Prior to its WTO accession in 2001, China’s external trade policy was rather restrictive, particularly so with respect to foreign investment, and more particularly so with respect to the influence foreign firms might exert on their Chinese affiliates.1 Keeping foreign investment under tight control was perceived as an inevitable necessity in order to stay on track with the main emphases of China’s investment policy — to produce exports and to promote technology transfer. Thus, the Catalogue of Industries for Guiding Foreign Investment, issued in 1995, classified FDI into the categories of ‘prohibited’, ‘restricted’, ‘permitted’ or ‘encouraged’, and imposed numerous restrictions and a cumbersome examination and approval system on it, particularly in certain key industrial sectors.
In the run-up to WTO accession, the Chinese government reverted to a policy of making investment in China more attractive to foreign investors. Successive amendments of the Catalogue increased the number of sectors open to foreign investment (from 186 sectors in 1997 to 262 in 2002), allowed for wholly foreign-owned enterprises and removed requirements for these such as exporting a stipulated percentage (which used to be 70%!) of their Chinese production.
To further accelerate the pace of introducing advanced technologies from abroad, China particularly encouraged FDI to its high-tech industries, including electronics and information, software, aeronautics and astronautics. It then turned out that promoting advanced technology and exporting seemed to be closely linked to each other, as the products incorporating the new technologies have become the key to the rapid expansion of China’s exports. And wholly foreign-owned enterprises have become the most popular form of FDI in China.
FDI, technology upgrading and exports
One question now is whether the specific FDI ownership structure, particularly the emergence of many wholly foreign-owned enterprises, has affected China’s technology upgrading and its exports. Two contradicting arguments could be raised. Either, one may expect a higher share of foreign ownership to lead to higher levels of investment in technology and skills, to technology upgrading and to higher export activity. The underlying idea would be that foreign high-tech firms might prefer introducing their valuable new technologies via wholly owned affiliates rather than via joint ventures. Or, one may make a case for a higher share of foreign ownership leading to lower levels of technology investment and skill upgrading. This could be the case because foreign firms might tend to cherry pick the best targets for wholly-owned takeovers, such as those with little need for technology upgrading, or because they might like to integrate wholly-owned affiliates completely into their international production network, stripping them of their R&D activities and relocating these to the headquarters. The theoretical expectation is, therefore, ambiguous and needs to be decided by empirical evidence.
Our recent study analyses such effects of foreign owned firms on China’s take-off after the lifting of the hitherto existing restrictions (Girma et al. 2015). Based on the Annual Reports of Industrial Enterprise Statistics for the period 2001 to 2007, by the China National Bureau of Statistics, some 27,500 Chinese firms are analysed that had no exporting and no R&D activities prior to 2002. The study then looks on the likelihood of these firms to take up exporting and R&D activities, conditional on five different types of ownership:
- Continuous domestic ownership (foreign ownership share 0%; 26,004 cases)
- Small minority acquisition (foreign ownership share < 25%; 152 cases)
- Minority acquisition (foreign ownership share 25 – 49%; 497 cases)
- Majority acquisition (foreign ownership share 50 – 99%; 349 cases)
- Whole acquisition (foreign ownership share 100%; 511 cases)
The data allow to further differentiate between foreign acquisition by ethnic Chinese or truly foreign investors, also between private or state ownership for the remaining domestic share of the acquired firm. The study then sets out to assess the specific effects on the exporting and R&D performing behaviour for the group of firms being acquired by foreign investors, as distinct from the control group of firms remaining completely in domestic ownership. The study takes selection biases into consideration, i.e. it considers that the very characteristics that may make a firm attractive for partial or whole acquisitions may also influence its general likelihood of starting to export or to perform R&D, irrespective of it being actually acquired by a foreign owner. These characteristics may distinguish it from the domestic-owned firms, may impair the utility of the latter to serve as control group, and may lead to biased estimations. The study deals with such selection biases in the observables by adopting propensity score reweighting in combination with covariate adjusted regressions, to obtain a so-called doubly robust estimator.
Figure 1 displays the results of the study, the differential effects of the various kinds of foreign ownership as compared to complete domestic ownership. These effects are estimated for the acquisition year, one year after, and two years after acquisition. Foreign-owned firms are more likely to start exporting after being acquired than firms that remain purely domestic-owned (blue columns being all positive). While this is true for all types of foreign-owned firms, the export starting effect is a little less for firms with majority ownership than for all others. With regard to the effect of whole ownership on exporting behaviour, no obvious advantage or disadvantage over ownership with smaller shares can be detected. The estimated effects are percentage changes, e.g., a wholly-acquired affiliate is about 8% more likely to start to export in the acquisition year than a purely domestic control group firm. Moreover, further calculations reveal that exporting behaviour is usually not increased if the foreign investor is of Chinese ethnicity, but is frequently spurred if the remaining domestic part of the firm is state-owned.
As to taking up R&D activities, the findings are more mixed (red columns being either positive or negative). In several cases, the likelihood of taking up R&D increases after the foreign acquisition, most strongly in the case of minority acquisitions. The estimated effects also tend to increase over time. Yet, at least for whole acquisitions, it is even less likely to start R&D activities than for the still domestic-owned firms. Thus, in this case, foreign investors seem to count on their home facilities of R&D rather than investing in their Chinese affiliate for this purpose. Chinese ethnicity of the foreign investor, again, mostly does not help improving the uptake of R&D activities in China, and state ownership of the remaining domestic part of the acquired firm gives mixed results.
Figure 1. Ownership effects on Chinese firms’ exporting behaviour and R&D behaviour
Note: Average treatment effects from propensity-score weighted doubly-robust regressions; all estimators are significant at the 1% level. Source: Girma, Gong, Görg, Lancheros, 2015.
Summarising, the liberalisation of FDI in China seems a success story from the perspective of Chinese policymakers. The inflows of foreign investment thereafter have contributed considerably to the policy objective of raising Chinese export activities. And they have also contributed to the policy objective of increasing R&D activities in China, though only in the case of joint ventures not in that of wholly-owned firms. The results provide solid evidence that joint ventures between foreign owners and Chinese firms can contribute positively to China’s science and technology take-off.
Text: Agenda By Sourafel Girma Jul 28 2015