Dissertations

Dissertation Committee:
Patrick Gaulé (chair)
Randall K. Filer
Daniel Münich
 

Abstract:

In the first chapter of the dissertation, two administrative datasets from the Targeted Social Assistance Program (unconditional cash transfer) and National Assessment and Examination Center in Georgia are merged in order to investigate the impact of an unconditional cash transfer on the university enrollment rate in Georgia. Given that the program recipients were selected by virtue of being below a certain quantitative poverty threshold, this feature of the program is exploited to implement a global regression discontinuity. The study finds a positive impact of cash transfers on enrollment in tertiary education. Specifically, being a recipient of the social assistance program significantly increases a student’s likelihood of enrollment, by 6.3%. More importantly, the findings suggest that the observed effect is gender specific: the impact is stronger for males. Male children of a beneficiary family have a 13.4% greater chance of being admitted to university. This marks the first attempt to study such a program in the context of education. The paper contributes to the growing literature on the long-run effects of cash transfers.

In the second chapter, the impact on a broad range of outcomes of the same social assistance program in Georgia is examined. An original household survey was developed and conducted in 2014, and a total of 340 households living in Tbilisi participated. A local regression discontinuity approach was employed to evaluate the unconditional cash transfer program in Georgia. In this study we found that receiving the transfer leads to a worsening in (self-reported) basic economic conditions, such as the ability to afford food. The recipients' worsening of economic conditions relative to the control is genuine, which begs the natural question: why? One possibility is that the program crowds out other sources of income or, alternatively, receiving the transfer could reduce incentives to work. Another possibility is that the recipients invest both the transfers and additional resources in investments in durable goods or human capital. This could then lead to a temporary lower ability to afford food (and similar patterns) in the time window of the survey.

The third chapter investigates how a unique education policy positively affected university enrollment rates of public school students in Georgia. In 2007, the Georgian government enacted legislation mandating the replacement of all public school principals under the assumption that the replacement of the principals with random assignation of qualified candidates to public schools would decentralize and improve school governance across Georgia in a fair manner. About half the public school principals were actually replaced with new candidates and a majority of them were assigned through a random allocation mechanism. Therefore, the standard difference-in-differences methodology is used to compare treated public schools with private schools that are not affected by the policy, in order to identify how this reform impacted education outcomes. Using the National Assessment and Examination Center university admissions data, the public schools with replaced principals increased university enrollment more than the control schools, by an average of 4%. The largest part of this increase comes from schools with randomly assigned principals. The positive findings herein could impact education policy in developing (and perhaps developed) countries. The statistically significant and strong effects of this type of reform could cause a positive domino effect in the developing world, especially in countries with similar characteristics and predicaments in their education system.

Full Text: “Essays on Economics of Education and Social Policy” by Zurab Abramishvili