Journal of Policy Modeling 32 (2010) 155–158

Available online at www.sciencedirect.com

Foreign direct investment inflows andeconomic growth of China

Jai S. Mah ∗Division of International Studies, Ewha Womans University, Seodaemun-ku, Seoul 120-750, South Korea

Received 1 November 2008; received in revised form 1 June 2009; accepted 1 September 2009Available online 11 September 2009

Abstract

This study examines the causality between FDI inflows and economic growth in case of China using asmall sample cointegration test. The empirical results show that since economic reform FDI inflows havenot caused economic growth, but the latter has caused the former.© 2009 Society for Policy Modeling. Published by Elsevier Inc. All rights reserved.

JEL classification: F21; F43; O53

Keywords: China; FDI; Economic growth

1. Introduction

It has often been argued that FDI inflow is one of the driving forces of economic growth indeveloping countries. Historical evidences are mixed. Of the so-called four east Asian tigers,Hong Kong and Singapore succeeded in attracting huge amount of FDI; however, South Koreaand Taiwan did not attract it so much.

Since economic reform in 1979, China has recorded remarkable economic growth rates; forinstance, the annual average real GDP growth rate was as high as 8.3 percent during 1979–2001.In the meantime, China has actively tried to attract FDI inflows. FDI flows into China increasedfrom US$ 0.9 billion in 1983 to US$ 46.9 billion in 2001 (Table 1). Chen, Chang, and Zhang(1995) evaluated that FDI had contributed to China’s post-1978 economic growth by augmentingresources available for capital formation and export earnings. Meanwhile, the issue on the causality

∗ Tel.: +82 2 3277 4683; fax: +82 2 365 0943.E-mail address: jsmah59@yahoo.com.

0161-8938/$ – see front matter © 2009 Society for Policy Modeling. Published by Elsevier Inc. All rights reserved.doi:10.1016/j.jpolmod.2009.09.001

156 J.S. Mah / Journal of Policy Modeling 32 (2010) 155–158

Table 1FDI inflows and economic growth.

Years FDI Inflows (US$ million) Annual average real economic growth ratea (%)

1983–1985 4,291 12.41986–1989 11,145 8.01990–1993 11,376 11.31994–1997 158,271 10.41998–2001 173,375 7.1

a Real GNP growth rate until 1990 and real GDP growth rate since 1991.

between FDI inflows and economic growth in case of China has seldom been analyzed rigorously.Although Qin, Cagas, Quising, and He (2006) revealed that economic growth caused investmentin China, the latter was not restricted to FDI. Using a simulation, Dees (2001) showed that FDIhas a long-run effect on output. The current study tests whether or not increase in FDI causedincrease in economic growth of China with a cointegration test procedure allowing for differentorders of integration and the Granger causality test using stationary data.

2. Factual background

The investment atmosphere in China had not been friendly to foreign investors until 1978. Anew policy towards foreign investment allowing foreign firms to operate in China was proclaimedin 1979, which granted foreign investment a legal status in China (Chen et al., 1995: 692). TheSpecial Economic Zones (SEZs) were established in the coastal areas, which granted the investorsadministrative support and tax benefits like profit tax reduction (Park, 2002: 21). The decision toopen up China to the world economy was formally included in the 1982 state constitution. In 1984,the concept of SEZs was extended to another fourteen coastal cities and Hainan Island. In 1985,development triangles were established for the encouragement of foreign investment. Severalinvestment incentives were provided to the SEZs (Chen et al., 1995). Despite such measures, theamount of FDI inflows was not substantial.

Since Deng Xiaoping’s Southern Trip in 1992, China began to strengthen the market econ-omy. Preferential taxation schemes were provided to foreign investors as well (Park, 2002: 23).Consequently, FDI flows into China began to increase substantially in 1992. In late 1997, Chinaintroduced various measures to attract FDI, including import tariff reductions and began to pro-vide preferential tax measure to the finance sector in 1999 (Park, 2002: 26–30). Although Chinaattracted huge amount of FDI inflows, the cumulative FDI amount per capita was only US$ 200in 1998, for instance, which was lower than those of most other transition economies.

3. Empirical evidence on the causality between FDI and economic growth

Alguacil, Cuadros, and Orts (2002) examined the Granger causality from exports and FDI tooutput in Mexico. Their empirical evidence supported not only the export-led growth hypothesisbut also the existence of an FDI-led growth relationship. Basu, Chakraborty, and Reagle (2003)found the bi-directional causality between FDI and GDP for open economies, while, for closedeconomies, although causality was bi-directional in the short run, it ran mainly from growth toFDI in the long run. Cuadros, Orts, and Alguacil (2004) examined both the export-led growthhypothesis and the FDI-growth nexus in Mexico, Brazil and Argentina. Their results based on the

J.S. Mah / Journal of Policy Modeling 32 (2010) 155–158 157

Johansen cointegration test and Granger causality test suggested it to be important to considerboth exports and FDI to ascertain the benefits associated with the outward orientation.

To examine the causal relationship between FDI inflows and economic growth in China, thissection uses the annual data for China during the period 1983–2001. The augmented Dickey–Fullertests show that the levels of FDI inflows and real economic growth rates are not stationary at 5percent level of significance. Optimal lags are chosen by Akaike’s final prediction error (FPE)criterion. FDI and real GDP growth rate are revealed to be integrated of order two and one,respectively, at 5 percent level of significance. The existence of a long-run equilibrium relation-ship among the concerned non-stationary variables is examined by Pesaran and Shin’s (1999)autoregressive, distributed lags model, which allows to test the cointegrating relationship amongthe underlying variables of different orders of integration with small sample sizes. It estimates thefollowing model and calculates the F-statistics of the lagged variables, FDI and real GDP growthrate expressed in terms of Y.

dY (t) = a0 +k∑

i=1

bi dY (t − i) +k∑

i=1

ci dFDI(t − i) + h1Y (t − 1) + h2FDI(t − 1) + w(t)

(1)

where d in front of the variables under consideration and w denote the first differenced formsand the conventionally assumed error term, respectively. Data for FDI inflows and real GDPgrowth rate are taken from Park (2002) and IMF, International Financial Statistics Yearbook2003, respectively.

The null hypothesis in Pesaran and Shin’s (1999) test is the non-existence of a long-run equi-librium relationship among the variables, which can be denoted as H0: h1 = h2 = 0, against thealternative hypothesis that each of hi is not zero. Assuming the lag number to be one in the firstdifferenced forms of the concerned variables in Eq. (1), the calculated F2,9 statistics, 2.324, isnot significant at any reasonable level of significance. Increasing the lag number up to four in thefirst differenced form of the concerned variables in Eq. (1) does not change the overall resultsqualitatively.

Table 2Granger causality test results.

Number of lag k H0: FDI does not cause Y H0: Y does not cause FDI

2b 3.516*

3b 4.512**

4b 4.784**

2a 0.0553a 0.0014a 0.001

a Y (t) =k∑

i=1

aiY (t − i) +m∑

j=1

bjFDI(t − j).

b FDI(t) =k∑

i=1

ciFDI(t − i) +m∑

j=1

djY (t − j).

* Statistically significant at 10 percent level of significance.** Statistically significant at 5 percent level of significance.

158 J.S. Mah / Journal of Policy Modeling 32 (2010) 155–158

Since the cointegration tests reveal that there does not exist any long-run equilibrium relation-ship, I use the Granger causality test based on the first differenced data for real economic growthrate and the second differenced data for FDI inflows to reveal the causal relationship between theformer and the latter. Table 2 shows the results of the Granger causality tests, where the numberof lags are chosen by Akaike’s FPE criterion. According to Table 2, the null hypothesis thatFDI inflow does not cause real economic growth rate is not rejected at any reasonable level ofsignificance, which is intuitively plausible in the sense that, although FDI inflows were not sosubstantial in the 1980s, the Chinese economy already recorded very rapid economic growth inthe same period. Meanwhile, there are evidences that economic growth caused FDI inflows. Thatis, the null hypothesis that the former does not cause the latter is rejected at 5 or 10 percent level ofsignificance. It means that the rapid economic growth of China since economic reform attractedforeign capital to realize profits in the rapidly expanding and promising market.

4. Conclusion

For the past rapid economic growth process, China has attracted huge amount of FDI. Althoughthe causality between FDI inflows and real economic growth has important policy implications, ithas seldom been analyzed with respect to China. The current study examines the concerned issueusing Pesaran and Shin’s (1999) small sample cointegration test procedure allowing for differentorders of integration and the Granger causality tests. The empirical results show that FDI inflowshave not caused real economic growth in China, but the latter is revealed to have caused theformer. It implies that it would not be necessary for the Chinese government to provide varioustypes of tax or financial incentives to attract foreign investors. Even without incentives to FDI,FDI inflows are expected to continue due to rapid economic growth. Up to now, reasons other thanFDI inflows, for instance, export promotion as well as guaranteeing private property rights andsmooth transition might have been more important in explaining rapid economic growth of China.Building efficient institutions not wasting resources would be necessary to lead FDI inflows toeconomic growth.

References

Alguacil, M. T., Cuadros, A., & Orts, V. (2002). Foreign direct investment, exports and domestic performance in Mexico:A causality analysis. Economics Letters, 77, 371–376.

Basu, P., Chakraborty, C., & Reagle, D. (2003). Liberalization, FDI, and growth in developing countries: A panelcointegration approach. Economic Inquiry, 41, 510–516.

Chen, C., Chang, L., & Zhang, Y. (1995). The role of foreign direct investment in China’s Post-1978 economic development.World Development, 23, 691–703.

Cuadros, A., Orts, V., & Alguacil, M. (2004). Openness and growth: Re-examining foreign direct investment, trade andoutput linkages in Latin America. Journal of Development Studies, 40, 167–192.

Dees, S. (2001). The opening policy in China: Simulations of a macroeconometric model. Journal of Policy Modeling,23, 397–410.

Park, S. (2002). Chinese economic growth and the role of FDI. Seoul: Korea Economic Research Institute.Pesaran, M. H., & Shin, Y. (1999). An autoregressive distributed lag modeling approach to cointegration analysis. In S.

Strom (Ed.), Econometrics and economic theory in the 20th century. Cambridge: Cambridge University Press.Qin, D., Cagas, M. A., Quising, P., & He, X. (2006). How much does investment drive economic growth in China? Journal

of Policy Modeling, 28, 751–774.