The divergence of growth and inflation rates across EMU in recent years has reignited the debate as to whether Europe is really an "optimum currency area" in which monetary union yields net benefits. But answering this question is complicated inter alia by the further controversy over whether economic integration and, in particular, monetary union tend to cause convergence or divergence of business cycles. Past studies have found convergence under the Exchange Rate Mechanism (ERM) regime of the 1980s, but the final verdict on the 1990s and especially on the period since the start of EMU is still pending. Several studies, such as Böwer and Guillemineau (2006), compute the unweighted average of countries' bilateral correlations and find convergence in the 1990s and divergence since then, but this is largely caused by an outlier in Greece. We argue that similar to the treatment of inflation in monetary policy, for which a country's inflation rates are weighted by the relative size of a country's private consumption, one has to look at weighted GDP growth rates. Using these, we find synchronisation, i.e. a further increase in correlation, both during the 1990s and since the start of EMU, but only the former change is significant. These findings are subsequently confirmed by the development of inflation dispersion over time. We infer that, unlike during the run-up to EMU, the introduction of a common monetary policy itself has not brought about a great reduction in business cycle heterogeneity, and synchronisation will probably be limited in the coming years as well. This means that policy-makers at the national level need to do more to improve their economies' flexibility, in order to make them better able to cope with the remaining heterogeneity in output and inflation.