Does the EU convergence machine still work?

Article published in the Economic Review of June 2020

Economic convergence has been one of the explicit goals of the EU from its very beginnings. The prospect of higher living standards has been a major attraction of EU membership and is arguably an antidote against rising scepticism with respect to the EU.

The article assesses the achievements of the EU with respect to income convergence, expressed in terms of GDP per capita (in purchasing power standards). Income or real convergence typically refers to the process of poorer countries and regions catching up towards the income levels of richer ones.

The article differs from the existing literature in that:

  • it covers all EU Member States;
  • it investigates both national and regional convergence;
  • it adopts a long-term perspective;
  • the focus is on large regions that correspond to countries’ first tier of sub-national government (régions/gewesten in Belgium, Länder in Germany, etc.);
  • the analysis takes the commuting zones of metropoles more closely into account. To attenuate the bias on the GDP per capita caused by the commuters in that they contribute to those regions’ GDP while being left out of their population head count, 12 metropolitan regions are extended to encompass their main commuting zones. These include notably London, Berlin, Amsterdam, Vienna, Prague and Brussels (whose commuting zone according to Eurostat includes Aalst, Halle-Vilvoorde and Walloon Brabant).

Results for the EU

The article finds evidence of relatively strong convergence of incomes across countries and regions in the EU over the long term. However, the process has not always been smooth. While the EU “convergence machine” has worked most of the time and for most regions, sometimes and for some places it has sputtered. Convergence has been strongest during high-growth periods, during the early stages of EU integration among the old Member States, and around the accession of the Central and Eastern European countries. Crisis periods were marked by slowing convergence or even divergence. The global financial crisis and European sovereign debt crisis heralded a period of severe economic underperformance in much of Southern Europe.

EU-wide convergence of regional incomes since 1996 has benefited from convergence of incomes between countries, while within-country income disparities remain substantial and have even widened slightly over time. Metropolitan regions – and even more so capital regions – have grown faster than average, thereby contributing to regional convergence across EU countries but also to within-country disparities. Agglomeration effects, such as a concentration of higher-productivity activities and innovation, have likely played a role here.

Insights for Belgium

Whether or not regions are corrected for commuting zones, disparities between Belgium’s richest and poorest region, Brussels and Wallonia respectively, do not appear exceptional relative to other EU countries’ regional disparities.

Over the period 1996-2018, Flanders was “pulling away”, Brussels was “cooling off” and Wallonia was somewhat “lagging behind”. Flanders succeeded in growing faster than the EU average, even from an above-average GDP per capita in 1996. Starting from a very high income level, Brussels did not keep pace with average EU growth. Nevertheless, when we include its commuting zone, its income growth was still faster than what could be expected based on its initial income level. The analysis suggests that this better-than-expected performance derives mostly from its status as a capital region. Starting from a below-EU average income, GDP per capita also grew less in Wallonia than the EU average. However, the extent of this underperformance is relatively limited, and even more so when the (relatively rich and dynamic) Walloon Brabant is not included in the commuting zone of Brussels. Wallonia’s lagging position can be partly explained by its relatively lower share of higher educated people of working age and the absence of a large metropole (more than 500,000 inhabitants) on its own territory. The analysis does not consider the effect of neighbouring metropoles outside of a region’s own borders (for example, the impact of the metropole of Brussels on Wallonia and Flanders).

Broader policy implications

First, investing in physical and human capital improves growth performance at both national and regional level. Second, convergence in the run-up to the global financial crisis proved to be unsustainable, pointing to the need to avoid excessive imbalances. Completing the EU integration process, including the Banking Union and the Capital Markets Union, could help in this regard.

Third, heterogenous patterns of regional development suggest a need for place-sensitive regional policies, tailor-made to the specific situation of each region. Indeed, optimal policy instruments arguably differ depending on whether regions are lagging behind or catching up. They may also differ between capital, metropolitan and non-metropolitan regions. Next to national and regional authorities, the EU could play a role by boosting the efficiency of its cohesion policy.