Abstract
Previous studies show that children in lower socioeconomic status families reveal higher rates of mortality. We complement the income-mortality literature by establishing a causal link between income and child mortality. Our instrument for income is based on time-series global shocks to oil prices combined with the cross-sectional share of employment in manufacturing across US states as their exposure to oil price changes. Using the universe of death records between the years 1975-2004, we find the OLS results of income-child-mortality relationships are under-biased. The 2SLS-IV results suggest that a $1,000 increase in income per capita at the state level reduces child mortality and infant mortality by 0.87 and 0.53 fewer incidences per 1,000 population of age-specific children.