The standard arrived after the money

In 2025, the licences of 8,636 Thai cannabis businesses expired.

1,339 applied to renew.

Fifteen and a half per cent.

The other 7,297 simply stopped. Not raided, not prosecuted, not refused — they declined to fill in the form. As at 28 December 2025 there were 18,433 cannabis establishments in the country, and the Ministry of Public Health’s own projection for what would remain was 11,136.

A further 4,587 licences expire in 2026 and 5,210 in 2027.

A renewal is a cheap, small, reversible decision. Six out of seven people who had already paid to be in this business, and had already built the shop, looked at the cost of staying in it and walked away. That number is the market’s own verdict and it is more honest than any survey.

The question worth asking is not why the shops are closing. It is what they were betting on when they opened.

Delisting is not a regime

In June 2022 the state removed cannabis from the narcotics schedule.

That is a subtraction. It is not the same thing as building a market. No licensing framework, no potency standard, no grading system, no retail zoning and no published price came with it. The plant stopped being a narcotic, and what followed was determined by whatever ordinary commercial law happened to apply.

And the immediate consequence is the one nobody plans for: the barrier to entry fell to almost nothing, on a single announced date, in front of everybody at once.

That last clause is the important one.

In an ordinary emerging market, entry is staggered. Some people learn of the opportunity earlier than others, capital arrives at different speeds, and the early entrants are visible to the late ones. A late entrant can count the competition before committing.

A delisting is a public event. It made the opportunity visible to the entire country simultaneously. Thousands of people made the same estimate of national demand, at the same time, on the same information — and not one of them could observe how many others were making it.

By the time you could count the shops on your own street, your money was already spent. A fit-out, stock, rent deposits and a licence are sunk costs. The competitor count — the single most important input to the decision — only became observable after the decision was irreversible.

Thai reporting carries the extreme version of this: an operator who put more than ฿10 million into a new shop, opened, and was two days into trading when the ministerial notification landed. He is not unusual in kind. He is only unusual in timing.

Why the margins were never going to hold

Now apply this site’s own test, from article 80 Money you can wear.

Article 80’s finding is that an informal market works when it has two things: a product standard and a published price. Gold has both — 96.5% purity and one national price revised dozens of times a day — and it serves the unbanked reliably across generations. The underground lottery and the motorcycle-vest market have neither, and they harm people.

Thai cannabis retail between 2022 and 2025 had neither.

No potency standard. No grading. Nothing that let a customer know what they were buying or a shop demonstrate that its product was better than the shop next door.

No published price. No benchmark, no reference, nothing against which a quoted figure could be judged.

Without a standard you cannot differentiate. Without a published price you cannot compete on value. What remains is location and discounting, and discounting in a market with free entry converges on the cost of goods.

So by this site’s existing test, the outcome was determined in June 2022. The overcrowding everybody blames is real, but overcrowding alone does not destroy a market — a crowded market with a product standard produces winners. A crowded market without one produces a race to the bottom, and everybody in it loses at approximately the same rate.

And then the standard turned up

Here is the part that extends article 80 Money you can wear rather than merely applying it.

Read what the new draft ministerial regulation actually requires. Approved by Cabinet and, as of January 2026, at the Council of State awaiting the Minister’s signature, it provides that sales may take place only in four kinds of establishment: medical facilities, pharmacies, licensed herbal product shops, and the premises of traditional practitioners.

And then the operational requirements. Effective odour and smoke extraction, so as not to be a nuisance to neighbours. Premises the applicant actually owns or lawfully possesses. A storage area of appropriate size, with cannabis kept separately from other goods, temperature and humidity controlled, kept out of sunlight, and never placed directly on the floor. At least one worker trained by the Department of Thai Traditional and Alternative Medicine on duty the entire time the premises are open.

That is a product standard. It is a good one. It is exactly what article 80 Money you can wear says this market needed.

It is arriving three years after eighteen thousand people spent their money.

And this is the correction to article 80 Money you can wear, and it is the third this run. Article 128 You can see the gearbox established that the test needs a way to handle used goods; article 157 The standard is for the next buyer established that a standard does two different jobs. This one adds a timing condition.

A standard introduced into an existing market does not merely improve it. It revalues every asset in it. A shop built to no standard, in premises the operator rents rather than owns, without a storage room or an extraction system or a trained member of staff, is not a shop that can be brought up to code. It is a shop that is now worth its fittings.

The standard that would have made this market work, arriving after the capital, is the thing that finishes the capital off.

The risk nobody could price

There is one more layer, and it is on the record from the Ministry itself.

A journalist at the January 2026 briefing asked the question that every operator in Thailand wants answered: if the government changes, does this change again?

The Ministry’s answer was that the ministerial regulation, once signed, would not change — but that government policy on cannabis would depend on what the next government looks like.

That is an honest answer and it is also the whole problem.

A business can price commercial risk. Competition, rent, demand, seasonality — these are estimable, and getting them wrong is the entrepreneur’s own fault. What cannot be priced is how long the legal basis of the product will last, because the state does not publish that and, in a parliamentary system, cannot credibly promise it.

These businesses were capitalised against a regime with no stated duration. The regime then changed twice in under four years — a delisting in 2022, and a re-regulation from 2025 that moved the flower to prescription and confined sales to four categories of premises.

The 15.5% renewal rate is what it looks like when a market prices that risk for the first time. The six-in-seven who walked away were not saying the business was unprofitable. They were saying they could no longer form a view about how long the rules would hold, and the renewal was not worth the option.

What this generalises to

The lesson is not about cannabis and it is not about Thailand.

Whenever a state removes a prohibition, it creates a market whose entry cost is near zero and whose participants cannot see each other in time. If it removes the prohibition without simultaneously publishing a standard and a licensing framework, the resulting market has no mechanism to reward quality, and the capital that enters is destroyed twice: once by competition it could not observe, and again by the standard that eventually arrives.

The sequencing is the policy. A standard published on the same day as the delisting would have raised the entry cost, reduced the number of entrants, given the survivors something to compete on, and cost the state nothing.

The Ministry is now drafting a full Cannabis and Hemp Act to cover the chain end to end. That is the right instrument. It is being written for a market that has already lost seven thousand of its participants.

---

Common misconceptions

“The shops closed because the government banned cannabis.” Cannabis was not re-banned. The flower moved to prescription and sales were confined to four types of premises. Most closures were non-renewals, not enforcement.

“There were just too many shops.” True, and insufficient. A crowded market with a product standard produces winners. A crowded market without one produces a race to the bottom.

“Operators should have seen it coming.” The competitor count was unobservable until after entry, and the regime’s duration was never published. One of those is a business misjudgement. The other is not.

“The new rules are unreasonable.” Storage, extraction, trained staff and defined premises are ordinary requirements for a controlled herbal product. The objection is to the timing, not the content.

“This settles it.” The Ministry’s own answer was that the regulation will hold but that policy depends on the next government.