How Algeria's Baladna project in the Sahara could cut milk powder imports, what a ton of Adrar powder really costs, and why water, heat and capital will decide the outcome.
For decades, Algeria's hydrocarbons paid for what its farms could not produce, including one of its most important staples: milk powder. That arrangement bought food security, but it also left the country exposed to international commodity prices, foreign suppliers, shipping costs and swings in export revenue.
The Baladna project in Adrar is an attempt to change that equation. It is a joint venture between Qatar's Baladna (51%) and Algeria's National Investment Fund, FNI (49%), with an announced investment of about USD 3.5 billion. The idea is to grow animal feed in the Sahara, build a very large Holstein herd, produce fresh milk and turn it into milk powder for the Algerian market.
At full capacity, the project targets about 1.7 billion litres of milk a year and 193,000 to 198,000 tonnes of powder, roughly half of national demand.
One point must be clear from the start. Construction is under way, the first powder is expected around the end of 2027 or early 2028, and the full herd is a multi-year target (around nine years). Every cost figure in this article is a projection for full-load operation, not a record of results.
The argument here is deliberately more modest than "the desert beats Europe". Adrar's advantage is real, but it lies in cheap land, cheap on-site feed and patient state capital, not in cheap gas for the dryer. Once capital is charged, the project lands somewhere between a modest cost edge and parity with imports. For a state that values foreign-exchange savings and food security, that can still be a rational bet. It is simply a different story from a commanding cost lead.
How dependent is Algeria on imported milk powder?
Algeria has a domestic fresh-milk sector, supported by subsidies and thousands of local breeders, but its industrial dairies rely heavily on imported powder. Published trade figures vary by year and source:
- USDA data for 2023 shows roughly 250,000 tonnes of whole milk powder and about 168,000 tonnes of skimmed powder, around 418,000 tonnes in total.
- Trade data for 2025 cited in the press points to about 256,000 tonnes of whole and 154,000 tonnes of skimmed powder, more than 410,000 tonnes combined, within total dairy imports of about 455,000 tonnes.
The two series are consistent: the country imports roughly 410,000 to 420,000 tonnes of powder a year. Every tonne produced locally can in principle replace a tonne bought abroad, although real substitution depends on quality, price, processing needs and demand. The goal is not to eliminate every dairy import, but to reduce strategic dependence on imported powder.
The Baladna Adrar model: feed, herd and factory in one chain
The agreement between the Algerian government and Baladna was announced in 2024. The model is vertically integrated:
- Agriculture. Barley, maize, forage and other feed crops are grown on land developed in Adrar.
- Livestock. A Holstein herd housed in climate-controlled barns with veterinary supervision and automated milking.
- Processing. A dedicated factory at Tamest turns fresh milk into powder, which can be stored and shipped across the country without a cold chain.
Published project figures are not interchangeable and should be read with care. Original documents describe about 117,000 hectares of leased land and a herd of 240,000 to 270,000 cattle. More recent company communications mention a total development area of 219,000 hectares and a target herd of 270,000. The gap between 117,000 and 219,000 hectares needs clarification: is the larger figure the wider project footprint or an expanded agricultural allocation? Powder capacity is quoted between 193,000 and 198,000 tonnes a year. All of these are planning figures.
Where the project stands today
- April 2026: a second wave of contracts worth more than USD 635 million covering livestock, construction and infrastructure.
- Livestock: a transport programme for 30,000 Holstein cattle from nine US states, using 109 cargo flights, with first deliveries expected from November 2026.
- First harvests: a first barley harvest at Tamest in May 2026, then, by October, about 530 tonnes of barley, 359 tonnes of wheat, 584 tonnes of forage and 746 tonnes of silage maize from around 360 hectares.
- Targets: up to 70,000 cultivated hectares equipped with about 1,400 irrigation pivots.
- Inspection: the Minister of Agriculture visited the livestock facilities, the Tamest powder plant and the irrigation installations on October 19, 2026.
- Timeline: first milk around the end of 2027, with powder starting around the end of 2027 or early 2028, depending on commissioning.
These volumes are small next to the long-term ambition, but they show that cultivation has begun.
How is the USD 3.5 billion financed?
The published plan combines foreign expertise, state participation and domestic banking:
- about USD 875 million from Baladna;
- about USD 840 million from the National Investment Fund;
- about USD 1.785 billion in local subsidised debt, reported at roughly 1% interest.
The financing documents also describe a 20-year offtake agreement with ONIL, the national dairy products office, a 20-year bank payment guarantee and a 10-year tax exemption. This reduces commercial uncertainty, but it also commits Algerian banks and the public purchaser for a long period. Equity ownership (51/49) should not be confused with the financing structure, where roughly half the money is domestic debt.
Is the dryer where the money is? No, the farm is
Turning milk into powder means evaporating about 7,000 to 7,500 litres of water per tonne. Even if we accept an unsubsidised European energy bill of USD 150 to 260 per tonne of powder, against USD 25 to 40 in Adrar, the saving is about USD 125 to 220 per tonne. That is only 3% to 5% of the USD 4,200 to 4,350 at which French whole milk powder lands in Algerian ports.
Milk is what drives the price. At about 8.7 litres of milk per kilogram of powder (1.7 billion litres over 194,000 tonnes, the project's own numbers), raw milk is more than three quarters of the cost. Any real advantage must therefore come from the farm. Adrar plans to grow its own forage, about 2.5 million tonnes a year, whereas Baladna's original 24,000-cow farm in Qatar imports hay and alfalfa by sea. Our estimate is that this takes raw milk from about USD 0.50 to 0.60 per litre in Qatar to USD 0.30 to 0.36 in Algeria. The whole case rests on that number, and it deserves an audited source before anyone treats it as settled.
What a tonne of Adrar milk powder might really cost
The table applies these cost layers to the project's milk-to-powder ratio and then adds the layer that is usually missing: capital. It is illustrative. It leaves out labour, packaging, logistics and marketing, and it assumes USD 0.30 to 0.36 per litre is a cash cost.
Scenario (per tonne of powder, full load) | Cost per tonne | Versus French powder landed at USD 4,200–4,350 |
|---|---|---|
A. Cash cost only: milk (USD 2,610–3,130), drying energy (25–40), farm electricity, cooling and pumping (70–95) | USD 2,705–3,265 | About USD 935–1,645 below |
B. A plus capital: USD 3.5 billion over 20 years, straight-line, zero interest (about USD 900 per tonne) | USD 3,605–4,165 | About USD 35–745 below |
C. A plus capital at a 6% cost of capital over 20 years (about USD 1,575 per tonne) | USD 4,280–4,840 | From about USD 70 below to USD 640 above |
Four cautions apply:
- The USD 3.5 billion covers the herd, the forage farm and the plant, so charging all of it to powder is conservative, although meat is only a by-product.
- If USD 0.30 to 0.36 per litre already includes farm depreciation, scenarios B and C double count.
- The farm electricity line may already sit inside the milk cost.
- Because about half the funding is subsidised debt at around 1%, the real blended cost of capital is probably below 6%, which places the true result between scenarios B and C.
The honest reading: Adrar has a large cash-cost advantage and, depending on how capital is priced, something between a modest edge and parity once capital is included. "Cheaper than Western Europe" is true on a cash basis. It is not yet proven on a fully loaded basis.
The benchmark also matters. Algeria buys from New Zealand, Argentina, Uruguay and the EU, so Adrar's costs should be set against Global Dairy Trade whole milk powder prices, not French quotes alone.
What import substitution is really worth
Covering about 200,000 tonnes at USD 4,200 per tonne avoids roughly USD 0.8 billion of imports a year. That is a gross figure, not a net one. Cows, genetics, German-built drying equipment, feed additives and spare parts are imported. Baladna holds 51% of the venture, so part of the profit leaves the country. And the price at which the joint venture sells powder to the state, in a market built around subsidised milk, will decide who captures the benefit. Total powder demand of about 410,000 to 420,000 tonnes implies an import bill of well over USD 1 billion, depending on the product mix, so even full output would cover only about half of it.
Water: the real constraint
The cost stack above contains no water charge, yet water is the project's largest long-term risk. Adrar draws on the Continental Intercalaire (Albian) aquifer, part of the North-Western Sahara Aquifer System shared with Tunisia and Libya. Much of this water is fossil groundwater with almost no modern recharge. Studies of the Adrar aquifer have found water more than 20,000 years old and documented pressure declines under intensive pumping.
The feed, not the cattle, is the largest water user. Agronomic benchmarks for irrigated forage in hot desert climates run on the order of 10,000 cubic metres per hectare per year or more. As an illustration, the 70,000-hectare target at that benchmark would imply up to about 700 million cubic metres a year, or several thousand cubic metres per tonne of powder. This is an order-of-magnitude illustration, not a measured figure. The number of pivots alone says nothing about consumption: what matters is the area irrigated, the crop mix, seasonal needs, system efficiency and the volume pumped per well.
Drip irrigation, drought-tolerant forage and closed-loop barn water, techniques Baladna developed in Qatar, will help. But no sufficiently detailed public water balance exists yet. The authorities and the company should publish annual abstraction targets, wellfield monitoring data, irrigation efficiency indicators, water use per tonne of powder, the pumping energy it implies and the aquifer's depletion horizon. The objective should be to maximise value per cubic metre of groundwater, not simply cultivated land.
Heat, energy and logistics
Heat is a second physical risk. At 45°C, Holsteins lose feed intake and fertility. Across the whole herd, 1.7 billion litres over 270,000 animals is about 6,300 litres per head per year. That is plausible, because the herd includes heifers and dry cows, but the yield per milking cow should be stated and depends on how heat stress is managed.
Large-scale dairy farming also needs reliable electricity for pumping, cooling, automated milking and processing. Algeria's gas-fired power system provides a foundation, but electricity still has to be generated, transmitted and paid for, so efficiency matters in a region where irrigation pumping and air-conditioning are heavy loads. Finished products must then travel from Adrar to dairies across the country.
The project also promises thousands of direct and indirect jobs and the transfer of skills in agronomy, veterinary science, irrigation and industrial operation. That workforce may prove as valuable as the powder.
Food sovereignty is not autarky
Food sovereignty does not mean producing every input at home. It means being able to secure essential food without excessive vulnerability to external shocks. Baladna can cut exposure to the powder market, but it will depend on imported genetics, equipment and technology, and probably on imported protein concentrates and premixes. "Eliminating imported feed" therefore needs care. The real measure of integration will be the share of feed produced locally, crop yield stability, irrigation efficiency and cattle productivity.
Qatar's experience shows that modern dairy systems can work in a hot, water-constrained setting, but Adrar is far larger, with a different aquifer, agriculture and market. It gives technology and know-how, not an independent assessment of Adrar's water and economics. The project also sits alongside Algeria's phosphate, fertiliser, port, rail and desalination investments. These can reinforce one another, or compete for the same energy and water.
The export story is the least proven part
For five to ten years, every tonne will stay in Algeria. Afterwards, the case for exporting to the Sahel and West Africa rests on Halal certification, the absence of anti-subsidy barriers and AfCFTA tariff preferences. Against that stand well over a thousand kilometres of desert road with security risks, while European, New Zealand and South American powder reaches Dakar, Abidjan and Lagos by sea at low freight cost. Exports are an option, not a base case, until logistics and tariff advantages are costed.
Bottom line
Baladna Adrar is best understood as a strategic import-substitution project backed by patient capital, with a large cash-cost advantage and a thin, still unproven margin once the USD 3.5 billion is charged. If it works, Algeria turns a billion-dollar annual import bill into domestic production, jobs and a food-security buffer. If it fails, the likely causes are heat, water or capital costs, not the price of gas in the dryer. The true test of food sovereignty will not be building the biggest dairy in the desert, but keeping it productive, affordable and sustainable for generations.
0 Comments