Industry relatedness, FDI liberalization and the indigenous innovation process in China

ABSTRACT

This paper employs a structural innovation model to study the process of indigenous innovation in China and the role of industry relatedness. To better take into account China's transitioning economy context, it further tests to what extent the relationship between relatedness and firms’ innovation process is influenced by the relaxation of foreign ownership controls, an arguably exogenous shock. Controlling for selection, simultaneity and unobserved heterogeneity, the results show that firm research and development (R&D) boosts innovation output, which in turn enhances firm productivity. Relatedness economies are positively related to each phase of innovation, although the size of the effects depends on the type of firm and the stage of innovation. Foreign direct investment (FDI) liberalization encourages firms to rely more on relatedness economies: (1) to complement R&D spending that is required to adapt foreign technologies to local applications; (2) to recombine knowledge from related industries in order to bring forth new proprietary ideas, processes or concepts; and (3) to help solve process or organizational problems faced in related industries.

KEYWORDS:

• relatedness
• foreign direct investment (FDI) liberalization
• innovation
• productivity
• China

JEL:

• D22
• O31
• R10

ACKNOWLEDGEMENTS

The author thanks Canfei He, Yang Rudai, David Rigby, Ron Boschma and other participants at the Evolutionary Economic Geography conference at Peking University, Beijing, for helpful comments.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author.

Notes

1. In models of active learning, for instance, high levels of risk get embedded into the innovation process as firms make successive investments in the innovative search process (Ericson & Pakes, 1995; Nelson & Winter, 1982).

2. Ghemawat and Spence (1985) were the first to caution that as a greater stock of knowledge generated external to the firm becomes freely available, the firm may avoid investing in learning opportunities, such as in-house R&D, as a cost-saving strategy.

3. See Arauzo-Carod, Liviano-Solis, and Manjón-Antolín (2010) for a recent review of the literature.

4. Boschma (2005), for instance, claims that due to the tacit and cumulative nature of knowledge, it is difficult for firms to learn from each other and replicate each others’ routines unless they share close geographical proximity as well as close proximity along other dimensions, including cognitive, social and technological.

5. By contrast, within-industry spillovers can imply excessive cognitive proximity between local firms, which can lead to a cognitive ‘lock-in’ (Nooteboom, 2000), while there may not be enough knowledge overlap for firms to learn from each other in unrelated industries.

6. Innovation risks are higher in transitioning economy contexts due to several issues, including poor intellectual property rights, lack of skilled workers and insufficient domestic demand for new products (Howell, 2015).

7. For an overview of the CDM framework, see Lööf et al. (2017).

8. A threshold of 0.35 is applied to the network, showing 1% of network ties, to better visualize actual relatedness between industries.

 

Additional information

Funding

This project received funding support from the School of Economics at Peking University and the National Natural Science Foundation of China [grant number 71603009].