Measurement and determinants of multidimensional poverty: the case of Hong Kong

Multidimensional poverty in urban cities has received growing attention. In this study, the Alkire-Foster method was applied to examine multidimensional poverty in Hong Kong using a random sample of 1,476 Hong Kong adults in a cross-sectional population study undertaken in 2016-2017. Multidimensional poverty was measured on seven dimensions: education, health, mental health, housing, employment, social network, and subjective poverty. The adjusted headcount ratio gradually decreased, and the poverty intensity climbed while the cut-off value of multidimensional poverty increased. Education, employment and subjective poverty made the greatest contribution to multidimensional poverty. Logistic regressions were used to investigate the groups at risk of multidimensional poverty with different deprivation thresholds. Study results show that being female, single, older, a low income and deprived of necessities were risk factors of multidimensional poverty with various poverty cut-offs. Low income and deprivation of necessities correlated but did not overlap with multidimensional poverty. Our findings imply that poverty measurement should be multidimensional, and anti-poverty policy should be more targeted and diversified according to different groups' risks of multidimensional poverty. The implications of these results and limitations are discussed.