Abstract
The aim of the paper is to assess the causes of spatial variations in labour productivity of Italian regions using the gravitational model of economic growth. The model is an extension of Robert Solow’s economic growth model. The model parameters are calibrated using historical data and numerical simulations of the long-run equilibrium states of the model are carried out. The scenarios considered in the paper vary in forecast investment rates, employment growth rates and urbanisation rates. Based on the results of numerical simulations, it is claimed that to achieve the full convergence in labour productivity, it is necessary to maintain higher investment rates in the south of the country than in Lombardy (by about 4%–11%), and to keep investment rates in central and northern Italy at a similar level as in Lombardy. The fall in investment has affected the poorest regions, Southern Italy, the most, followed by central Italy and the richest regions of the north of the country the least.