Abstract
This study examines the patterns and determinants of multidimensional poverty and related welfare interventions in Hong Kong. These patterns unveil which dimension or combinations of dimensions contribute the most to multidimensional poverty. These results are useful in informing poverty-alleviation policies as they help to identify who should be targeted and which welfare transfer programme(s) should be enhanced. Data were drawn from the first wave (2015) of the Hong Kong Panel Survey for Poverty Alleviation (N = 1458). A latent-class analysis revealed four types of multidimensional poverty: severely deprived, housing-led poor, socially and status excluded and neighbourhood poor. A multinomial logistic regression was conducted and identified distinctive determinants of the four types of poverty. The groups that require paramount attention from policymakers are older adults over age 70, households with members with disabilities or chronic diseases and households with five or more members. The assessment of the coverage and intensity of cash and in-kind transfers showed that the coverage of cash transfers for the severely deprived and housing transfers for the housing-led poor was inadequate. The approach proposed in this study exemplifies ways to transform multidimensional poverty research into evidence-based policymaking.