Latin America Animal Feed Market Size, Share & Forecast 2026–2033

The market in one paragraph

Latin America's feed mills put 204.4 million tonnes of compound feed into the region's animals in 2025 — 5.5 million tonnes more than in 2024 and 14.2% of the 1.44 billion tonnes fed worldwide (Alltech, 2026 Agri-Food Outlook, published 21 April 2026, covering 142 countries and 38,837 mills). On our revenue-per-tonne derivation that tonnage is worth USD 91.7 billion in 2025 and USD 125.5 billion in 2033, on 247.2 million tonnes. Three numbers explain the market: 89.9 million tonnes of the total is Brazilian, 41.9 is Mexican, and feed is 62.5% of a Paraná broiler producer's cost and 72.5% of a Santa Catarina pig producer's. The industry's growth is therefore decided as much in the grain belt as in the nutrition laboratory.

• Volume: 204.4 Mt (2025) → 247.2 Mt (2033), 2.40% CAGR
• Value: USD 91.7 bn (2025) → USD 125.5 bn (2033), 4.00% CAGR
• Concentration: Brazil 44.0% of regional tonnage, Mexico 20.5%, Argentina 9.1%, Ecuador 3.0%
• Fastest species: aquaculture, +11.4% in 2025, on Ecuadorian shrimp and Chilean salmon
• Slower than the world: 2.8% regional volume growth against 2.9% globally — Latin America is no longer the fastest-growing feed region; Africa grew 11.5%

What this report counts, and what it deliberately leaves out

Regional feed statistics are a minefield because four credible institutions measure four different things. This report is built on the narrowest, most comparable definition — industrial compound feed — and then says where the wider national measures differ, because the differences are large enough to change a growth rate.

The valuation is ours, and it is derived, not surveyed. No institution publishes a dollar value for Latin American compound feed. We therefore take the revenue per tonne implied by the global market — USD 645.80 billion of 2025 value against 1,440.077 million tonnes of 2025 volume, or USD 448.45 per tonne — and apply it to the region's tonnage. That gives USD 91.68 billion. The check is bottom-up and independent: Sindirações reports that Brazilian mills moved R$180 billion of macro and micro ingredients in 2025, which at roughly R$5.45 to the dollar is USD 33.0 billion of raw material for 94 million tonnes, or about R$1,915 per tonne of ingredient cost. Ingredient cost is three-quarters or more of a mill's selling price, which lands Brazilian feed turnover in the low-to-mid USD 40 billions — consistent with Brazil's 44.0% share of a USD 91.7 billion region.

What is deliberately excluded: feed grains traded as commodities rather than consumed as feed (only the feed they become is counted); on-farm mixing outside the commercial channel, except where a national measure like CAENA's explicitly captures it; veterinary pharmaceuticals, which are a separate market with separate regulators; feed machinery and silos; and pet food at retail value, of which only the feed tonnage is included.

One warning about public estimates. Published third-party figures for this specific region range from a USD 194 million "market" — a unit error of roughly three orders of magnitude — to a USD 176.8 billion figure that includes the whole feed-crop complex. We use neither, cite neither, and flag the discrepancy because analysts do reuse both.

Market forecast snapshot

The forecast separates volume from price, because conflating them is how regional feed forecasts go wrong. Volume grows 2.40% a year: below the region's 2019–2024 pace, above its 2025 outcome of 2.8% only in the out-years, and anchored on Brazilian mill output of 94 Mt in 2025 rising toward Sindirações' own 2026 expectation of roughly 97 Mt. The extra 1.56 percentage points of value growth is a mix-and-price wedge: aquaculture and pet feed taking share from commodity broiler rations, additive and premix intensity rising per tonne, and nominal prices recovering with the cost base.

How much the price assumption matters. If revenue per tonne stays flat at USD 448.45 for eight years, 2033 value is USD 110.9 billion, not USD 125.5 billion — a 12% difference from one assumption. Readers who want a pure-volume view should use that lower figure. Every USD 25 per tonne of error in the 2025 base moves the market by USD 5.1 billion, or 5.6%.

The protein basket: how one region came to hold 14% of the world's feed

Latin America grew 2.8% in 2025 while the world grew 2.9%, which is the first year in recent memory that the region has not outgrown the global average. The composition is what matters. Asia holds 559.297 million tonnes (38.8% of world volume) but feeds mostly for its own consumption; Latin America holds 204.446 million tonnes and exports a disproportionate share of what it feeds. That is the "protein basket" thesis, and it is a trade thesis rather than a consumption thesis.

The export evidence is unusually clean. In 2025 Brazil produced 15.289 million tonnes of chicken — third in the world — and exported 5.324 million tonnes to 153 countries for USD 9.8 billion, or 34.82% of output, with per capita consumption of 46.7 kg. It produced 5.592 million tonnes of pork and exported 1.510 million tonnes to 94 countries for USD 3.6 billion, or 27.01% of output. ABPA puts the two chains at R$205 billion of gross production value and USD 13.9 billion of foreign-exchange revenue. Ecuador's shrimp and Chile's salmon replicate the pattern in aquaculture, which is precisely why Latin American aquafeed grew 11.4% in 2025 against 4.7% globally.

Two structural forces sit under that. The first is grain. Brazil's 2025/26 grain crop reached a record 361.7 million tonnes (+2.6%), including 180.4 million tonnes of soybeans (+5.2%, with exports forecast at a record 116.2 million tonnes) and 144 million tonnes of corn (+2.0%, of which 112.1 million tonnes from the second crop). Cheap local corn and soybean meal are the region's comparative advantage in feed — and, as the cost chapter shows, also the reason its feed mills have little pricing power.

The second is the shift from on-farm mixing to commercial feed. Alltech attributes a meaningful part of Chinese volume growth to exactly this transition; in Latin America the equivalent is visible in Brazil's confinement census, where the number of confined cattle rose from 7.96 million head in 2024 to 9.25 million in 2025 (+16%), heading toward 10 million in 2026. Every animal that moves from pasture to a feedlot is a tonne of feed that moves from the farm's own mixer to a registered mill.

Brazil, Mexico, and the third that is neither

Brazil is a feed industry with a country attached. At 44.0% of regional tonnage it is larger than the next two countries combined, and its species mix is the most balanced in the region: 37.85 Mt of broiler feed (+2.5%), 22.5 Mt of swine feed (+4.2%), 7.76 Mt of beef cattle feed (+7.5%), 7.66 Mt of dairy feed (+7.9%) and 7.43 Mt of layer feed (+3.5%) under the Sindirações perimeter. Note the direction of travel: the two fastest-growing Brazilian segments are the two cattle segments, which is a feedlot-and-confinement story rather than a chicken story. Sindirações represents about 140 companies that it says account for roughly 90% of the national market.

Mexico is the import-dependent half of the region's feed complex, and that is its defining feature. It produces 41.883 Mt of feed on roughly 24.5 million tonnes of domestic corn. USDA's Mexico City post forecasts corn imports of 26.8 million tonnes in MY 2026/27, up 1%, against production forecast down 3%; total 2026 corn availability is 53.7 million tonnes against consumption of 49.0 million. Mexico's feed industry buys its energy in dollars and sells its protein in pesos, which is why Mexican feed values per tonne sit above the regional average and why the country's largest integrators hedge grain rather than feed.

Argentina is the region's most under-measured large market. CAENA's feed-equivalent method puts everything fed at 35.1 million tonnes in 2025 (+1.1%), of which industrial balanced feed and concentrates are 18.7 million tonnes (+2.1%). The gap between current feeding and balanced-diet potential is the largest in the region: beef cattle consume a feed equivalent of 12.34 Mt against a theoretical 152.8 Mt, and dairy 10.39 Mt against 12.12 Mt — a total theoretical potential of 191.3 Mt. Argentina is also the region's grain exporter, with corn production forecast at 54 million tonnes and exports at 37 million tonnes in MY 2025/26, so its feed industry competes for corn with its own export terminals.

Ecuador is the price outlier and the cautionary case. Balanced feed output rose 10.7% to 6.2 million tonnes, of which shrimp feed is 2.47 Mt (+10.3%), poultry 2.11 Mt (+7.6%) and swine 1.29 Mt (+8.4%). APROBAL values the whole complex at about USD 5.14 billion — roughly 4% of national GDP — which implies USD 829 per tonne, some 85% above the regional average, because Ecuador imports its corn and its wheat. It also carries the clearest weather risk in the region: Alltech attributes a 23–29% collapse in Ecuador's 2025 corn harvest to extreme weather, which paralysed domestic poultry and pig production, and APROBAL puts the resulting corn overcost at about USD 150 million a year.

The residual third is where the growth is. Colombia's broiler feed rose 14.7% in 2025 on technology upgrades and rising per-capita consumption, while its layer feed fell on weaker demand and Salmonella pressure. Peru's layer feed rose 36% and Bolivia's 27.3%, both technical rebounds after severe avian influenza in 2024. Chile's layer feed grew 3.7% and its salmon output roughly 8%. The Dominican Republic's pig feed was constrained by disease. These are 47.8 million tonnes with almost no comparable public data — the single biggest evidence gap in this report.

Five species, five different markets

Broiler feed is the anchor and it had a poor year. 74.766 Mt, up 1.9% against 3.7% globally, held back by avian influenza trade restrictions. The internal divergence is the story: Colombia grew 14.7% while Brazil — with 37.900 Mt, half the regional broiler total — held roughly flat. A region that exports a third of its chicken is a region whose feed demand is hostage to other countries' veterinary decisions.

Pig feed is the compounding growth story. 43.042 Mt, +4.0%, concentrated in Brazil on export demand and better production economics, with industrialisation and intensification in Chile and Colombia adding volume. Disease is the limiter: African swine fever pressure in the Dominican Republic constrained the Caribbean. Pig feed is also the most additive-intensive tonne in the region, which makes it the commercial battleground for premix and gut-health suppliers.

Dairy is the region's most mature and least dynamic segment. 23.221 Mt, +0.7%. Stronger milk prices supported feeding in Brazil and Uruguay; favourable forage in Mexico held programmes stable; Colombia contracted as margins tightened and production extensified. Growth here comes from consolidation and per-cow yield, not from animal numbers.

Beef feed is the surprise. 16.206 Mt, +4.1%, with Brazil up 7.1% to 7.730 Mt as feedlot margins improved and finishing periods lengthened. Supplementation is spreading where pasture volatility persists — the Southern Cone and Chile — and feedlot infrastructure is expanding in Paraguay. This is the segment where a 16% jump in Brazilian confinement numbers shows up directly in mill volumes.

Aquaculture is small, fast, and the best leading indicator of value. 7.641 Mt, +11.4%, the fastest of any region bar Africa. Ecuador led with shrimp production rebounding about 15%; Chile's salmon rose roughly 8%; Brazil contributed through tilapia. The watchpoint into 2026 is fishmeal price and availability, the one input where Latin America is a price taker rather than a price maker.

Layers are a recovery trade, and the residual matters. Layer feed across the region recovered in 2025 after 2024's avian influenza culls, with Peru and Bolivia rebounding hardest and Chile adding 3.7%. The 39.570 Mt residual in the table above is layer feed plus pet food, equine and non-species tonnage; we show it as a residual rather than inventing a split, because the species-level regional table is not published in a form that reconciles to the total.

Where the money goes: 62.5% and 72.5%

Embrapa's CIAS cost indices give the region's cleanest view of what feed is worth to its customer, and the answer is almost everything. In July 2026, a Paraná broiler producer's cost was R$4.77 per kilogram of live weight, of which feed was 62.51% and one-day chicks 19.83%; feed cost rose 1.67% in the month and 1.35% over twelve. A Santa Catarina pig producer's cost was R$6.29 per kilogram, of which feed was 72.52%; feed rose 1.19% in the month and had fallen 1.81% over the year to July.

Three consequences follow, and they shape every commercial decision in this market.

• Feed mills can pass through cost, but not margin. When feed is 62–73% of the customer's cost, the customer's first response to a price rise is to reformulate, not to pay. That caps the mill's ability to earn from formulation and pushes it toward selling services, additives and financing instead.
• Least-cost formulation is the product. A nutritionist who can cut one percentage point of soybean meal out of a broiler ration is worth more to a Brazilian integrator than a percentage point of market share. This is the mechanism that makes phytase, synthetic amino acids and alternative proteins grow faster than the tonnage they sit in — the argument developed in Lysine Below Cost and in our Global Feed Phosphates report.
• Cheap grain is a demand subsidy that expires. Today's volumes are being bought with a record Brazilian harvest. When corn and soybean meal revert toward their means, the feed cost line rises before the protein price does, and the mill's customer cuts stocking density rather than feed quality. That lag — one to two quarters — is the region's main cyclical risk, and it is not in any forecast that extrapolates 2025.

The ratio also explains why the region's feed industry is being built from both ends at once: integrated protein companies building captive mills to control the 62.5%, and additive and premix companies building local plants to sell into the 37.5% that remains.

What the data will not support

We would rather state the limits than imply a precision the sources do not have.

• No monthly regional series exists. The region has an annual survey (Alltech, first quarter each year) and a handful of national monthly measures (Sindirações in Brazil, APROBAL in Ecuador). There is no LatAm equivalent of the EU's feed manufacturers' federation data, so quarterly commentary on this market is inference, not measurement.
• Country-level dollar values are mostly unavailable. Of the four countries in our table, only Ecuador publishes a market value (USD 5.14 billion, via APROBAL). Brazil can be reconstructed from an ingredient bill. Mexico and Argentina cannot be valued from primary sources at all, which is why we present them in tonnes and do not publish a country value split.
• The species table does not reconcile by construction. Five species are published for the region; the remaining 39.570 Mt is a residual that contains layer feed plus pet, equine and non-attributed tonnage. Anyone presenting a clean eight-species regional split is estimating.
• No mill-level market shares are published. Sindirações' roughly 140 members represent about 90% of the Brazilian market, but no company discloses Brazilian feed volume. Our competitive chapter therefore describes business models and disclosed investments, not shares — and we have not constructed a share table from inference.
• Value per tonne is a derived number, not a surveyed one. The USD 448.45 figure is the global market's average, and it demonstrably understates Ecuador (USD 829) and probably overstates parts of Brazil. Treat the USD 91.7 billion as a central estimate within a band of roughly USD 87–101 billion.

What would close these gaps: a harmonised MERCOSUR feed registry with volumes and values, published additive-approval registers in machine-readable form, and any national statistical office treating compound feed as an industrial product rather than an agricultural input. None of that exists today.

Companion analysis on this market

Four shorter pieces sit alongside this report and each takes one part of the argument further.

• Brazil Is 44 Per Cent of the Region and Mexico Buys Its Corn in Dollars — why four countries in one region have four different cost structures, from a corn exporter pricing in reais to an importer buying 26.8 million tonnes in dollars.
• Feed Is 62.5 Per Cent of the Bill. That Is Why the Mill Cannot Raise Its Price — the cost ratio that governs pricing power, and why the industry's profit moved into formulation and additives.
• The Meat Merger That Was Really a Feed Order — CADE's clearance of BRF into Marfrig, and why it concentrated a feed order book rather than a milling market.
• Press release: Latin America's Feed Market to Reach USD 125.5 Billion by 2033 — the headline numbers, the valuation method and the regional context in one page.

Table of Contents

• The market in one paragraph
• What this report counts, and what it deliberately leaves out
• Market forecast snapshot
• The protein basket: how one region came to hold 14% of the world’s feed
• Brazil, Mexico, and the third that is neither
• Five species, five different markets
• Where the money goes: 62.5% and 72.5%
• What the data will not support

Competitive Landscape

Who owns the tonne

Latin America’s feed industry is often described as consolidating. It is consolidating at the ingredient and protein layers, and it remains stubbornly fragmented at the mill. Those are different industries with different economics, and reading one as the other is the commonest mistake made about this market.

No company publishes Latin American feed volume, so we do not publish shares. What can be described with evidence is the structure: who owns feed capacity, what they do with it, and where the disclosed capital has gone.

Tier one: integrated protein companies with captive mills

The largest feed producers in the region are not feed companies. They are chicken, pork and beef processors who mill for their own animals and sell the surplus into the merchant market. Brazil’s 2025 consolidation made this tier larger: on 5 September 2025 CADE cleared the merger of BRF into Marfrig without restrictions, creating a company renamed MBRF Global Foods Company S.A. that combines Marfrig’s beef and processed-protein business with BRF’s poultry, swine and processed-food operations. CADE’s reasoning is worth noting for what it says about the structure of the market: horizontal overlaps were below 20% and each vertical-stage share below 30%, thresholds that would have been impossible to meet in a genuinely concentrated milling market. Minerva Foods intervened in the case, raising concerns about SALIC’s cross-holdings in Marfrig and Minerva and the risk of coordinated effects; CADE’s tribunal unanimously rejected the appeal.

JBS, the world’s largest beef producer, and Minerva operate their own feed and supplement capacity for the same reason: when feed is 62.5–72.5% of your cost, owning the mill is a margin strategy, not a supply strategy. Captive mills are also a competitive weapon in procurement — an integrator with its own mill can buy corn at harvest, store it, and price its own feed below the merchant market for as long as its balance sheet allows.

Tier two: international agri-feed groups

De Heus is the clearest example of a European feed group building a Brazilian position rather than buying one. It has operated in Brazil for more than a decade and runs six plants — two in Rio Claro (SP), plus Apucarana (PR), Toledo (PR), Guararapes (SP) and Itaberaí (GO) — supported by an administrative office in Campinas, an experimental pig-nutrition centre and a distribution centre in Bezerros (PE). Its Toledo unit was rebuilt through 2025 as a pig-feed reference plant, with new exhaustion systems, enlarged premix and medication weighing rooms, external silos and two dedicated pelleting lines, and it is deliberately ractopamine-free because several export markets prohibit the additive.

Cargill, Bunge and ADM compete one layer up, in origination and processing, and selectively in feed and premix. ADM’s Paraná nutrient facility lifting production by 40% is the pattern: the internationals invest where their ingredient position can be converted into a formulated product, not in commodity milling. Bunge’s completed Viterra combination changes the procurement map more than the milling map.

Tier three: national independents with real capacity

Brazil’s independent millers are larger than outsiders assume. Agroceres Multimix opened its ninth plant at Quatro Pontes (PR) with more than R$100 million invested, 65,000 square metres of built area and capacity above 10,000 tonnes a month; phase one makes poultry and swine feed and concentrates, with ruminant nutrition, núcleos and premixes planned next. Guabi, 50 years old in 2024, operates five plants — Sales Oliveira (SP), Além Paraíba (MG), Pará de Minas (MG), Anápolis (GO) and Rondonópolis (MT) — and completed a refurbishment programme across all five to raise capacity and efficiency. Sindirações’ roughly 140 member companies account for about 90% of the Brazilian market, which tells you both that the association is representative and that no single member dominates it.

The additive and premix layer

This is the most concentrated part of the chain and the one where the region’s dependency has been falling. dsm-firmenich leads the Brazilian cattle-supplement market through its Tortuga brand and has localised production: a plant at Sete Lagoas (MG), opened in October 2024, with 100,000 tonnes a year of capacity for beef and dairy supplements, 40,000 square metres of site and 12,000 square metres built. Its FarmTell platform — used by more than 15,000 users across 6,000 farms monitoring over 5 million animals — is the clearest evidence that feed selling in the region is becoming a data business. Beyond dsm-firmenich, the amino acid, vitamin and enzyme layer is supplied by a small group of European and Chinese producers through distributors, which is why additive availability in the region is a function of global plant economics rather than local demand.

What the structure means for competition

• The milling layer competes on cost of capital and logistics, not on product. Rations are similar; delivered grain cost is not.
• The additive layer competes on registration and formulation data. A molecule approved in Brazil is not approved in Mexico, and a premix matrix validated on Brazilian corn is not directly transferable to Mexican corn.
• Integration is the region’s answer to thin margins. Every merger in 2025 — Bunge-Viterra, BRF into Marfrig — moved a company closer to owning an input or an outlet, not closer to owning a mill.
• New entrants should expect to buy distribution rather than build it. The independent millers are family-controlled, capacity-rich and valuation-conscious; the additive distributors control the technical relationship with the nutritionist.

Value Chain

From the Cerrado to the feeder: the chain behind a tonne of feed

A tonne of Latin American compound feed is, by mass, mostly Brazilian or Argentine corn and soybean meal, plus a small and expensive layer of minerals and additives that determines whether the animal performs. The chain’s economics sit in the first part; its margins sit in the second.

Energy. Corn is the base of nearly every ration in the region. Brazil’s 2025/26 crop reached 144 million tonnes (+2.0%), of which 112.1 million came from the second crop — the safrinha that has turned the Center-West into the world’s most price-competitive corn origin in the second half of each year. That second crop is the single most important supplier relationship in the region’s feed industry: it arrives close to the mills, it is cheap, and it disappears entirely in a bad year.

Protein. Soybean meal comes from the region’s own crush. Brazil’s 2025/26 soybean crop hit a record 180.4 million tonnes (+5.2%), with exports forecast at a record 116.2 million tonnes (CONAB). The residual after export and crush is what the feed industry buys, and the crush margin is being pulled in two directions at once: food and feed demand on the meal side, and biodiesel and renewable diesel demand on the oil side. A strong oil pull raises crush volumes and cheapens meal, which is good for feed buyers; a weak one does the reverse.

The three import-dependent markets behave differently. Mexico produces about 24.5 million tonnes of corn and forecasts imports of 26.8 million tonnes in MY 2026/27, so its feed industry is structurally long dollar and short peso. Argentina produces 54 million tonnes and exports 37 million, meaning its feed mills compete for corn with their own port terminals. Ecuador’s 2025 corn harvest fell 23–29% on extreme weather, and its feed industry responded the way an import-dependent system always does: substitution. Wheat imports are forecast at 1.78 million tonnes, and roughly half of the wheat destined for feed now goes to shrimp — a ration change that would have been unthinkable a decade ago.

Minerals and additives. This is the part of the chain where margin lives and where the region has been building capacity. Dicalcium and monocalcium phosphate supply the phosphorus and calcium that cereal and oilseed rations lack; the pricing mechanism behind them is set by sulfur, phosphate rock and Chinese export policy rather than by feed demand, as set out in our Global Feed Phosphates report and in Feed Phosphate Is a Sulfur Price With a Quota Attached. Amino acids are the other half of the substitution story — the mechanism by which a nutritionist removes soybean meal from a ration, analysed in Lysine Below Cost. Both categories are supplied by a small number of global producers selling into a region that until recently imported almost all of its premix.

Trade and origination. The ingredient layer consolidated sharply in 2025. Bunge closed its USD 34 billion merger with Viterra on 2 July 2025, creating a crop-trading and processing group explicitly positioned to rival ADM and Cargill in connecting origination to consumption — and Brazil is one of the origination markets that mattered most in the deal’s logic. For feed mills, that means fewer, larger counterparties for meal and a shorter list of companies able to offer origin-to-mill supply agreements.

Logistics is the invisible tax. The region’s feed is produced close to where grain is grown and consumed where animals are raised and processed, and in Brazil those two geographies have been moving apart for a decade as poultry and pig production concentrates in the South while corn and soy expand in the Center-West. Freight and storage, not formulation, are where the least efficient mills lose the most money. We treat this as a structural read rather than a measured figure: no regional body publishes delivered-cost breakdowns by mill.

Where the value pool sits

The asymmetry worth noticing: the layers with pricing power are the ones with a patent, a registration or a port. Milling, which is where the tonnage is, operates on thin conversion margins because its customers can reformulate.

Investment Activity

Where the capital is going

Three years of disclosed investment in this region tell a consistent story: money is going into premix and supplement capacity, into pig and poultry feed in Brazil’s South and Center-West, and into consolidation of the ingredient layer. Almost none of it is going into new commodity milling capacity in Argentina or Mexico.

Announced and completed projects

The two transactions that reset the map

Bunge and Viterra closed on 2 July 2025, two years after announcing a USD 34 billion deal that creates a crop trading and processing group explicitly built to rival ADM and Cargill in linking origination to consumption. For a Brazilian feed mill, the practical consequence is fewer independent counterparties for soybean meal and a longer contractual chain between the farm that grew the bean and the mill that buys the meal.

BRF merged into Marfrig, cleared by CADE on 5 September 2025 and renamed MBRF Global Foods Company. This is the most consequential single event in the region’s feed demand structure in a decade: it concentrates poultry and swine feed buying, plus beef processing, inside one balance sheet. Whether it increases or reduces merchant feed demand depends entirely on how much of the combined company’s feed stays captive — and that is a decision management has not disclosed.

What the capital is telling us

• Localising additives beats importing them. dsm-firmenich’s 100,000 tonnes at Sete Lagoas only makes sense if premix and supplement demand is expected to grow faster than the feed pile itself. That is a direct bet on additive intensity per tonne — the same bet embedded in our value CAGR.
• Brazil’s South and Center-West is where capacity is being added. Quatro Pontes (PR), Toledo (PR), Guararapes (SP), Itaberaí (GO) — every disclosed plant is inside the corridor where poultry and pig production is densest and where second-crop corn is cheapest.
• Headroom is modest, so growth must be taken. Brazil produced 94 Mt in 2025 and Sindirações expects about 97 Mt in 2026. Against that, each new plant of 10,000 tonnes a month is a little over 1% of national output — enough to matter locally, not nationally. New capacity in this market is a share fight.
• Argentina and Mexico remain under-invested relative to their animal populations. Argentina’s feed-equivalent demand runs at 35.1 Mt against a theoretical 191.3 Mt of balanced-diet potential (CAENA); Mexico imports 26.8 Mt of corn a year. Both are large opportunities that capital has so far declined, for reasons of currency, policy and margin — not of demand.

What to watch next

MBRF’s post-merger capital plan; whether De Heus adds a seventh Brazilian plant; whether any global additive producer follows dsm-firmenich into local premix manufacturing; and whether EUDR-driven traceability investment (see the regulatory chapter) turns into a cost advantage for larger exporters or simply a cost.

Technology & Innovation

Formulation is the product

In a market where feed is two-thirds of the customer’s cost, technology is not a differentiator appended to the product — it is the product. The competitive frontier in Latin America runs along four lines: removing expensive nutrients from the ration, measuring what is actually in the ingredients, replacing antimicrobial growth promotion, and turning feed into an emissions and traceability product.

1. Least-cost formulation, and the additives that make it work

The core technique is ordinary and the leverage is not. A nutritionist who can cut a percentage point of soybean meal from a broiler ration by adding crystalline amino acids and phytase saves the integrator more money than any purchasing negotiation. This is why amino acid and enzyme volumes in the region grow faster than feed tonnage, and why the molecules’ prices behave so strangely — as we set out in Lysine Below Cost. Phytase releases phosphorus bound in plant phytate, which reduces the inorganic phosphate a ration needs; the mechanism and its price consequences are covered in our Global Feed Phosphates report.

The Latin American twist is ingredient variability. Brazilian second-crop corn and Argentine corn differ materially in density, moisture and mycotoxin load from the US corn that most formulation models were calibrated on. That makes near-infrared analysis and rapid mycotoxin testing worth more here than in mature markets, and it makes local formulation data a genuine asset.

2. Digital and data services

dsm-firmenich’s FarmTell platform — more than 15,000 users, 6,000 farms and over 5 million animals monitored — is the clearest regional evidence that feed companies now sell software alongside supplements, with expansion from Brazil into Mexico and Paraguay planned. The commercial logic is straightforward: a supplement company that can show a farm its own performance data can defend a price that a commodity premix cannot. Expect the same move from genetics and animal-health companies, and expect consolidation of these platforms.

3. Gut health and the antimicrobial transition

Probiotics, organic acids, essential oils, prebiotics and zinc alternatives are the fastest-moving additive classes in the region, driven less by consumer sentiment than by export-market requirements. A Brazilian exporter selling to the EU or to stricter Asian buyers faces specifications on antimicrobial use that its domestic customers do not, so gut-health adoption tracks export exposure almost exactly. This is also where the technical service moat is deepest: these products only work inside a formulation and management system, which favours suppliers with field nutritionists.

4. Methane, and the feed carbon footprint

Feed is becoming an emissions product. dsm-firmenich’s Bovaer — approved by food-safety authorities in the EU, UK and Japan and commercially available in more than 70 countries — reduces enteric methane by about 30% in dairy cows and up to 45% in beef cattle. The Brazilian evidence is local: at the DSM Beef Center GF in Rio Brilhante (Mato Grosso do Sul), trials reported a 38.2% reduction over 86 days in 30 Nellore animals and a 40% average reduction over 100 days in 20 cows. For a region with the world’s largest commercial cattle herd and a beef export trade under pressure to document deforestation-free and low-carbon supply, a feed additive that produces a measurable abatement number is a commercial instrument, not a cost. It also creates a new revenue logic: feed sold with a verified emissions outcome can be priced partly on the carbon market rather than only on weight gain.

5. Novel proteins and the circular economy

Insect protein and black soldier fly meal are the region’s most credible partial substitutes for fishmeal and soybean meal in aquafeed and pet food, and Brazil’s combination of organic waste volumes and aquaculture growth makes it a natural production base. Our black soldier fly and insect protein reports size the global opportunity; in Latin America the near-term use case is shrimp and tilapia feed, where fishmeal price volatility is the binding constraint.

6. Biosecurity as process technology

The post-2023 investment cycle has been as much about contamination control as about capacity. Agroceres Multimix’s new plant uses drag conveyors with self-cleaning mechanisms and lot-level traceability from raw material to delivery; De Heus’s Toledo line was authorised under Brazil’s Decreto 12.031/2024 and Portaria SDA 798/2023 specifically because it handles medicated premixes. In a region where one avian influenza case in a commercial flock can close export markets, process technology is risk management priced as capex.

The test for the next five years

Does additive and service intensity per tonne rise faster than tonnage? If it does, our 4.00% value CAGR is conservative and the 1.56-point gap between value and volume growth is too narrow. If it does not — if cheap grain lets formulators compete purely on price — the region converges on volume growth of 2.4% and value growth barely above it. Every disclosed investment above is a bet on the first outcome.

Market Risk

The five risks that matter

Latin American feed demand is unusually exposed to decisions made outside the region. The risks below are ordered by how quickly they can change tonnage, not by how likely they are.

1. Trade policy — the fastest transmission channel

The 50% US tariff imposed on Brazilian goods in 2025 is the clearest case. Before it took effect, beef packers’ association Abiec — representing companies including JBS and Marfrig — estimated losses of USD 1 billion on the trade. The reallocation that followed is visible in the trade data: by August 2025 Mexico had overtaken the United States as Brazil’s second-largest beef export destination, with analysts expecting shipments to the US to be replaced by Mexico and Australia. Every tonne redirected changes where an animal is finished and therefore where feed is milled. A tariff is a feed demand shock wearing a trade costume.

2. Animal disease — the binary risk

Brazil recorded a single outbreak of highly pathogenic avian influenza in a commercial flock, in Montenegro (Rio Grande do Sul), notified on 16 May 2025. The sanitary void began on 22 May, the flock was destroyed, and on 18 June 2025 the Ministry of Agriculture notified the World Organisation for Animal Health of the end of the void and self-declared the country free of HPAI — after which it began notifying the countries that had restricted purchases. The episode cost weeks of export flow and demonstrated that a single farm can move a national trade variable.

The wider pattern is worse than the Brazilian one. Newcastle disease has generated repeated rounds of export restrictions, with MAPA issuing updated restriction lists as recently as October 2026. African swine fever pressure constrained pig feed in the Dominican Republic through 2025. Salmonella pressure hurt Colombian layer demand. And the layer segment’s 2025 rebound in Peru (+36%) and Bolivia (+27.3%) is a reminder that 2024’s outbreaks were severe enough to leave a measurable base effect a year later. Disease is the one risk that can move a species’ feed demand by double digits in a single quarter.

3. Input and weather shocks

Ecuador’s 2025 corn harvest fell 23–29% on extreme weather, which Alltech describes as paralysing domestic poultry and pig production, and APROBAL prices the resulting corn overcost at roughly USD 150 million a year. Argentina’s 2024/25 corn crop was cut by corn stunt and drought. These are not tail events in this region; they are the operating environment. The specific risk for 2026–2033 is the reverse of the current one: today’s demand is being supported by a record Brazilian crop and unusually low grain prices. A return to average grain prices without a matching rise in chicken and pork prices compresses the customer’s margin, and the customer responds by cutting stocking density.

4. Currency and macroeconomic pass-through

The region runs three monetary regimes at once: a floating real, a managed and historically volatile peso in Argentina, and a dollarised economy in Ecuador. A weaker real helps exporters and hurts importers of additives and phosphates; a stronger one does the opposite. Ecuador’s experience shows the cost of the wrong side of that trade — its feed complex is worth about 4% of GDP and pays an import premium on every tonne. Macro risk also operates through consumer demand: 25.2% of respondents to Alltech’s 2025 survey named product prices, inflation and the economy as the biggest expected disruptors in 2026, the single largest response category.

5. Capacity, consolidation and the captive-feed question

Brazil is expected to produce about 97 Mt of feed in 2026 against 94 Mt in 2025. Against that growth, every disclosed plant expansion is a share fight, and the merger of BRF into Marfrig adds a structural uncertainty: if MBRF internalises more of its feed, merchant millers lose volume without losing a customer relationship they can see. The same logic applies at the ingredient layer after Bunge-Viterra.

Two risks we are not weighting heavily, and why

Feed substitution away from animal protein. Plant-based alternatives have not moved the region’s per-capita meat consumption, and Brazil’s chicken consumption of 46.7 kg per head and pork of 19.1 kg are still rising. We see substitution as a long-horizon risk with no near-term tonnage effect.

Additive supply crises. Amino acid and vitamin markets have had violent price episodes, but the region’s mills respond by reformulating rather than by buying less feed. Price risk in additives moves value, not volume — which is precisely why it belongs in the value CAGR and not in the tonnage forecast.

What would falsify this forecast

Two consecutive quarters of declining Brazilian mill output; a grain price rise of more than 25% without matching protein price pass-through; a multi-state HPAI outbreak in Brazil’s commercial flock; or a US–Brazil trade settlement that removes the tariff and pulls significantly more Brazilian protein into the US market faster than the region can rebuild breeding stock.

Regulatory Landscape

Four regulators, one continent

There is no Latin American feed regulator. There are national regulators with national registers, and the practical consequence is that an additive approved in Brazil is not approved in Mexico, and a premix registered in Argentina cannot be sold in Colombia without a separate filing. Registration, not capacity, is the real barrier to entry in this market.

Brazil: MAPA, and a shift to self-control

Brazil’s Ministry of Agriculture and Livestock (MAPA) regulates feed through its animal-inputs department. Any establishment that manufactures, fractions, imports, exports or sells feed, supplements, premixes, co-products, núcleos, pet food or feed additives must be registered with MAPA and appear on its published lists of registered establishments and products, which MAPA updates and republishes (the current product list was updated on 8 September 2026).

The legal architecture was rebuilt between 2022 and 2025, and the direction of travel is deregulation by displacement — the state moving from pre-approval to audit:

• Lei 14.515/2022 set the framework for agricultural defence inspection and the administrative process around it.
• Decreto 12.031/2024 (28 May 2024) regulated the mandatory inspection of products intended for animal feed, covering both establishments and products, and is the instrument under which new or modified lines handling medicated premixes are authorised.
• Decreto 12.126/2024 (31 July 2024) regulated self-control programmes for private agents and introduced risk-based inspection and surveillance — the single most consequential change for compliance cost, because it makes the firm responsible for its own controls and the state responsible for verifying them.
• Decreto 12.502/2025 (11 June 2025) set the rules for the administrative enforcement process, revoked two articles of Decreto 12.031 and provided for a conduct-adjustment term.
• Portaria SDA 798/2023 governs the specific authorisation of lines handling medications, which is why a plant making medicated piglet feed needs its own license conditions.

What this means commercially: self-control favours scale. A company with a validated quality system, traceability from lot to delivery and its own laboratory can absorb the compliance burden; a small mill without those systems faces the same obligations with a fraction of the revenue. Expect the regulatory framework itself to drive consolidation in Brazilian milling over the next five years, independently of any merger.

Mexico: SENASICA and the NOMs

Mexico regulates feed through SENASICA under the Federal Animal Health Law (Ley Federal de Sanidad Animal), supported by a set of official standards:

• NOM-012-ZOO-1993 sets specifications for the formulation of balanced feed — prohibited substances, and limits on mycotoxins, heavy metals and medicinal residues.
• NOM-061-ZOO-1999 covered animal-feed specifications and has been integrated into the broader Federal Animal Health Law framework rather than remaining a standalone instrument.
• NOM-030-ZOO-1995 regulates medicinal residues in animal products, which is the standard that matters if a Mexican producer wants to export.

Establishments wishing to export livestock products must be registered with SENASICA and comply with the relevant NOMs; in practice Mexican feed and integrator businesses also carry ISO 22000, FSSC 22000 or HACCP certification because their customers’ export markets require it. Mexico’s regulatory burden is therefore export-driven: a purely domestic producer faces a lighter regime than one selling into the United States, which is most of the large ones.

Argentina, Colombia and Chile: the second tier

Argentina’s SENASA administers animal health and traceability, and its livestock stock data are the foundation of the CAENA feed-equivalent methodology that produces the country’s only credible feed statistic — a case of private statistical governance filling a public gap. Colombia’s ICA and Chile’s SAG perform the equivalent animal-health and input-registration functions in their markets. None of these regulators publishes feed volumes, which is why the region’s production data rests so heavily on Alltech’s industry survey.

The external regulator: EUDR

The European Union Deforestation Regulation (EU) 2023/1115 covers seven commodities — cattle, cocoa, coffee, palm oil, soy, wood and rubber — and therefore reaches two of this region’s feed inputs directly: soybean meal and cattle. The European Parliament and Council adopted a revised regulation on 19 December 2025 that added a further period before application and made substantive amendments, giving all companies inside and outside the EU another year to prepare, with application now expected from the end of 2026. The Commission’s April 2026 simplification report concluded that the 2024–2025 measures plus the December 2025 amendments and further simplifications substantially reduce administrative burden, particularly for operators sourcing from low-risk countries, and declined to propose further changes to the legal text.

For Latin American feed, the operative point is that compliance is documentary and traceable rather than chemical: a mill buying soy from a trader must be able to show the plot, not just the silo. That favours large, vertically integrated supply chains and adds a fixed cost to small exporters.

Divergent product standards inside one market

Ractopamine is the clearest example of a divergence with capital consequences. It is permitted in some markets and prohibited in others, and a Brazilian mill serving both cannot switch formulations per truckload, so it builds dedicated ractopamine-free lines — as De Heus did at Toledo. The same logic applies to antimicrobial growth promoters, to maximum residue levels and to feed-additive approvals. The region’s mills are therefore building compliance into plant design rather than into product labelling.

What to watch

Whether MERCOSUR moves toward mutual recognition of feed-additive registrations, which would cut the cost of a regional launch by a factor; how aggressively MAPA audits against Decreto 12.126’s self-control obligations; whether EUDR application from end-2026 survives without a further delay; and whether Mexico’s NOM framework is revised to match the Federal Animal Health Law’s consolidated structure.