Cluster analysis is used to explore the performance of key macroeconomic variables in European countries that share the euro, from the inception of the currency in 2002 through to 2013. An original applied statistical approach searches for a pattern synthesis across a matrix of macroeconomic data to examine if there is evidence for country clusters and whether there is convergence of the cluster patterns over time. A number of different clusters appear and these change over time as the economies of the member states dynamically interact. This includes some new countries joining the currency during the period of examination. As found in previous research, Southern European countries tend to remain separate from other countries. The new methods used, however, add to an understanding of some differences between Southern European countries, in addition to replicating their broad similarities. Hypotheses are formed about the country clusters existing in 2002, 2006 and 2013, at key points in time of the euro integration process. These hypotheses are tested using the rigour of a bivariate analysis and the multivariate method of Qualitative Comparative Analysis (QCA). The results confirm the hypotheses of cluster memberships in all three periods. The confirmation analysis provides evidence about which variables are most influencing cluster memberships at each time point. In 2002 and 2006, differences between countries are influenced by their different Harmonised Index of Consumer Prices (HICP) and labour productivity scores. In 2013, after the crisis, there is a noticeable change. Long term interest rates and gross government debt become key determinants of differences, in addition to the continuing influence of labour productivity. The paper concludes that in the last decade the convergence of countries sharing the euro has been limited, by the joining of new countries and the circumstances of the global economic crisis. The financial crisis has driven divergences from pre-existing integration. Country convergence needs to be understood as a dynamic and multivariate concept. This is a significant development of convergence theory and is an addition to how the concept has been understood previously.
Bibliographical noteThis is an open access article distributed under the Creative Commons Attribution License (CC BY 4.0).
- cluster analysis