Empirical essays on development economics : the case of Vietnam

The thesis comprises three empirical essays. Using up-to-date econometric techniques, they exploit data from the nationally representative Vietnam Living Standard Surveys (VLSS), conducted between 1993 and 2008. My goal is to analyze some important development problems in the Vietnamese economy over the transition period, defined as urban-rural inequality, the gender wage gap, child labour and its impact on children's educational success. The chapters also draw out policy implications as to how Vietnam can better protect its vulnerable groups during the economic transition. The first essay, Chapter 2, joint work with Professor Alison Booth, investigates inequality in Vietnamese urban-rural living standards between 1993 and 2006. We use real per capita household expenditure for living standard measurement. We find that urban-rural expenditure inequality is the most important single factor in explaining national inequality in Vietnam. Urban-rural expenditure inequality increased in the initial years of economic reform then decreased after 2002. We use the Oaxaca-Blinder decomposition method, applied to the unconditional quantile regression of (Firpo et al., 2009), to isolate factors contributing to the urban-rural expenditure gap across the whole expenditure distribution. Our results suggest that inter-group differences in education, household demographic structure, industrial structure and related returns are the major contributors to these markedly high urban-rural expenditure gaps. Adjusting the average characteristics of rural households to those of urban households will reduce the overall urban-rural expenditure gap by 40%. While expenditure growth is partly explained by changes in average characteristics, it is mainly due to increasing returns and improvements in unobservable factors. The second essay, Chapter 3, examines wages and the gender wage gap between 1993 and 2008. During this period, real wages of Vietnamese workers doubled and the gender wage gap declined by 58%. The decomposition results using the Oaxaca-Blinder decomposition applied for unconditional quantile regressions (Firpo et aI., 2009) suggest that the major part of the gender wage gap is attributed to gender differences in the return to observed characteristics and to unobserved factors, also known as 'unexplained gap' or 'discrimination'. Wage growth is partly explained by changes in average characteristics but is mainly due to increasing returns. The third essay, Chapter 4, studies the determinants and impact of child labour on children's educational performance in rural Vietnam in 1998. Of all six available waves of the VLSS, only the 1998-VLSS provides suitable information for my study purposes. I model the determinants of child labour and children's educational performance as functions of variables observed at individual, household, school, commune and regional levels. I use provincial temperature variation as an instrument for child labour. The joint equation results suggest that child labour has a significantly negative impact on students' academic performance. Household economic conditions are one of the most important determinants of child labour and children's educational performance. There exists gender discrimination against girls in the burden of child labour. Children from minority ethnic groups have to work more. Among school infrastructures, the presence of electricity, a library and a good blackboard enhance student's study results