1. Capital or Know-how: The Role of Multinationals in Sino-Foreign Joint Ventures Chong-En Bai, Tsinghua University Jiangyong Lu, Peking University Zhigang Tao, HKU

2. The Importance of FDI FDI flow into developing countries: $3.1b in 1770, $10.9b in 1980, $23.7b in 1990, and $185.4b in 1999 (WDI 2001). In 1999, FDI accounted for 84% of total net private capital flow into developing countries (WDI 2001). In 2007, FDI accounted for 95% of total net private capital flow into China (China Statistical Yearbook, 2008). The amount was 75 billion dollars.

3. China’s Open Door Policy Between 1979 and 2007, China attracted more than US $760 billion FDI (actually utilized) involving over 632,000 projects (China Statistical Yearbook, 2008). Between 1979 and 1997, 51% of FDI in China was invested in equity joint ventures.

4. Debate on the roles of FDI in China’s economic transition One view is that FDI supplied by multinationals has brought China much-needed knowhow, a term broadly defined to include advanced technology, modern management practices, and access to international markets. On the other extreme is the critical view that China’s huge influx of FDI is in fact a reflection of its discriminating financial system, under which less privileged domestic enterprises such as China’s private enterprises face difficulties in access to external finance and are forced to sell their companies for capital (Huang, 2003). Debate on the roles of FDI in China’s economic transition has become increasingly widespread and contentious, given that China has maintained a huge trade surplus invested in ever-depreciating U.S. treasury bonds yet continues to rely on FDI for economic development.

5. Our objectives Consider the roles played by the foreign partners in joint ventures. Do they bring in know-how? What kinds of know-how do they bring in? Do they bring in capital to relax the liquidity constraints of the domestic firms?

6. Why is the answer important? If foreign investors do not bring in any know-how (including the know-how to choose investment target and reduce incentive costs), then their only role is to relax the liquidity constraint of domestic firms. If so, they may buy Chinese assets too cheaply. Huang’s argument: The reason for the large flow of FDI to China is because domestic firms face unnecessary liquidity constraints, and it is harmful for the Chinese economy. Liquidity constraints should be relaxed through other means.

7. Why is the answer important? If foreign partners in JVs do bring in know- how, what kind? Some know-how has spillover effects and its transfer should be particularly encouraged. Some know-how can be learned or developed by domestic firms. If foreign ownership has its costs, then such learning and development of such know-how should be encouraged.

8. Why is the answer important? We can also find out what roles domestic partners play in the JVs. To attract more FDI, the enhancement of the ability for domestic firms to play their roles is important.

9. Why is the answer important? In the process of our research, we also test the empirical relevance of theories of the firm, in particular theories about the joint venture.

10. Existing evidence Aitken and Harrison (1999, AER): In Venezuela, foreign equity participation increases the productivity of the joint venture but does not have spillover effect on other firms. Aitken, Hanson, and Harrison (1997, JIE): In Mexico, domestic firms located near multinational exporters are more likely to export than other firms.

11. Existing evidence Borensztein, De Gregorio, and Lee (1998, JIE): –FDI contributes relatively more to growth than domestic investment, suggesting FDI helps the transfer of technology. –FDI may crowd in capital accumulation, but the result is not robust.

12. Our method According to the incentive theory of team production, the partner who plays a more important role should be given more equity shares. By estimating the determinants of the foreign partner’s equity share, we can infer what roles the foreign partner and the domestic partner play respectively.

13. Data used ( 中国对外经济贸易年鉴 Almanac of China’s Foreign Economic Relations and Trade) 5217 joint ventures in manufacturing industries in China between 1985 and 1996 Equity shares of the foreign and domestic partners Location of the JV: 28 provinces are represented Home country of the foreign partner: 51 countries and regions 29 two-digit manufacturing industries are represented

14. The roles of the foreign and domestic partners Bai, Tao, and Wu, 2004, Rand

15. FDI as a transfer of knowhow Drawing insights from the team production theory and the theory of the firm (see, for example, Holmstrom (1982), Grossman and Hart (1986), Hart and Moore (1990), Hart (1995), and Bai, Tao and Wu (2004)), we know that equity sharing is determined by the relative importance of the partners’ tasks or investments. For an equity joint venture to be successful, the partners need to cooperate in various stages of the operation, ranging from input procurement, production processes up through final sales or provision of services.

16. Traditional proxies for knowhow It is generally agreed that foreign multinationals (foreign partners of equity joint ventures) are strong in technology while local firms (local partners of equity joint ventures) have expertise in sales and services in their local market. Hypothesis 1: The equity shares of foreign partners increase with industry R&D intensity. Hypothesis 2: The equity shares of foreign partners decrease with industry advertising intensity.

17. Variations in the relative importance of the foreign partners’ contributions, Qualifications of hypothesis 1 Even in a given industry, multinationals coming from different countries may have different levels of technological capabilities and hence different contributions to the success of the equity joint ventures. To capture these variations we construct a variable called revealed comparative advantage (or RCA). It is the ratio between the share of country i’s total export contributed by industry j and the share of China’s total export contributed by the same industry j. Hypothesis 3: The equity shares of foreign partners increase with the revealed comparative advantage of their countries of origin in the concerned industries.

18. Qualifications of Hypothesis 2 Hypothesis 2 is qualified by the fact that local partners generally contribute to the sales and services of joint venture output only in their home markets. In practice, FDI in China could be both vertical (China as the “world’s factory”) and horizontal (China as a huge domestic market). Correspondingly, the contribution of local partners in the sales and services of joint venture output varies from horizontal FDI (significant) to vertical FDI (not significant). As a result, the relation between the equity shares of the foreign partners and the importance of marketing (represented by advertising intensity) could be ambiguous.

19. Chinese partners help JVs navigate through bureaucracy While local partners traditionally help joint ventures market output in their local economies, an arguably more important role in the emerging economies is navigating the joint ventures through an imperfect market environment. Following Djankov et al (2002), we construct an index of bureaucracy for various regions of China, using data on the average time needed for the registration of joint ventures after signing joint venture contracts. Hypothesis 4: The equity shares of the local partners are higher in regions with worse bureaucracy.

20. Chinese partners help JVs deal with SOEs Dominance of state ownership State-owned enterprises may have objectives (such as social stability) different from profit maximization The local Chinese partners of equity joint ventures, often state-owned, have advantages over their foreign partners in dealing with state-owned suppliers, customers, and even other competitors. Hypothesis 5: The equity shares of the local partners are higher in regions and/or industries with higher degrees of state ownership.

21. FDI as a transfer of capital In general, developing countries suffer from a lack of capital for economic development. However, it is not clear why the transfer of capital should be carried out through FDI as opposed to private debt flows. The case of China stands out among developing countries due to the fact that it has attracted a substantial amount of FDI in the form of equity joint ventures, even though China has simultaneously accumulated a vast quantity of foreign reserves invested mostly in low-yield U.S. treasury bonds. So the question arises as to whether China is really in need of foreign capital, especially in the later years of economic reform.

22. Inefficient domestic financial institutions One possible explanation is that China’s financial system is inefficient and incapable of channeling hard-earned foreign reserves and huge domestic savings to local firms in need of capital. Discriminations against China’s private enterprises in their access to external finance [see Bai, Lu and Tao (2006b) on bank loans, and Du and Xu (2008) on listing in the stock market]. As a result China’s private enterprises have to sell their equity shares to foreign multinationals for needed capital (Huang, 2003).

23. Proxies for the need of capital The industry capital-labor ratio captures the variations across industries in the need for capital, and size of project investment captures possible variations within industries in the need for capital. It is expected that foreign multinationals have greater bargaining powers and hence more equity shares in more capital intensive industries and in those projects of larger investment size. Hypothesis 6: The equity shares of foreign partners increase with the industry capital-labor ratios. Hypothesis 7: The equity shares of foreign partners increase with the size of the project investment.

24. Investment returns The industry average profit margin is indicative of how quickly foreign multinationals can recoup their investments, affecting the relative bargaining powers between foreign and Chinese partners. Hypothesis 8: The equity shares of foreign partners are lower in those industries with higher profit margins. One may argue that industries with high profit margins such as resource-based industries are often protected industries and foreign equity investments are restricted. In our empirical study, the observations in those resource-based industries are excluded to make sure that there are unambiguous interpretations of our results.

25. Capital endowments Different home countries of foreign multinationals have different capital-labor endowments. It is expected that multinationals from more capital- endowed countries have greater bargaining power in negotiating with local firms in China, and they have even more leverage against the local firms in those industries of higher capital-labor ratios. Hypothesis 9: Equity shares are higher for those multinationals from more capital-endowed countries. Hypothesis 10: The positive impact of capital endowment of FDI source countries on foreign equity share is greater in industries with higher capital labor ratios.

26. Empirical results

28. FDI as a transfer of knowhow R&D intensity and Revealed comparative advantage have positive coefficients with 1% statistical significance, which imply that foreign equity shares are higher in industries with higher R&D intensity and for multinationals whose countries of origin have higher revealed comparative advantage in the concerned industries. Bureaucracy, Share of SOE by region, and Share of SOE by industry all have negative coefficients with 1% statistical significance. These results suggest that equity shares of the local partners are higher in poorer business environment. The coefficient of advertising intensity is negative but not statistically significant, because much of FDI in China is vertical FDI (“China as a factory”).

29. Strong support for FDI as a transfer of knowhow Our results on R&D intensity and advertising intensity are largely consistent with the results reported in the literature focusing on industry-level variations in the relative contributions of multinationals vis-à-vis local firms. Our result related to revealed comparative statics, however, introduces a novel way of measuring within-industry variations in the relative contribution of multinationals vis- à-vis local firms. More importantly, our findings on state ownership and bureaucracy highlight the importance of local partners in poor institutional environments, which are characteristic of developing and transition economies. Taken together, the regression results summarized in column 1 of Table 3 lend strong support to the hypotheses on FDI as a transfer of knowhow as a whole and to hypotheses 1, 3, 4, and 5 in particular.

30. Partial support for FDI as a transfer of capital Project investment has a positive coefficient with 1% statistical significance, while industry capital labor ratio has a positive coefficient with 5% statistical significance. These results suggest that multinationals enjoy more bargaining power and hence obtain higher equity shares in those industries that require more capital per labor and in projects of larger investment (hypotheses 6 and 7). Profit margin is negative as predicted, but statistically insignificant. The impact of home country capital labor endowment is positive but statistically insignificant; however, its interaction term with industry capital labor ratio is positive & statistically significant. In other words, multinationals from countries with more capital endowment enjoy more bargaining powers and hence obtain higher equity shares only in those industries that require more capital per labor.

31. The degree of joint control Our data also allow us to investigate the determinants of control rights allocation in the joint venture. When the foreign partner’s equity share is closer to 50%, it is more likely that the two partners jointly control the JV. The degree of joint control is defined to be the distance between the foreign partner’s equity share to 50%.

32. The degree of joint control Bai, Tao, and Wu (2004, Rand) Joint control is desirable when it is difficult to prevent one partner from expropriating the other partner. Expropriation is less of a problem when –The foreign partner comes from a country sharing the same cultural background as the domestic partner. –The legal environment is favorable. –The JV does not have much profit to be expropriated.

34. Conclusion Increasingly contentious debates in China on the roles of FDI after 30 years of economic reform Our empirical analysis offers strong evidence for FDI as a transfer of knowhow, but also lends limited support for FDI as a transfer of capital. The support for FDI as a transfer of knowhow is reassuring for China’s open-door policy, but the limited support for FDI as a transfer of capital points to the need for further reforms in China’s financial system.

35. Conclusion Indeed, due to the poor protection of property rights as well as state ownership of financial institutions including banks and stock exchanges, privately owned enterprises have had difficulties in accessing external finance (Bai, Lu and Tao, 2006b; Du and Xu, 2008), and as a result they have to sell their equity shares to multinationals for capital (Huang, 2003). Our study thus calls for improvements in China’s economic institutions including property rights protection and contract enforcement, which would further unleash the competitiveness of the Chinese economy.