Linguistic competition and education spending in Spain 1992–2008

We thank Andy Baker, David Brown, participants at the Pizza and Politics working group at Emory University, Scott Carson, and the two anonymous Social Science Journal reviewers for their comments. We also thank Kerstin Hamann and Carol Mershon for sharing data with us.

Abstract

Linguistic competition occurs when two or more linguistic groups vie against each other for resources from the same state. What are the effects of this competition on education spending? In this paper, we examine two competing explanations. On the one hand, there is the claim that increasing levels of ethno-linguistic diversity can decrease education spending. On the other hand, there is also the argument that education spending is higher when there is electoral competition. Using a newly assembled dataset of education spending at the subnational level for Spain (1992–2008), we test these two arguments. We find (1) while ethno-linguistic diversity matters for spending, the effect is not in the expected direction and (2) electoral competition can affect education spending. We also find that the type of education curriculum (monolingual versus bilingual) can moderate the effects of ethno-linguistic diversity. These results shed light on the commonly held belief that diversity stunts education spending.

Keywords:

• Ethnic politics
• Party politics
• Spain

Notes

We thank Andy Baker, David Brown, participants at the Pizza and Politics working group at Emory University, Scott Carson, and the two anonymous Social Science Journal reviewers for their comments. We also thank Kerstin Hamann and Carol Mershon for sharing data with us.

2 Andalucía, Aragón, Asturias, Balearic Islands, Basque Country, Canary Islands, Cantabria, Castilla-La Mancha, Castilla y León, Catalonia, Extremadura, Galicia, La Rioja, Madrid, Murcia, Navarra, and Valencia.

3 While Spain's federal system is not of the purest constitutional form, an argument can be made for the country as being pseudo-federal at a minimum. As Brzinski, Lancaster, and Tuschhoff (1999) acknowledge, citing Hamann, Spain's system can be considered federal based on Linz's institutional argument. Moreover, referring to Keating, they remark Spain's federal structure, while asymmetrical, is still fundamentally compounded representation. Other works have treated Spain as federal in large N analyses of federal systems (Rodden & Wibbels, 2002).

4 Given the uniqueness of Convergence and Union (CiU) in Catalonia—a nationalist electoral alliance composed of a larger liberal Democratic Convergence of Catalan (CDC) and a moderate Democratic Union of Catalonia (UDC)—we also re-run our models with Catalonia removed. The substantive results do not change.