Abstract
Foreign direct investment (FDI) has played a pivotal role in the transformation of the Chinese economy. It is the largest host of FDI among the developing world, attracting over one-third of total world investment. China’s success in attracting FDI appears to have been linked with its large domestic market, openness, quality of physical infrastructure and lower wage rates. While China remains an attractive location for foreign investors, our findings point to the need to remove the bias against private sector by privatizing state-owned enterprises to facilitate foreign investment.
Introduction
For over a decade, foreign direct investment (FDI) has played a pivotal role in China’s spectacular growth performance. Since 1990s China has attracted over one-third of the total world FDI and it is now the second largest recipient of FDI in the world after the USA. While China’s ability to attract foreign capital and technology is largely attributed to its open door policy, large domestic market, availability of cheap labour and high quality physical infrastructure, to the best of our knowledge, studies examining the determinants of FDI in China are sparse and they are mainly based on data before China acceded to the World Trade Organization (WTO) in November 2001. 1 As China has witnessed a significant structural change with the accession to the WTO, findings from these studies can provide misleading policy prescriptions. Accession to the WTO has not only enhanced China’s image as a reliable country for investment by reducing the country’s risk for foreign investors, but also provided opportunities to enjoy the most-favoured nation (MFN) status in the world markets, leading to a significant rise in FDI since then. In this study, using provincial data from 1985 to 2006, we investigate the determinants of FDI in China. We do not cover period after 2006 as the 2007 financial crisis has significantly distorted FDI inflows not only in China but globally.
This article is organized as follows: The second section documents investment climate and recent trends in FDI in China, while the third section presents hypotheses and discusses econometric procedures. In the fourth section, we report empirical results and discuss the policy implications of our findings. The article concludes with policy remarks in the last section.
Investment Climate and Recent Trends of FDI in China
Investment Climate in China
China had an inward-oriented Soviet-style economic and political regime until the middle of 1970s. Under the Soviet-style planned economy, foreign-owned firms did not exist on Chinese soil until the middle of 1970s and foreign trade was regulated by government-owned enterprises. It was only after 1978 that foreign firms began to invest in China when Deng Xiaoping opened up the economy for export-oriented firms, a response following the shortage of foreign currency brought about by the relaxation of import restrictions for domestic firms. To facilitate the development of export-oriented foreign firms, four special economic zones (SEZs) were created along the southeast coast—Shenzhen, Zhuhai, Xiamen and Shantou—in 1980. In 1984, another 14 SEZs were established as they became increasingly attractive for export-oriented foreign firms. 2 The 1982 Constitution committed to protect property rights of foreign firms. By the middle of 1980s, export subsidies were introduced and the Chinese currency (RMB) was significantly devalued—from RMB 1.7 to the US dollar in 1981 to RMB 2.9 to the dollar in 1985 (Dwight, 1994). 3 In the early 1980s and early 1990s, SEZs were extended to three deltas—the Pearl River Delta, the Minnan Delta and Yangzi River Delta—and the Hainan province and the Pudong New Area in Shanghai. Foreign firms located in SEZs were granted preferential tax treatment (income tax holiday for two years and thereafter 50 per cent discount on income tax for another three years). In addition to this, a number of provinces offered additional tax incentives and lower land-use fees for foreign firms. By the early 1990s, the central government permitted 100 per cent ownership for foreign enterprises and amended the joint venture (JV) law which permitted a JV agreement beyond 50 years. Furthermore, foreign firms were allowed to operate in more sectors than domestic private firms. 4
In 1992 the government relaxed a number of sectoral and regional barriers relating to foreign investment, decentralized approval authority from the central government to local governments and entered into agreements with the United States to open up its market and protect intellectual property rights vigorously (Lardy, 1992, Lardy 1994). These, together with gradual reduction in tariffs, removal of non-tariff barriers and privatization of state-owned enterprises (SOEs), enabled China to gain the WTO membership in November 2001. 5 Accession to the WTO membership was a major development in sending correct signals to investors that China is committed to integrate its economy with the rest of the world. It not only provided opportunities to China to enjoy the most-favoured nation (MFN) status in the world markets, but also helped attract export-oriented investment by reducing the country’s risk for foreign investors.
It should be noted that while the foreign investment climate has been significantly liberalized over the years, domestic private sector firms in China face several problems (including credit constraint and restrictions in business expansion), which have significantly limited their ability to grow and enjoy the benefit of economies of scale. While there is a bias against domestic private firms, the state-owned enterprises (SOEs) have easy access to subsidize finance and there is excessive investment in technology and capital in these inefficient SOEs, making them an attractive target for foreign firms (Huang, 2003). These SOEs are protected from competition and they also have exclusive market access (Siebert, 2007). Discrimination against private firms and favourable incentives for foreign investment have led to huge inflow of FDI despite high corruption and a poor legal system—factors that deter inflows of foreign firms in other developing countries (Wei, 1996).
Recent Trends and Patterns of FDI
Although China formally opened up its economy in December 1978, it did not attract much foreign investment until the mid-1980s, which appears to be largely due to a bias inherited in the Soviet-style planned economy. However, as China liberalized its FDI policy in the early 1990 FDI inflows gradually increased—from just over US$ 3.2 billion in 1990 to US$ 45.2.5 billion in 2000 to US$ 102.4 billion by 2006 (Table 1), making it the largest recipient of FDI among developing countries and the second largest in the world after the USA. More than 100 per cent increase in FDI during 2000 to 2006 appears to be partly due to China’s accession to the WTO accession in 2001 and partly due to its ongoing commitments to improve physical infrastructure to attract investment in globally integrated production network. The membership to the WTO provided opportunities to China to enjoy the most-favoured nation (MFN) status in the world markets and further attract export-oriented foreign investment especially in assembly of electric and electronic goods whose shares in China export basket rose to about 67 per cent by the mid-2000s from about 30 per cent in the mid-1990s (Athukorala, 2009).
FDI Inflows to China by Region (in constant US$ million)
As shown in Table 1, foreign investment has been heavily concentrated in the Eastern (Coastal) region which provides an excellent infrastructure for export-oriented FDI. Even today, over 79 per cent of foreign investment is located in the coastal regions and, as expected, they are export-oriented. Gungdong—which is located in Eastern region—remains the largest host of FDI. 6 Although a vast majority of foreign firms are located in the coastal area to take advantage of export opportunities, they are also finding inland regions an attractive location. The attractiveness of inland regions for FDI is largely due to a large and growing internal market, and FDI located in these regions are mainly engaged in producing goods for domestic market. As for the central region, the share in attracting FDI has increased from 5 per cent of the total in the mid-1980s to about 15 per cent of the total FDI till now, while the share of Western region in attracting FDI remains very low, attracting only about 5 per cent of the total inward FDI.
It must be mentioned that, over the years, the share of Eastern region in total FDI inflows has gradually declined (from 91 per cent in the early 1990s to less than 80 per cent by 2006) as other regions provide lucrative opportunities for growing domestic market. For instance, the Central region has becoming an attractive location for foreign investors, attracting about 15 per cent of total FDI by the mid-2000s, while the share of Western region in attracting FDI remains very low, less than 6 per cent.
A large proportion of FDI to China has come from Hong Kong, Macao and Taiwan and they are in small-scale labour-intensive industries, namely textiles and clothing sectors. By 1990 the combined share of FDI coming from Hong Kong, Macao and Taiwan was just over 60 per cent, while the share of Japan, US and EU was about 31 per cent. In recent years, however, the combined shares of Hong Kong, Macao and Taiwan have fallen to 47 per cent, while the shares of US, EU and Japan have risen to 38 per cent (see Appendix I). The shares of ASEAN countries (namely Thailand, Indonesia, Malaysia and Singapore) and South Korea in FDI in China have risen from 9 per cent in 1990s to 15 per cent by 2006 (UNCTAD, 2009).
Most foreign investment in China has come in the form of equity joint ventures, contractual joint ventures and fully-owned foreign-owned enterprises. FDI realization rate has improved significantly since 1986 in response to the liberalization of the investment climate for foreign firms (Appendix II). The realization rate reached as high as 98 per cent in 1999, but fell to about 67 per cent by the early 2000, largely due to increased competition in attracting FDI from other developing countries. 7 By the mid-2000s, it remained at above 85 per cent.
Hypotheses and Econometric Model
In this section, we develop an eclectic model to investigate the determinants of FDI in China using theoretical and empirical literature. The proponents of market-size hypothesis have long been argued that the size of domestic market (MS) is a major attraction for foreign investors as large domestic market provides opportunity to exploit economies of scale (Coughlin and Segev, 2000; Wang and Swan, 1995). Available evidence also suggests that countries with a large domestic market tend to attract FDI, leading us to believe a positive link between the MS and FDI flows (Ali and Guo, 2005).
In addition to market size, growth in domestic market (in terms of rising purchasing power) is also an important determinant of FDI (Filippaios et al., 2003; Scaperlanda and Mauer, 1969; Wang and Swan, 1995). However, the nature of the relationships between FDI and growth in domestic market (MG) depends on the motive behind such investments. If the motive behind FDI is to capture export market then we do not expect statistically significant and a positive link between FDI and market growth. However, if the motive for foreign investment is to capture domestic market, then the MG is an important consideration. Thus, the nature of the link between FDI and MG must be investigated empirically.
In a vertically integrated production network, multinational enterprises tend to produce parts and components in different countries (rather than in one single country) based on each country’s intrinsic comparative advantage, which are brought together in a single location for final assembly and distribution. This practice is best achieved in economies that are closely integrated to global supply chain through openness and high-quality physical infrastructure. This suggests that openness (OP) and the quality of physical infrastructure (QI) contribute positively to FDI inflows.
Besides physical infarstructure and openenss, lower labour costs also appear to be an important factor in attracting foreing investment. Lower production costs arise mainly from cheap labour (WAG), the quality of human resouces, an innovative environment and domestic investment in physical infrastructure. This leads us to believe that there is a negative link between WAG and FDI, but a positive link between the quality of human resources (HR), an innovative environment (IE) and the domestic investment in physical infrastructure (DI). Note that labour cost per se may not be a dominant factor once the country moves out of the ‘cheap labour cost’ development model as the economy matures. This suggests the existence of threshold effects: cheap labour cost is important in attracting FDI only in the early stages of development. But in the long run, when wage rates increase, other factors such as strength of governance and the quality of human capital, become more dominant factors in sustaining FDI flows. A relatively stable political environment is an important factor which may result in low production cost. However, we are unable to test this hypothesis due to unavailability of data, and hence it is not included in the model. According to the institutional imperfections hypothesis, the government interference appears to have distorted investment climate in China (Huang, 2003). As state-owned enterprises (SOEs) are not competitive in global market, government tends to support these inefficient enterprises by providing subsidized finance and special privileges, creating a bias against private sector including foreign investment. Thus, we expect a negative link between FDI inflows and the share of SEO/GDP.
Traditionally, Chinese coastal regions had well-developed infrastructure in comparison to inland regions, which has further been developed with the pursuit of open door policy in 1979. This led us to believe that coastal regions have attracted more FDI than inland regions. Thus, we expect a positive sign for the coastal region dummy (CRD).
Based on the above discussions we develop the following baseline model. The expected signs of the parameters in the model are given in parentheses.
where
t = 1985, …., 2006, represents the time period.
FDI = value of FDI inflows, measured by US$ million;
MS = market size, defined by FDI inflows to each province in real terms (US$ million);
MG = market growth defined by the level of real GDP in constant price;
OP = openness, defined by the ratio of total trade (import + export) to GDP;
QI = quality of physical infrastructure, defined by the length of constructed highway;
WAG = wage per person in real term;
HR = human resources, defined as the number of student’s enrollment in higher education;
IE = innovative environment, defined as total patent application certified;
DI = domestic investment, as defined by total domestic investment in infrastructure in constant price;
SOE = value share of State-owned Enterprises to GDP;
CRD = Coastal region dummy, the value of which is 1 for the coastal regions and 0 otherwise.
Empirical Results
The model just specified is estimated using provincial panel data from 1985 to 2006. With the exception of Tibet, all 30 provinces are included in the data set. In total, we have 630 panel observations (e.g., 21 years × 3 regions × 10 variables for each year). Source of data is Chinese Statistical Yearbook, various issues. We have used Pooled Least Squares method to estimate the model specified earlier. To see the sensitivity of our results, we used three different measures of dependent variable. They are: (a) value of contracted FDI inflows (in US$ million), (b) per capita FDI inflows (FDI/population) and (c) contracted FDI per cent of GDP (FDI/GDP). The explanatory power of our models (in terms of R 2 ) varies from 83 per cent to 53 per cent. Our preferred model is per capita FDI inflows as a dependent variable (see, column two in Table 2) because it yields high R 2 than the other two models. However, for comparison purpose we present results from all three models in Table 2. Eview econometric software is used for estimating the model. Results are reported in Table 2.
In all three models, the coefficient for market growth variable (MG) is statistically insignificant. In addition to market growth, the coefficients for human capital (HR) and innovative environment (IE) were also statistically insignificant in our preferred model (see column two of Table 2).
Results of the Determinants of FDI Flows in China, 1985–2006
As expected, the coefficient for market size (MS) variable has an expected positive sign and is statistically significant indicating that a large domestic market is an attraction for foreign investment in China, although most foreign firms in China are aimed at export markets. This finding about the size of market in attracting FDI in China is similar to earlier findings by Ali and Guo (2005).
The statistically significant and positive signs for the coefficients of openness (OP) and the quality of infrastructure (QI) variables tend to suggest that liberalization of trade and investment regime and the quality of physical infrastructure have contributed to FDI inflows to China in the last three decades. This finding is particularly important in the context of China’s rising importance in global production sharing. As Athukorala (2008) has argued, most FDI in China are involved in assembly activities and they need a high-quality infrastructure and a greater degree of openness to be competitive—a finding consistent with our results.
As expected, we find that lower wage rates (WAR) and capital formation through investment in infrastructure (DI) help attract FDI. This finding is consistent with our earlier interpretation that in a globally integrated production network, lower wages and efficient infrastructure are crucial to be competitive. We find that the participation of state-owned enterprises (SOEs) tends to reduce FDI inflows, because of the policy bias in favour of such enterprises in China. Hence, reducing the participation of SOEs in the areas where private sector also operates would facilitate FDI inflows. The statistically significant and a positive sign for the coefficient of coastal region dummy (CRD) suggests that these regions have been major destinations for foreign investment probably because of the easy access to world market. The earlier studies on the determinants of FDI in China by Cheng and Kwan (2000) also found similar results.
Conclusion
In just less than five decades, China has transformed its economy from the rural to urban-based and primary to manufacturing-based economy. This has occurred against the backdrop of China’s open door policy, providing appropriate investment climate for foreign investors. By now China is the largest host of FDI among the developing world, attracting over one-third of the total world FDI. It is now the second largest recipient of FDI after the USA. Our empirical analysis suggests that large domestic market, openness, quality of physical infrastructure and lower wage rates have been key factors in attracting FDI to China in such a huge scale. There is evidence to suggest that coastal regions have been key destination for FDI which appears to be mainly due to easy access to export markets. We also find the evidence that investment in developing infrastructure helps attract FDI, probably by creating favourable business climate. While China remains an attractive location for foreign investors, our findings point to the need to remove the bias against private sector by privatizing state-owned enterprises to facilitate foreign investment. We did not find any statistical evidence to support or reject the hypotheses that market growth, human capital and an innovative environment are crucial in attracting FDI. While these results are interesting, it is important to note the limitations of our study. For instance, we have used a crude proxy for human capital and we were unable to include the role of political stability in attracting FDI to China mainly due to the lack of data. As more data becomes available, future researchers would be better placed to address these limitations.
Footnotes
Appendix I
Realized FDI and Major Source: China, 1986–2006
| Year | Realized FDI (US$ billion) | % Share |
||||
| Hong Kong/Macao | Taiwan | Japan | US | EU | ||
| 1986 | 2.2 | 59.2 | – | 11.7 | 14.5 | 7.9 |
| 1987 | 2.3 | 69.1 | – | 9.5 | 11.4 | 2.3 |
| 1988 | 3.2 | 65.6 | – | 16.1 | 7.4 | 4.9 |
| 1989 | 3.4 | 61.2 | 4.6 | 10.5 | 8.4 | 5.5 |
| 1990 | 3.5 | 54.9 | 6.4 | 14.4 | 13.1 | 4.2 |
| 1991 | 4.4 | 56.9 | 10.7 | 12.2 | 7.4 | 5.6 |
| 1992 | 11.0 | 70.0 | 9.5 | 6.5 | 4.6 | 2.2 |
| 1993 | 27.5 | 64.9 | 11.4 | 4.8 | 7.5 | 2.4 |
| 1994 | 33.8 | 59.7 | 10.0 | 6.2 | 7.4 | 4.6 |
| 1995 | 37.5 | 54.6 | 8.4 | 8.3 | 8.2 | 5.7 |
| 1996 | 41.7 | 50.9 | 8.3 | 8.8 | 8.3 | 6.6 |
| 1997 | 45.3 | 46.5 | 7.3 | 9.6 | 7.2 | 9.2 |
| 1998 | 45.5 | 41.6 | 6.4 | 7.5 | 8.6 | 8.8 |
| 1999 | 40.3 | 41.4 | 6.4 | 7.4 | 10.5 | 11.1 |
| 2000 | 40.7 | 38.9 | 5.6 | 7.2 | 10.8 | 11.0 |
| 2001 | 42.2 | 36.1 | 7.1 | 8.2 | 11.9 | 12.3 |
| 2002 | 44.2 | 34.2 | 7.3 | 8.3 | 12.1 | 12.4 |
| 2003 | 45.01 | 33.1 | 8.1 | 8.4 | 12.8 | 13.4 |
| 2004 | 47.01 | 32.1 | 9.1 | 9.1 | 13.3 | 13.7 |
| 2005 | 47.60 | 31.8 | 6.1 | 9.3 | 13.9 | 14.0 |
| 2006 | 49.29 | 31.0 | 7.1 | 9.0 | 14.5 | 14.9 |
Appendix II
Number of FDI Projects and FDI Realization Rate: China, 1979–2006
| Year | Number of Projects | Contracted FDI (US$ billion) |
Realized FDI (US$ billion) |
Realized FDI % of Contracted |
| 1979–82 | 920 | 5.0 | 1.8 | 36.0 |
| 1983 | 638 | 1.9 | 0.9 | 47.4 |
| 1984 | 2,166 | 2.9 | 1.4 | 48.3 |
| 1985 | 3,073 | 6.3 | 2.0 | 31.7 |
| 1986 | 1,498 | 3.3 | 2.2 | 66.7 |
| 1987 | 2,233 | 3.7 | 2.3 | 62.2 |
| 1988 | 5,945 | 5.3 | 3.2 | 60.4 |
| 1989 | 5,779 | 5.6 | 3.4 | 60.7 |
| 1990 | 7,273 | 6.6 | 3.5 | 53.0 |
| 1991 | 12,978 | 12.0 | 4.4 | 36.7 |
| 1992 | 48,764 | 58.1 | 11.0 | 18.9 |
| 1993 | 83,437 | 111.4 | 27.5 | 24.7 |
| 1994 | 47,549 | 82.7 | 33.8 | 40.9 |
| 1995 | 37,011 | 91.3 | 37.5 | 41.1 |
| 1996 | 24,556 | 73.3 | 41.7 | 56.9 |
| 1997 | 21,001 | 51.0 | 45.3 | 88.8 |
| 1998 | 19,799 | 52.1 | 45.5 | 87.3 |
| 1999 | 16,918 | 41.2 | 40.3 | 97.8 |
| 2000 | 22,347 | 62.4 | 40.7 | 65.2 |
| 2001 | 26,140 | 69.2 | 46.9 | 67.8 |
| 2002 | 27,921 | 72.3 | 49.8 | 72.4 |
| 2003 | 28,900 | 75.4 | 65.0 | 87.8 |
| 2004 | 29,978 | 76.9 | 67.8 | 88.0 |
| 2005 | 30,872 | 77.8 | 69.3 | 89.6 |
| 2006 | 35,672 | 80.5 | 70.1 | 87.5 |
