A Study On Technological Spillover Effects Of FDI In China's Industrial Sector

This dissertation aims to study FDI spillover in China's industrial sector. Besides aggregate analysis of FDI spillover, it focused on FDI spillover based on linkage effect perspective. Based on spillover direction, the FDI spillover effects can be classified into horizontal (intra-industry) and vertical (inter-industry) spillover. In this dissertation, we use industry-level and firm-level data, as well as a case study of Chinese automotive industry. As for methodologies, it combines the analysis of quantile regression, bootstrap simulation, input-output analysis, logit model, panel data analysis, and etc.The result of aggregate analysis shows that absorptive capacity of domestic country is crucial to FDI spillover. Research based on linkage effect perspective shows that inter-industry spillover is more likely than intra-industry spillover as for the mechanism of FDI spillover. Furthermore, various factors in domestic country are proven to be important for inter-industry spillover.Firstly, using quantile regression method, this dissertation examines the spillover of FDI in China without differentiating intra-industry and inter-industry spillover. The result of nationwide data suggests that improvement of human capital and investment environment is favorable to FDI spillover, while market competition, technological gap and openness play an opposite role. The positive effect of human capital and investment environment is more significant at the lower and upper limit of the conditional distribution respectively. The negative effect of openness and technological gap is most obvious at the bottom and top of the distribution. The negative effect of market competition is more obvious in the center part of the conditional distribution. The result of regional data suggests that the effect of investment environment varies at different regions. At eastern region, positive effect of soft investment environment on FDI spillover is more significant than hard investment environment. At western region, the opposite is true. At middle region, both hard and soft investment environment play important roles. Secondly, considering the difference between intra-industry and inter-industry spillover, this dissertation investigates FDI spillover by employing industry and firm-level data. The findings show that horizontal spillover of FDI is negative but insignificant, while vertical spillover is significantly positive. In other words, inter-industry spillover is more likely than intra-industry spillover as for the mechanism of FDI spillover. The case study of Chinese automotive industry also verifies the importance of vertical linkage channel.Using industry-level and firm-level data, this dissertation discusses various factors related to inter-industry spillover. The result shows that the great gap of productivity between domestic and foreign firms is unfavorable to both backward and forward linkage. The linkage potential also varies by industry. It is easier to source externally when the technology is divisible into discrete stages and services than when it is a continuous process. It is also found that foreign partnership, located in an industrial park or export processing zone, joining in a business association, managers with foreign educational background, as well as higher R&D intensity, are all favorable to vertical linkage. And export orientation of local firms has opposing effects on vertical linkage, which means the effect of export orientation of local firms on vertical linkage needs to be considered synthetically. Finally, the location of domestic firms in the industrial chain can also influence vertical linkage. Forward linkage is more likely for industries with larger diffusion coefficient, while backward linkage is more likely for industries with larger impetus coefficient.Compared with existing literature, the innovative points of this dissertation are as follows: Firstly, this dissertation is innovative in terms of methodology. Utilizing quantile regression, this dissertation analyzes FDI spillover in China's industrial sector, which is the first trial in China. Unlike OLS, quantile regression model allows for a full characterization of the conditional distribution of the dependent variable, so as to avoid partial understanding of economic issue. And this enriches the literature of FDI spillover in China. Secondly, this dissertation is innovative as for research perspective. Using survey data of Chinese firms conducted by World Bank, this dissertation discusses FDI spillover in China's industrial sector based on linkage effect perspective, which is complementary to existing literature. Most of the literature on FDI spillover in China uses macro aggregate data. However, models which apply macro aggregate data have many disadvantages, such as obscuring micro mechanism, small sample size, as well as problem of identification. And models which apply micro data can overcome above disadvantages. Thirdly, the technique of bootstrap is widely used in this dissertation. Even in small sample case, the estimate of bootstrap would be robust, which has better properties regarding both size and power. As a result, the conclusions of this dissertation are more credible.