Poverty dynamics in Indonesia: empirical evidence from three main approaches

Abstract

Although the number of studies on poverty dynamics in Indonesia is growing, the findings remain inconclusive. This study aims to reexamine the poverty dynamics in Indonesia by using three main approaches: the spell approach, the component approach, and the equally distributed equivalent (EDE) poverty gap approach. The study employed household panel data from the National Socio-Economic Survey of Indonesia (Susenas) 2008 and 2010 and also the FGT index. The analysis shows that the spell approach tends to underestimate chronic poverty (households remain poor in two consecutive periods of observations). It estimates that 6.7% of the total households experienced chronic poverty. Meanwhile, the conclusion of poverty dynamics under the component approach is sensitive to the choice of poverty parameter of the FGT index. By employing the poverty parameter of α = 0, around 11.07% of total households were classified as chronically poor (average expenditure below the poverty lines). However, applying the poverty gap (α = 1) and poverty severity (α = 2) indicates that the chronic component reached 63.16% and 54.15% of the total poverty, respectively. Likewise, the EDE poverty gap approach also suggests that poverty in Indonesia is mainly chronic at 92% of the total component of poverty. The high percentage of chronic poverty is contributed by the significant cost of inequality.