Unemployment Volatility and Regional Specialization in the European Union

Abstract

Ezcurra R. Unemployment volatility and regional specialization in the European Union, Regional Studies. This paper examines the link between unemployment volatility and the sectoral composition of economic activity in the regions of the European Union over the period 1980–2004. To that end, different spatial econometric techniques are used that allow the investigation of the role played in this context by spatial effects and geographical spillovers. The results show that unemployment volatility is positively related to regional specialization in the European Union. This finding does not depend on the use in the analysis of absolute or relative specialization measures.

Ezcurra R. L'instabilité du chômage et la spécialisaiton régionale dans l'Union européenne, Regional Studies. Cet article cherche à examiner le lien entre l'instabilité du chômage et la structure sectorielle de l'activité économique dans les régions de l'Union européenne entre 1980 et 2004. Dans ce but, on emploie diverses techniques économétriques qui permettent un examen du rôle joué dans ce contexte par les effets et les retombées géographiques. Les résultats laissent voir une corrélation étroite entre l'instabilité du chômage et la spécialisation régionale dans l'Union européenne. Ce résultat ne dépend pas de l'emploi dans l'analyse des mesures de spécialisation absolues ou relatives.

Chômage Instabilité Spécialisaton Régions Union européenne

Ezcurra R. Volatilität der Arbeitslosigkeit und regionale Spezialisierung in der Europäischen Union, Regional Studies. In diesem Artikel wird die Beziehung zwischen der Volatilität der Arbeitslosigkeit und der sektoralen Zusammensetzung der Wirtschaftstätigkeit in den Regionen der Europäischen Union im Zeitraum von 1980 bis 2004 beschrieben. Zu diesem Zweck kommen verschiedene räumliche ökonometrische Techniken zum Einsatz, die eine Untersuchung der Rolle von räumlichen Effekten und geografischen Übertragungen in diesem Kontext ermöglichen. Aus den Ergebnissen geht hervor, dass die Volatilität der Arbeitslosigkeit in der Europäischen Union in einer positiven Beziehung zur regionalen Spezialisierung steht. Dieses Ergebnis hängt nicht davon ab, ob bei der Analyse absolute oder relative Maßstäbe der Spezialisierung eingesetzt werden.

Arbeitslosigkeit Volatilität Spezialisierung Regionen Europäische Union

Ezcurra R. Volatilidad de las tasas de desempleo y especialización regional en la Unión Europea, Regional Studies. Este trabajo estudia la relación entre la volatilidad de las tasas de desempleo y la composición sectorial de la actividad económica en las regiones de la Unión Europea a lo largo del período 1980–2004. Para ello el autor utiliza diferentes técnicas de econometría espacial que permiten determinar la importancia en este contexto de la localización geográfica de las distintas regiones. El análisis llevado a cabo muestra que la volatilidad de las tasas de desempleo está correlacionada positivamente con el nivel de especialización regional. De hecho, este resultado no depende de la utilización en el análisis de medidas de especialización absolutas o relativas.

Desempleo Volatilidad Especialización Regiones Unión Europea

Keywords:

• Unemployment
• Volatility
• Specialization
• Regions
• European Union

JEL classifications:

• E32
• R11
• R12

Acknowledgements

The author is grateful to the Editors of Regional Studies and two anonymous referees for their insightful comments on earlier versions of this paper. This research benefited from the financial support of the Spanish Ministry of Science and Innovation (Project Number ECO2008-05072-C02-02/ECON).

Notes

During the last fifteen years, different studies have examined the links between economic integration, productive specialization, and the degree of symmetry in the economic cycle (for example, Krugman, 1993; Frankel and Rose, 1998; and Kalemli-Ozcan et al., 2001).

The analysis of the different sources of economic instability is beyond the scope of this paper. For further details, see Siegel et al. (1995).

NUTS is a hierarchical classification of sub-national spatial units established by Eurostat. In this classification, NUTS-0 corresponds to country level, while increasing numbers indicate increasing levels of sub-national disaggregation.

Spatial autocorrelation can be defined as the coincidence of value similarity with locational similarity (Anselin, 2001).

The author decided to use the level of employment as the reference variable instead of production to avoid problems of currency conversion inherent in value data (Brülhart and Traeger, 2005).

As a sensitivity analysis, an alternative index similarly defined, except using the square rather than the absolute value, was also employed in expression (4). The results, however, were very similar in both cases.

The participation rate was not employed due to the lack of data on the working-age population for the sample regions.

In order to check the validity of the decision rule used to choose this specification, an alternative model including a spatial autoregressive structure in the error term (spatial error model) was also estimated. The results are presented in Table A3 in the Appendix. It is worth noting that the spatial-lag model achieves a better fit than the spatial error model in all cases, thus confirming that the spatial-lag model is the preferred specification in this context.

Although there are some exceptions, most of these studies use a variance-based statistic applied to employment data. For further details on this issue, see Dissart (2003).

The possible presence of a non-linear relationship between the specialization indexes and the dependent variable was also investigated. To that end, different alternative specifications, including the square of the two specialization measures employed, were considered. Nevertheless, the coefficients of these variables were not statistically significant in any case.