Identifying national level education reforms in developing settings : an application to Ethiopia

Increasing enrollment in primary education has been at the center of international education policy for well over a decade. In developing parts of the world, significant increases in primary enrollment are often generated by large national level programs, which can simultaneously promote overcrowding and reductions in education quality. However, to analyze the trade-off between increased enrollment and potential reductions in quality one must first identify and evaluate the impact of the national reform on schooling. This paper provides a method with which these types of reforms can be identified in developing settings using both temporal and geographic variation, and readily available data. The method is applied to an early 1990s reform in Ethiopia based around the release of the Education and Training Policy, which removed schooling fees from grades one to ten. The model estimates that the reform led to an increase in schooling of at least 1.2 years, and provides initial evidence that the increased enrollment in Ethiopia outweighed any cost due to reductions in quality.