Date of Award


Degree Type


Degree Name

Doctor of Philosophy



Major Professor

Georg Schaur

Committee Members

Robert A. Bohm, Scott M. Gilpatric, Roland K. Roberts


Abstract The enormous costs incurred to government for foreign direct investment (FDI) inflows raised question whether its benefits are worthwhile. In this dissertation, I use productivity estimates as outcomes to explore the direct and indirect impacts of FDI inflows on local firms in manufacturing sector of Thailand during 2001 to 2006.

Chapter 1, I introduce the overview of the entire dissertation.

Chapter 2, I briefly reviewed investment climates and FDI conditions in Thailand. Then I constructed a comprehensive firm-level dataset from several data sources for FDI examination. The main dataset offers quantity and capacity outputs along with revenues at product-level.

Chapter 3, I modify two existing productivity estimation approaches to compute firm productivity in terms of value added, quantity and full capacity outputs. The new productivity estimation approach corrects for endogeneity, multicollinearity and allows for multi product firm assumption. I estimate productivity for three output types: value added, quantity and full capacity. The production function coefficients exhibit unskilled labor intensive technology in all sectors.

Chapter 4, with productivity estimates from chapter 3, I examine the FDI direct impacts on productivity of local affiliates. I adopt two selection bias correction methods: average treatment effects on the treated based on propensity score matching (ATT), and a control function based on second moment conditions proposed by Farre, Klein and Vella (2009). The results from ATT exhibit the existence of FDI direct effects in ten sectors. The results from the second method are significant in eleven sectors. This finding suggests that FDI direct effects depend on specific factors across sectors.

Chapter 5, I investigate whether FDI spillovers exist through horizontal and vertical relationships. The agglomeration-spatial weights are introduced in spillover variables to capture the contributions of clusters and geographical distances. When controlling for agglomeration and geographical distance, the results indicate positive spillover effects through backward linkages, and negative effects through forward linkages and horizontal relationships. This finding reflects the positive benefits from FDI to Thai suppliers, negative impacts on Thai buyers and competitors who located in close to clusters. This means FDI spillover exists only when FDI firms are located close to clusters.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."