Stefano Filauro (Bocconi University) - Incorporating Housing Costs, and their variation at local level, into EU Measures of Poverty: Advantages, Disadvantages, Assumptions, and Implications


The largest expenditure for most households is housing costs, including rents, mortages, and utility bills. The median housing cost in the EU in 2020 amounted to around 13% of the median income, although this ratio can peak at over 30% in some countries (e.g. Greece). Moreover, housing costs vary extensively across regions, especially in countries with secular territorial disparities. Within the same country, the median housing cost in the highest-cost region can be as high as 200% larger compared to the lowest-cost region (in the case of Spain, the Madrid area as opposed to Extremadura). Housing costs also vary meaningfully across regions in Germany, Italy and France. Despite this, housing costs are not usually taken into account into official poverty measurements. However, recent analyses have shown to what extent relative poverty changes when housing costs are somehow factored in a concept of income, whether imputing rents for outright owners or substracting housing costs for all households (Raitano 2022, Tormahleto and Sauli 2017). The cross-country evidence points to a different country ranking in terms of relative poverty as homeownership in widely diffused in Southern-European countries, and this practice improves the overall poverty rates of these countries vis-à-vis Northern EU countries. In this study we take a different approach: we look at the traditional poverty lines, such as the 60% of national median income, and we test their effectiveness against revised poverty thresholds that take into account the within-country variation in housing costs. We discuss advantages, disadvantages, assumptions, and implications of incorporating local housing costs into EU poverty thresholds. We then evaluate whether such adjustments ‘improve’ poverty measures by reducing their mismatch against material deprivation and subjective poverty rates. We use EU-SILC data for all countries reporting housing costs, income dimensions and material deprivation at regional level (NUTS1 or NUTS2). As some housing variables, such as housing quality and type of housing, have been recorded for a sufficient number of countries only since 2010, our analysis focuses on the last decade.

19 Luglio 2023, 14.00-15.00, Aula Marrama



© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma