Reversing the depressive dynamics of the Portuguese peripheral areas : DEMOSPIN model

Portugal is divided in two major parts: i) the coastal zone, with a higher demographic density and where the main economic activities are concentrated; ii) the periphery, where a low economic performance and a depressed demographic dynamics coexist. Economy and demography interact, reinforcing each other, and many European peripheral regions are characterised by a negative circle of cumulative causation between the economic and demographic evolution. So, policies to reverse such a dynamics must address simultaneously the demographic and economic dimensions, requiring models able to simulate this interactive process. This is the main objective of DEMOSPIN project which develops a model combining an input-output growth approach with population forecasts where migrations are driven by the positive or negative employment opportunities generated by economic growth. The model has two integrated modules: an economic component and a demographic one. The first component corresponds to regional input-output matrices, one for each NUTS III regions under study; these matrices are estimated for 2007, based on the national input-output matrix and assuming three types of products: i) regionally non-tradable products, ii) regionally tradable products and iii) fully tradable products. The regional 2007 matrices are the basis for projections up to 2030, based on a set of economic scenarios which assume several levels of exogenous demand growth and hypotheses concerning sectoral productivity growth. The use of the input-output matrices allows the estimation of employment growth rates, which in turn will feed the demographic component of the integrated model, in order to estimate net migrations for each region, age-group and sex, linking the economic and the demographic components. The demographic component estimates, in a first step, population growth for each region, age group and sex, assuming zero net migrations: they are based on mortality and fertility projections. Such estimates give us values about the supply of labour force which would be expected if the employment rates were made constant. These rates are calculated from the age groups employment distribution given by census data. The comparison between this (demographic) employment with the figures provided by the economic component is the basis for migration estimates, which are the last element of demographic projections. The net migrations estimations are calculated for each region, age group and sex, using a simultaneous estimation technique. This paper will present a detailed description of the used methodology and of the obtained results concluding with the policy guidelines which such results suggest.