Dynamic Free Trade Networks: Some Numerical Results

Authors


  • The authors are grateful to Richard Baldwin and the participants of the Hitotsubashi COE Conference on International Trade and FDI 2010, Asia Pacific Trade Seminars 2011, Midwest International Group Meeting (Fall 2011), and North American Summer Meeting of the Econometric Society 2011 for helpful comments. They have also benefitted from discussions with Jin Zhang at the early stage of the research. Part of the analysis was conducted when Furusawa was a research associate of the Program on U.S.-Japan Relations at Harvard University. Furusawa gratefully acknowledges their hospitality and financial support from the Abe Fellowship. Finally, the authors are grateful to a reviewer and an associate editor for helpful comments.

Abstract

To help predict whether the proliferation of free trade agreements (FTAs) continues until global free trade is effectively attained, this paper investigates dynamic paths of FTAs, generated by numerical simulations of a model of an FTA network formation game with many countries. The characteristics of the final FTA network naturally depend on how the proposer of an FTA is chosen in each period. The paper finds that if the country that has the largest incentive to form an FTA is chosen as a proposer in each period, the network evolution always leads to a unique final FTA network, which may or may not be the complete network of FTAs. FTA networks often evolve to a partition of the world into a small number of groups of asymmetric size owing to the negative network externality caused by preference erosion.

Ancillary