Is foreign direct investment good for growth? Evidence from sectoral analysis of China and Vietnam

Abstract

We estimate the impact of FDI on growth using sectoral data for FDI inflows to China and Vietnam. Previous empirical studies, using either cross-country growth regressions or firm-level micro-econometric analysis, fail to reach a consensus. Our paper is the first to use sectoral FDI inflow data to evaluate the sector-specific impact of FDI on growth. Our results show that, for the two developing-transition economies we examine, FDI has a statistically-significant positive effect on economic growth operating directly and through its interaction with labor. Intriguingly, we find the effects seem to be very different across economic sectors, with most of the beneficial impact concentrated in the secondary industries. Other sectors appear to see much less growth benefit from sector-specific FDI.

Keywords:

• foreign direct investment
• growth
• China
• Vietnam. JEL Classifications: F21
• F23

Acknowledgments

The authors wish to thank audiences at the ACE conference in Hong Kong and the ASSA meeting in New Orleans and two anonymous referees for many insightful comments. All remaining errors are ours.

Notes

* 10%,

** 5%, and

*** 1% level.

* 10%

** 5%

*** 1% level. In Columns 5.1 and 5.3, significance levels are for the null hypothesis that an industry's effect is the same as that of the base sector, manufacturing. In Columns 5.2 and 5.4, significance levels are for the null hypothesis that an industry's effect is not significantly differently from zero.

*10 %

**5 %

***1 % level. In columns 6.1 and 6.3, significance levels are for the null hypothesis that an industry's effect is the same as that of the base sector, manufacturing. In columns 6.2 and 6.4, significance levels are for the null hypothesis that an industry's effect is not significantly differently from zero.

1. With the availability of better data, the last few years have seen an especially large number of empirical papers devoted to this question (e.g., Alfaro et al. 2004, Bengoa and Sanchez-Robles 2003, Durham 2004, Hsiao and Shen 2003, and Li and Liu 2005).

2. For a critical look at these domestic tax/subsidy policies, see Hanson (2001) and Mooij and Ederveen (2003). Gastanaga et al. (1998) analyze other host-country policies that aim to encourage FDI inflows.

3. The reasons for this are not entirely obvious. The ‘fire-sale’ FDI that followed the large depreciations of 1997–98 was not sustained; the difficulties in the domestic financial systems of some of the target countries after 1998 and the worldwide slump of 2001–2002 might all be possible explanations.

4. Details for Vietnam can be found at http://www.vietnamlaws.com/legal_updates.aspx. There are numerous sources for China. See, for example, http://www.chinatoday.com/law/a0.htm.

5. See, for example, Tran (1997), and Do (1993).

6. Few countries received a 5. Examples for these are Zimbabwe, Turkmenistan, Sierra Leone, Libya, and North Korea. For details see: http://www.heritage.org/research/features/index/downloads.cfm#scores.

7. A related channel is the ‘creative destruction’ hypothesis raised by Aghion and Howitt (1992). If the competition from the foreign investors results in the destruction of inefficient firms, the FDI effect will turn out to be positive.

8. Lensink and Morrissey (2006) arrive at a similar negative conclusion by including FDI volatility measures in both cross-country and panel specifications.

9. See Mukand and Rodrik (2005) for insights into this problem that are relevant to the policy-applicability of estimation results.

10. For a recent survey of wage and productivity spillovers from FDI see Lipsey and Sjöholm (2005).

11. See, for example, International Labor Office (1984, pp. 71–74).

12. In Vu (2008), the accumulated stock of FDI is used instead of the inflow of FDI. For a discussion of time-varying coefficient models, see Griffiths et al. (1993, pp. 412–413), and Greene (2003, pp. 132–133).

13. Packard and Thurman (1996) suggest interaction terms for capital with several other variables: human capital, infrastructure, and research and development (R&D). Infrastructure and capital are highly correlated, and data on R&D by sector is not available for either country. Packard and Thurman (1996) also suggest an interaction term of labor with human capital, but this would cause high multicollinearity with the labor–FDI interaction term, and so we do not add this additional interaction term.

14. This is a more accurate expression of Vietnamese and Chinese educational systems than just secondary school enrollments. The secondary schools have to follow nationally drafted curricula, which focus heavily on political doctrine and abstract sciences. This leaves the vocational, technical schools or worker training schools the responsibility of providing technologically skilled labor.

15. Since exchange rates did not change much for Vietnam during 1991–2003 and for China during 1985–2004, we do not convert FDI data to local currencies.

16. The endogeneity t-test is a form of the Hausman (1978) specification test. A right-hand side variable is treated as the instrument in a first-stage regression, and the resulting error is introduced as a regressor in the second-stage regression. If the coefficient on this error term is significantly different from zero, this is taken as evidence of the existence of endogeneity.

17. We have also performed Granger causality tests as described in Geweke et al. (1983).We regress the FDILAB variable on its own lags, GDP lags, and other variables, using the FGLS estimation, and then test the significance of GDP lags. The t statistics for individual lagged GDPs and p values of the F-statistic for jointly lagged GDPs are all insignificant. We fail to reject the null hypothesis of no reverse causality.