on schedule?
How does your own country handle health coverage for self-employed people?
Is health insurance a normal purchase or something different?
Imagine you are 62, self-employed, and just over the income cutoff. What are your options?