Global macroeconomic interdependence: a minimum spanning tree approach

This paper measures the evolution of the interdependence level of the global economic system using a novel approach that combines network analysis and time-varying correlations. A Scalar-BEKK model is employed to determine the dynamic conditional bilateral correlations of 102 economies for the period 1952-2011. A subdominant ultrametric space is then defined for each year by applying a single-link cluster analysis to filter and highlight first-order connections in the network. I find the core of the global economic network to be dominated by advanced economies while the periphery is composed by nearly all of the developing economies and some of the emergent market economies; France occupies the central hub of the network, which coincides with recent findings from studies on European financial markets. I also find that the global macroeconomic interdependence rose over 61 percents between 1952-2011, which is in line with the majority of the empirical research on business cycle synchronization; however, this paper reveals an accelerating convergence process during the phase 1996-2011 that exceeds results from previous studies.