Answers are grounded in this paper and cite their sources.
We examine how household characteristics relate to unconditional quantiles of per-capita expenditure and how their contributions to explained variation differ between urban and rural Bangladesh. Using the 2022 Household Income and Expenditure Survey, we combine recentered influence function-based unconditional quantile regressions with a Shapley decomposition of explanatory power. The results show that household size contributes most to explained variation in rural locations, particularly at the lower quantile, whereas schooling dominates in urban locations and becomes increasingly important towards the upper tail. These findings reveal substantial distributional and urban-rural heterogeneity in the correlates of household expenditure. These heterogeneous expenditure-related associations may inform context-sensitive policy discussions.
No takes yet. Share an insight, caveat, or question.
Kataoka et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: