HCA Healthcare

IThe business

HCA Healthcare runs hospitals. At the end of September 2026 it operated 189 of them, 117 freestanding surgery centers and about 2,600 ambulatory sites of care in all, including emergency rooms, urgent care centers and physician clinics, in 19 states and England.⁵⁶ It began in 1968, is based in Nashville, and employs about 320,000 people, 90,000 of them part-time or on call.² Its biggest markets are Texas and Florida. At the end of September 2026 it had 103 hospitals there, which produced 58% of its admissions and 51% of its revenue.⁶

The business is simple to describe and hard to run. A patient arrives, usually through the emergency room or on a doctor’s referral. The hospital provides care, and someone pays. Who pays decides the price. Medicare pays a rate the government sets. Medicaid pays a rate the state sets, plus extra “supplemental” money in some states. Commercial insurers pay a rate HCA negotiates, and that rate is the profit engine. A patient with no cover pays little or nothing, and the hospital still has to treat anyone who arrives in an emergency. So the three numbers that drive profit are how many patients come, how much each is worth, and the mix of who pays C (calculated).²

Who pays

The table below is the first thing to read. Medicare and Medicaid together paid 45.4% of 2025 revenue.² Commercial insurers, a group the company calls “managed care and other insurers”, paid 48.9%, and the people on the Affordable Care Act exchanges sit inside that line. Patients with no insurance sit in “other”, together with small items. The 2026 column is the first nine months.

Two lines moved. Medicaid revenue rose 36% in the first nine months of 2026, to $6.0B from $4.4B, mostly because Florida’s directed payment program won federal approval and HCA booked earlier periods at once.⁶³ Commercial revenue, where hospitals generally earn the most, rose 0.9% in nine months, and fell from 48.8% to 46.7% of the total.⁶ I come back to both below.

The exchange subsidies

The Affordable Care Act exchanges sell individual health plans. From 2021 through 2025 the government paid enhanced premium tax credits that cut what most buyers paid. They expired on 31 December 2025.² KFF counted 19.2 million people paying for exchange plans in February 2026, down 12% from 21.8 million a year earlier, and 20.5 million in the whole individual market in June, down 18%.¹⁰ Many who left have no other cover.

For HCA the effect shows up as uninsured patients. They were 8% of admissions in the third quarter of 2026 against 6% a year earlier, and uninsured admissions at the same hospitals rose 15.5%, 23.4% and 27.4% in the first three quarters of 2026.⁶ Most of the growth is in Texas and Florida, which together had 73% of uninsured admissions.⁶ The cost of uncompensated care, meaning charity care plus the discounts and price concessions on uninsured accounts, was $1.477B in the third quarter against $1.188B, a rise of 24%.⁶ The company estimates that the move from insured to uninsured cut pre-tax income by about $400M in each of the second and third quarters, and it now puts the effect for 2026 at $1.0B to $1.2B, up from $600M to $900M in April.⁵³ That is a lot of money for what management calls a payer mix shift, and it is the largest negative in the July cut to guidance, partly offset by the Medicaid gain above.

Medicaid and the 2025 budget law

Medicaid pays hospitals less than they would get from Medicare, and often less than the care costs. But several states, Texas and Florida among them, pay extra through supplemental payments and “state directed payments”, which require Medicaid managed care plans to pay hospitals at set rates. HCA took in about $6.2B this way in 2025, $5.5B in 2024 and $4.4B in 2023, which is 8.2% of 2025 revenue.² The states fund their share partly with taxes on hospitals, so the net gain is less than the gross figure. The company does not publish the net figure for the whole programme. It gave a guide for the change: an extra net benefit of $300M to $500M in 2026, up from a loss of $50M to $250M it had assumed in April, mostly because of the Florida catch-up.³

The 2025 budget law, which the company calls the FBA, tightens these programs. New directed payments are capped at Medicare rates. Programs approved before 4 July 2025 are grandfathered, but from rating periods beginning on or after 1 January 2028 they step down toward the cap.⁶ CMS issued a final rule on provider taxes in February 2026 and a further proposed rule in July.⁶ The company says it cannot predict the effect. I can’t either. The step-down is the largest single input to my range, and I model it as a loss of $1.2B a year of pre-tax income by 2032 in the base, $0.3B in the bull and $2.0B in the bear E (estimate).

Volume and price

Underneath, the business is growing. Same-facility admissions rose 2.7% and equivalent admissions 3.0% in the third quarter of 2026, and emergency room visits rose 5.1%.⁵ Surgery is the soft spot: inpatient surgeries fell 1.4% and outpatient surgeries 3.7% at the same hospitals, which matters because surgery is a rich source of commercial revenue.⁵ Revenue per equivalent admission, the price-and-mix measure, rose 3.4% over nine months and 0.6% in the third quarter.⁵ That gap is the story of 2026. Price was strong until the payer mix turned.

Medicare’s annual rate update gives a floor and a ceiling. For the federal year starting October 2025 it raised inpatient rates by about 2.6%, a market basket update of 3.3% less a 0.7 point productivity cut.² Same-facility salaries per full-time employee rose 2.7% in the third quarter, so the update roughly matches wage growth. A 2% sequester on Medicare payments continues into 2033.²⁶ CMS is also phasing out the inpatient-only list over three years from 2026 and extending site-neutral payment to more drug administration, both of which move work and money out of the hospital setting.²

Where the profit is

HCA runs three geographic groups. The National Group has 52 hospitals, the Atlantic Group 64 and the American Group 66, with seven hospitals in England in corporate and other.⁶ In the first nine months of 2026 the National Group earned $3.9B of adjusted segment EBITDA, up 4.6%, the Atlantic Group $4.6B, up 9.7%, and the American Group, which holds most of Texas, $4.3B, down 5.2%.⁶ In the third quarter the American Group earned 1.6% less and the Atlantic Group 3.8% less than a year before.⁶ Texas and Florida are where the growth is and also where the uninsured are, and the segment numbers show it C (calculated).

IIThe moat

An advantage lasts when a rival with money and talent cannot copy it in a reasonable time. A hospital is a poor place to find one, because the price of the product is set by two payers, the government and the insurer, and neither is a friend. HCA’s advantages are real, but they are mostly about cost and position, and I rate them below those of the software and branded-goods companies on this site.

The idea is simple. A hospital in a metropolitan area with a strong emergency department, a large physician network and long-term contracts with insurers is hard to replace. A patient who has a heart attack goes to the nearest hospital, not the cheapest. Insurers need the hospital in the network. HCA owns many of these, so when it negotiates with a large insurer it negotiates for dozens of hospitals at once, and the insurer’s members in that city have nowhere else good to go.

The advantage has limits. HCA does not set most of its prices, and 45% of its revenue comes from Medicare and Medicaid, where governments set or approve the price. The Medicare inpatient update for the year from October 2025 was 2.6%; same-facility salaries per full-time employee rose 2.7% in the third quarter.²⁶ The 10-K lists the payer controls: managed care, value-based purchasing, site-neutral payment and the push to move work out of the hospital.² The test of an advantage is whether it has widened or narrowed. Here is the evidence both ways.

Evidence it is widening

• Adjusted EBITDA margin of 20.6% in 2025 was the second highest of the ten years, behind 2021, and the nine months of 2026 were 19.8% even with a $1B-plus drag from the exchanges C (calculated).⁴⁵
• Labour as a share of revenue fell from 45.5% in 2016 to 43.5% in 2025, while supplies fell from 16.7% to 15.0% C (calculated).⁴
• Capital spending rose from $2.8B to $4.9B while the company bought back 214 million shares in ten years.⁴

Evidence it is narrowing

• Other operating expenses rose from 18.1% of revenue to 21.0% in 2025 and to 23.3% in nine months of 2026 C (calculated). The company names contract services, professional fees, insurance and non-income taxes as the content.⁴⁶
• Price per admission slowed to 0.6% in the third quarter, and commercial revenue rose 0.9% in nine months.⁵⁶
• Surgeries are falling at HCA’s hospitals while its stated plan is to grow its networks.⁵ I read part of the loss of insured patients as a loss of better-paying ones, though the company does not split it.

On balance the advantage is intact and narrower at the margin than it looked a year ago. HCA has shown it can absorb a $1B-plus shock to payer mix and still land in the upper half of a guidance range.⁵ What that advantage cannot do is write the rules that decide who pays.

IIIThe competitive advantage

The honest comparison is the other listed hospital operators. Tenet Healthcare runs hospitals and a large surgery-center business, Universal Health Services runs acute care and behavioral hospitals, and Community Health Systems runs smaller hospitals in smaller markets and has sold many of them. HCA is more than three times the size of Tenet by revenue.

Three things stand out. First, HCA earns 20.6% on this measure and Tenet 21.0%, about the same, so size alone has not produced a wide margin gap C (calculated).⁸² Second, HCA earns it on $75.6B of revenue, 3.5 times Tenet’s, and Universal earns 15.8% on its own mix of acute and behavioral care C (calculated).⁸ Third, the group trades together. On 4 November HCA rose 10.3%, Tenet 9.6% and Universal 10.9%, so on that day the market was pricing the sector, not the company.¹⁷

Where it wins

• Size. HCA’s revenue is 3.5 times Tenet’s and 4.4 times Universal’s, which gives it better terms on supplies, insurance and debt.⁸²
• Markets. It is the leader in Texas and Florida, where population is growing faster than in most of the country.²
• Capital. It can spend $5B a year on beds, equipment and technology, nearly five times Tenet’s $1.0B of 2025 capital spending C (calculated).²⁸
• Cost. Salaries and benefits per equivalent admission fell 0.2% in the third quarter, and supply cost per equivalent admission fell 1.4%.⁶

Where it loses

• Exposure. Texas and Florida are also where uninsured patients are; they were 73% of its uninsured admissions.⁶
• Leverage. Debt is 3.2 times my 2026 EBITDA estimate and stockholders’ equity is a deficit of $8.1B, so a miss in earnings lands on a thin base C (calculated).⁶
• Dependence on Medicaid add-ons. $6.2B of 2025 revenue, 8.2% of the total, came from supplemental and directed payments.²

IVManagement

HCA’s leadership is long-tenured and stable. Hazen has spent four decades at the company, and the chairman belongs to the founding family.⁹ The pattern in the executive team is promotion from within, a recent CFO change and one departure, the chief clinical officer. I read none of it as a warning.

Pay, and what it pays them to do

Hazen was paid $26.5M for 2025: a $1.58M salary, $5.4M of annual bonus, $8.5M of share units and $8.5M of options-style rights, and $1.9M of pension change and $0.5M of other pay.⁹ The company says 93% of it is performance based.⁹ Two things drive the pay. Eighty per cent of the annual bonus depends on EBITDA, defined to exclude interest, taxes, depreciation, the partners’ share, gains on sales, legal costs and the expense of share-based pay. The other 20% depends on quality and patient experience. EBITDA came in at or above the top of the range, so the financial portion paid at 200% of target and the whole bonus at 195.78%.⁹ The long-term awards are half share units that vest on cumulative diluted earnings per share over three years. For 2023 to 2025 the target was $55.50 and the adjusted result was $69.66, 25% above target.⁹

That design pays the managers to grow EBITDA and earnings per share. It does not charge them for the capital used, the debt carried or the stock paid out. The two measures make it attractive to spend borrowed money on buybacks: a buyback shrinks the share count and lifts earnings per share even when the business itself grows no faster. From 2016 to 2025 net income grew 9.9% a year and earnings per share 16.3% a year, because the diluted share count fell 5.4% a year C (calculated).⁴ About a third of the earnings-per-share growth came from the smaller share count C (calculated). That is a legitimate way to create value if the shares are bought below their worth, and an average price of $200 against $510.70 today suggests many of the purchases were. It is also a way to turn a good year into an excellent pay cheque when debt rises.

Their own money

Hazen owns about 1.40 million shares outright, worth about $714M at $510.70 C (calculated), plus about 0.77 million exercisable stock appreciation rights.⁹ The company’s guideline is five times salary, about $8M, and he holds many times that.⁹ All 16 directors and executive officers together owned 3.35 million shares, 1.5% of the company.⁹ The bigger owner is the Frist family. The Frisco and Hercules holding partnerships owned 38.9 million and 34.6 million shares, 17.4% and 15.5%, and the proxy puts the Frist group at about 32% of the stock.⁹ I find that owner reassuring on strategy, since the family has been the controlling hand since the company’s founding, and it also matters for supply. In February 2026 the family’s holding company swapped 36.6 million shares with HCA for the same number of new shares, in a tax-free reorganization the company says helps estate and charitable planning. The new shares carry limits on how fast they can be sold, which last until they qualify under the securities rules for non-affiliates or for six years.⁹ A family that plans estates will sell shares one day. The limits are written to make that gradual. I do not count it in my values.

What they did with the money, over a full cycle

Over the ten years 2016 to 2025 HCA generated $84.7B of operating cash flow. It spent $38.9B on capital projects, $6.4B on distributions to partners in its joint ventures, $7.9B on acquisitions, $4.5B on dividends and $42.9B on buybacks C (calculated).⁴ That is $100.7B, $15.9B more than the operating cash flow, and debt rose $15.1B, from $31.4B to $46.5B C (calculated).⁴² The company paid out more than it earned, the difference came from lenders, and the stockholders’ equity line went from a deficit of $7.3B to a deficit of $6.0B, and to $8.1B in September 2026 C (calculated).⁴⁶

I give the managers credit for four things. They cut buybacks and the dividend in 2020, when $0.4B was spent against $1.0B the year before, and debt fell by $2.7B C (calculated).⁴ They bought heavily in 2021 and 2022 at an average of $217 and $228, prices that look cheap against $510.70.⁴ They kept leverage inside a band of 2.7 to 4.0 times EBITDA through the decade, and it was down to 2.7 in 2021 C (calculated).⁴ And they raised capital spending from $2.8B to $4.9B, cutting it only in 2020. The $214 million shares bought cost $42.9B, an average of $200; they would be worth $109B now C (calculated).⁴¹

I mark them down for three. The recent buybacks have been larger and dearer: $376 a share in 2025 and $430 in the first nine months of 2026, when the price was higher and debt was rising by about $4.0B in nine months.⁴⁶ In those nine months the company paid $7.1B in buybacks and dividends, against $3.2B left after capital spending, partners, stock pay and deals C (calculated).⁶ And the bonus formula pays for EBITDA and says nothing about the debt that supports it. None of this is reckless. It is a strategy that works well while hospitals earn what they earn now, and gets expensive if they do not.

VFinancials

Ten years

Ten years of reported figures give the shape. Revenue grew 6.9% a year, from $41.5B to $75.6B. Adjusted EBITDA grew 7.0% a year and net income 9.9%. Earnings per share went from $7.30 to $28.33, 16.3% a year, helped by a diluted share count that fell from 396 million to 239 million C (calculated).⁴² The one dip was 2020, when the pandemic cut volume in the spring; revenue was flat and EBITDA margin held at 18.9% C (calculated).⁴

Margins

The adjusted EBITDA margin ran from 18.8% to 21.5% across the decade, and averaged 19.9% in 2022 to 2025 C (calculated).⁴ The 2025 figure of 20.6% was the second highest. In the first nine months of 2026 it was 19.8% against 20.4% a year earlier, and the third quarter was 19.7% against 20.2%.⁵ The company said the exchange shift cost about $400M in that quarter, which is 2.0 points of revenue C (calculated).⁵ Without it, the margin would have been above last year’s.

Costs tell where the margin went and where it came from. Labour is 43.5% of revenue, down 2.0 points in nine years, while other operating expenses are 21.0%, up 2.9 points C (calculated).⁴ In the third quarter of 2026 labour was 43.2% of revenue and other operating expenses 23.0%, up from 21.8%.⁶ Higher “other” costs include contract services, professional fees and non-income taxes. Part of the 2026 rise may be the cost of funding Medicaid payments, such as provider taxes, but the filings do not give the split, so I do not read the margin on the payer lines alone.⁶ The 10-K says contract and other premium-rate labour costs declined in 2025 from the year before, though they are still in use.²

The first nine months of 2026

The nine-month figures show an operating business doing fine and a financial one doing less well. Revenue is up 5.5% and EBITDA 2.6%. Net income is flat. Earnings per share are up 9.0% only because the share count is 8.4% lower. And operating cash flow fell 14%, the biggest part of it a swing in cash taxes of $1.2B, which suggests 2025 was flattered by timing and 2026 is the return to normal C (calculated).⁶

Guidance, and how often it moved

The company first guided in January for EBITDA of $15.55B to $16.45B, cut the range in July to $15.40B to $16.10B and said in October that it expects the upper half.³⁵ So the January midpoint of $16.0B is now a year-end estimate of about $15.9B, 0.6% lower, in a year when the estimated exchange effect rose by about half from the April view C (calculated). That is a good sign of resilience. HCA offset a $1B-plus shock with Medicaid payments, cost trends and volume, and kept to the range. It is also why I treat the Medicaid payments as the thing to watch. If the net benefit fades faster than I assume, EBITDA falls with it E (estimate).

Cash and capital

Operating cash flow was $12.6B in 2025, $1.9 times net income to HCA. After capital spending of $4.9B, partners’ distributions of $0.8B and stock pay of $0.4B, $6.5B was left. Take off finance-lease principal of $0.17B and acquisitions of $0.40B, and the reported owner cash of 2025 is about $5.9B C (calculated).²⁴ That year’s cash flow was helped. Earnings alone would give about $11.7B of operating cash, and the reported figure was $1.0B higher, the product of working capital and the timing of taxes. A normal 2025 is about $4.9B of owner cash C (calculated). I use that figure as the reference for my 2026 estimate.

Debt

Debt at 30 September 2026 was $50.5B: $5.7B of commercial paper at an average rate of 4.2% and $44.2B of senior notes at 5.1%, with an average life of 11.6 years.⁶ Of the total, $8.1B falls due within a year, mostly the commercial paper, which is backed by an $8.0B credit facility with $2.2B available after backing the paper.⁶ On 2 November the company proposed to issue new senior notes, and said the money may redeem $1.2B of 4.5% notes due in February 2027 and repay commercial paper.¹² Interest took $14% of EBITDA in 2025; the third quarter of 2026 interest expense was $607M, 8% higher than a year earlier on a higher average debt balance of $49.7B at 4.8%.⁶ A one-point rise in rates on all the debt, once refinanced, would cost about $0.5B a year, 3% of EBITDA C (calculated).

Earnings quality

Four points affect how much of the reported figure an owner can take. First, EBITDA includes the profit that belongs to minority partners in surgery centers and hospital joint ventures, $0.99B in my 2026 estimate, and about $0.8B of it is paid out each year in cash E (estimate). I count it as a cost throughout. Second, revenue is stated after estimated price concessions for patients who will not pay, and $7.7B of such concessions had been recorded against receivables and revenue at the end of 2025.² With uninsured admissions up 27%, how well those estimates hold is a live question. Third, the 2026 total includes a Florida catch-up of about $400M for earlier periods, which does not repeat C (calculated).³ Fourth, the company reports adjusted EBITDA after stock-based pay, not before it, which I count as a good feature. Stock pay was $0.40B in 2025.²

VIValuation

Start with what would go wrong

Before any number, I ask what would most likely hurt an owner of this stock, and work back from there. There are three ways it goes badly.

First, the coverage loss does not reverse, and it spreads to the commercial book. The enhanced exchange subsidies are gone, KFF counts fewer people with paid plans, and the company says uninsured admissions are growing 27% a year.¹⁰⁶ If the people who lost cover stay uninsured, and if payers resist price increases on a weaker patient mix, EBITDA could be $2B to $3B lower than I assume. Each $1B a year of lasting EBITDA is worth about $46 a share in my base, 9% of the price C (calculated).

Second, the Medicaid supplemental money shrinks faster than I model. It was $6.2B of 2025 revenue and the net gain, after provider taxes, is not published.² If the 2028 step-down takes $2.5B a year of pre-tax income, my base falls to $304; at $4B a year, to $248 E (estimate).

Third, debt turns a modest miss into a large one. The company owes $50.5B against about $15.9B of EBITDA, 3.2 times.⁶ Equity is what is left after the debt, so a 10% fall in enterprise value is a 15% fall in the stock C (calculated). The debt is long-dated, with an average life of 11.6 years and a rate of 5.1%, so the danger is not a maturity wall.⁶ It is that buybacks and capital spending are sized for a world in which EBITDA keeps rising.

The price is a bet against all three. That is the thing to keep in mind when reading the numbers.

What an owner can really take out

Reported earnings are not what an owner can take out of a hospital company. The business must keep spending on buildings and equipment just to stay as good as it is, it pays its joint-venture partners, and it pays staff partly in stock. I start from EBITDA and take off what leaves the company in cash.

So an owner’s cash is about $4.8B a year, or $22 a share, against reported earnings of about $30 a share E (estimate). That is a gap of 27%. On the $510.70 price, owner cash is a yield of 4.4%, in a range of 3.9% to 6.1%, against an earnings yield of 6.1% and a Treasury yield of 4.90%.¹³ Operating leases cost about $0.5B a year, which is already in EBITDA, so I do not add the $2.2B of lease liabilities to debt.⁴ Doing both would count the same cost twice.

The gap between owner cash and what the company pays out is the debt story. In the first nine months of 2026 HCA paid $6.6B for buybacks and $0.5B in dividends, $7.1B in all, against $3.2B of owner cash, and debt rose by $4.0B C (calculated).⁶ Buybacks at that pace work for owners only if the shares are bought below their worth.

What the price implies

The price to earnings ratio at the end of each October ran between 11.3 and 16.3 times. It is 17.0 times now, above any October reading in ten years. The stock rose 10.3% on 4 November, and the value I am comparing to is the close after that rise. At the 3 November close of $462.86 the ratio was 15.4. What is known: Democrats won both chambers of Congress in the 3 November midterms, and Reuters reported that hospital and insurer shares rose the next morning, with the three hospital operators up 6% to 8% early. Analysts it quoted said the result could delay the Medicaid enrollment cuts and state payment limits in the 2025 budget law and raise the odds of extending the ACA subsidies; J.P. Morgan said major change stays unlikely because the president can veto bills.¹⁴ What is not known: how much of the 10% is the election and how much is something else, and whether any of it lasts. The new Congress does not sit until January 2027, no bill has passed, and no one has shown a number for what a delay is worth to HCA. I take the day as a sign that the market links these shares to Medicaid and subsidy rules, not as new information about HCA's own earnings. I have not seen a sourced explanation of why the three moved the same amount. ¹⁷

Reverse DCF: what the price needs

A reverse DCF starts from the price and asks what it assumes. I keep my discount rate of 8.5% and my base case for everything except one input at a time.

Put plainly, the price needs HCA to repeat the last nine years. Revenue has to grow about 7% a year, which is 6.9% history, for ten years from a base that is already $79B, or the margin has to sit at 23% every year. At the 2025 margin of 20.6% growth must be 6.3%. My base has 4.4% growth, a rate that fits volume of 2.5% to 3% and price of 1.5% to 2%, and a margin that gives back part of the 2025 gain. The market is pricing something closer to the past decade, and it is doing so after a year in which the subsidies expired and the Medicaid rules changed.

The discount rate

The ten-year Treasury yielded 4.90% on 4 November.¹³ Damodaran’s implied equity risk premium for the market was 4.41% on 1 November.¹⁵ HCA’s beta against the S&P 500, measured on monthly returns to October, is 1.10 over five years and 0.20 over three C (calculated).¹⁶¹ The first gives a cost of equity of 9.8%, the second 5.8% C (calculated). They disagree because the stock has moved on company news and on policy, not with the market. I use 8.5%, 3.6 points above the Treasury, the same rate I use for UnitedHealth, because both are large companies whose profits depend on federal rules.¹⁷ I apply it to cash flows before debt costs, so it also stands in for the cost of the debt. The table shows how much the answer depends on it.

Three cases

The bear case, $154, has revenue growing 2% to 2.5% a year, the exchange loss spreading and the Medicaid step-down costing $2.0B a year by 2032. EBITDA ends at $16.4B, a little above today, on revenue of $100B. The equity is $33B because the $49.6B of debt is 60% of the enterprise value in that world E (estimate). It is the kind of number leverage produces; the stock fell to $361.32 on 8 June 2026, and the bear is lower than that.

The base case, $352, has volume of about 2.5% and price of about 2% a year, an EBITDA margin that dips from 20.0% to 19.4% in 2027 and recovers to 19.7% before the step-down, and a Medicaid loss that builds to $1.2B a year. Revenue is $122B in 2036 and EBITDA $22.8B E (estimate).

The bull case, $569, has revenue growth of 4.5% to 6%, a margin that climbs to 21%, a level reached only in 2021, and little loss from the Medicaid rules. Revenue is $133B and EBITDA $27.5B in 2036 E (estimate).

What I can’t judge

I cannot judge four things well. The first is the net size of the Medicaid supplemental money. The gross figure is $6.2B; the provider taxes and local payments that fund it are not published as a single number, and the step-down is set by rules still being written.²⁶ The second is how many of the people who lost exchange coverage come back. That depends on Congress and on the states. The third is the wage cycle: salaries per employee rose 2.7% in the quarter, and a tight nurse market could push that up quickly.⁶ The fourth is collection. Revenue is net of estimates of what uninsured patients will pay, and uninsured volume is up 27%.²⁶ Instead of a single number for each, here is what moves the base value.

The table has an uncomfortable message for me. No single change brings the base to the price. Two or three together do. A margin of 20.6% and growth of 6.3% would do it, which is a repeat of 2025 and of the decade. Single changes run from $248 to $490, and the price is above all of them.

This is the other way to read the price. If the market still pays 17 times earnings in 2036, the stock returns about 9.2% a year on my base earnings. If it pays 12 times, which it paid in four of the last ten Octobers, the return is 6.6%. The gap between this table and the 7.4% of my DCF is the multiple. My DCF values the business at about 7.7 times EBITDA in 2036, against 10.0 times today C (calculated). I think a lower multiple is right if growth is 4% and leverage is 3 times, but it is a judgement, and I can be wrong on it.

Multiply it out

It helps to multiply the case out. My base needs five things to go right.

• Revenue grows about 4.4% a year, with the exchange loss absorbed by volume: 65%
• The Medicaid step-down from 2028 costs no more than about $1.2B a year: 65%
• Margin holds near 19.5% to 19.7% without a labour shock: 70%
• Debt stays near 3.1 times EBITDA and rates stay near 5%: 80%
• No new federal or state cut beyond those already law: 70%

Those are my guesses, and they multiply to 17% E (estimate). They overlap, so in practice it is a little higher. The price needs more than my base on the first and third, growth near 6.3% and a margin back to 20.6%, say 40% and 45%, with the other three as above. That multiplies to about 7% before overlap E (estimate). A stock priced for a case that is less likely than my base is, for me, the main reason to place the price above the base and not below it.

Why the market could be right

My base is below where HCA is guided to finish 2026. The upper half of the guide is a margin of 20.0%; my base starts 2027 at 19.4% and ends near 19.7%. I take that off because uninsured admissions are growing, wages move with inflation, and the Medicaid supplemental net is guided to shrink. If those fade, each point of margin on 2027 revenue is about $0.8B of EBITDA, or about $37 a share E (estimate). My Medicaid step-down, up to $1.2B a year, is also a judgement; if the new Congress delays the 2025 law, that line is too harsh. Add the 8% fall in the share count and the case where the price is fair is not odd. It is the 2025 margin and the decade growth rate holding together. I think that is less likely than not, but I cannot judge the Medicaid rules, which is where most of the gap sits.

Quality and price are two questions

HCA is a very good business. It has the largest set of hospitals in the country, a long record of operating them well, a family owner with a long view, and a ten-year record that includes 6.9% revenue growth and 16.3% growth in earnings per share C (calculated).⁴ That is the quality question, and the answer is high. The price is a different question. At $510.70, 17.0 times this year’s earnings, 10.0 times EBITDA and 4.4% owner cash, the price assumes the quality survives the exchange losses and the Medicaid rules at a level above my base. It is a high-quality business at a price that asks a lot of it, and I say so plainly, without calling the stock a good or poor choice.

The margin of safety

My margin-of-safety price is $229, 35% below my base value. I use 35% because my range is wide: the bear is 56% below the base, leverage amplifies every error, and three of the inputs, the Medicaid net, the exchange recovery and the wage cycle, cannot be known. At $229 my bear case returns 7.5% a year, below the 8.5% I require but above the Treasury, and the base returns 9.9% C (calculated). The price is 123% above that. The stock last closed below $229 in November 2023.¹ HCA has not been cheap on this measure for three years, and the price I would want for a margin of safety is a long way from where it trades.

It helps to compare with the next best use of the same dollar. A ten-year Treasury pays 4.90% with no risk to the dollars promised.¹³ HCA returns about 7.4% a year on my base, 2.5 points more, with every risk of a levered share. On this site, CVS Health is 20% below its base value of $111, and UnitedHealth is 7% above its base value of $376, so the nearest health-care comparables sit closer to my value than HCA does.¹⁸¹⁷ McKesson is 9% above its base.¹⁹ HCA is 45% above its own. These are different businesses, with different discount rates, so I use them only as a gauge of what I am being asked to pay for the uncertainty.

VIIThe case for the price

What most people miss

Three things cut in HCA’s favour. First, volume did not fall. Same-facility equivalent admissions rose 3.0% in the third quarter and emergency room visits 5.1%, in the quarter when uninsured admissions rose 27.4%.⁵⁶ Second, unit costs are falling: salaries and benefits per equivalent admission were down 0.2% and supply cost per admission down 1.4%.⁶ Third, lenders still lend at about 5%: in April HCA sold $3.0B of notes at 4.7% to 5.3%.⁶

The bull case

The exchange loss is a one-time reset, volume is growing, cost per case is falling, and Medicaid programs are still being approved. The share count keeps shrinking, 8% fewer diluted shares in the third quarter than a year earlier, so earnings per share outgrow profit. The company made this argument on 27 October. It said growth in equivalent admissions and in insured revenue per admission had accelerated, cost results had improved and the Medicaid supplemental net benefit had modestly increased, despite a payer mix shift of about $400M. Its chief executive said HCA has strong staying power and expects to end the year positively.⁵

Few named investors have written this case. Viking Global Investors held 999,180 shares at 31 December 2025, 1,075,133 at 31 March 2026 and none at 30 June 2026; Markel Group held 84,972 at 30 September 2026. Those are filings, not stated views.²⁰²¹ The Frist group owns about 32%; in February 2026 it exchanged shares with the company for estate and charitable planning.⁹

Against reality

If volume grows 2.5% to 3%, price 2%, the margin holds near 19.5%, and the Medicaid step-down is no more than $1.2B a year, my base of $352 is possible; with a 21% margin and growth of 4.5% to 6%, my bull of $569 is. The price of $510.70 needs more than the base: revenue growth of 6.3% a year at a 20.6% margin, or a margin of 23% at my growth E (estimate). Today’s numbers are short of both. Revenue grew 3.6% in the third quarter, with a margin of 19.7%.⁵ Nine months are 5.5% and 19.8%, helped by the Florida catch-up.⁶³ So the price needs faster growth and a margin 0.9 points higher, as the Medicaid rules tighten.

Multiplied out as in Section VI, the base needs five things at about 17% together, and the price needs more, about 7% E (estimate). It needs HCA to repeat the last decade from a larger base.

What a 10x would take. Ten times $510.70 is $5,107 a share, or $1.09T of market value on today’s shares. At 20 times earnings that is $55B of net income, eight times the 2026 estimate. At a net margin of 10%, above any year in my table except 2021, that is $545B of revenue, 6.9 times 2026. To get there by 2036 revenue must grow 21% a year, against 6.9% in the last nine and about 4% in my base. HCA’s own share price rose 6.8 times in ten years, from $75.33 in November 2016. I know of no hospital operator that has made a tenfold move from this size. A tenfold gain from here is close to impossible, and I would not build a case on it.¹

Where I come down

The cost record moved me: a $1B-plus payer-mix hit, no loss of volume, lower unit costs and guidance kept. The price did not: $510.70 is 45% above my base, after a one-day jump of 10.3% that followed the midterm result, a link the reporting suggests but does not prove. What would raise my values most is proof that the exchange loss is a one-time reset, with uninsured admissions flattening and the Medicaid net holding past 2027; each $1B a year of lasting EBITDA is worth about $46 a share E (estimate).

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• Federal and state policy. The company lists the end of the enhanced exchange credits, the 2025 budget law, changes to Medicaid financing, and site-neutral and inpatient-only rules in its 10-K as risks it cannot control.² Any of these can be changed again, in either direction, by Congress or CMS.
• Wages and staffing. Salaries and benefits are 43.5% of revenue, and a 1% rise in labour cost with no matching price is about $0.33B of EBITDA C (calculated). The 10-K flags nurse-staffing ratios in the states where HCA operates, union activity and limits on student loans and immigration.²
• Concentration. Texas and Florida produced 51% of revenue, and both are exposed to hurricanes. Hurricanes Helene and Milton cost about $250M in 2024.⁶²
• Interest rates. $5.7B of commercial paper reprices in weeks, and $8.1B is due within a year, so a one-point rise on the $8.1B alone would cost about $80M a year C (calculated).⁶
• Technology and cyber. HCA is rolling out a new electronic health record across its facilities, and the 10-K lists the project and cybersecurity incidents as risks. A failed rollout or breach would be costly and public.²
• Investigations and claims. The company is subject to government investigations, qui tam suits and professional-liability claims. It expects to pay $568M in net claims over the next 12 months and has $1.9B in self-insured reserves.²⁶
• Leadership. The chief executive is 65 and, under his award terms, eligible for retirement, and the chief clinical officer left in 2026.⁹¹¹ I found no named successor in the proxy.
• Supply of stock. The founding family owns 32%, and the February 2026 exchange sets limits on its sales that do not stop them.⁹

What would raise my base value: an outlook for 2027 with a margin of 20% or more, uninsured growth flattening, and the Medicaid net holding past 2028. What would lower it: a second year of rising uncompensated care, a final provider-tax rule that cuts payments, or a wage shock with debt still at 3.2 times EBITDA.⁶⁵