Regional Patterns of Industry Location in Transition Countries: Does Economic Integration with the European Union Matter?

Abstract

Resmini L. (2007) Regional patterns of industry location in transition countries: does economic integration with the European Union matter?, Regional Studies 41, 747–764. Recent developments in international trade theory predict that increased globalization will be associated with increased locational concentration of economic activities, and hence increased specialization of national and regional economies. Relatively little empirical evidence exists about whether these predictions are correct, especially as regards Central and Eastern Europe. The paper analyses the interplay between economic integration and industry location in four candidate countries during the 1990s. It demonstrates that economic integration with the European Union has changed industry re-location processes within candidate countries, giving rise to a spatial organization of manufacturing productions which is less inward-oriented and more evenly distributed across regions than it was at the beginning of the transition process.

Resmini L. (2007) La distribution régionale de la localisation industrielle dans les pays en voie de transition: l'intégration économique à l'Ue, importe-t-il?, Regional Studies 41, 747–764. Les développements récents dans la théorie du commerce international prévoient que la course à la mondialisation s'associe à la concentration géographique des activités économiques et, par la suite, à la spécialisation accrue des économies nationales et régionales. Rares sont les preuves empiriques de la justesse de ces prévisions-là, surtout pour ce qui est de l'Europe centrale et orientale. Cet article cherche à analyser l'interaction entre l'intégration économique et la localisation industrielle dans quatre pays-candidats pendant les années 1990. On démontre que l'intégration économique à l'Ue a modifié les processus de relocalisation industrielle au sein des pays-candidats, ce qui a entraîné une organisation géographique du tissu industriel qui s'avère moins nombriliste et distribuée plus uniformément à travers les régions qu'il ne l'était au début de la période de transition.

Localisation industrielle Intégration économique Pays en voie de transition

Resmini L. (2007) Regionale Muster der industriellen Umsiedelung in U¨bergangsländen: kommt es auf eine wirtschaftliche Integration in die EU an?, Regional Studies 41, 747–764. Die jU¨ngsten Entwicklungen in der Welthandelstheorie haben zur Prognose gefU¨hrt, dass eine erhöhte Globalisierung mit einer erhöhten lokalen Konzentration der Wirtschaftstätigkeiten einhergehen wird, wodurch sich die Spezialisierung der nationalen und regionalen Wirtschaften erhöhen wird. Hinsichtlich der Richtigkeit dieser Prognosen liegen nur relativ wenige empirische Belege vor, insbesondere was Mittel- und Osteuropa anbelangt. In diesem Beitrag werden die Wechselwirkungen zwischen Wirtschaftsintegration und Industriestandort in vier Kandidatenländern während der neunziger Jahre untersucht. Gezeigt wird, dass die wirtschaftliche Integration in die EU zu einer Veränderung der industriellen Umsiedelungsprozesse in den Kandidatenländern und zu einer räumlichen Organisation der produzierenden Industrie gefU¨hrt hat, die weniger nach innen orientiert und gleichmäßiger U¨ber die Regionen verteilt ist als zu Beginn des U¨bergangsprozesses.

Industriestandort Wirtschaftsintegration U¨bergangsländer

Resmini L. (2007) Modelos regionales de ubicación industrial en países de transición: ¿Es importante la integración económica en la UE?, Regional Studies 41, 747–764. Los últimos avances sobre la teoría del comercio internacional vaticinan que una mayor globalización estará vinculada a una mayor concentración de ubicación de actividades económicas, y por ende ocurrirá una aumento en la especialización de las economías nacionales y regionales. Existen pruebas empíricas relativamente escasas para saber si estas predicciones son correctas, sobre todo en lo que concierne a Europa central y oriental. En este artículo analizo la interacción entre la integración económica y la ubicación industrial en cuatro países candidatos durante la década de los noventa. Demuestro que la integración económica en la UE ha cambiado los procesos de reubicación de la industria en los países candidatos, causando la aparición de una organización espacial de producción manufacturera que está menos orientada hacia dentro y distribuida más equilibradamente en las regiones de lo que estaba al principio del proceso de transición.

Ubicación de la industria Integración económica Países de transición

Keywords:

• Industry location
• Economic integration
• Transition countries

Notes

1. The figures on trade and FDI in transition countries have been drawn from Eurostat (various issues) and UNCTAD (various years).

2. Besides trade reorientation and FDI, the other ‘stylized facts’ of the transition years in Central and Eastern European and former Soviet Union countries are: a fall in output, internal capital shortage, labour force adjustments (from public to private sector, from employment to inactivity, among occupations), structural changes in the economy, the collapse of the institutions, and increased poverty and inequality, often associated with a sharp deterioration in some social indicators, such as life expectancy and school enrolment rates. Campos and Coricelli (2002) took stock of the first 10 years of transition in the CEECs and CIS and of the various theories and explanations proposed. For another comprehensive survey on the economic and social costs of transition in the CEECs, see UN/ECE (2002).

3. Scale economies, ‘love for variety’ and imperfect competition are the main ingredient of these models. Fujita et al. (2000) and Fujita and Thisse (2002) provide extensive surveys of these models.

4. In a sense, this corresponds to a revival of the concepts of ‘circular and cumulative causation’ developed by Myrdal (1957) and formalized by Kaldor (1970) and of ‘backward and forward linkages’ by Hirschman (1958).

5. For a comprehensive survey on how foreign firms affect host country economies at the theoretical and empirical level, see Blomstrom and Kokko (1997) and Alfaro and Rodriguez-Clare (2003), respectively. UN/ECE (2000) summarizes the existing empirical evidence on FDI and spillovers in the case of CEECs.

6. Structural change usually relates to a broader set of issues, including institutional changes, changes in the composition of output, employment and exports, as well as changes in income and wealth distribution. All these issues are of great relevance for a fuller analysis of structural changes in transition economies. Such analysis, however, is beyond the scope of this paper. Those interested in these topics can refer to Journal of Economic Perspectives (2002) and UN/ECE (2000).

7. In the present paper, regions are administrative units at the NUTS III level, as classified by Eurostat. Border regions are those eligible for PHARE-CBC programs (European Commission, 2001).

8. Tallinn is not considered as a separate district. Thus, the whole region in which it belongs, Pohja-Eesti, is considered as the capital district. Therefore, Tallinn's performance is not completely comparable with those of the other capital districts included in the sample.

9. A finer sectoral classification was not possible because of the lack of homogeneous data among countries. See the Appendix for regions' classification and manufacturing sectors' description. The REGSPEC database covers five candidate countries and includes several variables at the regional level (employment, gross domestic product, the number of domestic and foreign firms, population, infrastructures, average earnings, etc.). It has been built on and updated with the financial support of the European PHARE-ACE Program 1998 and 5th Framework Program.

10. In Hungary the highest concentration of manufacturing activity was in Budapest, which accounted for about 30% of total manufacturing employment. Other internal regions lagged behind, with a share of less than 20%. The opposite trend characterized Bulgaria and Romania.

11. The region's geographical characteristics have indubitably contributed to the strengthening of manufacturing agglomeration patterns in Tallinn and its surroundings.

12. Firms are here accounted for in numbers, since data on employment and or output are not available at the desired level of disaggregation for all the countries included in the sample and throughout the 1990s. This entails that one cannot consider the relative economic weight of foreign firms in each region and country, but only discuss their agglomeration patterns.

13. Although most of the economic linkages inherited from the centrally planned period have been destroyed by the transition process, cross-border issues are of critical importance in some countries because of the historical peculiarities of the border configurations (UN/ECE, 2003).

14. Given the time span of the sample, requiring integration with the EU to eliminate the effects of distance to the capital cities would be a too strong hypothesis.

15. The available data could not be used to distinguish the type of externality generated by FDI, i.e. pecuniary versus technological externalities. However, this finer distinction is not relevant given the purpose of the study.

16. This problem was more severe for small countries, such as Estonia and Hungary. Estonia's only international airport is located in Tallinn, from which main national and international rail, road and maritime lines depart.

17. Residuals, by definition, are the portion of the variation of the dependent variable not explained by the explanatory variables. Thus, in the present case, they pick up the effects of FDI not related to distance and accessibility variables on industry location processes.

18. In this respect, equation (4) does not seem able to explain industry location patterns within BAC regions. The goodness-of-fit of the model is poor relative to the other subsamples of border regions, and the explanatory variables are either not significant or erroneously signed. Controlling for country-specific effects does not improve the results.

19. This explanation holds for Estonia and Hungary, as demonstrated in the third section.

20. The estimation results for BEU regions are very weak because of the lack of sufficient degrees of freedom.

21. Complete estimations on the parameters to the explanatory variables are not shown, but are available from the author upon request.

22. Since capital districts are located internally in Hungary, Bulgaria and Romania, while Tallinn is located in a region bordering on the former EU-15, the results for BEX and BAC regions are analogous to those in columns (5) and (6) in Table 3.

23. The estimations in cumulative differences did not change when capital cities were not considered. The results are therefore not reported but are available from the author upon request.