The theory of optimum currency areas states that members of a monetary union avoid regional or national concentrations of unemployment by relying on factor mobility rather than real exchange rate flexibility to adjust to asymmetric shocks, while countries with separate currencies must achieve more of the necessary adjustment through real exchange rate changes. In Discussion Paper No. 555, Research Fellow Paul De Grauwe and Wim Vanhaverbeke present a statistical analysis of European regions and countries that sheds some light on whether the European Community is an optimal currency area. They find that labour mobility plays a major role in adjustment for regions within countries, while there is virtually no labour mobility at the national level, where the real exchange rate plays a much greater role. Divergences among output and employment growth rates are typically larger and more sustained at the regional than at the national level.

Optimists argue as in the European Commission's recent report, `One Market, One Money' that asymmetric shocks at the national level are now infrequent and that their frequency and hence the adjustment costs they entail will decline as the Community moves towards monetary union and closer economic integration. Pessimists argue in contrast that regions of the individual countries are clearly more economically integrated with each other than the Community's member countries, so the frequency of regional asymmetric shocks may even increase with economic integration. If a future European monetary union faces divergences in national output and employment similar to those observed today within individual countries, labour mobility may therefore have to play a greater role. Pessimists stress economic integration's regional concentration and agglomeration effects and cite the experience of the US, where shocks to particular industries concentrate their effects on particular regions.

De Grauwe and Vanhaverbeke distinguish a `Northern' model of regional development, typified by (West) Germany, in which relatively large inter-regional labour mobility and low divergences in output and employment lead to relatively uniform regional unemployment rates, and a `Southern' model, characterized by relatively immobile labour, pronounced divergences in output and employment and large regional concentrations of unemployment. They do not predict which model will prevail, but they note that either will entail costs for individual countries. The Northern model appears more attractive overall, but it would require relatively substantial movements of labour between countries; while the Southern model would lead to large regional divergences in unemployment, in which case not all European regions and countries will benefit from monetary union.

Is Europe an Optimum Currency Area? Evidence from Regional Data
Paul De Grauwe and Wim Vanhaverbeke

Discussion Paper No. 555, May 1991 (IM)