京都大学 大学院経済学研究科・経済学部

タイトル: ”Decomposing Inclusive Growth: Application to Household Survey Data”
 

Abstract: This paper develops a decomposition framework for the upward mobility index of Ray and Genicot (2023) and applies it to repeated cross-sectional household survey data from Vietnam over 2002–2020. Building on the log-sum-exp structure of the index, we derive an exact subgroup decomposition that separates within-group inclusive growth from changes in population shares. We also conduct reweighting decomposition in the spirit of DiNardo et al. (1996) and a recentered influence function (RIF) regression approach, enabling both aggregate and detailed decompositions without requiring long household panels. Applying this framework, we find that rural households and ethnic minorities, which were initially left behind, experienced stronger inclusive growth over the period and contributed disproportionately to the rise in the national index. Aggregate decomposition results show that changes in observed household characteristics explain only a modest share of overall inclusive growth, while the structure effect and regional economic conditions account for the bulk of the improvement. Detailed decomposition further reveals that educational upgrading contributes little to inclusive growth in the composition effect, partly because the poorest households are disproportionately older and low-educated and thus less directly connected to the gains from schooling expansion and labor-market transformation.