Evaluating short- and long -term impact of education policy: Evidence from Indonesia
Endogenous growth theories link a country's stock of human capital to growth and development. Thus, raising the stock of human capital is a major policy priority of developing countries. Yet, the existing literature remains mostly inconclusive on what the governments can do to raise the educational attainment of its people. This dissertation makes use of education policies in Indonesia in the 1970s to understand the effectiveness of those policies in raising the educational attainment. As it adds to the existing literature on the immediate and long-term impact of educational policy by looking at evidence from Indonesia, it is intended that this knowledge will help developing country governments make better educated policy decisions. This dissertation consists of three closely related yet stand-alone substantive essays. The first essay evaluates the impact of the school construction program on children's educational attainment. This paper re-estimates the same effect as in Dulfo (2000) utilizing additional information about the program and a better program measure. The study suggests a program measure that takes into account the number of schools completed each year and estimate the average number of new program schools one is exposed to in that persons primary school life. Further, this study is the first to estimate the program impact on women. The study finds the program had a significant impact on the education of both men and women exposed the program and the estimate was slightly higher than what was estimated by Duflo (2000). The second essay examines how the same educational policy has altered the observed strong educational association between parents and children. The study finds that the program was successful in weakening the observed strong relationship as the greatest impact of the program was on the children of poorly educated parents. The purpose of the third essay is shed light on the intergenerational educational transfers within families. Using an instrumental variables approach the causal link of the educational attainment of one generation on the next is examined. Findings indicate that even though both mothers' and fathers' education matter positively, it is the paternal education that matters more.