| Utilizing FDI to promote economic growth is an important strategy for China. The "Two-Gap Model" can't explain our country's current condition. To have a systematic research on the impact of FDI on China's economic growth for the past twenty years, this paper is going to rethink about the impact and its mechanism quantitatively and qualitatively on the basis of theoretical analysis, considering the dimensions of FDI used by China and FDI's sub-effects.Different theories have different ideas about how FDI can promote economic growth. "Two-Gap Model" is not a proper theory to explain current condition in China. The new-classical growth theory only takes FDI as a kind of capital and external shock to promote growth for short-run. While the Endogenous growth theory emphasizes the effect of technological diffusion of FDI on long-run growth, the Institutional diffusion effect of FDI is not explained until the Neoinstitutionalists.In this paper, many dimensions of FDI into China and its latest changes, including amount, scale, means, original countries, distribution among different industries and regions are analyzed. The effects of FDI on China's fixed assets formation, industrial output, foreign trade, employment, taxes and institutional transitions are also explored. An FDI-induced endogenous technological progress model is constructed by making slight changes to the Cobb-Douglas production function in the context of the endogenous growth theory. In this model the contributions of FDI on technical progress are divided into two parts: one is the higher productivity owned by FDI relatively to domestic capital, which is direct effect; the other is technological diffusion of FDI on domestic sector, which is indirect effect. To consider the effect of different restraints in host country on technological diffusion, a model with restraints is also constructed. Since all variables are unstable, EG cointegration analysis is widely used. Using time series data from 1984 to 2002, the econometric analyses to the model without restraints implied that capital and the direct technical contribution of FDI are very robust to economic growth, while the technological diffusion of FDI and the marginal contribution of Labor are not significant. The model with restraints explores the effect of China's economic development level, human capital, R&D expenditure, market development degree, facilities and open degree on technological diffusion and economic growth. Results still show that FDI has promoted China's technical progress and economic growth mainly by direct effect instead of technological diffusion. Furthermore, the economic development level and human capital seem to have hampered technological diffusion, while other restraints seem to have facilitated domestic sector to absorb FDI's technological diffusion. Since all regression coefficients on the technological diffusion of FDI are not significant, further detailed explorations are needed. Chow's breakpoint test shows that the relationship between FDI and economic growth of China has changed since 1992 and later. Cointegration test implies that real GDP and annual FDI flow is cointegrated. Granger causality test shows that the growth of GDP has caused the growth of FDI flowed into China, instead of the contrast.Our research also implies that technological diffusion effect is a very promising mechanism which is worthy of further development for FDI to contribute more to China's future economic growth. To make full use of the technological diffusion of FDI, some measures can be taken: firstly, to upgrade the quality of FDI into China; secondly, to strengthen the absorptive capabilities of domestic sectors; thirdly, to construct and fulfill relevant laws and principles. |