UnitedHealth Group: Bidding for Medicare Advantage Under a Changing Model

Topic

Risk Management, Growth Strategy, Pricing Strategy

Industry

Managed Health Care

Internal ID

S&F-STR-564

Source

Strategy & Finance

Publication date

5 October 2026

Pages

7

License

Not specified

Download case PDF (7 pages)Request teacher’s pack

Case information cutoff: June 2, 2023

On Friday, June 2, UnitedHealthcare's Medicare Advantage team had one last weekend before submitting its 2024 bids to the Centers for Medicare & Medicaid Services (CMS). The deadline was Monday, June 5. [11] The company had to decide which county offerings to retain and how to set premiums and benefits for the next contract year. Should it preserve broad availability and keep growing membership, or selectively reprice, redesign, or leave offerings where expected economics no longer justified them?

The choice was familiar, but the payment outlook had shifted. CMS's final March announcement projected average Medicare Advantage revenue growth of 3.32% for 2024, compared with 1.03% in its January advance notice. That national estimate combined several moving parts, including a revised risk-adjustment model and lower average star-rating effects. It did not tell UnitedHealthcare what any specific contract or county would receive. The bids were due before 2024 claims experience or member response could be observed.

This case puts students in the role of advisers to a hypothetical UnitedHealthcare bid committee on June 2, 2023. The committee must choose a county-by-county portfolio posture and define the evidence and safeguards needed to carry it out. The case does not claim that the company held such a meeting or made an undisclosed decision.

A large book with uneven local economics

UnitedHealth Group reported 7.105 million UnitedHealthcare Medicare Advantage members at year-end 2022, an increase of 615,000, or 9%, from 2021. Medicare & Retirement reported $113.671 billion of revenue that year. Those figures show scale and growth; they do not reveal the margin of the MA book or the economics of an individual county. The broader UnitedHealthcare segment reported $249.741 billion of revenue and $14.379 billion of operating earnings, a 5.8% margin. It includes Employer & Individual, Medicare & Retirement, and Community & State, so its margin cannot be used as an MA margin. [1]

UnitedHealthcare's filing described county participation and selective changes to benefits, premiums, and county offerings as tools when benchmarks or other funding come under pressure. It also noted that bids depend on assumptions about enrollment, utilization, medical costs, and competitors. A bid that proves too low can expose the plan to costs it cannot quickly reprice; a less attractive offer can lose members or weaken local scale. The public filing did not identify which 2024 counties were at risk or disclose their margins. [1]

The organization also brings together businesses with different roles. UnitedHealthcare is the insurance business; Optum includes health services, data and technology, and pharmacy services. UnitedHealth Group's first-quarter 2023 release reported growth in Medicare & Retirement revenue and discussed value-based care within Optum Health. Consolidated and segment results include multiple customers, products, and payment arrangements. They cannot establish that an Optum service will reduce costs for a particular UnitedHealthcare contract. [2][3]

The 2024 payment bridge

CMS's final announcement raised its projected average MA revenue change for 2024 to 3.32%, from the 1.03% estimate in the advance notice. CMS attributed the final average to a 2.28% effective growth rate, a −1.24% average star-rating effect, a −2.16% combined risk-model and normalization effect, and a 4.44% average MA risk-score trend. The components interact. CMS described risk-score trend as a national estimate and said plan-specific experience varies. The 3.32% figure is not a UnitedHealthcare rate increase, profit forecast, or amount available for extra benefits. [4]

For 2024, CMS finalized a transition: 67% of risk scores would use the 2020 model and 33% the updated model. The updated model used ICD-10 diagnoses and more recent fee-for-service diagnoses and costs, and it revised categories subject to coding variation. CMS planned to increase the updated model's share to 67% in 2025 and 100% in 2026. A 33% share does not mean any one plan's payment falls by 33% or by a fixed percentage. The effect depends on its members, diagnoses, benchmarks, and other bid inputs. [4][5]

Star ratings also affect the offer's economics and appeal. UnitedHealth's filing described four stars as the threshold for quality bonuses. CMS reported that the average rating effect on 2024 payments was −1.24%, but contract-level effects differ. A bonus may support benefits and marketability, yet the national evidence on whether the Quality Bonus Program itself improved measured quality is mixed. Neither a group average nor a bonus alone indicates a particular contract's quality or clinical performance. [1][4][8]

What risk adjustment can and cannot tell the committee

Risk adjustment is meant to account for differences in expected health needs and reduce incentives to avoid people with greater needs. Diagnoses supported through care can improve planning and payment accuracy. At the same time, payment tied to diagnoses creates incentives for more intensive coding, and CMS's updated model specifically addressed conditions subject to coding variation.

Research adds context, but not a UnitedHealth forecast. Geruso and Layton estimated diagnosis-based risk scores in Medicare Advantage to be 6%–16% higher than the same enrollees would have produced in fee-for-service Medicare, using two research designs to address selection. They also found more intensive coding associated with vertical integration. Their estimate is program-wide, based on data ending in 2011, and does not show that a given diagnosis is inaccurate or measure UnitedHealth's coding. [6]

Jacobs and Kronick studied Medicare Advantage and Part D records from 2008 through 2015. Their abstract reports that each dollar of potential coding-related revenue was associated with $0.10–$0.19 lower bids and a reduction in medical loss ratios associated with $0.21–$0.45 in higher profitability; bid effects were larger in more competitive counties. The authors' full text was not available for this case research, and the findings are historical average estimates, not a current issuer pass-through rate. They suggest that competitive conditions can affect how payment changes reach benefits and plan finances, but cannot be applied as a UnitedHealthcare bid coefficient. [7]

MedPAC estimated that MA risk scores in 2021 were 10.8% above comparable fee-for-service scores; after its estimated 5.9% coding adjustment, the remaining difference was 4.9%. It estimated $17 billion in 2021 MA payments attributable to coding intensity. These are commission estimates with methodological assumptions, not a measurement of UnitedHealthcare or of the 2024 model. [9]

The distinction matters for strategy. If the committee assumes past risk-score growth will fully offset the model transition, it may price too optimistically. If it treats every additional diagnosis as mere payment capture, it may disregard useful care documentation and outreach. The public information available to the committee does not establish UnitedHealthcare's own coding intensity, the clinical value of its outreach by county, or how each contract will score under both models.

Growth, discipline, and the local offer

UnitedHealthcare can argue for broad availability. A larger member base may support local networks, brand visibility, provider relationships, and learning across markets. Its scale and affiliated services could help coordinate care or document genuine health needs. Those are plausible mechanisms to test. The public data do not identify which counties have enough provider capacity, lower avoidable utilization, or incremental returns from those services.

Selective repricing or redesign has a different logic. County benchmarks, risk mix, star bonuses, provider costs, and competition vary. A plan with weak expected economics might improve its position through a premium adjustment, a narrower or differently structured benefit, or an exit. Each choice may affect member retention, care access, network position, and future scale. Those responses are not quantified in the public evidence.

Competitors also combine scale and care delivery in different ways. Humana reported 4.566 million individual and 0.565 million group MA members at year-end 2022, and said 70% of its individual MA membership was in value-based relationships. It attributed part of its lower 2022 individual MA benefit ratio to higher premiums and lower inpatient utilization, among other factors. This is management's account, not proof that integration caused lower utilization. The membership definitions and business mix also differ from UnitedHealth's. [10]

The committee therefore needs a rule for local choices, not a single national answer. It must weigh the county bid and benchmark, projected utilization and medical costs, model-specific risk scores, star-related payments, competitor offers, member response, benefit expense, administrative cost, and delivery capacity. It must also decide what minimum service and access standards apply if financial pressure leads to changes.

Decision assignment

Recommend a 2024 county and benefit portfolio posture as of June 2, 2023. Choose broad availability, selective repricing/redesign/exits, or a defined hybrid. Explain:

1. What conditions make a county or contract eligible for each action?
2. Which assumptions should determine the bid, and how should the committee test downside cases for medical costs, risk-model effects, star ratings, and member response?
3. What care, access, compliance, and financial guardrails should apply to coding outreach and benefit changes?
4. What evidence should management collect after bids are submitted to decide whether to expand, revise, or exit an offering in a later cycle?

Do not infer local unit economics from UnitedHealthcare's consolidated margin, Medicare & Retirement revenue, CMS's national estimate, Optum Health's margin, or academic coding estimates. County-level bids, margins, diagnosis mix, competitor offers, and enrollment elasticity are unknown. Any numerical county inputs or hurdle rates in your recommendation must be labeled as teaching assumptions.

Exhibit 1. CMS's CY2024 average payment estimate

CMS's components are presented as an average payment impact and can interact. The difference between the final and advance estimates is arithmetic. These values do not estimate any UnitedHealthcare contract's payment, bid, revenue, margin, or benefit capacity. [4]

Exhibit 2. Scale and reporting boundaries

Exhibit 3. CY2024 risk-model transition known at the cutoff

CMS said the updated model uses 2018 diagnoses and 2019 costs, relies on ICD-10, and revises categories subject to coding variation. CMS's average risk-score trends were 5.00% for the current model and 3.30% for the revised model, blended to 4.44%; CMS cautioned that individual plan experience varies. These are program estimates and model features, not UnitedHealthcare forecasts. [4][5]

Exhibit 4. Illustrative county bid worksheet

Use the following structure for a county chosen by the committee. No company or county values are supplied; students may propose labeled assumptions or specify what data are required.

A simplified contribution view is: payment and member premiums + permitted rebate value − expected medical cost − administration − incremental benefit and service cost. This is a decision aid, not a complete bid formula or statutory accounting measure. Rebate rules, quality bonuses, and the medical-loss-ratio requirement constrain how the expression relates to cash profit. [1][4]

Sources

1. UnitedHealth Group, 2022 Form 10-K, Items 1, 1A, and 7, especially pp. 4–5, 16–18, and 25–27. SEC filing.
2. UnitedHealth Group, Q1 2023 earnings release, April 14, 2023, opening results and UnitedHealthcare / Optum Health discussion. SEC exhibit.
3. UnitedHealth Group, Q1 2023 Form 10-Q, MD&A and Note 8. SEC filing.
4. CMS, CY2024 MA and Part D Rate Announcement, March 31, 2023, pp. 63–67 (printed pp. 64–68); “Net Payment Impact” table and notes. Final announcement PDF.
5. CMS, Fact Sheet: CY2024 Medicare Advantage and Part D Rate Announcement, March 31, 2023, sections “Net Payment Impact,” “Part C Risk Adjustment Model Revision,” and “MA Risk Score Trend.” CMS fact sheet.
6. Michael Geruso and Timothy Layton, “Upcoding: Evidence from Medicare on Squishy Risk Adjustment,” Journal of Political Economy 128(3), 2020, pp. 984–1026. Published-article abstract and author working-paper sections 1–4 and conclusion reviewed. Published article; author working paper.
7. Paul D. Jacobs and Richard Kronick, “The effects of coding intensity in Medicare Advantage on plan benefits and finances,” Health Services Research 56 (2021), pp. 178–187. Abstract and bibliographic record reviewed; full text access not confirmed. DOI.
8. Adam Markovitz et al., “The Medicare Advantage Quality Bonus Program Has Not Improved Plan Quality,” Health Affairs 41(2), 2022, pp. 230–237. Methods, results, sensitivity analyses, and limitations reviewed. Article.
9. Medicare Payment Advisory Commission, Report to the Congress: Medicare Payment Policy, March 2023, Chapter 11, pp. 353–357. Chapter PDF.
10. Humana, 2022 Form 10-K, Item 1 and Item 7, including membership and benefit-ratio discussion. SEC filing.
11. CMS, CY2024 Medicare Advantage application and bid calendar, section 1.8, printed p. 10, “June 5, 2023 — Bids due to CMS.” Application PDF.

All decision evidence is limited to information available by June 2, 2023. Later bids, ratings, enrollment, financial results, policy changes, and outcomes are excluded.

Back to strategy case studies