West Africa produces more agricultural output than its own markets can absorb. Very little of it ever reaches a structured market. That gap is infrastructural before it is agricultural. This is a paper about the specific constraint that prevents West African supply from reaching institutional buyers at scale, why previous solutions have failed to address it, and what the architecture looks like when you actually solve it.
There is enough commodity
Ghana produces maize in volume every season, drought years included. Nigeria’s maize production is considerably larger. Together, the Guinea Savannah Belt production zones contain enough commodity to supply institutional buyers across the region indefinitely.
Yet institutional buyers, particularly Gulf sovereign food security programmes, source from established origins with demonstrated reliability. Black Sea origins have carried a disproportionate share of the wheat that moves in international trade, and Gulf buyers were concentrated there until supply disruption forced them to diversify.
West Africa’s absence from institutional procurement isn’t due to production capacity. It’s due to the absence of the infrastructure that converts field-grade commodity into institutional-grade supply.
They require certified supply
Saudi Arabia imports most of the food it eats. The General Food Security Authority (GFSA) is the exclusive importer of subsidised milling wheat. Almarai and Savola Group are the primary institutional feed grain and oilseed processors.
These entities require certified supply: commodities meeting phytosanitary standards, moisture specifications, aflatoxin limits, protein and oil content thresholds. They require consistent volume, quarterly or annual commitments with reliable delivery schedules. They require documented traceability, a field-to-port commodity chain with full certification at each stage. And they require specification alignment, processing standards that match the buyer’s end-use requirements.
West African production has never been formatted to these specifications at scale. That’s the constraint. Not productivity. Not quantity. Format.
Three disconnected problems
Agro-industrial parks in Ghana, Nigeria, and Côte d’Ivoire run well below designed capacity. The common explanation is poor management or inadequate aggregation. Both are wrong. Parks fail because they separated market linkage into three disconnected problems and expected each to solve itself.
On the supply side, thousands of smallholder farmers, seasonal production, highly variable quality. Parks assumed farmers would self-organise into supply chains. They don’t. Farmers lack storage, drying infrastructure, and price certainty. Without guaranteed buyers and price floors, there’s no incentive to produce to a specification rather than to volume.
On the processing side, parks built equipment and assumed processors would occupy space and add value. Processors are capital-constrained and risk-averse. They won’t invest if input supply is uncertain or output buyers haven’t committed. Why would they?
On the market access side, parks assumed that processing to export-grade specification would automatically create buyer interest. It doesn’t. Buyers don’t know the facility exists. They have existing supply chains with established origins. The burden of proof is entirely on the new entrant.
No single facility can solve all three problems alone. Yet most parks tried. Each problem remained partially solved. The entire system failed.
Contracts that specify outcome
Market linkage fails when supply, processing, and export are designed as separate functions with separate actors, separate capital, and separate risk. It works when they’re architecturally integrated.
Supply-side contracts specify outcome, not process. Farmers are compensated for commodity meeting a specification (moisture, aflatoxin, protein content) not for producing a crop. This flips the risk model entirely. Farmers know exactly what price they’ll receive for commodity meeting the spec. Processors know exactly what specification they’ll receive. Buyers know exactly what they’re buying.
Processing infrastructure is built to buyer specification. The processor doesn’t design the facility and hope buyers show up. The buyer specifies the commodity form, quality standards, and delivery schedule. The processor builds to that specification. No guessing. No overcapacity.
Export logistics are pre-contracted. Quarterly departure windows are scheduled 12 to 18 months in advance, coordinating production schedules with aggregation network capacity, processing capacity, port allocation, and buyer procurement calendars. No demurrage. No quality disputes at port.
De-risking all three at once
An Integrated Agro-Industrial Platform (IAIP) is not an agro-industrial park. It’s a vertically integrated system where production contracts align farmer incentives with facility requirements, processing operations are built to buyer specification before construction starts, and export logistics operate on a scheduled model rather than spot booking.
Fifteen years in West African agriculture, much of it at Volta Presentation Farms, refined it into an architecture that de-risks all three components simultaneously: supply aggregation, processing, and export.
Who carries the risk
The architecture only holds if someone is accountable for each failure point. Supply is the farmer’s risk until the contract moves it. Throughput is the operator’s risk, which is why the operator has to hold equity rather than a management fee. Quality at port is the platform’s risk, because the platform is the only party that touches the commodity from field to vessel. Where those lines are blurred, everybody hedges. Farmers side-sell the moment a trader turns up at the gate with cash. Operators run the line at whatever volume protects their fee. The platform finds out at the point of shipment, which is the most expensive place to find out anything.
The failures are rarely mechanical. Rain arrives late and the aggregation window compresses into three weeks. A road goes out and trucks queue at the gate while grain heats in the trailer. A buyer moves a specification a month before departure and the whole grading protocol has to move with it. Infrastructure gives you the capacity to absorb that. Incentives decide whether anyone bothers to. Market access decides whether absorbing it was worth doing in the first place. Execution is what happens when all three are tested in the same week, which in West Africa is most weeks.
No buffer, no overcapacity
Previous agro-industrial models fail because they optimise for generic infrastructure. The IAIP can’t become a white elephant because supply contracts exist before construction, processing equipment is built to buyer specification with no excess capacity for non-existent demand, export logistics are pre-contracted with quarterly departures scheduled 18 months in advance, and governance participation by APDC Holdings ensures operational integrity across all SPVs via Class B voting control.
There’s no buffer. No overcapacity. No hope that buyers will eventually show up. Everything is built to answer a specific buyer requirement before the first shovel touches ground.
