Party Competition and Preferences for Inter-regional Redistribution in Spain

Abstract

Abstract The main contribution of the article is to highlight the fact that preferences for inter-regional redistribution in Spain are crucially shaped by the dynamics of party competition. First, I show that it is not regional income but the saliency of the second dimension of party politics that drives preferences for regional redistribution. Second, I analyse whether the relationship between partisan attitudes and redistributive preferences is conditional on the saliency of the second dimension across various regions. Third, I also investigate whether the relationship between partisan attitudes and redistributive preferences changes between cohorts within a given party system. Keywords: Party CompetitionHerestheticsInter-regional RedistributionParty Preferences Acknowledgements I would like to thank Elias Dinas, Ignacio Sánchez-Cuenca, Tim Müller, Elna Roig, Michael Donnelly, Erik Wibbels, David Rueda, Pablo Beramendi and two anonymous reviewers for their very helpful comments and suggestions on my work. Notes [1] The Spanish government published the regional fiscal balances in 2008, although they refer to the regional fiscal (im)balances in the year 2005. There are various methods to calculate the fiscal (im)balance for each region, but the basic idea is always to approximate the difference between the revenues raised by central government in a region, and the expenditure in the same region. Here I use the percentage of the fiscal (im)balance calculated using the 'monetary flow' methodology. [2] The variable Saliency of the second dimension of electoral competition in region (i) is calculated by taking the mean position of political parties in the territorial/decentralisation dimension and averaging over the sum of quadratic distances between the mean position for each party and the centre of the scale. Both the mean positions of parties and the scale are centred around 0. The mean position of political parties at the regional level are obtained from the individual-level data. The formula is as follows: