The single market and geographic concentration in the regions of the European Union

Abstract

This note examines industrial concentration in the regions of the European Union during the period following the implementation of the Single Market. To achieve this objective, we have applied a new methodology in this literature based on the second degree inverse stochastic dominance concept. The results obtained reveal an increase in geographical concentration in most industrial activities between 1992 and 1999.

Acknowledgement

The author wishes to acknowledge the financial support from Spanish MEC (Project SEJ2005-08738-C02-01 and Programme Juan de la Cierva).

Notes

¹ Exceptions we could mention are Hallet (2002) and Brülhart and Traeger (2005).

² Various authors have used absolute concentration indices (Midelfart-Knarvik et al ., 2002). This, however, implies the assumption that lack of concentration corresponds to a situation where the employment and/or the value added shares of each sector are exactly the same across all the territorial units considered, irrespective of their size. As reported in Combes and Overman (2004), however, this is hard to reconcile with most market-based location models.

³ Lack of data forced us to omit from our analysis the countries incorporated into the EU in 2004 and 2007, the Länder of the former East Germany, the French overseas departments and the Spanish territories in North Africa.

⁴ With respect to this issue, it must be borne in mind that as the value of the parameter δ increases, so does the importance attributed in the analysis to the lower end of the distribution. However, as shown by Cowell and Flachaire (2001), the dispersion measurement can be strongly affected when there exists high sensitivity to lower values of the variable under consideration. To overcome this problem, we chose to consider values of δ between 1.5 and 15.