AltProtein VietnamProtein thay thế tại Việt NamNghiên cứu xác định phạm vi phía cung · Phiên bản 1.0

AltProtein Vietnam · Phiên bản 1.0, tháng 11/2026

Protein thay thế tại Việt Nam

Nghiên cứu xác định phạm vi phía cung

Lập luận

Toàn bộ báo cáo trong 7 bước. Mỗi bước có con số chính và liên kết đến chương chứa bằng chứng. Nội dung bằng tiếng Anh.

  1. Why Vietnam

    Vietnamese people eat more meat each year: 136.4 g per person per day in 2020, against a recommended 50 to 80 g. Most of that meat, and the eggs and farmed fish, is produced in Vietnam on imported feed. About 99% of the soy protein in feed was imported in 2025, for a feed-ingredient bill of about USD 10 billion (Why Vietnam).

    about 99%

    Tỷ lệ đạm đậu tương nhập khẩu trong thức ăn chăn nuôi

    2025Tính toán của nhóm nghiên cứuBằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy trung bình: nguồn thứ cấp uy tín, hoặc đọc một phần nguồn gốc.

    Bối cảnh và nguồn

    In 2025 Vietnam imported 2.61 Mt of soybeans (98.4% of imports plus domestic output) and 5.70 Mt of soybean meal; about 99% is our calculation. Domestic crushing uses imported beans, so it does not reduce protein import dependence. 1,2,3

    Why Vietnam

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

    Imports as share of imports plus domestic production, 2025 (our calculation)

    How much of Vietnam's feed protein and grain is importedBar chart: about 99% of soy protein, 98% of soybeans, 79% of soybean meal, 74% of maize and 100% of wheat and DDGS used in Vietnam in 2025 were imported.0%25%50%75%100%Share of 2025 supply imported (%)100% importedSoy protein (beans andmeal)Soy protein (beans and meal): 99% imported Note: about 99% of soy protein in feed (our calculation from USDA supply data); domestic crush uses imported beans Main origins: Argentina, Brazil, United States99%SoybeansSoybeans: 98.4% imported, 2.61 Mt imported in 2025 Note: imports over imports plus domestic production Main origins: Brazil 45.8%, United States 43.6%98.4% (2.61 Mt)Soybean mealSoybean meal: 79% imported, 5.70 Mt imported in 2025 Note: imports over imports plus local crush Main origins: Mostly Argentina (origin shares not re-checked)79% (5.70 Mt)MaizeMaize: 74% imported, 11.91 Mt imported in 2025 Note: about 74%: imports over imports plus domestic production Main origins: Argentina 46.5%, Brazil 34.8%, United States about 13%74% (11.91 Mt)Wheat (all uses)Wheat (all uses): 100% imported, 5.45 Mt imported in 2025 Note: no domestic production Main origins: Australia, Brazil, United States100% (5.45 Mt)DDGSDDGS: 100% imported, 1.5 Mt imported in 2025 Note: no domestic production Main origins: United States 72%100% (1.5 Mt)

    Nguồn 1,2,3,4,5 Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy cao: đã đọc nguồn gốc.

    Xem số liệuẨn số liệu
    How much of Vietnam's feed protein and grain is imported
    IngredientShare imported, 2025 (%)Imports, 2025 (Mt)NoteMain origins
    Soy protein (beans and meal)99about 99% of soy protein in feed (our calculation from USDA supply data); domestic crush uses imported beansArgentina, Brazil, United States
    Soybeans98.42.61imports over imports plus domestic productionBrazil 45.8%, United States 43.6%
    Soybean meal795.70imports over imports plus local crushMostly Argentina (origin shares not re-checked)
    Maize7411.91about 74%: imports over imports plus domestic productionArgentina 46.5%, Brazil 34.8%, United States about 13%
    Wheat (all uses)1005.45no domestic productionAustralia, Brazil, United States
    DDGS1001.5no domestic productionUnited States 72%
  2. Food security: real, but narrow

    Counting feed, about two thirds of the animal protein people eat (61% to 71% in 2023) rests on imports (our calculation). Disease is the shock consumers feel: pork prices rose 57% in 2020 after African swine fever. But alternative protein made from imported soy only moves the dependence. Only domestic inputs reduce it, and at the scale demand supports, the effect is small (Why Vietnam).

    61% to 71%

    Tỷ lệ đạm động vật trong khẩu phần phụ thuộc nhập khẩu, trực tiếp hoặc qua thức ăn chăn nuôi

    2023Tính toán của nhóm nghiên cứuBằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy thấp: một nhận định đơn lẻ, báo chí, tiếp thị của doanh nghiệp hoặc suy ra từ dữ liệu yếu.

    Bối cảnh và nguồn

    Imported meat, offal, milk and fish, plus domestic pork, poultry, eggs and farmed fish attributed to imported compound feed (82% of feed ingredients by weight and about 93% by protein were imported in 2025). FAO food balance sheets, supply basis. Our calculation. 1,2,3

    Why Vietnam

  3. What it could make first

    Three of seven plays lead (Plays):

    • domestic textured plant protein, to replace imports from China (T1);
    • a hub that turns cassava starch into fermentation sugar, so that step stays in Vietnam (T4);
    • shared pilot and tolling fermentation, where firms rent time on someone else's fermenter (T5).

    Fungal foods such as koji (T9), and mung bean and rice protein for plant drinks (T10), are worth proving.

    Seven plays, scored with weights you can change

    Weighted score (1 to 5) by audience preset

    Seven plays, scored with weights you can changeBar chart of weighted scores for seven plays: under balanced weights the cassava-starch-to-sugar hub (3.90) and domestic textured plant protein (3.80) lead, followed by shared pilot and tolling fermentation (3.45), mung bean and rice protein for plant drinks (3.05) and koji and fungal foods (3.00).012345Weighted score, balanced weights (1 to 5)T4 Cassava starch to fermentation sugarhubT4 Cassava starch to fermentation sugar hub: 3.90 (balanced weights, 1 to 5) Must be true: Cassava prices stabilise after the 2025 to 2026 spike and the E10 mandate; at least one fermenter commits to co-locate; wastewater and biogas infrastructure can serve fermentation Kill tests: No fermenter commits within two years; cassava price volatility makes long-term sugar contracts impossible Horizon: 1 to 3 years3.90T1 Domestic textured plant proteinT1 Domestic textured plant protein: 3.80 (balanced weights, 1 to 5) Must be true: Ex-factory cost at or below landed Chinese textured protein (USD 1.24/kg FOB in 2025; USD 1.29 to 1.38/kg landed; 0% duty under ACFTA and RCEP) (TIC-12; BUY-02; FTR-32); a food-grade, low-temperature defatted soy flour supply below the breakeven price (about USD 700 to 940 per t delivered for a 1 t/h line at 70% utilisation; our calculation, TIC-12; TIC-13); buyers value freshness, customisation and local supply Kill tests: Delivered food-grade flour costs more than about USD 700 to 940 per t (1 t/h at 70%) or USD 590 to 900 per t (0.5 t/h at 70%); fewer than three buyers sign conditional orders totalling at least 2,000 t a year; Chinese landed price (USD 1.29 to 1.38/kg in 2025) stays below Vietnamese full cost at 2,800 t a year (0.5 t/h at 70%) or 5,600 t a year (1 t/h at 70%), on 8,000 operating hours a year; India-origin textured soy (USD 0.98/kg FOB in 2025) becomes the reference price; Version 0.10: no bank agrees to finance the supplier's receivables against the anchor buyer (supply-chain finance or factoring); without an anchor the working capital of USD 0.6 to 3.5 million has no named source Horizon: 0 to 2 years3.80T5 Shared pilot and tolling fermentationT5 Shared pilot and tolling fermentation: 3.45 (balanced weights, 1 to 5) Must be true: An anchor host (industrial producer, Saigon Hi-Tech Park, FIRI or a university) and an operator; enough users to fill the line; public or DFI co-funding Kill tests: No host agrees to tolling terms; utilisation below about a third in the first two years Horizon: 2 to 4 years3.45T10 Mung bean and rice protein for plantdrinksT10 Mung bean and rice protein for plant drinks: 3.05 (balanced weights, 1 to 5) Must be true: Functional and sensory parity in drinks; cost near imported pea protein; a buyer specification; a reliable mung bean supply (65% of imports come from Myanmar); Version 0.10: raw material must fit the price. At 2023 to 2025 import prices, mung beans alone cost USD 3.0 to 4.9 per kg of isolate (60 to 80% protein recovery assumed), above Chinese pea protein at USD 2 to 3 per kg, so the line needs a paid starch co-product or a higher-priced job such as egg-white replacement (TPP-05); rice protein from broken rice costs USD 5.1 to 6.6 per kg of product in raw material alone, so the rice arm works only as a co-product of a rice syrup or starch line, and we found no such line in Vietnam (our calculation) Kill tests: Buyer rejects on taste or stability; cost above imported pea protein by more than a local-sourcing premium buyers will pay; Version 0.10: no buyer for the mung bean starch co-product at a price that brings raw-material cost inside the TPP-05 band; for the rice arm, no rice syrup or starch host within two years Horizon: 2 to 4 years3.05T9 Koji and fungal foods on food-grade sidestreamsT9 Koji and fungal foods on food-grade side streams: 3.00 (balanced weights, 1 to 5) Must be true: A characterised strain with no cyclopiazonic acid or 3-nitropropionic acid risk; a food-grade substrate supply; consumer acceptance in chay and mainstream dishes Kill tests: Strain safety data cannot be obtained; sensory trials fail with target buyers Horizon: 1 to 3 years3.00T7 High-value, low-doseprecision-fermentation ingredientsT7 High-value, low-dose precision-fermentation ingredients: 2.10 (balanced weights, 1 to 5) Must be true: A product where titers of 1 to 10 g/L are economic; freedom to operate (OPENPichia or licensed strains); a foreign approval or an ingredient use outside novel-food rules; a clear GMO pathway under Decree 43/2026 Kill tests: No customer letter of intent after a sample programme; contract-manufacturing cost above the target price at pilot scale Horizon: 2 to 5 years2.10T8 Cultivated seafood researchT8 Cultivated seafood research: 1.55 (balanced weights, 1 to 5) Must be true: A food-suitable cell line; research funding over 5 years or more; a partner with scale-up know-how; Version 0.10: the published cost model assumes an engineered cell line with efficient metabolism; that line must exist for a Vietnamese species Kill tests: No serum-free, suspension-capable line after 3 years of work; Version 0.10: no hydrolysate-based medium supports fish-cell growth over several passages within three years Horizon: 5 to 10 years or more1.55Top three under each presetPresetFirstSecondThirdBalancedT4 (3.90)T1 (3.80)T5 (3.45)InvestorsT1 (3.80)T4 (3.50)T5 and T9 (3.00)Policy makersT4 (3.95)T5 (3.85)T1 (3.40)StartupsT1 (4.15)T4 (3.85)T10 (3.25)Food manufacturersT1 (4.15)T4 (3.75)T5 (3.20)Research bodiesT4 (4.15)T5 (4.00)T1 (3.50)International bodies and DFIsT4 (4.10)T5 (3.85)T1 (3.65)

    Bằng chứng Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy trung bình: nguồn thứ cấp uy tín, hoặc đọc một phần nguồn gốc.

    Xem số liệuẨn số liệu
    Seven plays, scored with weights you can change
    RankPlayBalanced score (1 to 5)Horizon
    1T4 Cassava starch to fermentation sugar hub3.901 to 3 years
    2T1 Domestic textured plant protein3.800 to 2 years
    3T5 Shared pilot and tolling fermentation3.452 to 4 years
    4T10 Mung bean and rice protein for plant drinks3.052 to 4 years
    5T9 Koji and fungal foods on food-grade side streams3.001 to 3 years
    6T7 High-value, low-dose precision-fermentation ingredients2.102 to 5 years
    7T8 Cultivated seafood research1.555 to 10 years or more
    Top three under each preset
    PresetFirstSecondThird
    BalancedT4 (3.90)T1 (3.80)T5 (3.45)
    InvestorsT1 (3.80)T4 (3.50)T5 and T9 (3.00)
    Policy makersT4 (3.95)T5 (3.85)T1 (3.40)
    StartupsT1 (4.15)T4 (3.85)T10 (3.25)
    Food manufacturersT1 (4.15)T4 (3.75)T5 (3.20)
    Research bodiesT4 (4.15)T5 (4.00)T1 (3.50)
    International bodies and DFIsT4 (4.10)T5 (3.85)T1 (3.65)
  4. Who would buy it

    Firms, not shoppers. Makers of chay (traditional vegetarian) food, meat processors, plant-milk makers, canteens and dumpling exporters would buy new protein as an ingredient. Branded meat replacements have little pull. On a benchmark path, demand reaches about 19,000 t of protein a year by 2035, or about 0.2% of meat protein (Demand sizing).

    about 19,000 t delivered; about 2,350 t (0.2%) displaced

    Kịch bản tham chiếu, 2035: lượng đạm cung ứng và lượng đạm thịt được thay thế

    2035Tính toán của nhóm nghiên cứuBằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy thấp: một nhận định đơn lẻ, báo chí, tiếp thị của doanh nghiệp hoặc suy ra từ dữ liệu yếu. Hướng tới tương lai: ước tính, tính toán của nhóm từ các giả định đã nêu. Bằng chứng về nhu cầu: suy luận, suy luận hoặc tính toán của nhóm.

    Bối cảnh và nguồn

    Our demand model, benchmark path. Most delivered protein replaces imported ingredients; about 11,000 to 17,000 t of it would be made in Vietnam. The Part IV alternative-protein scenario assumes 39,300 t (3%) displaced in 2035. Our estimate, not a forecast. 1,2,3

    Demand sizing

    What the routes deliver

    Protein delivered by domestic or novel protein, kt a year, by route, scenario and year (our estimates)

    What the routes deliverStacked bars: on the benchmark path, domestic or novel protein delivers about 6 kt of protein in 2030, 19 kt in 2035 and 67 kt in 2050, mostly as import substitution; the stretch path delivers 14, 46 and 172 kt; incumbent drift 1, 3 and 11 kt.R1 Import substitutionR2 Chay occasionsR3 HybridsR4 InstitutionsR5 Household analoguesR6 Plant milksR7 ExportsS-ALT (meat replaced)0306090120150180Protein delivered, kt a yearD-DRIFT 2030, R1 Import substitution: 0.63 kt of protein All routes: 1.29 ktD-DRIFT 2030, R2 Chay occasions: 0.19 kt of protein All routes: 1.29 ktD-DRIFT 2030, R3 Hybrids: 0.03 kt of protein All routes: 1.29 ktD-DRIFT 2030, R4 Institutions: 0.04 kt of protein All routes: 1.29 ktD-DRIFT 2030, R5 Household analogues: 0.06 kt of protein All routes: 1.29 ktD-DRIFT 2030, R6 Plant milks: 0.22 kt of protein All routes: 1.29 ktD-DRIFT 2030, R7 Exports: 0.12 kt of protein All routes: 1.29 kt1.3S-ALT (meat replaced), 2030: 11.79 kt D-DRIFT delivers 1.3 kt2030D-DRIFT 2035, R1 Import substitution: 1.92 kt of protein All routes: 3.25 ktD-DRIFT 2035, R2 Chay occasions: 0.38 kt of protein All routes: 3.25 ktD-DRIFT 2035, R3 Hybrids: 0.07 kt of protein All routes: 3.25 ktD-DRIFT 2035, R4 Institutions: 0.12 kt of protein All routes: 3.25 ktD-DRIFT 2035, R5 Household analogues: 0.11 kt of protein All routes: 3.25 ktD-DRIFT 2035, R6 Plant milks: 0.4 kt of protein All routes: 3.25 ktD-DRIFT 2035, R7 Exports: 0.25 kt of protein All routes: 3.25 kt3.2S-ALT (meat replaced), 2035: 39.28 kt D-DRIFT delivers 3.2 kt2035D-DRIFT 2050, R1 Import substitution: 6.93 kt of protein All routes: 10.68 ktD-DRIFT 2050, R2 Chay occasions: 0.99 kt of protein All routes: 10.68 ktD-DRIFT 2050, R3 Hybrids: 0.14 kt of protein All routes: 10.68 ktD-DRIFT 2050, R4 Institutions: 0.41 kt of protein All routes: 10.68 ktD-DRIFT 2050, R5 Household analogues: 0.26 kt of protein All routes: 10.68 ktD-DRIFT 2050, R6 Plant milks: 1.2 kt of protein All routes: 10.68 ktD-DRIFT 2050, R7 Exports: 0.75 kt of protein All routes: 10.68 kt10.7S-ALT (meat replaced), 2050: 144.6 kt D-DRIFT delivers 10.7 ktS-ALT (meat replaced)2050D-BENCH 2030, R1 Import substitution: 3.48 kt of protein All routes: 5.93 ktD-BENCH 2030, R2 Chay occasions: 0.56 kt of protein All routes: 5.93 ktD-BENCH 2030, R3 Hybrids: 0.24 kt of protein All routes: 5.93 ktD-BENCH 2030, R4 Institutions: 0.22 kt of protein All routes: 5.93 ktD-BENCH 2030, R5 Household analogues: 0.2 kt of protein All routes: 5.93 ktD-BENCH 2030, R6 Plant milks: 0.48 kt of protein All routes: 5.93 ktD-BENCH 2030, R7 Exports: 0.75 kt of protein All routes: 5.93 kt5.9S-ALT (meat replaced), 2030: 11.79 kt D-BENCH delivers 5.9 kt2030D-BENCH 2035, R1 Import substitution: 11.64 kt of protein All routes: 19.06 ktD-BENCH 2035, R2 Chay occasions: 1.6 kt of protein All routes: 19.06 ktD-BENCH 2035, R3 Hybrids: 0.79 kt of protein All routes: 19.06 ktD-BENCH 2035, R4 Institutions: 0.81 kt of protein All routes: 19.06 ktD-BENCH 2035, R5 Household analogues: 0.43 kt of protein All routes: 19.06 ktD-BENCH 2035, R6 Plant milks: 1.29 kt of protein All routes: 19.06 ktD-BENCH 2035, R7 Exports: 2.5 kt of protein All routes: 19.06 kt19.1S-ALT (meat replaced), 2035: 39.28 kt D-BENCH delivers 19.1 kt2035D-BENCH 2050, R1 Import substitution: 44.64 kt of protein All routes: 66.65 ktD-BENCH 2050, R2 Chay occasions: 3.21 kt of protein All routes: 66.65 ktD-BENCH 2050, R3 Hybrids: 2.17 kt of protein All routes: 66.65 ktD-BENCH 2050, R4 Institutions: 2.29 kt of protein All routes: 66.65 ktD-BENCH 2050, R5 Household analogues: 1.05 kt of protein All routes: 66.65 ktD-BENCH 2050, R6 Plant milks: 5.79 kt of protein All routes: 66.65 ktD-BENCH 2050, R7 Exports: 7.5 kt of protein All routes: 66.65 kt66.7S-ALT (meat replaced), 2050: 144.6 kt D-BENCH delivers 66.7 kt2050D-STRETCH 2030, R1 Import substitution: 7.64 kt of protein All routes: 14.13 ktD-STRETCH 2030, R2 Chay occasions: 1.22 kt of protein All routes: 14.13 ktD-STRETCH 2030, R3 Hybrids: 0.88 kt of protein All routes: 14.13 ktD-STRETCH 2030, R4 Institutions: 0.74 kt of protein All routes: 14.13 ktD-STRETCH 2030, R5 Household analogues: 0.59 kt of protein All routes: 14.13 ktD-STRETCH 2030, R6 Plant milks: 1.06 kt of protein All routes: 14.13 ktD-STRETCH 2030, R7 Exports: 2.0 kt of protein All routes: 14.13 kt14.1S-ALT (meat replaced), 2030: 11.79 kt D-STRETCH delivers 14.1 kt2030D-STRETCH 2035, R1 Import substitution: 25.26 kt of protein All routes: 45.62 ktD-STRETCH 2035, R2 Chay occasions: 3.45 kt of protein All routes: 45.62 ktD-STRETCH 2035, R3 Hybrids: 2.95 kt of protein All routes: 45.62 ktD-STRETCH 2035, R4 Institutions: 2.27 kt of protein All routes: 45.62 ktD-STRETCH 2035, R5 Household analogues: 1.28 kt of protein All routes: 45.62 ktD-STRETCH 2035, R6 Plant milks: 2.91 kt of protein All routes: 45.62 ktD-STRETCH 2035, R7 Exports: 7.5 kt of protein All routes: 45.62 kt45.6S-ALT (meat replaced), 2035: 39.28 kt D-STRETCH delivers 45.6 kt2035D-STRETCH 2050, R1 Import substitution: 106.84 kt of protein All routes: 171.51 ktD-STRETCH 2050, R2 Chay occasions: 8.32 kt of protein All routes: 171.51 ktD-STRETCH 2050, R3 Hybrids: 7.59 kt of protein All routes: 171.51 ktD-STRETCH 2050, R4 Institutions: 5.72 kt of protein All routes: 171.51 ktD-STRETCH 2050, R5 Household analogues: 2.63 kt of protein All routes: 171.51 ktD-STRETCH 2050, R6 Plant milks: 15.41 kt of protein All routes: 171.51 ktD-STRETCH 2050, R7 Exports: 25.0 kt of protein All routes: 171.51 kt171.5S-ALT (meat replaced), 2050: 144.6 kt D-STRETCH delivers 171.5 kt2050D-DRIFT (incumbent drift)D-BENCH (benchmark path)D-STRETCH (stretch path)Note: delivered protein is not displaced meat; see the next chart.

    Nguồn 1,2,3,4 Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy thấp: một nhận định đơn lẻ, báo chí, tiếp thị của doanh nghiệp hoặc suy ra từ dữ liệu yếu.

    Xem số liệuẨn số liệu
    What the routes deliver
    Scenario and yearR1 Import substitution (kt)R2 Chay occasions (kt)R3 Hybrids (kt)R4 Institutions (kt)R5 Household analogues (kt)R6 Plant milks (kt)R7 Exports (kt)All routes (kt)S-ALT food assumption (kt)
    D-DRIFT 20300.630.190.030.040.060.220.121.2911.79
    D-DRIFT 20351.920.380.070.120.110.40.253.2539.28
    D-DRIFT 20506.930.990.140.410.261.20.7510.68144.6
    D-BENCH 20303.480.560.240.220.20.480.755.9311.79
    D-BENCH 203511.641.60.790.810.431.292.519.0639.28
    D-BENCH 205044.643.212.172.291.055.797.566.65144.6
    D-STRETCH 20307.641.220.880.740.591.062.014.1311.79
    D-STRETCH 203525.263.452.952.271.282.917.545.6239.28
    D-STRETCH 2050106.848.327.595.722.6315.4125.0171.51144.6
  5. What stands in the way

    There is no route to approve new foods, and we found no open pilot fermentation capacity. Chinese and ASEAN protein ingredients enter without tariffs, and risk capital is scarce (Rules, Industrial base, Business buyers, Capital).

    Plant protein from China nearly tripled

    China-reported exports to Vietnam, tonnes, by customs line, 2019 to 2025

    Plant protein from China nearly tripledStacked bars: plant protein shipped from China to Vietnam rose from 18,375 t in 2019 to 53,976 t in 2025, with protein substances (HS 3504) jumping from 10,028 t in 2023 to 25,267 t in 2024.Protein substances (HS 3504)Textured protein and concentrates (HS 2106 10)Wheat gluten (HS 1109)010,00020,00030,00040,00050,00060,000Tonnes2019, Protein substances (HS 3504): 2,140 t Total: 18,375 t2019, Textured protein and concentrates (HS 2106 10): 6,615 t Total: 18,375 t2019, Wheat gluten (HS 1109): 9,620 t Total: 18,375 t18,375 t20192020, Protein substances (HS 3504): 2,274 t Total: 20,778 t2020, Textured protein and concentrates (HS 2106 10): 9,741 t Total: 20,778 t2020, Wheat gluten (HS 1109): 8,763 t Total: 20,778 t20,77820202021, Protein substances (HS 3504): 2,516 t Total: 19,347 t2021, Textured protein and concentrates (HS 2106 10): 7,151 t Total: 19,347 t2021, Wheat gluten (HS 1109): 9,680 t Total: 19,347 t19,34720212022, Protein substances (HS 3504): 3,935 t Total: 23,203 t2022, Textured protein and concentrates (HS 2106 10): 5,639 t Total: 23,203 t2022, Wheat gluten (HS 1109): 13,629 t Total: 23,203 t23,20320222023, Protein substances (HS 3504): 10,028 t Total: 28,323 t2023, Textured protein and concentrates (HS 2106 10): 7,957 t Total: 28,323 t2023, Wheat gluten (HS 1109): 10,338 t Total: 28,323 t28,32320232024, Protein substances (HS 3504): 25,267 t Total: 51,130 t2024, Textured protein and concentrates (HS 2106 10): 10,029 t Total: 51,130 t2024, Wheat gluten (HS 1109): 15,834 t Total: 51,130 t51,13020242025, Protein substances (HS 3504): 24,363 t Total: 53,976 t2025, Textured protein and concentrates (HS 2106 10): 10,202 t Total: 53,976 t2025, Wheat gluten (HS 1109): 19,411 t Total: 53,976 t53,976 t2025Note: the 2024 jump in HS 3504 isprobably partly feed-grade.

    Nguồn 1 Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy cao: đã đọc nguồn gốc.

    Xem số liệuẨn số liệu
    Plant protein from China nearly tripled
    YearProtein substances (HS 3504) (t)Textured protein and concentrates (HS 2106 10) (t)Wheat gluten (HS 1109) (t)Total (t)
    20192,1406,6159,62018,375
    20202,2749,7418,76320,778
    20212,5167,1519,68019,347
    20223,9355,63913,62923,203
    202310,0287,95710,33828,323
    202425,26710,02915,83451,130
    202524,36310,20219,41153,976
  6. What to do first

    • Add a clause on new foods to the Food Safety Law before the vote considered for May 2027.
    • Clarify the genetically modified organism (GMO) status of purified proteins made by engineered microbes, and set a proportionate novel-food procedure in the implementing decree.
    • Share the expensive parts: a food-grade pilot fermentation line and a protein-quality laboratory.
    • Firms: get delivered quotes for food-grade soy flour, win conditional orders from chay makers, and test plant protein mixed into meat products against the soy recipe buyers already use.

    Policy options by effort and impact

    Ranked options; colour shows tier

    Policy options by effort and impactScatter of 19 ranked options: most tier 1 options sit at low effort and high impact; the pilot fermentation line is high effort and high impact.Tier 1Tier 2Tier 3Low effortLow impactMedium effortMedium impactHigh effortHigh impactRank 12 (tier 3): Low effort, Low impact Option: Adopt the Codex list of priority allergens for food labels; on names, no new rule: no rule reserves thịt, sữa or phô mai and names must only not mislead, so keep this permissive regime in the new Food Safety Law decree and issue guidance (not a rule) on neutral state and media vocabulary for new foods (a 2025 provincial scheme uses thịt nhân tạo for cultivated meat) Owner: MOST (labelling decree owner) with MOH Timeline: 2,02712Rank 15 (tier 3): Low effort, Low impact Option: Publish official grid emission factors for 2024 and 2025, and signal capacity-charge levels of the two-part power tariff early Owner: MAE (Department of Climate Change); MOIT Timeline: 3 to 6 months15Rank 19 (tier 3): Low effort, Low impact Option: Review MFN duties on food protein inputs not made in Vietnam at scale (wheat gluten 10%, protein concentrates 5%), while weighing the effect on local producers of SCP and yeast (2102.20 at 7%) Owner: MOF Timeline: Next MFN amendment19Rank 5 (tier 2): Low effort, Medium impact Option: Publish the closed-production certificate timeline, fee and issuing level, and a model Form 33 for industrial fermentation strains Owner: MAE Department of Nature Conservation and Biodiversity Timeline: 3 months5Rank 6 (tier 2): Low effort, Medium impact Option: Tax guidance confirming that precision-fermentation and biomass-fermentation plants qualify as "new projects applying high technology" (Decision 23/2026 items 60, 92, 100) and for the 3-year first-technology exemption (Decree 320/2025 Art 4(4)(b)); clarify the 30% agri raw-material test for fermentation on agricultural sugars or cassava Owner: MOF (Tax Department); MOST Timeline: 3 to 6 months6Rank 9 (tier 2): Low effort, Medium impact Option: Name alternative protein and biomanufacturing as priorities in the rules of the national, Ho Chi Minh City and Hanoi venture funds, and add a first-loss window for pre-revenue pilot plants Owner: MOF; MOST; Ho Chi Minh City and Hanoi People's Committees Timeline: 6 to 12 months9Rank 11 (tier 3): Low effort, Medium impact Option: Publish a yearly national by-product atlas (volume, location, price, composition of cassava pulp, bran, okara, spent grain, seafood by-products) Owner: MAE with the National Statistics Office; provincial departments Timeline: 12 months11Rank 13 (tier 3): Low effort, Medium impact Option: Name alternative and novel protein ingredients for food and feed explicitly in biotech programmes (Resolution 189/NQ-CP tasks, Decision 429 follow-up) and the next high-tech product list update Owner: MOST; MAE; MOH13Rank 14 (tier 3): Low effort, Medium impact Option: Transition rules in the new Food Safety Law: grandfather existing self-declarations and set realistic deadlines Owner: MOH/VFA14Rank 16 (tier 3): Low effort, Medium impact Option: Standard licence and benefit-sharing terms for strains supplied by public microbial culture collections (Decree 43/2026 already delivered a simplified access procedure and a standard contract form, Form 12) Owner: MAE (biodiversity); provincial People's Committees (access licences); VAST, VNU, FIRI collections16Rank 17 (tier 3): Low effort, Medium impact Option: Negotiate pilot-plant access and regulatory cooperation with Thailand and Singapore (EECi access for Vietnamese firms; observer role in regional novel-food and cultivated-food safety work) Owner: MOST; MOH; Ministry of Foreign Affairs Timeline: 12 months17Rank 1 (tier 1): Low effort, High impact Option: Put a short delegation clause in the new Food Safety Law: "The Government shall regulate the management of new foods and new food ingredients" (thực phẩm mới, nguyên liệu thực phẩm mới), with a definition based on history of use. This keeps the law short and lets the decree carry the detail Owner: MOH/VFA (drafter); National Assembly Committee on Science, Technology and Environment (review) Timeline: First comments at the October 2026 National Assembly session; passage considered at the May 2027 session; the window for the clause runs into early 20271Rank 2 (tier 1): Low effort, High impact Option: Joint MAE and MOH guidance on precision-fermented proteins: no GMO food-use certificate is needed for a purified protein made in contained use with no viable GMO or recombinant DNA in the product; and state whether a recombinant protein identical to the native protein is a detectable product of the modified gene (Joint Circular 45/2015 Art 1(2)(a)), which decides whether the GM label applies above 5% of ingredients, and whether a product with such a protein below 5% may still claim non-GMO; the note could name the evidence a firm must show: no viable production cells, no recombinant DNA and no antibiotic-resistance markers in the product Owner: MAE Department of Nature Conservation and Biodiversity and Department of Science and Technology; MOH/VFA Timeline: 3 to 6 months2Rank 4 (tier 2): Medium effort, Medium impact Option: Use the Hanoi sandbox for one supervised novel-food pilot (e.g. precision-fermented dairy protein or cultivated seafood tasting), with MOH in the appraisal. In parallel, a Government decree under Law 93/2025 Art 21(6) naming MOH as the approving agency for food tests Owner: Hanoi People's Committee and Hanoi Department of Science and Technology; MOST; MOH Timeline: 6 to 12 months (Hanoi); 12 to 24 months (national)4Rank 10 (tier 3): Medium effort, Medium impact Option: Accredit amino-acid profiling and protein-digestibility (DIAAS or PDCAAS) testing at a national reference laboratory, and extend it to protein function: a standard functional panel (solubility, water and oil holding, emulsifying, foaming, gelling) on Embrapa's 2022 open methods, buyer tests (gel strength in a sausage model, UHT stability) and an open dataset for mung bean, broken rice, rice bran, peanut, sesame and okara linked to the by-product atlas (P5) Owner: MOH (NIFC); MOIT (FIRI NACEFA); candidate hosts for the functional panel: HUST, Can Tho University, Nguyen Tat Thanh University Timeline: 12 to 24 months10Rank 3 (tier 1): Medium effort, High impact Option: Novel-food procedure in the implementing decree of the new law: (a) a "not novel" determination on request; (b) a traditional-food notification lane for foods with long safe use abroad, with a fixed objection window; (c) a reliance lane for foods approved by 2 or more reference regulators (Singapore, US, EU, Australia/New Zealand, Thailand); (d) full assessment by recognised assessment units, with a public positive list Owner: MOH/VFA; Government Timeline: Drafting from 2026; in force 2027 to 2028 (after the law is voted at the May 2027 session)3Rank 8 (tier 2): Medium effort, High impact Option: Shift biomanufacturing incentives towards capital grants, loan guarantees and feedstock-cost support rather than further tax holidays Owner: MOF; MOST; MPI functions now in MOF Timeline: 12 to 24 months8Rank 18 (tier 3): High effort, Medium impact Option: Vietnam to co-lead ASEAN work on common principles and reliance for novel foods Owner: MOH; MOST (ASEAN ACCSQ)18Rank 7 (tier 2): High effort, High impact Option: Shared food-grade pilot fermentation line plus a tolling programme on existing industrial tanks, open to startups and researchers Owner: MOST; MAE; Ho Chi Minh City People's Committee; Saigon Hi-Tech Park or FIRI as host Timeline: 24 to 36 months7

    Bằng chứng Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy trung bình: nguồn thứ cấp uy tín, hoặc đọc một phần nguồn gốc.

    Xem số liệuẨn số liệu
    Policy options by effort and impact
    RankTierOptionEffortImpactOwnerTimeline
    11Put a short delegation clause in the new Food Safety Law: "The Government shall regulate the management of new foods and new food ingredients" (thực phẩm mới, nguyên liệu thực phẩm mới), with a definition based on history of use. This keeps the law short and lets the decree carry the detailLowHighMOH/VFA (drafter); National Assembly Committee on Science, Technology and Environment (review)First comments at the October 2026 National Assembly session; passage considered at the May 2027 session; the window for the clause runs into early 2027
    21Joint MAE and MOH guidance on precision-fermented proteins: no GMO food-use certificate is needed for a purified protein made in contained use with no viable GMO or recombinant DNA in the product; and state whether a recombinant protein identical to the native protein is a detectable product of the modified gene (Joint Circular 45/2015 Art 1(2)(a)), which decides whether the GM label applies above 5% of ingredients, and whether a product with such a protein below 5% may still claim non-GMO; the note could name the evidence a firm must show: no viable production cells, no recombinant DNA and no antibiotic-resistance markers in the productLowHighMAE Department of Nature Conservation and Biodiversity and Department of Science and Technology; MOH/VFA3 to 6 months
    31Novel-food procedure in the implementing decree of the new law: (a) a "not novel" determination on request; (b) a traditional-food notification lane for foods with long safe use abroad, with a fixed objection window; (c) a reliance lane for foods approved by 2 or more reference regulators (Singapore, US, EU, Australia/New Zealand, Thailand); (d) full assessment by recognised assessment units, with a public positive listMediumHighMOH/VFA; GovernmentDrafting from 2026; in force 2027 to 2028 (after the law is voted at the May 2027 session)
    42Use the Hanoi sandbox for one supervised novel-food pilot (e.g. precision-fermented dairy protein or cultivated seafood tasting), with MOH in the appraisal. In parallel, a Government decree under Law 93/2025 Art 21(6) naming MOH as the approving agency for food testsMediumMediumHanoi People's Committee and Hanoi Department of Science and Technology; MOST; MOH6 to 12 months (Hanoi); 12 to 24 months (national)
    52Publish the closed-production certificate timeline, fee and issuing level, and a model Form 33 for industrial fermentation strainsLowMediumMAE Department of Nature Conservation and Biodiversity3 months
    62Tax guidance confirming that precision-fermentation and biomass-fermentation plants qualify as "new projects applying high technology" (Decision 23/2026 items 60, 92, 100) and for the 3-year first-technology exemption (Decree 320/2025 Art 4(4)(b)); clarify the 30% agri raw-material test for fermentation on agricultural sugars or cassavaLowMediumMOF (Tax Department); MOST3 to 6 months
    72Shared food-grade pilot fermentation line plus a tolling programme on existing industrial tanks, open to startups and researchersHighHighMOST; MAE; Ho Chi Minh City People's Committee; Saigon Hi-Tech Park or FIRI as host24 to 36 months
    82Shift biomanufacturing incentives towards capital grants, loan guarantees and feedstock-cost support rather than further tax holidaysMediumHighMOF; MOST; MPI functions now in MOF12 to 24 months
    92Name alternative protein and biomanufacturing as priorities in the rules of the national, Ho Chi Minh City and Hanoi venture funds, and add a first-loss window for pre-revenue pilot plantsLowMediumMOF; MOST; Ho Chi Minh City and Hanoi People's Committees6 to 12 months
    103Accredit amino-acid profiling and protein-digestibility (DIAAS or PDCAAS) testing at a national reference laboratory, and extend it to protein function: a standard functional panel (solubility, water and oil holding, emulsifying, foaming, gelling) on Embrapa's 2022 open methods, buyer tests (gel strength in a sausage model, UHT stability) and an open dataset for mung bean, broken rice, rice bran, peanut, sesame and okara linked to the by-product atlas (P5)MediumMediumMOH (NIFC); MOIT (FIRI NACEFA); candidate hosts for the functional panel: HUST, Can Tho University, Nguyen Tat Thanh University12 to 24 months
    113Publish a yearly national by-product atlas (volume, location, price, composition of cassava pulp, bran, okara, spent grain, seafood by-products)LowMediumMAE with the National Statistics Office; provincial departments12 months
    123Adopt the Codex list of priority allergens for food labels; on names, no new rule: no rule reserves thịt, sữa or phô mai and names must only not mislead, so keep this permissive regime in the new Food Safety Law decree and issue guidance (not a rule) on neutral state and media vocabulary for new foods (a 2025 provincial scheme uses thịt nhân tạo for cultivated meat)LowLowMOST (labelling decree owner) with MOH2027
    133Name alternative and novel protein ingredients for food and feed explicitly in biotech programmes (Resolution 189/NQ-CP tasks, Decision 429 follow-up) and the next high-tech product list updateLowMediumMOST; MAE; MOH
    143Transition rules in the new Food Safety Law: grandfather existing self-declarations and set realistic deadlinesLowMediumMOH/VFA
    153Publish official grid emission factors for 2024 and 2025, and signal capacity-charge levels of the two-part power tariff earlyLowLowMAE (Department of Climate Change); MOIT3 to 6 months
    163Standard licence and benefit-sharing terms for strains supplied by public microbial culture collections (Decree 43/2026 already delivered a simplified access procedure and a standard contract form, Form 12)LowMediumMAE (biodiversity); provincial People's Committees (access licences); VAST, VNU, FIRI collections
    173Negotiate pilot-plant access and regulatory cooperation with Thailand and Singapore (EECi access for Vietnamese firms; observer role in regional novel-food and cultivated-food safety work)LowMediumMOST; MOH; Ministry of Foreign Affairs12 months
    183Vietnam to co-lead ASEAN work on common principles and reliance for novel foodsHighMediumMOH; MOST (ASEAN ACCSQ)
    193Review MFN duties on food protein inputs not made in Vietnam at scale (wheat gluten 10%, protein concentrates 5%), while weighing the effect on local producers of SCP and yeast (2102.20 at 7%)LowLowMOFNext MFN amendment
  7. By 2050

    On trend, the meat protein Vietnamese people eat rises from about 1.01 Mt (million tonnes) in 2025 to about 1.45 Mt in 2050 (our model). The imported soybean meal behind it rises from 7.2 Mt to about 10.4 Mt. Seven moves pay off in every scenario we tested (Protein balance 2050, Moves for any 2050).

    10.4 Mt (+44% on 2025)

    Nhu cầu khô đậu tương năm 2050 theo xu hướng

    2025 baseTính toán của nhóm nghiên cứuBằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy thấp: một nhận định đơn lẻ, báo chí, tiếp thị của doanh nghiệp hoặc suy ra từ dữ liệu yếu. Hướng tới tương lai: ước tính, tính toán của nhóm từ các giả định đã nêu.

    Bối cảnh và nguồn

    Our balance model on trend diets and trend efficiency: 7.2 Mt (2025), 8.5 (2030), 10.0 (2040), 10.4 (2050), nearly all imported. Published projections stop at 2035. 1,2

    Protein balance 2050

Đọc theo vai trò của bạn

Mỗi bản tóm lược nói nghiên cứu có ý nghĩa gì với một nhóm độc giả. Chọn vai trò của bạn để xem lộ trình đọc gợi ý.

Chúng tôi chắc chắn đến đâu?

Mỗi nhận định trong báo cáo có một dấu cho loại bằng chứng và mức độ tin cậy. Trên màn hình rộng, các dấu nằm ở lề cạnh nhận định; chọn một con số để xem nguồn.

Bằng chứng đến từ đâu

Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy cao: đã đọc nguồn gốc.
bằng chứng về Việt Nam (VN-direct)
Bằng chứng: bằng chứng từ một nước tương đồng, có nêu giả định chuyển đổi (VN-adjacent); độ tin cậy cao: đã đọc nguồn gốc.
bằng chứng từ một nước tương đồng, có nêu giả định chuyển đổi (VN-adjacent)
Bằng chứng: bằng chứng chung, không riêng Việt Nam (general); độ tin cậy cao: đã đọc nguồn gốc.
bằng chứng chung, không riêng Việt Nam (general)

Độ tin cậy

Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy cao: đã đọc nguồn gốc.
độ tin cậy cao: đã đọc nguồn gốc
Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy trung bình: nguồn thứ cấp uy tín, hoặc đọc một phần nguồn gốc.
độ tin cậy trung bình: nguồn thứ cấp uy tín, hoặc đọc một phần nguồn gốc
Bằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy thấp: một nhận định đơn lẻ, báo chí, tiếp thị của doanh nghiệp hoặc suy ra từ dữ liệu yếu.
độ tin cậy thấp: một nhận định đơn lẻ, báo chí, tiếp thị của doanh nghiệp hoặc suy ra từ dữ liệu yếu

Nhận định về tương lai

Hướng tới tương lai: xu hướng, thay đổi đã quan sát được và dự kiến tiếp tục.
thay đổi đã quan sát được và dự kiến tiếp tục
Hướng tới tương lai: dự phóng, kết quả mô hình đã công bố hoặc mục tiêu chính thức.
kết quả mô hình đã công bố hoặc mục tiêu chính thức
Hướng tới tương lai: ước tính, tính toán của nhóm từ các giả định đã nêu.
tính toán của nhóm từ các giả định đã nêu
Hướng tới tương lai: tín hiệu, diễn biến mới nổi, có thể lan rộng hoặc không.
diễn biến mới nổi, có thể lan rộng hoặc không
Hướng tới tương lai: biến cố hiếm, biến cố xác suất thấp, tác động lớn.
biến cố xác suất thấp, tác động lớn
Hướng tới tương lai: tầm nhìn, chuẩn tắc: điều chúng tôi cho là nên xảy ra, không phải dự báo.
chuẩn tắc: điều chúng tôi cho là nên xảy ra, không phải dự báo

Nhận định về nhu cầu

Bằng chứng về nhu cầu: được nêu, điều người ta nói: khảo sát, thảo luận nhóm, ý định.
điều người ta nói: khảo sát, thảo luận nhóm, ý định
Bằng chứng về nhu cầu: bộc lộ, điều người ta hoặc doanh nghiệp làm: doanh số, giá, thương mại, thực đơn, kệ hàng.
điều người ta hoặc doanh nghiệp làm: doanh số, giá, thương mại, thực đơn, kệ hàng
Bằng chứng về nhu cầu: được thử, nếm thử, đấu giá bằng tiền thật, thí nghiệm lựa chọn hoặc thử nghiệm thực địa.
nếm thử, đấu giá bằng tiền thật, thí nghiệm lựa chọn hoặc thử nghiệm thực địa
Bằng chứng về nhu cầu: suy luận, suy luận hoặc tính toán của nhóm.
suy luận hoặc tính toán của nhóm

Giải thích đầy đủ, kèm bảng mã và tên kịch bản

2,669

Số nguồn tài liệu đã ghi nhận

Oct 2026Tính toán của nhóm nghiên cứuBằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy cao: đã đọc nguồn gốc.

Bối cảnh và nguồn

Our count of sources cited across thirteen research rounds, including press, peer-reviewed papers, company pages, statistics and legal texts. Every source is in the source list.

R4. Sources

170

Số giải pháp đã công bố được sàng lọc cho Việt Nam

Oct 2026Tính toán của nhóm nghiên cứuBằng chứng: bằng chứng về Việt Nam (VN-direct); độ tin cậy trung bình: nguồn thứ cấp uy tín, hoặc đọc một phần nguồn gốc.

Bối cảnh và nguồn

Our count (October 2026): the 112 entries of the Good Food Institute's Solutions Database and 58 publications, research outputs and grants from three other sources. Three candidates are listed but not ranked; none displaces the leading plays. 1

Về nghiên cứu này