Essays on International, Development and Environmental Economics

Abstract

This dissertation consists of three empirical essays in three chapters, which sheds light on select research questions on foreign inflows on the income inequality in low-income countries, the agricultural productivity of donor countries induced by the growth in recipient countries in biodiversity, and non-market analysis of the environmental degradation and externalities. Chapter 1 is to extend de Janvry and Adult’s (1988) analysis of the relationship between foreign aid and agricultural imports by including Foreign Direct Investment (FDI) and Official Development Assistance (ODA) as explicit policy instruments affecting growth rates in the industrial and agricultural sectors of less developed countries (LDCs). An equilibrium displacement model is specified to deduce hypotheses about increased FDI and ODA’s effects on agricultural imports. The hypotheses are tested using unbalanced panel data for 28 least developed countries from 1994 to 2018. The elasticities are obtained by estimating the structural model using three-stage least squares (3SLS). For robustness checks, we use different subsamples, African and Asian LDCs, and the three- and five-year averages. Results of the structural model suggest FDI to LDCs deters the economic growth of recipient countries, while the opposite is true for increased ODA. Our findings match de Janvry and Sadoulet, although the magnitude of effect is smaller than they found. The frontier of harmony between LDCs’ agricultural development and U.S. agricultural export interests is reached when the change in agricultural imports with respect to the change in the agricultural sector is zero. Aid and trade are in harmony when reduced-form elasticity of agricultural imports with respect to foreign inflows has a positive sign and negative otherwise. Therefore, there will be harmony on the side of the frontier for FDI and ODA. Chapter 2 investigates the impact of Chinese Outward Directed Investment (CODI) into Africa on income inequality, measured by the Gini coefficient. We use data for 21 low-income African countries over the period 2003 – 2014. The dynamic GMM/FE and 2SLS/FE are implemented by incorporating several lags as instrumental variables to correct endogeneity. Results suggest that CODI exacerbates income inequality. In line with the dependency hypothesis, an isolated 100% increase in CODI is significantly associated, on average, with a 5 % increase in income inequality in low-income African countries. Consistent with the modernization hypothesis, a 100% increase in non-CODI sources of FDI is estimated, on average, to decrease income inequality by 2%, reflecting the historical trade partnership between Africa and countries, such as EU and USA. The increase in trade openness and employment rate leads to better income distribution. In addition, an isolated 10% increase in secondary school enrollment rate relative to total population is significant and increases income inequality by 2.3%, widening the gap between population groups who are eligible to work. Therefore, income inequality is more sensitive to secondary school enrollment than to CODI. Increased natural resources rent results in higher income inequality in low-income African countries. We controlled for GDP and the inflation rate, which were found not to affect income inequality. Lastly, the relationship between the Gini coefficient and CODI is in line with the inverted U-curve - the Kuznets curve, however, the magnitude is small. Results are consistent with findings for other countries and are robust to model specification. Chapter 3 estimates the effects of environmental degradation on residential prices and locations in Rwanda (Eastern Africa). The main hypothesis is that average residential prices are negatively associated with extensive biomass use, and individuals are less likely to pay for degraded environmental conditions. The cross-sectional dataset includes five provinces, subdivided into 30 districts of Rwanda, and retrieved from the Statistics Integrated Household Living Conditions Survey, 2013-2014. Two stages of spatial quantile regression are employed to capture spatial spillovers and deal with spatial heterogeneity. The presence of negative externalities from the extensive use of biomass will cause housing prices to decrease and the willingness to pay for such houses to decline. Findings suggest that the use of inferior cooking and lighting fuel sources in residential dwellings, such as firewood and crop wastes, has a negative and significant impact on housing values, across quantiles, though different magnitudes. On the other hand, there are indications of willingness to pay for quality-of-life improvements. For example, living in a modern planned area and houses made from cement and bricks positively affect the dwelling values compared to houses made from different low-quality materials. As expected, housing values are also positively affected by floor area.