Working Paper: NBER ID: w29810
Authors: Frances R. Lu; Tom Vogl
Abstract: We study the intergenerational persistence of inequality by estimating grandmother-mother associations in the loss of a child, using pooled data from 119 Demographic and Health Surveys in 44 developing countries. Compared with compatriots of the same age, women with at least one sibling who died in childhood face 39% higher odds of having experienced at least one own-child death, or 7 percentage points at age 49. Place fixed effects reduce estimated mortality persistence by 47%; socioeconomic covariates explain far less. Within countries over time, persistence falls with aggregate child mortality, so that mortality decline disproportionately benefits high-mortality lineages.
Keywords: No keywords provided
JEL Codes: I14; I15; J62; O15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| sibling mortality (J17) | own-child mortality (J13) |
| maternal sibling death (J12) | child mortality risk by age 49 (J13) |
| maternal sibling death (J12) | child dying under 5 (J13) |
| aggregate mortality decline (J17) | persistence of child mortality across generations (D15) |
| place fixed effects (C23) | estimated mortality persistence (J17) |
| socioeconomic covariates (I14) | variation in mortality persistence (C41) |