Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.773007
Title: Human capital outcomes for children : the impact of school subsidies and natural disasters
Author: Sulistyaningrum, Eny
Awarding Body: Lancaster University
Current Institution: Lancaster University
Date of Award: 2013
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
This thesis contains a series of related studies about the human capital outcomes of children as consequences of school subsidy reform and natural disasters. All the studies in this thesis rely on the Indonesia Family Life Survey data 2000 and 2007. The first study is an evaluation of the impact of the school operational assistance program (BOS) on child test scores. This study uses three different methods: Ordinary Least Square (OLS), Instrumental Variable (IV) and Propensity Score Matching (PSM) estimation. OLS is used as conventional method and estimates the effect of school subsidy on average by assuming that BOS is exogenous, while IV estimation is used to deal with endogeneity of BOS and also correct for selection bias based on unobservable characteristics. PSM is used to estimate the average treatment effect in the absence of selection on unobserved characteristics. The results confirm that the BOS can increase student performance. Paradoxically it does not help very poor students, yet it helps less poor students. The second study is an examination of the impact of natural disasters on child test scores and child health. Arguably natural disasters are exogenous events, so we use the exogenous variation from natural disasters as a natural experiment design to estimate the effect of disasters on child test scores and child health. When a certain group is exposed to the causal variable of interest, such as a disaster, and other groups are not, Difference in Difference model (DID) can be used in estimation. For child health, we use two types of data: child height under 5 and self-reported general health condition, while test scores are obtained from the national child test scores at age 11. In conjunction with the DID model, for analysis of the impact of disasters on child health, we also used zero inflated negative binomial and an ordered logit model for analysing self-reported data on child health. The results confirm that child test scores are significantly affected by disasters, but there are no serious impacts of disasters on child health, conditioned on survived. The third contribution is a study on the impact of disasters on household expenditures and food demand. We employ a DID model to estimate the total impact of disasters on various types of household expenditures. For comparison, we use a Linear Approximate Almost Ideal Demand System (LA-AIDS) model to estimate the net impact of disasters on expenditures. In the LA-AIDS model we control for the price of foods, while in the DID model we do not. Furthermore, the LA-AIDS model also estimates the price elasticities of demand and the expenditure elasticity of demand. The findings show that there is a negative impact of disasters on educational expenditures, and no impact of disasters on total household expenditures. Yet, we found a net negative impact of disasters on expenditures when controlling for prices.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.773007  DOI: Not available
Share: