1. Why Vietnam: a protein-processing economy built on imported plant protein

In one paragraph. Vietnam is one of Asia's largest producers of pork, poultry, eggs, farmed fish and farmed shrimp, and exports more than USD 11 billion of seafood a year. Almost all of the plant protein that feeds these animals arrives by ship: about 99% of the soy protein in Vietnamese feed and about three-quarters of the maize supply are imported, for a feed-ingredient import bill of roughly USD 10 billion in 2025. Feed demand is forecast to keep growing, domestic soybean and maize area is shrinking, and 2026 brought a new feed-price shock and a fishmeal price at its highest level this decade. This is the structural case for alternative protein in Vietnam: not a consumer fashion or a climate headline, but a protein supply chain that depends on a few exporting countries and a few commodity markets. The rest of this report asks what Vietnam could make at home, from what, at what cost, under which rules, and who could do it.

---

#1.1 A large and growing animal-protein economy

Vietnam's livestock and aquaculture sectors are large by any regional standard, and still growing.

• Meat. In 2025 Vietnam produced 5.39 Mt of pork and 2.60 Mt of poultry meat (live weight), plus 0.50 Mt of beef and 0.13 Mt of buffalo meat. Pork grew 3.9% and poultry 5.9% on 2024 ^7,8 VN-directHigh.
• Eggs and milk. 21.4 billion eggs (+5.3%) and 1.3 Mt of fresh milk (+5.5%) in 2025 ⁷ VN-directHigh.
• Fish and shrimp. Fisheries output reached 9.95 Mt in 2025, of which aquaculture was 6.12 Mt. Farmed pangasius was 1.94 Mt, whiteleg shrimp 0.99 Mt and black tiger shrimp 0.30 Mt ⁹ VN-directHigh.
• Animals on farms. At the end of 2025 there were about 31.4 million pigs and 584.9 million poultry in Vietnam ⁸ VN-directMedium. About 53.5 million pigs were slaughtered in 2024 according to one press report ¹⁰ VN-directLow.
• Still growing in 2026. In the first half of 2026 pork output rose 4.8% and poultry 5.6% year on year ¹¹ VN-directMedium.

Pork consumption was reported at about 39 kg per person in 2025, up from about 30 kg in 2021 ¹² VN-directLow. Our own check from production and population gives a similar figure (about 39.5 kg carcass weight per person at a 75% carcass yield; our calculation from ^7,13).

Vietnam's animal protein output, 2025

Thousand tonnes (meat live weight; aquaculture harvest)

#1.2 Running on imported feed

The animal-protein economy runs on compound feed, and compound feed runs on imports.

Feed output. Vietnamese feed mills made about 22 Mt of livestock and poultry feed in 2025, after 21.5 Mt in 2024 ^4,14 VN-directHigh. This official figure excludes aquafeed. Aquafeed estimates disagree: industry sources put 2025 aquafeed at 3.9 to 4.8 Mt by species, while USDA's post estimates 6.5 Mt ^15,4 VN-directLow. We do not average them (see Feed and aquafeed market and R. Disagreements).

Imports of the two ingredients that set feed cost:

Vietnam grew only 42,000 to 43,000 t of soybeans in 2025 on about 26,000 ha, and domestic maize output was flat to falling at about 4.1 Mt ^1,4 VN-directHigh. Even if every domestic soybean were crushed, it would supply about 2% of the crush (our calculation from ¹). Domestic crushing capacity is expanding, led by one Bunge and Wilmar plant at Phu My in Ho Chi Minh City (former Ba Ria-Vung Tau) with 2.6 Mt a year of capacity ¹⁶ VN-directMedium. But crushing imported beans at home shifts imports from meal to beans; it does not reduce protein import dependence.

The import bill. Adding four customs lines (maize, soybeans, wheat and the "animal feed and ingredients" group) gives about USD 10.2 billion of imports in 2025 (our calculation from ^5,3,2,6). This includes some milling wheat and food soybeans. The livestock department's narrower 2024 figure for feed ingredients is 22.4 Mt and USD 7.7 billion ¹⁴ VN-directMedium. USDA's 2025 feed-ingredient table shows 23.57 Mt imported out of 28.62 Mt used, about 82% ¹⁷ VN-directHigh.

How much of Vietnam's feed protein and grain is imported

Share of 2025 supply imported, by ingredient

#1.3 Exposure: prices, disease and policy shocks

A protein supply chain that depends on imports transmits external shocks directly into Vietnamese farm costs.

• Feed is most of the cost. Industry sources put feed at 60% to 70% of livestock production cost, and above 70% for pigs ^18,19 VN-directLow. For shrimp, 2026 trade press puts feed at 50% to 60% of production cost ²⁰ VN-directMedium.
• 2022. At the peak of the post-Ukraine price spike, wheat import prices reached USD 424/t in September 2022, 37% above a year earlier ²¹ VN-directMedium. A review estimates that about 30% of small pig farmers left the sector during the COVID-19 and 2022 disruptions ¹⁹ VN-directLow.
• 2026. Feed prices rose 10% to 14% between January and April 2026, which the poultry association attributed to the Middle East conflict. Farm-gate egg prices fell below production cost ¹⁸ VN-directMedium. Shrimp feed makers raised prices by VND 1,200 to 1,500/kg in June 2026 ²⁰ VN-directMedium.
• Fishmeal. The World Bank fishmeal price reached USD 2,500/t in August 2026, 43% above a year earlier and the highest in the 2020 to 2026 series ^22,23 generalHigh. The fishmeal-to-soybean-meal price ratio widened from 3.8 in 2024 to 6.2 in August 2026 ²⁴ generalHigh.
• Disease. African swine fever killed or forced the culling of about 6 million pigs in 2019, more than 20% of the herd, and returned from the third quarter of 2025, when 1.2 to 1.3 million pigs were culled ^25,4,8 VN-directMedium.
• Tax. Animal feed became "not subject to VAT" from 1 July 2025, so feed makers can no longer deduct VAT paid on inputs ^26,27 VN-directHigh. This matters for any new feed-protein producer (see Economics).

Fishmeal at its highest price this decade

World Bank fishmeal series (Peru 65%), USD per tonne; soybean meal for comparison

#1.4 Growth ahead, and where it will come from

• USDA expects total feed demand (livestock, poultry and aquafeed) to rise from 28.6 Mt in 2025 to 30.6 Mt in 2027, and raised its forecast of maize imports for 2026/27 to 15 Mt ^4,28 VN-directMedium.
• The national livestock strategy targets 30 to 32 Mt of industrial feed output and 6.0 to 6.5 Mt of carcass meat by 2030 ²⁹ VN-directHigh. If the feed target covers livestock and poultry feed only, it implies growth of about 6% to 8% a year from 2025 (our calculation; see O. Outlook data for the scope question).
• Vietnam's population is projected at 104.3 million in 2030 and 106.5 million in 2035 (UN medium variant) ³⁰ VN-directMedium. Population will rise only about 5% between 2025 and 2035, so protein demand growth will come mostly from income and diet.
• Nothing in the published strategies sets a numerical target for localising feed raw materials. The livestock strategy encourages domestic feed from by-products such as brewer's grain and cassava pulp, and biological products that replace antibiotics, but sets no localisation number ^29,31 VN-directHigh.

On current trends, every additional tonne of feed demand will be met by imports.

#1.5 An exporter of animal protein and an importer of the plant protein behind it

Vietnam exported USD 11.3 billion of seafood in 2025 (+12.4%), including about USD 4.6 billion of shrimp and USD 2.2 billion of pangasius ^9,4 VN-directHigh. In January to August 2026 seafood exports reached USD 8.0 billion, with China and Hong Kong now the largest market ³² VN-directMedium. Vietnam also exported about USD 1.4 to 1.5 billion of animal feed in 2025 ^4,8 VN-directMedium, and it is a net exporter of fishmeal by value (USD 430 million exported, USD 260 million imported in 2023) ^33,34 VN-directMedium.

The shape of the economy is therefore clear: Vietnam imports plant protein and energy, converts them into animal and aquatic protein, and exports a large share of the result. Its competitive strength is processing. Its vulnerability is the input.

A second, related pattern appears on the carbohydrate side. Vietnam supplied about 39% of China's cassava starch imports by value in 2024 (1.56 Mt, USD 757 million, China-reported), but only about USD 0.16 million of glucose ^35,36 VN-directHigh. In 2025 Vietnamese starch exports rose to about 2.28 Mt in 11 months, and Vietnam became China's largest cassava starch supplier ^37,38 VN-directHigh. The step that turns starch into fermentable sugar, and sugar into higher-value products, mostly happens after the starch leaves Vietnam. Asset map and Region return to this.

#1.6 What the case for alternative protein rests on, and what it does not

For readers deciding whether Vietnam belongs in an alternative-protein strategy, four arguments carry different weight.

1. Supply security and the import bill (strong). About 99% of soy protein in feed and about 74% of maize supply are imported, feed demand is growing, and the 2022 and 2026 shocks show the exposure. Any domestically made protein that can match imported protein on price for a defined use reduces that exposure. This is the argument with the most evidence behind it.
2. Industrial value added (strong, but conditional). Vietnam already processes cassava, rice, sugar, soy and seafood at scale, and exports much of the carbohydrate as starch. Converting more of it to protein or fermentation feedstock at home is an industrial-policy argument that aligns with Resolution 36 on biotechnology and the 2026 high-technology priority list (see Rules). It is conditional on costs (Economics) and on a legal route to market.
3. Environment (real, but narrower than often claimed). Agriculture produced 104.5 MtCO2e in 2020, 19.8% of national emissions ³⁹ VN-directHigh. But in the 2016 inventory, livestock (enteric fermentation plus manure) was 18.5 MtCO2e, 5.8% of the national total, against 49.7 MtCO2e from rice cultivation ⁴⁰ VN-directMedium. The environmental case for alternative protein in Vietnam therefore rests more on the land footprint of imported feed, manure pollution and antimicrobial use than on national methane totals. About 71.7% of antimicrobials used in Vietnam in 2015 went to animals ⁴¹ VN-directMedium.
4. Animal welfare (a motivation for some actors, including the publisher of this report). The scale is large: tens of millions of pigs and hundreds of millions of poultry on farms at any time, and billions of farmed fish and shrimp. Proteins that displace animal products in food, or that displace forage fish in feed, reduce these numbers. Proteins that make aquaculture cheaper could also expand it. This report does not model welfare effects; the sign depends on the use (see Plays).

A fifth argument, consumer demand, is outside this report's scope. We note only that the traditional đồ chay (vegetarian) category already exists in every supermarket we visited, and that the first thing it lacks is protein, not shelf space (Alt protein today).

#1.7 What this means for each reader

• Investors: the addressable demand is B2B (feed mills and food manufacturers), large, concentrated and price-disciplined by duty-free imports of soybean meal and maize (0% MFN duty since March 2025 ⁴² VN-directMedium). Size theses on specific uses, not on total feed tonnage.
• Policy makers: the protein import dependence is structural and unmeasured in policy targets. A feed-and-protein security indicator (share of feed protein from domestic inputs) would make it visible.
• Startups: your first customers are among about 10 to 15 feed groups and a similar number of food manufacturers. The import price is your ceiling.
• Manufacturers: input-cost volatility (2022, 2026) is the business reason to qualify a second source of protein, even at parity.
• Researchers: the missing basic data (feed balances, Vietnam-specific fishmeal inclusion rates, aquafeed volumes that match output statistics) are cheap public goods.
• International bodies: Vietnam's feed-security exposure is a regional issue; the same import chains supply its neighbours.

Related: Feed and aquafeed market (all feed and fishmeal data), O. Outlook data (projections), Economics (the prices to beat).