Trade liberalisation and economic geography in CEE countries: the role of FDI in the adjustment pattern of regional wages

Abstract

This article studies the within-country regional effects of trade liberalisation in Central and Eastern European countries. CEE countries liberalised their trade with the European Union from the mid-1990s, while also receiving substantial foreign investment in the process. The first part of the period witnessed strong agglomeration effects in all of the countries, leading progressively to core–periphery type specialisation, and increasing regional wage differentials. In the second part of the period, however, there is notable evidence of a reversal in the relative regional specialisation, pointing to a U-shaped pattern of relative regional wages. Using the regional data for five CEE countries in 1990–2004 we argue that FDI inflows can be an important factor accelerating the observed regional adjustment process in the host country. First, we show that in four out of five CEE countries there is a significant U-shaped adjustment pattern of regional wages after they opened up to foreign trade. Second, we find robust econometric confirmation that in three of the five countries FDI has contributed significantly to faster adjustment of relative regional wages.

Acknowledgements

We gratefully acknowledge financial support from the European Commission within the PHARE ACE Programme ‘European integration, regional specialisation and location of industrial activity in accession countries’ (Contract No. P98-1117-R) and within the RTD 5th FP research project EURECO ‘The impact of European integration and enlargement on regional structural change and cohesion’ (Contract No. HPSE-CT-2002-00118).

Notes

 1. Similar results, with trade liberalisation favouring the development of EU-bordering regions, are also found by Resmini (2003) and Brülhart et al. (2004).

 2. Further support for the proposed divergence – convergence type of adjustment of relative wages after trade liberalisation is found also at the national level. Polanec (2004) examines the hypotheses of absolute and conditional convergence for a sample of 25 transition countries over 1990–2002. He finds a negative relationship between productivity growth and the pace of liberalisation at the initial stage of transition (1990–94), while at the later stages (1999–2002) evidence is found in favour of convergence of productivity levels among the countries under examination.

 3. A comprehensive overview of alternative NEG models is given in Fujita et al. (1999).

 4. See for instance Fidrmuc (2005) and Transition Report (2003) where very low migration intensity within transition countries is observed.

 5. The empirical specification also includes higher order values of time trends to test for the possible concavities of the temporal response of relative regional wages.

 6. Note that relative FDI inflows into the economic centre would further stimulate the initial divergence (increase the relative wage difference), while inflows targeting primarily the N/W border regions would stimulate the convergence in wages.

 7. Estonia could also be considered an EU-bordering country, but only conditionally as it shared only a sea border with the EU (Finland and Sweden).

 8. PHARE ACE Programme research project ‘European integration, regional specialisation and location of industrial activity in accession countries’ (Contract No. P98-1117-R) and the RTD 5th FP research project EURECO ‘The impact of European integration and enlargement on regional structural change and cohesion’ (Contract No. HPSE-CT-2002-00118). Both projects were financed by the European Commission.

 9. Another advantage of this study is that we have access to longer panels of regional data than previous studies on transition countries. In this way we believe we have the opportunity to study the complete adjustment pattern of regional economic activity and wages and the underlying factors affecting these processes.

10. We consider all enterprises where foreign ownership constitutes at least 10% of the ownership structure as foreign-owned enterprises.

11. Table A2 in the Appendix for more detailed descriptive statistics on the regional pattern of FDI.

12. In the late 1990s some governments (e.g. Bulgaria and Romania) started to actively attract foreign capital into disadvantaged and poorer regions, which is likely to benefit all regions.

13. In order to assess the robustness of the estimated coefficients on the impact of FDI on wages (rFDI _it ) we also ran FE estimates. We find the sign and significance (but not the size) of the rFDI _it coefficients very closely resemble those obtained by the RE estimations.

14. Time trend variables to the third and fourth power have also been included in various specifications further confirming the empirical relevance of the proposed U-shaped response curve.

15. Note however that Brülhart and Koenig (2006) estimate wage gradients for a different set of transition countries (Czech Republic, Hungary, Poland, Slovenia and Slovakia) and for a different period (1996–2000).

16. In Table 5 we omit reporting the coefficients on the lagged relative regional FDI (rFDI _it − 1 ).