Search

This dissertation includes three empirical essays as three chapters selecting different research questions in Land Economics and Financial and Development Economics. I contribute to the literature by extending and implementing methods of dynamic panel regression, nonlinear and nonparametric model estimation, and estimating causal effects to address the issues of heterogeneity, endogeneity, and confounding factors. Chapter 1 determines the association between agricultural land-use change, cropping patterns, and land and irrigation determinants in Alabama, using the United States Department of Agriculture (USDA) National Agricultural Statistics Service (NASS), Forest inventory analysis (FIA), USDA soil survey, and United States Geological Survey (USGS) aggregate panel data for the period from 1992 to 2017 on the county level. This study integrates the dynamic panel regression model using Arellano-Bond (AB) estimator comparing other models such as fixed effects (FE), seemingly unrelated regression (SUR), and ordinary least squares (OLS) and finds that major land attributes such as Agriculture and forest revenues and expenses show negative and positive effects on land use shares and irrigated land shares show positive effects for cropland, whereas more detailed irrigation variables are insignificant with land use shares. Overall, crop shares are less associated with the explanatory variables as the distribution of crop acreage within the cropland in the state with different local climates and soils and crop shares of cotton and corn are more associated with land determinants compared to peanut and soybean shares. Principal component analysis confirms the initial assumption that land-specific economic and social attributes are associated with land-use change and cropping decisions in Alabama. Chapter 2 contains an impact analysis of financial inclusion of unbanked rural households after the Dodd-Frank act of 2010 using the Quantile treatment effect (QTE) with the nonlinear Changes-in-changes (CIC) model utilizing 2009, 2011, and 2019 household data from the National household survey conducted by FDIC. The discrete CIC model is developed using the DID approach and estimated with quantile treatment effects for the unbanked rural and urban households’ variables. Results indicate that the Dodd-Frank act is associated with a higher likelihood of opening a bank account by the unbanked rural households’ in both the short and long term with an expected smaller long-term magnitude. Relative to urban unbanked households, rural unbanked households are more likely to use AFS for transaction purposes in short term, indicating a strong potential effect related to substituting transactions via banks with transactions via AFS. However, relative to urban households, rural households are more likely to have used more AFS for credit purposes, in the long run in 2019, which may be related to the resulting closures of banking infrastructure from 2009 to 2019. Chapter 3 determines the factors associated with prepaid debit card usage of unbanked households’ impact on financial inclusion entering formal banking services and additionally, investigate the use of alternative financial services as a substitute for traditional banking in the US. The results show that unbanked prepaid debit card users are less likely to open a bank account. In further investigation, adding AFS credit and transaction services as outcome variables, results show that unbanked prepaid card users are more likely to use alternative financial services. These findings suggest that unbanked households who use prepaid cards have a tendency to use more alternative financial services compared to the more valuable and typically cheaper banking services offered by traditional banks for their financial needs in the household system in the US. Overall, the study confirms that unbanked households who use prepaid debit cards are open to using financial instruments and services that meet their needs outside of the formal financial system with a propensity to use alternative financial services.

This study examines morphological diversification of the late Oligocene ampullariid species from the Rukwa Rift Basin of Tanzania. Six new species of ampullariids are described, including five species of Lanistes and one species of Carnevalea. The high-spired Lanistes species described record the earliest appearance of this morphotype in the fossil record. Carnevalea santiapillai records the first appearance of this genus outside the Eocene of Oman. Paleoecological interpretations suggest a paludal to lacustrine ecology for C. santiapillai and a lacustrine ecology for L. nsungwensis. Lanistes songwensis and L. songweellipticus were interpreted as fluviate species whereas L. microovum and L. songweeovum were capable of inhabiting both lentic and lotic habitats. Morphological variations among paleoenvironments provide evidence for tectonically induced local radiation of Lanistes. Overall, this study provides valuable insight onto the taxonomy, evolution and the biogeographic affinities of the Paleogene freshwater gastropods of Afro-Arabia, and thus contributes significantly to closing the “African Gap.”