Abstract
In the first chapter, I examine both the often-claimed foreign ownership advantages and the technological spillover effects of FDI on domestic firms in the host country. Specifically, using a matched dataset of the Annual Survey of Industrial Enterprises and official firm patent data from 1999 to 2007, I provide a general overview of how FDI affects three types of firms in the Chinese manufacturing sector: those that are directly owned by foreigners, those that are acquired by foreigners, and those that are owned by domestic investors. I document three facts: First, compared to wholly-privately-owned firms, FDI firms via greenfield investment, be they wholly-foreign-owned enterprises (WFOEs) or international joint ventures (IJVs), are larger, more productive, have better financial conditions, and pay higher wages. They also have a significantly higher skilled labor share. Second, foreign acquisitions lead to performance improvements. Foreign-acquired firms significantly outperform their domestic-acquired counterparts after acquisition along the following dimensions: firm size growth, financial condition growth, and workforce growth. Third, there is evidence hinting at positive spillovers from FDI to domestically owned firms in one particular measure, patent applications. These facts also motivate the skill premium analysis in Chapter 2.
In the second chapter, I examine how inward FDI affects the relative demand for skilled labor and, thus, the skill premium in Chinese local labor markets. First, I develop a local labor market model that accounts for the empirical findings from Chapter 1. I introduce technology choices in a Melitz-type model with heterogeneous domestic firms in a small local economy, which is endowed with a fixed supply of skilled labor and unskilled labor. In the model setting, only the most productive local firms adopt the skilled-labor intensive technology, while the less productive local firms use the unskilled-labor intensive technology. FDI from developed countries goes into the creation of foreign-invested firms, and they always use the skilled-labor intensive technology and are more productive than domestic firms. The presence of foreign-invested firms has positive technological spillovers to local firms in the sense that the more foreign-invested firms there are, the lower are fixed costs for local firms to adopt the skilled-labor intensive technology. With a lower fixed cost for FDI to set up plants abroad, the direct channel through foreign-invested firms and the indirect channel through domestic firms upgrading their technology work together to deliver a positive relationship between FDI shocks and skill premium changes in the local labor market. Then I test the model prediction using individual-level survey data, prefecture-level statistical yearbooks, and prefecture-level innovation index constructed by scholars. Using a sample of 37 prefectures with available skill premium changes, I find a significantly positive effect of FDI shock on prefectural skill premium. I further supplement my skill premium analysis using a larger sample (225 vs. 37) that features in prefectures' innovation and average wages. My results show that a 3 percentage point increase in FDI share (the average increase rate from 2001 to 2007) leads to a 10\% increase in innovation index and 1.6\% increase in the average local wage, contributing to 4\% of the innovation index increase and 1.6\% of the wage increase during the period. As a further supplement, I perform a cross-section analysis using Census2005 only and find that a one-standard-deviation increase in FDI share results in a 15\% increase in the skill premium. All these findings depict a coherent picture that FDI exposure shock drives up the skill premium in Chinese local labor markets.
In the third chapter, I look at the other driving force of globalization --- trade liberalization. I explore the effects of input trade liberalization on imported input and exported output prices. Using the highly disaggregated transaction-level customs data and firm-level industrial survey data from 2001 to 2006, I first document three facts about the evolution of tariffs, imported input prices, and exported output prices over the trade liberalization period. Next, I estimate the causal relationships between exogenous input tariff reductions and within-firm changes in prices of imported input bundles as well as prices of exported final outputs. My results show that firms import inputs at higher prices and quality when product-specific tariffs fall during the period of input tariff reductions, and these effects are more prominent for products imported from developed countries. Firms increase import prices of their input bundles and their import share from developed countries when firm-specific input tariffs fall during the period of 2001 to 2006. Such effects are more prominent for exporting firms than non-exporting firms during the period of 2001 to 2004. On the export side, firms export products at higher prices with higher quality when firm-specific tariffs fall during the period of 2001 to 2006. The impacts of the input tariff reductions on export prices and quality are specific to differentiated products. Overall, the empirical results are consistent with a scenario in which firms exploit the input tariff cuts induced by China's accession to the WTO in 2001 to access high-quality inputs in order to upgrade the quality of their exports.
