Affordable Care Act enrolment falls by nearly three million as US insurance costs rise

Affordable Care Act marketplace enrolment in the United States fell by nearly three million people in February 2026 compared with the same month in 2025, marking the first major reversal after several years of rapid growth in subsidised health coverage.

New federal data showed that approximately 19.2 million people had active Affordable Care Act exchange plans in February, down from 22.1 million a year earlier. The 13 percent decline provides the clearest evidence so far of how the expiration of enhanced federal premium subsidies and tighter enrolment controls are reshaping the individual insurance market.

The United States Department of Health and Human Services presented stronger action against fraudulent, improper and so-called phantom enrolment as an important explanation for the decline. Independent health policy analysts have placed greater emphasis on affordability, arguing that many legitimate customers dropped their plans after the enhanced premium tax credits expired on January 1, 2026.

The competing explanations matter because they produce very different policy conclusions. A decline driven mainly by improper enrolment would indicate that the marketplace is becoming more accurate and financially controlled. A decline driven mainly by unaffordable premiums would indicate that millions of self-employed workers, small-business owners and other Americans without workplace insurance are losing access to coverage.

Why did Affordable Care Act enrolment fall from 22.1 million to 19.2 million people?

The federal enrolment figures were compiled in April but measure coverage that remained active in February. This timing is important because customers who selected a plan during the annual open-enrolment period may not complete their first premium payment or may later lose coverage after failing to pay subsequent bills.

The February data capture the market after the initial nonpayment grace period had expired. They therefore provide a more reliable picture of effectuated enrolment, meaning people whose insurance had actually taken effect, than the larger headline totals announced during open enrolment.

Approximately 800,000 fewer people had selected Affordable Care Act plans during the enrolment period compared with the previous year. By February, however, the year-on-year decline had widened to approximately 2.9 million, indicating that a much larger group either did not activate coverage or was unable to keep paying for it.

The United States Department of Health and Human Services maintained that enforcement and eligibility reviews had removed or prevented significant numbers of improper enrolments. Federal officials have raised concerns about individuals being registered without their knowledge, brokers switching customers between plans and people receiving subsidies despite incomplete or inaccurate eligibility information.

Those concerns are not necessarily incompatible with an affordability-driven decline. Stronger verification could remove improper accounts at the same time that higher monthly payments cause legitimate customers to leave. The central unresolved question is how much of the 2.9 million reduction belongs to each category.

The federal report did not establish that all people removed from the enrolment total had been fraudulent customers. It also did not demonstrate that every enrollee who left the market became uninsured, since some may have obtained employer coverage, Medicaid, Medicare or another form of insurance.

How did the expiration of enhanced Affordable Care Act subsidies increase premiums?

The enhanced premium tax credits were introduced through the American Rescue Plan Act of 2021 and later extended through the end of 2025 by the Inflation Reduction Act of 2022.

The expanded assistance lowered the share of income that eligible households were expected to contribute towards marketplace premiums. It also temporarily removed the subsidy cliff that had prevented households earning more than 400 percent of the federal poverty level from receiving premium assistance.

That structure made marketplace coverage substantially cheaper for many middle-income households and reduced some monthly premiums to very low or zero-dollar amounts. Affordable Care Act enrolment subsequently rose to record levels as more people found that insurance was within their budgets.

Congress did not extend the expanded credits beyond December 31, 2025. The marketplace therefore returned to the Affordable Care Act’s earlier subsidy structure at the beginning of 2026, requiring many customers to pay a larger share of their premiums and ending assistance for some households above the income threshold.

The Congressional Budget Office had previously projected that allowing the enhanced credits to expire would increase the number of uninsured people and raise average benchmark premiums in the individual marketplace. The budget office also expected the market to lose some healthier customers, which can place further upward pressure on gross premiums when the remaining insurance pool has higher average medical needs.

Customers experienced widely differing increases depending on age, income, state, insurer and selected plan. Some faced moderate increases, while others encountered double-digit or larger jumps in the amount they were required to pay each month.

For a household already balancing housing, food, energy, transport and debt costs, even a relatively modest premium increase can make insurance difficult to retain. Customers may technically remain eligible to purchase a plan while being unable to afford both the premium and the deductible required before substantial benefits become available.

Is the enrolment decline mainly a fraud crackdown or an affordability crisis?

The federal government and independent analysts are interpreting the same enrolment decline through different institutional priorities.

The United States Department of Health and Human Services has focused on programme integrity. Its position is that previous marketplace totals included substantial improper, fraudulent or phantom participation and that stronger verification has reduced government spending on accounts that should not have received subsidies.

Phantom enrolment can occur when a policy is created or renewed but the named person is unaware of the coverage, does not intend to use it or has been improperly enrolled by an intermediary. Zero-premium plans can increase this risk because an account may remain active without the consumer receiving a monthly bill that would alert them to the enrolment.

Independent health researchers have not dismissed the existence of enrolment fraud. Their concern is that the federal explanation may understate the number of genuine customers who left because their premiums rose after the enhanced tax credits expired.

KFF has pointed to consumer evidence showing that real households faced double-digit or even larger increases in what they paid. The health policy organisation expects marketplace participation could continue declining during 2026, potentially falling to approximately 17.5 million people.

The most balanced interpretation is that both forces are operating. Eligibility enforcement may be removing improper accounts, while the loss of enhanced subsidies is simultaneously pushing legitimate consumers out of the market.

A more definitive answer will require detailed information showing why individual policies ended, whether former enrollees obtained other coverage and how losses varied across income groups, states and types of marketplace.

Which Americans are most exposed when Affordable Care Act coverage becomes more expensive?

Affordable Care Act exchanges primarily serve working-age people who do not receive affordable insurance through an employer and do not qualify for Medicare, Medicaid or another public programme.

The enrolment base includes self-employed professionals, gig workers, independent contractors, farmers, ranchers, hairstylists, early retirees and employees of small businesses that do not provide comprehensive health benefits. Many have incomes that fluctuate, making annual subsidy calculations and premium payments particularly difficult.

People who earn too much to qualify for Medicaid but lack affordable job-based insurance are especially dependent on the marketplace. Residents of states that have not expanded Medicaid also face a higher risk of falling into a coverage gap in which their incomes are too low for marketplace subsidies under the standard rules but too high or otherwise ineligible for their state’s Medicaid programme.

KFF estimated that 26.7 million Americans under the age of 65 were uninsured in 2024, an increase of more than 1.3 million from 2023. More than six in 10 uninsured working-age adults identified affordability as the primary reason they lacked coverage.

The uninsured population is also closely connected to employment. More than eight in 10 uninsured people lived in a family with at least one worker, while many employed adults were not offered insurance or were ineligible for their employer’s plan.

The loss of marketplace coverage can therefore affect people who are economically active but work outside the traditional large-employer benefits system. Their decision to remain uninsured may reduce monthly expenses in the short term while increasing exposure to medical debt and delayed treatment.

What happens when people lose marketplace insurance but still need medical care?

Losing insurance does not remove the need for healthcare. It changes when people seek care, which providers they can access and who ultimately pays the cost.

Uninsured patients are more likely to postpone routine appointments, prescription refills, diagnostic tests and treatment for chronic conditions. Delayed care can allow manageable illnesses to become more serious, increasing the likelihood that patients eventually require emergency or hospital treatment.

Hospitals and community health centres may experience more uncompensated care when patients cannot pay their bills. Providers can absorb some of those losses, receive limited government support or shift costs through higher charges elsewhere in the healthcare system.

Households without coverage are also more exposed to medical debt. A single emergency admission, cancer diagnosis, serious accident or complicated pregnancy can generate costs far beyond what most families can pay from savings.

The enrolment decline may consequently affect healthcare providers, state budgets and insurance markets in addition to the individuals leaving coverage. A smaller and less healthy marketplace population can make premiums harder to control, particularly if healthier customers are more willing to risk remaining uninsured.

The financial consequences will vary by state. States operating their own marketplaces may adopt supplementary subsidies or other affordability measures, while states relying heavily on the federal exchange may have fewer resources or less political support for replacing the expired federal assistance.

Could Affordable Care Act enrolment continue falling through the end of 2026?

The February total is unlikely to be the final measure of 2026 marketplace participation. Enrolment can decline during the year as customers stop paying premiums, obtain other insurance, experience income changes or are removed following eligibility reviews.

KFF’s estimate that enrolment could fall to approximately 17.5 million suggests that an additional 1.7 million people could leave after the February measurement. That outcome is not guaranteed, but it illustrates the potential scale of continued attrition if higher costs remain the dominant pressure.

The eventual total will depend partly on whether Congress revisits the enhanced premium tax credits. Restoring the subsidies could reduce customer payments, but any legislation would face disputes over federal spending, programme integrity and whether assistance should be targeted differently.

Federal administrative policy will also matter. More frequent income verification, restrictions on automatic re-enrolment and stronger broker oversight may further reduce enrolment while improving the accuracy of subsidy payments.

The 19.2 million February total nevertheless remains higher than marketplace participation before the most recent enrolment surge. The Affordable Care Act exchanges have not collapsed, but they are entering a period in which coverage growth can no longer be assumed.

Insurers will watch whether the departing population is younger and healthier than those who remain. A deteriorating risk pool could influence proposed premiums for 2027 and create a cycle in which rising prices cause additional healthy customers to leave.

Why could health insurance affordability become a major issue in the November elections?

The enrolment decline has emerged as American voters are already concerned about the cost of healthcare, groceries, housing and other household necessities.

Democrats are likely to present the figures as evidence that Congress should have extended the enhanced subsidies. Their argument will be that the tax credits produced record coverage levels and that allowing them to expire made insurance unaffordable for millions of working households.

Republicans are likely to emphasise the federal government’s findings concerning improper and phantom enrolment. Their argument will be that previous enrolment growth overstated genuine participation and allowed taxpayer-funded subsidies to flow through a system with inadequate verification.

The policy debate will therefore turn on whether the 2026 decline is framed as a correction or a coverage failure. That framing could influence proposals involving subsidy restoration, eligibility checks, broker regulation and the future cost of the Affordable Care Act.

The immediate numbers support neither an entirely celebratory nor an entirely catastrophic interpretation. Programme integrity may have improved, but genuine consumers also faced higher costs and left the market.

The political test will be whether lawmakers can reduce fraud without making legitimate insurance unaffordable. Treating those goals as mutually exclusive risks leaving the marketplace vulnerable to both improper enrolment and preventable coverage losses.

What are the key takeaways from the fall in Affordable Care Act marketplace enrolment?

• Affordable Care Act marketplace enrolment fell from approximately 22.1 million people in February 2025 to 19.2 million in February 2026, representing a decline of roughly 2.9 million people or 13 percent.
• The February data are more meaningful than initial open-enrolment selections because they measure active coverage after customers had time to make premium payments and after the early nonpayment grace period had expired.
• The United States Department of Health and Human Services linked much of the decline to stronger action against improper, fraudulent and phantom enrolment, although the report did not establish that every person leaving the total was improperly enrolled.
• Independent health policy analysts emphasised the January 1, 2026, expiration of enhanced premium tax credits, which increased the amount many legitimate marketplace customers had to pay for their health insurance.
• Self-employed workers, gig workers, farmers, ranchers, early retirees and small-business employees are among the groups most dependent on Affordable Care Act plans because they often lack access to affordable employer-sponsored coverage.
• KFF expects marketplace enrolment could continue falling during 2026 and potentially reach approximately 17.5 million, although the final total will depend on premium payments, eligibility reviews and changes in customers’ employment or income.
• The decline does not prove that 2.9 million people became uninsured because some former marketplace customers may have moved to employer coverage, Medicaid, Medicare or another insurance arrangement.
• Affordable Care Act affordability is likely to become a major issue before the November elections, with policymakers divided between restoring enhanced subsidies and strengthening controls against improper enrolment.