Medicare Advantage plans make money by collecting a fixed monthly payment from the federal government for each enrollee and then spending less on that person’s medical care than the payment covers. The gap between what comes in and what goes out is the basic profit engine, but the real story is in how plans widen that gap. From aggressive diagnosis coding to prior authorization requirements to sophisticated enrollment dynamics, the strategies are layered and, in some cases, worth tens of billions of dollars a year to the industry.
How the Government Payment Gets Set
Every year, the Centers for Medicare and Medicaid Services (CMS) calculates a benchmark payment rate for each county. This is roughly what it would cost traditional Medicare to cover a beneficiary in that area. Medicare Advantage insurers then submit a “bid” representing what they believe it will cost them to provide the same standard Medicare benefits. If a plan’s bid comes in below the benchmark, the plan doesn’t just get its bid amount. It also receives a share of the difference between the benchmark and the bid, known as a rebate. Plans are required to use that rebate money to offer extra benefits to enrollees, like dental coverage, vision care, or fitness programs, but these perks also serve as marketing tools that attract more members and generate more per-person payments.
The bidding process itself reveals something important about how much pricing power these insurers have. Research examining how plans respond when CMS raises the benchmark found that for every dollar the benchmark increased, plans raised their bids by about 53 cents on average.1PubMed Central. Competitive bidding in Medicare Advantage: Effect of benchmark changes on plan bids That means plans were pocketing roughly half of every benchmark increase as additional rebate, rather than passing it all through to enrollees in the form of richer benefits. This pattern suggests meaningful market power: insurers can absorb government generosity into their margins rather than compete it away.
Risk Adjustment as the Biggest Revenue Lever
The monthly payment a plan receives for each enrollee isn’t a flat amount. It’s adjusted up or down based on how sick that person is, as measured by a formula that converts diagnosed conditions into a numerical risk score. Sicker patients with more conditions generate higher scores and therefore higher payments. The intent is straightforward: plans shouldn’t be penalized for enrolling people who need expensive care. In practice, though, this system has created the single most powerful financial lever in the Medicare Advantage business.
Plans have every reason to make sure that every diagnosable condition a member has ends up documented and submitted to CMS. One of the primary tools for this is the Health Risk Assessment, an annual check-in (often conducted in a member’s home) where a clinician reviews the enrollee’s health and records diagnoses. A study using 2019 encounter data found that about 44 percent of MA beneficiaries had at least one such assessment. Among those who did, their risk scores increased by an average of 12.8 percent as a result, and restricting the risk-score impact of these assessments could have reduced Medicare spending by somewhere between $4.5 billion and $12.3 billion in 2020 alone.2PubMed. Medicare Advantage Health Risk Assessments Contribute Up To $12 Billion Per Year To Risk-Adjusted Payments
The broader phenomenon is known as “coding intensity,” and it goes beyond health risk assessments. When the same person is enrolled in a Medicare Advantage plan versus traditional Medicare, their diagnoses tend to generate meaningfully higher risk scores under the private plan. Research has shown that MA enrollees generate risk scores about 6 to 16 percent higher than they would under traditional Medicare, where diagnoses do not directly affect most provider payments.3PubMed Central. Upcoding: Evidence from Medicare on Squishy Risk Adjustment The financial stakes are enormous. By one estimate, higher risk scores driven by coding intensity cost CMS roughly $40 billion per year in overpayments.4Health Affairs Scholar. Streamlining oversight in Medicare advantage: measuring coding intensity via contract rankings
Not every plan codes at the same intensity. When researchers measured coding inflation rates across MA contracts, the average was about 8.4 percent, but the range was wide, spanning from a negative rate (meaning some plans actually coded less intensively than traditional Medicare) up to nearly 38 percent. About two-thirds of MA enrollees were in plans whose coding inflation exceeded the adjustment CMS applied at the time to account for it.5PubMed Central. Coding intensity variation in Medicare Advantage In other words, CMS was trying to claw back some of the coding-driven overpayment, but for most enrollees, the clawback wasn’t large enough to offset what plans were actually doing.
There’s an additional twist that makes risk adjustment particularly profitable. When CMS introduced the current risk-adjustment system starting in 2004, the idea was to discourage plans from cherry-picking healthy enrollees. But research found that because the cost variation around any given risk score increases as the predicted cost goes up, plans could target people who had high risk scores but whose actual costs fell well below what the formula predicted. After risk adjustment was implemented, MA plans did shift toward enrolling people with higher scores, but those enrollees still had lower costs than their scores would suggest.6American Economic Review. How Does Risk Selection Respond to Risk Adjustment? New Evidence from the Medicare Advantage Program The system, intended to neutralize selection, had partly redirected it.
Who Enrolls and Who Leaves
Even beyond coding, the composition of a plan’s membership matters to its bottom line. Historically, Medicare Advantage plans attracted healthier enrollees than traditional Medicare. Before risk adjustment existed in a meaningful form, this favorable selection meant plans were collecting payments calibrated to the average Medicare beneficiary while serving a population that was cheaper than average.7PubMed Central. New risk-adjustment system was associated with reduced favorable selection in medicare advantage The new risk-adjustment system and a lock-in provision limiting midyear disenrollment were designed to reduce this advantage, and they did narrow the gap.
But disenrollment patterns still tilt in plans’ favor. When enrollees leave MA and return to traditional Medicare, they tend to be sicker and more expensive than those who stay. One study found that disenrollees cost Medicare about $1,021 per month in actual payments, compared with $798 in predicted payments, meaning their costs ran roughly 28 percent above what the formula expected.8PubMed Central. Impact of continued biased disenrollment from the Medicare Advantage Program to fee-for-service This pattern persisted despite substantial policy changes aimed at curbing it.
The dynamic is especially visible among the sickest patients. Among people newly starting dialysis, those enrolled in lower-rated MA plans were substantially more likely to disenroll and shift to traditional Medicare. Compared to plans rated four stars or higher, plans with 2.5 or fewer stars saw disenrollment rates that were about 12 percentage points higher among dialysis patients.9PubMed Central. Medicare Advantage Ratings And Voluntary Disenrollment Among Patients With End-Stage Renal Disease When the most expensive patients leave, the plan’s remaining membership is cheaper to serve, and the costs of those patients shift onto traditional Medicare. Plans don’t need to actively push anyone out for this pattern to benefit them financially; it happens partly because sicker patients discover that narrower provider networks and prior authorization barriers make MA less workable for complex, high-cost care.
Controlling What Gets Spent on Care
Once premiums and capitation payments are collected, the other side of the profit equation is controlling how much gets spent. Medicare Advantage plans have tools that traditional Medicare mostly lacks. The most prominent is prior authorization, a requirement that a plan approve certain medical services before they’re delivered. Traditional Medicare generally doesn’t use prior authorization for medical services, but MA plans apply it widely. Research comparing prior authorization across insurers found that its use for medical services was pervasive in MA and that the financial savings from these programs appeared to exceed the administrative costs of running them.10BMJ. Comparison of prior authorization across insurers: cross sectional evidence from Medicare Advantage
Prior authorization is controversial precisely because it works. By requiring advance approval, plans can deny or delay services they consider unnecessary or low-value. Critics argue the process also blocks medically necessary care and creates burdens for patients and doctors. From the plan’s perspective, though, it is a direct mechanism for keeping spending below the capitation payment.
Plans also manage costs through their provider networks. By contracting selectively with hospitals and physician groups, and negotiating rates below what traditional Medicare pays in some cases, plans can reduce per-service costs. Some plans go further by transferring financial risk to physician groups themselves, paying them a fixed amount per patient and letting the group absorb any surplus or deficit. These delegated risk arrangements give doctors direct incentives to manage utilization and avoid unnecessary high-cost services.11JAMA Network Open. Medicare Risk Arrangement and Use and Outcomes Among Physician Groups When the risk sits with the physician group, the insurer has effectively outsourced cost control while locking in a predictable expense.
Where the Money Actually Shows Up
Medicare Advantage is, by a meaningful margin, the most profitable line of business in health insurance. At the end of 2024, gross margins per enrollee in MA averaged about $1,655, compared with roughly $850 in the employer group market and about $990 in the individual market. The Medicaid managed care market trailed at around $610 per enrollee. Simple loss ratios, the share of premiums spent on medical claims, were about 90 percent in MA.12KFF. Health Insurer Financial Performance in 2024 That 90 percent figure might sound like plans are barely breaking even, but on a per-enrollee basis, the 10 percent left over adds up fast across the more than 30 million people enrolled.
Federal rules require MA plans to spend at least 85 percent of revenue on medical claims and quality improvement, a threshold known as the medical loss ratio (MLR). Plans that fall below it must issue rebates to enrollees. But the MLR requirement creates its own strategic dynamics, especially for vertically integrated insurers that own both the insurance plan and the medical providers. When an insurer pays its own hospitals and clinics, those payments count as medical spending for MLR purposes even if the prices are set internally. Research examining hospital-owned MA plans found that they consistently reported higher MLRs than non-integrated plans, raising the possibility that some organizations are shifting profits to the provider side of the business rather than keeping them visible on the insurance side.13Health Affairs Scholar. Hospital–Medicare Advantage vertical integration and medical loss ratios In effect, the MLR rule can be partly neutralized when the insurer is also the provider.
Vertical integration has become an increasingly important feature of the MA landscape. Several of the largest insurers now own physician practices, outpatient clinics, home health agencies, and even pharmacies. A study comparing care experiences found that legacy-integrated plans (those with long-standing integration between the insurer and provider) scored meaningfully higher on multiple quality measures, including getting appointments quickly, care coordination, and customer service, compared with non-integrated plans.14JAMA Network Open. Vertical Integration and Care Experiences Among Medicare Advantage Beneficiaries Integration can genuinely improve care delivery. But it also creates a closed loop where the insurer controls both the payment and the delivery of services, making it harder for regulators to see where profit is being earned.
How Regulators Are Trying to Rein It In
CMS has been aware of the coding intensity problem for years and has gradually tightened the screws. The most significant recent change is the V28 risk-adjustment model, phased in starting in 2024, which excludes some of the most aggressively coded diagnoses from the payment formula. The new model reduced overall risk scores by about 5.8 percent compared with the previous version, though the impact varied dramatically by insurer. Some saw their scores drop by nearly 18 percent, while others were barely affected.15Health Affairs Scholar. Exposure to the new Medicare Advantage risk adjustment model varies across insurers Insurers with higher estimated coding intensity saw the biggest score reductions, which is exactly what the model was designed to do.
The industry response has been revealing. When researchers examined how plans adjusted their benefits, cost sharing, and premiums after V28 took effect, they found that only about 17 to 24 percent of the anticipated revenue reduction was passed through to beneficiaries. Plans largely absorbed the payment cuts rather than trimming the extra benefits that attract enrollees.16PubMed. CMS’s New Risk-Adjustment Model Had Limited Impact On Medicare Advantage Benefits, 2024-25 That willingness to absorb the hit suggests that the margins were wide enough to take it, which in turn suggests the overpayments had been substantial. The researchers noted that this finding should be reassuring to policymakers considering further tightening.
CMS also conducts Risk Adjustment Data Validation (RADV) audits, which sample enrollees from a plan and check whether the submitted diagnoses are actually substantiated in medical records. But these audits cover only a small fraction of contracts and have historically been slow and contentious. The scale of the coding intensity problem, potentially $40 billion a year in excess payments, dwarfs the enforcement capacity.
The Bigger Cost Question
An underappreciated aspect of Medicare Advantage profitability is that the program has, for most of its history, cost the government more per beneficiary than traditional Medicare. A review of the program’s economic history found that while higher payments attracted greater plan availability and enrollment, Medicare Advantage generally cost more than the traditional program, with the overpayment growing over time.17PubMed Central. An economic history of Medicare part C This might seem counterintuitive, since plans claim to deliver care more efficiently. But the combination of benchmarks set above traditional Medicare costs, coding-driven payment inflation, and favorable enrollment selection has historically meant that taxpayers pay more when a beneficiary chooses MA than when the same person stays in traditional Medicare.
Plans respond strongly to the payment environment. Research on Dual-Eligible Special Needs Plans, which serve people eligible for both Medicare and Medicaid, found that higher benchmark payments influenced whether insurers chose to enter a market at all.18PubMed Central. The effects of plan payment rates on the market for Medicare Advantage Dual-Eligible Special Needs Plans When the payment is generous enough, insurers show up. When it tightens, they pull back. This sensitivity underscores that MA plan participation is fundamentally a business decision driven by expected margins, not a commitment to serve a population.
What This Means if You’re Choosing a Plan
None of this means Medicare Advantage is a bad deal for every enrollee. The extra benefits funded by rebates, like dental, vision, hearing, and fitness memberships, have real value, especially for healthier retirees on fixed incomes. And some integrated plans genuinely deliver well-coordinated, high-quality care. But understanding how these plans make money can change how you evaluate them. The free gym membership and the zero-dollar premium exist because the plan expects to collect more from the government than it spends on your care. If your health is relatively straightforward, that math tends to work for both sides. If your health becomes complex, involving multiple specialists, expensive drugs, or frequent hospitalizations, you may find that the plan’s cost-control tools start working against you.
Prior authorization denials, narrow networks that exclude your preferred specialists, and the friction of getting approvals for treatments all serve the same function: they keep spending below the capitation payment. The disenrollment data makes this tension visible. The sickest patients are the most likely to leave MA and return to traditional Medicare, not because anyone forces them out, but because the plan’s structure becomes harder to navigate when care needs escalate. If you’re considering MA, the plan’s star rating, its specific network, and its prior authorization policies are worth more scrutiny than the glossy benefits brochure. The benefits attract you in; the utilization controls are what determine whether the plan works when you actually need it.