Identifying people in poverty: a multidimensional deprivation measure for the EU

In this article, I propose a multidimensional deprivation measure of poverty for the EU. The paper stands on the claim that a deprivation measure can be adequate, both conceptually and empirically, to capture poverty in the EU defined in Townsendian terms. Yet existing deprivation scales have three conceptual problems such as data-driven specification, neglected dimensionality and missing dimensions, and four data problems such as limited extent, cross-cultural equivalization, behavioral choices and reporting error. To address conceptual problems, I offer a concept-led methodology for constructing a multidimensional measure. To address data problems, I apply a post-hoc adjustment strategy using dual criteria of income poverty and financial strain. The proposed measure has four dimensions, namely needs for basic goods, health, education, leisure and social relationships, where each dimension is evaluated separately with relevant scales. When compared to the formal EU 2020 poverty target measure, the proposed measure is more likely to capture people with needs and lower resources as well as those in less affluent countries than those in more affluent countries. The (adjusted) proposed measure can be used as a stand-alone indicator to identify a target population for policy; or the unadjusted proposed measure can be combined with an income poverty measure to identify a worst-off group within that target population.