Centene Corp CNC

Thesis

Centene is a government-payor managed care utility: states pay it a per-member-per-month (PMPM) capitation fee to administer Medicaid (and to a lesser extent Marketplace, Medicare Advantage, TRICARE, and correctional health) for roughly one in twelve Americans. The thesis is not that this is a great business — it is mediocre, with a 10-year average ROIC of 8.77% — but that it is a predictable, scaled, low-quality utility trading at distressed prices because of three layered fears: (1) the 2023–2025 Medicaid redetermination cycle that purged ~25% of pandemic-era enrollees and left a sicker residual book; (2) state rate updates lagging the resulting acuity shift, compressing the Medicaid Health Benefit Ratio (HBR); (3) the looming 2026 expiration of enhanced ACA Marketplace subsidies (eAPTCs), which threatens the Marketplace book where Centene grew aggressively into the #1 position.

The scorecard reflects that mediocrity honestly: composite 68/100, capital allocation 15/25 (10-year share count up 13.77%, mostly from the WellCare and Magellan deals), valuation 22/25. Yet the math is striking. TTM owner earnings are $4.17B against a $26B market cap. P/E is 7.7x vs a 10-year average of 20.4x. The reverse-DCF embeds −9.4% perpetual decline in owner earnings — a verdict that the entire Medicaid franchise is melting. Even widening the IV range for maintenance capex uncertainty and clamping base CAGR from 36.5% to 14.0%, the scorer outputs IV_low $198, IV_base $287, IV_high $447 against a $53.34 print — a 0.19 price/IV ratio. Q1 2026 actuals contradict the death narrative: GAAP EPS $3.11 vs $2.63 prior year, premium revenue $43.9B vs $41.7B, $21B of cash on the balance sheet, net-debt-to-EBITDA of −1.52x (net cash). This is a Hold-becoming-Buy at $53 if you believe rate adequacy returns within 24 months. It is a Trim if you don't.

Moat

Centene's moat must be evaluated against Munger's standard: 'a $10 billion competitor with five years' — could Aetna/CVS, UnitedHealthcare's Optum, Humana, Elevance, or Molina meaningfully take share? The honest answer is partial moat with uneven durability.

Cost advantages (NARROW): Scale matters enormously in Medicaid managed care because the contractually-fixed PMPM rate is the same for every plan in a state — the lowest-MLR operator wins. Centene serves ~28M+ members across all programs and ~14M Medicaid lives, the largest dedicated Medicaid book in the country. This gives it pharmacy and provider negotiating leverage, claims-processing scale, and a Magellan behavioral-health backbone (now being divested per the Q1 2026 10-Q) [3,4]. But scale economies plateau: regional operators like Molina and integrated systems like Kaiser/CVS-Aetna often operate at lower MLR in their footprints. The 8.77% 10-year average ROIC is the verdict — this is a thin-margin business where fixed-cost leverage exists but is not decisive.

Intangibles — regulatory licenses (NARROW-to-WIDE in pockets): Each state Medicaid contract is a multi-year procurement won on RFP, with strict requirements for solvency, network adequacy, and regulatory track record. Buffett's framing of MidAmerican applies in spirit: 'There is no hiding your history when you stand before these regulators... they call their counterparts in other states' [5]. A new $10B entrant cannot simply buy share — they must win RFPs state by state, build provider networks, demonstrate Star ratings (CMS quality scores), and survive incumbency advantages. Centene operates in 30+ states; that breadth is genuinely hard to replicate. But the same intangible cuts the other way: contract losses (e.g., recent state RFP losses) are public, episodic, and material. The moat is real but porous.

Switching costs (NONE on the consumer side, NARROW on the state side): Medicaid members don't 'switch'; they're auto-assigned or annually re-elect. There is no lock-in. State governments, however, face genuine switching costs: re-procuring a Medicaid contract is politically and operationally disruptive, which is why incumbents win re-procurements at high rates. This is a weak structural moat, more akin to government-contractor stickiness than insurance switching costs.

Pricing power (NONE): Centene cannot raise prices — rates are set by state actuaries and CMS. This is the inverse of GEICO's 'low costs permit low prices' loop [6]: when costs rise (post-redetermination acuity shift), Centene must absorb the gap until states catch up, which can take 12–24 months. Buffett's insurance commandments [4] — particularly the willingness to walk away if the appropriate premium can't be obtained — are not really available to Centene. Walking away from a Medicaid state means forfeiting the franchise. The product is commoditized at the regulatory level, much like Buffett's description of insurance: 'they sell a commodity-like product' [3].

Network effects (NONE): No two-sided marketplace dynamics; provider networks are built bilaterally.

Competitor stress test: Could Optum, with $10B and five years, take Centene's Medicaid book? Probably not wholesale — Optum's strategic priority is provider integration and Medicare Advantage, not state Medicaid bid-and-build. Could Molina take share state by state? Already does — Molina recently won/expanded contracts where Centene lost. Could AI-driven utilization management compress Centene's SG&A advantage? Plausibly, over five+ years.

Erosion risks: (1) Federal block-granting or per-capita caps (a perennial GOP proposal); (2) state-level disenrollment of MCO model in favor of fee-for-service — rare but exists; (3) CMS Star ratings degradation in Medicare Advantage (a known issue Centene flags in 10-K).

The honest read is that Centene has a narrow but genuine moat anchored in regulatory licensing, multi-state operating breadth, and Medicaid-specific scale. It is not a Coke or a railroad. It is a regulated low-margin utility whose competitive position is durable enough to deliver high-single-digit ROIC across cycles, not double-digit excellence.

Moat verdict: NARROW.

Management & Capital Allocation

Capital allocation is where the bear case has the most legitimate purchase. The 10-year share count is up 13.77%, not down — net dilution despite ~$26B market cap and substantial buybacks. The reason: two large stock-funded acquisitions, Health Net (2016, ~$6B) and WellCare (2020, ~$17B, the transformative deal), plus Magellan (2022). The thesis was scale-driven Medicaid leadership; the execution has produced a company that is bigger, more diversified, and arguably no more durably profitable than the pre-WellCare Centene on a per-share basis.

The five capital-allocation choices:

Reinvest in the business: Centene's organic reinvestment is essentially state RFP bid expense, IT, and clinical infrastructure. Returns are modest (10y avg ROIC 8.77%). Maintenance capex is uncertain — the scorer flags >50% spread, hence widened IV range. Free cash conversion of 145% over 5 years is very strong, reflecting the float-like dynamic of premium receivables and medical claims liabilities (premiums collected before claims paid, similar to insurance float, though state-regulated and not as durable as Berkshire's [1,2]).

Acquisitions: Mixed. WellCare added critical Medicaid scale and Medicare Advantage entry. Health Net was a Marketplace and California Medicaid play. Magellan (behavioral + Magellan Rx) was justified as vertical integration but is now being unwound — Magellan Rx sold in 2024, and per the Q1 2026 10-Q the remaining Magellan Health businesses are held for sale. Buying expensive and selling cheap a few years later is a capital allocation tell. Goodwill on the balance sheet is large ($17B+) and impairment risk exists.

Debt: Centene runs a moderate gross debt load but the scorecard shows net debt/EBITDA of −1.52x, meaning the company has more cash than debt at the holding company. $21.3B cash and equivalents at March 31, 2026 against ~$18B long-term debt. This is genuinely strong and gives management optionality. Interest coverage is unreported in the scorecard but Q1 2026 interest expense of $164M against $1.86B operating earnings suggests roughly 11x — comfortable.

Buybacks: Centene is in a 'net capital return period' per scorer notes. Q1 2026 saw modest repurchases (treasury count rose from 124.3M shares at YE 2024 to 131.7M at March 31, 2026). At $53 vs an $287 IV_base, every dollar of buyback is meaningfully accretive if the IV holds. The risk: management has historically been a serial issuer for deals, then a buyer at lower prices — pattern suggests inconsistent discipline. Average P/IV when buying is unknown; we should give partial credit, not full.

Dividends: None. Reasonable — government healthcare is too lumpy, and the company prioritizes M&A optionality.

Communication quality: Centene's disclosures are dense, lawyerly, and risk-factor-heavy — typical for a regulated payor. CEO Sarah London (took over 2022 after the Neidorff era ended) has emphasized portfolio simplification (Magellan divestitures, exit from PBM in-housing pivots) and operational discipline. Q1 2026 results — adjusted EPS up to $3.37 from $2.90, premium revenue +5.2%, GAAP EPS up 18% — suggest the operational reset is taking hold. But guidance has been volatile in recent years and the Marketplace business adds optical earnings noise.

The honest grade: Operationally competent, capital-allocation history mediocre. Net dilution despite size, expensive M&A followed by divestiture, but a strong cash position, accretive optionality at current prices, and credible recent simplification.

Capital allocator: C.

Industry Structure

Porter's Five Forces applied to government-sponsored managed care:

1. Bargaining power of suppliers (HIGH). Healthcare providers — hospitals, physician groups, pharmaceutical manufacturers — are the dominant cost driver. Provider consolidation (the rise of regional hospital systems and PE-rolled-up physician groups) has steadily shifted negotiating leverage away from payors. Pharmacy costs (especially GLP-1s, oncology, gene therapies) are a structural drag that Centene cannot pass through except via slow state rate updates. Behavioral health and specialty drug supply tightness is acute. This is the single most important force, and it is moving against Centene.

2. Bargaining power of buyers (VERY HIGH). Centene's 'buyers' are state Medicaid agencies and CMS — monopsony purchasers. They set the rate, write the contract, and audit performance. Members are not the buyer; they are the assigned subscriber. State budgets are pro-cyclically squeezed in recessions even as Medicaid rolls grow, and rate-setting actuaries lag medical cost trend. The current Medicaid HBR pressure is precisely this dynamic: post-redetermination acuity rose, but state rates were set on pre-redetermination data. Centene has no negotiating leverage other than threatening to exit a state — a costly threat rarely used.

3. Threat of new entrants (LOW–MEDIUM). Regulatory licensing, RFP track records, multi-state operating breadth, and capital requirements (state risk-based capital rules) are real barriers. New entrants are typically existing payors expanding into new states (e.g., Molina, Elevance) rather than de novo entrants. Tech-driven entrants (Bright Health, Oscar) have largely failed in Medicaid. This is the most favorable force for Centene.

4. Threat of substitutes (LOW for Medicaid; HIGH for Marketplace). Medicaid is a federally-funded entitlement; the substitute is fee-for-service (state Medicaid run directly without MCOs), which is politically disfavored and operationally inferior. Marketplace, however, faces direct substitution risk if eAPTC subsidies expire end of 2025/2026 — uninsured rates would rise and Centene's Marketplace book (where it's #1) would shrink materially. Medicare Advantage faces partial substitution from traditional Medicare as MA reimbursement is squeezed.

5. Rivalry among existing competitors (HIGH). UnitedHealthcare (with Optum vertical integration), Elevance, CVS/Aetna, Humana, Molina, and regional Blues all compete for the same state RFPs. Pricing is bid-driven; reprocurement losses are public and material. The industry is not consolidating to oligopoly — it is fragmenting at the state level. Recent Centene losses in specific states underscore that no incumbency is permanent.

Value pool location and trajectory: The value pool sits with (a) state Medicaid agencies (rate-setters), (b) consolidated provider systems, and increasingly (c) PBMs and specialty drug manufacturers. MCOs like Centene capture a thin spread (1.5–3% net margin) on enormous gross premium volumes ($43.9B Q1 alone). Trajectory: pressure on the spread from suppliers and buyers simultaneously, with relief possible only via state rate normalization (12–24 months) and benign medical cost trend (uncertain).

Industry Verdict: Average. This is a regulated commodity-utility business with structural buyer power and rising supplier power. Returns will average single-digit ROIC across cycles. Centene is a leader in a mediocre industry, not a leader in a great industry.

Inversion (Bear Case)

I am now playing the short-seller. The bull thesis is wrong, and here is why.

1. The single event that kills this: Federal Medicaid block-granting or per-capita caps. Medicaid is currently an open-ended federal-state matching program. Republican administrations and Congresses have repeatedly proposed converting it to block grants or per-capita caps (e.g., the 2017 AHCA, 2025 reconciliation discussions). Under a per-capita cap, federal payments grow at a fixed index (e.g., medical CPI minus 1%) rather than tracking actual costs. States respond by squeezing MCO rates and tightening eligibility. Centene's earnings power could be permanently re-based 30–50% lower. This isn't a tail risk — it's an active legislative agenda. With unified Republican government control as of 2025, this is the highest-probability structural threat in 20 years. The eAPTC expiration is a smaller, more proximate version of the same risk — Marketplace enrollment could shrink 30%+, gutting Centene's #1 Marketplace position.

2. Why the moat is narrower than bulls think. Centene is not like a railroad or a Coke. It is a regulated procurement winner. Recent state RFP losses (multiple states have moved business to Molina, Elevance, and others over 2023–2025) demonstrate that incumbency is not permanent. The 'multi-state breadth' moat is also a curse: every state contract is a separate political-regulatory exposure, and a single high-profile member-care failure (think the 2018 Iowa Medicaid scandal) can cascade into bid disqualification across states. The 8.77% 10-year ROIC is below the cost of equity for a healthcare insurer (~9–10%) — by Buffett's definition, this is a value-destroying business at the corporate level even before considering dilution. Bulls confuse 'big' with 'wide moat.'

3. Why management is worse than it appears. The 13.77% 10-year share count increase is the single most damning capital allocation fact. Centene used stock as currency to buy WellCare at peak Medicaid optimism and Magellan at peak vertical-integration enthusiasm. Both look mistaken in retrospect (Magellan is being divested at a loss). The compensation structure rewards revenue and adjusted-EPS growth, not per-share owner-earnings. The Neidorff era ended with regulatory settlements (PBM transparency lawsuits across multiple states cost ~$1B+). The new CEO has been operationally credible but has not yet faced a true crisis. The Magellan held-for-sale designation in Q1 2026 confirms management is admitting prior errors — fine, but the errors cost shareholders.

4. What bulls are extrapolating that won't hold. Bulls extrapolate (a) state rate updates fully catching up to acuity, (b) Marketplace remaining a profit center post-eAPTC, (c) Medicare Advantage Star ratings recovering, and (d) buybacks at $53 being highly accretive. Each is questionable. State actuaries are systematically conservative — they'll under-true-up to protect state budgets. Marketplace post-eAPTC could go from +profit to −profit as healthier enrollees leave first. CMS has been tightening MA payment policy; Centene's MA Star ratings have been below peers. Buybacks are accretive only if IV holds — if structural Medicaid reform happens, the IV anchor itself moves down 30%+. Bulls are running the bull-case state-catch-up scenario without weighing the Medicaid reform tail.

5. Valuation trap (multiple compression / regime change). The 7.7x P/E vs 20.4x 10-year average looks like a screaming bargain. It isn't necessarily. The 10-year average covers a regime of Medicaid expansion (ACA), pandemic enrollment surge (2020–2022), and easy fiscal policy. The 2026+ regime is the opposite: redetermination overhang, eAPTC expiration, Republican fiscal restraint, and provider cost inflation. Multiples in regulated buyer-monopsony industries don't mean-revert when the regime changes — they re-rate permanently lower. UK utilities post-Ofgem reform, US tobacco post-MSA, and rural hospital ownership stocks all show this pattern. Centene's 'normalized' P/E may be 9–11x, not 20x. Apply 10x to $4.17B owner earnings × $26B = right where we are. No upside from multiple expansion in the bear regime.

Bear-case price target math: TTM owner earnings $4.17B. Bear normalization: assume eAPTC expiration and Medicaid reform reduce normalized owner earnings by 35% to $2.7B. Apply a regulated-utility 9x multiple = $24B equity value. Divide by 494M diluted shares = $49/share. If a true Medicaid block-grant passes, owner earnings could drop 50% to $2.1B, 8x multiple = $17B = $34/share. Add operational execution risk (a missed quarter, a state contract loss, a Star ratings downgrade), and a 6–9 month panic could drive the stock to $30–35.

If I am right, the stock could be worth $30–40 within 2 years.

Lollapalooza Bias Check

Several biases are active in me right now as I analyze this position:

Anchoring (HIGH): The IV_base of $287 is anchoring my thesis. Once you see a 0.19 price/IV ratio in the scorecard, the brain wants to find reasons the IV is right rather than reasons it's wrong. I have to consciously remind myself the IV is a model output that compounds assumptions about owner-earnings durability, capex normalization, and discount rates. The scorer's own notes — 'Maintenance capex uncertain (>50% spread); widen IV range; base CAGR clamped from 36.5% to 14.0%' — are warnings I should take seriously, not edge cases.

Recency / availability (MEDIUM, in BOTH directions): Recent Q1 2026 strong results ($3.11 GAAP EPS, beating prior year by 18%) make the 'inflection has arrived' narrative feel obvious. But one quarter is small evidence; medical cost trend is volatile; HBR can swing 100bps quarter to quarter. Conversely, the bear narrative (redetermination, eAPTC, state losses) is also recency-loaded — these stories peaked in 2024 and may already be partially priced in.

Confirmation bias (HIGH): The setup — 'beaten-down healthcare insurer at 7.7x P/E with $21B cash' — is exactly the value-investor dopamine pattern. Munger would call this a 'pattern recognition' trap: matching this superficially to past distressed-financial winners (post-2008 banks, etc.) without verifying the deep mechanics are similar. They are not. Banks have pricing power on net interest margin; Centene does not.

Authority bias (MODERATE): The scorecard is presented as 'ground truth' and 'deterministic Python.' That doesn't mean the IV outputs are accurate — it means the math is consistent. The IV is only as good as the owner-earnings normalization. With maintenance capex ambiguous and TTM owner earnings of $4.17B partially reflecting the favorable receivables/claims-reserve dynamics of a particular accounting period, treating IV_base as a hard number would be authority-deference.

Deprival super-reaction (LOW for me, HIGH for existing holders): I don't own the stock, so I don't experience the 'I've lost 50% from highs' deprival pain. Holders do, and that explains some of the panic selling. Inverting: if I were a holder, I would be irrationally tempted to average down to recover dollar-cost basis. Recognizing this lets me decouple position-sizing from cost-basis recovery thinking.

Incentive bias (MEDIUM): Sell-side analysts are paid to publish theses; the consensus has migrated bearish over 2024–2025, which creates a herd from which contrarian rewards can come — but also a herd that can be right. Compounder analysts are incentivized to find the 'misunderstood quality compounder.' Centene is not a quality compounder. The analyst-side incentive is to talk myself into a Buy.

Net effect: I should down-weight my conviction on this name. The IV math says strong-buy; the qualitative business quality says Hold-to-Buy at most. The right answer threads that needle.

10-Year Outlook

10-year outlook test:

Same fundamental business model in 10 years? Probably yes — states will still administer Medicaid, will still contract with MCOs to manage utilization and provider networks, and will still pay PMPM capitation. The Medicaid managed care penetration rate is ~75% of Medicaid lives nationally; that's unlikely to reverse to fee-for-service. But the rate-setting regime, eligibility scope, and federal funding structure could all change materially, and any of those would re-base earnings power.

Customer base larger? Probably yes in absolute terms — U.S. demographics (aging, dual-eligible growth) push government healthcare program enrollment higher over decades. But Centene's Medicaid book specifically depends on (a) federal eligibility rules, (b) state contract retention, and (c) competitive RFP outcomes. Not guaranteed.

Profit per customer higher? Uncertain. Provider cost inflation, specialty drug costs, and aging acuity push medical costs up. Rate adequacy is reactive. Without scale-driven SG&A leverage or vertical integration breakthroughs, per-member margins are flat at best.

Moat wider? Probably no. The narrow regulatory-licensing moat erodes as competitors (Molina, Elevance, Optum-affiliated plans) match operating breadth. Tech-driven utilization management could compress one of Centene's relative cost advantages. The moat is more likely to stay narrow than to widen.

Single biggest threat? Federal Medicaid structural reform — block grants or per-capita caps. This is not a tail risk; it's an active policy debate with unified GOP government control. Probability over 10 years: 30–45%. Impact: 30–50% earnings re-basing.

Confidence: I can model the next 18–24 months reasonably (state rate normalization, eAPTC outcome). I cannot confidently model the 10-year arc because federal Medicaid policy is exactly the kind of regulatory outcome prediction that Munger's circle-of-competence test flags as auto-fail-adjacent. The business shape is roughly the same; the earnings power is regime-dependent.

CONFIDENCE: medium