Exchange rate policy and economic convergence in the European Union

This paper tests for long-run macroeconomic convergence among European Union countries according to the various exchange rate regimes that have prevailed over the last forty years. Applying a recently developed test to monthly index of industrial production data, output convergence is confirmed or rejected depending on whether or not the first largest principal component based on benchmark deviations with respect to Germany is stationary or not. It is argued that this methodology has key advantages over existing cointegrating and common trends procedures. For most European Union countries, there is evidence of increased macroeconomic convergence during the 1990s where evidence is particularly strong for Belgium, France and the Netherlands. The evidence also indicates that the Snake era of the 1970s was more conducive towards convergence than the initial ERM period of 1979-92. Evidence of convergence is lacking for Austria, Finland and Sweden who joined the EU in 1995 and for a sample of non-EU countries.