Measurement and Determinants of Multidimensional Poverty: Evidence from Taiwan

Poverty is multifaceted. In this study, multiple data sources and analytical approaches were used to analyze multidimensional poverty profiles in Taiwan. The Alkire–Foster method and a multilevel modeling technique were applied to examine multidimensional poverty and multilevel factors, respectively. The resulting estimates of multidimensional poverty indicated disparate multidimensional poverty profiles for the four districts of Taiwan. This study’s findings on the multilevel determinants of multidimensional poverty demonstrate that the levels of multiple dimensions of deprivation vary across cities and counties according to microlevel factors including age, socioeconomic status, marital status, household income, and household size. At the macrolevel, level of urbanization and service-to-manufacturing ratio were found to significantly correlate with level of multidimensional poverty. Finally, the study presents the implications of these findings for antipoverty policy and provides suggestions for future research.