This dissertation consists of three chapters. The first chapter studies how the rise of China affects innovation in other countries. I emphasize the importance of export competition, which means the competition in third countries, in answering this question for three reasons. First, export competition with Chinese firms should be prevalent since Chinese exports have grown worldwide. Second, competition could be more intense in export markets due to the lack of home bias. Third, innovation is skewed toward high-productivity firms, many of which are exporters. To explore the export competition channel, I develop a multi-country model of innovation incorporating the quality preferences of consumers and the heterogeneous productivity of firms. The model predicts that more intense competition increases innovation of high-productivity firms, whereas it decreases innovation of low-productivity firms. The model also suggests that export competition could be more important than import competition in explaining innovation since high-productivity firms are exposed to export competition in more markets. These predictions are confirmed by the evidence from South Korean patent data using a novel firm-level measure of export competition developed in this chapter.
The second chapter studies whether regional development policy generates knowledge spillover effects. Causal inference on this question is not straightforward due to the endogenous location choice and the difficulty in measuring knowledge spillover. This chapter overcomes these challenges in four directions. First, a winner-loser comparison is conducted using a quasi-experimental South Korean Innovation City project, which relocates 112 public agencies with 41,364 employees from Seoul metropolitan area to provincial regions. Second, South Korean patent data are classified by the relevance with relocated public agencies to distinguish the direct effect of relocation and its spillover effect. Third, the patenting history of relocated agencies is used to measure the magnitude of shock precisely. Fourth, the physical distance and the traffic volume between municipalities are used to examine whether spillover effects are local. The empirical evidence shows that innovation increases in Innovation Cities both directly by the relocated agencies and by their co-work with local agencies. However, knowledge spillovers beyond Innovation Cities are limited to very close regions.
The third chapter, joint work with Robert Feenstra, explores a new mechanism through which the labor share in GDP falls in general equilibrium. We develop a general equilibrium model with non-CES preferences, occupational choice of ex-ante identical individuals, and the heterogeneous productivity of firms that explores the fiscal origin of the decline. The model suggests that (i) corporate-friendly fiscal policy decreases the labor share; and (ii) the labor share declines more when the entry of firms is restricted. This is because rigid entry adjustment prevents new entrants from entering the market, resulting in weaker competition between firms.
