On 22 March 2018, the Trump administration announced its intention to impose tariffs of $60 billion on Chinese imports. In doing so, it escalated its rhetoric in the simmering dispute over China’s official policy towards the intellectual property of foreign investors. US Trade Representative (USTR) Robert Lighthizer claimed that China was guilty of “state-led efforts to force, strong-arm, and even steal US technology and intellectual property.” The US government’s claims of unfair Chinese trade practices are detailed in the findings of a Section 301 investigation (executive summary here).
The USTR report argued that China uses onerous restrictions on ownership and technology licensing to compel foreign investors to transfer advanced technology to domestic firms. Part of this is the requirement in specific industrial sectors to form international joint ventures (IJVs), which are legal partnerships between a domestic firm and a foreign investor, to form a new operation in the domestic market. Those sectors typically make widespread use of proprietary technologies, intellectual property, and advanced production methods – for example pharmaceuticals or automobile production.
China’s requirement that foreign investors form a joint venture with a domestic partner in particular industries has been relaxed in recent years, leading to more wholly foreign-owned enterprises, but IJVs still account for billions of dollars of annual investment inflows and offer a unique environment for the diffusion of foreign technology and practices.
In this column we unpack some of the issues pertaining to joint ventures in China – one of the foremost recipients of foreign direct investment (FDI) conducted via IJVs. We examine the motives for foreign firms to form such partnerships, and the effects of the partnerships on Chinese firms, both inside and outside the joint venture.
FDI, joint ventures, and technology transfer
There are clear incentives for foreign investors to form joint ventures. A domestic partner can help the foreign partner navigate the complexities – regulatory, cultural, and others – when entering a foreign market. Existing research (e.g. Kogut 1988, Geringer 1991) suggests the characteristics of the ideal domestic partner in a joint venture: established market share, plus a well-developed capacity for innovation or a strong export orientation. But none of this early work conducts a comprehensive empirical analysis.
For the host country’s government, the policy arguments for mandating joint ventures (or encouraging FDI in general) are also well-established: foreign multinationals bring advanced technologies and managerial expertise. Policymakers hope that domestic firms will benefit from these, through direct or indirect channels.
The landmark studies on the extent of technology transfer that takes place with FDI (e.g. Javorcik 2004, Keller and Yeaple 2009) argued that FDI generates broad learning externalities at the industry level, as multinational enterprises transfer their technology and methods to their operations in destination markets. These innovations diffuse directly or indirectly to domestic firms. As penetration of FDI increases, these firms tend to exhibit higher levels of productivity or engage in more innovation.
FDI may also lead to negative effects. In this case, foreign competitive pressure saps the market share and innovative capacity of domestic firms.
Should we expect IJVs, an unusual mode of FDI, to generate the same sorts of outcomes? In our recent work (Jiang et al. 2018), we posit several channels through which the international diffusion of technology might take place (Figure 1).
Figure 1 Joint venture formation and technology transfer
Source: Jiang et al. (2018).
We might expect the newly established firm created as the result of an IJV to benefit from the technology and practices imparted by its foreign benefactor. This first channel is internal technology transfer: a foreign partner firm imparts its know-how, technology, or intellectual property for the benefit of the joint venture. It does this either because it wants the new operation to succeed, or because of legal requirements.
The Chinese partner in the joint venture might benefit indirectly from this internal effect, as technology ‘leaks’ to the Chinese partner (perhaps through the transfer of intellectual property or labor turnover). IJVs generate spillovers on their domestic ‘parents’. We call this the intergenerational technology transfer effect.
Finally, in light of the prolific literature on externalities arising from FDI, we consider the external effects of IJV formation. We ask whether firms outside of the joint venture are themselves impacted by the penetration of joint ventures into the industry in which they operate through the spread of foreign technology. The existence of these channels, and the magnitude of their effects, are empirical questions that we are able to investigate.
Empirical evidence for partner selection and technology transfer
To address the questions detailed above in an econometric framework, we use a novel administrative dataset accounting for all joint ventures in China from 1998 to 2007. Our dataset contains detailed attributes of the firms created by any IJV, the domestic Chinese partner firms, and all other Chinese firms with at least RMB 5 million ($795,000) in annual sales operating during the time period of our sample.
We explore which Chinese firms are most likely to have been selected as IJV partners. In line with predictions from the early literature on the determinants of joint venture formation, we find that the Chinese firms most likely to be chosen as partners in IJVs are larger, more productive, and engaged in more innovation and patenting. This is not surprising, but these large, productive firms at the frontier of innovation will be the firms that possess the absorptive capacity (Cohen and Levinthal 1990) to benefit from the intergenerational technology transfer effect.
Next, we examine the evidence for the three channels of technology transfer.
- Internal: If there is an internal technology transfer channel between foreign investors and joint ventures, we might expect to observe joint venture firms performing highly along dimensions of productivity, innovation, and market share, among others. On average, they do. Controlling for other firm characteristics, joint ventures tend to be 30% more productive, as measured by total factor productivity. They have higher sales, and engage in more patenting than non-joint venture Chinese firms. This is evidence consistent with the beneficial transfer of technology.
- Intergenerational: A similar effect is apparent for Chinese joint venture partners, though much more muted than for joint ventures. The partners are only around 5% more productive than other Chinese firms.1 While foreign investors have a strong incentive to transfer technology directly to their joint ventures, no such incentive exists with regard to their Chinese partners. Nevertheless, foreign partners are able to indirectly benefit from the technology conferred on their joint venture offshoots.
- External: The overall external effect of IJVs might manifest as technology spillovers, enhancing the productivity of domestic firms through broad learning effects, or as market share rivalry effects, in which high-performing foreign competition inhibited the performance of domestic firms. We do not isolate the effects of these two mechanisms, but our estimates suggest that the learning effect generally dominates. When joint venture firms (or the domestic partners) account for a larger share of the market in a particular industry, other Chinese firms have substantial gains in their own productivity. On average, when the share of industry sales by joint ventures increases by 10 percentage points, other firms in that industry are 10% more productive. For the effect arising from domestic partners, an increase of 10 percentage points in IJV partners’ share of industry sales yields around a 4.5% increase in productivity of other firms. The origin of the foreign investor matters. We compare investors from Hong Kong, Macau, and Taiwan to Japan, and also to the US. We find that US joint ventures account for the bulk of these externalities.
Joint ventures and FDI policy
US policymakers have aired their grievances over Chinese foreign investment policy. But the underexplored benefits to China of encouraging or requiring joint ventures are clear. We show that these arrangements between domestic firms and foreign partners generated far-reaching impacts, for firms inside and outside the joint venture. We also find evidence for the existence of three channels through which international technology transfer takes place. This is an important consideration for any country attempting to formulate optimal policy towards foreign investment.
Bloom, N, M Schankerman, and J Van Reenen (2013), “Identifying Technology Spillovers and Product Market Rivalry,” Econometrica 81(4): 1347–1393.
Cohen, W and D Levinthal (1990), “Absorptive Capacity: A New Perspective on Learning and Innovation,” Administrative Science Quarterly 35(1): 128–152.
Geringer, J (1991), “Strategic Determinants of Partner Selection Criteria in International Joint Ventures,” Journal of International Business Studies 22(1): 41–62.
Javorcik, B (2004), “Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillovers through Backward Linkages,” The American Economic Review 94(3): 605–627.
Jiang, K, W Keller, L D Qiu, and W Ridley (2018), “International Joint Ventures and Internal versus External Technology Transfer: Evidence from China,” CEPR Discussion Paper 12809.
Keller, W and S Yeaple (2009), “Multinational Enterprises, International Trade, and Productivity Growth: Firm-level Evidence from the United States,” The Review of Economics and Statistics 91(4): 821–831.
Kogut, B (1988), “Joint Ventures: Theoretical and Empirical Perspectives,” Strategic Management Journal 9(4): 319–332.
 We controlled for the fact that highly productive firms are more likely to be chosen as domestic joint venture partners, which would reverse the direction of causality in our empirical exercise, using a methodology known as inverse probability weighting with regression adjustment. This process first estimates the propensity that a given firm is chosen as a joint venture partner as a function of its firm-level characteristics. These propensity scores were used as weights in an ordinary least squares regression. Firms were weighted more heavily if they were unlikely to be picked to form an IJV but were, or if they were likely to be picked to form an IJV but were not. This allows us to account for the nonrandom selection into treatment and control groups.