A Multidimensional poverty analysis: evidence from Italian data

Conventional poverty measures, showing that poverty and inequality have increased in Italy over the past fifteen years, are based on household income. The main drawback of this method is that it does not include other non-monetary variables relevant for defining households’ necessities. It is now widely agreed that poverty should be conceptualised as a multidimensional phenomenon, more related to the standard of living of the person or household than to the simple inability of satisfying basic subsistence needs. In this paper we propose to measure poverty in Italy by complementing income information with non-monetary indicators. To this end a multidimensional poverty analysis is performed by using a representative sample based on the first wave (2004) of the Italian component of the European Statistics on Income and Living Conditions (EU-SILC). Starting from the concept of deprivation, a non linear principal component analysis is applied to selected items in order to reveal underlying latent dimensions to be interpreted as deprivation indicators. We then examine how such measures can be combined with income measures in order to obtain a better identification of the poor. Finally we examine the overlapping between the income poor and the deprived and provide an analysis of deprivation profiles. Our results show that a more comprehensive poverty measure, combining deprivation criteria and income poverty, leads to a different identification of poor people, compared to analyses based only on income measures.