In 2021, China imported 28.35 million tonnes of corn, a record that more than doubled the previous year’s total and briefly made the country the world’s most important swing buyer of feed grain. Four years later, according to customs data compiled by Mysteel, the figure had collapsed to about 2.65 million tonnes, a fall of roughly 81 per cent in a single year. Some of that was weather, some was tariffs and some was a weak pig economy. But a growing part of the story is happening in the fields of Heilongjiang, Jilin and Inner Mongolia, where Chinese farmers are now planting genetically modified corn on a commercial scale for the first time.
Soybeans tell a different, more stubborn tale. China bought a record 111.82 million tonnes in 2025, according to the General Administration of Customs, and Brazil supplied nearly three quarters of it. Domestic output, despite years of subsidy and political exhortation, covers only around one sixth of demand.
Put those two numbers side by side and you have the shape of Beijing’s seed strategy, and the reason it matters to everyone from a soybean exporter in Mato Grosso to a feed miller in Ho Chi Minh City. China is not trying to stop importing. It is trying to decide how much it imports, from whom, and on whose genetics. As 2027 planting decisions are made and the 15th Five-Year Plan sets its priorities, that ambition is moving from policy documents into seed bags.
From “seed revitalisation” to commercial planting
The modern chapter began in 2021, when the leadership elevated seeds to a matter of national security and launched a campaign to revitalise the domestic seed industry. The logic was explicit: a country that feeds a fifth of humanity on less than a tenth of the world’s arable land cannot afford to rent its core genetics from foreign firms. A revised Seed Law that took effect in 2022 strengthened plant variety protection and introduced the concept of essentially derived varieties, a change designed to reward genuine breeding rather than cosmetic copies of existing hybrids.
Biotech followed. In October 2023 China approved its first batch of GM corn and soybean varieties for commercial registration, and in December that year the Ministry of Agriculture and Rural Affairs (MARA) licensed 26 companies to produce and sell GM corn and soybean seed in designated provinces. The names on that list were telling: Beijing Dabeinong Technology, subsidiaries of Yuan Longping High-Tech Agriculture, and China National Seed, now part of the Syngenta Group, which is itself owned by state-controlled Sinochem.
The pipeline then widened quickly. In March 2024, MARA added 27 GM corn varieties and three GM soybean varieties to the approved list. In April 2025, the national variety registration committee published a further preliminary list of 97 GM corn varieties and two soybean varieties for public comment, according to the US Department of Agriculture’s Foreign Agricultural Service (FAS). Corn, plainly, is where Beijing sees its quickest win.
Planting has followed the paperwork, if more cautiously. Industry sources expected around 670,000 hectares of GM corn in 2024. For 2025, USDA attaché reporting cited industry contacts estimating about 3 million hectares of GM corn and soybeans, more than four times the 2024 area. Even so, that represents only around 7 per cent of China’s corn acreage on estimates reported by Reuters, compared with more than 90 per cent in the United States and Brazil.
Gene editing moves to the centre of the plan
If GM corn is the near-term lever, gene editing is the long-term bet. China issued its first agricultural gene-editing biosafety certificate in April 2023, to Shandong Shunfeng Biotechnology for a high-oleic soybean. According to Caixin, MARA issued ten gene-editing biosafety certificates across four batches in 2023 and 2024: five for soybean, two for corn, two for wheat and one for rice. A further round of approvals at the very end of 2024 covered both gene-edited and GM events, with Dabeinong and China National Seed among the beneficiaries, and USDA reporting noted that the first gene-edited rice event was cleared for local cultivation in December 2024.
The start-up layer is real, too. Beijing Qi Biodesign Technology, founded only in 2021, holds five certificates and has raised more than 500 million yuan, Caixin reported. Its high-oleic soybean became the first Chinese gene-edited product to receive a US regulatory exemption in 2024, a small but symbolic reversal of the usual direction of agricultural technology flows.
Policy has now caught up with the laboratories. The State Council’s 15th Five-Year Plan for accelerating agricultural and rural modernisation, released in June 2026, designates gene editing as a critical core agricultural technology for the first time, according to Caixin. The broader plan outline approved by the National People’s Congress pledges to keep promoting the industrialisation of biological breeding. The same planning cycle sets a binding target of around 725 million tonnes of grain production capacity by 2030, up from the 714.9 million tonnes MARA reported for 2025, and aims to lift science and technology’s contribution to agricultural growth from just over 64 per cent to 67 per cent.
Regulation matters here. Under China’s framework, gene-edited plants without foreign DNA face a lighter safety evaluation than transgenic crops, which shortens the road from laboratory to variety approval. That is precisely the pathway Beijing wants its seed champions to exploit.
Table: China’s biotech seed milestones, 2021 to 2026
|
Date |
Milestone |
Scale |
|
2021 |
Seed industry revitalisation elevated to national security priority |
Policy campaign |
|
2022 |
Revised Seed Law takes effect, adding essentially derived variety protection |
Legal framework |
|
April 2023 |
First agricultural gene-editing biosafety certificate (Shunfeng, high-oleic soybean) |
1 certificate |
|
October 2023 |
First GM corn and soybean varieties approved for registration |
First batch |
|
December 2023 |
MARA licenses firms to produce and sell GM corn and soybean seed |
26 companies |
|
2023 to 2024 |
Gene-editing biosafety certificates issued in four batches |
10 certificates |
|
March 2024 |
Additional GM varieties approved |
27 corn, 3 soybean |
|
2024 |
GM corn planted commercially |
About 670,000 ha (industry expectation) |
|
December 2024 |
First gene-edited rice cleared for cultivation; new GM and gene-edited certificates |
Multiple events |
|
April 2025 |
Preliminary GM variety registration list published |
97 corn, 2 soybean |
|
2025 |
GM corn and soybean planted |
About 3 million ha (USDA industry estimate) |
|
June 2026 |
15th Five-Year Plan names gene editing a critical core technology |
National plan |
The import dependence equation
So how far can this go? The honest answer differs sharply by crop.
For corn, the arithmetic is favourable to Beijing. Analysts and one seed official cited by Reuters suggested GM traits for pest resistance and herbicide tolerance could lift yields by 6 to 13 per cent. Applied across a corn crop of close to 300 million tonnes, gains of that order would amount to roughly 18 to 39 million tonnes at full adoption, at the upper end more than China imported even in its record year. Adoption at today’s 7 per cent share delivers a fraction of that, which is why planting area is the number to watch. China also controls the tap directly: its corn tariff-rate quota is 7.2 million tonnes a year at a 1 per cent duty, with out-of-quota imports facing 65 per cent.
For soybeans, the gap is structural. Domestic output reached 20.91 million tonnes in 2025, according to figures reported by the Western Producer, against imports of 108 million tonnes on that publication’s count, leaving self-sufficiency at about 16 per cent. China spends roughly US$3 billion a year on soybean subsidies to keep production above 20 million tonnes, and retired USDA economist Fred Gale has argued self-sufficiency is unlikely ever to reach even 20 per cent. Higher-yielding GM soybeans help at the margin; they cannot conjure the extra land that closing a 90 million tonne gap would require.
Table: China’s import dependence in corn and soybeans
|
Indicator |
Corn |
Soybeans |
|
Peak import year |
2021: 28.35 million t |
2025: 111.82 million t (record) |
|
2025 imports |
About 2.65 million t (down about 81% year on year) |
111.82 million t (up 6.5%) |
|
Leading 2025 supplier |
Brazil, over 60% share |
Brazil, 73.6% share (82.33 million t) |
|
2025 domestic production |
Record national grain harvest of 714.9 million t |
20.91 million t |
|
Approximate self-sufficiency |
High; imports a small share of use |
About 16% |
|
Policy lever |
7.2 million t tariff-rate quota |
US purchase pledge of 25 million t a year, 2026 to 2028 |
|
Biotech status |
GM area scaling; dozens of varieties registered |
GM varieties registered; gene-edited high-oleic lines certified |
Sources: China Customs via Mysteel and Hellenic Shipping News; Western Producer; Reuters; MARA.
The friction nobody puts in the press release
None of this is frictionless. Reuters reporting in September 2025 described a rollout hampered by tight state controls over which seeds may be sold where, seed companies sitting on unsold inventory, and farmer and consumer scepticism about GM food. Some early regional trials in 2022 and 2023 reportedly saw yield declines of 10 to 20 per cent, attributed to rushed releases with insufficient backcrossing into locally adapted lines. Illegal GM seed use, a long-running problem in the north-east, complicates both regulation and the business case for licensed suppliers.
Weather also reminds everyone that genetics are only one input. Heavy rain during the 2025 harvest left part of northern China’s record corn crop mouldy, Reuters reported in February 2026, pushing domestic corn prices up about 10 per cent year on year. Chinese buyers responded by booking at least 2.5 million tonnes of US sorghum over three months and taking around one million tonnes a month of Australian barley, roughly double the previous year’s pace. Neither sorghum nor barley is subject to import quotas, which makes them the pressure valve when domestic corn disappoints.
Then there is geopolitics. After the Trump and Xi meeting in late October 2025, Beijing resumed US soybean purchases and committed to buy at least 25 million tonnes a year from 2026 to 2028. By September 2026 China was more than halfway to that year’s pledge, The Edge reported, but USDA data showed no sales of US corn for the 2026 to 2027 season. Soybeans are a negotiating chip; corn, increasingly, is not.
What changes for you
Grain and oilseed exporters (Brazil, the US, Argentina, Russia). The corn opportunity in China is shrinking and becoming more volatile. Expect demand to spike in poor-harvest years and fade in good ones, rather than the steady 20 million tonne-plus offtake of 2021 to 2023. Soybeans remain a volume business for at least the rest of the decade, but with political allocation built in: the US share is negotiated, and Brazil’s is the default.
Australian and Southeast Asian feed grain suppliers. Barley and sorghum demand from China will behave like an insurance policy: lucrative when Chinese corn quality slips, thin otherwise. Exporters should price that optionality rather than plan around it.
Seed and trait companies across the region. China is building national champions with a domestic market, state backing and, increasingly, gene-editing intellectual property that faces fewer regulatory hurdles than transgenic events. Expect those firms to look outward, particularly to Southeast Asia, Central Asia and Africa, where corn hybrid markets are growing and Western trait licensing has been expensive. Companies operating in Vietnam, the Philippines and Indonesia should anticipate new Chinese competition in hybrid corn genetics.
Feed buyers and integrators in Southeast Asia. If China imports less corn, more Brazilian, Argentine and Black Sea corn is available for Vietnam, Thailand, Indonesia and the Philippines, which could soften delivered prices in good Chinese harvest years. The flip side is that a weather-hit Chinese crop now sends buyers rushing into sorghum and barley markets that Southeast Asian feed mills also use.
Why the seed bag is now a trade instrument
The deeper shift is conceptual. For two decades, the global grain trade treated China’s demand as a function of pig herds and incomes. Increasingly, it is also a function of Beijing’s seed policy: how many GM varieties are registered, which provinces are allowed to plant them, and whether gene-edited lines clear safety review. Those are administrative decisions, made on an annual cycle, and they can shift import demand by millions of tonnes.
That makes MARA’s approval calendar as relevant to trading desks as the USDA’s monthly balance sheets. It also means the countries most exposed are not only the big exporters but the mid-sized suppliers and the feed-importing economies of Southeast Asia, whose price environment now depends partly on what grows in Jilin. The seed revolution will not end China’s role as the world’s largest food importer. It will make that role more discretionary, and discretion, in commodity markets, is a form of power.
What to watch in 2027
- Whether MARA’s 2027 variety registration lists continue the pace of 2024 and 2025, particularly for herbicide-tolerant and insect-resistant corn.
- Official or credible industry estimates of the 2026 and 2027 GM corn area, and whether adoption breaks beyond the north-east into the North China Plain.
- The first commercial planting of gene-edited soybean or corn lines under the 15th Five-Year Plan’s biological breeding push.
- Progress on China’s 25 million tonne annual US soybean pledge, and whether any equivalent commitment emerges for corn or sorghum.
- Outbound moves by Chinese seed champions such as Longping High-Tech and Dabeinong into Southeast Asian and Central Asian corn markets.
- Domestic corn quality after the 2026 harvest, the clearest signal of whether China will lean on Australian barley and US sorghum again.
Ankit Kankar | ankit.kankar@mmactiv.com