Growth, integration, and regional disparities in the European Union

In this paper we challenge the ability of the conventional methods initiated by Barro and Sala-i-Martin in the early 1990s to detect actual convergence or divergence trends across countries or regions and suggest an alternative dynamic framework of analysis, which allows for a better understanding of the forces in operation. With the use of a SURE model and time-series data for eight European Union (EU) member states, we test directly for the validity of two competing hypotheses: the neoclassical (NC) convergence hypothesis originating in the work of Solow and the cumulative causation hypothesis stemming from Myrdal's theories. We also account for changes in the external environment, such as the role of European integration on the level of regional disparities. Our findings indicate that both short-term divergence and long-term convergence processes coexist. Regional disparities are reported to follow a procyclical pattern, as dynamic and developed regions grow faster in periods of expansion and slower in periods of recession. At the same time, significant spread effects are also in operation, partly offsetting the cumulative impact of growth on space. Similar results are obtained from the estimation of an intra-EU model of disparities at the national level, indicating that the forces in operation are independent of the level of aggregation. Our findings challenge the view of economic growth as the main driver for a reduction of regional disparities and contribute to the growing scientific evidence that points towards the need to rethink current EU-wide regional development policies.