This study tests whether Ricardian comparative advantage is valid for value added in exports that does not include intermediate inputs imported from various industries in a number of countries. Using a panel data on valued added contents of bilateral exports, we find that changes in the labor productivity lead to growth of value added in exports. This implies that Ricardian comparative advantage is an important determinant of exports in longitudinal changes. The estimated coefficients of the observed productivity turn out to be larger than those of CDK (2012), implying that Ricardian comparative advantage has greater influence on determining the patterns of trade in a world with global value chains.
  This study also investigates the role of offshoring and transaction costs in comparative advantage. We use data on value added in exports, offshoring in materials and services, and transaction costs at the country and industry-level for the period 1995-2009 calculated from World Input-Output Table which covers 40 countries and 14 manufacturing industries. Employing a system GMM estimator to alleviate the potential endogeneity problem, we find that services offshoring has positive effects on comparative advantage while material offshoring affects it negatively. We also find that transaction costs have a negative effect on comparative advantage. Moreover, it turned out that there is a magnification effect of transaction costs on the induced value added in exports.
 

Executive Summary

1. Introduction

2. Construction of Data Sets

2-1. Deriving Net Value Added in Exports
2-2. Offshoring
2-3. Transaction Costs

3. Value Added in Exports and Ricardian Comparative Advantage

3-1. Empirical Model and Data
3-2. Empirical Results
3-3. Robustness Check

4. Offshoring and Transaction Costs as a Source of Comparative Advantage

4-1. Empirical Model and Data
4-2. Estimation Results

5. Conclusion

Appendix

References