Since the 1990s, most countries, especially the developing countries, have formulated all kinds of generous policies to attract foreign investment, because on one hand foreign investment is an important original funds for their economy growth, on the other hand they believe FDI can improve the domestic productivity through spillover effects. A great number of studies have been done on FDI spillovers. However, with more scientific methods and more suitable data used, more and more negative results are found in spillovers. This put a great question on front of scholars and policy makers. Therefore, in recent years, many scholars have begun to shift their attention to the inter-industry spillovers. In this paper we try to explore the all-round FDI spillovers both from the intra-industry and inter-industry ways.Both theoretical and empirical analyses are used in the study on existence of intra- and inter-industry spillovers and factors that may affect spillovers in our paper. In the theoretical part, a simple game model is used to prove that comparing to intra-industry spillovers, foreign firms prefer the inter-industry spillovers happen. On this base, we next analyses the impact of market structures on inter-industry spillovers in this part. In the empirical part, based on the firm level panel data in Yangtze River Delta Region's manufacturing sectors, we testify our former theoretical conclusion with OLS, FE, LP and DEA methods. The main findings from this study are as follows:First, compared to intra-industry spillovers, inter-industry spillovers is a more important way for FDI spillover effects.Intra-industry spillovers happen from the foreign firms to domestic firms in the same industry. The relationship between them are competitive, so foreign firms will try to prevent such kind of spillovers happen. However, inter-industry spillovers may mean a kind of benefit for foreign firms, because it can cut down the cost or increase the production. So if the spillover costs are low enough, foreign firms are willing to supply necessary conditions for inter-industry spillovers.Second, the conclusions from theoretical and empirical analysis on effect factors for FDI spillovers are as follows:From the study on how technological gap impacts FDI spillovers, we find that intra-industry spillover effects become more obvious with the technological gap increases; spillovers through backward linkages happen only when technological gap is low; the results for spillovers through forward linkages are positive under any condition.From the study on how foreign firms' market orientation impact FDI spillovers, we find that intra-industry spillovers happen from export-oriented foreign firms; positive results are found in the domestic-market-oriented sample group both through backward and forward linkages which don't exist in the export-oriented sample group.From the study on how foreign participation impact FDI spillovers, we find that no obvious intra-industry spillovers happen in all sample groups; positive spillovers through backward linkages are found in non-foreign-controlled firm group, however there is none in foreign-controlled firm group and negative ones in foreign-owned firm groups; consistent with previous analysis, positive spillovers are found through forward linkages in all sample groups, especially in foreign-controlled firm group.From the study on how FDI's original countries impact spillovers, we find that there are negative intra-industry spillovers in all sample groups; positive spillovers through backward linkages are found in the HMT (Hong Kong, Macao and Taiwan) group, and negative in the other group; positive spillovers through forward linkages are found in both groups, and they are more obvious in the other original FDI group.From the study on how market structure impact FDI spillovers, we find that intra-industry spillovers are more prone to happen with higher market concentrations of the industries which FDI entry; however, with lower market concentrations of upstream and downstream industries, the inter-industry spillovers are more prone to happen.Compared to the previous studies, this main innovation of this study is, first, we explore both the intra- and inter-industry spillovers with theoretical and empirical analysis; second, the data we used is firm-level panel data from Yangze River Delta Region's manufacturing sectors. On one hand, this complement the China's present studies almost all of which are done on industry-level data; on the other hand, the linkages between firms are more closely under industry clusters, so FDI spillovers are more obvious under such environment. Third, we not only use OLS and FE estimation methods, but use the latest LP and DEA ones which can effectively solve the bias estimation problems and make the results more convincing. |