CMS 2027 Medicare Advantage Rate Reversal: What 2.48% Means for Plan Actuaries

From modeling MA bid scenarios across rate announcement cycles since 2022, the gap between advance notice and final rate has never been this large. On April 6, 2026, CMS released the final CY 2027 Medicare Advantage and Part D Rate Announcement with a net average payment increase of 2.48%, translating to over $13 billion in additional MA plan payments relative to CY 2026. When combined with estimated risk score trend from population aging and coding practices, the total effective increase reaches 4.98%. The January Advance Notice had projected just 0.09%, or roughly $700 million. The 239 basis point swing between proposed and final represents the widest gap Georgetown's Center on Health Insurance Reforms has documented across more than a decade of tracking these announcements.

The headline number matters, but the mechanism behind it matters more. CMS reversed course on its proposed risk adjustment model recalibration, kept the existing V28 model calibrated with older data, softened the chart review diagnosis exclusion with a new switching exception, and finalized separate Part D risk adjustment model segments reflecting Inflation Reduction Act benefit changes. Each of these decisions has distinct downstream consequences for how plan actuaries construct bids ahead of the June 1, 2026 submission deadline.

The Risk Adjustment Model Reversal: Why CMS Kept the 2024 Model

The single largest driver of the advance-notice-to-final swing was CMS's decision not to finalize its proposed update to the CMS-HCC risk adjustment model's underlying data. In January, CMS had proposed recalibrating the V28 model using 2023 diagnosis data and 2024 expenditure data from the Original Medicare (fee-for-service) population. The existing model, which CMS retained for CY 2027, uses 2018 diagnoses and 2019 expenditures as its calibration base.

That five-year gap in calibration data is consequential. Healthcare spending patterns shifted meaningfully between 2018/2019 and 2023/2024, driven by post-COVID utilization normalization, the emergence of high-cost specialty pharmaceuticals (particularly GLP-1 receptor agonists and oncology therapies), and changes in diagnostic coding density. Had CMS finalized the update, the recalibrated model would have redistributed predicted spending across HCC categories in ways that reduced aggregate risk-adjusted payments. The Georgetown Medicare Policy Initiative analysis estimates this single decision avoided an approximately 3.32% reduction in payments, replacing it with only a 1.12% drag from the normalization factor alone.

CMS cited a desire to "allow the MA market more time to adjust" following the recent completion of the V28 phase-in. The V28 model reached its full 100% weighting in CY 2026, ending the multi-year transition from the prior V24 model. Layering a data recalibration on top of a model that plans had only operated under at full weight for one year would have introduced compounding uncertainty into the CY 2027 bid cycle. From a regulatory stability perspective, the decision is defensible. From a program integrity perspective, it prolongs reliance on pre-COVID calibration data, and as Georgetown noted, "the longer the agency waits to update the V28 model, the larger the year-over-year bottom line repercussions" when the recalibration eventually lands.

For plan actuaries, the practical implication is that CY 2027 risk scores will continue to reflect 2018/2019 spending relationships. Plans with populations whose condition profiles have shifted since 2019 (heavy oncology utilization, metabolic disease concentration, behavioral health) may be systematically over- or under-predicted by the retained model relative to what a 2023/2024-calibrated model would produce. That misalignment does not affect aggregate payments, since the normalization factor calibrates to total FFS spending, but it does redistribute revenue across plans with different population compositions.

Chart Review Diagnosis Exclusion: The $7 Billion Policy With a New Exception

The second major policy change CMS finalized for CY 2027 is the exclusion of diagnoses from unlinked chart review records (CRRs) from risk score calculations. Unlinked CRRs are diagnosis codes that MA plans submit to CMS without an associated beneficiary encounter; they typically result from retrospective health risk assessments and chart audits conducted by plans or their vendors to identify conditions that were documented in medical records but not reported through claims or encounter data.

CMS first proposed this exclusion in the January Advance Notice, estimating the payment impact at approximately $7 billion in CY 2027. The agency framed the change as addressing "longstanding concerns about how MA coding intensity leads to higher MA payments than under Medicare FFS, worsening the financial sustainability of the program." The MedPAC March 2026 Report to the Congress had separately documented that MA plans' reliance on chart reviews and health risk assessments has grown steadily, with some plans deriving a meaningful share of their total risk-adjusted revenue from diagnoses that would not appear in FFS claims data.

The final rule retained the exclusion but introduced a critical modification: diagnoses from unlinked CRRs will still count for beneficiaries who switch from one MA organization to another. The rationale is continuity of care: when a member moves between plans, the receiving plan does not have prior-year encounter data to capture existing conditions through its own provider network. Allowing CRR-sourced diagnoses for switchers prevents a payment cliff for the receiving plan and reduces incentives to avoid enrolling members with complex conditions.

This switching exception reduced the negative payment impact from the originally proposed 1.78% to a finalized 1.53%, a difference of approximately 25 basis points in aggregate. But the plan-level variation around that average is substantial. Plans that rely heavily on in-home health risk assessments and retrospective chart reviews to supplement encounter-based diagnoses will absorb a larger negative impact. Plans that capture most diagnoses through routine clinical encounters will see minimal effect. The Crowell & Moring analysis noted that CMS expects "the payment impact to be greater for MA organizations that heavily rely on unlinked chart review records to report risk-adjustment eligible diagnoses for their enrollee population."

Plans should also note that CMS is finalizing the exclusion of diagnoses from audio-only encounters from risk score calculations starting in CY 2027. While the dollar impact is smaller than the CRR exclusion, this change closes another pathway through which plans have historically captured diagnoses outside of standard in-person or video-based clinical encounters. The combined effect of both exclusions is to tighten the definition of what constitutes a valid diagnosis source for risk adjustment, pushing plans toward encounter-based coding models and away from supplemental assessment strategies.

Advance Notice to Final: A Component-Level Walk

Georgetown's analysis provides the clearest public decomposition of what moved between January and April. Three components account for nearly all of the 239 basis point swing.

Component	Advance Notice (Jan 2026)	Final Rate (Apr 2026)	Change
Effective growth rate	4.97%	5.33%	+0.36 pp
Unlinked CRR exclusion impact	−1.78%	−1.53%	+0.25 pp
Risk adjustment model decision	−3.32% (proposed recalibration)	−1.12% (normalization only)	+2.20 pp
Net effective change	0.09%	2.48%	+2.39 pp
With estimated risk score trend	2.54%	4.98%	+2.44 pp

The effective growth rate increase of 36 basis points reflects updated Part A and Part B per-capita spending data through Q4 2025. Stronger-than-expected inpatient utilization during winter 2025/2026 respiratory season and continued Part B drug spending growth, particularly in oncology and GLP-1-adjacent metabolic therapies, pushed FFS spending above the January trajectory. Milliman's MA Rate Notice Insights series had flagged both as upside risks during the comment period.

The CRR switching exception added 25 basis points by narrowing the scope of the diagnosis exclusion. And the risk model decision, by far the largest factor, contributed roughly 220 basis points by avoiding the payment reduction that would have resulted from recalibrating the model to post-COVID spending patterns. Patterns we have seen across prior rate announcement cycles show that upward revisions from advance notice to final are common; Georgetown documented an average upward revision of 1.26 percentage points over the past eleven years. But the 2.39 point swing in CY 2027 sits well outside that historical range, driven primarily by the risk model reversal rather than by routine data refreshes.

Part D Risk Adjustment: Separate Segments and IRA Alignment

The CY 2027 rate announcement also finalized significant changes to the Part D risk adjustment model, reflecting both technical improvements and the structural benefit changes mandated by the Inflation Reduction Act (IRA). These changes interact directly with MA-PD bid construction and deserve close attention from plan actuaries building integrated Part C and Part D benefit packages.

CMS is finalizing separate RxHCC model segments for MA-PD plans and standalone prescription drug plans (PDPs). The existing model applied the same relative risk factors across both plan types, despite well-documented differences in their enrollee populations, formulary structures, and utilization patterns. The segmented approach aims to "improve the model's predictive accuracy for MA-PD plans and PDPs compared to a model that uses the same relative factors," according to the CMS press release.

For MA-PD plans, the practical impact depends on how the new MA-PD-specific factors redistribute predicted drug spending across the plan's population. Plans with high specialty drug utilization (oncology, autoimmune, hepatitis C) may see improved prediction at the member level, while plans with predominantly generic-heavy formularies may see smaller benefits from the segmentation. The segmentation also creates a new source of plan-level variance in Part D risk-adjusted revenue, which actuaries should incorporate into the Part D component of the integrated bid.

The Part D model updates also account for the IRA's restructured benefit design. CMS is adjusting baseline parameters to reflect the $2,000 annual out-of-pocket cap, updated manufacturer discount obligations, modified catastrophic phase cost-sharing, and the Maximum Fair Price provisions for selected drugs. These adjustments are technically complex because the underlying actuarial value calculations shift when the benefit phases (deductible, initial coverage, coverage gap, catastrophic) are fundamentally redesigned. The Avalere Health Advisory analysis noted that the advance notice "materially alters Part D risk adjustment" in ways that interact with the first full year of post-IRA claims experience.

Importantly, the Part D model also aligns its diagnosis sourcing policies with Part C. Diagnoses from unlinked chart review records and audio-only encounters will be excluded from Part D risk adjustment calculations, with the same MA-switching exception applied to Part C. This alignment simplifies compliance for integrated MA-PD plans but imposes an additional revenue adjustment for plans that had relied on CRR-sourced diagnoses in their Part D risk score calculations.

Bid Strategy Implications: How the Rate Environment Reshapes 2027 Benefits

The June 1, 2026 bid deadline gives plan actuaries roughly eight weeks from the April 6 rate announcement to finalize revenue projections, benefit designs, premium pricing, and supplemental benefit packages for CY 2027. The 239 basis point swing between advance notice and final changes the available budget for every downstream decision.

Revenue projections need recalibration. Plans that built preliminary bid models against the 0.09% advance notice baseline now have an additional 2.39 percentage points of per-member-per-month revenue to allocate. For a plan with 100,000 members and a $1,200 monthly benchmark, that swing translates to roughly $34 million in additional annual revenue. The decision of where that revenue goes, whether toward enriched benefits, restored margins, premium reductions, or provider rate increases, is the central strategic choice of this bid cycle.

Supplemental benefits face expansion pressure. Historically, favorable rate years have prompted plans to compete on supplemental benefit richness: dental, vision, hearing, over-the-counter allowances, transportation, meals, and fitness benefits. Oliver Wyman's April 2026 analysis notes that plans should "conduct a thorough review of supplemental benefits and SSBCI offerings to ensure they deliver measurable value while meeting heightened documentation standards." CMS has strengthened program integrity requirements for Special Supplemental Benefits for the Chronically Ill (SSBCI) in the CY 2027 final rule, including eligibility standards and new safeguards around debit card-based benefit delivery. Plans expanding supplemental benefits must therefore balance competitive positioning against tighter compliance expectations.

The margin restoration versus benefit expansion trade-off is plan-specific. Not all plans enter the 2027 bid cycle from the same position. UnitedHealthcare, the largest MA carrier with roughly 9.4 million members as of February 2026, reduced enrollment by approximately 9% from its October 2025 peak, reflecting deliberate county exits and benefit trims to restore margins after the challenging 2024/2025 cycle. Humana, by contrast, expanded its MA presence during 2026 and may be positioned to surpass UnitedHealthcare as the largest MA insurer. CVS/Aetna also reduced service areas for 2026. For carriers that cut markets to stabilize profitability, the 2027 rate environment raises a question: does the higher baseline justify re-entering exited counties, or is the margin buffer better held in reserve?

Part D formulary and benefit design require joint optimization. CY 2027 is the second year operating under the IRA's $2,000 Part D out-of-pocket cap. First-year claims experience from CY 2025 and early CY 2026 data, as our analysis of year-one Part D redesign data documented, shows catastrophic phase utilization running roughly 22% above base-case projections from Milliman and Wakely, driven by GLP-1 adherence improvements and specialty drug utilization under the reduced cost-sharing structure. MA-PD plans constructing integrated Part C and Part D bids must reconcile the higher Part C revenue with the ongoing Part D cost variance. The segmented Part D risk adjustment model may partially address this misalignment, but first-year experience remains the best available guide for 2027 pricing assumptions.

The Chart Review Revenue Cliff: Quantifying Plan-Level Exposure

The unlinked CRR exclusion deserves a deeper actuarial look because the plan-level impact distribution is highly skewed. National health plans and vertically integrated organizations that operate large-scale retrospective chart review programs, including those that contract with specialized risk adjustment vendors for in-home health risk assessments, face the largest revenue exposure.

Consider the mechanics. Under the prior policy, a plan could submit a diagnosis code identified through a chart review of a member's medical records even if that diagnosis was not associated with a specific clinical encounter during the measurement year. These supplemental diagnoses flow into the member's risk score, increasing the plan's capitated payment. The CY 2027 exclusion breaks that pathway for all diagnoses not linked to a specific encounter, except for the switching exception.

Plans with mature encounter-data capture, where the vast majority of diagnoses are reported through primary care and specialist visits, through emergency department claims, or through inpatient discharge coding, will see minimal impact. These plans may lose 0.5% or less in risk-adjusted revenue. Plans where 5% to 15% of total risk-adjusted revenue derives from CRR-sourced diagnoses face a proportionally larger reduction. Industry estimates suggest the plan-level range spans from near zero to 3% or more of total revenue for the most CRR-dependent organizations.

The strategic response is already visible in the market. Plans are accelerating investment in prospective coding programs, where providers identify and document conditions during routine clinical encounters rather than through retrospective chart review. This shift aligns with CMS's long-term goal of moving MA coding practices closer to the FFS encounter-based model, but the transition requires investment in provider education, electronic health record (EHR) integration, and workflow redesign. Plans that have not already built prospective coding infrastructure face a compressed timeline to implement it before CY 2027 diagnoses begin accruing.

Competitive Landscape: Winners and Losers in the 2027 Rate Cycle

The stock market reaction to the April 6 announcement offered a blunt signal of how investors read the rate decision. UnitedHealth Group, Humana, and CVS Health each climbed more than 8% in the session following the release, according to STAT News. The market's read was straightforward: $13 billion in additional payments flows directly to plan revenue lines.

But the competitive dynamics underneath that headline are more complex. The rate environment creates an asymmetric benefit across plan types and market positions.

Broad-market HMOs benefit most. Plans with general-enrollment populations, strong encounter-based coding, and presence in counties with above-average FFS cost growth receive the full benefit of the effective growth rate increase and face minimal exposure to the CRR exclusion. These plans have the widest latitude to expand supplemental benefits, reduce premiums, or build margin cushions.

Dual-focused plans benefit less. D-SNPs and other dual-eligible-concentrated plans face a smaller net revenue benefit for three reasons: the V28 model redistributes HCC weight away from conditions prevalent in dual populations; dual enrollees concentrate in lower-income urban counties where FFS growth often runs below the national average; and the policy tightening on look-alike D-SNP contracts and integrated care requirements adds compliance costs that offset revenue gains. Plans modeling dual-focused books may see net effective increases in the range of 0.8% to 1.5%, well below the 2.48% headline.

CRR-dependent plans face a structural adjustment. Organizations that have built revenue models around large-scale retrospective chart review operations, whether in-house or through vendor partnerships, must absorb the CRR exclusion's revenue impact while simultaneously investing in prospective coding alternatives. This creates a temporary margin squeeze for plans in transition.

Market re-entry decisions loom. Carriers that exited counties or reduced plan offerings during the 2025 and 2026 annual enrollment periods to stabilize profitability now face a decision window. The 2027 rate environment, with its higher baseline and revenue certainty, makes re-entry more attractive than the January advance notice implied. But plans considering re-entry must weigh whether the rate improvement is sustainable. Georgetown's documentation that upward revisions from advance notice to final are "structurally embedded in the MA payment process" suggests the current rate announcement is not an anomaly, but the specific composition of the 2027 swing, heavily dependent on a one-time risk model reversal, may not repeat in CY 2028.

The Structural Risk: Deferred Model Update and the 2028 Cliff

CMS's decision to retain the 2024 risk adjustment model for CY 2027 solves a near-term stability problem but creates a medium-term risk. The V28 model will eventually be recalibrated to more recent data, and every year the update is deferred increases the magnitude of the eventual adjustment.

The 2018/2019 calibration data predates several structural shifts in Medicare spending: the COVID-19 pandemic's lasting impact on utilization patterns, the explosive growth of GLP-1 and oncology spending, the expansion of telehealth as a permanent care delivery channel, and shifts in post-acute care utilization following the Patient-Driven Payment Model (PDPM) reform. A model recalibrated to 2023/2024 data would capture all of these dynamics. A model calibrated to 2018/2019 data does not.

Georgetown's analysis warned that "the longer the agency waits to update the V28 model, the larger the year-over-year bottom line repercussions" when the recalibration eventually occurs. If CMS proposes the recalibration in the CY 2028 Advance Notice (expected January 2027), plans will face the full accumulated recalibration impact in a single year. That impact could be significantly larger than the 3.32% reduction proposed for CY 2027, because it would incorporate an additional year of spending pattern divergence.

Plan actuaries building three-year strategic projections should scenario-test the recalibration in both CY 2028 and CY 2029, with the understanding that deferral increases severity. The appropriate response is not to assume the model will never be updated, but to reserve or provision for the recalibration in internal projections even though the CY 2027 rate environment provides short-term relief.

Network Adequacy and Provider Contracting

The rate environment has secondary effects on provider contracting that plan actuaries should incorporate into medical cost assumptions. A 2.48% payment increase, or 4.98% including risk score trend, provides plans with additional negotiating room in provider rate discussions. Plans that had been constraining provider rates during the 2024/2025 margin compression may face pressure to restore rates, particularly from hospital systems and specialist groups that point to the rate increase as evidence that plans can afford higher reimbursement.

Provider contracting also interacts with the CRR exclusion. Plans shifting from retrospective chart review to prospective encounter-based coding need provider cooperation. Persuading primary care physicians and specialists to document all active conditions during each visit, including conditions previously captured through retrospective review, requires workflow changes, EHR template modifications, and sometimes financial incentives (coding quality bonuses embedded in provider contracts). These costs are incremental to the plan's medical loss ratio and should be modeled explicitly in the bid.

Network adequacy takes on additional significance because CMS finalized stricter network adequacy standards in the CY 2027 final rule, building on the CY 2026 requirements. Plans expanding into new service areas or restoring coverage in previously exited counties must meet updated time-and-distance standards, provider type requirements, and essential community provider thresholds. The cost of building compliant networks in new geographies partially offsets the revenue benefit of re-entry.

Why This Matters for the Actuarial Profession

The CY 2027 rate reversal is not just a pricing event. It is a test case for how CMS manages the tension between program integrity (updating risk adjustment to reflect current spending) and market stability (avoiding disruptive payment swings during model transitions). The resolution CMS chose, deferring the recalibration while tightening diagnosis sourcing rules, reflects a regulatory philosophy that treats the risk model and the data integrity rules as separate levers.

For actuaries working on MA plans, the June 1 bid deadline concentrates the practical consequences into a six-week window. Revenue projections, benefit design, supplemental benefit strategy, Part D formulary optimization, provider contracting assumptions, and Star Ratings revenue expectations all need to reflect the final rate parameters rather than the advance notice baseline. Plans that began bid construction against the 0.09% assumption in January have meaningful rework ahead.

For actuaries in consulting or advisory roles, the rate reversal creates an unusually rich engagement cycle. Plan-level impact analysis, CRR revenue exposure quantification, prospective coding strategy development, Part D risk model segmentation modeling, and competitive positioning analysis all flow from the April 6 announcement and all need completion before June 1.

And for the profession broadly, the Georgetown finding that upward advance-notice-to-final revisions are "structurally embedded in the MA payment process" raises questions about whether the current two-step announcement process (January advance notice, April final) produces information-efficient price signals. When plans can reliably expect upward revisions, bid behavior adjusts, and the advance notice's signaling function diminishes. Whether that structural dynamic persists in CY 2028, when the deferred risk model recalibration may re-enter the policy conversation, is the most consequential open question for MA plan actuaries heading into the next cycle.