Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Jorge Martinez-Vazquez

Second Advisor

Sally Wallace

Third Advisor

Andrew Feltenstein

Fourth Advisor

Charles Hankla

Fifth Advisor

Yongzheng Liu


Chapter 1 revisits the relationship between fiscal decentralization and economic growth by addressing the endogeneity. We use an instrumental variable approach based on two geography variables, namely a Geographic Fragmentation Index (GFI) and country size. We find that both instruments are strong and valid in the first stage regression and that a ten percent increase in decentralization measured by the expenditure or revenue share of subnational government in total government expenditures or revenues increases GDP per capita growth by approximately 0.4 percentage points. Moreover, we find that the results are more pronounced in the case of developed countries with a higher magnitude of the impact of revenue decentralization and lower impact of expenditure decentralization, while for a sample of developing countries both decentralization measures are insignificant.

Chapter 2 explores the role of the tax structure and its key elements on the volatility of output growth. We account for both embedded automatic stabilizers measured by progressivity of the tax system and discretionary policy by accounting for the actual levels of revenue and its composition measured by tax mix ratio or the ratio of direct taxes to indirect taxes. We find that higher reliance on direct taxes versus indirect taxes is a significant stabilizing factor for output volatility for the whole sample of all countries and the subsample of lower income countries. For the subsample of high-income countries, we find a significant stabilizing impact of progressivity in the income tax structure, especially when there is higher reliance on personal income tax revenue.

Chapter 3 reexamines the causal link between institutional quality and economic development using "Malaria Endemicity" as an instrument for institutions. This instrument is superior to the previously used instruments in the literature which suffered from measurement error. Because the Malaria Endemicity measure captures the malaria environment before the discovery that mosquitoes transmit the disease and before the successful eradication efforts that followed, it is exogenous to both institutional quality and economic development. We find Malaria Endemicity a valid strong instrument which yields larger significant effects of institutions on economic development than those obtained in the previous literature.