The Political Origins of Education Decentralization in Latin America

Abstract

For much of the 1990s Latin American countries experimented with transferring the responsibility of managing public schools from the central government to subnational units such as states, municipalities, and schools. Why? Conventional wisdom stresses that these countries decentralized education because central governments were too inefficient to resolve longstanding problems related to access and quality (for example, see Grindle 2004; Kaufman and Nelson 2004). But a careful review of the evidence reveals that the education systems of Latin America generally were not in crisis. The data show that these countries achieved notable gains in enrollment and completion rates between 1950 and 1990—gains that were not lost during the severe debt crisis that spanned the 1980s (World Bank 2018). Rather than a technical explanation, then, we need a political explanation to account for the reasons why Latin American countries decentralized such an important public service. My central claim is that countries decentralize education governance for two political, rather than technical, reasons. First, international pressures shape both the demand and supply for education reform ideas. The 1980s debt crisis made budget cuts necessary, but the importance of education for development meant finding better ways of improving its provision. In this context, decentralization emerged as one, but not the only, choice of reform. Second, the partisan affiliation of teachers determines whether or not incumbents push for decentralization. The reason is that decentralization strengthens the electoral position of incumbent political parties whenever their opponents are affiliated with teachers’ unions because decentralization tends to weaken teachers’ organizational advantages, which are often used to help their partisan allies drum up votes. In a decentralized system, teachers are less able to act as political brokers for the opposition. These political factors, not the underlying strengths and weaknesses of the education systems, drove education decentralization. To support this argument, this dissertation provides case study evidence from four countries (Colombia, Costa Rica, El Salvador, and Paraguay). It also deploys statistical tools on an original dataset on education decentralization and teachers' union partisan affiliations to show the argument's generalizability.