We use data from the 1985, 1987 and 1991 United States Vital Statistics Linked Infant Birth and Death Records to assess the relationship between state-level economic inequality and an infant's probability of death. We find that economic inequality is associated with higher neonatal mortality even after we control mother's age and race and state characteristics that are likely to be associated with both inequality and infant death. Inequality is not associated with post-neonatal mortality. We assess three mechanisms that could link income inequality and infant deaths: non-linearity in the relationship between parental income and infant death, economic segregation, and state health care spending. Our evidence suggests that non-linearity in the relationship between family income and infant health accounts for little of the relationship between inequality and infant death. However inequality is associated with greater economic segregation, which in turn is associated with a higher probability of infant death. This effect is partially offset by the fact that inequality is also associated with state spending on health care, which is in turn associated with lower death rates. The increase in economic segregation increased infant deaths more than the increase in health care spending reduces them, so the net effect of economic inequality is to increase infant deaths especially in the first month after birth.