Thailand cannabis: the 49% cap isn't the real ceiling

Most foreign investors read the ownership cap, conclude Thailand is closed to them, and move on. That reaction is the single most common strategic error we see in this market, and it is based on reading only half the rulebook.

The half everyone reads is real. Foreign equity in a licensed cannabis operating company is capped at 49%. Extract production, the highest-margin layer of the value chain, is barred to foreign participation outright, full stop, no structuring around it. Read no further than that and Thailand looks like a market you watch from outside. That conclusion is wrong, and it is wrong for a specific, structural reason: ownership and participation are not the same variable, and Thai law only constrains one of them.

What is actually capped, and what is not

The 49% figure applies to the licensed core - the entity that holds the cultivation, processing, or dispensing license itself. A foreign investor cannot hold a majority equity stake in that licensed entity, and cannot use a nominee arrangement to get around it. That much is settled and enforced.

What is not capped is everything a foreign brand or capital provider actually wants to do in this market: supply inputs, buy output, license intellectual property, provide financing, or build a genuine joint venture where a Thai partner holds the majority and the foreign partner holds the economics that matter to them. None of that requires foreign ownership of the license. All of it is legal today, and almost none of it is being used at scale.

The four routes that do not touch the cap

Structuring around a legal restriction is a bad idea. Structuring within one, using instruments the law already permits, is simply deal-making. Four routes carry the weight for most credible entrants:

• Offtake agreements. Commit to purchasing a Thai-majority licensee's output on agreed terms. The foreign party never touches the license and never needs to.
• Toll-manufacturing. Pay a licensed Thai producer to process or manufacture to a foreign brand's specification. The IP, the formulation, and the brand stay with the foreign party; the license stays Thai.
• IP and brand licensing. License a brand, formulation, or process to a Thai-majority operator in exchange for royalties. This is the standard playbook in dozens of other regulated consumer categories, and it transfers cleanly here.
• Finance. Debt, convertible instruments, and structured finance into Thai-majority operators, priced for the risk, without requiring equity control of the license itself.

Layer two or three of these into a single relationship with one Thai-majority JV partner, and a foreign entrant can hold most of the commercial value of a position while a Thai company holds 51% or more of the license, exactly as the law requires. The cap constrains the org chart. It does not constrain the deal.

Why the nominee shortcut is not a gray area

The obvious shortcut - park a Thai national's name on the license while a foreign party retains actual control through a side agreement - is not a compliance risk to be managed. It is a Section 36 crime, and it is now prosecuted. This is worth stating plainly because the nominee model was tolerated, in practice, in adjacent Thai sectors for years. That tolerance does not extend here. Any structure whose sole function is to disguise foreign control of a licensed entity is the wrong side of a criminal statute, not a gray zone.

The distinction that matters legally is between control and economics. A genuine Thai-majority JV, where the Thai partner has real decision rights and real capital at risk, paired with an offtake or licensing agreement that gives the foreign party its return, is a normal commercial structure. A Thai name with no real authority, installed to satisfy a registrar, is not. The line between them is not always comfortable to draw, but the law draws it, and enforcement now follows the law.

What this means for sizing an entry

None of this changes what the market is actually worth, and we are not going to pretend otherwise here. The national retail figure, the channel splits, and the segment-level revenue bands that would tell an investor how large a JV or offtake position could realistically become are Likely reconstructed with full confidence tagging in the report. see the segment breakdown →

What we can say without redaction is the shape of the opportunity: the market has consolidated hard since its 2024 peak, oversupply and price compression are the live risk rather than recriminalization, and no Thai authority publishes cultivation area, production tonnage, or lab-testing throughput - the "triple absence" that makes structuring capital into this market materially harder than structuring the legal entity. Getting the corporate structure right solves a legal problem. It does not solve the measurement problem, and the two are equally capable of sinking a deal.

How we know this

This view is built from direct exposure to the licensing process itself, not from summarizing statutes at a distance: 850+ dispensaries visited, 100+ owner and operator interviews across ten-plus nationalities, 30+ farm operators, and the full legal licensing process completed first-hand, end to end, by the team. Structuring guidance that has not been tested against an actual license application tends to miss the parts of the process that are unwritten but decisive.