Negative correlations between income and behaviors such as smoking, drinking, and drug use suggest that these habits may be a possible mechanism for transmission of health to infants. The decision to drink or smoke may be related to other maternal behaviors or characteristics that could affect infant health; therefore, an extensive set of control variables or a natural experiment that changes smoking behavior independent of maternal characteristics is necessary to isolate the impact of these behaviors’ outcomes such as birth weight and infant mortality. Numerous studies have linked maternal drinking and smoking with reduced infant health and long-term human capital outcomes.
In a survey of Danish mothers who had recently given birth, women who reported drinking four or more drinks per week while pregnant were more likely to have a preterm delivery than women who reported drinking no alcohol. This finding may be a result of omitted variable if women who choose to drink during pregnancy are negatively selected on other attributes. Accordingly, there has been a shift to the use of quasiexperimental approaches to unraveling the alcohol and child outcome relation.
Variation in the legal drinking age across states and over time has been used to identify the causal effect of maternal drinking on infant health. A lower drinking age is associated with more alcohol consumption during pregnancy, an increase in premature births, and an increase in the probability of low birth weight. The reduction in health at birth can partially be attributed to changes in the composition of births, increasing the number of births without a father listed and suggesting that more unplanned pregnancies occur when drinking laws are less stringent.
Maternal alcohol consumption can have long-term effects on human capital development, as demonstrated by a policy experiment in Sweden. In 1967, grocery stores in certain regions were temporarily allowed to sell strong beer that was previously only available in government-run liquor stores. Children exposed the longest to the policy while in utero had lower completed education, lower earnings, and higher rates of welfare participation than children that were not exposed to the policy experiment.
Smoking during pregnancy increases health risk for both the mother and infant in the form of complications such as miscarriage, membrane ruptures, ectopic pregnancy, pneumonia, and stillbirth. Women who smoke during pregnancy have lower birth weight babies on average and are at a greater risk for having an infant classified as low birth weight. The seminal study of the impact of smoking on infant health is the randomized-controlled trial of Sexton and Hebel (1984), in which pregnant smokers were randomized into a treatment group receiving assistance quitting smoking and a control group receiving no intervention. Babies whose mothers were in the treatment group were on average 92 g heavier than control group babies.
The 1964 Surgeon General Report on Smoking and Health alerted the nation to the health hazards of smoking resulting in a reduction in smoking among pregnant women that was concentrated among higher-educated mothers. A study comparing birth outcomes of children before and after the release of the Surgeon General Report reveals that higher smoking rates are associated with lower birth weight. However, no effect of smoking was found on gestation, prematurity, or the likelihood of having a low birth weight baby. These results are similar to studies that use increases in cigarette excise taxes to estimate the impact of smoking on birth weight.