Chương 11 · Phần III
11. Ten plays and six public goods
We turn the evidence into ten concrete plays, each with what must be true, the tests that would kill it, first customers and a time horizon, and score them on seven criteria with weights that each audience can change. Domestic textured plant protein, functional microbial feed ingredients and a cassava-to-fermentation-sugar hub lead under almost every weighting. Shared pilot and tolling capacity is the enabler most of the others need. Bulk fishmeal replacement is marginal, precision fermentation fits only as contract-made niche products, and cultivated seafood is research. Six public goods would lower the cost of every play.
Nội dung trang này bằng tiếng Anh. Đọc bản tóm tắt tiếng Việt
Trong trang này
In one paragraph. A "play" here is a specific business or programme that someone could start in Vietnam, with a named product, buyer and partner. We describe ten, from a textured-protein line that could sell within two years to cultivated-seafood research that will not reach a plate this decade. Each play states what must be true for it to work and the tests that would kill it, so that readers can check it rather than take it on trust. We score the plays on seven criteria and publish the weights, so each audience can re-rank them. Three plays lead under almost any weighting: domestic textured plant protein (T1), functional microbial feed ingredients for shrimp and pangasius (T2) and a cassava-starch-to-fermentation-sugar hub (T4). Shared pilot and tolling fermentation (T5) is what most fermentation plays need first. Six public goods, mainly rules, trial capacity and data, would lower the cost of all of them.
11.1 How the plays are scored
| Criterion | What a 5 means | What a 1 means |
|---|---|---|
| Market pull | Several named buyers already buy the imported or incumbent equivalent | No identifiable buyer or need |
| Vietnam advantage | A clear, hard-to-copy Vietnamese edge | Any neighbour does it as well or better |
| Technical readiness | Commercial elsewhere, skills present in Vietnam | Laboratory stage with major unknowns |
| Route to market | Ordinary food or listed feed rules apply | No legal route exists |
| Capital efficiency | First revenue for under a few USD millions, or via tolling | Tens to hundreds of USD millions before revenue |
| Defensibility | Function, logistics or IP that imports cannot easily match | Commodity with cheaper imports |
| Public value | Large contribution to protein security or national capability | Little wider benefit |
Weight presets (percent; each row sums to 100):
| Preset | Market | Advantage | Readiness | Route | Capital | Defensibility | Public value |
|---|---|---|---|---|---|---|---|
| Balanced | 15 | 15 | 15 | 15 | 15 | 10 | 15 |
| Investors | 25 | 10 | 15 | 10 | 20 | 15 | 5 |
| Policy makers | 10 | 20 | 10 | 5 | 10 | 10 | 35 |
| Startups | 20 | 10 | 20 | 20 | 20 | 5 | 5 |
| Manufacturers | 30 | 10 | 25 | 15 | 10 | 5 | 5 |
| Research bodies | 10 | 25 | 10 | 10 | 5 | 5 | 35 |
| International bodies and DFIs | 10 | 20 | 15 | 10 | 10 | 5 | 30 |
11.2 The scores
| Play | Market | Advantage | Readiness | Route | Capital | Defensibility | Public value | Balanced score |
|---|---|---|---|---|---|---|---|---|
| T1 Domestic textured plant protein | 4 | 3 | 5 | 5 | 4 | 2 | 3 | 3.80 |
| T2 Functional microbial feed ingredients | 4 | 4 | 3 | 3 | 3 | 3 | 4 | 3.45 |
| T3 Bulk microbial protein for fishmeal replacement | 3 | 3 | 3 | 3 | 1 | 1 | 4 | 2.65 |
| T4 Cassava starch to fermentation sugar hub | 2 | 5 | 5 | 5 | 3 | 3 | 4 | 3.90 |
| T5 Shared pilot and tolling fermentation | 2 | 4 | 4 | 4 | 2 | 3 | 5 | 3.45 |
| T6 Duckweed on aquaculture effluent for feed | 2 | 4 | 3 | 4 | 4 | 3 | 3 | 3.30 |
| T7 High-value, low-dose precision fermentation | 2 | 2 | 2 | 1 | 3 | 3 | 2 | 2.10 |
| T8 Cultivated seafood research | 1 | 2 | 1 | 1 | 2 | 2 | 2 | 1.55 |
| T9 Koji and fungal foods on side streams | 2 | 3 | 3 | 4 | 4 | 3 | 2 | 3.00 |
| T10 Mung bean and rice protein for plant drinks | 3 | 2 | 3 | 5 | 3 | 2 | 3 | 3.05 |
Source: AltProtein Vietnam scoring (Technology fit and the evidence cited in each play) .
Top three under each preset.
| Preset | First | Second | Third |
|---|---|---|---|
| Balanced | T4 (3.90) | T1 (3.80) | T2 and T5 (3.45) |
| Investors | T1 (3.80) | T4 (3.50) | T2 (3.40) |
| Policy makers | T4 (3.95) | T5 (3.85) | T2 (3.65) |
| Startups | T1 (4.15) | T4 (3.85) | T2 (3.35) |
| Manufacturers | T1 (4.15) | T4 (3.75) | T2 (3.45) |
| Research bodies | T4 (4.15) | T5 (4.00) | T2 (3.70) |
| International bodies and DFIs | T4 (4.10) | T5 (3.85) | T1 (3.65) |
Reading. T1, T2 and T4 are robust: they stay in the top three or four under every preset. T4 scores highly because it carries little technical or legal risk and uses Vietnam's clearest advantage, but its weak point is market pull: no fermenter has yet committed to buy the sugar. T5 rises when public value is weighted, which is the case for public or development money. T7 and T8 stay at the bottom under every preset; that does not mean they are worthless, only that they are not near-term plays for Vietnam.
11.3 The plays
T1. Domestic textured plant protein
- What. Dry textured soy protein and soy-gluten granules and slices, made in Vietnam for chay brands, frozen-food makers and co-packers. High-moisture whole cuts come later.
- Why Vietnam. Every audited Vietnamese product with 10 g of protein or more per 100 g runs on imported textured soy, soy protein or gluten, mostly Chinese; we found no commercial extrusion in Vietnam; ordinary food rules apply 1,2,3 .
- Must be true. Ex-factory cost at or below landed Chinese textured protein (about USD 1.6 per kg before freight and duty in 2023); buyers value freshness, customisation and local supply; a food-grade soy flour or concentrate supply.
- Kill tests. Three or more named buyers decline a trial order at a price that covers cost; Chinese landed price stays below Vietnamese ex-factory cost at 3,000 t a year.
- First customers. An Nhien, Au Lac, Xuan Hong, LC Foods, Cholimex, CJ Foods Vietnam, Vissan, Thoai An, GN Foods and other co-packers 1 .
- Partners. Extrusion-line suppliers; a soy crusher for a food-grade flake stream 4; FIRI for pilot trials.
- Horizon and capital. 0 to 2 years; indicative equipment USD 1.5 to 4 million for a 1 t/h line (assumption; see Economics).
T2. Functional microbial feed ingredients for shrimp and pangasius
- What. Yeast, fungal or bacterial biomass used at 1 to 5% of shrimp or fish diets and sold on survival, feed efficiency or palatability, made on Vietnamese carbohydrate.
- Why Vietnam. Abundant carbon; aerobic fermentation skills in amino-acid plants; a concentrated shrimp-feed industry; inactive yeasts already on the livestock feed list 5,6 ; trial evidence of functional benefits 7,8 .
- Must be true. Repeatable species-specific trials show at least a 3 to 5% gain in survival or feed conversion at 1 to 5% inclusion; a legal route for aquafeed use; cost below the functional willingness to pay (about USD 1,000 to 4,250 per tonne in total for pangasius; USD 1,280 to 4,800 per tonne above fishmeal parity for shrimp; Economics).
- Kill tests. Two independent tank or pond trials show no gain at 5% inclusion or less; Chinese by-product biomass 9 delivers the same function for less.
- First customers. Sheng Long, CP Vietnam, Uni-President Vietnam, Tongwei, Grobest, Gromax; integrated pangasius groups such as Vinh Hoan 10,11 .
- Partners. Amino-acid producers for tolling or co-product streams; shrimp trial laboratories; large feed groups for distribution.
- Horizon and capital. 1 to 3 years; low if tolled, otherwise in the fermentation capex range.
T3. Bulk microbial protein for fishmeal replacement
- What. Fungal or bacterial protein at 10,000 t a year or more, replacing 25 to 50% of fishmeal in shrimp feed.
- Why Vietnam. The displaceable fishmeal pool is about 120 to 245 kt a year on central assumptions, of which about 25 to 90 kt is realistically contestable 10,12,13 .
- Must be true. Capex below about USD 6,000 per annual tonne; near-free carbon; fishmeal near its 2026 high; a price below Chinese by-product and gas-fermentation imports.
- Kill tests. After a pilot, measured cost stays above 1.2 times fishmeal-protein parity at 2025 prices; no mill signs an offtake at 70 to 90% of fishmeal-protein parity.
- First customers. As T2.
- Horizon and capital. 3 to 6 years; tens of USD millions; a candidate for development-finance first-loss capital, as in the Entobel template (Capital).
T4. Cassava starch to fermentation sugar hub
- What. Turn part of the cassava starch Vietnam exports into fermentation-grade glucose next to starch factories, and co-locate fermenters there.
- Why Vietnam. Vietnam shipped 2.15 Mt of cassava starch in 2023, 93.5% to China, while China imported only USD 0.16 million of glucose syrup from Vietnam in 2024 14,15 . Starch factories cluster in Tay Ninh and the Central Highlands (Asset map).
- Must be true. Cassava prices stabilise after the 2025 to 2026 spike and the E10 fuel-ethanol mandate 16,17 ; at least one fermenter commits to co-locate; wastewater and biogas infrastructure can serve fermentation.
- Kill tests. No fermenter commits within two years; cassava volatility rules out long-term sugar contracts.
- First customers. Domestic fermenters; foreign biomass or precision-fermentation firms looking for a lower-cost ASEAN site (Region).
- Horizon and capital. 1 to 3 years; capital not estimated (standard starch-industry equipment; Vedan already built about 140,000 t a year of glucose capacity 18 ).
T5. Shared pilot and tolling fermentation
- What. A food-grade pilot line (hundreds of litres to a few cubic metres) plus negotiated tolling on existing industrial tanks, open to startups and researchers.
- Why Vietnam. Aerobic tanks exist inside Vedan, Ajinomoto and Daesang, but we found no published tolling offer and no open pilot fermenter (Industrial base) 19,20 .
- Must be true. An anchor host and an operator; enough users; public or development co-funding.
- Kill tests. No host accepts tolling terms; use below about a third of capacity in the first two years.
- First customers. Teams working on T2, T7 and T9; research groups; foreign firms testing Vietnamese feedstocks.
- Partners. Saigon Hi-Tech Park 21, FIRI, the amino-acid producers, and an access arrangement with Thailand's EECi pilot plant 22.
- Horizon and capital. 2 to 4 years; pilot-scale capital, not estimated.
T6. Duckweed on aquaculture effluent for feed
- What. Duckweed grown on pond or processing effluent and used in fish, poultry or pig feed; a food protein concentrate later.
- Why Vietnam. Duckweed is a permitted livestock feed material 5,6 ; there is a Vietnamese collection of more than 100 wild strains 23 ; the climate suits year-round growth.
- Must be true. Yields of 10 to 30 t of dry matter per hectare per year on Vietnamese sites 24,25; heavy metals and manganese within limits 26,27; cheap harvesting and drying.
- Kill tests. Field yields below 10 t per hectare per year; contaminants above feed limits on typical effluent.
- First customers. Integrated pangasius and tilapia farms; small feed mills; livestock farms.
- Horizon and capital. 1 to 3 years; low.
T7. High-value, low-dose precision-fermentation ingredients
- What. Enzymes, heme or sweet proteins and specialty functional proteins developed by Vietnamese teams and made under contract abroad until a domestic option exists.
- Why Vietnam. Vietnamese groups already express proteins in yeast (Knowledge and talent); low-dose products escape the bulk-protein titer benchmark 28,29 .
- Must be true. A product that is economic at 1 to 10 g/L; freedom to operate 30; a foreign approval or an ingredient use outside novel-food rules; a clear GMO pathway (Rules).
- Kill tests. No customer letter of intent after a sample programme; contract-manufacturing cost above target price at pilot scale.
- Partners. ScaleUp Bio, EECi or Chinese CDMOs 31,22,32.
- Horizon and capital. 2 to 5 years; asset-light at first.
T8. Cultivated seafood research
- What. Serum-free adaptation of striped catfish and shrimp cells, growth-factor expression and regulatory preparation, with foreign partners.
- Why Vietnam. Vietnam exports about USD 2.2 billion of pangasius a year (Why Vietnam); Vietnamese groups express growth factors and culture stem cells (Knowledge and talent); Vinh Hoan has invested in regional cultivated-seafood firms 33 .
- Must be true. A food-suitable cell line (none exists 34,35 ); five or more years of research funding; a scale-up partner.
- Kill tests. No serum-free, suspension-capable line after three years.
- Horizon and capital. 5 to 10 years or more; research grants.
T9. Koji and fungal foods on food-grade side streams
- What. Solid-state fermented foods and ingredients using characterised Aspergillus oryzae or other food moulds on okara, broken rice or brewer's spent grain.
- Why Vietnam. A tương tradition; local gene tools for A. oryzae 36,37 ; koji on food substrates goes through ordinary self-declaration (Rules).
- Must be true. A characterised strain without cyclopiazonic acid or 3-nitropropionic acid risk 38 ; a food-grade substrate supply; acceptance in chay and mainstream dishes.
- Kill tests. Strain safety data cannot be obtained; sensory trials fail with target buyers.
- Horizon and capital. 1 to 3 years; low.
T10. Mung bean and rice protein for plant drinks
- What. Concentrates or isolates from mung bean and rice, replacing imported pea protein in high-protein plant drinks and yogurts.
- Why Vietnam. A large plant-milk industry whose high-protein lines use pea protein, for which no Vietnamese producer was found 1 ; ordinary food rules.
- Must be true. Functional and sensory parity; cost near imported pea protein; a buyer specification; reliable mung bean supply (65% of imports from Myanmar 39); honest protein-quality claims (human mung bean digestibility is about 20% below pig values 40 ).
- Kill tests. Buyer rejects on taste or stability; cost exceeds imported pea protein by more than buyers will pay for local sourcing.
- First customers. Vinamilk, TH and other plant-milk makers.
- Horizon and capital. 2 to 4 years; moderate, not estimated.
When each play could deliver
Years from now to first commercial result
Bằng chứng:
Xem số liệuẨn số liệu
| Play | Name | Segment | Horizon | Start (years) | End (years) |
|---|---|---|---|---|---|
| T1 | Domestic textured plant protein | food ingredient | 0 to 2 years | 0 | 2 |
| T2 | Functional microbial feed ingredients for shrimp and pangasius | feed ingredient | 1 to 3 years | 1 | 3 |
| T4 | Cassava starch to fermentation sugar hub | enabler | 1 to 3 years | 1 | 3 |
| T6 | Duckweed on aquaculture effluent for feed | feed ingredient | 1 to 3 years | 1 | 3 |
| T9 | Koji and fungal foods on food-grade side streams | food product | 1 to 3 years | 1 | 3 |
| T5 | Shared pilot and tolling fermentation | enabler | 2 to 4 years | 2 | 4 |
| T10 | Mung bean and rice protein for plant drinks | food ingredient | 2 to 4 years | 2 | 4 |
| T7 | High-value, low-dose precision-fermentation ingredients | food ingredient | 2 to 5 years | 2 | 5 |
| T3 | Bulk microbial protein for fishmeal replacement | feed ingredient | 3 to 6 years | 3 | 6 |
| T8 | Cultivated seafood research | research | 5 to 10 years or more | 5 | 10 |
11.4 Six public goods
These are not businesses. They lower the cost and risk of every play above, and most are cheap. Policy options sets out who would do each.
| ID | Public good | What it unlocks | Policy options |
|---|---|---|---|
| P1 | Proportionate new-food procedure | T7, T9, and food uses of T6 and fungal biomass | PO-015, PO-019 |
| P2 | Aquafeed raw-material clarity and generic microbial-biomass entries | T2, T3, T6 | PO-017, PO-018 |
| P3 | Public feed-trial capacity with standard protocols | T2, T3, T6 | PO-006, PO-021 |
| P4 | Protein-quality laboratory (amino acids, DIAAS) | T1, T9, T10, and any novel-food dossier | (Policy options) |
| P5 | Yearly by-product atlas (volume, location, price, composition) | T3, T4, T6, T9 | (Policy options) |
| P6 | Open cost model and public feed-ingredient price series | All fermentation plays; investor screening | (Policy options) |
11.5 Sequencing: a portfolio view
- Now to 2028. T1 and T2 as businesses; T4 if a fermenter commits; T9 and T6 as small pilots; P1 and P2 through the Food Safety Law and the next feed-list updates; T5 designed and funded.
- 2028 to 2031. T5 in operation; T10 if a drinks buyer writes a specification; T7 through contract manufacturing; T3 only if T2 and T5 have proved the process and fishmeal stays high.
- 2031 onward. T8 if a food-suitable cell line exists and a regional approval route is working.
11.6 What we do not recommend, and why
- Replacing soybean meal in pig and poultry feed with fermentation protein. The price gap is 4.6 to 5.1 times at the low end of our cost range (Economics).
- Premium imported-style consumer analogues as a first business. They sell at an order of magnitude above chicken and egg protein and appear only in a few premium stores (Alt protein today, Economics).
- Commodity precision-fermented dairy or egg proteins made in Vietnam. Titers, rules and missing infrastructure all point the other way (Technology fit).
- Cultivated-meat production. No cell line, no route, and costs 7 to 15 times retail meat (Technology fit).
- Insect protein. Out of this report's scope as a play; an established incumbent and benchmark (Alt protein today).
11.7 What this means
- Investors: T1 and T2 are the investable near-term theses; T3 and T4 need a strategic partner and development capital. Use the kill tests as diligence questions.
- Startups: T1, T2, T6 and T9 can reach revenue on today's rules. Build the trial data (T2) or the buyer specification (T1, T10) before the plant.
- Manufacturers: you are the first customer for T1, T2 and T10. A letter of intent or a joint trial is the cheapest way to start one of them.
- Policy makers: P1 and P2 cost little and unlock several plays; T4 and T5 are where provincial and national investment promotion can act.
- Research bodies: T6, T8, T9 and the public goods P3 to P6 are research-led.
- International bodies: T3, T4 and T5 fit development finance and technical assistance.
Related: plays.csv (all fields), Policy options, Investors, Startups, Manufacturers.
Nguồn được trích dẫn trong trang
- FORM-01 AltProtein Vietnam. AltProtein Vietnam retail field audit, Nha Trang 6 Sep 2026 and Ho Chi Minh City 16 and 20 Sep 2026, 186 SKUs, 11 stores. 6 to 20 Sep 2026
- FORM-05 World Bank WITS / UN Comtrade. China Protein concentrates and textured protein subst exports by country, 2023. 2023
- FORM-11 Phụ Gia Việt Mỹ (VMC Group). Top 3 địa chỉ phân phối đạm đậu nành uy tín chính hãng
- FORM-10 Dân Trí. TPHCM có tổ hợp ép dầu đậu nành quy mô hàng đầu Đông Nam Á. 10 Dec 2025
- REG-32 MARD. Thông tư 21/2019/TT-BNNPTNT hướng dẫn Luật Chăn nuôi về thức ăn chăn nuôi. 28 Nov 2019
- REG-33 MARD. Thông tư 31/2022/TT-BNNPTNT sửa đổi Thông tư 21/2019. 30 Dec 2022
- FM-20 Frontiers in Immunology. Effects of dietary Clostridium autoethanogenum protein on the growth, disease resistance, intestinal digestion, immunity and microbiota structure of Litopenaeus vannamei. 2022
- FM-31 Journal of the World Aquaculture Society. Mitigating the impact of winter temperatures on striped catfish (Pangasianodon hypophthalmus) using functional feed additives. 2023
- FM-12 CK Vietnam Group / V-Group Vietnam. Mycoprotein
- FM-04 Mardiana. Aquafeeds in 2025: Disrupted by tariffs. 30 Jun 2026
- ECO-05 The Asset. Vinh Hoan strikes insect protein deal with Entobel. 23 Feb 2024
- FM-05 Naylor, R. L., Hardy, R. W., Buschmann, A. H., et al. A 20-year retrospective review of global aquaculture. 2021
- FM-08 Xie, S., Niu, J., Zhou, W., et al. Developing a low fishmeal diet for juvenile Pacific white shrimp, Litopenaeus vannamei, using the nutritional value of FM as the reference profile. 2018
- RGN-53 Vietnam exports HS 110814 by partner. World Bank WITS. 2023
- RGN-55 China imports HS 170230 (glucose) by partner. World Bank WITS. 2024
- FS-11 Nông nghiệp và Môi trường. Nguồn cung giảm mạnh, giá sắn gần gấp đôi cùng kỳ 2025. 9 April 2026
- FS-12 Doanh nghiệp và Đầu tư. Thị trường sắn quý I/2026: giá tăng, xuất khẩu bứt phá nhưng tiềm ẩn rủi ro. 2 April 2026
- VCO-03 Vedan International. Milestones. undated
- INF-02 FIRI. FIRI, Food Production Pilot and Technology Transfer Center (FTET) page. undated
- INF-05 HCMC Biotechnology Center, technology transfer services list. undated
- INF-06 SHTP Management Board. Saigon Hi-Tech Park, home page (units, area, projects). undated
- RGN-36 EECi. Biorefinery Pilot Plant
- RD-44 Tran N.B.T. et al. Morphological variation, chromosome number and DNA barcoding of giant duckweed (Spirodela polyrhiza) in Vietnam. 2022
- SCI-36 Salian A et al. Lemnaceae as a poultry feed supplement: a review. 2026
- SCI-37 Pagliuso D et al. Duckweeds as Promising Food Feedstocks Globally. 2022
- SCI-28 EFSA NDA Panel. Safety of Wolffia globosa powder as a Novel food. 2021
- SCI-34 Appenroth KJ et al. Nutritional Value of the Duckweed Species of the Genus Wolffia as Human Food. 2018
- SCI-01 Nielsen MB, Meyer AS, Arnau J. The Next Food Revolution Is Here: Recombinant Microbial Production of Milk and Egg Proteins by Precision Fermentation. 2024
- SCI-05 Chen H et al. Integrated Fermentation Engineering Enables High-Level Leghemoglobin Production in Kluyveromyces marxianus. 2026
- SCI-08 Claes K et al. OPENPichia: licence-free Komagataella phaffii chassis strains and toolkit for protein expression. 2024
- RGN-10 ScaleUp Bio. ScaleUp Bio, home page
- RGN-24 van der Kley, D. China Biomanufacturing Study Report. 31 Jul 2026
- CAP-16 Bao Dau Tu. Dau tu vao start-up nuoc ngoai, Vinh Hoan huong den hai san, thit duoc nuoi cay tu te bao. 2021
- SCI-10 Soni P, Pradhan PK, Sood N et al. Development, characterization and in vitro applications of a thymus cell line from Pangasianodon hypophthalmus. 2023
- SCI-12 Anoop BS, Puthumana J et al. Immortalization of shrimp lymphoid cells by hybridizing with the continuous cell line Sf9 leading to the development of 'PmLyO-Sf9'. 2021
- RD-33 Thai H.-D. et al. Development of a new Agrobacterium-mediated transformation system based on a dual auxotrophic approach in Aspergillus oryzae. 2021
- RD-34 Trinh M.T. et al. Efficient generation of uridine/uracil auxotrophic mutants in homokaryotic Cordyceps militaris strains for constructing a food-grade expression platform. 2024
- SCI-19 Frisvad JC et al. Safety of the fungal workhorses of industrial biotechnology. 2018
- FORM-08 World Bank WITS / UN Comtrade. Vietnam imports of HS 071331 (dried mung/black gram beans) by country, 2023. 2023
- SCI-42 Shivakumar N et al. Protein-quality evaluation of complementary foods in Indian children. 2019