Private capital flows to developing countries: the role of the domestic financial sector

Abstract

The relationship between private capital flows and growth has been examined extensively in the literature, yet numerous controversies remain. This study examines the relationships among private capital flows (foreign direct investment and portfolio investment), financial development and economic performance in a panel of developing countries over the period 1983–2006, by employing generalized method of moments (GMM) panel data analysis. We find that these private capital flows have a positive impact on growth with a well-developed financial sector but have a negative effect in the presence of poor financial sector development. Moreover, foreign direct investment promotes economic growth via efficiency effect, while portfolio investment stimulates economic growth via investment effect. Hence, well-developed financial sectors are ones that are crucial for economic growth.

Keywords:

• foreign direct investment
• portfolio investment
• financial development
• eco-nomic growth
• panel data analysis

JEL classifications:

• F21
• F34
• O57

Acknowledgements

The authors are truly thankful for the insightful comments and suggestions by the anonymous referee, which resulted in a much-improved version of the paper. We would also like to thank the Editor for useful suggestions. All remaining errors are ours. This paper is a product of a research sponsored by the Fundamental Research Grant Scheme (FRGS) 2007, Ministry of Higher Education, Malaysia.

Notes

1. In this regard, Levine (1997), Andersen and Tarp (2003) and Wachtel (2003) have provided comprehensive surveys on the relationship between financial development and economic growth.

2. Wang and Blomstrom (1992) also found a significant positive relationship between the degree of spillovers from FDI and the size of the technology gap between domestic and foreign firms.

3. The selection of the macroeconomic variables is further discussed extensively in Section 4.

4. This method is fully described in Arellano and Bond (1991), Arellano and Bover (1995) and Blundell and Bond (1998).

5. This is not a surprise because the model is using annual data, and business-cycle effects may propagate for more than one year.

6. For a more complete description of GMM rules and instruments, see Arellano and Bond (1991) and Arellano and Bover (1995).

7. Arellano and Bond (1991) have called this test statistic as m2 test. For the test statistic, if the residuals ϵ_it were first-order correlated, then y_{i , t – 2} would be correlated with Δϵ_it , and it could not be used as an instrument. The same is true with any variable from X_it that is temporarily correlated with ϵ_it .

8. Demetriades and Luintel (1996), for example, have revealed that the measure is less appropriate as financial deepening in Nepal because the monetization of transactions can increase without evolution of financial development. In their argument, this is particularly true in Nepal where the non-monetized sector still plays a crucial role in promoting economic performance.

9. The indicators, however, do not clearly reflect the proportion of credits allocated to both the private and public sectors.

10. The size of the government and inflation rate serve as indicators of macroeconomic stability (see, for example, Easterly and Rebelo 1993, Fischer 1993).

11. See Newey and McFadden (1994, p. 2231) for details on this test.

12. Bekaert et al. (2001) have also found a positive relationship between economic growth and a dummy variable for the years in which domestic equity markets are open to foreign investors.