It arrived, and it was not enough

The question is usually asked as an accusation. Where did the state budget go, and why does a famous person have to collect money for a hospital?

The first half has an answer, and it is not the one the question expects.

The money did not vanish

For the 2026 financial year the universal coverage fund is ฿265,295.58 million, of which ฿198,227.75 million is per-head capitation for 47.502 million eligible people — ฿4,173.04 each. The figures are in the National Health Security Board’s announcement on the fund’s management for the 2026 financial year, published in the Royal Gazette and in force. The year before, the fund was ฿235,842.80 million with capitation of ฿181,297.44 million for 47.15 million people, or ฿3,844.55 each. Read the budget breakdown.

Now the other side of the same system, and this is where the article originally went wrong.

In October 2025 the Permanent Secretary of the Ministry of Public Health set out the financial position of the ministry’s hospitals — 908 of them: 116 regional and general hospitals under the Office of the Permanent Secretary, 49 under the Departments of Medical Services, Disease Control and Mental Health, and 743 community hospitals. In the 2024 financial year those hospitals took ฿179,885 million in universal-coverage revenue and spent ฿198,335 million treating universal-coverage patients. Expenditure was 110% of revenue and the scheme’s EBITDA was negative ฿18,479 million.

But the hospitals themselves did not lose money, and the Ministry said so explicitly. Over the same year the civil servants’ medical benefit scheme returned a positive ฿11,510 million and social security a positive ฿2,130 million, and every Ministry of Public Health hospital taken together made an EBITDA profit of ฿5,327 million. The ฿18 billion is a hole in one scheme, not a hole in the hospitals.

Which is a more interesting fact than the one this article first reported, and a more uncomfortable one.

The universal scheme is not merely underfunded. It is being cross-subsidised by the categorical ones — by the scheme for civil servants and the scheme for formal-sector employees, both of which pay hospitals more than the care costs. Every universal-coverage patient is, in accounting terms, partly paid for by a civil servant’s entitlement.

The reason is a published rate that has not moved. Between 2017 and 2023 the true cost of a general inpatient episode ran from ฿10,558 to ฿15,919 per adjusted relative weight, while the NHSO reimbursed between ฿7,000 and ฿8,350. The share of cost actually reimbursed fell from 66% in 2017 to 52% in 2023. For three of those years the cost stood at ฿15,919.79 and the rate paid stayed at ฿8,350. The Ministry is asking for ฿10,000, and later ฿13,000.

So “งบรัฐไปไหนหมด” has a boring and unwelcome answer: it went where it was supposed to go, and the rate it went at was set below what the care costs.

That distinction matters enormously, because the two diagnoses have opposite remedies. If money is being stolen, the answer is investigation and prosecution. If a universal entitlement is reimbursed at half the cost of delivering it, the answer is arithmetic — and no amount of anti-corruption work fixes it.

Which is this site’s finding with figures attached

Article 89 The wrong way round established that Thai protections work when universal and fail when categorical, because a category is always drawn around people who already had options.

Article 92 The first toll added the limit, and this is the clearest illustration of it anywhere in the archive.

Universal is not sufficient. It has to be universal and adequate. A universal entitlement funded below the cost of the thing it entitles you to fails in the same way a categorical one does — it simply fails at the point of delivery rather than at the point of eligibility.

Nobody in Thailand is excluded from universal healthcare. 908 hospitals are delivering it at 52% of what it costs them.

And the cross-subsidy sharpens article 89 The wrong way round rather than softening it. The finding there was that categorical protections are drawn around people who already had options. Here the categorical schemes are not merely better funded — they are what keeps the universal one solvent at the point of delivery. The civil servant’s entitlement is paying, in part, for the universal patient’s treatment. That is not a scandal and nobody designed it as a transfer. It is what a below-cost rate produces in a hospital that treats everybody.

And the gap between what an entitlement promises and what it is funded to deliver has to be closed by something. In Thailand, one of the things closing it is donation.

Now the harder question

Why a famous person, and not a foundation, or the state?

Start with what a donor can actually verify, because that turns out to be the whole of it.

Almost nothing.

You give ฿500. You cannot follow it. You cannot see the equipment it contributed to, or confirm it was bought, or establish that it was needed, or check what it cost, or find out whether the same item was donated three times over. Nothing about a donation is observable to the person making it.

This is article 80’s test, and donation fails both halves. There is no product standard — nobody can tell you what ฿500 buys. There is no published price — there is no benchmark against which any appeal’s efficiency can be judged.

So a donor cannot verify. What a donor does instead is substitute a proxy for verification.

And the proxy is a person, not a document

Here is the part that ought to be uncomfortable, because the obvious remedy already exists and is not what people choose.

A registered มูลนิธิ in Thailand is not an informal thing. Section 110 of the Civil and Commercial Code constitutes it as property specifically appropriated to a purpose — public charity, religion, art, science, literature, education or other public benefit — and not for sharing out gains. Section 128 gives the registrar power to inspect and supervise, to order directors, staff or agents to explain the foundation’s affairs, to require the books of account and documents to be produced, and to enter the office between sunrise and sunset.

The annual reporting duty is not in the Code itself. It sits in the ministerial regulation on foundation registration, operation and records, B.E. 2545 (2002), made under the Code: the foundation reports to the registrar each year with its accounts and the auditor’s report.

That is a real verification mechanism. It exists, it is compulsory, and it is more than most countries require of small charities.

And appeals fronted by a recognisable face routinely raise money faster than foundations that have been filing audited accounts for decades.

Which looks irrational until you ask the right question: verified by whom?

An audited financial statement is verification for a regulator. It is not verification for a donor, because a donor cannot read it, would not know what a reasonable ratio looked like if they could, and has no way to tell a well-run foundation from a badly-run one from its balance sheet.

So from where the donor is standing, both options are unverifiable. And given two unverifiable options, the donor picks the one where somebody identifiable would be blamed if it went wrong.

That is not stupidity. It is a rational response to an information problem.

Reputation is the collateral

Article 144 The channel is the collateral of this collection found an influencer’s audience functioning as undisclosed collateral for a product claim — the channel is what is at risk if the claim fails.

The charitable version is the same instrument, used openly.

A person with public standing who fronts an appeal is staking something the donor can observe. Not accounts — reputation. If the money goes astray, the donor knows exactly whose name is attached, exactly where to find them, and exactly what they stand to lose.

And this archive has now met the underlying principle repeatedly, always as a burden: obligation attaches to whoever can be found. Article 170’s forged-licence buyer, carrying the offence because he has an address. Article 143’s venue. Article 146’s traveller.

This is the same principle operating as an asset. The celebrity is trusted because they can be found. Findability, which everywhere else in this archive is a liability, is here the entire value proposition.

Which is why it works, and why it will keep working. A foundation is a name on a registration document. A person is somebody you could shout at.

What this means, said plainly

Three things, and none of them is a complaint about anybody.

The donation appeals are not evidence of theft. They are evidence of a gap between an entitlement and the rate at which it is reimbursed, and the figures show that gap directly.

They are not a substitute for the budget, and cannot become one. ฿198 billion of capitation is the scale of the state’s contribution this year. No amount of fundraising is going to be relevant at that scale, which means donation closes specific, visible, local gaps — a machine, a ward, a bill — and cannot close the structural one.

And the reason donors choose a face over accounts is a verification problem, not a character flaw. If audited accounts were legible to ordinary donors, they would be a better proxy than fame, and the fact that they are not is a design failure in charitable disclosure rather than a failure of public judgement.

Which suggests the useful reform, and it is not more auditing. It is disclosure a donor can read — what was raised, what was bought, what it cost, published in the form a person actually looks at. The audit already exists. What is missing is the version for the audience.

---

Common misconceptions

“The budget was stolen, that’s why hospitals need donations.” The universal coverage fund is ฿265,295.58 million for the 2026 financial year, with ฿198,227.75 million of it capitation. The published critique is a reimbursement rate set below cost, not missing money.

“Ministry of Public Health hospitals are losing money.” Not overall. In the 2024 financial year they made an EBITDA profit of ฿5,327 million. What ran a deficit was the universal-coverage line, at negative ฿18,479 million — offset by the civil servant and social security schemes. The Ministry stated the distinction itself, and this article previously blurred it.

“Thailand doesn’t have universal healthcare.” It does. The question is whether it is reimbursed at the cost of delivering it. Between 2017 and 2023 the reimbursed share of inpatient cost fell from 66% to 52%.

“Foundations aren’t regulated.” Registered foundations file annual financial statements audited by a licensed auditor.

“Celebrity appeals are just vanity.” They raise money quickly for real needs, and they do it by staking something the donor can observe.

“Donations could replace the budget.” At ฿198bn of capitation, no.