Unveiling multidimensional poverty across Italian Provinces using small area estimation and penalized power means
利用小区域估计和惩罚幂均值指标，基于2018-2021年意大利家庭调查数据，分析新冠疫情前后各省多维贫困的变化，发现多数省份贫困加剧，南部地区尤为严重。
Abstract This paper aims to examine the impact of the Covid-19 pandemic on multidimensional poverty in Italy and its provinces by comparing household poverty levels before and after the outbreak. To capture the multidimensionality of poverty, we analyze various dimensions, including economic well-being, health status, education, neighborhood quality, and subjective well-being. The empirical analysis relies on micro-data from Istat’s aspects of daily life (AVQ) survey, covering the years 2018–2021. As the survey’s direct estimates are reliable only at the regional level (NUTS 2), we apply small area estimation techniques to produce accurate estimates of provincial (NUTS 3) deprivation incidences. Subsequently, we aggregate the deprivation headcounts across the elementary indicators using penalized power mean composite indicators. The empirical findings indicate that overall multidimensional poverty worsened in most of the Italian provinces, particularly during the second year of the pandemic, with higher levels persisting in southern areas. The various dimensions of poverty exhibited different trends, with education, subjective well-being, and health emerging as the most negatively affected in numerous provinces.