Year: 2026
Type: preprint
Abstract: Conventional distributional analysis evaluates resources among individuals who are alive, even though survival is strongly socially graded. This paper shows that differential mortality affects distributional analysis through two distinct channels. Unequal survival prospects widen inequality in the lifetime opportunities attached to current resources, while selective mortality subsequently changes the population over which inequality is measured. I develop an age-standardized lifetime equivalent-income measure that combines material resources with predicted remaining longevity, derive exact identities for the effect of mortality selection on fixed baseline distributions, and characterize how lifetime-welfare inequality aversion changes marginal social welfare weights on annual resource flows. Using the U.S.\National Health Interview Survey linked to mortality records, I find that the Gini coefficient rises from 0.4043 for equivalized family income to 0.4306 for lifetime equivalent income, a 6.51 percent increase in relative inequality (95 percent CI: 6.10--6.92). The preferred mortality model closely reproduces the observed D10--D1 survival gradient: at a 20-year horizon the observed gap is 13.28 percentage points and the predicted gap is 12.52 points. In an age-40 landmark exercise that holds baseline resources fixed, the original bottom income decile accounts for only 7.12 percent of predicted survivors by age 85, compared with 12.86 percent for the original top decile. Finally, the longevity correction to marginal social welfare weights has an exact normative threshold: it is neutral at lifetime-welfare inequality aversion \(\eta=1\) and raises the relative priority of shorter-lived individuals for \(\eta>1\). At the illustrative calibration \(c=1,\eta=2\), incorporating differential longevity raises the D1-to-D10 marginal social welfare weight ratio by 15.14 percent (95 percent CI: 14.47--15.82). Socially graded mortality therefore affects lifetime inequality, redistributive priority, and the composition of the population on which later-life distributional statistics are computed.
Source: Research Square
Author: Felix Dopchie
Institution: Tilburg University
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7233964992
