Investigating Multidimensional Child Poverty in Indonesia: A Spatial Regression Approach

Abstract

Child poverty is defined as poor children who experience poverty because of the lack of basic needs (nutrition, education, shelter, drinking water, or health services) that can affect their development or future opportunities (UNICEF 2019). The traditional approach to measure child poverty based on household income can not provide a comprehensive overview of children’s welfare. Therefore, multidimensional child poverty (MCP) that defines poor children by deprivation levels in several dimensions supporting children’s living standards is vitally important for more representative measurement to monitor child poverty. The traditional approach measuring child poverty based on household income can not provide a comprehensive overview of children’s welfare. Therefore, multidimensional child poverty (MCP) that defines poor children by deprivation levels in several dimensions supporting children’s living standards is vitally important for a more representative measurement of child poverty. The thesis is motivated due to several reasons. First, several provinces in Indonesia had a medium level of human capital, while these provinces experienced high MCP rates in 2016. Second, a limited study includes district-level interactions in Indonesia’s child poverty study. It seems plausible to assume that socioeconomic factors generating MCP in one district are correlated to or even influenced by socioeconomic factors in neighbouring districts. Therefore, this thesis investigates the linkage between MCP and its determinants by incorporating spatial analysis. Indonesia’s national socioeconomic survey in March 2020 is used to construct multidimensional poverty experienced by children. Then this thesis analyses the factors influencing MCP at regional levels employing spatial regression analysis. The results find that district-level MCP in Indonesia and its driven factors are spatially correlated. Local household characteristics, industrial structure, and infrastructure strongly determine the MCP at regional levels. The thesis shows that female-headed households and infrastructure have spillover effects towards child poverty. It suggests that reducing childhood poverty at district levels has to consider spatial effects of the local determinants of MCP, such as by investing in women’s empowerment, internet access, and educational infrastructure.