We are regularly sent board papers whose regional expansion case rests, somewhere in the assumptions, on the African Continental Free Trade Area. The line is usually short and confident: tariffs come down, the addressable market becomes continental, and the distribution model scales accordingly.
The ambition behind the agreement is real and the direction of travel is correct. But the treaty is a framework for negotiation, not a completed customs union, and a plan that treats it as delivered will misprice its landed costs and mis-time its expansion.
What is actually in force
The agreement entered into force in 2019 and trading under it formally began in 2021. Since then the practical work has been in the annexes rather than the headline: tariff concession schedules, rules of origin, and the mechanics of how a preference is actually claimed at a border post. Progress on those has been real but uneven, and the products where negotiation has been hardest — textiles, automotive, sugar, processed agriculture — are precisely the ones where a preference would be commercially decisive.
Meanwhile, the volume moving under the pilot and early-implementation arrangements remains small relative to total intra-African trade. That is not a criticism of the project. It is a statement about what your first three years of operation will look like.
The better guide is the scheme you already have
For anyone operating in West Africa, there is a more informative precedent than the continental agreement: the ECOWAS Trade Liberalisation Scheme. It has existed since the late 1970s, it promises duty-free circulation of approved originating goods across the community, and it is administered by a bloc with a common external tariff and, for eight of its members, a common currency and central bank.
It still does not work smoothly. Approved-product certification is slow. Recognition of another member state's certificate is inconsistent. Road checkpoints, informal levies and corridor delays impose costs that never appear in a tariff schedule. Some member states continue to apply charges that are difficult to reconcile with the scheme.
If forty-five years of regional integration among fifteen neighbouring states has not removed friction from a single corridor, a fifty-four-state agreement will not do it on your investment horizon.
The Sahel departures have changed the map
Any regional trade assumption written before 2024 needs revisiting. The withdrawal of Mali, Burkina Faso and Niger from ECOWAS, and the formation of a separate alliance among them, has introduced a genuine discontinuity in what had been treated as a single free-movement zone.
For coastal operators the practical exposure is corridor exposure. Landlocked demand served through Abidjan, Lomé, Cotonou or Tema now crosses a political boundary whose tariff, customs and security treatment is no longer settled. If your volume model assumes northbound distribution, that assumption needs its own scenario, not a footnote.
How to write the assumption honestly
None of this argues against regional expansion. It argues for modelling it on observed conditions rather than on treaty text. In practice that means:
- Price landed cost on currently applied tariffs, not on the preferential rate you expect to be able to claim. Treat any preference you do obtain as upside.
- Budget explicitly for non-tariff costs. Corridor time, checkpoint friction, port dwell, certification and documentation frequently exceed the duty saving the preference would have delivered.
- Test your rules-of-origin position early. If your product is assembled from imported inputs, whether it originates under the applicable rule is a factual question with a documentable answer. Get it answered before it is load-bearing.
- Model the corridor, not the country. In this region logistics, not tariffs, is usually the binding constraint on regional scale.
- Give the Sahel routes their own scenario with explicit trigger indicators, rather than averaging them into a regional number.
The point
A board paper that says "we assume continental tariff liberalisation delivers by year three" is making a political forecast dressed as a financial assumption. A board paper that says "our base case assumes no preference, and here is what changes if we obtain one" is doing the same work honestly — and is far more likely to survive its first year of operation.