Spillover Effects of FDI in China: From the Perspective of Technology Gaps

On January 1, 2008, the Chinese government partly reduced the privileges enjoyed by FDI firms. This policy change again put the effects of FDI into public focus. Using Chinese industry-level panel data, this paper analyzes the spillover effects of FDI from the perspective of technology gaps (GAP). Unlike most previous studies that only analyze two or three levels of GAP, we instead treat it as a continuous explanatory variable. Also, we propose a more accurate measure for GAP. To overcome the difficulty of measuring spillover effects, we transform the spillover regression into the output regression. The method of DEA and a new set of instrumental variables are also employed to solve the problems of misspecification and endogeneity. We find that the spillover effects are negative and have a U-shaped relationship with GAP. These results are robust to various model specifications and estimation methods. We interpret our seemingly counter-intuitive findings as follows: the overall spillover effect can be divided into three components--the increasing "learning-room effect," the decreasing "learning-ability effect" and the "crowding-out effect," which is uncorrelated with GAP. Mainly because of the strict controls that FDI firms place on their core technologies, the "brain drain" from domestic firms to FDI firms, and the GDP-oriented behaviors of Chinese municipal officials, the negative "crowding-out effect" dominates the other two positive effects, and as GAP decreases from a small initial value, the increasing "learning-ability effect" dominates the decreasing "learning-room effect" (and vice versa). Consequently, a policy of reducing the privileges of FDI firms in industries with middle-sized technology gaps is suggested.