Business

Know the Business

Centene is a government-payer aggregator: it collects fixed per-member premiums from state Medicaid agencies, CMS, and the ACA Marketplace, then tries to deliver care for less. In a normal year, that math nets out to a 1.5–2% operating margin and a ~12% ROE — there is no second act if the medical-cost line moves against you. 2025 was that year: HBR jumped 360 bps to 91.9%, the company took a $6.7B goodwill impairment, and the equity now trades near book at ~14× a 2026 recovery EPS the market does not yet trust.

1. How This Business Actually Works

Centene is paid a fixed monthly amount per member by a government program, and keeps the spread between premium and medical cost. It does not own hospitals or doctors; it buys risk from the government and re-prices it to the provider network. Profit shows up only when the rate the government set 12–24 months ago exceeds what members actually consume today.

Over 95% of premium dollars flow from a government check — Medicaid agencies in 30 states, CMS for Medicare Advantage and PDP, and CMS subsidies for the Ambetter Marketplace book. That single fact dominates the economics. Pricing is set once a year through a regulated bid or rate cell, with limited mid-year flexibility. Centene cannot raise prices when costs spike; it can only file better rates next cycle.

The economic engine is brutally simple: 92 cents goes back out the door as medical claims, ~7 cents covers all corporate overhead, and what is left — sometimes 1 cent, sometimes 3 — is operating profit. A 100-bp miss on HBR wipes out roughly half of normalized operating income on a $175B premium base.

The bargaining power asymmetry is also unusual. Centene's customers are 30 state governments and CMS, all sophisticated, regulated, and in some cases politically motivated. Its suppliers — hospitals, specialists, pharma — set list prices and increasingly use the No Surprises Act independent dispute resolution process to extract higher reimbursements. Centene's only real lever between rate cycles is utilization management and network design, which is why the 2025 turnaround program reads like a fraud-and-waste manual: ABA outlier termination, formulary control, payment integrity algorithms, and litigation against a New York provider for fraudulent NSA claims.

Where scale actually helps: state-by-state regulatory expertise, fixed-cost SG&A leverage (now 7.4% of premium, down 110 bps), and pharmacy spend ($60B annually) that Centene leverages through a transparent third-party PBM contract rather than owning one. Where it does not help: medical-cost inflation, which is fundamentally local and provider-driven.

2. The Playing Field

CNC is a top-three managed care company by revenue, but the lowest-margin name in the peer set — by design. The trade-off Centene made was scale in low-acuity, low-margin government programs in exchange for de-emphasizing the high-margin commercial group business that UNH, ELV, and CI dominate.

Three things the peer set reveals.

The valuation gap is real but explained. UNH normalizes around 7-8% operating margin because Optum (services, pharmacy, data) sits inside the company; the insurance arm is a feeder for higher-margin businesses. CNC has no Optum equivalent — it is a pure insurance carrier — so its terminal margin is structurally capped 200-300 bps below UNH. The market prices that gap correctly through EV/Revenue (CNC: 0.09x, UNH: 0.83x).

MOH is the cleaner Medicaid comp. Among the large names, only Molina runs the same Medicaid + Marketplace + Medicaid-adjacent Medicare playbook. MOH's 2025 numbers blew up similarly (op margin collapsed from ~5% to 1.7%), confirming that 2025 was a sector-wide Medicaid acuity event, not a Centene execution failure. UNH and ELV held up better because commercial group balanced the government drag.

Humana is the cautionary mirror. HUM's pure Medicare Advantage strategy looks like CNC's Medicaid bet but in a different segment — a single-customer, regulated, low-margin business that compounds beautifully until rates compress. HUM was at 4-5% op margin three years ago and is now at 1.1%. CNC investors should study HUM's 2023-2025 path before assuming Medicaid is structurally different.

3. Is This Business Cyclical?

Managed care is not GDP-cyclical, it is rate-cycle and regulation-cycle. The downturn looks nothing like a recession — it looks like a bad two-year HBR. The cycle hits margins, not revenue. Membership and premium dollars are sticky; what swings is the spread between rates locked in 18 months ago and medical cost trend today.

The pattern is clear: 2020-2022 was a windfall (COVID suppressed elective utilization, states kept paying continuous-enrollment rates), 2023-2024 the gap with reality opened as redeterminations stripped out healthier members and trend climbed, and 2025 was the rip — Medicaid HBR hit 94.9% in Q2 before recovering to 93% in Q4. Marketplace took a separate hit from a higher-than-priced morbidity baseline once enhanced APTCs were known to be expiring.

Where the cycle does NOT hit: revenue (CNC grew 19% in 2025), working capital (operating cash flow was actually $5.1B, up from $154M because of timing of state-directed payments and PDR releases), or balance sheet liquidity (medical claims liability is 46 days, well-funded). The pain is entirely in the income statement, and the recovery vehicle is the next rate cycle. Centene closed 2025 with a Medicaid composite rate increase of ~5.5% and is guiding to ~4.5% net trend for 2026 — a flat HBR is the bull case, sequential margin recovery is the call's promise.

The 2027-2028 risk is policy, not cycle. The One Big Beautiful Bill Act (passed July 2025) introduces work requirements and more frequent eligibility redeterminations starting 2027, which by management's own admission will raise the morbidity of remaining Medicaid Expansion members. The expiry of enhanced APTCs already cut Marketplace membership from 5.5M to ~3.5M heading into 2026. These are man-made cycle accelerants the 2026 guidance does not fully absorb.

4. The Metrics That Actually Matter

For an insurance carrier, P/E and revenue growth are decorative. The only metrics that explain value creation are HBR by segment, days in claims payable, the rate-trend gap, and the Star rating mix that drives Medicare revenue.

Binder Error: Set operations can only apply to expressions with the same number of result columns

Two practical interpretations. First, the Marketplace number tells you that pricing in this business is fundamentally an annual reset — the segment ran 70-71% HBR in Q1 (priced for one risk pool) and 89-90% by Q2 (the actual risk pool). That dynamic resets every January, which is why Centene's 2026 commentary keeps emphasizing the +30% rate increase already filed in 95% of states. Second, Medicaid moves through a slower waterfall: Q2 was the worst quarter, Q3-Q4 each took ~150 bps off through network and program actions. The 2026 base is whether 93% is the new normal or a way station to 91%. The street is split.

5. What I'd Tell a Young Analyst

Five things that matter more than any model you build.

Watch HBR sequentially, not the headline. Annual numbers are accounting averages; the cycle lives in the quarterly delta. CNC told you in October 2025 that Medicaid Q3 was 93.4% — that single data point was worth more than the full year guide. If Q1 2026 Medicaid HBR comes in flat at 93% with positive PYD (prior-period development), the recovery is real. If it slips back to 94%, the rate cycle did not catch acuity and the entire 2026 EPS bridge breaks.

The market's two errors here are symmetric. Bears assume 2025 broke the model — it did not, the model has always been this thin and rates always lag, the question is duration. Bulls assume 2026 is a clean snap-back — it is not, because OBBBA work requirements (2027) and APTC expiry (2026) introduce structural morbidity worsening that the rate cycle has to chase a second time. The right framing is "what is normalized adjusted EPS three years out," and consensus has moved from $9 (mid-2024) to ~$5 (today). Real value-creation hinges on whether $7 is reachable.

The goodwill impairment was not a one-off accounting line — it was a re-rating of the WellCare deal. That $6.7B writedown corresponds to the Magellan + WellCare-era M&A premium that the market had already discounted. The book value reset is now closer to the running businesses, which is helpful for clean ROE math but also says management has stopped pretending the next decade will look like the last.

Two things would change the thesis. Upside: Medicaid HBR in 2026 trends toward 91% (vs guided 93%) as rate cycles catch up — that puts adj EPS at $5+ and the stock at a 12x multiple instead of 14x of $3. Downside: A second leg down in Marketplace risk adjustment (the 2024 surprise was -$1B+) or work requirements rolled out faster than 2027 with no rate offset — that takes 2026 EPS below the $3 floor and the recovery story dies.

Don't trust the "moat" framing here. Centene has scale in state Medicaid contracting and a structural advantage in Marketplace pricing through Ambetter, but moats in regulated insurance erode fast when policy shifts. The real durable advantage is operational: 75 fraud algorithms, payment integrity, network design, and a tight third-party PBM contract on $60B of pharmacy spend. That is grinder economics, not flywheel economics. Price it accordingly.