Unemployment Insurance, Birth Rates, and Infant Health


This figure is a bar graph titled, Birth Outcomes and UI Benefits. It is subtitled, Effect of moving from o to 100 percent income replacement for UI with unemployment rate at 6.01 percent. The y access represents a percent change and ranges from negative 6 to positive 2 percent. The x-axis has two categories: Low birth weight (less than 25000 grams) and Preterm births (less than 37 weeks). Each category has 3 corresponding bars: non-Hispanic White, non-Hispanic Black, and Hispanic. For low birth weight, the values are – 1.1 percent, negative 6 percent, and positive 2.4 percent for White, Black and Hispanic respectively. Only the value for Black is statistically significant. For preterm births, the values are negative 3.2 percent, negative 6.4 percent, and negative 3 percent for White, Black and Hispanic respectively. The values for White and Black are statistically significant.  The note on the figure reads, percent changes are calculated using the means for each subsample. Results with asterisks are statistically significant at 99% confidence intervals. The source line reads, Source: Researcher’s calculations using data from Vital statistics, SEER, BLS LAU, Zillow, Michigan SRC, and SCF.

Falling birth rates in the US and other advanced economies have raised questions about the links between economic conditions, government safety nets, and fertility and infant health. In The Cyclicality of Births and Babies’ Health, Revisited: Evidence from Unemployment Insurance (NBER Working Paper 30937), Lisa Dettling and Melissa Kearney find that fertility rates and infant health are influenced by the mother’s financial circumstances during economic downturns. Unemployment insurance (UI), which provides liquidity following job loss, can mitigate financial stress and improve birth outcomes.

The researchers use birth certificate data from the National Center for Health Statistics, which contains all births from 2000–19. The dataset includes information on the mother’s race and ethnicity, age, marital status, infant birth weight, length of gestation, and maternal health behaviors such as prenatal care use and smoking. Their analysis groups the data by county, month of conception, race/ethnicity, and age group (18–34 and 35–49). To construct fertility rates, the researchers divide the births by the female population in each group according to Centers for Disease Control and Prevention Surveillance, Epidemiology, and End Results Program data. The resulting dataset on fertility rates is linked with unemployment rates from the Bureau of Labor Statistics Local Area Unemployment Statistics, data on the proportion of families who are liquidity constrained from the Survey of Consumer Finances, and other measures of local economic conditions.  

A higher replacement rate for UI benefits attenuates the negative effect of unemployment on fertility and improves infant health, especially for groups with above-average rates of low birth weight and preterm births.

The researchers calculate the median potential UI replacement rate in each state-year-demographic group using information on state benefit rules from the Department of Labor along with data on the wage and salary income of women aged 18 to 49 from the 2000 Census. Potential UI replacement rates range from less than 50 to greater than 60 percent.   

The analysis suggests that each percentage point increase in the local unemployment rate reduces the fertility rate by about 1 percent, controlling for other local economic conditions, county, demographic group, and month. This effect appears to be due to liquidity constraints. When UI is more generous, the effect of rising unemployment rates in depressing fertility is attenuated. With no UI, a 1 percentage point increase in the unemployment rate is associated with a 4.2 percent reduction in fertility rates. Each 10 percent increase in the UI replacement rate is associated with a reduction of about 0.5 percent in the negative impact of unemployment on fertility, so that when UI replaces 100 percent of lost income, there is no effect of unemployment rates on fertility rates.  

Infants in utero during times of high unemployment suffer worse health outcomes than those who are born in better economic times. A 1 percentage point increase in the unemployment rate increases the share of babies born prematurely by 0.14 percentage points, relative to the mean of 13 percent. However, babies conceived during times of high unemployment are healthier: a 1 percentage point increase in the unemployment rate reduces the percentage of babies born with low birth weights by 0.04 percentage points and premature births by 0.07 percentage points. This is consistent with positive selection into childbearing during downturns, since liquidity-constrained women — who tend to be in lower socioeconomic strata and have less healthy babies, on average — are less likely to choose to bear children during periods when unemployment is high.

The analysis finds that more generous UI mitigates the negative effects of unemployment on infant health. Absent UI, each 1 percentage point increase in the unemployment rate would increase the proportion of low-birth-weight infants by 0.17 percentage points and the proportion of preterm births by 0.43 percentage points. A UI replacement rate of about 75 percent would fully offset these negative effects. On average, this would require replacing $383 per week, or $17,000 over a full-term 44-week pregnancy. The cost of birth hospitalization for a Medicaid-insured preterm infant is $42,000 higher than for a full-term infant, implying that the net social return to providing such UI benefits would be positive.   

UI replacement most positively improves infant health for Black women and for women aged 35 to 49, who have a higher incidence of low birth weight and preterm births. Moving from 0 to 100 percent income replacement is calculated to decrease low birth weight and preterm births by 6 percent among Black women. The link between the unemployment rate and infant health is strongest in the first trimester of pregnancy, while the effect of UI replacement is consistent across trimesters. These results are partly driven by changes in maternal behaviors that are associated with economic conditions. Higher in utero unemployment rates are associated with higher rates of maternal smoking and drinking, and with fewer daily servings of fruits and vegetables. UI, by replacing lost income, offsets these effects.

—Whitney Zhang