Endogenous optimal currency areas: The case of the Central African Economic and Monetary Community

The Central African Economic and Monetary Community (CAEMC) has been a monetary union for several decades now. According to the hypothesis of endogenous optimal currency areas (OCA), the degree of business cycles synchronization across its member states should be significantly higher today than 40 years ago. Investigating the empirical validity of this hypothesis is important in the context of the African economic integration process. If currency unions are endogenous, then quick monetary integration is a worthwhile option that can be used to accelerate economic integration. On the contrary, if currency unions were not endogenous, then a speedy monetary unification would not benefit countries collectively and might therefore jeopardize the whole regional integration initiative. This paper assesses the endogeneity of CAEMC as an OCA by examining the cross-country synchronization of business cycles along three statistical dimensions: bilateral correlation of cyclical co-movements, similarity of cycle statistical properties, and concordance of cyclical phases. Its innovative contribution is threefold. First, it provides a direct test of the endogeneity hypothesis on a specific currency union. Most previous studies instead rely on panel estimates of global datasets. Second, it expands the existing literature on the monetary geography of Africa. Indeed, there are several papers that study whether or not specific African regions are OCA. However, these papers generally look at the ex-ante conditions for optimality, leaving the issue of endogeneity of OCA criteria unexplored. The paper fills in this gap. Third, the paper presents a business cycle chronology for the six CAEMC members, thus opening up new opportunities to understand the cyclical characterization of economic systems and policies in the region. The main result of the analysis is that (i) the degree of synchronization of business cycles across CAEMC countries has remained low throughout the period 1960-2007, but (ii) it has somewhat increased over time. This increase is however marginal in both economic and statistical terms, thus implying that CAEMC currency union is not as endogenous as one would expect from previous empirical results obtained from global datasets. The reason why the endogeneity effect is so weak is that its channels of transmission are not work: intra-regional trade is very low and macroeconomic policies across union members do not seem to converge. Furthermore, increasingly different productive structures also reduced the intensity of synchronization. The policy implications of the analysis then concern the design of policy and institutions in the CAEMC and the way forward for monetary unification in Africa.