The Long Term Effects of Caciquismo: A Regression Discontinuity Approach

Autor principal:
Jaime Bordel Gil (Universitat Autònoma de Barcelona)
Programa:
Sesión 4, Sesión 4
Día: jueves, 10 de septiembre de 2026
Hora: 09:00 a 10:45
Lugar: 21

This article examines the long-term effects of caciquismo on political participation and electoral coordination during the Spanish Second Republic (1931–1936). We argue that the cacique system weakened local associative networks and political organization, generating persistent patterns of demobilization and coordination failures among political forces. To test this argument, we propose a Regression Discontinuity Design (RDD) based on two institutional features that make such a design possible. First, the 1907 electoral law provides a reliable way to operationalize caciquismo in the period before 1931. The article operationalizes caciquismo as the proportion of elections resolved through Article 29 of the 1907 Electoral Law, which allowed a single candidate to be proclaimed elected without holding the vote, thereby offering a clear measure of the power and control exercised by local caciques. Second, whereas during the Restoration electoral districts were subprovincial, the Second Republic established the province as the electoral district. This institutional shift, combined with the availability of municipal-level electoral data, allows us to exploit a large number of discontinuities between municipalities and identify the causal effect of historical exposure to caciquismo on electoral behavior in 1931. Preliminary results show that exposure to caciquismo during the Restoration period reduced electoral coordination in 1931. 

Palabras clave: Regression Discontinuity Design, Caciquismo, Participation, Electoral Coordination.