Global Outsourcing and Wage Inequality in Middle-Income Countries: Evidence from South Korea

Abstract

A substantial number of studies have suggested that global outsourcing can induce wage inequality. As Feenstra and Hanson [(1996a) Foreign investment, outsourcing, and relative wage, in: R. C. Feestra, G. M. Hanson, and D. A. Irwin (Eds.) Political Economy of Trade Policy: Essays in Honor of Jagdish Bhagwati (Cambridge: The MIT Press), pp. 89–127] argued, global outsourcing is comparable to skill-biased technological change in that global outsourcing is more likely to increase the wage of skilled workers rather than their unskilled counterparts. We examine the effects of outsourcing on wage of skilled and unskilled workers in Korea's manufacturing sector with a focus on the dissimilar effects of outsourcing to developed countries (DCs) and less developed countries (LDCs) on relative wage. The results of system and difference GMM estimation based on manufacturing data from 1992 to 2006 indicate that outsourcing to DCs and LDCs have opposite (and significant) effects on relative wage, that is, outsourcing to DCs (LDCs) decreases the wage of skilled (unskilled) workers.

Key Words:

• Global outsourcing
• wage inequality
• middle income country
• system GMM

Jel Classification:

• F16
• F23

Acknowledgements

We would like to thank seminar participants at the Singapore Economic Review Conference (SERC) in Singapore in 2011. All responsibility for any errors remains our own. This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (NRF-2011-332-B00064).

Notes

1. Precisely, .

2. in Equation (8) is to be substituted for the unit-cost function in Equation (2).

3. Neutral technology change or capital flow is regarded as major driving force behind global outsourcing. See Feenstra and Hanson (1996a) and Chongvilivan et al. (2009).

4. This is obvious because,

5. If it is not easy to grasp that capital flows from the Middle to North, we can think of technological advances in north countries, which shift downward.

6. For the proof, see Feenstra and Hanson (1996a).

7. The subscript t (time) is omitted for simplicity.

8. The list of industries is summarized in the appendix.

9. Since Korean Manufacturing Census has stopped providing wage of skilled and unskilled workers since 2007, we could not obtain recent data on the wage share of skilled labor. In addition, the industrial classification system changed in 1992. Thus, we only considered the 1992-2006 period for the analysis.

10. In addition, both the first and second approaches have crucial disadvantages. Regarding the first approach, Feenstra and Hanson (1996a) and Krugman (1994) argued that MNCs’ transactions cannot explain a substantial portion of outsourcing because they leave much room for underestimating actual outsourcing activity. The second approach, which uses the international IO table, is considered the best method for measuring global outsourcing because it can accurately capture the flow of imported inputs. However, the international IO table is not published every year and does not have worldwide coverage.

11. The list of countries is available in the appendix.

12. By referring to Feenstra and Hanson (1996a), the dependent variable can be written as,

13. We did not covert KRW into USD because R&D intensity is a ratio as well.

14. If skilled wage share is regressed on employment share, simultaneous correlation occurs, which leads to biased results.