6533b826fe1ef96bd1284936

RESEARCH PRODUCT

Centres and peripheries in Finland: Granger causality tests using panel data

subject

description

Abstract Despite their importance from a policy point of view, empirical studies on the effects of growth centres in their regions are rare. This paper analyses mutual relationships between growth processes in centres and their surrounding hinterlands in 19 Finnish regions. Annual population data from the period 1970–2004 are used. A novel testing procedure based on an extension of the Granger causality definition in a panel data context is applied. Heterogeneity between regions is allowed. Both the homogeneous non-causality hypothesis and the homogeneous causality hypothesis are rejected. Causal processes prove to be heterogeneous. Causality from centres to peripheries is found for nine regions and causality from peripheries to centres for twelve regions. Rapidly growing and large centres, in particular, have negative effects on their hinterlands. Centres et banlieues en Finlande: tests de causalite de Granger faisant usage de donnees de panel RESUME En depit de leur importance sur le plan de la politiqu...