Why Vietnam at all

In one paragraph. Vietnam is an unusual case: a country with one of the world's largest animal feed industries, almost none of the protein that industry runs on, very cheap industrial labour and power, enormous concentrated crop residues, and no legal framework for selling a new kind of protein as food. The first four facts make it one of the better places in Asia to make bulk protein cheaply. The fifth is why nobody has.

The import bill is the opportunity#

Vietnam produced 21.5 million tonnes of industrial compound feed in 2024 and imported 21.75 million tonnes of feed grain in the same year. Those are different quantities and their near-equality is a coincidence rather than an identity, but the direction is not in doubt.

The soy position is starker. Vietnam grew 41,000 tonnes of soybeans and imported 3.2 million tonnes, which is about 98.5 percent of supply. Soybean meal consumption for feed was 7.0 million tonnes, met by imported meal plus meal crushed domestically from imported beans. Essentially every gram of soy protein in Vietnamese animal feed arrived by ship.

Corn tells the same story: 4.2 million tonnes grown against 12 million tonnes imported.

A protein made in Vietnam from Vietnamese material would be substituting for that. It is the largest and hardest price target in this report at the same time.

The inputs are genuinely cheap, within limits#

The published cost models for microbial protein are priced on European inputs. The one techno-economic assessment in this study that resolves its cost structure line by line assumes electricity at EUR 0.18 per kilowatt-hour and labour at EUR 20 per hour. Vietnam's comparable figures are USD 0.08 per kilowatt-hour and a manufacturing wage of USD 342 a month, which at a 40-hour week is about USD 1.97 an hour.

Vietnamese power is therefore around 40 percent of the modelled price, and Vietnamese manufacturing labour around a tenth of it.

What that is actually worth, which is less than it sounds. Revised in this version. Utilities are 18 percent of production cost in that model. Cutting the power portion of an 18 percent line by 60 percent is worth roughly 6 to 7 percent of total production cost, not 60 percent of anything. And a tropical site pays part of it back. Southern Vietnam's humidity means evaporative cooling towers cannot deliver water cold enough to cool a fermenter passively, so mechanical refrigeration has to run continuously; a first external estimate puts that at USD 0.095 to 0.12 per kilogram of biomass, which against a product selling at USD 1.20 to 2.07 per kilogram is 5 to 10 percent of cost. Those two numbers are close enough that cooling may consume the whole of the electricity advantage on the fermentation step.

What survives is labour. At roughly a tenth of the modelled price, and with no chiller to pay it back, cheap skilled manufacturing labour is the durable part of Vietnam's cost case. The electricity argument should be made carefully or not at all.

Two further qualifications. This is an indicative transfer of Vietnamese prices into a European model, not a costed Vietnamese result, and no Vietnam-specific study exists. And the cooling estimate above is itself a desk calculation that nobody has checked against a real plant. See chapter 10 and open question F-G2.

The residues are large and concentrated#

Vietnam harvests about 10.5 million tonnes of cassava roots a year, most of which passes through roughly 120 industrial starch factories in 27 provinces. Those factories leave behind wet pulp that is 72 percent carbohydrate and currently sells cheaply as cattle feed. Four brewery groups produce somewhere on the order of 600,000 to 860,000 tonnes of spent grain a year, already 15 to 30 percent protein. Rice milling generates 43 to 47 million tonnes of straw and several million tonnes of bran. Seafood processing generates about 1 million tonnes of nitrogen-rich by-product.

Nobody has published how much cassava pulp those 120 factories actually produce. That is the single most important unmeasured number in this report, and it is one phone call away. See open question CF-G1.

The general rule that emerged is that dry, concentrated residues are available and wet, scattered ones are not, and that a wet residue has to be used where it is made. Chapter 8 maps all of them.

There is a buyer already waiting, though a smaller one than it first looks#

Vietnam's farmed fish and shrimp industry consumes 3.4 to 4.0 million tonnes of feed a year, and its most expensive protein ingredient, fishmeal, reached USD 1,600 per tonne in late 2025. Microbial protein modelled at USD 1,200 to 2,070 per tonne is in range for fishmeal and not for soybean meal.

But tonnage is not the market. Revised in this version. Fishmeal inclusion is concentrated in shrimp and marine finfish. Pangasius, which is more than half of that 3.4 to 4.0 million tonnes, is formulated around cheap plant and by-product meals and carries very little marine protein to displace. The pool of fishmeal a new ingredient could actually compete for is on the order of 350,000 to 450,000 tonnes a year. That is the number to quote, and it is still a large market.

That price comparison, against that volume, decides the commercial thesis of this report. Chapter 9 sets it out.

And one fact cancels a great deal of the above#

Vietnamese food safety regulation contains no novel food category. There is no defined route to market for a fermentation-derived or otherwise new protein ingredient intended for human food, no application to make and no process to be refused by. Vietnam scored zero on regulatory readiness in the only comparative Asia-Pacific site-selection study available, and does not appear in the academic literature on ASEAN novel food regulation at all.

Animal feed is different, though less different than the first version of this report claimed. Revised in this version. That regime has a named authority, a permitted-ingredient list and a defined dossier, which is three things more than the food regime has. What it does not have is a published timeline for a genuinely new ingredient. The thirty-working-day clock that circulates applies to feed additives and to raw materials already on the permitted list. Treat the feed route as a defined process of unknown duration, not a fast one.

That is still the second reason, independent of price, to start with feed: there is something to apply for. It is no longer a reason to expect a quick answer.

Chapter 7 is the whole picture, including two routes around the food gap that can each be tested with a letter.