Regional productivity network in the EU

We develop a unified stochastic frontier model which controls for the local spatial correlation and the global factor dependence as well as parameter heterogeneity, simultaneously. We then propose the regional productivity network analysis to examine the diffusion impacts of the capital intensity on the labour productivity in the EU. We apply the proposed approach to the dataset consisting of 202 regions in the EU15 countries over 1980-2019, and convincingly unveil that the technological shock diffuses from efficient regions operating on or near the frontier to inefficient regions. This suggests that policies to enhance domestic absorption capacity appear better suited to net receivers of technological shocks whilst policies to attract more R&D investments are appropriate to their transmitters. In this regard we stress the importance of investing European funds in peripheral regions to address regional inequality and polarisation.