What This Paper Covers

In last year’s The Great Obamacare Enrollment Fraud, we uncovered the Obamacare exchanges were riddled with fraud stemming from improper enrollment. Using the latest release of government data, we have found that this improper enrollment and fraud surged over the past year—rising by more than one-quarter to now total nearly 6.4 million ineligible recipients. Improper enrollment and fraud are particularly egregious in states using the federal exchange (HealthCare.gov), especially those that have not expanded Medicaid.

This paper reviews the basic enrollment mechanics and perverse incentives of Obamacare subsidies. From its inception, Obamacare provided large subsidies for lower-income people to buy coverage in the exchanges. COVID-era legislation, signed into law by President Biden, substantially increased these subsidies and scheduled them to expire after 2025. The Biden subsidy boost made plans fully taxpayer-funded for enrollees who claim income between 100 percent and 150 percent of the federal poverty line (FPL). Enrollees in this income range also qualify for a cost-sharing reduction program that significantly lowers deductibles and other out-of-pocket expenses. In addition, the Biden administration took a host of administrative actions that prioritized higher enrollment numbers over eligibility verification, allowing fraud to grow unchecked.

This paper describes the powerful incentives for individuals, brokers, and insurers to misestimate applicant income to qualify for larger subsidies. Insurers benefit from larger enrollment and government subsidies, and brokers benefit from higher commissions. As a result, millions of enrollees appear to be enrolled without their knowledge, already have other coverage, or are unaware they remain enrolled.

This paper breaks down the impact of this faulty design and its financial consequences by state. We do this with a clear side-by-side comparison of the number of people reporting income between 100 and 150 percent FPL and the estimated number actually eligible. Moreover, this paper explores factors contributing to the significant growth in fraud, identifying concerning trends.

The paper concludes with a set of recommendations to reduce fraudulent enrollments, which would save billions of taxpayer dollars that could be directed to those most in need.

What We Found & Why It Matters

Exchange enrollment fraud is driven by a combination of four factors: (1) the enhanced Obamacare subsidies, enacted as a temporary pandemic-era policy, which fully subsidize plans for people who report income between 100 and 150 percent of the FPL; (2) Biden administration policies that prioritized enrollment over eligibility verification and program integrity; (3) limits on how much the federal government can recover if insurers receive excess advance subsidies because of improper eligibility determinations; and (4) automatic re-enrollment, which perpetuates improper enrollments year after year.

We also found that improper exchange enrollment significantly increased from 2024 to 2025. We estimate, conservatively, that improper enrollment—defined as enrollees who claimed, but did not actually have, income between 100 and 150 percent FPL—increased from 5.0 million enrollees in 2024 to 6.4 million enrollees in 2025. We estimate that the taxpayer cost of improper enrollment will exceed $27 billion this year. Under more expansive and likely realistic assumptions, improper enrollment reaches 7.1 million people.

More than half of exchange sign-ups during the 2025 open-enrollment period in HealthCare.gov states reported income between 100 percent and 150 percent FPL, qualifying for fully-subsidized, 94 percent actuarial value plans. The percentage of people signing up who report income in this range has increased substantially since the enhanced subsidies took effect. Among enrollees reporting income between 100 and 150 percent FPL in HealthCare.gov states, we estimate that 62.3 percent are not eligible. In other words, for every two eligible enrollees in this income category, there are more than three who are ineligible.

In 29 states, the number of sign-ups reporting income between 100 percent and 150 percent FPL exceeds the number of potential enrollees. ¹

The problem is particularly acute in Florida, where we estimate there are nearly five times as many enrollees reporting income in that range as residents who are actually income-eligible. In 14 other states—Arizona, Georgia, Indiana, Louisiana, Michigan, Mississippi, Missouri, North Carolina, Ohio, Oklahoma, South Carolina, Tennessee, Texas, and Utah—there are more than twice as many enrollees as eligible individuals.

In all states, there is an incentive for people who have income between 200 and 400 percent of the FPL to report income between 100 and 150 percent of the FPL. First, they qualify for a larger advance subsidy and a plan with much lower cost-sharing. Second, the Internal Revenue Service only recoups a small portion of the excess subsidy when they file their taxes.

As we reported in our first paper, the perverse incentives of the system have attracted unscrupulous brokers. These brokers understand how to game Obamacare. They are driving fraudulent enrollments by exploiting the enhanced direct enrollment feature of HealthCare.gov, which allows third parties to enroll consumers without their direct interaction with the exchange. In addition, the temptation to run scams only grew as the Biden administration sought to sign up as many as possible. For many brokers, this opportunity for fraud proved too easy and too profitable to pass up. All they needed was a person’s name, date of birth, and address to enroll them in coverage. There has been widespread media coverage of this problem, and recently, the Department of Justice has uncovered two major fraud rings in Florida.

Another group of major beneficiaries of the surge in improper enrollments are health insurers. By misestimating income, enrollees get larger subsidies. The larger subsidies mean that consumers are less sensitive to prices of plans and are more likely to enroll. It is also much easier for insurers to collect subsidies from the U.S. Treasury than from customers. In many cases, enrollees are actually unaware of their enrollment in these plans.

In 2025, 45 percent of Obamacare exchange enrollees—nearly 11 million people—took no affirmative action during open enrollment; instead, they were automatically re-enrolled. The soaring levels of automatic re-enrollment mean that improper eligibility determinations are perpetuated from one year to the next. We also discuss a skyrocketing number of enrollees for whom the government does not know their race or ethnicity, with the share of unknown responses nearly twice as high in HealthCare.gov states. Such a high number of enrollees with missing race or ethnicity information is consistent with widespread improper enrollment.

Only two states—North Carolina and Virginia—saw a decline in fraudulent exchange enrollment from 2024 to 2025. North Carolina’s decline—despite still having substantial improper enrollment—is attributable to its recent Medicaid expansion, while Virginia’s is linked to its transition to a state-based exchange.

What We Recommend

Our analysis of the latest data from the exchanges underscores the urgency of our recommendations from a year ago. Fortunately, the Trump administration is taking a different approach from the Biden administration and has proposed a regulation to reverse the previous administration’s enrollment-at-any-cost policies. And The One Big Beautiful Bill, which passed the House of Representatives in May, would codify the provisions of the Trump administration’s rule and implement additional vital integrity measures, including ending automatic re-enrollment and requiring full subsidy recapture if insurers receive too much advance subsidy.

If the provisions of The One Big Beautiful Bill become law, five of Paragon’s six recommendations from The Great Obamacare Enrollment Fraud would be implemented. Congress should ensure that these important reforms are included in the final legislation that reaches President Trump’s desk. We also recommend that Congress consider ways to penalize insurers and brokers who facilitate improper enrollment, which would better align their financial incentives to comply with the law’s requirements. The remaining reform, and the most important, is for Congress to permit the enhanced subsidies to expire after 2025. In total, this policy combination would significantly reduce improper enrollment in the exchanges, reducing waste, fraud, abuse, and corporate welfare, and protecting taxpayer resources so they are preserved for those most in need. The integrity of federal health programs depends on restoring accountability to the exchanges.

Last year, Paragon research exposed widespread fraud in the Obamacare exchanges resulting from fraudulent enrollments nationwide.² This paper shows that the scandal grew significantly worse in 2025. Using government data, we found that more than one-quarter of exchange enrollees are fraudulently enrolled this year. The cost of this fraud for taxpayers will likely exceed $27 billion this year. To protect taxpayers and ensure that eligible Americans get the health care support the program is intended to provide, Congress and the Trump administration should act now to clean up this mess.

Obamacare exchange enrollment fraud is overwhelmingly concentrated in states that use HealthCare.gov, the federal exchange, to enroll individuals—as opposed to states that operate their own exchanges. Enrollment fraud is also prevalent in states that have not expanded Medicaid. Previous Paragon research has documented extensive improper enrollment in Medicaid expansion programs.³

Exchange enrollment fraud is driven by a combination of four factors:

1. the enhanced Obamacare subsidies, enacted as a temporary pandemic-era policy, which fully subsidize plans for people who report income between 100 and 150 percent of the FPL;
2. Biden administration policies that prioritized enrollment over eligibility verification and program integrity;
3. limits on how much the federal government can recover if insurers receive excess advance subsidies because of improper eligibility determinations; and
4. automatic re-enrollment, which perpetuates improper enrollments year after year.

As a result of these four powerful forces, applicants, brokers, and insurers have significant financial incentives to misrepresent information or lie to maximize subsidies, commissions, and profits.

Last year, Paragon estimated that there were about 4.8 million people enrolled in the 100-150 percent FPL category who did not have that income. This fraud cost taxpayers more than $20 billion in 2024. In its recent proposed rule, the Centers for Medicare and Medicaid Services (CMS) cited and confirmed our methodology, concluding that “the number of plan selections for people with household incomes between 100 and 150 percent FPL exceeds the population of people at that income level based on U.S. Census Bureau surveys.”⁴ CMS’s methodology estimated total improper enrollment at a level consistent with our study. Based on updated Census Bureau data, we have revised our estimate of improper enrollees in 2024 to 5.0 million people, as there were somewhat fewer 19-64-year-olds with incomes between 100 and 150 percent FPL.

Since releasing our original research a year ago, fraud in the exchanges has worsened significantly. Using the same conservative assumptions as in last year’s report, we now estimate about 6.4 million people improperly claimed income between 100 and 150 percent FPL during the 2025 open enrollment period, with improper expenditures likely exceeding $27 billion in 2025 alone. Using more expansive assumptions, improper enrollment reaches about 7.1 million people.

According to open enrollment data, 55 percent of exchange sign-ups reported income between 100 and 150 percent FPL in the 2025 open enrollment period. Although CMS took steps in mid-2024 to block unscrupulous brokers from signing people up for coverage, these efforts have proven ineffective given policies that incentivize fraud. This is likely due, in large part, to automatic re-enrollment in 2025 of people who were first improperly or fraudulently enrolled in 2024 or prior years.

Congress and the Trump administration owe it to hard-working American families—and to those truly eligible—to reduce rampant fraud and restore program integrity in the Obamacare exchanges. At the end of this paper, we include a series of recommendations to reform the program.

The Trump administration has already taken some important steps forward. The administration proposed a rule to address the harmful legacy of several Biden administration policies. In the One Big Beautiful Bill (OBBB), Congress would codify the key elements of the proposed rule, including tightening enrollments and requiring exchanges to check applicant eligibility prior to enrollment. Congress proposes important additions to that rule, including prohibiting automatic re-enrollment and requiring that applicants repay their subsidies if they misestimated income and received more subsidy than they were entitled to. Congress should also consider holding brokers and insurers that profit from improper enrollment financially liable for high error rates. And perhaps most importantly, Congress should let the enhanced Obamacare subsidies expire after 2025.

During open enrollment, people sign up for Obamacare plans sold through the federal exchange (HealthCare.gov) or state exchanges. Obamacare significantly increased the cost of coverage, making subsidies essential for most enrollees to afford coverage. These subsidies, which take the form of premium tax credits (PTCs), limit the portion of a benchmark plan’s premium that a household must pay to a percentage of its income.⁵ As a result of legislation first enacted by only Democrats in 2021 during the pandemic, someone in a household with incomes between 100 and 150 percent FPL qualifies for subsidies that pay for the entire premium for a health insurance plan that has a very low deductible and low cost-sharing amounts.

The subsidies come primarily in the form of advance PTCs (APTCs). The APTC is based on estimated income. During the annual open enrollment period, applicants estimate their household income for the following year, often with the assistance of brokers or navigators. The government then sends a monthly APTC to the insurer offering that plan.

When a person files his or her subsequent tax return (generally in April of the year after the coverage), the APTC an enrollee received is reconciled with the amount they were actually entitled to based on reported income on their tax return. People who received excess subsidies must repay a portion back when they file their taxes, subject to significant limits. Those who received smaller APTCs than they were eligible for would receive additional credit against their taxes.

In Medicaid expansion states, able-bodied, working-age adults with income below 138 percent FPL are eligible for Medicaid. Obamacare prohibited anyone eligible for Medicaid from receiving a PTC to purchase a plan in the exchange. Therefore, in expansion states, only enrollees who estimate their income between 138 percent and 150 percent FPL are eligible for fully subsidized benchmark plans, meaning that the entire cost is borne by taxpayers (now and in the future given the massive federal budget deficits). In states that have not expanded their Medicaid programs, enrollees with income between 100 and 150 percent FPL are eligible for fully subsidized benchmark plans, because able-bodied, working-age enrollees are typically not eligible for Medicaid in those states. If their reported income is below 100 percent FPL, they are ineligible for PTCs.

The law limits the amount that people need to repay when they file their taxes. For 2025, the amount that single filers must repay is capped at $375 if their income is between 100 and 200 percent FPL, $975 for those between 200 percent and 300 percent FPL, and $1,625 for those between 300 percent and 400 percent FPL.⁶ People with income above 400 percent FPL would need to fully reconcile the APTC amounts with the PTC amounts to which they were entitled.

The combination of enhanced subsidies and low subsidy recapture amounts creates significant incentives for people to misstate income—many times at the direction of unscrupulous brokers. Figure 1 demonstrates the incentive to misstate income and how it varies by age. The figure includes an estimate of the taxpayer cost for enrollees, or brokers or agents, to misstate enrollee income to between 100 and 150 percent FPL to qualify for a fully subsidized plan with an actuarial value of 94 percent.⁷

People generally qualify for larger subsidies if they underestimate their income. This figure illustrates the benefits of income underestimation beginning at 200 percent FPL, because the incentive to underestimate income for households below 200 percent FPL is minimal. The benefit gradually increases as household income increases until the benefit ceases at 400 percent FPL, where there is full subsidy recapture.

Here are some examples. For a 40-year-old enrollee at 290 percent FPL, the incentive for estimating income at just under 150 percent FPL is $1,471 on average. He would receive an APTC of $5,958 to cover the full premium of insurance coverage with an actuarial value of 94 percent.⁸ At 290 percent FPL, he was eligible for a PTC of $3,512—receiving $2,446 of excessive subsidy for a much less generous coverage with a 70 percent actuarial value. He would need to repay $975, which would leave him better off by $1,471 in premium subsidies due to underestimating his income—and from the added benefit of coverage with significantly reduced cost-sharing.⁹ For enrollees over 200 percent FPL, the only differential in the benefit of misstating income based on age occurs because the value of the cost-sharing reduction subsidy is greater for older enrollees than younger enrollees.

In non-expansion states, however, there are also incentives for people to overstate their income, as Obamacare bars PTCs to people with incomes below the poverty line in those states. In this circumstance, the government deems them to be qualified for the PTC so long as the income estimates were not made “with intentional or reckless disregard for the facts.”¹⁰ By overstating their income, these individuals get generous coverage at zero cost to themselves—instead of being ineligible for any subsidies.

Given how the subsidy structure works, there is a massive incentive for older people in non-expansion states who earn income below 100 percent FPL to overestimate income in order to qualify for subsidies. The PTC structure limits premium payments to a certain percentage of household income, regardless of the premium amount. Because premiums are three times more for enrollees near age 65 than for enrollees in their 20s, the subsidies are also much larger. Nationally in 2025, the average PTC for a 21-year-old at 150 percent FPL is $4,662, and the average PTC for a 64-year-old is $13,986.¹¹ Older enrollees typically use more medical services, all else being equal, and some may be looking to retire before the age of 65. These factors create a stronger incentive for them to misreport income to qualify for greater subsidies.

Brokers and Insurer Engagement and Tolerance of Fraud

Health insurers profit substantially from improper enrollment. Obamacare facilitates premium collection for insurers by sending subsidies directly from the U.S. Treasury. If a plan is fully subsidized, the only cost to enrollees is the paperwork burden. This means that people have the incentive to enroll even if they receive very low benefit from the plan. Automatic re-enrollment exacerbates this problem by re-enrolling people for an additional year, in many cases, even if they get other coverage, move out of state, or have died. For enrollees in all states, 45 percent were automatically re-enrolled in coverage in 2025, up from 31 percent of all enrollees in 2024.¹² All this leads to large payments to health insurers on behalf of many people who are likely receiving low or no value from the coverage.

Obamacare also forces the enrollees—not the insurers—to bear the cost of paying back excessive subsidies. Even though the payment goes directly from the U.S. Treasury to the insurer, the payment is considered a tax credit for the enrollee. So the liability for subsidy overpayments is on the enrollees when they reconcile their taxes (if they file their taxes).

Many brokers and agents have exacerbated the problem of fraudulent enrollment. These entities maintain contracts with insurers that entitle them to a commission for each person they enroll. The more people they enroll, the greater their commissions from the insurers.

Fraudulent activity can often be found as approved brokers work within the system, such as through enhanced direct enrollment. In enhanced direct enrollment, brokers host their own eligibility applications and send enrollment information to Healthcare.gov.¹³ This is unlike traditional direct enrollment platforms, where brokers send enrollees to Healthcare.gov to file their eligibility application directly, then return to the broker’s site to select a health plan. Enhanced direct enrollment platforms often miss critical information — like Social Security numbers — yet made up 81 percent of all broker-assisted enrollments in 2023.¹⁴

In online ads on Facebook and other platforms, unscrupulous agents and brokers lure Americans to hand over their personal information with promises like: “[P]ut $6,400 in your pocket right now for free.”¹⁵ The agents then use that information to enroll people in ACA coverage, even if they did not consent to signing up for insurance. One consumer, Angela Wells, clicked on one ad that promised cash cards for groceries. Even though she refused to sign up for insurance over the phone, the agent switched her plan anyway. She found out only when her pharmacy said her insurance had been cancelled and replaced with a plan—one with significantly higher copayments.¹⁶

The Wall Street Journal reported that CMS had received 208,000 complaints of unauthorized sign-ups in 2024—just through September.¹⁷ In April 2024, individuals improperly enrolled in plans brought a class action lawsuit against insurance agents.¹⁸ More recently, in the Trump administration’s proposed rule, CMS noted that in 2024, HHS received 44,151 complaints (including 12,954 “medically urgent” ones) that Medicaid beneficiaries were not able to access their benefits because rogue brokers had signed them up for ACA plan without their knowledge.¹⁹ The Biden administration’s policies made this type of fraud easier. Special enrollment periods (SEPs) let people sign up for coverage outside of the normal open enrollment. While SEPs are usually limited to certain events, such as losing coverage or getting married, the Biden administration created a SEP that allowed individuals with incomes between 100 and 150 percent FPL to enroll at any time.²⁰ The Wall Street Journal reported that the expansion of SEPs “created a continuous opportunity to make sales” and “the boost in subsidies gave agents an enticing sales pitch: zero-premium health insurance.”²¹

CMS responded to the public outcry around broker fraud with actions to curb plan switches by requiring a three-way phone call with the broker and a CMS official. CMS also increased enforcement, suspending “200 agents or brokers for suspicion of fraud.”²² Our new findings show that such efforts, although well-intentioned, do not appear to have had an impact on stemming the tide of fraudulent enrollments.

Over the past few months, there has been significant legal action related to this fraud. In February 2025, the Department of Justice (DOJ) unsealed an indictment against two executives—a president of an Obamacare brokerage firm and a CEO of an insurance marketing company—alleging that the two executives had orchestrated a massive enrollment fraud scheme that swindled taxpayers out of “at least $161.9 million in government subsidies.”²³ The indicted individuals face more than 50 years in prison for charges related to wire fraud, conspiracies to defraud the United States government, and money laundering. DOJ alleges that the defendants deliberately “conspired to enroll consumers in ACA [Affordable Care Act] plans that were fully subsidized by the federal government by submitting false and fraudulent applications for individuals whose income did not meet the minimum requirements to be eligible for the subsidies.”²⁴

In April 2025, DOJ announced that a Florida insurance brokerage executive, Dafud Iza, pled guilty to a scheme to “fraudulently enroll ineligible individuals into ACA plans that offered tax credits to eligible enrollees.”²⁵ According to DOJ, “Iza and his accomplices deceptively marketed subsidized ACA plans to ineligible consumers and falsely inflated consumers’ incomes to obtain the federal subsidies.” They even targeted vulnerable individuals, such as “low-income individuals experiencing homelessness, unemployment, and mental health and substance abuse disorders.”²⁶

Figure 2 demonstrates the shift in overall enrollment to the lowest income category in the states that use HealthCare.gov. The first open enrollment period with fully subsidized plans was 2022. By 2025, 55 percent of people who signed up for coverage during open enrollment reported that their income was between 100 and 150 percent FPL. This figure shows only the federal exchange sign-ups, because not all states with state-based exchanges reported sign-ups by income grouping prior to 2022.

In May, CMS released the 2025 Obamacare open enrollment report.²⁷ The report sheds light on the growth of fraudulent and improper enrollment. One piece of data is on applicant race and ethnicity. The number of enrollees who did not report race or ethnicity skyrocketed, which reflects how many enrollees were most likely signed up for coverage without their consent by rogue brokers.

In 2020, 28 percent of enrollees did not report their race or ethnicity. This steadily increased to 50 percent of enrollees by 2024, a percentage that remained the same in 2025. Figure 3 shows enrollment from 2020 to 2024 based on applicant race or whether race is unknown. It shows a surge in the number of enrollees who do not report their race or ethnicity. The sudden spike in non-reporting of race or ethnicity is suspicious and strongly suggests fraud or improper enrollment, often involving brokers enrolling individuals without their consent.

In HealthCare.gov states, the race and ethnicity is unknown for 57.4 percent of enrollees. This is significantly higher than state exchanges, where 34.1 percent of enrollees’ race or ethnicity was not reported or was unknown.

Table 1 shows the sign-ups during open enrollment broken down by expansion states versus non-expansion states and states using the federal exchange (HealthCare.gov) versus those states that established their own exchanges in 2024 and 2025. Table 1 also shows the estimate of people eligible for exchange plans with incomes between 100 and 150 percent FPL. The estimate includes only people ages 19-64, as children in that income range are eligible for either Medicaid or the Children’s Health Insurance Program, and seniors are eligible for Medicare. We exclude people who report having either Medicare or Medicaid coverage, but we did not exclude people who report that they have employer coverage. In 15 states—Arizona, Florida, Georgia, Indiana, Louisiana, Michigan, Mississippi, Missouri, North Carolina, Ohio, Oklahoma, South Carolina, Tennessee, Texas, and Utah—there are more than twice as many enrollees as eligible individuals.

We excluded three states—Minnesota, New York, and Oregon—that use the Basic Health Program (BHP).²⁸ The BHP is a state-administered, federally funded Obamacare option that provides affordable insurance to low-income individuals who do not qualify for Medicaid but earn up to 200 percent FPL. In at least one of those states, New York, there is evidence of far more enrollees than are eligible. Bill Hammond of the Empire Center estimated that in 2023, 5.5 million individuals were eligible for either Medicaid or the BHP-equivalent program in New York based on census income data, but the programs’ enrollment records show an additional 3.4 million individuals receiving coverage. Hammond estimates that this improper enrollment could cost $20 billion or more per year to New York and federal taxpayers.²⁹ Thus, by excluding states that use the BHP, we are undercounting the extent and expense of the fraud problem nationally.

There is improper enrollment in every state, given the complexities of Obamacare’s advance subsidy structure, but we find substantial problems in 29 of the 47 states. Overall, there were 10.89 million sign-ups in 2025 that reported income between 100 and 150 percent FPL. We estimate that 6.37 million of these enrollees—or 58.5 percent— were improperly classified. Total fraudulent enrollment in this income grouping is up from 5.01 million enrollees in 2024. The taxpayer cost of fraudulent enrollment in the exchanges likely exceeds $27 billion in 2025 alone. To put that $27 billion in perspective: It would fully fund the National Institutes for Health for more than six months.

A significant share of the fraudulent enrollment problem is almost certainly due to the automatic re-enrollment policy, which puts fraudulent initial determinations on autopilot. In 2025, this had already reached an astounding level as 10.84 million enrollees were automatically re-enrolled in coverage from 2024 to 2025—an increase of 65 percent over previous year. Figure 4 demonstrates the growth in automatic re-enrollment over time. Automatically re-enrolled people represented 45 percent of all 2025 enrollees, up from 29 percent of all enrollees in 2020.

Estimates of Improper Enrollment in Non-Medicaid Expansion States

We make separate estimates for improper enrollments in Medicaid expansion states versus non-Medicaid expansion states.³⁰ In non-Medicaid expansion states, estimating improper enrollment is a simple calculation—the number of 100 percent to 150 percent FPL sign-ups minus the total potential enrollees (i.e., the number of 19 to 64-year-olds with income in that category as reported by the American Community Survey who do not report having Medicare or Medicaid).³¹ Our estimates are conservative since we also assume that all those who meet these criteria are enrolled in a plan and some people in every non-expansion state with income between 100 and 150 percent FPL will not be enrolled in coverage.

In eight of the 10 states that have not expanded Medicaid, there are more sign-ups claiming income between 100 and 150 percent FPL than are eligible. As discussed above, in states that did not expand their Medicaid programs, there is a significant incentive for enrollees, brokers, and insurers to have applicants inflate their estimated income to be at or above 100 percent FPL in order to qualify for fully subsidized, low cost-sharing plans. There is also a significant incentive for applicants, brokers, and insurers to have higher-income enrollees underestimate income to gain more subsidy payments.

Overall, in the 10 non-expansion states, there were 7.88 million enrollees claiming income between 100 and 150 percent FPL compared to 2.87 million people who fit the eligibility criteria for subsidized exchange plans with income in that category. These respective numbers have increased from last year—6.89 million enrollees and 2.83 million, respectively. In non-expansion states, we estimate 5.04 million fraudulent enrollees in 2025—an increase from 4.12 million fraudulent enrollees in 2024.

Florida is a clear outlier, with nearly five times as many people enrolled in the 100 to 150 percent FPL category than are eligible. We estimate 2.46 million ineligible enrollees in Florida. There are more than twice as many enrollees than are eligible in six other states—five of them are non-expansion states (Georgia, Mississippi, Texas, South Carolina, and Tennessee). There are two non-expansion states where there is no evidence of widespread improper enrollment—Wisconsin and Wyoming.

Estimates of Improper Enrollment in Medicaid Expansion States

In expansion states, the estimation is more difficult. The most conservative approach—as well as one consistent with the methodology in “The Great Obamacare Enrollment Fraud”— is to define improper enrollment as any enrollment exceeding half the number of potential enrollees. This is the number of potential enrollees shown in Table 1. In states that expanded Medicaid to able-bodied, working-age adults, people between 100 percent and 138 percent FPL are eligible for Medicaid, so only those with income between 138 percent and 150 percent FPL should be signed up for coverage in the fully subsidized category. Assuming an even distribution of enrollees in that income category, only about one quarter of such people would be eligible for the exchanges.

Under the methodology of half of people between 100 and 150 percent FPL being eligible (consistent with Table 1), improper enrollment in expansion states increased from 900,000 enrollees to 1.32 million enrollees from 2024 to 2025. Our assumption that one-half are eligible is thus very conservative, and the improper enrollment problem in the exchanges in expansion states is almost certainly higher than what we report.

Improper enrollment is growing in these states—possibly indicating dual enrollment in Medicaid and subsidized exchange plans. Among expansion states, the problem is most severe in the following:

• North Carolina, with an estimated 303,001 improper enrollees
• Ohio, with an estimated 141,578 improper enrollees
• Utah, with an estimated 118,094 improper enrollees
• Missouri, with an estimated 112,764 improper enrollees
• Michigan, with an estimated 108,178 improper enrollees
• Louisiana, with an estimated 104,184 improper enrollees
• Indiana, with an estimated 91,074 improper enrollees
• Oklahoma, with an estimated 85,554 improper enrollees
• Illinois, with an estimated 46,369 improper enrollees
• New Jersey, with an estimated 38,806 improper enrollees
• Virginia, with an estimated 26,084 improper enrollees

Improper Exchange Enrollment Is Much Greater in HealthCare.gov States

Table 2 breaks fraudulent enrollment into whether states use HealthCare.gov or whether they operate their own exchanges as well as whether they expanded Medicaid or not. Consistent with the discussion above, enrollment fraud is much more severe in non-expansion states because of the incentives to enroll people who do not otherwise qualify for Medicaid. Table 2 also shows how much more severe the problem is in HealthCare.gov states. In fact, large-scale improper enrollment is virtually absent in states with their own exchanges—all of which expanded Medicaid.

The concentration and scale of improper enrollment in HealthCare.gov states is an indictment of the Biden administration’s management of the program over the prior four years. The Biden administration’s mismanagement has left the Trump administration with a massive clean-up operation. In states using HealthCare.gov, there were 6.29 million fraudulent enrollees in 2025, up from 4.97 million in 2024. In states using HealthCare.gov, a staggering 62.3 percent of enrollees reporting income between 100 and 150 percent FPL are improper. In other words, for every two eligible enrollees in this income category, there are more than three who are ineligible.

Some might argue that improper enrollment is less of a HealthCare.gov problem, and more of a refusal-to-expand Medicaid problem. However, exchange enrollment is a substantial problem in states that both expanded Medicaid and use HealthCare.gov. Controlling for Medicaid expansion shows that improper enrollment is much more severe in HealthCare.gov states. In fact, in expansion states that use HealthCare.gov, there were more enrollees reporting income between 100 and 150 percent FPL than have such income in 2025.

As evidence of the greater enrollment fraud in HealthCare.gov states, the government does not know the race or ethnicity for 57.4 percent of enrollees in HealthCare.gov states compared to 34.1 percent of enrollees in states with state-based exchanges. In Florida, 2.95 million enrollees have an unknown race or ethnicity (62.3 percent of all enrollees). In Georgia, it is 67.6 percent of all enrollees; in South Carolina, it is 63.2 percent; in Missouri, it is 62.1 percent, and in Tennessee, it is 60.5 percent.

We adopted a conservative methodology that likely undercounts fraudulent enrollment in state-based exchanges, all of which are in Medicaid expansion states. We find that the only states with state exchanges that had improper enrollment in 2025 (according to our conservative methodology) were New Jersey, Virginia, and Massachusetts—and the total improper enrollment was just 72,000 people.

According to CMS, in 2025, 96 percent of HealthCare.gov applicants were deemed eligible versus 49 percent of state-based exchange applicants requesting coverage.³² This strongly suggests serious eligibility determination problems in HealthCare.gov states.

Biden Policies That Exacerbated Improper Enrollment and Fraud on Federal Exchanges

This greater improper enrollment is likely due to several policies that make the federal exchanges more vulnerable to fraudulent activity. The Biden administration made several policy changes that increased improper enrollment in all states. For example, it required all exchanges to accept enrollees’ self-attestation of income if they could not confirm reported income through alternate data sources, like tax filings.³³ Since individuals who earn less than 100 percent of FPL usually do not file taxes, such individuals could misstate their income to qualify for the fully subsidized plans, and exchanges generally take them at their word, i.e., self-attestation.

However, the Biden administration’s rules particularly impacted federal exchanges, which helps explain why fraudulent enrollment is so much more extensive in those states.

Chief among those policy changes is the lack of pre-enrollment verification for SEPs on the federal exchanges. Starting in 2023, the Biden Administration stopped requiring applicants on federal exchanges to submit proof that they qualify for almost all SEPs.³⁴ By contrast, most state-run exchanges require some form of documentation. Some states—including Pennsylvania,³⁵ Connecticut,³⁶ and Virginia³⁷—require documentation before coverage even starts for most SEPs.

On top of this lack of pre-enrollment verification, all federal exchanges have a low-income SEP that specifically targets individuals earning between 100 and 150 percent FPL. Although most state exchanges chose to implement this policy (except Virginia, Maryland, and Nevada),³⁸ some of them have at least some form of pre-enrollment verification for SEPs that could help prevent fraudulent enrollment.

Similarly, the Biden administration implemented a more aggressive automatic re-enrollment policy on the federal exchanges. If an individual’s bronze plan is no longer available, the federal exchanges will automatically move individuals to a silver plan if the silver plan has a lower premium.³⁹ While states are allowed to implement that policy after the 2024 Notice of Benefit and Payment Parameters (NBPP),⁴⁰ some states, such as Virginia, have not.⁴¹ This automatic shifting to zero-premium plans increases the number of people who never receive a bill, raising the likelihood that enrollees will be re-enrolled automatically—even if their circumstances have changed and they now have other coverage.

Finally, the Biden administration also implemented several policies to make it more difficult for federal exchanges to remove ineligible enrollees. During the pandemic, CMS stopped marking people as ineligible if they failed to file taxes and reconcile their APTCs, as they are required to do by law.⁴² During that time, state exchanges had the discretion to enforce that requirement. In the 2024 NBPP, the Biden administration prohibited all exchanges (both state and federal) from deeming enrollees ineligible until they had failed to file taxes or reconcile for two consecutive years.⁴³ However, CMS stated that federal exchanges would still allow consumers to get coverage, even after they failed to file and reconcile for two straight years. They simply have to self-attest that they filed.⁴⁴

Table 3 shows the fraction of exchange enrollees who report their income between 100 and 150 percent FPL by state. Nationally, the rate was 45.8 percent. It was 55 percent in HealthCare.gov states, and much lower in states with their own exchanges. Unsurprisingly, non-expansion states have the highest percentage of exchange enrollees who report income between 100 and 150 percent FPL. But there are several Medicaid expansion states in which at least 40 percent of exchange enrollees report income in this range, which is suspicious, as enrollees with income between 100 percent and 138 percent FPL qualify for Medicaid, not exchange plan subsidies. These states are Louisiana (52.2 percent of all enrollees reporting income between 100 and 150 percent FPL), Oklahoma (48.3 percent), Missouri (46.3 percent), North Carolina (45.8 percent), Indiana (43.8 percent), and Ohio (41.1 percent). Some of these states expanded Medicaid recently, which suggests that automatic re-enrollment may be a key factor keeping improper enrollment levels high.

Georgia, North Carolina, and Virginia

In November 2024, Georgia transitioned from HealthCare.gov to a state-based exchange. The transition did not reduce improper enrollment. In fact, estimated improper enrollment increased from 504,524 in 2024 to 543,642 in 2025. Much of the problem in Georgia (and other states) stems from fraudulently enrolled people being automatically re-enrolled in subsequent years.

Even though Georgia is now a state-based exchange, for this paper, we still classify Georgia as a HealthCare.gov state to prevent biased results. If we classified Georgia as a state-based exchange state, it would dramatically raise improper enrollments in state-based exchange states. This would bias the results, particularly given that the current high level of fraud in Georgia’s program was perpetrated before 2025 and carried forward through automatic re-enrollment.

From 2024 to 2025, the share of fraudulent enrollees relative to potentially eligible individuals declined in only two states: North Carolina and Virginia North Carolina stands out in the fraud analysis. The 2025 open enrollment period was only the second since the state expanded Medicaid in November 2023. Exchange enrollees with income between 100 percent and 138 percent FPL should have transitioned to Medicaid. This should have occurred in the 2024 open enrollment period, but we estimate there were 365,707 improper exchange enrollees that year. Despite being in the second year of Medicaid expansion, we estimate there are still 303,001 improper enrollees in the 100-150 FPL category in North Carolina in 2025.

In 2025, there were more people enrolled in North Carolina’s exchange reporting income between 100 and 150 percent FPL and Medicaid expansion than were eligible. In 2025, there were 589,371 Medicaid expansion enrollees; 446,367 exchange enrollees with incomes between 100 and 150 percent FPL; and 874,469 people with incomes below 150 percent FPL. Thus, even under the extreme assumption that the state enrolled every eligible person with income below 150 percent FPL into Medicaid expansion or the fully subsidized exchange plan category, there are still 161,269 more enrollees than potentially eligible. The combined enrollment-to-eligibility ratio for Medicaid and exchange plans below 150 percent FPL for those programs grew from 92.1 percent in 2024 to 118.4 percent in 2025.⁴⁵ In fact, exchange enrollment in North Carolina climbed from 2023 to 2025, from 800,850 to 975,110 despite the expansion adding 600,000 to Medicaid. Moreover, despite Medicaid expansion, there are over 100,000 additional enrollees in the 100 percent to 150 percent FPL category in 2025 than in 2023. All of this is staggering evidence of improper enrollment in both Medicaid expansion and the exchanges in North Carolina.

A notable shift in reported income in North Carolina occurred between 2024 and 2025, with those reporting income between 100 percent and 138 percent FPL dropping from 22 percent to 12 percent, and a commensurate increase in those reporting income between 138 percent and 150 percent—from 11 percent to 20 percent. This illustrates how those who may otherwise be ineligible for exchange plans, and unscrupulous brokers and insurers, are exploiting the system’s lack of integrity for financial gain.

Virginia offers the best example of improvement. Fraudulent enrollment decreased from 36,989 enrollees after the 2024 open enrollment period to 26,084 enrollees after this past open enrollment period. Total exchange enrollment also decreased in Virginia from 2024 to 2025, the only state other than North Carolina (which is explained by the state adopting Medicaid expansion). Virginia’s performance is likely, in part, due to using a fully state-based exchange starting in 2024.⁴⁶ As a result, Virginia has had the flexibility not to adopt the Biden administration’s policies, but also to implement and control its own program integrity measures.⁴⁷ For example, for plan year 2023, Virginia still used HealthCare.gov,⁴⁸ which was subject to the Biden administration’s policy of no longer requiring pre-enrollment verification for all but one SEP.⁴⁹ However, when Virginia switched to becoming a state-based exchange for the 2024 plan year, it was able to resume pre-enrollment verification. Virginia currently indicates it uses pre-enrollment verification for multiple SEPs.⁵⁰ This policy may have helped reduce fraudulent enrollments.

Total Cost of the Fraudulent Exchange Enrollment

Using conservative assumptions, we estimate that 6.37 million people improperly enrolled for subsidized health coverage on the exchanges in 2025. We estimate that in non-expansion states, two thirds of fraudulent enrollees have income below 100 percent FPL. That equates to 3.36 million people. The remaining improper enrollees (those with income above 100 percent FPL) equal 3.01 million people. The average subsidy for a 45-year-old with income between 100 and 150 percent FPL is $6,732, which is the cost of improper enrollment for people with income below 100 percent FPL. For people above 150 percent FPL who underestimated their income, we assume an average loss to the taxpayer of $1,500. Based on this back-of-the-envelope estimate, the total projected loss in exchange enrollment fraud for people claiming income between 100 and 150 percent FPL who do not have that income is $27.1 billion. That estimate would be $28.3 billion under a scenario in which only one-quarter of 100 to 150 percent FPL enrollees were eligible for subsidized exchange plans in expansion states.

Our estimate of improper spending may be too low for several additional reasons. First, we use a $6,732 subsidy for a 45-year-old, but the subsidies are much higher for older people, so we think the average we chose is conservative, particularly given that the incentive to cheat is so much higher for older enrollees and for insurers who collect the subsidies from Treasury, as well as brokers who collect commissions from insurers. In fact, unscrupulous brokers may inflate applicant age to capture higher compensation. Our estimate does not consider any improper enrollment in two non-expansion states—Wisconsin and Wyoming—as well as 16 other expansion states that do not meet the criteria of more than 50 percent enrollment among people between 100 and 150 percent FPL. Our analysis does not include Minnesota, New York, and Oregon, because they have the BHP. The District of Columbia is also excluded, as the majority share of plan selections did not report income information.

Obamacare was constructed with subsidies targeted on household income—less than 138 percent FPL in Medicaid expansion and between 138 percent and 400 percent FPL. The Supreme Court’s 2012 decision that made Medicaid expansion optional⁵¹ meant that in states that did not adopt expansion, people in households with income above the FPL would be eligible for the subsidies and that people with incomes below that level could not receive federal premium subsidies. When researchers and modelers projected the impact of Obamacare’s coverage expansion, they generally based their estimates on the assumption that insurers, brokers, and enrollees would comply with those rules. It has turned out that the law is not working anywhere close to what was intended.

This paper demonstrates that the subsidy eligibility categories are essentially meaningless, with nearly 6.4 million more enrollees claiming income between 100 and 150 percent FPL than the number of people who actually live in households with that income. And automatic re-enrollment annually perpetuates and expands the fraud.

Previous Paragon research has demonstrated that there is also massive improper enrollment in Obamacare’s Medicaid expansion, in part because states have enormous incentives to classify enrollees as eligible under the expansion.⁵² There are almost certainly millions of people enrolled in these programs now, particularly in states using HealthCare.gov, that are enrolled simultaneously in multiple forms of coverage—Medicaid, subsidized exchange plans, or employer-sponsored insurance.⁵³

As we discussed in “The Great Obamacare Enrollment Fraud,” there has been improper enrollment in the exchanges—particularly in Florida—since shortly after the program took effect. But the fraud exploded during the Biden administration from both the expanded subsidies and the administration’s enrollment-at-any-cost policies. The exchanges are so rife with fraud and abusive spending that correcting the status quo will require fundamental reforms to eligibility review and enforcement.

First, proper program integrity steps would cause enrollment to fall. In fact, enrollment should fall by several million people just by checking eligibility. Second, a large percentage of the disenrolled would not be without coverage, as they are already enrolled in or are eligible for other forms of coverage. As the Congressional Budget Office has noted, the number of individuals with multiple sources of coverage skyrocketed after Congress expanded the ACA subsidies—from 18 million in 2021 to 29 million in 2023.⁵⁴

Third, the decline in fraudulent enrollment would reduce government expenditures. Fourth, the decline in average enrollment would cause average premiums to rise, because insurers are now covering millions in their plans who are not using any health care services, in part because they do not know they are enrolled. However, as CMS notes, some program integrity measures like reforming SEPs should put downward pressure on premiums by improving the risk pool.⁵⁵

Biden administration policies enriched the health care industrial complex and made it too easy for fraudsters and scammers to rip off Americans. The Trump administration and Congress owe it to the American people—those who are eligible for these programs and the hard-working taxpayers who finance them—to take aggressive action to fix the problems and to eliminate policies that make it so easy to cheat. Fortunately, the administration and Congress are taking key initial steps to reduce the fraud.

Congress can enact legislation to clean up a substantial part of the fraudulent enrollment and improper spending in the exchanges. First, Congress should ensure that the Obamacare program integrity provisions in the One Big Beautiful Bill (OBBB), most of which were provisions in the Trump administration’s proposed Marketplace Integrity and Affordability rule, become law. Second, Congress should permit the enhanced Obamacare subsidies to expire on schedule after 2025. Third, Congress should consider financial penalties on brokers and insurers for massive improper eligibility determinations.

The OBBB contains many policies that tighten the enrollment periods and require exchanges to better ensure that only those eligible are enrolled:

• Reducing the open enrollment period back to the timeframe originally proposed by the Obama administration: from November 1 to December 15.⁵⁶
• Removing the SEP for anyone claiming income between 100 and 150 percent FPL. This SEP enables year-round abuse and incentivizes people to wait until they are sick before enrolling, worsening the risk pool and driving premiums up.
• Prohibiting exchanges from accepting self-attested income.
• Requiring exchanges to verify eligibility for PTCs before enrollment.
• Requiring individuals to confirm their coverage decisions each year and ending the automatic re-enrollment process.
• Eliminating the caps on repayment of excess advance PTCs. By requiring exchanges to verify eligibility earlier in the process—before enrollment takes place—this provision would also reduce the risk of unexpected charges due to unresolved data-matching inconsistencies.

In addition to enacting all these important program integrity reforms, the most important action that Congress can take is permitting the enhanced Obamacare subsidies to expire. The expanded subsidies present a host of public policy problems, including significantly increasing the cost of Obamacare, discouraging small employers from offering coverage, and incentivizing cheating and fraud.⁵⁷ The expanded subsidies also reduce insurers’ incentives to provide coverage that enrollees value. The best way to increase the overall value from the program and reduce both fraud and wasteful expenditures on people who place extremely low value on the coverage is for Congress to permit the expanded subsidies to expire.

CMS should do a top-to-bottom audit of the entire exchange regime. Every ACA-related policy decision from the Biden administration should be closely scrutinized. One area where CMS should examine is whether children are part of any applications for exchange plans between 100 and 150 percent FPL. They should not be. Children in households with income in that range should be in Medicaid or the Children’s Health Insurance Program, making them ineligible for subsidized exchange plans.

In order to properly align incentives to ensure proper enrollment, policymakers must address the large financial incentives brokers and insurers receive from enrolling people in exchange plans with greater subsidy. One way to do that would be for Congress to require that insurers repay the Treasury at least half of any improper subsidy and to require brokers to surrender their commissions to the Treasury. In many cases, the consumers on whose behalf subsidies are paid are often unaware that they are enrolled in plans. Holding insurers and brokers accountable for fraud is the best way to reduce it, because that is where the money flows. It is also the best way to ensure that taxpayer dollars get to Americans who need it most.