Four claims, tested

In one paragraph. This study began with four claims about Vietnam, all of them treated at the outset as unverified. This chapter records what the evidence did to each. One is strongly supported and was understated. One is supported with both of its constraints now specified rather than asserted. One is only half testable within this project's scope. One is partly supported and partly not.

Claim 1. The barrier to Vietnamese plant-based adoption is product shortfall, meaning nutrition profile and taste, not cultural resistance. Chay food culture already normalises plant-based eating.#

Not resolvable within this project's scope. The supply-side half is qualified rather than supported.

This study can speak only to whether the products could be better, not to why people buy them.

• On taste: mung bean protein is reported as milder and cleaner-flavoured than soy, and commercial processing rather than the choice of crop is identified as the main driver of beany, grassy and oxidised notes. The ceiling on taste is therefore set by processing, which is a solvable engineering problem. This supports the product-shortfall reading.
• On nutrition: the claim that plant proteins are nutritionally inferior is true on average and much softer than usually stated. Mung bean DIAAS spans 57.7 to 93 against animal references at or above 100. A well-made mung bean isolate at the top of that range is not meaningfully worse than soy at 91.
• The qualification that matters: neither half of claim 1 can be resolved without demand-side evidence, which is out of scope. What this study establishes is narrower and still useful: there is real headroom to improve Vietnamese plant-based products on both taste and protein quality, and the constraint on doing so is not knowledge but the absence of a locally available protein crop at scale.

Claim 2. Vietnam is a credible Asia-Pacific biomanufacturing location, constrained by feedstock and regulatory immaturity.#

Supported, with both constraints now specified rather than asserted, and one of them larger than first thought.

The feedstock constraint is specifically a sugar and molasses constraint, confirmed independently by Vietnam being a net importer of molasses. It is real for precision fermentation and wrong for biomass fermentation, where cassava processing residue is the answer.

The regulatory constraint is a scored zero on the food side, meaning no framework at all rather than a slow one. Revised in this version: it is also larger on the feed side than the first version of this report allowed. The feed regime has a process, but the route for a genuinely new raw material is undocumented and may involve formal trials over 12 to 18 months. See R-G2 and R-C2.

Vietnam's credibility is higher than the source study shows, because the published cost models are priced on European electricity and labour. Revised in this version: how much higher is smaller than the first version implied. The power saving is worth about 6 to 7 percent of production cost and a tropical chilling duty plausibly offsets most of it; the durable advantage is labour, feedstock and co-location. See 10.7.

Claim 3. Import reliance makes a locally-fed novel protein a food security strategy.#

Strongly supported, and the figure the claim originally carried understated it.

Soybean is about 98.5 percent imported by supply, not the 93 percent the claim originally carried. Soybean meal feed consumption was 7.0 million tonnes in MY2024/25 against 41,000 tonnes of domestic soybean production, met by 5.7 million tonnes of imported meal plus meal crushed domestically from imported beans. Essentially all of Vietnam's soy protein is imported, in one form or the other. Total feed grain imports of 21.75 million tonnes in CY2024 sit alongside compound feed output of 21.5 million tonnes, though as section 8.11 notes these are different quantities and their near-equality is a coincidence rather than evidence in itself. Strongest source: USDA FAS GAIN VM2025-0011.

One reconciliation note. Earlier drafts stated that 3.2 million tonnes of imported beans were crushed domestically, while also reporting 3.2 million tonnes as total soybean imports and 550,000 tonnes as soybean food use. Those cannot all be true at once, and the crush figure is the least certain of the three. The conclusion does not depend on which is right, because domestic production of 41,000 tonnes is small against any of them.

Qualification: this also sets the competitive bar. The delivered Vietnamese prices of imported soybean meal and corn remain unestablished, which is open question CF-G4, and until they are known the feed-grade case cannot be closed either way.

Claim 4. Post-2023 conditions have shifted feasibility.#

Qualified, and now fully tested including the artificial intelligence component.

• Supported on policy, weakly: Resolution 36-NQ/TW, Decree 182/2024, and the corporate income tax law effective October 2025. The practical value of the investment support fund is much lower than headline figures suggest, and the eligibility thresholds are higher than the first version of this report stated; see R-C1 and section 7.6.
• Supported on regional infrastructure: ScaleUp Bio licensed in May 2024, and Singapore approvals in 2022 and 2023.
• Supported on intellectual property: OPENPichia, published in 2024, removing the chassis licensing barrier. Note that this removes a barrier to entry rather than a cost of production.
• Supported narrowly on artificial intelligence: measured gains in variant screening and a 40 to 80 percent reduction in media cost in proof-of-concept work, but end-to-end acceleration claims rest on benchmarks rather than production outcomes.
• Not supported on domestic capability, which remains flask-scale and open-pond with no submerged fermentation pilot.
• Not supported on titers, which remain one to three orders of magnitude below the economic threshold.