Ph.d.-forsvar: Fiseha Haile Gebregziabher: Essays in Empirical Development Economics: Insights from Time-Series and Panel Data Analysis
This dissertation comprises three self-contained chapters focusing on empirical questions in development economics. The first chapter investigates the long-run effects of foreign aid in Ethiopia for the period 1960-2009. More specifically, we analyze whether aid from different sources (multilateral and bilateral aid) and different aid modalities (grants and loans) exert different long-run impacts on several macroeconomic variables. Using a well-specified cointegrated VAR model as a statistical benchmark, we show that aid affects income, investment, and imports positively, whereas it is negatively linked with government consumption. Disaggregation of aid flows unveils that this effect is due to aid from bilateral donors and aid in the form of grants.
The second chapter examines the link between adjustment programs and long-run economic performance in 18 Sub-Saharan African (SSA) countries for the period 1960-2009. During the 1980s and 1990s, IMF and World Bank-sponsored structural adjustment programs have dominated policy-making in SSA. Most SSA countries adopted a wide array of policy reforms to arrest macroeconomic imbalances and foster long-term growth. Applying our multivariate cointegration model to each of these countries, we find that only a handful of countries have shown positive and sustained results. The first-generation adjustment package was associated with resurgence of growth in GDP, export, and investment only in two countries. Taken as a whole, countries in the CFA franc zone fared much worse than their non-CFA counterparts.
The third and final chapter (joint with Miguel Niño-Zarazúa) assesses the impact of government spending on social sectors (health, education, and social protection) on two major indicators of aggregate welfare (the Inequality-adjusted Human Development Index (IHDI) and child mortality), using a panel dataset comprising 55 countries from 1990 to 2009. Notwithstanding the widespread consensus in policy circles, there is surprisingly little rigorous empirical work on the effectiveness of social sector spending and the existing evidence remains inconclusive. Applying a broad range of econometric methods, we find fairly strong and robust evidence that social spending has a significantly positive causal effect on the IHDI, while health spending has a significant negative impact on child mortality rate.