Here We Go Again with Adverse Selections

It would be an understatement to say that a lot happened in the 2025 rate filing season (for 2026 policy year) for the individual Affordable Care Act (ACA) plans. In short, policy year 2026 was expected to have higher premiums for the ACA enrollees along with potential enrollment shifts in the markets. The premium subsidy was set to be cut tangibly when the enhanced premium subsidy provision under the American Rescue Plan Act of 2021 (later extended under the Inflation Reduction Act of 2022) legislation expired at the end of calendar year (CY) 2025, affecting millions of enrollees in the plans. This anticipated legislative expiration resulted in the issuers projecting erosion in enrollments accompanied with expected increases in morbidity for the market as a whole, reflecting an anticipation of adverse selection to the issuers’ experience.[1] Other regulatory changes (called Marketplace Integrity and Affordability, or MIA, rules) were proposed and later adopted by the Centers for Medicare and Medicaid Services (CMS) that could result in further enrollment decreases under strengthened eligibility rules and income verification processes. Some of the key provisions of the MIA rules were later blocked by a Maryland District Court order.[2] However, by then pricing for 2026 was already finalized and approved under assumptions of further decreased enrollments and higher morbidity under the restrictive rules on eligibility.

Issuers’ Reactions to Looming Adverse Selections

Issuers were handed the classic adverse selection situation where rising premium rates would trigger healthier enrollees to drop their coverage, leaving the single risk pool to become more costly. The aim of this article is to observe, as an episode, how seven issuers in a marketplace have reacted to anticipated changes in the composition of the risk pool.

In a nutshell, adverse selection impact is measured by comparing morbidity or average claims cost level of a population before and after an adverse selection event takes a place. The modeling generally involves segmenting the population into various risk cohorts and projecting lapses for each risk cohort under rate increase scenarios. Normally higher lapse rates are assumed for healthier cohorts of enrollees, leaving the remaining pool more costly per person. Varying degrees of adverse selection impact are therefore expected depending on assumed distribution of the risk cohorts and lapse rates assigned to the cohort distributions. For summary and comparison of the adverse selection estimates of the seven issuers observed, the issuers were aggregated into two groups (Group A and Group B) depending on the market size of each issuer. Table 1 summarizes the projected enrollment decreases and the corresponding morbidity increases due to adverse selection by the two groups.

Table 1: Projected Enrollment Decreases and Morbidity Increases Due to Adverse Selection

Notes:
1. CY2024 member months.
2. Based on straight average.
3. One of the three issuers in Group A withdrew from the market after pricing was finalized and rate filing was approved.

Source: North Carolina Department of Insurance, Rate filings for individual ACA plans for policy year 2026, https://www.ncdoi.gov/documents/life-and-health/healthcare-law/initial-aca-rates-plan-year-2026.

The table shows the average lapse rates and average morbidity increases by the two issuer groups. The average lapse rates showed a small difference between the two groups, 23% for Group A versus 28% for Group B. The Group B average morbidity increase is estimated to be 14.8%, materially higher than the Group A estimate of 8.9%. This shows that the issuers with more membership (Group B) in this state used more conservative adverse selection assumption in their pricing. Group B used, on average, 0.6% morbidity deterioration per 1% decrease in enrollment, whereas Group A used 0.4% deterioration per 1% decrease in enrollment. It makes sense given the risk of greater losses faced by the issuers with larger memberships, particularly if their previous membership significantly benefitted from enhanced subsidies. Whether that is the case in general or not, what is clear is that all issuers are anticipating a substantial morbidity deterioration.

Pricing Impact

The higher morbidity assumptions led to higher premium rate increases proposed by the issuers for policy year 2026. In this particular market, the proposed rate increases ranged from 16.9% to 36.4%, averaging 28.1%, much higher than any increases seen since 2018 as a result of the Tax Cuts and Jobs Act of 2017.

The large rate increases include other incremental factors that are incidental to the underlying rate increases. The non-benefit expenses portion of the premium went up for all seven issuers, reflecting higher administrative expenses per person due to reduced membership and adding margins for risk of increased uncertainty. The target loss ratios went down by 2.5% on average. Some issuers projected population demographics to become older as younger and likely healthier members are more likely to lapse. This increases their base rates to compensate for premium shortfall resulting from using the compressed age factors that limit the adult age factor ratio to 3 to 1, instead of allowing the full range of actual claim cost differences by age. These incremental rate increases reflect an expectation of more adverse selection in the new member cohort.

Challenges in the 2026 Policy Year Pricing Cycle

Issuers faced extra difficulties in this pricing cycle due to regulatory uncertainties. They had to develop the 2026 rates without knowing whether the enhanced premium subsidy would be extended or not. Not extending it meant a significant increase in morbidity assumption. The final rates were developed assuming that the enhanced premium subsidy would expire, as that scenario was more likely at the time (indeed, it expired at the end of the year despite partisan effort to extend it).

CMS finalized and released Marketplace Integrity and Affordability rules on June 20, 2025. The key aim of these rules was to address improper enrollments through stronger income verification and eligibility requirements, but they had the effect of restricting all enrollments. The final rules’ negative impact on enrollments and morbidity were included in the morbidity deterioration assumptions and reflected in pricing. However, many of the key provisions of the final rules were blocked by a court order in August 2025.[3] Issuers adjusted pricing to reflect the narrower actuarial value (AV) de minimis range (reverting to the prior AV range as a result of the court order) but did not roll back the added morbidity increases built into the pricing.

This article has focused primarily on the causes and impact of adverse selections on the 2026 premium rates. Other factors were also in play in setting 2026 premium rates. A July 2025 Issue Brief from the American Academy of Actuaries provides a good summary of these factors.[4]

Projection vs Early Enrollment Data

ACA plans are subject to a risk adjustment program administered by CMS. The risk adjustment program spreads financial risk undertaken evenly by transferring funds from plans with lower-risk enrollees to plans with higher-risk enrollees.[5] Leveling the playing field helps to mitigate adverse selection between issuers, promote healthy competition and stabilize premiums. This zero-sum game result presumes that issuers as a whole are pricing at the market-wide morbidity. Issuers are exposed to underpricing risk when there is an anticipated market-wide morbidity deterioration. Conservative pricing may be one approach to counter the risk of underestimating morbidity increases, as can be seen from the wide range of morbidity increases (5%–23%) shown in Table 1. Reinsurance could be another method to reduce such risk exposure.

Substantial premium subsidies for low-income individuals are still available, even after expiration of the enhanced premium subsidy, and the cost-sharing subsidies continue. That might be enough incentives to provide a floor to the enrollment shrinkage and prevent the adverse selection spiral. The individual ACA market went through a similar situation in the past when the individual mandate penalty was repealed (starting in 2019) and the CSR subsidy was defunded under the Tax Cuts and Jobs Act of 2017 (the per member per month premium increases from 2017 to 2018 were 27% on average nationwide[6]). The market survived these large increases and increased its membership substantially when the premium subsidy was enhanced under the American Rescue Plan Act of 2021.

The Congressional Budget Office (CBO) projected the individual ACA enrollments would decrease by about 17% from 2025 to 2026 if the enhanced premium subsidy expired.[7] The seven issuers in Table 1 projected an approximate 26% decrease in the enrollments on average. This decrease also reflects the stricter eligibility requirements imposed under the MIA rules issuers assumed in their pricing. The gap between the two projections (17% versus 26%) would be narrowed when adjusted for the MIA impact. Reporting on 2026 actual enrollments to date indicates a smaller drop in enrollments than projected—about 5%, with 23 million in 2026 versus 24.2 million in 2025.[8] The year-over-year enrollment changes (as of January 3, 2026) varied by state, from +19.5% in New Mexico to –21.1% in North Carolina. The unexpected enrollment gain in New Mexico is due to state funding that replaced the expired federal subsidies.[9] The three states with the largest enrollment decreases (West Virginia with –15.0%, Ohio with –18.6% and North Carolina with –21.1%[10]) have high concentrations of members (relative to other states) in the 138%–150% Federal Poverty Level (FPL).[11] Members in this FPL segment are particularly hard-hit by the subsidy expiration (with premiums increasing from $0 to 3%–4% of their income), likely causing many of them to drop coverage. The termination number is expected to grow when the effectuated enrollees (those who actually paid their premiums) are counted. Recent CBO projections indicate more attrition of the ACA enrollments in the near future as provisions of the One Big Beautiful Bill Act are implemented.[12] It seems there is still a bumpy road ahead for the ACA marketplace.

This article is provided for informational and educational purposes only. Neither the Society of Actuaries nor the respective authors’ employers make any endorsement, representation or guarantee with regard to any content, and disclaim any liability in connection with the use or misuse of any information provided herein. This article should not be construed as professional or financial advice. Statements of fact and opinions expressed herein are those of the individual authors and are not necessarily those of the Society of Actuaries or the respective authors’ employers.

William Ju, ASA, MAAA, is a Life & Health actuary at North Carolina Department of Insurance. William can be reached at William.ju@ncdoi.gov.

Endnotes

1. Matthew Buettgens, et al, “4.8 Million People Will Lose Coverage in 2026 If Enhanced Premium Tax Credits Expire,” Urban Institute, September 17, 2025, https://www.urban.org/research/publication/48-million-people-will-lose-coverage-2026-if-enhanced-premium-tax-credits. 
2. EBIA Checkpoint News Staff, “Court Delays Key Provisions of Marketplace Integrity Regulations; Cost-Sharing Increases Remain on Schedule,” Thomson Reuters, August 27, 2025, https://tax.thomsonreuters.com/news/court-delays-key-provisions-of-marketplace-integrity-regulations-cost-sharing-increases-remain-on-schedule/.  
3. EBIA Checkpoint News Staff, “Court Delays . . .”  
4. American Academy of Actuaries, “Drivers of 2026 Premium Changes,” Issue Brief, July 2025, https://actuary.org/wp-content/uploads/2025/07/brief-Drivers-2026-Premium.pdf.  
5. Centers for Medicare & Medicaid Services, “Summary Report on Individual and Small Group Market Risk Adjustment Transfers for the 2024 Benefit Year,” CMS, June 30, 2025, https://www.cms.gov/files/document/ra-report-by2024.pdf.  
6. Centers for Medicare and Medicaid Services, “Appendix A to 2017 Benefit Year Risk Adjustment Summary Report – HHS Risk Adjustment Program State-Specific Data (XLSX),” July 9, 2018, and “Appendix A to 2018 Benefit Year Risk Adjustment Summary Report – HHS Risk Adjustment Program State-Specific Data (XLSX),” June 28, 2019, https://www.cms.gov/marketplace/health-plans-issuers/premium-stabilization-programs.  
7. Jared Ortaliza, et al, “Inflation Reduction Act Health Insurance Subsidies: What Is Their Impact and What Would Happen If They Expire?” KFF, July 26, 2024, https://www.kff.org/affordable-care-act/inflation-reduction-act-health-insurance-subsidies-what-is-their-impact-and-what-would-happen-if-they-expire/
8. Tami Luhby, “Obamacare Enrollment Drops After Enhanced Premium Subsidies Expire,” CNN, January 28, 2026, https://www.cnn.com/2026/01/28/politics/aca-subsidies-insurance-obamacare-enrollment.  
9. Jakob Emerson, “New Mexico Sees Record ACA Enrollment After State Replaces Expired Subsidies,” Becker’s Payer Issues, January 2, 2026, https://www.beckerspayer.com/payer/aca/new-mexico-sees-record-aca-enrollment-after-state-replaces-expired-subsidies/.  
10. “Marketplace Enrollment Snapshot for Open Enrollment for 2026,” KFF, January 28, 2026, https://www.kff.org/affordable-care-act/state-indicator/marketplace-enrollment-snapshot-for-open-enrollment-2026/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D.  
11. Centers for Medicare and Medicaid Services, “2025 Marketplace Open Enrollment Period Public Use Files,” CMS, last modified May 12, 2025, https://www.cms.gov/data-research/statistics-trends-reports/marketplace-products/2025-marketplace-open-enrollment-period-public-use-files.  
12. Ortaliza, et al, “How Will the One Big Beautiful Bill Act Affect the ACA, Medicaid, and the Uninsured Rate?”