The Real Economics of Running a Hospital in India
A hospital is a capital project that happens to treat patients. Three numbers, capex per bed, revenue per occupied bed and occupancy, decide whether it ever pays back.
Summary
- A hospital's economics are set by three numbers: what each bed cost to build, what an occupied bed earns per day (ARPOB), and how many beds are full.
- ICRA's sample of 11 listed chains grew revenue 18% in FY2026 at 63.5% occupancy and a 24.1% operating margin, while many standalone hospitals are squeezed by overbuilt capex and slow government payments.
- The promoter's real decisions are made before opening: the capex envelope, the specialty mix that sets ARPOB, and how much scheme revenue the balance sheet can afford to wait for.
Ask a hospital promoter what the business runs on and you will usually hear about doctors, equipment and reputation. All three matter. None of them shows up on the P&L until it has passed through three numbers that are far less glamorous.
The first is capex per bed, which is fixed the day you sign the construction contract. The second is ARPOB, average revenue per occupied bed per day, which your specialty mix and payer mix decide. The third is occupancy. Multiply the last two and you have revenue. Set that against the first and you have your return on capital.
The Indian market shows both ends of the result. ICRA's sample of 11 listed hospital companies grew revenue 18% in FY2026, on occupancy of 63.5% and a 9.2% rise in ARPOB, and held operating margins at 24.1%. In the same period, standalone hospitals in smaller cities have been suspending government-scheme admissions because they cannot carry the receivables. Same sector, same demand. Different arithmetic.
Capex per bed: the decision you live with for twenty years
Industry estimates put the build cost of a bed in a listed Indian hospital anywhere between about ₹30 lakh and ₹1.5 crore. The spread is not waste or thrift. It follows the clinical positioning: a -care hospital in a Tier-2 town sits at the bottom, a quaternary centre with transplant programmes and high-end imaging at the top.
Consultants who cost these projects put a 50-bed, NABH-ready secondary hospital in Delhi at roughly ₹40 crore to ₹95 crore all in, with land alone taking ₹30 crore to ₹40 crore of that. Excluding land, a bed runs from about ₹70 lakh at Tier-2 specification to ₹1.2 crore with CT and MRI in the building. Construction of a 50-bed facility takes two to three years, and accreditation usually follows a year or more later.
Two line items are routinely missed in feasibility reports. The first is licensing, compliance and hospital IT, which add something like a tenth to the project cost when they are bolted on late rather than planned. The second is the ramp: a new hospital spends its first year at 40% to 50% occupancy, and the working capital to survive that year is part of the project, not an afterthought.
The costliest mistake in the sector is a building specified for premium care and priced for the mass market.
It happens more than promoters admit. The hospital is designed to impress, the catchment will pay a secondary-care tariff, and the gap between the two is carried as depreciation and interest for the next decade. Feasibility models commonly show break-even inside three to five years only once occupancy crosses about 65%. At 45% the same model stretches to seven or eight years.
ARPOB: what an occupied bed earns
Once a hospital is open, ARPOB is the single most informative number about its business. It captures the acuity of the cases you treat, the prices you can hold, and how much of your revenue comes from cash and insured patients rather than scheme packages.
Revenue per occupied bed per day at listed Indian hospital chains
| Chain | ARPOB per day, FY25 | Positioning |
|---|---|---|
| Max Healthcare | about ₹73,900 | Metro tertiary and quaternary |
| Fortis Healthcare | about ₹66,300 | Metro multi-specialty |
| Medanta (Global Health) | about ₹62,000 | Quaternary, strong international mix |
| Apollo Hospitals | about ₹60,600 | Pan-India scale |
| Narayana Hrudayalaya (India) | about ₹46,300 | Volume and cost discipline |
| KIMS Hospitals | about ₹39,200 | Tier-2 South India |
The spread is almost two to one, and the lesson from it is not that higher is better. Narayana Hrudayalaya earns well under two-thirds of Max's ARPOB per bed and still reports an margin of about 23%, in the same band as chains charging far more. It gets there through throughput and cost control: standardised protocols, high surgical volumes per theatre, procurement at scale, and a relentless focus on what each procedure costs to deliver.
For a promoter outside the metros, that is the more useful model. Patients in a Tier-2 catchment are price-sensitive, the volume is there, and a strategy built on premium pricing tends to hit a ceiling that a strategy built on cost per case does not. Where you sit on the ARPOB range is a set of choices: specialties, geography, payer mix and clinical positioning. Make them on purpose. Our piece on hospital pricing and payer mix works through the tariff side of that choice.
Occupancy and the fixed-cost base
A hospital's costs are mostly fixed. Nurses are rostered whether the beds are full or not, the building depreciates at the same rate, and the specialists you hired to build a department expect to be paid in a slow month. That is why margins move so sharply with occupancy. Below a threshold every empty bed is a loss; above it, most of each additional bed-day's revenue reaches EBITDA.
ICRA's listed sample ran at 63.5% in FY2026, with the best-run metro chains well above that: Max reported about 74% for FY25. For a typical standalone hospital, the difference between 55% and 70% occupancy is the difference between covering costs and making real money.
Below EBITDA the capex decision comes back. Depreciation and interest sit between operating profit and net profit, and for a heavily borrowed single hospital they can consume most of what operations produce.
Every crore of capex that did not earn its place is still being paid for in year eight.
The scheme question: revenue you have to wait for
Ayushman Bharat PM-JAY covers up to ₹5 lakh per family per year and has brought hospital care within reach of tens of crores of people. For the hospitals that treat them, the difficulty is timing and rate, not intent.
ThePrint, citing Health Ministry figures, reported that pending PM-JAY bills had crossed ₹1.21 lakh crore by February 2025, with more than 63 lakh claims open. Read that number with care: it is large against the scheme's cumulative authorisations, so it likely includes claims still in normal processing rather than only overdue payments. The ground-level signal is clearer. In August 2025 more than 600 empanelled private hospitals in Haryana suspended scheme admissions over roughly ₹500 crore of unpaid bills, and similar disputes have surfaced in other states.
A chain can absorb that. A 60-bed hospital in a district town, where scheme patients may be the majority of admissions, often cannot. Its choice is between carrying months of receivables at a package rate that barely covers cost, or leaving the scheme and losing much of its patient base. The fix is financial discipline on the revenue cycle, which our revenue cycle management monograph covers in detail.
What investors are pricing in
Capital has been generous to hospitals that have crossed a maturity threshold. Listed chains trade at multiples well above most Indian sectors, and ICRA expects its sample to grow revenue 13% to 15% in FY2027 as new beds ramp up. The thesis is straightforward: India has about 1.3 hospital beds per 1,000 people against a commonly cited benchmark of 3, the population is ageing, and insurance coverage is widening.
The premium is paid for what sits behind the numbers: scale, a deliberate specialty mix, a diversified payer base and management that does not depend on one person. Hospitals without those traits do not get the multiple, and the next two pieces in this series look at why: why most hospitals never scale past one site and why expansion now favours brownfield beds.
What to do if you run or are building a hospital
- Fix the capex envelope against the tariff you can charge, not the building you want. Work backwards from realistic ARPOB and a 65% occupancy year, and cap cost per bed at what that revenue can service.
- Fund the ramp explicitly. Put twelve to eighteen months of below-break-even operation into the project cost and the debt structure, so the first slow year is planned rather than survived.
- Track ARPOB and occupancy by specialty every month. A blended number hides departments that lose money on every bed-day.
- Cap scheme exposure at what your working capital can carry. Know your receivable days by payer and set a ceiling on the share of beds given to slow payers.
- For CFOs: model depreciation and interest per occupied bed. It is the clearest way to show a board what an extra floor or an extra scanner will cost the P&L for a decade.
None of these numbers is hard to know in advance. The hospitals that struggle are rarely surprised by demand. They are surprised by their own cost base. If you want a second view on a project model or a monthly FP&A pack for an operating hospital, that is the work SRF does in hospitals and diagnostics.
How useful was this article?
One tap. It tells us what to write more of.
About the author
SRF Capital Studio
The next one
Get what we publish next, by email.
Working notes on raising, borrowing, protecting, growing and structuring capital in India. One email a week at most, and you can leave any time.
We use your address only to send this. See our privacy policy.
We store your address to send you these emails and nothing else. See our privacy policy.
Related reading
