Working Paper: CEPR ID: DP15955
Authors: Valrie Lechene; Krishna Pendakur; Alexander Wolf
Abstract: We develop a method to estimate resource shares — the fraction of total householdexpenditure allocated to each household member — using OLS estimation of Engelcurves.The method reframes the nonlinear model of Dunbar, Lewbel and Pendakur (2013),extended to allow single parent households, scale economies in assignable goods, complementaritiesbetween non assignable goods, and supplemented by a linear identificationtest.We apply the model to data from 12 countries, and investigate resource shares,gender gaps, and poverty. We reject equal sharing, and find large gender gaps inresource shares, and consequently in poverty rates, in some countries.
Keywords: Collective households; Resource shares; Intrahousehold inequality; Poverty
JEL Codes: D13; D63; I32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Resource shares identified by OLS estimation (C51) | Gender gaps in expenditure within households (D13) |
| Gender gaps in expenditure within households (D13) | Women have lower resource shares than men (J16) |
| Women have lower resource shares than men (J16) | Higher poverty rates among women (I32) |
| Households have sufficient resources (D10) | Women may still experience poverty (F63) |
| Gender gaps in expenditure within households (D13) | Challenges standard World Bank poverty measurement strategy (I32) |
| Recognizing within-household inequality (D13) | Accurate assessment of poverty and resource allocation (I32) |