An Estimate of the Degree of Interconnectedness between European Regions: A Bayesian Model Averaging Approach

This paper provides a methodology based on General Variance Decomposition and Bayesian Model Averaging to estimate the degree of economic interconnectedness across different regions, and applies such methodology to a sample of 199 European NUTS2 regions in the period 1980-2008. The estimated connectedness appears very heterogeneous and not symmetric. The idiosyncratic component is not very significant, as well as the common component. A clear pattern of core-periphery exists but not defined in geographical terms. The country component is not very significant, very heterogeneous across countries, and proportional to countries' size. The degree of interconnectedness positively depends on the time horizon of the analysis. Finally, the comparison of the estimated connectedness matrix with two spatial matrices generally used in spatial econometrics (a firstorder contiguity and a distance-based matrix) reveals that both are far from representing the actual interconnectedness between European regions.