Plans for European monetary union, the break-up of monetary unions in Central and Eastern Europe and German unification have renewed interest in the literature on `optimum currency areas', which exhibit high inter-regional factor mobility, trade relatively little with the rest of the world and produce diverse ranges of products. In Discussion Paper No. 968, Research Associate Tamim Bayoumi develops a formal model in which regions with sticky nominal wages and producing different goods choose whether to keep their own currencies or join together in currency unions. Joining a union reduces the transaction costs of trade with its other members but risks substantial output losses from the loss of the exchange rate instrument to offset asymmetric disturbances. His model embodies such criteria for optimum currency areas as the size of the underlying disturbances, the degree of their correlation across regions, the costs of transactions between different currencies and the level of factor mobility and the interrelationship of demand across regions.

Bayoumi finds that a currency union may raise the welfare of its member regions but unambiguously reduces the welfare of non-member regions: while the gains from lower transaction costs are restricted to the union's members, the losses from reduced output affect everybody, which serves as a useful reminder that the a currency union's impact on non-members need not be benign. A region's incentives to join an existing currency union also contrast with the union's incentives to admit it. While the entrant gains from reduced transaction costs with the entire existing union, its incumbent regions only gain in their trade with the potential entrant. A small region's incentive to join will therefore always be greater than the union's incentive to admit it, so even a region that would prefer a free float everywhere is likely to join a currency union that others plan to form: it will incur most of the welfare losses from the nascent union will occur whether it joins or not. These findings may account for the concern of EU member countries that remain sceptical about plans for monetary union not to be relegated to the second division of a two-speed EMU.

A Formal Model of Optimum Currency Areas
Tamim Bayoumi

Discussion Paper No. 968, June 1994 (IM)