The socio-demographic and geographical dimensions of fuel poverty in Italy

Fuel poverty is a multi-scale and a multidisciplinary topic, as it relates the demand-side to the supply-side, the "household-scale" to the "country-scale", and it involves social and economic boundary conditions. This paper focuses on the Italian case and contributes to the present-day discussion by introducing a novel measure of fuel poverty that considers details on household energy expenditure and its impacts on aggregated regional and national levels. The proposed measure of fuel poverty couples the "household-scale" to the "country-scale" and is based on an integrated statistical-lumped parameter approach. This approach compares the theoretical household thermal requirements for heating purposes-obtained by a lumped parameter model-with the real thermal energy expenditure-obtained by a national-representative survey. Subsequently, the proposed measure of fuel poverty is related to the socio-demographic and geographic variables by using by a statistical approach, structured in four techniques: ordinary least squares method, (b) variance inflation factor, (c) least absolute shrinkage and selection operator and (d) classification and regression tree approach. It is found that fuel poverty, in the Italian case, is mainly related to the geographical dimension, rather than being related to the sociodemographic dimension. This results is of practical importance and provide a rational basis for policy makers when planning strategies to tackle the incidence of fuel poverty in Italy.