Abstract
This dissertation investigates the effect of direct and indirect taxes on poverty in developing countries, which are characterized by higher level of poverty and low level of total tax revenue as share of GDP. We use an annualised panel data of 37 developing countries for the period 1995-2016. Panel cointegration, Dynamic Ordinary Least Squares (DOLS) and Fully Modified Ordinary Least Squares (FMOLS), the Dumitrescu-Hurlin causality test and the Pooled Mean Group (PMG) were employed to determine the short- and long-run impact of direct and indirect taxes on poverty, and to assess the direction of the causal effects among the variables. The results from the FMOLS and DOLS show that only tax on goods and services and corporate taxes are negative and significant in explaining poverty in the long run in developing countries. From the Dumitrescu-Hurlin causality test, the findings indicate that there is a causality running from corporate taxes to poverty, while tax on goods and services cause poverty and vice versa. Finally, the PMG demonstrates that while the long-run estimates show a negative and significant relation among our variables in developing economies, the short-run relationship indicates that the link is statistically insignificant, with an error correction term of 0.059. Therefore, the short-run deviations from the long-run equilibrium are corrected at the speed of 6% each year. The overall findings support that argument that taxes on goods and services combined with corporate income taxes play a key role in reducing poverty in a long-run in developing economies. Therefore, the policy recommendation of this is that transfer and tax system should be designed in the way that income received from transfer should be more than taxes paid by the poor. And the revenue mobilized from taxes on good and services and corporate income taxes should be allocated to education at the early stage.
M.Com. (Development Economics)