Please use this identifier to cite or link to this item:
Hartgen, Kenneth
Klasen, Stephan
Misselhorn, Mark
Year of Publication: 
Series/Report no.: 
Courant Research Centre: Poverty, Equity and Growth - Discussion Papers 8
Spurred by international commitments and expanded funding at the national and international level, attendance in education and associated years of schooling have expanded substantially in developing countries in recent years. But has this expansion in enrolments reduced existing inequalities in educational access and achievements? This paper analyzes differences in improvements in the access to the education system and in educational outcomes across the welfare distribution between and within countries, and also by gender and regions for a sample of 37 developing countries using Demographic and Health Surveys (DHS). For the analysis, the toolbox of pro-poor growth analysis is applied to several educational indicators. We find drastic inequalities in educational attendance across the income distribution. Interestingly, inequalities in attendance declines with rising average attendance, while inequality in completion rates or schooling years increases with rising completion rates or schooling years. We find great heterogeneity in the distribution of progress of education, with very little pro-poor progress in educational achievement indicators. Also, progress appears to be less pro-poor in countries with low initial educational achievement and high overall educational progress. We find no correlation between pro-poor progress and free education policies or initial inequality in education. At the regional level, educational progress was generally more pro-poor in Asia and Latin America, while in Africa the experience is very heterogeneous. While gender inequality has decreased slightly, large differences by region tend to persist over time.
human capital
pro-poor growth
Document Type: 
Working Paper

Files in This Item:

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.