Aquafeed: the market that would buy it first

In one paragraph. Substantially revised in this version. A new bulk protein made in Vietnam has to displace something already on the market, and the price of that something decides whether the idea works. Fishmeal, the high-protein ingredient in feed for farmed fish and shrimp, reached USD 1,600 per tonne in late 2025. Soybean meal trades in the hundreds. A microbial protein modelled at USD 1,200 to 2,070 per tonne is competitive against the first and not against the second. That is the commercial thesis of this report in three numbers. The correction in this version is about which aquafeed: the first version pointed at shrimp and pangasius together, on a combined tonnage of 3.4 to 4.0 million tonnes. Pangasius barely uses fishmeal. The fishmeal-displacing market is shrimp, marine finfish and part of tilapia, and it is worth roughly 350,000 to 450,000 tonnes of fishmeal a year.

9.1 The market, by what it actually contains#

Tonnage of feed is the wrong denominator. What matters is how much fishmeal each species' ration contains, because that is what a new protein is competing to replace. [VN-direct] on the market, [general] on the feeding trials

The displaceable fishmeal pool is roughly 350,000 to 450,000 tonnes a year. That is the market this report recommends aiming at. It is a great deal smaller than the 3.4 to 4.0 million tonne headline the first version used, and it is still large.

Provenance note. These inclusion rates come from an external review of this report rather than from the study's own research, and they have not been checked against the underlying aquafeed census. The species-level feed volumes also differ slightly between the two sources. Both should be confirmed in a single reading of the census, or better, by asking the mills. See contradiction CF-C6 and action 5.

9.2 The price to beat#

• Fishmeal reached USD 1,600 per tonne by November 2025. An external reviewer reports a subsequent rise to USD 1,825 per tonne on catch-quota fluctuation in the southeastern Pacific, which if confirmed widens the parity corridor. Unverified, so USD 1,600 remains the figure quoted here.
• Soybean meal trades far below that, and an external reviewer puts delivered prices at Cai Mep and Ho Chi Minh City at USD 350 to 450 per tonne over 2024 and 2025. Also unverified.
• Microbial protein modelled at USD 1.20 to 2.07 per kilogram is competitive against fishmeal and not against soybean meal. A fungal or bacterial protein costs roughly three to four times soybean meal per unit of crude protein, which is why terrestrial pig and poultry rations, costing USD 300 to 400 per tonne in total, are not a viable target.

The industry is already substituting: poultry by-product meal as a partial fishmeal replacement, plus fermented soy, soy protein concentrate and rapeseed meal. A new ingredient is not asking the industry to do something it has never done.

9.3 The biology supports it, and points at two different products#

In whiteleg shrimp, yeast products, bacterial single-cell protein and microbial floc meals work well at low to moderate inclusion, with some studies reporting full fishmeal replacement where the rest of the diet and the culture system are favourable. Limits at higher inclusion track digestibility, amino acid balance and palatability rather than crude protein. In black tiger shrimp the evidence is thinner but positive, strongest for the commercial microbial product Novacq and for one yeast study showing better survival and immune response at 10 percent of the diet.

In pangasius and other catfish, yeast culture and yeast single-cell protein maintain or improve growth, feed conversion, immunity and disease resistance at moderate replacement levels.

Those two findings support different businesses. Clarified in this version. In shrimp, microbial protein competes on price against an expensive ingredient and wins on function as well. In pangasius, there is no expensive ingredient to displace, so the same material would be sold as a functional gut-health ingredient at low inclusion, against a feed cost base of plant meals, on a claim about survival rather than protein. Different inclusion rate, different price ceiling, different buyer, different regulatory dossier. The first version of this report ran them together and produced a headline market larger than its own evidence supported. Price and position them separately.

A co-benefit worth pricing, in shrimp. At the right inclusion rate these ingredients improve gut condition, general immunity and survival after disease challenge. An external reviewer cites trials in whiteleg shrimp showing improved survival against acute hepatopancreatic necrosis disease and Enterocytozoon hepatopenaei at 3 to 8 percent hydrolysed yeast or bacterial biomass, on a thin source that should be checked. Given that disease loss in Mekong Delta shrimp farming routinely exceeds 30 to 40 percent, the functional claim may be worth more than the protein itself, and may support a price above bare fishmeal parity.

9.4 The second reason to start here#

Price is only half of it. Vietnam's animal feed regime has something to apply to: a named authority, a permitted-ingredient list and a defined dossier. The food regime has no category for a new protein at all.

What changed in this version is the strength of that argument, not its direction. The first version described a thirty-working-day clock. That timeline applies to feed additives and to ingredients already on the permitted list, not to a new microbial biomass, and the route for a new raw material is undocumented. So the feed regime offers a defined process of unknown duration, which is still better than no process, and is no longer a reason to expect speed. Chapter 7 sets out both regimes, and action 3 is the letter that resolves this.

9.5 What is still missing#

The decisive number has not been established by this study: the delivered Vietnamese price of fishmeal and of imported soybean meal, against the modelled cost of microbial protein at Vietnamese input prices. That is one conversation with a purchasing manager at an aquafeed mill, and it is open question CF-G10. The same call should ask each mill for its own fishmeal inclusion rate by species, which is the number section 9.1 turns on and which the mills hold.

Until both are answered, the case in this chapter is directionally strong and not yet costed.