Export Spillovers to Chinese Firms: Evidence from Provincial Data

Abstract

Multinational firms are important conduits of managerial skills, foreign market linkages, and technology. Foreign export spillovers associated with multinational firms have the potential to reduce entry costs for local exporting firms. This paper examines whether exports by multinational firms increase the probability of exporting by domestic Chinese firms. The findings from the Probit estimation highlight the varying relationships between multinational exports and local foreign entry based on the type of ownership. The results from separating foreign-invested enterprises into overseas Chinese companies and OECD-based multinational firms suggest that the export activity of the former does not increase the probability of exporting by local firms, whereas the latter positively influence the export decision of local firms, particularly under processing trade.

Keywords:

• Export spillover
• multinational firm
• China
• foreign direct investment

Keywords:

• JEL Classifications: F1
• F14
• F23

Acknowledgements

This paper is part of the author's dissertation requirement at the University of California, Davis. The author is grateful to Robert C. Feenstra, Deborah L. Swenson and Giovanni Peri for guidance, and would like to thank an anonymous referee and seminar participants at the University of California, Davis for helpful valuable comments.

Notes

Notes

¹Motorola trained local engineers in its factory in China, while BMW and DaimlerChrysler transfer know-how to their local suppliers involved in their car production in South Africa (The Economist 2001).

²In Costa Rica, for example, Intel was given a bundle of tax concessions to build a factory (The Economist 2001).

³Cheung & Lin (2004: 1) find that FDI does ‘benefit innovation activity in the host country via spillover channels such as reverse engineering, skilled labor turnovers, demonstration effects, and supplier-customer relationships.’

⁴It is widely noted that, unlike non-state-owned firms, SOEs have access to soft budgets, operate with excess workers and outdated technology, and are poorly managed (Qian & Xu 1993; Sachs & Woo 2001; US Department of State 2001).

⁵ World Investment Report (2003).

⁶See Blomström & Kokko (1998) for a survey.

⁷Refer to Aitken et al. (1997) for more information.

⁸AHH argue that Γ _MNE captures positive spillovers on exports as long as the effects outweigh the potential positive impact on domestic sales.

⁹See Rivers & Vuong (1988) for more information regarding the estimator.

¹⁰AHH notes that estimating only local-owned firms in the empirical analysis further controls for the issue concerning the endogeneity of MNE activity.

¹¹The correlation of the investment per worker between the different firm types is less than 0.30, except for SOEs and COEs, which is 0.67.

¹²The highest correlation between the values of provincial-industry exports by firm type across the dual trade regimes is between PT for COEs and OT for SOEs at 0.60.

¹³Tibet is excluded owing to the lack of data.

¹⁴Data on the preferential policy index run until 1998; therefore, later years are repeated using 1998 values.

¹⁵The results are invariant to the inclusion of the lag of MNE export activity.

¹⁶However, it may also be driven by the endogeneity of the export concentration variables.

¹⁷The variable capturing POEs’ export concentration in the SOEs’ estimation is dropped owing to the lack of variation.

¹⁸Results of the first-stage estimation are available upon request.

¹⁹Rauch & Trindade (1999) find that ethnic Chinese networks facilitate international trade by providing information regarding the international market.

²⁰The OECD export activity excludes the Czech Republic, Hungary, Korea, Mexico, Poland, and the Slovak Republic since these countries gained membership during the period of analysis.