The Determinants of Multidimensional Poverty in Indonesia: A Multilevel Analysis
No Thumbnail Available
Date
2022-01-14
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Addressing poverty through a monetary approach means inadequately capturing its multidimensional aspects. This paper aims to analyze multidimensional poverty in Indonesia by adopting the Alkire-Foster method, consisting of three dimensions, namely, health, education, and standard of living. It also explores the cause of multidimensional poverty at both households and regional levels, including examining the variation between regions. The result is that about 19.98% of households are classified as multidimensionally poor; moreover, about 12.06% of them experienced overlapping deprivation in all dimensions. The eastern regions were discovered to have a higher incidence and intensity of poverty. Since the data has a hierarchically nested structure, whereas households as a level 1 unit analysis is nested in the municipalities/regencies as level 2, it can utilize the multilevel model to analyze the effects of both household and regional characteristics variables of the multidimensional poverty status. The findings are that the multilevel model presents an improvement in the fit relative in estimating multidimensional poverty rather than the standard logistic model. Multidimensional poverty varies across the municipalities or regencies even after controlling household and regional variables. The result presented that the probability of being multidimensionally poor is higher when households are in rural areas, household heads have lower education levels, the occupants work in the agriculture sector, and people have lower incomes. For regional variables, the multidimensional poverty significantly corresponds with the Human Development Index.
Description
Keywords
Multidimensional poverty, AF method, Household and regional variables