Obamacare’s subsidy cliff: How many enrollees are actually affected?

By Jeremy Nighohossian of CEI.

"Democrats in Congress have put Obamacare front and center in their opposition to the Republicans’ temporary budget. One provision of the American Rescue Plan Act of 2021 intended to provide relief during the pandemic expanded the health insurance premium subsidies created by the Affordable Care Act (ACA or Obamacare) for two years. The goal was to make subsidies more generous for existing recipients and available to those above the original income limit. The Inflation Reduction Act of 2022 extended those subsidies beyond the original expiration date, setting a new expiration date at the end of 2025.

With that deadline looming, Democrats in Congress are seeking another extension and are using the broader budget debate as leverage to win a full or partial renewal of those subsidies.

Because this expiration date is approaching, Americans enrolled in Obamacare and receiving subsidies face a reduced contribution to their premiums from taxpayers. Those below the income cutoff will continue to receive the original level of subsidies set by the ACA, while those above the income threshold will no longer receive subsidies at all. In both cases, the subsidies will revert to pre-pandemic levels.

Because many Americans will see reductions in subsidies, the number of affected individuals has become politically relevant. According to the Centers for Medicare and Medicaid Services (CMS), the agency that administers much of the Affordable Care Act, 24.3 million people signed up for ACA plans in 2025, and 22.4 million were eligible for subsidies.

News reports on the ACA and the shutdown have claimed repeatedly that 22.4 million people would be affected by the expiration of expanded subsidies, but there are many compelling reasons to doubt the accuracy of this figure.

Foremost among these reasons is that both CMS and the Kaiser Family Foundation have clarified that the 22.4 million figure refers to the number of people who signed up during the annual open enrollment period, rather than the number who are actually enrolled at any given time.

Open enrollment numbers can differ from actual participation for several reasons. Some may sign up believing they need coverage but later choose not to use it, and there have also been accusations of improper or fraudulent enrollment. Bloomberg investigated the ACA broker market in Florida and found numerous examples of unscrupulous brokers and other participants exploiting the ACA subsidy system by signing people up without their full consent in order to collect commissions. The system was vulnerable to such abuse because it combined fully subsidized, zero-cost enrollment with minimal oversight.

Under the Trump administration, CMS Director Mehmet Oz has made weeding out abuse one of his priorities, announcing efforts to address 3 million multiple enrollments.

Additionally, Paragon Health Institute has conducted several studies on ACA enrollment. In one study, it found that for certain income groups and states, ACA enrollment exceeded the number of people in those categories. A follow-up study found that the number of ACA enrollees with no claims has surged, which is an indicator, though not conclusive proof, of fraudulent enrollments. Paragon estimates that there are 6.4 million fraudulent enrollments in ACA in 2025.

How these fraudulent enrollments manifest is still an unanswered question. One possibility is that individuals sign up despite being ineligible for subsidies, misrepresenting their circumstances to secure the subsidies. Another possibility, as the Bloomberg article noted, is that brokers sign people up without their full knowledge. As with many government programs, there are strong incentives to exploit vulnerabilities to receive government funding, and the president has motivations that conflict with ensuring program integrity.

Another valuable source of information in this discussion is survey data. While CMS relies on applications from enrollees and brokers, several surveys ask Americans whether they’re insured and where they obtain coverage. The Current Population Survey (CPS), used to calculate the official uninsured rate, also asks respondents annually if they are enrolled in an ACA plan, and even if they receive a subsidy. Survey data cannot always capture the full picture, however, as an individual who lied to qualify may or may not tell a surveyor that they have ACA coverage.

From 2018 to 2021, CMS enrollment numbers tracked with the number of people in CPS’s survey who claimed they were covered by ACA plans, demonstrating the reliability of CPS’s methods.

However, in 2022, the first year CMS’s data would include enrollments based on the expanded subsidies, that alignment broke down. From 2018 to 2020, the two sources differed by only 2-3 million. In 2022, CMS reported 5 million more enrollees than CPS; in 2023, it was 5.3 million; and by 2024, it jumped to 9.6 million. CPS hasn’t released its 2025 survey yet, but the trend suggests an even larger discrepancy. Based on the difference from 2024, the CMS enrollment totals may be overstated by 10.9 million. This means the true number affected by the subsidy expiration would be 11 million, roughly half of the commonly cited number.

Of course, whether the true number is 22 million or 11 million doesn’t affect the broader discussion of who should receive subsidies and how large they should be. Yet the correct number is relevant to the program’s overall cost and significance to the economy and government. In the interest of accurate public debate, the 22 million estimate is highly doubtful and should carry a bold asterisk, if it’s to be used at all."