Monetary Union
Optimal currency areas

Many authors use the theory of optimal currency areas (OCAs) to assess the potential costs and benefits of monetary integration, but the theory often suggests that existing currency areas are already too large. In Discussion Paper No. 590, Research Fellow Jacques Mélitz proposes focusing initially on a small currency area and considering whether it should be enlarged. In his model, the size of the common currency area is a continuous variable ranging from zero (if there is no enlargement) to one (where all sources of imports and competition in trade are included in the union). Once this rises above zero, a fixed exchange rate entails no costs if prices and wages are perfectly flexible; but price and wage inertia impair real exchange rate adjustment in the short run, which gives rise to costs from two sources.
First, an individual country can only adjust its intra-union terms of trade by changing its prices and wages; it can minimize the resulting costs by choosing partners that are geographically proximate and similar in industrial structure. Second, the need to equilibrate the current accounts of the individual country in the long run and the union as a whole subjects each member to real exchange rate changes that it would not have experienced outside the union. These costs may be minimized by choosing partners that are similar in age structure, propensity to save and long-run factor productivity. Since the impacts of the latter are difficult to assess and the shocks affecting them infrequent, Mélitz assumes that the first source dominates and then considers the optimal order for new members to join the union.
Mélitz considers three implications of this analysis. First, not all asymmetric shocks argue against monetary union: if a nation's exportables resemble those of its geographical neighbours, the propensity for its output to be generally mispriced is much reduced. Second, Mundell's criterion for an OCA as a zone of labour mobility region with a single long-run Phillips curve loses much of its appeal once the long-run trade-off between unemployment and inflation is abandoned in favour of long-run price flexibility. Labour mobility unambiguously improves the case for a fixed rate, but so do capital mobility and short-run price flexibility. Third, the only decision an individual government can take unilaterally is to peg its exchange rate externally. Achieving an OCA in practice therefore requires concerted action, and its membership is usually determined from the start and largely by non-monetary considerations.
Although the applicability of this approach to potential monetary unions in the real world is clearly circumscribed by wider economic and political considerations, Mélitz nevertheless maintains that the progressive enlargement of a given political unit to form a wider common currency area provides stronger support than the more traditional approach for the common argument that most countries are too small to form OCAs by themselves.

A Suggested Reformulation of the Theory of Optimal Currency Areas
Jacques Mélitz

Discussion Paper No. 590, October 1991 (IM)