FDI and the spillover effect analysis: the case of Ethiopia

Abstract

This thesis presents the spilling over effect resulting from the foreign direct investment with a focus on the manufacturing firms. It covers extensive econometric analysis based the Central Statistics Agency's (CSA) survey on the manufacturing firms and an Input-Output matrix done by the Ethiopian Development Research Institute (EDRI). A pooled, Fixed and Random Effect estimation techniques are employed for estimating the log transferred production function augmented for the spillover proxies: Backward, Forward and Horizontal. Yet, as is stated in a lot of literatures like that of Javorcik (2004), the Cobb-Douglas production function suffers from the endogeneity problem and there is a need for a better estimation technique that can capture and solve this problem. As a result, I also used the Levinsohn-Petrin estimation technique, which used intermediate inputs as a proxy for unobservable shocks and the residuals from this estimate used as a measure of total factor productivity (TFP) of the firm. The TFP analysis from the LP estimation suggests that a one percentage point increase in the foreign presence in the downstream sectors is associated with the 1.1 percent rise in the total output of each supplying industries. Likewise, a one percentage point increase in the weighted share of output in the...