Essays of empirical research on international outsourcing of United States manufacturing industries, 1972--2001

The existing literatures have proposed some measures to quantify the increase in international outsourcing in recent years, however, there are limitations in the current measures and several empirical issues have not been adequately addressed. This paper proposes a new approach to compile measures of U.S. manufacturing outsourcing at the disaggregated levels by host country and by industry from 1972 to 2001. The measures have confirmed the rapid increase in international outsourcing and the geographic orientation of outsourcing to the developing countries. The paper then empirically tests Grossman & Helpman's model of the determinants of the location choice of outsourcing and confirms the theoretical predictions. It also extends to incorporate the factors of the Gravity model and international openness into the panel regression for robustness test. The paper also takes the controversial question whether outsourcing has caused domestic job losses to the data and examine the effects of outsourcing on employment at the 2-digit SIC level using the framework of Hamermesh's labor demand model. Contrary to the popular belief, the paper finds that outsourcing does not have a significant effect on employment. The alternative specifications in terms of change and with the dynamics of the lagged values of employment also fail to detect any significant effects. Therefore outsourcing is not to blame for domestic job losses.