An explanatory study of bilateral FDI relations: The case of China

Abstract

A bilateral FDI relationship should include two dimensions, size and intensity. This study explores the factors that contribute to an explanation of both the size and the intensity of bilateral FDI relations, using the case of China. The results suggest that the countries, which are geographically, politically, ethnically, and economically close to China, have a firm intention of investing in China and enjoy a high FDI intensity. The countries with a large market, high productivity and an advanced structure are capable of investing more in China, and they enjoy a higher FDI value than other countries. The two factors, the overseas Chinese and the distance from China, are significant influences on both the size and the intensity of the FDI relationship.

Keywords:

• China
• FDI
• FDI intensity
• gravity model
• Jel Classifications: F21

Acknowledgments

The author is grateful to the Editor, two anonymous referees, Arjen van Witteloostuijn, Robert Lensink and Angela Ashworth for their valuable comments.

Notes

 A similar index was used by Dunning (1997)

 In the 45-country sample of this study, 14 countries levy an export tariff. Among the 14 only two developed countries levy an export tariff, the share of the export tariff is 0.006% and 0.0002%, in contrast to the average of 1% of the other 12 developing countries.

 The paper use the corrected trade intensity with formulas as following (see e.g. Braga et al., 1994; Kunimoto, 1977): exTI_cf  = ((x_cf /m_f )/(x_c /(m_w − m_c ))), imTI_cf  = ((m_cf /x_f )/(m_c /(x_w − _xc ))), where x _cf is Chinese exports to country f, m _f is total imports of country f, x _c is Chinese total exports and x _w is world exports. m _cf is Chinese imports from country f, x _f is total exports of country f, m _c is Chinese total imports and m _w is world imports.

 I calculated the Pearson correlation coefficients between the explanatory variables in the two models. For the intensity model, all the correlation coefficients are lower than 0.7. But for the volume model, the three variables are higher than 0.7.

 The data of the countries that invest less than $10,000 are not available.

 Although Hong Kong and Taiwan are the most important FDI sources in China, the paper does not include them in the estimation. The reason is twofold, (1) Hong Kong and Taiwan are special parts of China; (2) the FDI and trade data of Hong Kong is misleading because of re-export and re-investment.

 The variables with absolute t-statistics smaller than 0.5 are first removed from model 2. Then the variables that are not significant in model 2 are removed from model 3.

 Tariff exemptions for foreign-invested enterprise (FIE) imports had been abruptly discarded for 20 months from April 1996 to January 1998. This change frustrated the FIEs' import of machinery, equipment, parts and other materials.