Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Dr. Michael Ryan

Second Advisor

Dr. Susan Pozo

Third Advisor

Dr. Christine Moser

Fourth Advisor

Dr. Rajib Paul


This dissertation uses spatial econometric technique to investigate the role of spatial interdependence/third-country effects on Multinational Enterprises (MNE’s) for affiliate location choice, spillover effects of foreign direct investment (FDI) determinants, and the economic growth of developing countries. Unlike non-spatial models, the advantage of spatial econometric models lies in their ability to capture cross-country interdependence and spillover effects of explanatory variables.

The first essay employs a spatial autoregressive model and data on U.S. outbound FDI to investigate whether U.S. FDI into Africa, Latin America and the Caribbean (LAC) depends on previous U.S. FDI in these regions. Additionally, the study investigates the prominent motivation for U.S. firms in these regions. The empirical results confirm third-country effects do matter and the coefficient estimates on the market potential and spatial lag indicate U.S. FDI into Africa and LAC is consistent with complex vertical FDI.

The second essay uses a spatial marginal effect technique to examine the direct and spillover effects of the FDI determinants from the first essay. In addition to finding spatial interdependence among countries in Africa and LAC, the empirical results provide evidence of both direct and spillover effects of FDI determinants in both regions. The empirical evidence confirms changes in policies that affect inward FDI in a particular country affect inward FDI in surrounding countries.

Finally, the third essay examines the effect of FDI, foreign aid, and remittances on the regional economic growth of Africa and LAC using a spatial dynamic panel data model. In addition to finding growth interdependence across the two regions, the result for Africa shows foreign aid and FDI affect economic growth positively while remittances do not. This result holds both when the growth impacts of these factors are estimated separately as well as together. However, for LAC, the results show only foreign aid affects growth when estimated separately, while foreign aid and remittances impact growth when all three factors are included in the same regression. Hence, controlling for the growth impact of the three factors produces reliable and unbiased estimates.

Access Setting

Dissertation-Open Access