Poverty is strongly associated with worse health across countries and within countries across individuals. However, not all poor individuals suffer from poor health: the effects of poverty on health vary across place and time. In this review, we discuss the evidence documenting these patterns, and the reasons for the associations. We then provide an overview of what is known about policies that may improve the health of the poor. We focus primarily on the modern-day United States, but also discuss evidence from historical experiences and low- and middle-income countries. Throughout we discuss areas in need of future research.
The Supplemental Security Income (SSI) program uses a birthweight cutoff at 1200 grams to determine eligibility. Using birth certificates linked to administrative records, we find low-income families of infants born just below the cutoff receive higher monthly cash benefits (equal to 27\% of family income) at ages 0-2 and small but significant effects on transfers through age 10. Yet, we detect no improvements in health care use and mortality in infancy, nor health and human capital outcomes as observed through young adulthood for these infants. We also find no improvements for their older siblings.
We use linked administrative data that combines the universe of California birth records, hospitalizations, and death records with parental income from Internal Revenue Service tax records to provide novel evidence on economic inequality in infant and maternal health. We find that birth outcomes vary non-monotonically with parental income, and that children of parents in the top ventile of the income distribution have higher rates of low birth weight and preterm birth than those in the bottom ventile. However, unlike birth outcomes, infant mortality varies monotonically with income, and infants of parents in the top ventile of the income distribution—who have the worst birth outcomes—have a death rate that is half that of infants of parents in the bottom ventile. When studying maternal health, we find that although mothers in the top and bottom income ventiles have similar rates of severe maternal morbidity, the former group are three times less likely to die than the latter. At the same time, these disparities by parental income are small when compared to racial disparities, and we observe virtually no convergence in health outcomes across racial and ethnic groups as income rises. Indeed, infant and maternal health in Black families at the top of the income distribution is markedly worse than that of white families at the bottom of the income distribution. Lastly, we benchmark the health gradients in California to those in Sweden, finding that infant and maternal health is worse in California than in Sweden for most outcomes throughout the entire income distribution.
Undocumented immigrants are ineligible for public insurance for prenatal care in most states, despite their children representing a large fraction of births and having U.S. citizenship. In this paper, we examine the short- and long-term effects of a policy that expanded Medicaid pregnancy coverage to undocumented immigrants using a novel dataset that links California birth records to Census surveys and administrative records on mortality, earnings, educational attainment, and public program participation. Using these records, we identify siblings born to immigrant mothers before and after the policy and implement a mothers' fixed effects design to estimate policy impacts. We find that the policy increased coverage for and use of prenatal care among pregnant immigrant women, and increased average gestation length and birth weight among their children. Later in life, these children experience better educational outcomes, are less likely to have children at young ages, and receive fewer public supports. The effects are robust to a variety of specification checks and are larger among women who are predicted to have undocumented status at the time of the policy change and their children. Placebo tests estimate effects among immigrants and later cohorts not affected by the policy change and find no effects, as expected. Calculations based on our estimates indicate that, over the long-run, the government more than recoups the cost of its initial investment of providing Medicaid coverage to these families.