Portal de Periódicos da CAPES

Increasing Returns and the Spatial Structure of French Wages

2010; Routledge; Volume: 5; Issue: 1 Linguagem: Inglês

10.1080/17421770903511338

1742-1780

Sylvain Barde,

Regional Economic and Spatial Analysis

Abstract Abstract New Economic Geography presents increasing returns to agglomeration as a central explanation for concentration of economic activity. The estimation of the size of these effects remains, however, a standing issue in the field. The focus of this study is to investigate the presence of increasing returns to agglomeration in the spatial structure of wages in France, using the methodology developed by Fingleton and initially used in the UK. The central finding is the statistically significant presence of such returns to density for France, as was the case for the UK in the original study. Compared to Fingleton's original work, it is shown that returns to density play a larger role in explaining French labour productivity, while commuting plays a smaller role than in the UK. Rendements croissants et structure spatiale des salaires français RÉSUMÉ La nouvelle économie géographique présente les 'rendements croissants d'agglomération' comme une variable explicative privilégiée de la concentration spatiale de l'activité économique. Un des enjeux empiriques de la nouvelle économie géographique reste cependant l'estimation de leur taille. Dans cette étude, nous cherchons à évaluer la présence de tels rendements croissants d'agglomération dans la structure spatiale des salaires français, en utilisant la méthodologie d'estimation développée pour le Royaume-Uni par Fingleton. Le résultat central de notre étude est la présence statistiquement significative de rendements croissants d'agglomération sur les zones d'emploi françaises, du même ordre de grandeur que ceux qui avaient été établis pour le Royaume-Uni dans l'étude originale. De plus, par rapport à l'analyse originale de Fingleton, nous montrons que les retours à la densité sont plus déterminants dans l'explication de la productivité du travail en France, tandis que les déplacements domicile—travail y jouent un rôle moins important qu'au Royaume-Uni. Rendimientos crecientes y la estructura espacial de los sueldos franceses RÉSUMÉN La nueva geografía económica presenta los rendimientos crecientes de aglomeración como una explicación central para la concentración de actividad económica. No obstante, la estimación de la magnitud de estos efectos continúa siendo una cuestión pendiente en el campo. El enfoque de este estudio es investigar la presencia de rendimientos crecientes de aglomeración en la estructura espacial de los sueldos en Francia, utilizando la metodología desarrollada por Fingleton y empleada inicialmente en el Reino Unido. El hallazgo central es la presencia estadísticamente significativa de tales rendimientos de densidad en Francia, como fue el caso del Reino Unido en el estudio original. En comparación con el trabajo original de Fingleton, se muestra que los rendimientos de densidad desempeñan un papel más importante en explicar la productividad laboral francesa, mientras que viajar largas distancias al trabajo diariamente tiene una función menor que en el Reino Unido. Keywords: Spatial econometricsincreasing returnsspatial autoregressive modelJEL CLASSIFICATION : C21R12R23 Acknowledgements The author has benefited from discussions with Bernard Fingleton and comments from Julie Le Gallo on an earlier version of this paper. Furthermore, the author is grateful to seminar participants at the ERSA Liverpool 2008 conference as well as two anonymous referees for their helpful comments. The usual disclaimers apply. Notes 1. For a more complete explanation of how the returns to scale stem from the intermediate good, the reader is referred to the initial theoretical model in the appendix in Fingleton (2003 Fingleton, B. 2003. Increasing returns: evidence from local wage rates in Great Britain. Oxford Economic Papers, 55: 716–739. [Crossref], [Web of Science ®] , [Google Scholar]). 2. The transformation of the original wage equation and its effects on the model properties is presented in Appendix B. 3. The specific sources for the wage, employment, education and technology variables are detailed in Appendix A, as well as scatter plots illustrating the relationships between these variables. 4. This truncation is also carried out in equation (9) of Fingleton (2003 Fingleton, B. 2003. Increasing returns: evidence from local wage rates in Great Britain. Oxford Economic Papers, 55: 716–739. [Crossref], [Web of Science ®] , [Google Scholar]). 5. There are commuting data for 16 bands of 5 km, covering a radius of 0–80 km and one additional band for 'more than 80 km', which we take to be the 80–100 km band, given the truncation of distance in the W matrix. 6. In order to stay consistent with Fingleton (2003 Fingleton, B. 2003. Increasing returns: evidence from local wage rates in Great Britain. Oxford Economic Papers, 55: 716–739. [Crossref], [Web of Science ®] , [Google Scholar]), the replication results in all the tables are presented in a similar fashion. In particular, the parameter estimates are not adjusted with the spatial multipler 1/(1–β). This is because the purpose of the estimation is to retrieve the direct estimate of γ–1, which corresponds to the underlying parameter on returns to density in the structural model presented in Subsection 2.1. 7. Anselin & Kelejian (1997 Anselin, L. and Kelejian, H. H. 1997. Testing for spatial error autocorrelation in the presence of endogenous regressors. International Regional Science Review, 20: 153–182. [Crossref], [Web of Science ®] , [Google Scholar]) show that under these conditions, the standard moments method of Cliff & Ord (1973 Cliff, A. D. and Ord, J. K. 1973. Spatial Autocorrelation, London: Pion. [Google Scholar]) gives inconsistent results. 8. The preferred parameters reported in Fingleton (2003 Fingleton, B. 2003. Increasing returns: evidence from local wage rates in Great Britain. Oxford Economic Papers, 55: 716–739. [Crossref], [Web of Science ®] , [Google Scholar]) are α = 5.5460; β=0.0014; (γ–1) = 0.016; δ 1=0.2929;δ 2=0.0503 and δ 3=0.7762. 9. The specification of these tests is provided in Appendix C. The author is grateful to an anonymous referee for suggesting this extension. 10. Estimates for the other years in the sample are very similar, and are therefore omitted. 11. The 2001 UK census measured the Inner London population at 2.766 million, for a surface area of 319 km2, while central Paris, with a similar population of 2.181 million in the 2006 French census, has a surface area of 105.4 km2.