Du franc CFA à l’ECO : géographie monétaire endogène et séquençage d’une union monétaire ouest-africaine[From the CFA Franc to the ECO: Endogenous Monetary Geography and the Sequencing of a West African Monetary Union]

The proposed creation of the ECO is typically assessed through a static question: do West African economies sufficiently satisfy the criteria of an optimal currency area to share a common currency? This paper shifts the focus to the composition and sequencing of the monetary union. We show that enlargement does more than add economies to a given monetary architecture. By changing union-wide aggregates, it alters the common monetary policy, external exposure, and the economic base for risk sharing. We refer to this mechanism as endogenous monetary geography. Using a panel of fifteen West African economies over 2000-2023, we construct a multidimensional index of macro-monetary divergence combining business-cycle synchronization, shocks, inflation, fiscal and external positions, public debt, and the real effective exchange rate. Hierarchical clustering identifies a highly stable core of seven WAEMU economies, with Côte d’Ivoire at its boundary. This structure is robust to alternative clustering methods, equal weighting of the underlying dimensions, and temporal resampling. The core’s average co-clustering frequency reaches 0.982 across 500 bootstrap replications. We then assess the consequences of enlargement using counterfactual national and common monetary policy rules. The mismatch between the common policy stance and national conditions rises sharply as the union expands from WAEMU to the full regional configuration but does so non-monotonically. Ghana and Nigeria generate the largest discontinuities. Nigeria reveals a key distinction between macro-monetary proximity and systemic weight. Its entry increases the RMSE of the mismatch for the representative country but reduces the GDP-weighted RMSE because its economic weight endogenously shifts the common monetary benchmark. An exhaustive decomposition of all 5,040 possible admission sequences confirms this mechanism. Nigeria’s average marginal contribution is 5.040 points for the equally weighted RMSE, compared with 0.014 points for the GDP-weighted RMSE, with the sign of the effect depending on its position in the sequence. Finally, enlargement simultaneously changes the structure of external exposure and the capacity to pool residual stabilization needs. These findings show that the ECO cannot be reduced to either a choice of membership or a binary test of monetary optimality. When member economies differ substantially in size, the composition of the union itself becomes a state variable of common monetary policy. Membership, sequencing, external anchoring, and stabilization mechanisms must therefore be analyzed jointly.