Regional integration and decoupling in the Asia Pacific: a Bayesian Panel VAR approach

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)
3 Downloads (Pure)

Abstract

Policymakers have been debating for over a decade whether Asia is decoupling from the USA. Increasingly, deepening regional integration is cited as a possible driver of this decoupling. Using large Bayesian Panel Vector Autoregressions, estimated over different subperiods, we jointly examine bilateral macro-financial interdependencies between Asia Pacific countries and between each Asia Pacific country and the USA. We uncover no evidence of decoupling. Instead, we find that both global and regional interdependencies deepened following the Asian financial crisis, before receding after the Global financial crisis. We also show that while US shocks are important, attention should also be devoted to regional shocks which play a large role in Asia Pacific countries across all subperiods considered. Our results also suggest that there have been shifts in the relative importance of different transmission channels over time. Following the Asian financial crisis, as regional interdependencies deepened, US financial shocks began to play a larger role than US macroeconomic shocks. These results support the view that rising intra-regional trade contributed to a fall in the importance of US macroeconomic shocks. They are also consistent with research suggesting that strong, common global financial linkages increase the synchronization of Asian regional business cycles.

Original languageEnglish
Pages (from-to)773-807
Number of pages35
JournalIMF Economic Review
Volume70
Issue number4
Early online date15 Jul 2022
DOIs
Publication statusPublished - Dec 2022

Keywords

  • regional integration
  • decoupling
  • economic fluctuations
  • stochastic search variable selection
  • Bayesian Panel VAR
  • Asia Pacific region

Fingerprint

Dive into the research topics of 'Regional integration and decoupling in the Asia Pacific: a Bayesian Panel VAR approach'. Together they form a unique fingerprint.

Cite this