A New FDI Potential Index: Design and Application to the EU Regions

Abstract

The aim of this paper is to propose a new way of computing the FDI Potential Index to address the issue of FDI attractiveness at the EU regional level. This new index employs a sound way of selecting the variables involved in its construction, for which a factor analysis is performed. Accordingly, six factors (“economic potential”, “market size”, “labour situation”, “technological progress”, “labour regulation” and “competitiveness”) are identified. Next, by applying the methodology of composite indicators and considering different weighting and aggregation schemes, three versions (un-weighted linear, weighted linear and weighted geometric) of the new FDI Potential Index are computed. Afterwards, the comparison of the weighted linear version of the Potential Index with the conventional FDI Performance Index allows us to apply the United Nations Conference on Trade and Development (UNCTAD) FDI typology. The results reveal considerable heterogeneity among EU regions in terms of FDI attractiveness, and that regions belonging to the same group of the UNCTAD classification are highly concentrated from a geographical perspective. In view of these findings, we compute an additional version of both the FDI Potential and Performance indices, in which the geographical location of each region plays a key role. Based on these spatial indices, some general policy implications are drawn.

Keywords:

• FDI Potential Index
• FDI Performance Index
• UNCTAD FDI typology
• factor analysis
• European regions

Acknowledgements

We would like to thank W. Polasek and R. Sellner for providing us with the FDI regional database, the Swedish Institute for European Policy Studies (SIEPS), as well as two anonymous referees for their helpful comments and suggestions. The usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. From here on all references to FDI do really mean inward FDI.

2. As shown later in Table 1, there are also political and socio-cultural factors that affect the location of FDI. The focus here, however, is just on economic factors.

3. Rodríguez et al. (2009) also try to improve the computation of the FDI Potential Index. They do not apply, however, an endogenous way of obtaining the weights for the factors.

4. To be precise, all versions of the FDI Potential Index are weighted as there is no other possibility of computing it. By referring to the un-weighted version we make use of its conventional naming in which all the variables involved in the calculation have the same weight.

5. For a historical review of the theories of FDI see chapters 4 and 5 of Dunning and Lundan (2008).

6. This paradigm is somewhat based on the preliminary work by Hymer (1960) in that it pays attention to the reasons why some firms engage in foreign production.

7. Here, it is probably mandatory to include short comments about two alternative, and relatively novel, theories: the new theory of international trade and the so-called institutional approach. Building on the OLI paradigm, in the first one FDI is linked to variables such as market size, barriers to entry, transport costs and factor endowments. In the institutional approach, however, variables such as financial incentives, fiscal incentives and other economic incentives play a crucial role in explaining FDI.

8. MNEs motivation can lie on augmenting innovation capacities through transfers of knowledge from local firms to MNEs and vice versa, that is to say, by exploiting know-how related assets (Dunning, 2002).

9. This is, for instance, the case of the percentage of employment in high-tech sectors. Although it tries to somewhat measure the quality of human capital, it can also be misleading because, as it is the case of many MNEs, most workers here just carry out simply assembling tasks.

Although there is also an additional problem with our specification of the FDI drivers —that it cannot handle the differences between vertical and horizontal FDI— this is of no relevance here as our measure of FDI includes both vertical and horizontal FDI.

10. Although in Table 1 there are more factors related to “business facilitation”, we restrict our analysis to investment as it is the only one for which there are homogeneous data.

11. Not only for lack of availability but also for technical reasons when it comes to carrying out factor analysis, some potential FDI drivers were removed from the analysis: energy-manufacturing share, services share, unit labour costs, population density, corporate tax rate, % researchers, patents, internet, urban rural typology index, legal structure and security property rights index, and business regulations index.

12. Anyway, and for the sake of robustness, the sum scores approach is also employed. The results are roughly the same.

13. In any case, and to give a good illustration to the reader, let us take a specific region (e.g. Brussels/Region de Bruxelles) throughout the rest of the paper to show the results. Accordingly, it could be said that Brussels is characterised by its high economic potential and degree of technological progress, although shows weakness on its labour situation.

14. Appendix shows the results for the un-weighted and geometric aggregation versions of the new Potential Index.

15. The dividing value is always the (population) weighted average of each index.

16. Getting back to Brussels, it is a “front-runner” region as it presents high levels of both “PEI” and “POI”.

17. Villaverde and Maza (2012), for example, found evidence for the Spanish case of the existence of negative spatial spillovers, namely if a region improves its situation in an attraction factor for FDI, nearby regions will undergo a reduction in FDI flows.

18. As for Brussels, it keeps being a “front-runner” region when its geographical situation is considered, although in this case it seems it is not exploiting all its potential as FDI hosting region. Regarding factors, its weakest point is related to the labour market.

19. This result is in line with that of a recent paper by Sass and Fifekova (2011) stating that Central and Eastern European regions are becoming an increasingly popular destination for service-based FDI.

20. Some recent papers showing the existence of geographical concentration on FDI are, for example, Yavan (2010) for the Turkish case, and Holl et al. (2012) and Villaverde and Maza (2012) for the Spanish case.