One the one hand the global economy seems more integrated than ever, but on the other, it is claimed that the BRICs are growing right through rich country cycles, so what is up with the de-coupling hypothesis?
A new paper by Kose Otruk & Prasad addresses this question empirically using a large dataset of over 100 countries from 1960-2005. They divide the countries into Industrial, Emerging, & Other and use Bayesian methods (Gibbs sampling with data augmentation) to estimate a dynamic factor model of what shocks drive cycles in these countries. When comparing the 1960-84 period with 1985-2005, they find that the global factor has declined in importance in all three groups, while the group factor has become more important in the Industrial and Emerging groups.
So globally, decoupling but within two of the three regions, increased syncronization while in the dreaded "Other" group (developing but not emerging!) idiosyncratic factors have become more important.
Very nice paper, but it seems to me that Tolstoy should get a shout out in the acknowledgements! After all, he said it first.