Two essays in international trade

Abstract

The first chapter of the thesis studies the impacts of capital goods imports on firm productivity. International sourcing has been a strong driving force for firm productivity growth, as documented in the literature. However, the distinct roles of capital and intermediates importing on firm productivity are much less explored. This paper argues that capital import, although accounting for only one-sixths of the total import, is the major source of productivity gains from international sourcing. We support this argument using evidence from a large dataset of Chinese manufacturing firms. While both types of imports improve the productivity of importers immediately, the effect is larger from capital import. Moreover, capital imports further enhance productivity dynamically through two additional channels: the direct learning by importing effect and induced R&D effect. For intermediate import, in contrast, the dynamic effects are either statistically insignificant or economically small. All together, capital goods import accounts for over 70% of the total productivity gains from international sourcing in the data period, among which about half is due to the dynamic effects. These findings emphasize the importance of tariff structure in maximizing the productivity gains from tariff liberalization. These effects, however, exist only for ordinary trade but not processing trade. The second chapter of the thesis investigates the impacts of export destinations on firms' innovation. It establishes a link between the average income of export destinations and the number of patents for Chinese manufacturing firms. We use detailed panel data to construct an instrumental variable using real-exchange-rate movements. We find robust and statistically significant evidence that, Chinese firms shipping products to high-income destinations do have fewer number of patents and vice versa. Our IV results indicate this is a causal relationship. This negative pattern is owing to the competition effect related with comparative advantage in innovating. We also find that the result is only observed in industries with high-technology, industries with high capital-labor ratio and within domestic firms. Moreover, we discover the operating mechanism through R&D involvement, which is usually regarded as the input of patents.