Approximately 2.9 million Medicare Advantage enrollees are switching plans in 2026 after carrier exits spiked to 10% of the MA population, a tenfold jump from the 1% annual average seen between 2018 and 2024 (JAMA, February 2026). Plans that absorbed those displaced members filed their 2027 bids in June 2026 before the membership shift was fully visible, and CMS's 2.48% effective rate (CMS, April 2026) against 8 to 10% medical cost trend locks in structural negative operating leverage before the V28 coding disruption on displaced members adds a second layer of exposure.
The Bid Cycle Cannot Absorb a Mid-Year Membership Shift
MA plans file benefit bids with CMS each June for the following calendar year. The 2027 bids were submitted in June 2026. The open enrollment period for 2026 ran October through December 2025, with plan assignments effective January 1. That timing means plans absorbing displaced members knew, when they filed their 2027 bids, that they had taken on new membership in 2026. What they did not yet have was a full picture of their H1 2026 utilization experience for those members, the chronic condition coding profile those members would carry under V28 at the new plan, or the geographic concentration of the absorbed cohort relative to their bid-year assumptions.
CMS finalized the 2027 effective rate at 2.48% on April 6, 2026, an increase from the 0.09% proposed in the Advance Notice after industry pushback. Oliver Wyman, in a March 2026 analysis titled "Medicare Advantage plan economics reset in 2027," described the environment as one of negative operating leverage: "Revenue growth will not keep pace with expense growth," projecting medical cost trend in the 8 to 10% range against that rate floor. Plans absorbing high-utilization cohorts from displaced membership face a compounded version of that gap, on top of a base that Oliver Wyman already characterizes as structurally insufficient for most participants.
The 2.48% figure represents what CMS estimates as over $13 billion in additional payments compared to 2026 (CMS, April 2026). That headline number is not evenly distributed. Effective rate growth varies by county benchmark, Star Rating status, and risk score trend. Plans in lower-benchmark rural counties, which are precisely the geographies where exit rates were highest, see below-average effective rate growth even in a year where the national average improved. The displacement wave and the rate geography are not independent: the hardest-hit counties are also where the remaining plans face the least favorable rate math.
Who Was Displaced and Why the HCC Data Surprises
The Johns Hopkins Bloomberg School of Public Health research letter, published February 18, 2026 in JAMA by Mark Meiselbach, Matthew Lavallee, Jianhui Xu, and Dan Polsky, analyzed 192 million enrollee-years of CMS Plan Finder data. Annual forced disenrollment had averaged just over 1% between 2018 and 2024 before climbing to 6.9% in 2025 and reaching 10% for 2026 (JAMA, February 2026). Smaller insurers accounted for 48.8% of forced disenrollments compared to 27.1% of retained enrollees, reflecting the scale disadvantage that makes smaller plans unable to absorb utilization volatility or negotiate competitive provider rates. Rural beneficiaries faced a 28% forced disenrollment rate against 15% for urban enrollees (JAMA, February 2026). PPO plan enrollees were more likely to be displaced than HMO members, consistent with the structural cost-management disadvantage of PPO products where out-of-network access raises the floor on medical loss ratios.
The HCC finding is the counterintuitive one. Meiselbach et al. found no statistically significant difference in fee-for-service HCC risk scores between members who were forced to disenroll and those who retained their plans. "We are probably overpaying for Medicare Advantage, and so efforts to reduce how much we pay are going to mean that some plans can no longer be profitable," Meiselbach told AJMC, framing the exit wave as a payment correction rather than a demographic self-selection event. On paper, the displaced population does not look sicker than the retained population.
That finding reframes the actuarial risk for receiving plans. They did not inherit an obviously higher-acuity cohort in aggregate. The problem is structural and mechanical, not compositional: new members whose coding history the receiving plan cannot access, arriving in a year where the annual bid cannot be retrospectively adjusted to reflect their presence, under a risk model that requires the receiving plan to generate its own encounter data before it can claim the full risk score those members warrant.
The V28 Coding Gap: Year-One Risk Scores Understate True Cost Exposure
The V28 HCC risk adjustment model completed its phase-in for payment year 2026, running at full weight after phasing in at one-third in 2024 and two-thirds in 2025. V28 uses 115 HCC categories and 7,770 diagnostic codes, narrowed from V24's 86 categories and 9,797 codes. CMS designed the tighter code set to reduce payment for conditions that had inflated coding patterns, but the narrower structure means each surviving code carries greater weight in a plan's risk score. Missing a code under V28 is more costly than missing a code under V24 because the denominator of surviving categories is smaller.
Annual Wellness Visits are the primary encounter type at which chronic conditions get coded under Medicare risk adjustment protocols. A member enrolled at Plan A who completes an AWV in February 2025 carries that coding into Plan A's 2026 risk scores. When that member is displaced to Plan B for 2026, Plan B does not inherit Plan A's AWV encounter data. Plan B needs to conduct its own AWV in 2026 to capture the member's chronic condition burden under its own NPI, and submit those encounter records to generate its V28 risk scores for 2027 payment. Until that happens, the member's contribution to Plan B's risk score pool reflects whatever partial coding the plan has from other 2026 encounters, which is systematically lower than the member's full chronic condition profile.
In practice, displaced members often delay or miss AWVs in their first year at a new plan. Establishing a new primary care relationship, navigating a different provider network, and resolving prescription continuity issues all compete for a newly enrolled displaced member's attention. The transition friction that characterizes the first enrollment year suppresses AWV completion rates. The result: a year-one V28 risk score at the receiving plan that understates the member's actual chronic condition burden.
| Stage | Mechanism | Actuarial Effect |
|---|---|---|
| Year 0 (exiting plan, 2025) | AWV and chronic condition codes submitted by exiting plan | Risk scores accrue to the plan that exits, not the receiving plan |
| Year 1 (receiving plan, 2026) | No inherited encounter history; AWV completion delayed by transition friction | V28 risk scores suppressed below member's true acuity; bid filed June 2026 cannot reflect this |
| Year 2 (receiving plan, 2027) | AWV and coding catch-up as member stabilizes; chronic conditions re-established in records | Risk score increases above year-1 baseline; looks like medical trend if not modeled separately |
The actuarial consequence compounds across the bid cycle. The year-two coding catch-up at receiving plans, where displaced members' chronic conditions get fully documented after a delayed AWV cycle, will show up in 2027 experience as apparent medical cost trend. A plan that models trend against its stable historical base will misattribute the 2027 increase to organic trend rather than a catch-up from year-one coding suppression. Plans that file 2028 bids using 2027 trend data without separating the displaced-member cohort will embed that misattribution into their forward pricing.
The 2027 Recalibration Adds a Third Layer
For 2027, CMS updated V28's calibration factors using 2023 diagnosis data and 2024 expenditure data, replacing a prior calibration anchored to 2018 diagnoses and 2019 expenditures. The recalibrated model reflects current cost patterns for each condition category. CMS estimated the combined effect of implementing the 2027 risk model and normalization would reduce payments to MA plans by 3.32% (CMS Advance Notice, 2026), before the offsetting effect of the rate increase is applied.
For plans that absorbed displaced members, the recalibration introduces a third actuarial uncertainty layer on top of the membership timing mismatch and the coding gap. The historical diagnosis codes that exist for displaced members from their time at the exiting plan were generated under that plan's coding practices and scored under the pre-2027 calibration. The receiving plan inherits members whose prior-year risk scores reflect a different plan's documentation culture, now subject to recalibration that updates the coefficient weights for each HCC category. A member whose prior-plan HCC profile was calibrated against 2018-2019 expenditure data may score differently under the 2023-2024 calibration, in either direction depending on the condition mix, introducing a reclassification effect that actuaries must model separately from pure membership mix changes.
CMS confirmed it will continue using the V28 clinical classification system for 2027, deferring more substantial model revisions that had been proposed in the Advance Notice. That decision maintained more continuity than the Advance Notice had signaled, but the recalibration itself is material: 3.32% reduction in plan payments from normalization and model updates runs in the same direction as the rate gap and the coding suppression, not opposite to them.
Rural Reversion Changes Pool Composition Without Changing Individual Risk Scores
The JAMA finding that individual HCC risk scores do not differ significantly between displaced and retained members does not mean that population composition is neutral for receiving plans. The geographic story shows why the aggregate-level finding masks a plan-level exposure.
Rural beneficiaries faced a 28% forced disenrollment rate versus 15% for urban enrollees (JAMA, February 2026). In many rural counties, carrier exits reduced plan availability significantly: Vermont saw 92.2% of MA enrollees facing forced disenrollment, and in six additional states, at least 40% of MA enrollees lost their plans. When MA plans exit rural counties and no remaining plan absorbs displaced rural members, those members revert to traditional Medicare. In rural geographies with limited plan choice, that reversion is not a hypothetical tail event. It is the default outcome for beneficiaries whose county no longer has a viable MA alternative.
This creates a pool composition effect distinct from individual-level adverse selection. If rural members disproportionately exit MA through TradMed reversion rather than re-enrollment in another MA plan, the surviving MA risk pool shifts toward a more urban composition. The remaining plan portfolio has a different geographic risk profile than the bids assumed, even without individual members becoming meaningfully sicker or healthier. Risk adjustment normalizes at the county and plan level. A plan whose membership becomes more concentrated in urban counties because rural members reverted to TradMed faces a geographic mix shift in its risk score distribution that does not surface in individual-level HCC tracking but does affect aggregate risk revenue.
The ACA market offers a historical precedent. When ACA carrier exits shrink the pool of remaining issuers, the membership absorbed by surviving plans is not simply a random draw from the exiting plan's enrollees. The members who actively re-enroll in alternative MA plans are those with the highest engagement with the MA product, often the chronically ill who depend on care coordination and supplemental benefits. Members who revert to TradMed tend to be healthier. That differential sorting, familiar from ACA market dynamics, plays out in MA disenrollment geography as a rural-urban pool composition shift rather than an individual HCC score shift.
What Plans Can Still Do Before 2027 Begins
The 2027 bid is filed. Benefit design is locked. But plans have H1 2026 utilization data on displaced members enrolled in January 2026, and that data can still drive internal actuarial assumption updates and operational responses before the 2027 benefit year begins.
Accelerating AWV completion for newly enrolled displaced members is the highest-leverage intervention. Plans that cannot revise their 2027 bids can still reduce the coding gap. Proactive AWV outreach to members enrolled from displaced-plan cohorts, ideally scheduled in H2 2026 before year-end, allows the plan to capture chronic condition codes under its own NPI and improve the 2027 risk score baseline. A displaced member who completes an AWV at the new plan in October 2026 generates encounter data that flows into 2027 risk scores; a member who does not complete an AWV until March 2027 generates data that may not affect 2027 payment at all, depending on submission timing. The AWV completion window matters.
H1 2026 utilization data from the absorbed population also enables triage. The JAMA aggregate data shows no difference in average HCC scores, but averages conceal distribution tails. The subset of displaced members who had complex chronic conditions managed at a now-exited plan, who experienced care disruption during the transition, and who presented to the new plan with delayed or deteriorated conditions is precisely the cohort that generates out-of-period IBNR surprises in H2 2026 and early 2027. Identifying that cohort via H1 utilization flags and routing them to proactive care management outreach reduces the claim volatility that feeds 2027 combined ratio uncertainty.
Provider contracting is a longer-duration lever, but one where the 2026 displacement data creates actionable signals. If the receiving plan inherited significant displaced enrollment from specific county exits, its network adequacy and provider contract rates in those counties may not reflect the increased member density. The 2026 utilization data will surface where displaced member care is concentrating geographically and by provider type, informing 2028 contracting strategy before those negotiations begin.
For 2028, the bid cycle provides the first opportunity to fully price in the 2026 displacement effects. Plans that track the coding catch-up in their 2027 experience data, separating it analytically from organic trend using a displaced-member cohort flag in their data warehouse, will have a more accurate trend baseline for 2028 bids than plans that blend the cohort into the general trend calculation. Oliver Wyman's negative operating leverage projection assumes the 2027 rate environment, not the 2028 one. How aggressively the coding catch-up dynamic distorts 2027 experience trend will determine whether 2028 offers plans a recovery window or compounds the exposure.
Why This Matters for 2027 Plan Economics
From tracking membership disruption patterns across prior MA exit cycles, absorbed displaced members consistently enter receiving plans with suppressed coding in their first enrollment year, creating what appears to be trend acceleration in year two that is partly a coding artifact. The 2026 displacement wave is larger by an order of magnitude than prior cycles, and it arrives on top of a rate environment that Oliver Wyman already characterizes as structurally insufficient for most plans before the displacement layer is counted.
The 2027 rate math is set: 2.48% effective rate, a projected 3.32% reduction from V28 recalibration and normalization, against 8 to 10% medical cost trend. The displacement math is not fully known: 2.9 million members moved plans, their year-one coding profiles are suppressed, their rural-urban composition is shifting the pool geometry in ways that aggregate HCC averages do not capture, and their coding catch-up will begin appearing in 2027 utilization data. Plans that model these effects separately from organic trend, accelerate AWV outreach in H2 2026, and flag displaced-member cohorts in their reserving and pricing data infrastructure will carry a more accurate picture of their 2027 exposure into the year. Plans that do not will find 2027 worse than the bid predicted, and will not understand why until 2028.