Keywords: head count ratio, linear mixed models, m-quantile models
In order to implement policies against poverty and inequality, policy makers all over Europe should have at their disposal information referring to appropriate domains, since the impossibility to access to goods and services varies with age, gender and zone of residence. However, the major Italian and European surveys collecting information on poverty and living conditions can be used to produce accurate indicators only at the regional level. To compute these indicators at the Provincial and Municipality levels there is the need to resort to small area methodologies. The aim of this paper is to produce and compare monetary indicators of poverty for the Provinces of Tuscany using parametric and nonparametric small area estimation methods. The target is the estimation of the "at-risk-of-poverty rate" (Head Count Ratio) and of the mean and the quantiles of the household income distribution using census data and data from the EUSILC survey. These measures can be considered as a starting point for more in depth analyses, such as the estimation of the income cumulative distribution function.