No Commercial GM Rice Exists Anywhere, and That Is an Export Story

There is no commercially grown genetically modified rice anywhere in the world. Not a small area, not a pilot, not a country holding an approval it has chosen not to exercise. Zero.

That single fact is the most consequential thing about the USD 6.35 billion global rice seed market, and it is routinely explained as a technology lag — as though the trait were harder to build in Oryza sativa than in maize. It is not. Insect-resistant and herbicide-tolerant rice have existed in breeding programmes and regulatory dockets for two decades. China has held biosafety certificates for GM rice without commercial release for many years, while the same regulator moved corn and soybean to routine commercial planting.

The correct explanation is not agronomic. It is that rice is the crop where a seed decision is a trade decision, and the countries capable of commercialising it first are the countries whose export franchise would be placed at risk by doing so.

The asymmetry that proves it is policy, not science

Look at what the same regulator did with the same technology in the same period.

In 2025 China’s GM cotton and maize sat inside a programme that reached thirteen provinces and passed 30 million mu. The 2026 season is the first conducted under ordinary commercial planting rules, with maize targeted at 57.85 million mu and 239 cumulative variety approvals granted by the fourth batch. Rice, in the same country, in the same regulatory system, with events already cleared at the biosafety stage, remains unplanted.

That is not a regulator that cannot approve a GM crop. It is a state that has decided, crop by crop, which approvals to convert into acreage — and it has drawn the line at the staple its own population eats and its neighbours import.

Why the export channel is the binding constraint

Rice is the most politically managed staple in the world, and the instrument of management is the export channel. India demonstrated the elasticity of that instrument in 2023–24, when export restrictions moved the paddy price — and with it, seed demand — within a single season. A state willing to close its export window to hold a domestic price is a state that treats its export franchise as a policy asset rather than a commercial one.

That changes the calculus on a GM release in three ways.

First, approval and market access are separate permissions. Destination markets authorise events individually. An event approved in the country of production but not in the country of consumption makes the shipment non-compliant, not merely discounted — which is why asynchronous approval and export rejection is a named restraint in our GM seed coverage rather than a footnote. A rice exporter cannot absorb that risk the way a maize exporter can, because rice moves as a commodity through a smaller number of national channels and carries a proportionally larger share of the exporting country’s agricultural trade reputation.

Second, segregation is not achievable at scale. Rice moves from millions of smallholders through thousands of small mills and a layer of aggregators. A country that plants GM rice for domestic use and continues to export non-GM rice has to guarantee identity preservation across that chain, every season, to every destination. No rice economy has demonstrated that capability, and the cost of building it falls on the exporter rather than on the trait developer.

Third, the precedent for approval is already contested. The Philippines was the first country to approve Golden Rice for commercial propagation, and then went through prolonged legal challenge over the biosafety permits. The lesson is not that approval is hard. It is that approval and acceptance are different hurdles, and that a legal challenge can hold a product out of the ground after the regulator has cleared it.

What the absence is worth

Rice seed earns USD 36.9 a hectare — USD 6.35 billion across roughly 172 million hectares. GM maize earns USD 163.9 a hectare. The gap is 4.44 times.

If rice converted at maize’s global penetration of 35.8%, that would be 61.6 million hectares. Valued at maize’s per-hectare GM revenue, it is USD 10.1 billion — larger than the entire current rice seed market, added on top of it.

That is why the GM rice question should be modelled as an option rather than as a base case, and why our forecast carries no commercial GM rice through 2033. It is also why the option is not priced like a technology option at all. A technology option is exercised when the technology works. This one is exercised when a government decides that the export risk is smaller than the yield gain — a judgment that depends on the world price of rice, on the stock position of the exporting country, and on whether its competitors move first.

Those are not conditions a seed company can engineer. They are conditions it can only anticipate, which is why the first GM rice launch, if it happens inside this forecast window, is most likely to occur in a country that exports little, exports to permissive markets, or has already accepted a segregation cost somewhere else in its agricultural trade.

The route that actually works

There is a way around every one of these constraints, and it is not transgenic.

Gene editing is the only technology in view that can create enforceable rights in a crop whose germplasm is largely public, and it is the only one that can sidestep the export problem by not being a GMO in the destination market. Japan has established a notification and consultation pathway for genome-edited food and agricultural products, with nine products cleared and four in domestic production and distribution. Regulation (EU) 2026/1388, adopted on 17 June 2026 and applying from 17 July 2028, creates a separate pathway for plants obtained by certain new genomic techniques — establishing, in the world’s most restrictive market, the precedent that an edited crop can be separated from a modified one.

A lodging-resistant, direct-seeding-adapted, anaerobic-germination-competent edited rice, approved in a permissive jurisdiction and legally not a GMO in its export destinations, would be the first genuinely proprietary rice product in the market’s history and the first defensible basis for a royalty in it. That is a realistic 2030s product. GM rice is not.

The judgement

The absence of commercial GM rice is not evidence that rice is a laggard. It is evidence that rice is the one major crop where the export market disciplines the seed market, and where the constraint operates through trade policy rather than through biosafety law.

Which means the thing to watch is not a regulatory docket. It is the world price of rice, the Indian stock position, and whether any exporter concludes that its trade franchise is worth less than a 10% yield gain. Nothing in the current record suggests it does before 2033 — and the 5.30% forecast, built on conversion of farm-saved and public seed rather than on a technology step, is the honest consequence.