|Abstract: ||In the United States of America, black mothers are at a much higher risk of giving birth to a baby that is of low birthweight (LBW), when compared to white mothers. This fact is true even when comparing black and white mothers within similar educational, and income levels (Lhila & Long, 2012).
The weight of a newborn can be a key determinant to the infant’s lifetime income and human capital attainment. The adverse health effects of being born LBW can be detrimental to the child’s future human capital potential (Baguet & Dumas, 2019). If a child’s health is affected adversely by being born with LBW, then it can affect their educational attainment, and consequently, their human capital potential (Almond & Currie, 2011). Those who experience a health insult at a young age may be unable to work or pursue education due to poorer health, their lifetime income will be affected, through both the loss in potential earnings and the direct costs of health care services (Petrou, 2003).
A natural question that arises due to this possibility is: What causes LBW? Studies have found that maternal stress during pregnancy increases the probability of having a child with LBW (Nkansah-Amankara et al. 2010). A well-known cause of stress is the risk of violence. Pregnant women who have been exposed to terrorist attacks and homicides experience stress and fear that can negatively affect the birthweight of their in-utero child (Camacho, 2008; Kane, 2011; Mansour & Rees, 2012; Koppensteiner & Manacorda, 2016; Brown, 2018).
Moving past just the lifetime of a child born with LBW, there is also evidence that people with LBW are at a higher risk to have children of their own who are of LBW and may have poorer health as a result (Currie & Moretti, 2007; Kreiner & Sievsersten, 2020). This intergenerational result implies that a group that experiences many LBWs will have their educational attainment severely curbed, which truncates potential earnings as a result. This relationship serves as the motivation for this paper. If there exists an intergenerational transmission of low birthweights and there is an impact of birthweight on education and earnings, then a community who is suffering from a high proportion of babies born with LBWs would see its social mobility affected. In a context where black mothers have a much higher chance of giving birth to LBW than white mothers, racial disparity in birthweights between white and black mothers could be seriously affecting the black community’s human capital potential.
To investigate this possibility, we utilize 1989-1991 Linked Birth/Infant Death Data from the U.S. Department of Health and Human Services, National Center of Health Statistics and 1988-1991 Uniform Crime Reporting (UCR) Program Data, Supplementary Homicide Reports from the U.S. Department of Justice Federal Bureau of Investigation. Our measure of violence in this case is homicides that occurred in counties during a woman’s pregnancy.
By regressing homicide intensity in each trimester of pregnancy on birthweight, we find that an increase in homicide intensity during a pregnancy’s first trimester is associated with an increase in birthweight, while an increase in homicide intensity during the second trimester is associated with a decrease in birthweight. The positive effect in the first trimester may be due to exposed mothers believing that the first trimester is essential in their child’s health and subsequently, seeking out additional, unobserved, protective measures to remedy the adverse effects. Our results in the second trimester are stronger for black mothers than white mothers, suggesting that black mothers may be more stressed about homicides than white mothers. We also obtain that this effect only holds among mothers who are married, presenting the possibility that black mothers may be more stressed that their spouse will fall victims to a homicide. This points to future research into how homicides affect women from different neighbourhoods within counties.
This paper proceeds as follows: Section 2 is a review of previous literature, Section 3 presents our data, Section 4 lays out our model specifications, Section 5 presents our results, and Section 6 concludes.|