The largest private health insurance companies often offer plans in multiple markets, including the Medicare Advantage, Medicaid managed care, individual (non-group), and fully-insured group (small and large employer) health insurance markets. Each market has unique features, including eligibility, payment, and coverage rules, which affect insurers’ overhead and potential profit. In recent years, private insurers are playing a growing role in public insurance programs, with more than half of eligible Medicare beneficiaries enrolled in a private Medicare Advantage plan and more than three-quarters of Medicaid enrollees obtaining coverage through a managed care plan (typically a private insurer).

This brief examines two measures of financial performance – gross margins and medical loss ratios – in the Medicare Advantage, Medicaid managed care, individual, and fully insured group health insurance markets using data reported by insurance companies to the National Association of Insurance Commissioners (NAIC) and compiled by Mark Farrah Associates, through the end of 2024 (the most recent year of annual data). The Medicare Advantage market is made up of around 33M people in 2024. In comparison, this is less than half of the population for Medicaid managed care and about 70% of the size of the fully insured group market. However, the Medicare Advantage market is about 1.4 times larger than the individual market.

In 2024, per enrollee gross margins in dollars were highest in the Medicare Advantage market, and medical loss ratios (measured as percentages) were lowest in the individual insurance market. In 2024, the Medicaid managed care market had both the lowest gross margins per enrollee and highest medical loss ratio. While both gross margins and medical loss ratios are indicators of financial performance, higher margins and lower loss ratios (as they are calculated in this analysis) do not necessarily translate into greater profitability since they do not account for administrative expenses or tax liabilities. Additionally, the increasingly complex structure of insurance companies, including the rise in consolidation and vertical integration, and role of subsidiaries, make it difficult to isolate the revenues and expenses associated with a particular insurance market. (A detailed description of each market is included in the Appendix).

Gross margins

The gross margin per enrollee is the amount by which total premium income exceeds total claims costs per person over a specified time period (i.e., per year).

At the end of 2024, gross margins per enrollee ranged from $608 in the Medicaid managed care market to $1,655 in the Medicare Advantage market. Gross margins per enrollee in the group market was $846, roughly half the level observed among Medicare Advantage plans on average. Per enrollee gross margins in the individual market in 2024 amounted to $987. The level of margins reflects, in part, the overall health needs and spending in a market segment. A similar margin in percentage terms will translate to a higher margin in dollars per enrollee when average health expenses are higher.

Medical loss ratios

Another way to assess insurer financial performance is to look at medical loss ratios (MLRs), or the percent of premium income that insurers pay out in the form of medical claims. Generally, lower MLRs mean that insurers have a higher share of income remaining after paying medical costs to use for administrative costs or keep as profits. Each health insurance market has different administrative needs and costs, so a lower MLR in one market does not necessarily mean that market is more profitable than another market.

MLRs are used in state and federal insurance regulation in a variety of ways. In the commercial insurance (individual and group) markets, insurers must issue rebates to individuals and businesses if their MLRs fail to reach minimum standards set by the ACA. Medicare Advantage insurers are required to report MLRs at the contract level (which typically combines multiple plans) and are required to issue rebates to the federal government if their MLRs fall short of the required level of 85% and are subject to additional penalties if they fail to meet MLR requirements for multiple consecutive years. For Medicaid managed care organizations (MCOs), CMS requires states to develop capitation rates for Medicaid to achieve an MLR of at least 85%. There is no federal requirement for Medicaid plans to pay remittances if they fail to meet their MLR threshold, but a majority of states that contract with MCOs require remittances in at least some cases. The MLRs shown in this issue brief are simple loss ratios (claims as a share of premium income) and may differ from loss ratios calculated using the definition of MLR in the ACA and in Medicaid managed care.

In 2024, MLRs were similar between the Medicare Advantage, Medicaid managed care, and group markets. However, individual market loss ratios were lower. Simple loss ratios were around 85% in individual market, 88% in the fully insured (group) market, 90% in the Medicare Advantage market, and 91% in the Medicaid managed care market.

While gross margins are not equivalent to profitability, changes in gross margins can be indicative of changes in profitability (assuming administrative costs and tax liability are stable). Gross margins have declined from increases that occurred in 2020 during the initial phase of the COVID-19 pandemic. In 2024, all markets saw decreases in gross margins compared to 2023.

Medicaid Managed Care: Per enrollee gross margins in the Medicaid managed care market increased during the pandemic as policies prohibited states from disenrolling people from Medicaid in exchange for additional federal dollars. Gross margins decreased by 19% to $608 from 2023 to 2024, which is slightly higher than in 2019, before the pandemic. Starting in April 2023, “continuous enrollment” in Medicaid ended and states began disenrolling individuals who were no longer eligible or who did not complete the renewal process, and Medicaid/CHIP enrollment declined by more than 16% (about 15 million people) from March 2023 to December 2024. As millions were disenrolled, states and plans faced considerable rate setting uncertainty. A shift in member risk, characterized by an increase in acuity (or health risk) of the remaining population, and increasing utilization patterns began to emerge by late 2023, which may have contributed to the decrease in per enrollee gross margins seen from 2023 to 2024. States may use a variety of risk mitigation strategies, including “risk corridors” (where states and plans agree to share profit or losses), to provide financial protection and limits on financial risk for states and plans that may not be accounted for in the data used in this gross margin analysis. States may also make capitation rate adjustments (with CMS approval) when substantial coverage changes occur mid-year or adjustments are necessary to address unforeseen circumstances that increase benefit costs.

Medicare Advantage: Through the end of 2024, gross margins in the Medicare Advantage market averaged $1,655 per enrollee, which is 17% lower than in 2023 ($1,986), consistent with reports by the largest Medicare Advantage insurers of increased utilization beginning in late 2023 that extended through 2024. Additionally, the phase-in of changes to Medicare Advantage payments, stemming from revisions to how the federal government makes adjustments for the health status of enrollees, began in 2024 and reduced the pace at which revenue per enrollee grew. Per enrollee gross margins have consistently been larger than those in the individual, fully insured, and Medicaid managed care markets since 2018.

Group Market: Gross margins per enrollee for fully insured group plans declined by 7% from $914 to $846 from 2023 to 2024. This is the first time per enrollee gross margins in the fully insured group market have declined from the year prior since 2021, when they were the lowest in the past decade (not shown).

Individual Market: Individual market gross margins were about 5% lower in 2024 compared to 2023, going from $1,042 to $987 per enrollee.  In 2018, following efforts to repeal the ACA and defunding of Cost Sharing Reduction subsidies, insurers raised individual market premiums substantially. These premium increases resulted in significantly higher margins than in earlier years. For context, gross margins per enrollee in 2024 were around 35% and 16% lower than in 2018 and 2019, respectively.

Each health insurance market has different administrative needs and costs, so similar MLRs do not imply that the markets are similar to each other in profitability. Additionally, simple MLRs examined in this brief do not incorporate the effects of changes in tax law, such as the health insurer tax, which has been permanently repealed starting in 2021, was in effect in 2018 and 2020, but was not in 2019. While MLRs alone cannot convey whether a market is profitable in a particular year, if administrative costs hold mostly constant from one year to the next, a change in the MLR could imply a change in profitability.

Individual Market: The average individual market MLR in 2024 was similar to 2023, but higher than those seen in the years following the end of cost-sharing reduction payments. As mentioned earlier, 2018 and 2019 were exceptionally lucrative years for the individual market. Many plans fell short of the ACA’s MLR requirements and were therefore required to issue large rebates to consumers based on their 2018 and 2019 experience.

Group Market: The average MLR for group plans was stable between 2022 and 2023 at 86% but rose to its 2021 value of 88% in 2024. These are all higher than in the years prior, when MLRs ranged from 83% in 2018 and 2020 to 85% in 2019.

Medicaid Managed Care: Relative to 2023, the average MLR in 2024 for the Medicaid managed care market increased from 88% to 91% (implying a potential decrease in profitability). This is the highest Medicaid managed care average medical loss ratio observed in the past decade (data not fully shown). As previously discussed, states and plans faced considerable rate setting uncertainty after millions of people were disenrolled during the unwinding of the pandemic-era Medicaid continuous enrollment provision, resulting in acuity and utilization shifts within the remaining population. These factors may have contributed to the change in MLR seen from 2023 to 2024. Looking ahead, implementation of the 2025 federal budget reconciliation law’s Medicaid coverage and financing provisions could affect Medicaid managed care plans.

Medicare Advantage: Average MLRs in the Medicare Advantage market rose to 90% in 2024. That is higher than before and during the onset of the COVID-19 pandemic, from 2018-2020, when MLRs ranged from 83% to 86%. The increase of the MLR in the Medicare Advantage market could imply decreased profitability, consistent with higher utilization and the effects of phasing in a new risk-adjustment model.  At the same time, it may be difficult to interpret changes in MLRs with increasing consolidation, driven in part by insurers purchasing related businesses, such as pharmacy benefit managers, physician groups, and post-acute care providers, because it is not entirely clear how insurers allocate expenses across different lines of business.

Medicare Advantage plans have both higher average costs and higher premiums (largely paid by the federal government), because Medicare covers an older, sicker population. So, even when Medicare Advantage insurers spend a similar share of their premiums on benefits as other insurers in other markets, the gross margins described above—which include profits and administrative costs—tend to be higher in Medicare Advantage plans.

We analyzed insurer-reported financial data from Health Coverage Portal^TM, a market database maintained by Mark Farrah Associates, which includes information from the National Association of Insurance Commissioners (NAIC). We used the “Exhibit of Premiums, Enrollment, and Utilization” annual report (accessed January 21, 2026) for this analysis. The dataset analyzed in this report does not include California HMOs regulated by California’s Department of Managed Health Care. Additionally, for Medicaid, there are four states (California, Delaware, New York, and Oregon) that have different reporting practices and therefore may only have partial or no NAIC data available for the years displayed.

We excluded plans in each segment of interest that filed negative values or have negative or zero dollars in premiums or claims. We also excluded plans reporting at least 1,000 hospital patient days incurred per 1,000 member months. We only included plans that were categorized as having a “medical” focus in our analysis and exclude “specialty” plans which are categorized as “ancillary or supplemental benefit plans.” We also excluded any plans from the U.S. territories. We corrected for plans that did not file “member months” or filed a zero “member month” value in the annual statement but did file current year membership by imputing these values. If, after imputing, plans still did not have “member months,” they were excluded.

The group market in this analysis only includes fully insured plans (but excludes Federal Employee Health Benefits Program plans and plans regulated by the California Department of Managed Health Care). NAIC defines “Medicaid” as “business where the reporting entity charges a premium and agrees to cover the full medical costs of Medicaid subscribers.” This explicitly excludes Administrative Services Only (ASO) plans. We only use “medical” focused plans to help exclude any specialty plans; however, prepaid ambulatory health plans (PAHPs), prepaid inpatient health plans (PIHPs), or Programs of All-Inclusive Care for the Elderly (PACE) plans may be included in the analysis due to NAIC’s definition of Medicaid.

Gross margins per enrollee were calculated by subtracting the sum of total incurred claims from the sum of unadjusted health premiums earned and dividing by the total number of members.

Premiums for Medicare Advantage plans primarily consist of federal payments made to plans and any additional amounts plans may charge their enrollees. Premiums for Medicare Advantage plans do not include payments for Medicare Part D benefits. Premiums for Medicaid may not reflect contractual adjustments related to risk corridors or other risk-sharing adjustments.

To calculate medical loss ratios, we divided the market-wide sum of total incurred claims by the sum of all unadjusted health premiums earned. MLRs in this analysis are simple loss ratios and therefore, may differ from loss ratios used to calculate rebates.

Individual Market: The individual market includes coverage purchased by individuals and families through the Affordable Care Act’s exchanges (Marketplaces) as well as coverage purchased directly off-exchange, which includes both plans complying with the ACA’s rules and non-compliant coverage (e.g., grandfathered policies purchased before the ACA went into effect and some short-term plans). The federal government provided subsidies for low and middle-income people in the Marketplace and includes measures, such as risk adjustment, to help limit the financial liability of insurers. Insurers in the individual market receive premium payments from enrollees, plus any federal subsidies for people in the Marketplaces.

Some plans submitting data on the Exhibit of Premiums Enrollment and Utilization appear to be including some Children’s Health Insurance Program (CHIP) data in their Individual market filings. In a previous version of this analysis, we used the Supplemental Health Care Exhibit to address this. However, in this analysis, we opted to use the EPEU to ensure comparability.

Group Market: The fully insured group market serves employers, their employees and dependents who are enrolled in fully insured health plans. This market includes both small and large group plans but excludes employer-sponsored insurance plans that are self-funded, which account for 63% of workers with employer-sponsored insurance in 2024. This analysis does not capture metrics for the Federal Employee Health Benefit Program or California Managed Health Care plans.  Roughly 25 million people from the fully insured group market in 2024 are accounted for in this analysis. Plans typically receive premium payments from both employers and their employees.

Medicaid Managed Care: The Medicaid managed care market includes managed care organizations (MCOs) that contract with state Medicaid programs to deliver comprehensive acute care (i.e., most physician and hospital services) to enrollees. As of July 2024, more than three-fourths (over 66 million people) of all Medicaid beneficiaries nationally received most or all of their care from comprehensive risk-based MCOs. There is significant variation across states with respect to services that are covered by MCOs.

In this analysis, the NAIC data we use defines “Medicaid” as “business where the reporting entity charges a premium and agrees to cover the full medical costs of Medicaid subscribers” and only explicitly excludes Administrative Services Only (ASO) plans from their reporting. While we only use “medically” focused plans to help exclude any specialty plans, PAHPs, PIHPs and PACE plans may not be excluded due to NAIC’s definition of Medicaid. Additionally, for Medicaid, there are four states (California, Delaware, New York, and Oregon) that have different reporting practices and therefore may only have partial or no NAIC data available for the years displayed. In other work, KFF defines comprehensive MCOs as managed care plans that provide comprehensive Medicaid acute care services and, in some cases, long-term services and supports as well. This excludes “limited benefit plans” including prepaid ambulatory health plans (PAHPs), prepaid inpatient health plans (PIHPs), and Programs of All-Inclusive Care for the Elderly (PACE) that may be included in this analysis.

Medicare Advantage: The Medicare Advantage market provides Medicare-covered benefits through private plans to around 33 million Medicare beneficiaries in 2024, which is over half of all Medicare beneficiaries in 2024. The federal government makes risk-adjusted payments (higher payments for sicker enrollees and lower payments for healthier enrollees) to plans (averaging nearly $14,823 per enrollee in 2024) to cover the cost of benefits covered under Medicare Parts A and B and supplemental benefits, such as dental, vision, hearing, and others, with additional payments for costs associated with prescription drug coverage. Some plans charge enrollees an additional premium.

Comprehensive, capitated managed care is the dominant Medicaid delivery system, accounting for about 78% of beneficiaries (over 66 million individuals as of July 2024) and 50% of total Medicaid spending (over $458 billion in FY 2024). States were contracting with over 290 individual Medicaid managed care organizations (MCOs) (as of July 2024), which represent a mix of private for-profit, private non-profit, and government plans. Medicaid managed care contracts are among the largest and most complex state contracts, frequently exceeding billions of dollars a year. While the majority of states contract with managed care plans, states decide which populations and services to include in managed care arrangements, leading to considerable variation across states. States have been largely responsible for monitoring and overseeing managed care plans, while federal rules have evolved over time. Historically, publicly available managed care performance data has been limited and inconsistently available across states, limiting transparency and accountability of individual plans.

Sweeping changes to Medicaid managed care rules and regulations in 2016 and 2024 included provisions related to beneficiary protections, access, and program oversight, including new managed care reporting and oversight requirements and efforts to promote transparency. The first Trump administration relaxed some managed care requirements in rules issued in 2020, but the administration left the managed care reporting requirements intact. To date, the Centers for Medicare and Medicaid Services (CMS) has continued to publicly post newly collected state managed care reports on Medicaid.gov. It remains uncertain whether the Trump administration will seek to roll back or revise provisions included in the 2024 managed care final rules.

This brief describes the Managed Care Program Annual Report (“MCPAR”), a relatively new, comprehensive report on state managed care programs that includes plan-level data, that must be submitted (to CMS) by states annually. It functions alongside other managed care reports aimed at improving state and federal managed care program monitoring, oversight, and transparency. (Future KFF analysis will explore policy relevant metrics from the data.)

By introducing the MCPAR reporting requirement, CMS intended to address the fragmented program information it received from states and to improve managed care program oversight. In finalizing the MCPAR requirement in 2016, CMS indicated it expected the reports would provide valuable and timely information to assess managed care program operations in each state and would improve transparency. In its June 2021 guidance, CMS noted the structured data collected from MCPARs (and other newly required managed care reports) would allow for the analysis of state-specific and national data and that CMS could use these data to assist states in improving their managed care programs while also ensuring compliance with federal laws and regulations. See Appendix for additional information on key administrative actions related to the MCPAR.

The first MCPARs were due to CMS between December 2022 and September 2023 (with due dates dependent on the managed care contract period). Following the release of CMS guidance in June 2021 that triggered the 2016 requirement, states are required to submit one MCPAR report for each managed care program the state administers, no later than 180 days after each program’s contract year (which typically operate on a calendar- or state fiscal year-basis). A managed care “program” is defined by a specified set of benefits and eligibility criteria identified in managed care plan contracts.¹ A state may run separate programs for different populations (e.g., children, adults, etc.). As of February 2026, CMS has made MCPAR reports from 2023 and 2024 available on Medicaid.gov. (States are also required to post MCPARs on their websites; however, it is unclear how many states have posted their MCPAR reports. The 2024 rule reaffirmed regulatory transparency requirements and sets a timeframe (within 30 calendar days of submitting to CMS) for posting MCPARs on state websites.)

The MCPAR collects data across many areas of managed care program operations, including plan-level data (Table 1). Notably, the MCPAR template is designed to collect plan-level data in specific areas including enrollment, prior authorization, grievances and appeals, quality measures, encounter data submissions, sanctions, program integrity, and financial performance. While there are some limitations (discussed in more detail below), plan-level data can be used to analyze and compare individual plan performance within and across states. In addition, the MCPAR template collects state- and program-level data – for example, providing information on certain state monitoring activities, contract requirements, and program characteristics.

The bullets below provide additional context and information involving select MCPAR data elements/domains that are often cited as key to improving managed care oversight and transparency:

• Grievances and appeals. Plan-level data on appeals and grievances can help reveal quality and access issues. MCPAR reported grievance and appeal data include the total number of appeals and grievances resolved, the service types and reasons for appeals and grievances, the total number of state fair hearings and their outcomes, and the outcomes of external medical reviews (if offered by the state). Previously, states were not required to report appeals and grievances data to CMS.

• Prior authorization. The original MCPAR reporting template did not require states to track the number of prior authorization denials or appeal outcomes. Beginning with MCPAR reports submitted in June 2026, states will be required to report new plan-level prior authorization data, including the total number prior authorization requests, denial and approval rates, the percentage of standard prior authorization requests that were approved after appeal, and average and median decision times. These new data can be used to assess timeliness of prior authorization decisions, the extent of plan denials, and plan-wide issues with access to care. CMS expanded prior authorization reporting requirements following the release of an Office of Inspector General (OIG) report that raised concerns about the high number and rates of denied prior authorization requests by some plans and concerns that states were not adequately overseeing and monitoring MCO decisions on prior authorization requests. MCPAR prior authorization reporting requirements align with public reporting requirements included in the 2024 Interoperability and Prior Authorization rule.

• Sanctions. States may impose monetary or non-monetary sanctions on managed care plans if they do not meet contractual obligations. Historically, publicly available data on states’ use of sanctions has varied across states. A recent MACPAC state scan indicated the most common sanction tools identified in managed care contracts include monetary penalties, corrective action plans, and administrative/corrective actions. The MCPAR template collects plan-level data on state-imposed sanctions. Reported elements include intervention type (e.g., civil monetary penalty, liquidated damages, suspension of new enrollment, correction action plan), performance issues (e.g., quality measure performance, network adequacy, timely payment to providers), and sanction amount (if applicable/a monetary penalty was imposed).
• Program integrity. Plans are required to conduct a variety of program integrity activities. For example, plans must have a system for routine internal monitoring and auditing of compliance risks, promptly report to the state all overpayments identified or recovered, and notify the state if they receive information regarding changes to enrollee and provider eligibility. The MCPAR collects data related to plan program integrity activities, including plan-level data on managed care provider overpayments recovered, the number of program integrity investigations opened and resolved by plans, and how frequently plans report changes in beneficiary circumstances to the state. The reports also collect data on state program integrity responsibilities including contract standards for overpayments, how the state monitors overpayment reporting, and links to audits of plan encounter and financial data.

While the data and information contained in MCPARs have the potential to facilitate assessment and comparison of plan performance within and across states, there are also limitations. For example:

• Relatively new requirements. MCPARs are still in the initial years of implementation. The MCPAR template has evolved since its initial release. CMS is likely to continue to refine MCPAR reporting requirements and instructions to ensure consistency across states and to improve data quality and completeness.

• Data variation & comparability. There is significant variation across states in managed care programs (populations served, benefits), plan contract requirements, and state monitoring and oversight approaches. As a result, direct comparison across states, particularly in certain areas, may be challenging. States may also use inconsistent methods for certain metrics or reporting may be inconsistent across states.

• Importance of contextualized data. Without context, MCPAR data may be difficult to interpret in some areas. For example, listing sanctions without context may make it difficult to determine the significance of the noncompliance or the state’s approach to enforcing contract compliance issues. Low numbers of appeals and grievances could reflect few adverse benefit determinations, that access to care met enrollees’ needs, or that the plan does not adequately categorize or track appeals and grievances.

• Scope. MCPARs do not capture all aspects of managed care program and plan performance. MCPARs are also not the only source of plan or program performance information. Additional data can be found in other reports, including the MLR summary report and external quality review (EQR) annual technical report.

• Report format. TheMCPAR reports are currently posted on Medicaid.gov as PDF files, which allows for a centralized, uniform report repository. While the reports are posted publicly, they are technical and may be difficult for the general public/consumers to understand or gain actionable insights.

• Report period. As states operate with different contract years, the timing of data submission and the scope of reported data can differ (including number of months covered).

• Timeliness. The lag in time between end of contract year and report posting (on Medicaid.gov or state website) may affect the immediacy and relevance of the data.

According to a March 2024 U.S. Government Accountability Office (GAO) report, CMS had developed a draft plan (as of April 2023) for processing and analyzing MCPAR data, including developing an analytic dashboard, publishing summary reports and state-level data, and implementing an oversight plan. While CMS began publishing state MCPAR reports on Medicaid.gov in July 2024, CMS indicated to GAO they planned to prioritize working with states on data quality and technical assistance before implementing next steps related to data use and oversight. In November 2025, CMS told GAO that the agency is working on developing an internal appeals and grievance dashboard, to facilitate using the data for oversight, with planned implementation by June 2026. MACPAC is also currently examining state and federal tools for ensuring accountability of managed care organizations and may issue additional policy recommendations in a future report.

Key Administrative Actions Related to the MCPAR

Federal oversight has primarily focused on reviewing and approving plan contracts and capitation rates, and states have generally had considerable flexibility in implementing federal managed care monitoring and oversight requirements. While states have historically produced a variety of data and reports on their managed care programs (e.g., encounter data, external quality reviews etc.), states were not required to produce or submit standardized, comprehensive managed care program reports. Publicly available managed care performance data has been limited and inconsistently available across states, limiting transparency. The following bullets track administrative action related to introducing and implementing the MCPAR reporting requirements:

• 2016. In 2016, under the Obama Administration, CMS issued sweeping managed care regulations to advance program goals related to access, quality, fiscal and program integrity, beneficiary protections, and monitoring and oversight. The rules created new managed care reporting requirements, requiring states to submit three annual reports: 1) an annual managed care program report, 2) a network adequacy and access report, and 3) a medical loss ratio (MLR) summary report. Implementation of the reporting requirements was phased in. However, the first annual managed care program reports were not due until CMS issued guidance on the content and form of the report.

• 2017-2020. In 2017, under the first Trump administration, CMS indicated it would use “enforcement discretion” to allow states to postpone compliance with certain managed care rules introduced in 2016 (although it’s unclear what enforcement discretion was offered and to which states). In 2020, CMS finalized revisions to Medicaid managed care rules in specific areas (e.g., relaxing network adequacy requirements and beneficiary protections) aimed at reducing state and federal burden, streamlining regulations, improving rate setting transparency, and promoting state flexibility. The new managed care reporting requirements from the 2016 rule were left intact, including the annual managed care program report requirement, though the administration did not issue further guidance to effectuate the reporting requirement.

• 2021-2024. In June 2021, the Biden administration released guidance that triggered the annual managed care program report (referred to now as the managed care program annual report or “MCPAR”) requirement, developed standard reporting templates (for the MCPAR, network adequacy and access report, and MLR Summary report) and a variety of toolkits, and released a series of informational bulletins on managed care monitoring and oversight. The first MCPAR reports were due to CMS in December 2022. To improve transparency, CMS began posting state-submitted MCPAR (and MLR Summary) reports on Medicaid.gov in 2024 (from “performance year” 2023). In 2024, the Biden administration also issued significant new Medicaid managed care regulations aimed at improving access, quality of care, and monitoring and transparency, including expanding MCPAR reporting elements and transparency requirements.

• 2025 and beyond. The second Trump administration has continued to post state-submitted MCPAR reports on Medicaid.gov (reports from 2024 were posted in 2025). The administration has also made minor revisions and additions to the MCPAR template² and released an updated MCPAR “FAQ” in August 2025. CMS has also given states the option to forgo completing the access and network adequacy sections of the MCPAR, noting this option was introduced to reduce state burden and duplication with the separate network adequacy and access report; however, it may also reduce transparency as the network adequacy and access report is not currently posted on Medicaid.gov. In November 2025, CMS told GAO that the agency is working on developing an internal appeals and grievance dashboard, to facilitate using the data for oversight, with planned implementation by June 2026. Beginning with MCPAR reports submitted in June 2026, states are required to report new plan-level prior authorization data including the total number prior authorization requests, denial and approval rates, appeal outcomes, and average and median decision times. The 2024 managed care rule requires states to include the results of enrollee experience surveys in the MCPAR (for contract rating periods that begin on/after July 9, 2027).

• On January 27, 2026, the Trump administration released details of the latest iteration of the Mexico City Policy (MCP), now part of a broader umbrella called “Promoting Human Flourishing in Foreign Assistance (PHFFA),” significantly expanding the policy to encompass more funding, more organizations, and new policy areas.
• First announced in 1984 by the Reagan administration, the Mexico City Policy has been rescinded and reinstated by subsequent administrations along party lines and has been in effect for 23 of the past 42 years.
• Historically, when in effect, the policy had required foreign non-governmental organizations (NGOs) to certify that they would not “perform or actively promote abortion as a method of family planning” using funds from any source (including non-U.S. funds) as a condition of receiving U.S. government global family planning funding. This was expanded by President Trump in 2017 to encompass the vast majority of U.S. bilateral global health assistance (including PEPFAR, maternal and child health, malaria, nutrition, and other U.S. global health programs). The latest expansion now includes most non-military foreign assistance and applies to U.S. NGOs, international organizations, and foreign governments, as well as foreign NGOs. In addition to abortion, it prohibits the promotion of “discriminatory equity ideology” and “gender ideology.”
• KFF estimates have found that almost $40 billion could now be subject to the policy restrictions, billions more than during the first Trump administration ($7.3 billion).
• While the full reach of this latest expansion will be shaped by broader changes to the U.S. foreign aid portfolio, potential legal challenges, and additional applications to other agencies and funding streams, it already extends well beyond the reach of what was in place during the first Trump administration.

The Mexico City Policy is a U.S. government policy that – when in effect – has required foreign NGOs to certify that they will not “perform or actively promote abortion as a method of family planning” using funds from any source (including non-U.S. funds) as a condition of receiving U.S. global family planning assistance and, as expanded by the first Trump administration, most other U.S. global health assistance. The latest expansion, by the second Trump administration, applies to significantly more funding (most non-military foreign assistance), many more organizations (beyond foreign NGOs), and new services and activities (beyond abortion).The policy was first announced by the Reagan administration at the 2nd International Conference on Population, which was held in Mexico City, Mexico, on August 6-14, 1984 (hence its name; see Box 1).¹ It was renamed “Protecting Life in Global Health Assistance” (PLGHA) in 2017, and now “Protecting Life in Foreign Assistance” (PLFA) as part of the broader “Promoting Human Flourishing in Foreign Assistance (PHFFA)” Policy umbrella. Among opponents, it is also known as the “Global Gag Rule,” because, in addition to prohibiting service delivery, it also prohibits the promotion of restricted activities, including through lobbying, referrals, and public information campaigns.

When first instituted in 1984, the Mexico City Policy marked an expansion of existing legislative restrictions that already prohibited U.S. funding for abortion internationally. Prior to the policy, foreign NGOs could use non-U.S. funds to engage in certain voluntary abortion-related activities as long as they maintained segregated accounts for any U.S. money received, but after the Mexico City Policy, they were no longer permitted to do so if they wanted to receive U.S. family planning assistance. During those years, U.S. funding for family planning ranged from approximately $300-$600 million annually.

When the policy was expanded by President Trump in 2017 to include the vast majority of U.S. bilateral global health assistance, including funding for HIV under the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR), maternal and child health, malaria, nutrition, and other programs, it increased the funding that could be subject to restrictions to more than $7 billion.³ At that time, the Trump administration also moved to further tighten restrictions in how it defined “financial support,” reaching other areas of U.S. development assistance beyond global health and other non-U.S. funding streams. See “What is the definition of ‘financial support’?” below.⁴

Although President Biden rescinded the policy, President Trump reinstated the MCP through a presidential memorandum on January 24, 2025, directing the Secretary of State to “to implement a plan to extend the requirements of the reinstated Memorandum to global health assistance furnished by all departments or agencies,” which would reach beyond the 2017 expansion. Some organizations and members of Congress had also called for the policy to be expanded to apply to all foreign assistance, and last year, it was reported that the administration intended to include new areas of restrictions related to DEI and gender.

On January 27, 2026, the policy details were released under three separate, but inter-related final rules (interim final rules were first posted on January 23) as part of the PHFFA umbrella (see Box 2). The rules cite the presidential memorandum on the MCP and eight other presidential actions issued in January and February of 2025 that addressed foreign aid, DEI, and gender issues to “ensure that foreign aid is aligned with administration policy and promotes human flourishing.” As such, the PHFFA applies to most non-military foreign assistance and most recipients of foreign aid (not just foreign NGOs). In addition to abortion, it prohibits promoting “discriminatory equity ideology,” including activities related to diversity, equity, and inclusion (DEI), and promoting “gender ideology,” including providing gender affirming care, seeking legal protections based on gender identity, and other related services and activities. The rules indicate that the State Department will work with other agencies that administer foreign assistance to expand the policy further.

With the latest expansion, KFF estimates have found that as much as $39.8 billion in U.S. foreign aid and almost 2,600 prime recipient organizations may be affected (this number should be considered a floor since the policy requirements are passed down when funding is sub-awarded to another organization).

The Mexico City Policy has been in effect for 23 of the past 42 years, primarily through executive action, and has been instated, rescinded, and reinstated by presidential administrations along party lines (see Table 1). 

The policy was first instituted in 1984 (taking effect in 1985) by President Ronald Reagan and continued to be in effect through President George H.W. Bush’s administration. It was rescinded by President Bill Clinton in 1993 (although it was reinstated legislatively for one year during his second term;⁵ see below).⁶ The policy was reinstated by President George W. Bush in 2001,⁷ rescinded by President Barack Obama in 2009,⁸ and reinstated and expanded by President Trump in 2017.⁹ It was rescinded by President Biden in 2021.¹⁰ President Trump then reinstated the policy from his first term in January 2025, noting an intention to expand it further, with final rules released in January 2026.¹¹

The Mexico City Policy has, for the most part, been instituted or rescinded through executive branch action (typically via presidential memoranda¹²). While Congress has the ability to institute the policy through legislation, this has happened only once in the past when  a modified version of the policy was briefly applied by Congress during President Clinton’s last year in office as part of a broader arrangement to pay the U.S. debt to the United Nations.¹³ (At that time, President Clinton was able to partially waive the policy’s restrictions.¹⁴)Other attempts to institute the policy through legislation have not been enacted into law,¹⁵ nor have legislative attempts to overturn the policy.¹⁶ See Table 1.

Under the PHFFA, the types of organizations now subject to the policy, albeit with different applications, include:

• Foreign NGOs,¹⁷ which include:
  • international NGOs that are based outside the U.S.,
  • regional NGOs that are based outside the U.S., and
  • local NGOs in assisted countries.
• U.S. NGOs,¹⁸
• Foreign governments¹⁹ and parastatals²⁰ (note: the State Department will assess whether or not to apply the policy based on foreign policy considerations), and
• International organizations (note: applies to voluntary contributions; international organizations who are recipients of only U.S. assessed contributions are not subject to the policy).²¹

A waiver of the policy or its elements in specific cases is allowed if, in the Secretary of State’s judgment, such a waiver is necessary for national security or foreign policy purposes; see “Are waivers or exemptions allowed?”.

When in place, the policy applies to these organizations as a condition of receiving most non-military U.S. foreign assistance, either directly (as the main – or prime – recipient of U.S. funding) or, for foreign NGOs, U.S. NGOs, and international organizations, indirectly (as a recipient of U.S. funding through an agreement with the prime recipient; referred to as a sub-recipient); see “To what assistance does the PHFFA apply?” below.

Importantly, some aspects of the policy are applied differently based on organization type:

• The strongest restrictions apply to foreign NGOs and international organizations, which are required to agree that outside the U.S. they will not provide or promote abortion as a method of family planning, promote gender ideology, promote discriminatory equity ideology, or engage in unlawful DEI-related discrimination with any funding, from any source; they also must agree not to provide financial support to any other foreign NGO or international organization that engages in such activities. See “What is the definition of ‘financial support’?” below.
• A U.S. NGO is required to agree that it will not provide abortion as a method of family planning outside the U.S. (including with non-Federal funds) and that it will not, “within the scope of any program, project, or activity funded by foreign assistance,” provide or promote abortion as a method of family planning, promote gender ideology, promote discriminatory equity ideology, or engage in unlawful DEI-related discrimination. It is also required to “ensure the physical and financial separation of their foreign assistance-funded programs, projects, and activities from the provision or promotion of” prohibited activities; if such separation is made, a U.S. NGO may carry out prohibited “promotion” activities (such as conducting public information campaigns, or using or teaching education materials) with non-Federal funds but still may not “provide” prohibited activities (such as providing abortion, providing “sex-rejecting procedures,”²² or engaging in unlawful DEI-related discrimination) with any funds. In addition, U.S. NGOs “are not subject to an additional requirement not to provide financial support using non-Federal funds to other organizations that provide or promote abortions outside the United States.”
• A foreign government or parastatal is subject to the policy if the State Department decides to apply the policy – in whole or in part – to an award based on foreign policy considerations; specifically, foreign governments and parastatals may have to agree to not use U.S. foreign assistance award funds to provide or promote abortion as a method of family planning, promote gender ideology, promote discriminatory equity ideology, or engage in unlawful DEI-related discrimination. If it is applied, foreign governments and parastatals “will be required to place any foreign assistance funds under the award in a segregated account to ensure that such funds may not be used to support such activity to the extent the foreign government conducts or supports such activity.” Thus, a foreign government may carry out prohibited activities with non-Federal funds, and they are not subject to the financial support requirement.

The PHFFA policy applies to most non-military foreign assistance, beyond U.S. global health assistance or U.S. family planning assistance alone (see Table 2).²³ Specifically, it applies to foreign assistance administered by the State Department under title III of the annual Department of State, Foreign Operations, and Related Programs (SFOPS) Appropriations Act as well as four other headings within the SFOPS appropriations act: International Narcotics Control and Law Enforcement; Nonproliferation, Anti-Terrorism, Demining and Related Programs; Peacekeeping Operations; and International Organizations and Programs.²⁴ This extends its reach to other sectors, beyond global health, to include humanitarian assistance; economic development, democracy, human rights, and governance; peace and security; education and social services; and environment.²⁵ Within these accounts, the policy currently applies to two main types of funding instruments used for foreign assistance – grants and cooperative agreements – and also to grants provided under contracts. The administration has stated its intention to apply the policy to contracts, subject to future rule-making.²⁶ This was also sought during President Trump’s first term when a proposed rule to this effect was issued, but it was never finalized.²⁷ The rules also indicate that the State Department will work with other agencies that administer foreign assistance to apply the restrictions to their funding streams.

Foreign NGOs, international organizations, U.S. NGOs, foreign governments, and parastatals must pass down (or “flow down”) the policy requirements under the award terms, as applicable, to sub-recipients of foreign assistance, which would then require the sub-recipients to agree to the policy’s restrictions and then flow down the requirements to other organizations, as applicable, if they sub-award the funding further.

The financial support²⁸ requirement governs any funding, no matter the source of funds, provided by foreign NGOs and international organizations to a foreign NGO or international organization for any purpose, if that foreign NGO or international organization happened to provide or promote prohibited activities (there are some differences based on type of organization; see “To what entities does the PHFFA Policy apply?” above). This approach codifies how the financial support requirement was expanded during the first Trump administration.²⁹

Accordingly, this approach to “financial support” may affect funding provided by other donors (such as other governments and foundations) and non-foreign assistance funding provided by the U.S. government for a wide range of purposes. For example, if this funding is first provided to a foreign NGO or international organization that has accepted the policy (as a recipient of U.S. global health or other non-military foreign assistance), the organization then could not, using another donor’s funding for any purpose, provide support to a foreign NGO that provides or promotes prohibited activities.

Under the expanded Protecting Life in Foreign Assistance (PLFA) policy under PHFFA, foreign NGOs and international organizations “agree that, during the period of the award, [they] will not, outside the United States, provide or promote abortion as a method of family planning, or provide financial support to any other foreign NGO or [international organization] that engages in such activities” (there are some differences based on type of organization; see “To what entities does the PHFFA Policy apply?” above. Restricted activities include (also see Box 3):

• providing abortion as a method of family planning,³⁰
• providing advice and information about and offering referral for abortion – where legal – as part of the full range of family planning options,
• promoting changes in a country’s laws or policies related to abortion as a method of family planning (i.e., engaging in lobbying), and
• conducting public information campaigns about abortion as a method of family planning.³¹

The prohibition of these activities is why the policy has been referred to by its critics as the “Global Gag Rule.”

The policy also provides information and clarification about other abortion-related matters:

• Allows for exceptions in cases of rape, incest, risk to life of mother: The policy does not prohibit, where consistent with local law, organizations from providing advice and information about, performing, or offering referral for abortion in cases where the pregnancy has either posed a risk to the life of the mother or resulted from incest or rape.³² The policy also states that “treatment of an ectopic pregnancy or spontaneous loss of pregnancy (a miscarriage) is not an abortion.”
• Allows for post-abortion care: The policy states that “treatment of injuries or illnesses caused by legal or illegal abortions,” which is known as post-abortion care, is not restricted.
• Prohibits passive referral: The policy now prohibits responding to a question about where a safe, legal abortion may be obtained when a woman who is already pregnant clearly states that she has already decided to have a legal abortion (passively providing information, versus actively providing medically-appropriate information). The policy does permit an exemption to be sought from the State Department where this, or another term, is in conflict with local law. This is an expansion compared to the prior policy where passive referral was permitted and the policy requirements did not apply when healthcare providers had an affirmative duty required under local law to provide counseling about and referrals for abortion as a method of family planning.

Under the new Combating Gender Ideology in Foreign Assistance policy of the PHFFA, foreign NGOs and international organizations “agree that, during the period of the award, [they] will not, outside the United States, promote gender ideology, or provide financial support to any other foreign NGO or [international organization] that promotes gender ideology” (there are some differences based on type of organization; see “To what entities does the PHFFA Policy apply?” above). Gender ideology is defined in the policy as “an ideology that replaces the biological category of sex with an ever-shifting concept of self-assessed gender identity.”³⁶ Restricted activities include:

• providing or promoting gender-affirming care;
• lobbying for legal protections based on gender identity, legalization or availability of gender-affirming care, and lobbying for the continued legality of such activities or to change policies;
• conducting a public information campaign in foreign countries regarding acceptance of “gender ideology” or the benefits and/or availability of gender-affirming care;
• using or teaching sex education materials that include gender ideology; and
• conducting drag queen workshops, performances.

The policy also defines sex as “person’s immutable biological classification as either male or female,” which will have likely implications for vulnerable populations served by U.S. foreign assistance programs, including transgender persons.

The policy allows procedures done to “to treat a person with a medically verifiable disorder of sexual development,” “for purposes other than attempting to align an individual’s physical appearance or body with an asserted identity that differs from the individual’s sex,” or “to treat complications of, including any infection, injury, disease, or disorder that has been caused by or exacerbated by, the performance of” gender-affirming care.

Under the new Combating Discriminatory Equity Ideology in Foreign Assistance policy of the PHFFA, foreign NGOs and international organizations “agree that, during the period of the award, [they] will not, outside the United States, promote discriminatory equity ideology, engage in unlawful [diversity, equity, and inclusion-related, or] DEI-related discrimination, or provide financial support to any other foreign NGO or [international organization] that conducts such activities” (there are some differences based on type of organization; see “To what entities does the PHFFA Policy apply?” above). Discriminatory equity ideology is defined in the policy as “an ideology that treats individuals as members of preferred or disfavored groups, rather than as individuals, and minimizes agency, merit, and capability in favor of generalizations,”³⁷ such as that “an individual, by virtue of the individual’s race, color, religion, sex, or national origin, should be discriminated against or receive adverse treatment to achieve diversity, equity, or inclusion.” Restricted activities include:

• promoting discriminatory equity ideology, which includes using or teaching education materials (including books, curricula, and media) that advance such ideology; and
• unlawful DEI-related discrimination, which the policy defines as “discrimination on the basis of race, color, religion, or national origin if such discrimination violates U.S. federal antidiscrimination law or would violate U.S. federal antidiscrimination law if it occurred inside the United States, including the use of those characteristics as a selection criterion or preference for, or basis for exclusion from, employment, contracting, program participation, resource allocation, or similar activities, opportunities, or benefits.”³⁸

The policy does not “address discrimination on the basis of disability or other protected classes, nor does it address … grounds other than race, color, religion, or national origin.”

The rules allow for waivers or exemptions in certain cases as follows:

• Waiver: A waiver of the policy or its elements in specific cases is allowed if, in the Secretary of State’s judgment, such a waiver is necessary for national security or foreign policy purposes. A waiver can be issued either by the Secretary of State or the Under Secretary of State for Foreign Assistance, Humanitarian Affairs, and Religious Freedom, and according to the rules, the State Department will issue guidance on the waiver process.
• Foreign governments/parastatals: The State Department is allowed to exercise discretion, based on foreign policy considerations, in determining whether, and how (in full or in part), to apply the policy terms to the awards of foreign governments and parastatals.
• Local law: The policy permits an exemption to be sought from the State Department where a specific PHFFA award term is in conflict with local law (see, for example, “Prohibits passive referral” above).

A KFF literature review of studies examining the policy, from 2001 to 2024, identified a range of impacts associated with the policy, when in effect, including: increases in abortion rates and reductions in contraceptive prevalence (among other health outcomes); disruption and gaps in services; reduction in service integration; over-implementation and chilling effects; confusion about the policy; loss of civil society/NGO coordination and partnerships; and increased administrative burden (see Box 4). In addition, several studies sought to calculate or estimate the reach of the policy, as measured by amount of funding, countries, and/or NGOs affected.

During the first Trump administration, two official assessments by the State Department were released:

• Initial six-month review of implementation of the policy: released in February 2018, it directed agencies to provide greater support for improving understanding of implementation among affected organizations and provided guidance to clarify terms included in standard provisions of grants and cooperative agreements.
• Second review of the policy, covering the first 18 months of implementation: released in August 2020, it found the awards declined spanned a variety of program areas,³⁹ in addition to cross-cutting awards; spanned geographic areas, but many were for activities in sub-Saharan Africa; and led to gaps or disruptions in service delivery sometimes being reported, despite U.S. efforts to transition projects to another implementer in order to minimize disruption.

Additionally, a report by the U.S. Government Accountability Office (GAO), released in March 2020, found that from May 2017 through FY 2018, the policy had been applied to over 1,300 global health projects, with the vast majority of these through USAID and CDC, and NGOs declined to accept the expanded policy during the first Trump administration in 54 instances, totaling $153 million in declined funding⁴⁰ across USAID and CDC. The State Department and DoD did not identify any instances where NGOs declined to accept the policy conditions.

Medicaid is a very complex program that involves millions of enrollees, hundreds of thousands of providers, and significant federal and state expenditures. Both the federal government and states are responsible for ensuring the proper management and functioning of the Medicaid program to ensure it is providing quality and efficient care while using funds–taxpayer dollars–appropriately with minimal waste. Program integrity efforts historically have worked to prevent and detect fraud, waste, and abuse, to increase program transparency and accountability, and to work on corrective action plans and recover improperly used funds. The Trump Administration has signaled reducing fraud, waste, and abuse across federal programs (including Medicaid) is a priority.  However, broader Medicaid cuts from 2025 reconciliation law may impede states’ ability to invest in oversight and other administrative initiatives to reduce fraud. The reconciliation law also includes changes that directly address the Medicaid Payment Error Rate Measurement (“PERM”) program, which over time could have financial and administrative implications for states. The Congressional Budget Office (CBO) estimates that Medicaid payment reductions related to PERM could reduce federal spending in Medicaid by $7.6 billion over ten years.

The PERM program measures “improper payments” in Medicaid to produce improper payment rates. Improper payments are payments that do not meet CMS program requirements, often the result of administrative or paperwork issues. PERM is designed to help states improve their operation of the program and reduce errors; it is not a measure of fraud against the program. Beginning October 1, 2029, the reconciliation law requires HHS to reduce federal Medicaid financial participation to states that exceed a three percent PERM eligibility error rate threshold. (SNAP policy changes will also require states to pay a portion of SNAP benefit costs, depending on the state’s payment error rate.) While Medicaid pays most outlays properly (93.9% in 2025) most improper payments are due to insufficient documentation (77.2% in 2025), including eligibility-related errors (68.1% due to insufficient documentation in 2025). State error rates vary, but nearly one quarter of states have eligibility error rates above the three percent PERM threshold based on their most recent audit result. This brief explains the PERM program and its three components as well as upcoming changes to PERM and possible state impacts.

The PERM program operates according to requirements set in law and CMS rulemaking. The Improper Payments Information Act (IPIA) of 2002 (replaced by the Payment Integrity Information Act (PIIA) of 2019) requires the heads of federal agencies to annually review programs they administer and identify those that may be susceptible to significant improper payments, to estimate the amount of improper payments, to submit those estimates to Congress, and to submit a report on actions the agency is taking to reduce the improper payments. The federal government has designated Medicaid and other programs (including Medicare, SNAP, and TANF) as “risk-susceptible.” Medicaid receives this designation due to its size and complexity, as the program involves millions of enrollees, hundreds of thousands of providers, 51 state agencies (including DC), different delivery systems, complicated eligibility rules, and significant federal and state expenditures. GAO monitors high-risk programs and their improper payments, including Medicaid.To comply with IPIA requirements in Medicaid, CMS developed the PERM program.

The PERM program measures improper payments in Medicaid by producing a national improper payment rate through cyclical audits of state Medicaid programs. Improper payments are payments that do not meet Medicaid program requirements. The improper payment rate is the estimated share of Medicaid claims that do not meet Medicaid program requirements. The error rate is not a “fraud” rate (or a waste or abuse rate), but a measurement of payments made that did not meet statutory, regulatory, or administrative requirements or are made in an incorrect amount including overpayments and underpayments. Each year (since 2009) HHS has reported an estimated national (or overall) Medicaid improper payment rate, which is based on reviews of three program components: fee-for-service (FFS) claims, managed care capitation payments, and eligibility determinations. In 2025, Medicaid paid an estimated 93.9% of outlays properly, representing $610.95 billion in proper federal payments (Figure 1). The overall estimated improper payment rate was 6.1%, or $37.39 billion in federal payments. (See Box 1 for discussion of improper payment rate trend over time.)

PERM audits a sample of state Medicaid FFS claims and managed care payments every three years. Federal contractors conduct standardized audits of states on a rolling three-year basis, meaning each PERM cycle measurement includes one-third of states (see Appendix). Annually, the most recent three cycles are combined to produce a national improper payment rate (weighted by state size). Audits include a random sample of FFS claims and managed care payments from a 12-month period. To create the 2025 improper payment rate, PERM program contractors reviewed about 48,000 claims and managed care payments.¹ States receive results of their audits and use the information to improve compliance during the next audit cycle. States are often required to develop and implement corrective action plans to reduce mistakes and improve recordkeeping.

Most improper payments are due to insufficient documentation or missing administrative steps. HHS reports the different types of errors that contribute to the improper payment rate each year. In 2025, 77.2% of improper Medicaid payments were the result of insufficient documentation or missing administrative steps (Figure 2). These payments were not necessarily for ineligible enrollees, providers, or services since they may have been payable if the missing information had been on the claim and/or the state had complied with requirements. Examples include state failure to retain documentation of enrollee eligibility or to appropriately screen enrolled providers, or medical records not submitted or missing required documentation to support the medical necessity of a claim. Errors that represent “monetary loss” made up the remaining one quarter (22.8%) of improper payments in 2025 (Figure 2). Monetary loss represents a subset of improper payments where CMS has sufficient information to conclude the payment should not have been made or was made in the incorrect amount. These errors include payments made for enrollees who are ineligible for the program or for services and providers not enrolled. 

The overall improper payment rate is calculated based on three component error rates: fee-for-service, managed care, and eligibility. The PERM program uses a random sample of Medicaid claims and managed care payments to calculate component error rates. Those component error rates are combined to produce an overall improper payment rate.

Fee-for-Service. Within a FFS delivery system, the state Medicaid agency pays providers or groups of providers directly. The FFS PERM component estimates the payment error rate for fee-for-service claims, where selected claims undergo a medical review (focused on documentation in the medical record) and a data processing review (checking if claims followed correct procedures –e.g., coding, billing and payment amounts).

Managed Care. States that contract with Medicaid managed care plans pay a set per member per month (“capitation”) payment to the plan for the services covered by the contract. The managed care PERM component estimates Medicaid managed care capitation payments made by the state in error (plan payments to providers are not reviewed under PERM). The PERM program’s first report (2009) estimated a managed care component error rate of 1.5%, and since 2011, the estimated managed care error rate has remained below one-half of one percent (Figure 3).

Eligibility. The eligibility component assesses the state’s application of federal rules and documented state eligibility policies and procedures. States must provide records or documentation to support eligibility determinations. Non-compliance with eligibility requirements may include not performing a required element (e.g., income verification); not providing or retaining sufficient documentation for a required element; failing to conduct timely eligibility redeterminations; determining eligibility under the incorrect eligibility category; or enrolling an individual who is not eligible. (Note that PERM does not attempt to identify or measure instances where a state incorrectly determines an individual is ineligible for Medicaid – i.e., where eligibility rules were applied inaccurately.) 

Insufficient documentation errors are the cause of most eligibility errors. While the national eligibility error rate decreased from 2021-2024, due in part to audits that accounted for the continuous enrollment provision that paused Medicaid eligibility renewals and disenrollments from March 2020 through March 2023 (e.g., excluding errors due to out-of-date redeterminations or changes in the enrollee’s circumstances), the eligibility error rate rose from 2024 to 2025 as states resumed normal eligibility operations (Figure 3 and Box 1). Insufficient documentation errors continue to be the cause of most eligibility errors (Figure 4). Eligibility determination errors caused by reasons other than insufficient documentation include enrollees receiving services for which they are financially or categorically ineligible (for example, an enrollee who is eligible for a certain eligibility group and receives services through a program for another eligibility group).

Recent federal rules increase recordkeeping requirements aimed at reducing insufficient documentation errors. Beginning in 2026, states must adhere to updated eligibility and payment recordkeeping standards required in the 2024 Eligibility and Enrollment Final Rule. This rule aimed to reduce “paperwork” errors that lead to the majority of eligibility-related improper payments by implementing new documentation and retention requirements and procedures. The Eligibility and Enrollment rule specifies that states must maintain records for all eligibility determinations in an electronic format, clarifies what information must be contained in the records, and requires that records be kept for a minimum of three years. States must comply with these requirements by June 2026.

PERM program changes in the reconciliation law focus on the eligibility component of the error rate. Beginning in 2029, states may be subject to a financial penalty if PERM eligibility error rates exceed three percent. Even before passage of the reconciliation law, HHS had the authority to recoup federal funds from states with eligibility-related improper payment error rates greater than three percent. This statutory recoupment authority was first implemented by CMS through the Medicaid Eligibility Quality Control (MEQC) program (which complements the PERM program). In 2020, GAO reported no recoupments due to MEQC eligibility errors had occurred since 1992. In 2017, CMS introduced new procedures to begin recouping funds based on PERM eligibility errors beginning in FY 2022, but there are no public reports of any recoupments. Prior law also gave HHS authority to grant states “good faith” waivers if they demonstrated a “good faith effort” to improve eligibility error rates. However, the reconciliation law effectively eliminates “good faith” waivers, requiring CMS to impose the penalty for states that exceed the PERM eligibility error threshold. Beginning October 1, 2029, states with eligibility error rates greater than three percent will have to repay the federal portion of the improper payment amount above the three percent threshold. The reconciliation law also expands the types of eligibility errors that are used to calculate the financial penalty. Insufficient documentation errors will be included in the federal financial recoupment calculation. Implementing federal financial recoupment will likely depend on further federal guidance to states.

Based on the most recent PERM audits, nearly one quarter of states have eligibility error rates above the new threshold for penalty.³ Twelve states had eligibility error rates above three percent in their most recent PERM audit (Figure 5). The most recent three PERM audit cycles used Medicaid claims and payments from July 2021 – June 2024. During most of this period, eligibility redeterminations and disenrollments were paused under flexibilities related to the COVID-19 PHE, which likely decreased the rate of eligibility errors. As a result, the audit results may not indicate future state performance. The amount of federal funds recouped will vary across states due to differences in eligibility error rates and because of federal match rate (FMAP) differences across states. States with a higher traditional “FMAP” rate risk losing a greater share of any payments subject to recoupment because a larger share of their Medicaid payments come from federal funds.

States will need to prepare for PERM program changes and implement other complex eligibility and enrollment policy changes required by the reconciliation law at the same time. State readiness to comply with increased recordkeeping and electronic documentation requirements in the Eligibility and Enrollment rule will also vary. At a time of increasing state budget pressures, states may need to invest in upgraded technology, expanded audits or staffing, or other costly increases to program integrity capacity to maintain a low rate of improper payments. At the same time, states are also implementing other complex eligibility policy changes required by the 2025 reconciliation law, including several changes by January 1, 2027:

• The reconciliation law requires states to implement work requirements (at application and renewal) for Medicaid expansion adults. States will need to make significant changes to eligibility systems in a short timeframe to verify compliance as well as to identify individuals who qualify for exemptions or optional exceptions. Failure to document compliance with complex work-related federal and state eligibility determination policy could lead to increased PERM eligibility errors.
• The reconciliation law increases the frequency of required eligibility redeterminations from every 12 months to every 6 months for expansion adults. An increased number of eligibility determinations could result in additional eligibility errors.
• The reconciliation law also decreases the time period that states must provide retroactive Medicaid coverage (from three months to one month of retroactive coverage for expansion adults and two months for traditional enrollees). Narrowed claims payable under new retroactive coverage rules could result in additional payment errors.

Beginning in 2029, states risk losing federal funds if their PERM eligibility error rate is above three percent – a threshold nearly one quarter of states met prior to implementing complex eligibility changes included in the reconciliation law (12 states in 2025, Figure 5). However, because these audits occurred during the PHE and unwinding periods, the results may not predict future improper payment rates. CBO estimated $7.6 billion over 10 years in Medicaid payment reductions related to the reconciliation law’s changes to PERM. Updated electronic documentation and recordkeeping requirements may help reduce some errors due to insufficient documentation or administrative missteps. However, to limit the risk of financial penalty due to eligibility errors, states may add administrative steps or impose more onerous paperwork requirements on enrollees or applicants.

The overall improper payment estimate for each state combines the state’s fee-for-service, managed care, and eligibility estimates, with a correction factor so that sampled payments that are improper in more than one of those components are not double counted.