Assessing European business cycles synchronization

Objectives: We analyzed the level of economic integration in Europe by analyzing the degree of growth cycle synchronization between 36 countries and its evolution over the past 17 years. Information whether the business cycles in a currency union are synchronized or not is of key importance for policymakers, because lack of synchronization will lead to sub-optimal common monetary policy. The article has three objectives: extend the literature on the business cycles synchronization by using dataset that includes countries that have never been analyzed before, test the robustness of the results to extraction and synchronization measures used and propose new method for assessing evolution of the synchronization over time. Data/methods: Quarterly GDP series from Eurostat database covering period 2000q1-2016q3 were used with two exceptions (industrial productions indexes for Bosnia and Herzegovina and Montenegro). Series were prepared by removing seasonal component using X13-ARIMA procedure. To assess robustness of synchronization tests results to alternative methods of detrending, business cycles were extracted using two filters: Corbae-Ouliaris ideal band filter and double Hodrick-Prescott filter. For assessing synchronization of the business cycles two methods were used: concordance index and cross-correlation function. Rolling cross-correlations at three lags were used to assess evolution of synchronization over time. Conclusions: Both concordance index and cross-correlations indicated that business cycles of most old EU members are synchronized with EU cycle. However, rolling cross-correlations suggested that this synchronization decreased after 2012. Majority of new EU members cycles were weakly or not at all synchronized with EU cycle until 2004/5. After 2004 most of them were synchronized in the same quarter but with greater variations between countries. For most of them after 2010/12 the degree of synchronization dropped significantly. These results are quite robust across the cycles extraction and synchronization measures used