Discussion and concluding remarks

This paper examines whether or not advanced knowledge of MNEs spills over to domestic firms and if so, under what conditions such spillovers take place, using a firm-level panel data set for a high-tech cluster in China. We are particularly interested in the role of MNE employment of educated workers and possible differences in spillover effects between Japanese and US MNEs.

The results from the estimations of this paper lead to the following conclusions. First, MNE knowledge in the Z-Park does not unconditionally spill over to domestic firms in the same industry. However, when MNEs employ highly educated workers, such spillovers in fact take place. A possible reason for this finding is that MNE's advanced knowledge remains in a black box for local workers when MNEs engage only in production activities that require unskilled labor. However, when MNEs engage in advanced activities, such as R&D, using local skilled labor, the black box opens and those educated workers can learn the advanced knowledge of MNEs. Then, the advanced knowledge spills over to domestic firms through labor turnovers, startups of new firms, and technological collaboration between MNEs and domestic firms, as suggested in Section 2. In addition, the size of such FDI spillovers is larger in more R&D-intensive industries.

Second, we find that Japanese MNEs in the Z-Park contribute less to spillovers to domestic firms than non-Japanese MNEs, which include US MNEs. This is most likely due to the fact that Japanese MNEs employ educated labor much less than non-Japanese MNEs.

Finally, we should emphasize a number of limitations of this paper. First, the analysis of this paper focuses on intra-industry knowledge spillovers. Although FDI may also benefit upstream industries through inter-industry spillovers, or vertical spillovers, as Javorcik (2004), Kugler (2006), Blalock and Gertler (2008), and Liu (2008) find, we generally disregarded such vertical spillovers. Therefore, our findings do not rule out the possibility of vertical spillovers from FDI through production activities. Second, our data set allows for only a short panel data which may underestimate the dynamic effect of FDI spillovers. Third, while the Z-Park provides us with a desirable context for examining technology diffusion via interpersonal interactions and labor mobility within a geographically concentrated unit, a study based on one park also limits the generality of our research results. We expect future research to be conducted which will address these limitations.

Download this Paper [ PDF 194.5KB| 27 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.