|dc.description.abstract||This dissertation is composed of three chapters examining the wage inequality in developing countries, and the effect of insurance subsidy on land use allocation.
Chapter 1 is about the distributional decomposition of gender wage gap in developing Countries. In this paper I investigate the gender wage gap across 12 developing countries. Based on the traditional Blinder-Oaxaca decomposition, I also consider the distributional effects on the wage gap decomposition. I apply the approaches which are similar to Machado and Mata (2005) and Firpo et al. (2007, 2009) to obtain both aggregate and detailed decomposition. The results show that gender wage gap exists at all wage levels and decreases as wage increase. The number of children contributes to the gender wage gap especially at low wage levels. Opening more mother friendly positions and making education and training more accessible to the low-income groups will be helpful to raise their earnings and reduce the wage inequality.
Chapter 2 investigates the effects of parental education on children’s income. In wage equation, education is an important factor that affects personal income. However, education of parents also has influence on children’s income, because of the intergenerational effects. In this paper, I investigate the effect of parents’ educational level on children’s income in 12 developing countries. I use the maximum educational level of parents as the independent variable and estimate the equation of children’s hourly income. As there is omitted ability in the wage equation, I use interruption of schooling as the instrumental variables to identify the educational years. In addition, I applied Heckman Selection Model to fix sample selection bias. The results show that high parents’ educational level has positive effect on children’s hourly earnings. Policy makers should consider the intergenerational effect to reduce social inequality.
Chapter 3 investigates the effect of crop insurance subsidies on agricultural land use allocation. Since the objective of crop insurance is to help farmers with risk management, the expected profit of crop production under crop insurance might be improved, leading farmers to allocate more land into crop production. In this paper, agricultural land use type is classified by irrigated/unirrigated farmland and cropland/woodland/pasture land. The data contain counties from all continental states. Considering the fractional outcome of land use share, I apply a Multinomial Fractional Logit Model to estimate the effects of insurance subsidies on land use. The results show that insurance subsidies have a significant effect on land use allocation. An increase in insurance subsidies increases farmland share, indicating insurance subsidies could be an efficient tool to adjust agricultural land use allocation.||en_US