Household equivalence scales, poverty and inequality in Indonesia : three essays

Abstract

This study addresses the issue of differences in resource requirements for adults and children in analyzing poverty and income inequality. Sensitivity of poverty and inequality measures for two different equivalence scales are discussed using Indonesian socio-economic survey. Essay one, "Estimation of Equivalence Scales for Indonesia," presents estimates for the two most widely used scales - Engel and Rothbarth - by urban/rural residence and sex of children. We found that scale values are higher in rural areas implying that rural parents have to forego a larger proportion of their resources for nurturing a child compared to their urban counterparts. But the absolute cost is larger in urban areas because of higher total household expenditures. The estimates also provide evidence of economies of scale in consumption. Like previous studies, we also found a large difference between the Engel and Rothbarth scales. Essay two, "Aspects of Poverty in Indonesia: A Decomposition Analysis," analyses the incidence of poverty and contribution to total poverty from different demographic and socio-economic groups using a decomposable poverty index. Three different poverty thresholds - per-capita, Engel scale-adjusted, and Rothbarth scale-adjusted - were used to take into account the effect of family size in poverty. We found that poverty measurement is sensitive to the choice of scaling parameter. The per-capita threshold and the Engel scale-adjusted threshold provide similar results whereas the Rothbarth scale-adjusted threshold is in conformity with other two thresholds only when the decomposition is based on the education of the household head. The third and final essay, "Aspects of Inequality in Indonesia: A Decomposition Analysis," endeavors to assess the impact of 'within-group' and 'between-group' inequalities on inequality in total household expenditure using two different decomposable inequality indices. The main conclusion is that the 'between-group' inequality accounts for little of the inequality observed in Indonesia with most inequality stemming from 'within-group' differences. As in the poverty exercise, the use of Rothbarth scale gives different results.