Chapter 17 · Part VNew in edition 1.1
17. The protein balance to 2050: how much feed protein Vietnam will need
On trend diets and trend efficiency, Vietnam's soybean-meal need rises from 7.2 Mt (2025) to about 10.4 Mt (2050), nearly all imported, and the country already farms about 4 million hectares abroad for feed. Across four scenarios the 2050 range is 7.0 to 11.7 Mt. Better feed conversion and lower soybean-meal inclusion cut 2050 imports about twice as much as a large alternative-protein build-out. Aquafeed is short of omega-3 more than of protein. The published baseline and official targets stop at 2030 to 2035; our 2040 and 2050 values are estimates from a reproducible model.
On this page
In one paragraph. No one has published a projection of Vietnamese meat, feed or feed-protein demand beyond 2035, and official targets stop at 2030. We therefore built a simple, reproducible balance model (tools/balance_model.py) that reproduces the published outlooks to 2035 within about 2% and extends them to 2050 under four scenarios. On trend diets and trend efficiency, soybean-meal need rises 44%, from 7.2 Mt in 2025 to about 10.4 Mt in 2050, almost all imported. The range across scenarios is 7.0 to 11.7 Mt. Two findings shape Part V. First, feed conversion and formulation move the 2050 result more than diets do, and a plausible efficiency programme cuts imports about twice as much as a large alternative-protein build-out; the two add up rather than compete. Second, the alternative-protein scenario, modest as it looks, implies about 78 times Entobel's current Vietnamese capacity by 2050, a level no country has come close to.
17.1 What published numbers exist
| Series | 2025 | 2030 | 2035 | Beyond 2035 | Type | Sources |
|---|---|---|---|---|---|---|
| Industrial livestock feed, Mt | about 22 | 30 to 32 | none | none | Official target | 1,2 |
| Meat output, Mt | about 6.0 | 6.0 to 6.5 | none | none | Official target | 3,2 |
| Aquaculture, Mt | 5.7 (OECD-FAO base) | 7.0 target; 6.12 OECD-FAO | 6.83 OECD-FAO | qualitative 2045 vision | Target and projection | 4,5 |
| Protein-meal feed use, Mt | 9.5 | 11.1 | 12.9 | none | Projection (OECD-FAO baseline) | 6 |
| Protein-meal imports, Mt | 7.4 | 8.6 | 10.3 | none | Projection (OECD-FAO baseline) | 6 |
| Maize imports, Mt | 12.0 | 14.0 | 16.6 | none | Projection (OECD-FAO baseline) | 6 |
| Fishmeal feed use, kt | 221 | 309 | 339 | none | Projection (OECD-FAO baseline) | 6 |
All forward values: . We found no FAO, IFPRI, GLOBIOM or peer-reviewed projection for Vietnamese meat, feed or feed protein to 2050 6,7,8 .
17.2 Four scenarios
| Scenario | Diets | Aquaculture | Feed efficiency and formulation | Alternative protein |
|---|---|---|---|---|
| S-BASE (trend) | OECD-FAO growth to 2035, then tapering: 66.5 kg of meat per person (2025) to 87.6 kg (2050) | OECD-FAO growth to 2035, then +1% a year: 8.5 Mt in 2050 | Feed conversion improves 0.4% a year (livestock) and 0.3% (aquaculture); soybean-meal inclusion constant | None |
| S-HIGH (official growth path) | Incomes follow the power plan's 10% then 7.5% a year: 101 kg of meat per person by 2050 | Official 7.0 Mt by 2030, 9.0 Mt by 2050 | As S-BASE | None |
| S-EFF (efficiency) | As S-BASE | As S-BASE | Feed conversion improves 1.0% and 0.8% a year; soybean-meal inclusion falls 1% a year (amino acids, low-protein diets, by-products) | None |
| S-ALT (alternative protein) | As S-BASE, with 1% (2030), 5% (2040) and 10% (2050) of meat demand replaced by plant-based and fermented foods | As S-BASE | As S-BASE | Microbial protein replaces 1%, 5% and 10% of soybean-meal protein and 10%, 25% and 40% of fishmeal protein (2030, 2040, 2050) |
The S-ALT shares are illustrative what-if levels, not targets. The fishmeal shares sit within what feeding trials support; the soybean-meal share implies large cost falls or policy support (Economics).
17.3 Results
Soybean-meal import need, Mt
| Scenario | 2025 | 2030 | 2035 | 2040 | 2050 | 2050 against 2025 |
|---|---|---|---|---|---|---|
| S-BASE | 7.20 | 8.46 | 9.47 | 9.95 | 10.40 | +44% |
| S-HIGH | 7.20 | 8.98 | 10.11 | 10.92 | 11.72 | +63% |
| S-EFF | 7.20 | 7.81 | 8.08 | 7.85 | 7.00 | minus 3% |
| S-ALT | 7.20 | 8.32 | 9.01 | 9.15 | 8.78 | +22% |
Soybean-meal need to 2050 in four scenarios
Million tonnes a year, our model; shaded band: the vision range of chapter 19
Show dataHide data
| Scenario | Year | Soybean-meal need (Mt) | Vision range low (Mt) | Vision range high (Mt) |
|---|---|---|---|---|
| S-BASE | 2025 | 7.2 | 7.2 | 7.2 |
| S-BASE | 2030 | 8.46 | 7.8 | 8.3 |
| S-BASE | 2035 | 9.47 | 8.1 | 9.0 |
| S-BASE | 2040 | 9.95 | 7.8 | 9.2 |
| S-BASE | 2050 | 10.4 | 7.0 | 8.8 |
| S-HIGH | 2025 | 7.2 | 7.2 | 7.2 |
| S-HIGH | 2030 | 8.98 | 7.8 | 8.3 |
| S-HIGH | 2035 | 10.11 | 8.1 | 9.0 |
| S-HIGH | 2040 | 10.92 | 7.8 | 9.2 |
| S-HIGH | 2050 | 11.72 | 7.0 | 8.8 |
| S-EFF | 2025 | 7.2 | 7.2 | 7.2 |
| S-EFF | 2030 | 7.81 | 7.8 | 8.3 |
| S-EFF | 2035 | 8.08 | 8.1 | 9.0 |
| S-EFF | 2040 | 7.85 | 7.8 | 9.2 |
| S-EFF | 2050 | 7.0 | 7.0 | 8.8 |
| S-ALT | 2025 | 7.2 | 7.2 | 7.2 |
| S-ALT | 2030 | 8.32 | 7.8 | 8.3 |
| S-ALT | 2035 | 9.01 | 8.1 | 9.0 |
| S-ALT | 2040 | 9.15 | 7.8 | 9.2 |
| S-ALT | 2050 | 8.78 | 7.0 | 8.8 |
Other results, 2050
| Indicator | 2025 | S-BASE 2050 | Range across scenarios, 2050 |
|---|---|---|---|
| Compound feed, Mt | 28.5 | 40.9 (livestock and poultry 31.6; aquafeed 9.2) | 35.4 (S-EFF) to 46.2 (S-HIGH) |
| Maize import need, Mt | 11.9 | 13.8 | 11.6 (S-EFF) to 16.3 (S-HIGH) |
| Fishmeal demand, kt | 262 | 253 | about 151 (S-EFF, S-ALT) to 267 (S-HIGH) |
| Farmland abroad for imported soy and maize, million ha | 4.0 | 4.7 | 3.4 (S-EFF) to 5.4 (S-HIGH) |
| Meat production, Mt carcass weight | 6.0 | 8.5 | 7.7 (S-ALT) to 9.9 (S-HIGH) |
Context and sources
S-EFF 7.0, S-ALT 8.8, S-BASE 10.4, S-HIGH 11.7 Mt. Only the efficiency scenario keeps imports at or below the 2025 level. 6
Readings.
- The near term checks out. Interpolated 2027 values (30.5 Mt of feed; 7.7 Mt of soybean meal) match USDA's forecasts (30.6 and 7.8 Mt), and protein-meal use is within about 2% of OECD-FAO for 2030 and 2035 (+2.3% and minus 2.2%) 11,10,6 .
- Vietnam already farms about 4 million hectares abroad for feed, about five times its own maize area, rising to 4.7 million by 2050 on trend (our calculation) .
- Maize imports depend more on domestic output and fuel use than on diets. If E10 ethanol shifted to maize and non-feed use reached the OECD-FAO 2035 level, 2050 imports would rise by 2.8 Mt .
- Demand is already steep. OECD-FAO puts Vietnamese pig, poultry and beef food use at 61.9 kg per person in 2035, above Korea (59.2 kg) and China (48.5 kg) in the same outlook 6 . Our taper after 2035 is a judgement; S-HIGH is a stress case, not a central view.
- Fishmeal starts from a different base and diverges. Our 2025 figure (262 kt) is above OECD-FAO's apparent feed use (221 kt) because it includes snakehead, marine fish and livestock uses; by 2035 the two diverge further (274 against 339 kt) because our inclusion rates fall 1.5% a year while OECD-FAO's use rises. We report both and keep ours (R. Disagreements) .
17.4 What moves the 2050 result
Varying one assumption at a time around S-BASE (10.4 Mt in 2050):
- Formulation and efficiency dominate. The soybean-meal inclusion trend (minus 1% to plus 0.5% a year) swings 2050 imports by 3.7 Mt; the rate of feed-conversion gains (0 to 1% a year) by 2.4 Mt .
- Demand assumptions matter less. Aquaculture growth after 2035, poultry demand per person, meat self-sufficiency and population each move the result by about 0.8 to 1.1 Mt; pork demand and pig feed assumptions by less.
- The two largest levers are also the least observed. We found no Vietnamese time series of average soybean-meal inclusion or feed conversion by species. Measuring them is the first job of a protein statistic (Robust moves).
Efficiency does twice the work of alternative protein at the shares tested. S-EFF cuts 2050 soybean-meal need by 3.4 Mt against S-BASE; S-ALT cuts it by 1.6 Mt (16%). The two are additive, not rivals . A combined case was not modelled, so we give no number for it.
Context and sources
Against trend, feed conversion improving 1.0% a year (livestock) and 0.8% (aquaculture) instead of 0.4% and 0.3%, and soybean-meal inclusion falling 1% a year, save 3.4 Mt in 2050; replacing 10% of soybean-meal protein, 40% of fishmeal protein and 10% of meat demand with alternatives saves 1.6 Mt. The two add up. 6
17.5 What an alternative-protein build-out would take
S-ALT looks modest on paper: 10% of soybean-meal protein and 40% of fishmeal protein by 2050. In physical terms it is industrial.
| Item | 2030 | 2040 | 2050 |
|---|---|---|---|
| Microbial feed protein, kt of protein | 56 | 265 | 514 |
| Microbial feed product (dry), kt | 102 | 460 | 863 |
| Multiples of Entobel's Vietnamese design capacity (11 kt a year, two plants) | about 9 | about 42 | about 78 |
| Food substitute protein, kt of protein | 12 | 69 | 145 |
| Glucose for the sugar route (feed and food), kt | 238 | 852 | 1,318 |
| as fresh cassava roots, kt (share of the 2025 harvest) | 968 | 3,473 | 5,368 (about 52%) |
| Electricity, TWh a year | 0.21 | 4.0 | 11.7 |
| Hydrogen for the gas route, kt a year | 0 | 57 | 185 |
| Soybean-meal imports avoided against S-BASE, Mt | 0.14 | 0.80 | 1.62 |
| Maize imports avoided, Mt | 0.13 | 0.70 | 1.37 |
Multiples: our arithmetic from Entobel's combined design capacity in Vietnam, 1 kt a year in Dong Nai and 10 kt in Ho Chi Minh City 12 (Feed and aquafeed market) .
Context and sources
S-ALT implies about 0.86 Mt of dry microbial feed product in 2050: about 78 times the combined design capacity of Entobel's two Vietnamese insect-meal plants (11 kt a year), or 43 plants of Calysseo's 20 kt size; about 9 times by 2030. Plus 0.29 Mt of food protein ingredient. Our arithmetic. 6,12
Readings.
- Carbohydrate is the binding physical constraint. The 2050 sugar-route need equals about half of Vietnam's 2025 cassava harvest or almost all of its sugar, competing with starch exports to China and E10 ethanol 13,14,15,16 . Residue carbon could cover 5% to more than 100% of it, depending on whether cellulosic sugar works (Z. Aquafeed and feedstocks 2050).
- Energy volume is not the constraint; energy carbon is. 11.7 TWh is under 1% of planned 2050 electricity 17 . The gas route beats soybean meal on CO2 only below about 21 g CO2 per kWh, a grid average the power plan reaches only around 2050 .
- The food route saves the most imports per tonne. Replacing 1 t of meat protein in diets avoids about 4.5 t of soybean-meal and 9.5 t of maize imports; microbial protein in feed replaces soybean meal but not maize . Diet change is outside this study's scope; we report the arithmetic only.
- Cassava-based microbial protein moves land rather than saving it: about 0.80 ha per t of protein against about 0.71 ha for imported soybean meal (our calculation) . Hydrogen gas fermentation needs a small fraction of that 21,19 .
17.6 Aquafeed: short of omega-3 more than of protein
Aquaculture growth after 2030 is the largest uncertainty for aquafeed, not formulation 22,23 . Our three species-level paths give 7.4, 8.9 and 11.3 Mt of aquaculture in 2050, needing 8.0, 9.7 and 12.5 Mt of compound aquafeed and 2.5, 3.1 and 4.1 Mt of feed crude protein (2.0 Mt in 2025) . These species-level paths are separate from the four model scenarios. The middle path uses the S-BASE index and sits 0.3 Mt above S-BASE only because marine finfish grows faster; the high path follows the 7.0 Mt target to 2030 and then grows 2% a year, well above S-HIGH's 9.0 Mt.
- The capture plan removes most of the trash-fish supply. The trawl fleet is to shrink to about 8,360 boats by 2030, about 59% below 2016 24,25 (official target). About 90% of south-central mariculture stakeholders still fed trash fish in 2021 26 . Replacing it adds little tonnage but protein-dense, fishmeal-heavy feed: 7 to 34 kt of extra fishmeal by 2030 at today's formulations .
- Vietnam is already a large by-product marine-ingredient producer. Pangasius off-cuts supplied 12% of the world's by-product fishmeal and 29% of its by-product fish oil in 2024 27 . But pangasius oil carries only 0.17 to 0.31% EPA plus DHA 28 , so it is not an omega-3 source.
- Omega-3 is the tighter constraint. Vietnamese aquafeed needs about 11 kt of EPA plus DHA today and about 21 kt by 2050 on the middle path (10 to 40 kt) ; the world supply gap is estimated at 0.4 Mt to over 1 Mt 29 .
- A credible 2050 share for alternative ingredients is 1 to 6% of aquafeed protein but 15 to 50% of aquafeed omega-3 (our estimate, from 30,31,32) . Algal and oilseed omega-3 oils are already commercial.
- Domestic soybean is not a lever. Even 300,000 ha at 2.5 t per ha would give 5.6% of the 2050 S-BASE bean-equivalent need, and domestic beans sell into food at about twice the import price 33,5 .
17.7 Reality check: what other countries have achieved
The model's alternative-protein scenario sits above every real-world analogue we found; its efficiency scenario sits at the low end of what China has done.
- Formulation moves faster than land. China cut the soybean-meal share of feed by 0.14 to 0.45 percentage points a year between 2017 and 2025; Norway cut the marine share of salmon feed from 90% (1990) to about 30% (2013), about 2.6 points a year, under price pressure 34,35,36 .
- Self-sufficiency targets are usually missed, then cut. Japan's feed self-sufficiency was 28% in FY2000 and 27% in FY2024, and its FY2030 target fell from 34% to 28% 37,38 .
- Novel feed ingredients stay tiny even in the most advanced aquafeed market. Insect meal, single-cell protein, fermented products and microalgae made up 0.4% of Norwegian salmon-feed ingredients in 2020 39,40 .
- S-ALT's microbial share (8.4% of high-protein feed protein by 2050) is unprecedented. Its 2030 share (1.0%) is ambitious but within reach, since Norway was at 0.4% in 2020 39 .
On this evidence, a credible ambitious range for Vietnam's soybean-meal need in 2050 is about 7 to 9 Mt, against 10.4 Mt on trend. Holding imports at the 2025 level (7.2 Mt) would be more ambitious still . Vision 2050 builds on this range.
17.8 What this means
- Investors: on trend, soybean-meal need grows by about 3.2 Mt between 2025 and 2050, but efficiency and formulation will take the largest share of any reduction. The sharper thesis is aquafeed omega-3 and fishmeal replacement.
- Policy makers: a feed-efficiency and formulation programme is the lever with a real-world pace behind it. The 2030 livestock-feed target needs a scope statement and a statistic to be judged.
- Manufacturers: soybean-meal inclusion and feed conversion are the variables that matter most, and nobody publishes them. Sharing anonymised data would make every plan better, including your own.
- Startups: plan volumes against real plants. Nine times Entobel's current Vietnamese capacity by 2030 is ambitious; 78 times by 2050 is a national programme, not a market outcome.
- Research and international bodies: a published Vietnamese protein and feed balance, updated yearly, is the missing public good (Robust moves).
Related: W. Balance model (model, assumptions and sensitivity), Z. Aquafeed and feedstocks 2050, Feed and aquafeed market (2025 feed market), Outlook.
Sources cited on this page
- NTS-23 Prime Minister. Quyết định 1625/QĐ-TTg, Đề án Phát triển công nghiệp chế biến thức ăn chăn nuôi đến năm 2030. 15 Dec 2023
- REG-54 Prime Minister. Quyết định 1520/QĐ-TTg 2020 chiến lược phát triển chăn nuôi 2021 to 2030. 6 Oct 2020
- MAC-23 Prime Minister of Vietnam. Decision 1520/QD-TTg approving the Livestock Development Strategy 2021-2030, vision 2045. 6 Oct 2020
- NTS-08 Prime Minister. Quyết định 339/QĐ-TTg, Chiến lược phát triển thủy sản Việt Nam đến năm 2030, tầm nhìn 2045. 11 Mar 2021
- AQF-02 OECD and FAO. OECD-FAO Agricultural Outlook 2026-2035, SDMX API, dataflow OECD.TAD.ATM DSD_AGR@DF_OUTLOOK_2026_2035 version 1.1. June 2026
- QNT-01 OECD and FAO. OECD-FAO Agricultural Outlook 2026-2035, data via the OECD SDMX API, dataflow OECD.TAD.ATM DSD_AGR@DF_OUTLOOK_2026_2035 version 1.1. June 2026
- QNT-02 USDA Office of the Chief Economist. USDA Agricultural Projections to 2035, OCE-2026-1. 13 February 2026
- QNT-15 ABARES. Food demand to 2050: opportunities for Australian agriculture. 2012
- MAC-28 VietDVM. San luong thuc an chan nuoi nam 2024 dat 21,5 trieu tan, tang 3,4% so voi nam 2023. 2024
- MAC-04 USDA FAS GAIN. Vietnam: Oilseeds and Products Annual, report VM2026-0006 (Nguyen Linh). 15 Apr 2026
- MAC-01 USDA FAS GAIN. Vietnam: Grain and Feed Annual, report VM2026-0012. 15 Apr 2026
- ECO-01 Entobel. Entobel Sets Industry Record with the Opening of the Largest Insect Protein Production Plant in Asia. 23 Nov 2023
- FS-01 Dân Việt. Việt Nam đang là nhà cung cấp lớn nhất một thứ tinh bột cho Trung Quốc, năm 2025 bán được gần 3,7 triệu tấn. January 2026
- FS-05 Phommakod, S., Lueangwattanapong, K., Chaiprasert, P., & Songkasiri, W. Valorization of biomaterial side streams: Kinetics of cassava pulp and its components degradation by Clostridium manihotivorum CT4T. 2026
- FS-31 Doanh nghiệp Hội nhập. Ngành mía đường sản xuất gần 1,3 triệu tấn, áp lực tiêu thụ vẫn đè nặng. 16 July 2026
- GT-10 MOIT. Thông tư 50/2025/TT-BCT quy định lộ trình áp dụng tỷ lệ phối trộn nhiên liệu sinh học với nhiên liệu truyền thống tại Việt Nam. 7 Nov 2025
- GT-11 Prime Minister. Quyết định 768/QĐ-TTg năm 2025 phê duyệt Điều chỉnh Quy hoạch phát triển điện lực quốc gia thời kỳ 2021-2030, tầm nhìn đến năm 2050. 15 Apr 2025
- QNT-10 Sillman, J., Uusitalo, V., Ruuskanen, V. and others. A life cycle environmental sustainability analysis of microbial protein production via power-to-food approaches. 2020
- QNT-11 Jean, A.B. and Brown, R.C.. Techno-economic analysis of gas fermentation for the production of single cell protein. 2024
- FTG-01 Fasihi, M., Jouzi, F., Tervasmaki, P., et al.. Global potential of sustainable single-cell protein based on variable renewable electricity. 2025
- QNT-09 Leger, D., Matassa, S., Noor, E., Shepon, A., Milo, R. and Bar-Even, A.. Photovoltaic-driven microbial protein production can use land and sunlight more efficiently than conventional crops. 2021
- AQF-25 Chan, C.Y., Tran, N., Dao, D.C., Sulser, T.B., Phillips, M.J. and others. Fish to 2050 in the ASEAN region. 2017
- MAC-15 NSO. Thuy san Viet Nam quy IV va ca nam 2025: Giu da tang truong, vuot thach thuc, huong toi phat trien ben vung. 8 Jan 2026
- AQF-15 Prime Minister. Quyết định 389/QĐ-TTg Quy hoạch bảo vệ và khai thác nguồn lợi thủy sản thời kỳ 2021-2030, tầm nhìn đến năm 2050. 9 May 2024
- AQF-06 Leadbitter, D.. Driving change in South East Asian trawl fisheries, fishmeal supply, and aquafeed. 2019
- AQF-07 Journal of Agriculture and Ecology Research International. Mariculture in Southern Central Region, Vietnam: Status and Orientation Toward Sustainable Development. 2021
- AQF-04 IFFO, The Marine Ingredients Organisation. Update on by-product marine ingredients. March 2026
- AQF-29 Aryani, N., Suharman, I. and Heltonika, B.. Changes in the fatty acid profile of fish oil derived from Pangasius catfish (Pangasianodon hypophthalmus) processing waste due to variations in fish size and heating temperatures. 2023
- AQF-16 Tocher, D.R., Betancor, M.B., Sprague, M., Olsen, R.E. and Napier, J.A.. Omega-3 long-chain polyunsaturated fatty acids, EPA and DHA: bridging the gap between supply and demand. 2019
- AQF-19 Vo, L.L.G., Galkanda-Arachchige, H.S.C., Iassonova, D.R. and Davis, D.A.. Efficacy of modified canola oil to replace fish oil in practical diets of Pacific white shrimp Litopenaeus vannamei. 2021
- AQF-23 Veramaris. Veramaris opens US$200m facility for EPA and DHA omega-3 algal oil. 10 Jul 2019
- AQF-24 Feed Strategy. Plant-based fish oil alternatives growing fast in aquafeed. 1 Jun 2021
- AQF-20 VnEconomy. Ngành đậu tương có nhiều 'nút thắt' cần tháo gỡ. 7 Nov 2024
- GEO-10 farmdoc daily (University of Illinois). Can China Reduce Soybean Import Demand? Evaluating Soybean Meal Reduction Efforts. December 2025
- GEO-11 South China Morning Post. China failed to hit 2025 soybean meal dependence target. US deal makes it harder. 9 Feb 2026
- VIS-19 Ytrestøyl T., Aas T.S., Åsgård T.. Utilisation of feed resources in production of Atlantic salmon (Salmo salar) in Norway. 2015
- VIS-11 MAFF Japan. Summary of the Basic Plan for Food, Agriculture and Rural Areas. March 2020
- VIS-12 MAFF Japan. 飼料をめぐる情勢（データ版）. August 2026
- VIS-18 Aas T.S., Åsgård T., Ytrestøyl T.. Utilization of feed resources in the production of Atlantic salmon (Salmo salar) in Norway: An update for 2020. 2022
- VIS-20 Nofima. Salmon feed is slowly changing. 19 Apr 2022