Industry mobility and concentration in the European Union* by Salvador Barrios and Eric Strobl ** DOCUMENTO DE TRABAJO 2002-19

October 2002

This research has benefited from financial support through the RTN research project “Specialization versus diversification: the microeconomics of regional development and the spatial propagation of macroeconomic shocks in Europe” of the European Commission (grant No. HPRN-CT-2000-00072). Thanks to Glenn Ellison for valuable advice and Antonio Teixeira for very useful suggestions. Also many thanks to Marius Brülhart, Martin Hallet and Colin Webb for the generous provision of the data.

CORE, 34 Voie du Roman Pays 1348 Louvain-La-Neuve, Belgium. E-mail: barrios@core.ucl.ac.be and strobl@core.ucl.ac.be

Los Documentos de Trabajo se distribuyen gratuitamente a las Universidades e Instituciones de Investigación que lo solicitan. No obstante están disponibles en texto completo a través de Internet: http://www.fedea.es/hojas/publicaciones.html#Documentos de Trabajo These Working Documents are distributed free of charge to University Department and other Research Centres. They are also available through Internet: http://www.fedea.es/hojas/publicaciones.html#Documentos de Trabajo

Abstract

We study the pattern of concentration of industries in EU countries and regions between 1972 and 1995. We find that changes in concentration levels were mainly due to industry mobility through randomness in growth rates rather than historical accidents and past levels of concentration as often argued by the New Economic Geography literature.

JEL classification: F14, F15, R12 Key words: Industry concentration, industry mobility, Europe.

1. Introduction

According to Krugman (1991), the concentration of industrial activities across space is primarily influenced by historical accidents together with agglomeration forces, such as increasing returns to scale, high final demand or strong backwardforward linkages, which systematically drive the growth and decline of industrial centers. This framework has frequently been used to assess the possible implications of the European integration process on the location of economic activities, see, for instance, Krugman and Venables (1996) for a theoretical analysis and Amity (1998), Brülhart (2000), Hallet (2000) and Midelfart-Knarvik et al (2000) for empirical evidence. Specifically, these authors argue that concentration is likely to rise due to economic integration. Importantly, however, Dumais et al. (2002) have recently shown for the US that industries are extremely mobile and that non-historical factors attributable to randomness in industries’ location play a major role in geographical concentration. We investigate whether such a conclusion can also be drawn for the European case.

2. Empirical Analysis

We utilize two data sources covering the EU 15 members: (1) employment data for a panel of 36 manufacturing industries over the period 1972-95 from the OECD’s STAN database¹, and (2) Gross value added (GVA) data at constant prices from the European Commission REGIO database for a panel of 17 broader industries, namely manufacturing (9 sectors), agriculture, construction, energy, and services (5 sectors), located in the 113 EU NUTS 2 regions². While the former allows more detailed sectoral classification for manufacturing and is comparable to that used in most other empirical studies of industry location in Europe, the latter data enables us to consider the location in the context of the continuing rise in services activities (from 57 per cent to 67 per cent of total gross value added between 1980 and 1995) in Europe relative to the decline of other traditional industries like manufacturing (from 26 per cent to 21 per cent) and agriculture (from 4 per cent to 2 per cent). Additionally, since the kind of agglomeration forces generally depicted in the economic literature are mostly local in nature, it seems appropriate to also use regional classifications that go beneath national boundaries.

¹ The data for Ireland was not available in STAN and we had to instead draw on the Forfás employment survey which is an exhaustive employment survey of all manufacturing firms in Ireland.

² Those regions are located in continental Europe only, see Hallet (2000) for more details.

We start from the simple geographic concentration index described by Dumais et al. (2002):

\[\mathbf {G} _ {\mathrm{it}} = \boldsymbol {\Sigma} _ {\mathrm{c}} (\mathbf {s} _ {\mathrm{ict}} - \mathbf {s} _ {\mathrm{ct}}) ^ {2}\tag{1}\]

where is the share of industry i in country c and is the country’s share of employment in the average industry at time t. is thus the sum of squared deviations of the industry’s country share of employment from the country’s share of employment in the average industry. To derive an aggregate measure of we consider its simple mean across industries, . The time pattern of this variable is given in Table 1 using both STAN and REGIO data. As one would naturally expect the level of concentration is higher when considering countries rather than regions. According to columns 1 and 2, and as been found by previous studies, the geographic concentration of manufacturing industry has increased over the whole period. In addition, the rise in the concentration appears to be deeper when using STAN rather than REGIO probably indicating the possibility that some important productivity adjustments have occurred over the period. The third column of table 1 shows in turn that the concentration of industries is lower and relatively more stable when including other non-manufacturing industries which simply translates into the fact that services activities are more evenly distributed than manufacturing industries. Aggregate agglomeration patterns can mask considerable equilibrating forces involving large changes in industry location as shown by Dumais et al (2002). Accordingly, one can treat the change in the share of industry i in country c as a function of (a) the difference between initial country-industry share and the country’s average share and (b) the growth of the country’s average employment share in a simple regression:

\[\mathrm {s_ {ict} - s_ {ict + 1} = \alpha+ \beta(s_ {ict} - s_ {ct})+ \gamma(s_ {ct + 1} - s_ {ct})+ \varepsilon_ {ist}}\tag{2}\]

where β, and γ are estimated coefficients and is an estimated error term orthogonal to the regressors. Using (1) and (2) one can easily show, see Dumais et al (2002), that the change in G over time can be decomposed as follows:

\[\mathrm{G} _ {\mathrm{t+1}} - \mathrm{G} _ {\mathrm{t}} = (2 \beta + \beta^ {2}) + (1 / \mathrm{I}) \Sigma_ {\mathrm{cs}} \varepsilon_ {\mathrm{ist}} ^ {2}\tag{3}\]

where I is the total number of industries in the data. The first term in (3), (2β + , can be thought of as the effect of mean reversion on changes in geographic concentration and depends on the extent of the gap between and . If β is positive then industry centers are growing, thus causing an increase in concentration. If, in contrast, β is negative then current industry centers are declining in importance (relative to non-centers), thus causing mean reversion. The second term of (3), , which is always positive, captures the effect of randomness on changes in geographic concentration and reflects the degree of heterogeneity in changes in country-industry employment for countries with initial similar shares. Our results of estimating (2) and calculating (3) are given in Table 2. For the entire sample period β is estimated to be –0.06 with the STAN data, thus suggesting that countries which in 1972 had a low share of a typical EU industry experienced on average a rise in their share by 6 per cent by 1995. While the fact that is negative implies that mean reversion took place in the location of European industry, this equilibrating process, reducing agglomeration by 12.34 per cent, was not enough to counteract the heterogeneity across countries’ and industries’ growth since randomness on its own increased agglomeration by 57.63 per cent. We also divided our data into four sub-periods and conducted the same exercise. We observe some differences across sub-periods, especially between 1985 and 1990 (which corresponds to the setting up of the Single Market Program) where mean reversion has had a larger impact than randomness. The results of our decomposition using REGIO data are shown in the lower two parts of Table 2. For the manufacturing industries the β coefficient is found to be negative and is larger in absolute terms relative to the one found with the country-level database for the overall period. The results by period also show that mean reversion has been taking place continuously - a result rather different to the one we got when using the country-level data. This evolution appears to be persistent and more pronounced here, implying that agglomeration forces and the role of history for industry location are especially important for regions rather than countries. Despite this, mean reversion appears to have been offset by randomness and industry mobility and the overall picture thus shows a rise in concentration. Mean reversion dominates only during the 1985-1990 period as for the country-level data. In the bottom part of table 2 we consider the results including the other nonmanufacturing industries. Here also the evolution of the β term indicates a decrease in concentration, even more pronounced than when considering manufacturing industry alone. However, the decomposition into mean reversion and randomness indicates that the later has been more than compensated by the former yielding to a relative stability in concentration.

Conclusion

Our results then bring two important conclusions: (1) there has been mean reversion in the geographical concentration of industries contrary to the general arguments made in the New Economic Geography literature concerning the possible impact of European economic integration; (2) the observed rise in concentration of manufacturing activities is generally due to randomness in the distribution of countries’ and regions’ industrial growth, a feature which has not been yet considered by the empirical literature concerning the European case.

References

1. Amity, M. (1999). “New Trade Theories and Industrial Location in the EU: A Survey of Evidence”, Oxford Review of Economic Policy, 14 (2), 45-53.
2. Brülhart, M., (2001). Evolving geographical concentration of European manufacturing industries, Weltwirtschaftliches Archiv, 137(2), 215-43.
3. Dumais, G., G. Ellison and E.L. Glaeser (2002). “Geographic Concentration as a Dynamic Process”, Review of Economic and Statistics, 84 (2), 193-204.
4. Hallet, M. (2000). “Regional Specialisation and Concentration in the EU”, Economic Paper 141, Directorate-General for Economic and Financial Affairs, European Commission.
5. Krugman, P. (1991). “Increasing Returns and Economic Geography”, Journal of Political Economy, 99 (3), 483-499.
6. Krugman, P. and A.J. Venables (1996). “Integration, Specialization, and Adjustment”, European Economic Review, 40 (3-5), 959-67.
7. Midelfart-Knarvik, K.H., H.G. Overman, S.J. Redding and A.J. Venables (2000). The Location of European Industry, Report prepared for the Directorate General for Economic and Financial Affairs, European Commission, Brussels.

Tables

Table 2- Decomposition of Geographical Concentration Index

DOCUMENTOS DE TRABAJO

References

1. 2002-19: “Industry Mobility and Concentration in the European Union”, Salvador Barrios y Eric Strobl.

References

1. 2002-18: “The Closed-Form Solution for a Family of Four-Dimension Non-Linear MHDS”, José Ramón Ruiz-Tamarit.

References

1. 2002-17: “Multiplicity, Overtaking and Convergence in the Lucas Two-Sector Growth Model”, José Ramón Ruiz-Tamarit

References

1. 2002-16: “A Matching Model of Crowding-Out and On-the-Job Search (with an application to Spain)”, Juan J. Dolado, Marcel Jansen y Juan F. Jimeno.

References

1. 2002-15: “Youth unemployment in the OECD: Demographic shifts, labour market institutions, and macroeconomic shocks”, Juan F. Jimeno y Diego Rodríguez-Palenzuela

References

1. 2002-14: “Modelling the linkages between US and Latin American stock markets”, José L. Fernández-Serrano y Simón Sosvilla-Rivero

References

1. 2002-13: “Incentivos y desigualdad en el sistema español de pensiones contributivas de jubilación”, Juan F. Jimeno.

References

1. 2002-12: “Price Convergence in the European Union”, Simón Sosvilla-Rivero y Salvador Gil-Pareja.

References

1. 2002-11: “Recent Trends in Occupational Segregation by Gender: A Look Across The Atlantic”, Juan J. Dolado, Florentino Felgueroso y Juan F. Jimeno.

References

1. 2002-10: “Demand- and Supply-Driven Externalities in OECD Countries: A Dynamic Panel Approach”, Salvador Barrios y Federico Trionfetti.

References

1. 2002-09: “Learning by Doing and Spillovers: Evidence from Firm-Level Panel Data”, Salvador Barrios y Eric Strobl.

References

1. 2002-08: “Interdependent Growth in the EU: The Role of Trade”, María García-Vega y José A. Herce.

References

1. 2002-07: “Export market integration in the European Union”, Salvador Gil-Pareja y Simón Sosvilla-Rivero.

References

1. 2002-06: “Early mortality declines at the dawn of modern growth”, Raouf Boucekkine, David de la Croix y Omar Licandro.

References

1. 2002-05: “Nearest-Neighbour Predictions in Foreign Exchange Markets”, Fernando Fernández-Rodríguez, Simón Sosvilla-Rivero y Julián Andrada-Félix

References

1. 2002-04: “Demografía, empleo, salarios y pensiones”, Juan F. Jimeno.

References

1. 2002-03: “La reforma de la negociación colectiva en España”, Samuel Bentolia, Juan F. Jimeno.

References

1. 2002-02: “Efficiency Spillovers from Foreign Direct Investment in the EU Periphery: A comparative study of Greece, Ireland and Spain”, Salvador Barrios, Sophia Dimelis, Helen Louri y Eric Strobl

References

1. 2002-01: “Non-Linear Forecasting Methods: Some Applications to the Analysis of Financial Series”, Oscar Bajo-Rubio, Simón Sosvilla-Rivero y Fernándo Fernández-Rodríguez.

TEXTOS EXPRESS

References

1. 2001-01: “La reforma de las pensiones en el contexto internacional”, José A. Herce y Juan F. Jimeno.

References

1. 2000-03: “Efectos sobre la inflación del redondeo en el paso a euros”, Mario Izquierdo y Simón Sosvilla-Rivero.