The Centers for Medicare & Medicaid Services has canceled about 315,000 Affordable Care Act Marketplace enrollments covering more than 760,000 people, saying its review confirmed the enrollments were unauthorized. The agency expects the cancellations to return roughly $2.2 bn in federal premium tax credit payments.
The action applies to the federally facilitated Marketplace and state-based exchanges using the federal platform. CMS said it worked with health insurers to investigate suspected unauthorized enrollments before canceling coverage on August 31, 2026.
Federal officials described the affected group as including people enrolled without their knowledge as well as people whose eligibility information did not satisfy Marketplace requirements. Officials have not published a detailed breakdown showing how many cases fell into each category.
The administration is also reviewing another 419,000 to 450,000 enrollees to verify residency and income eligibility, according to federal officials. Those reviews form part of a wider effort focused on improper enrollment and broker activity across HealthCare.gov.
CMS said unauthorized enrollments might generate as much as $6.6 bn in improper federal spending during the 2026 plan year.
The $2.2 bn estimate attached to the August cancellations refers specifically to advance premium tax credits associated with the 315,000 terminated enrollments.
The agency has also tightened oversight of insurance agents and brokers selling ACA plans. Since January 2026, CMS has sent termination notices to more than 200 agents and brokers for alleged non-compliance with Marketplace standards.
Another 569 agents and brokers received notices of intent to terminate after submitting 2026 applications lacking applicant identification information such as Social Security numbers. Of the first 100 cases reaching the end of the response period, 66 agents or brokers received termination notices.
CMS also imposed a temporary freeze on new registrations for agents and brokers seeking to participate in the 2027 federal Marketplace without an active 2026 Exchange Agreement. Reuters reported the restriction will remain in place through February 1, 2027.
Federal data show recently registered brokers represented a relatively small share of broker-assisted enrollment but produced a disproportionate amount of activity CMS classified as high risk.
Applications associated with brokers who first registered for 2026 were 2.8 times more likely to contain unresolved income verification issues and 2.7 times more likely to lack Social Security numbers.
They were also 2.6 times more likely to contain unresolved citizenship or immigration verification issues. CMS found higher rates of unverified special enrollment periods and overlapping Medicaid or Children’s Health Insurance Program coverage among the same group.
The new controls extend beyond broker registration. Existing agents and brokers must complete identity verification again through Login.gov or ID.me, while broker-assisted applications now require verifiable Social Security or immigration document numbers for applicants other than newborns.
CMS is also restricting agents from attaching themselves to applications consumers are completing independently through HealthCare.gov. Electronic consumer authorization will become mandatory before a broker takes action on an application or enrollment.
The government says the changes address cases where brokers enrolled consumers without permission, switched plans without authorization or submitted inaccurate eligibility information to collect insurance commissions.
Unauthorized ACA enrollment has drawn bipartisan scrutiny for several years, with enforcement beginning before the current administration.
The review process has also raised questions about legitimate policyholders losing coverage. Health policy researchers have argued that the available federal disclosures don’t provide enough information to determine how many canceled enrollments involved nonexistent applicants, consumers signed up without consent or people who were eligible but failed to complete verification requirements.
Insurers involved in the review attempted to contact flagged enrollees before coverage was terminated.
Critics of the process argue that failure to respond doesn’t necessarily establish fraudulent enrollment, especially among lower-income households facing language barriers, housing instability or difficulty receiving official correspondence.
People whose coverage was canceled have a process to challenge the decision and restore enrollment after establishing eligibility. The practical effect of additional documentation requirements remains under scrutiny because administrative steps themselves often reduce participation among people who otherwise qualify for public programs.
The broader ACA Marketplace has already contracted sharply during 2026. Effectuated enrollment fell from about 22.1 mn people in February 2025 to 19.2 mn in February 2026, a decline of nearly 3 mn, according to federal enrollment data analysed by the Center on Budget and Policy Priorities.
That decline followed the expiration of enhanced premium tax credits at the end of 2025. The temporary subsidies had reduced net premiums for Marketplace customers and expanded financial assistance above the previous income threshold.
KFF reported that 2026 marked the first decline in ACA Marketplace enrollment in seven years after the enhanced credits expired. The 19.2 mn enrollment figure still remains above levels recorded before the recent expansion of subsidies.
The administration has separately argued that rapid enrollment growth after 2021 included substantial improper or phantom enrollment. A June report from the Department of Health and Human Services estimated 2.6 mn improper or phantom enrollments remained in 2026, including more than 1 mn enrollments without a Social Security number.
Those estimates remain disputed. Researchers critical of the methodology argue that comparisons between Marketplace administrative records and Census income data don’t always use equivalent definitions of household size or income, potentially overstating discrepancies.
Another point of disagreement concerns enrollees who never submit an insurance claim. Federal officials have cited high levels of zero-claim enrollment as evidence supporting further investigation, especially among fully subsidized plans arranged through brokers.
Health insurance researchers note that having no claims during a year doesn’t by itself establish fraudulent enrollment. Healthy policyholders routinely pay premiums without using medical services, and such members generally lower rather than increase insurers’ average claims costs.
The broker moratorium has drawn concern from the insurance distribution sector as well. Industry representatives argue that restricting all new federal Marketplace broker registrations risks affecting legitimate agents alongside firms involved in unauthorized enrollment.
Analysts have also raised a risk-pool issue. If enforcement disproportionately removes legitimate low-utilization members, the remaining insured population would become more expensive on average, putting upward pressure on premiums. The size of that effect depends on how many terminated enrollments belonged to genuine policyholders rather than unauthorized or ineligible accounts.
The ACA still contains a separate coverage gap for some low-income adults in states that have not expanded Medicaid.
Marketplace premium subsidies generally begin at 100% of the federal poverty level, leaving some adults below that threshold without subsidized Marketplace coverage or Medicaid eligibility under their state’s rules.
Federal officials said eligibility verification will include closer scrutiny of income information for people receiving Marketplace subsidies. Critics argue that stricter verification in this income range risks removing eligible consumers alongside improper enrollments if documentation problems aren’t resolved quickly.
The latest enforcement action therefore combines two separate issues: documented unauthorized broker activity and a broader dispute over the scale of improper ACA enrollment. CMS says the August cancellations followed insurer review and confirmation, while outside health policy researchers are seeking more information about how affected cases were classified.
