The Future of European Regional Inequalities: Box-Cox Transformed Arma Process Trend Smoothing (Bats) Forecasting

Abstract

The vast majority of the empirical studies on regional economic inequalities has analyzed the past evolutions, while the future trajectories are often ignored. Despite, no methods exist to predict the future precisely, it is worthwhile to shed light on the prospective tendencies in order to plan and formulate the policies at the present time. The current study addresses the following questions; Will regional convergence continue in Europe? Which regions will become more prosperous? What are the future determinants of regional growth? Our dataset covers 236 NUTS-2 regions belonging to the 28 European Countries for the period 2000-2022. In terms of methodology, we use a nonlinear forecasting technique BATS ("Box-Cox Transformation, ARMA errors, Trend and Seasonal Components") model and Spatial Durbin Regressions along with explorative maps and descriptive statistics. As an outcome of the analyses, we obtained several remarkable results. First, regional inequalities are expected to widen by 2050 indicating the evidence of regional divergence. Second, spatial poles of prosperity are likely to change substantially. Most of the regions belonging to the countries in the "Mediterranean Basin" are predicted to remain relatively backward while many Eastern European regions are expected to rise in prosperity. Northern and Central European regions are likely to keep their prosperous position. Third, several crucial determinants of future growth patterns are detected. It appeared that younger demographic profile, industrialization and cohesion policies (particularly for CEE regions) have become key factors of future growth performance.