A test of the market potential equation in Spain

Abstract

In this article, we examine the relationship between regional wages in Spain and the market potential of these regions in the period 1955 to 1995. We demonstrate the existence of a spatial wage structure, in which wages fall with increasing distance from the highest income regions. This result strengthens the hypothesis that agglomerative forces were operating in Spain during the second half of the twentieth century.

Acknowledgements

The authors would like to thank the participants at the VIII Encuentro de Economía Aplicada, Murcia, June 2005, for their comments, the Spanish Ministry of Science and Technology for the financial support granted through research projects BEC2002-00423, SEC2002-03212, SEJ2005-02498/ECON and SEJ2005-03196/ECON and the Catalan government (Generalitat de Catalunya) for the grants 2005SGR-00460 and 2005 SGR 00854.

Notes

¹ Pons et al. (2002) present evidence about the origins of this long-term process during the second half of the nineteenth century.

² In fact, this relationship is amplified to real salaries through the forward linkage or price index effect. An increase in the number of consumers is associated with a decrease in the price index of the differentiated good. So if income is high in some region, we expect the real wage to be high both because the nominal wage is high and because the price index is low. (Fujita et al., 1999). However, in this article we focus only on the demand linkage and thus, exclusively in the relationship between nominal wages and market potential.

³ Fingleton (2005) or López-Rodriguez and Andrés-Faíña (2006) have analysed other related predictions of this kind of models such as the existence of spatial structures for regional incomes or prices.

⁴ In Paluzie et al.(2005), this relationship is estimated as a test of one of the predictions of the Krugman's (1991) model.

⁵ Other indicators such as the Hirschmann–Herfindahl index show the same tendency (Paluzie et al. 2004).

⁶ The estimation of the structural wage equation is not possible as we do not have all the necessary data. However, Brakman et al. (2002) and Hanson (2005) show that the results of the reduced estimation do not differ greatly from those obtained using the structural estimation.