Date of Award

Spring 2026

Abstract

Fertility rates in the United States have declined by 21 percent since the most recent peak in 2007. Despite more childbearing-aged women in the United States, this fertility rate decline has resulted in fewer births. Tracking these shifts is not a purely demographic exercise. Rather, a declining number of births has wide-ranging effects on society, including school closures, loss of labor and delivery centers, and eventually labor force implications as the number of workers shrinks. To examine the causes of modern fertility decline, this dissertation contains three interrelated analyses. I first examine population-wide fertility trends using data from the Center for Disease Control (CDC WONDER), plus the U.S. Census Bureau’s Population Estimate Program and American Community Survey (ACS). This analysis shows wide-ranging fertility rate declines across different sociodemographic groups. The second and third analytical chapters provide a period study and a cohort study, respectively, of changing fertility patterns by examining past-year childbirth at the person-level. In both chapters, logistic regression models estimate the relationship between past-year childbirth and multiple sociodemographic variables, in isolation and after controlling for other measures. The period study examines three time periods around the Great Recession and the Covid-19 pandemic, both of which had outsized cultural impacts on the United States. These time periods include pre-recession era (2005–2007), pre-Covid era (2017–2019) and Covid-era (2021–2023). Results suggest that the predicted probability of childbirth in each of the three periods is substantively similar, suggesting that declining fertility rates are the result of compositional changes within the population: that is, increasing shares of women in sociodemographic groups that are less likely to give birth. Namely, more childbearing-aged women have a college education, are not currently married, and go longer without having a first child. The cohort study compares the probability of childbirth between the ages of 20–44 for two cohorts: Younger Generation X (born 1973–1980) and Older Millennials (born 1981–1988). I find that despite an overall decline in the probability of childbirth for women under 25, Older Millennials still have a higher predicted probability of childbirth than Younger Generation X. I further contextualize cohort differences by focusing on past-year childbirth at ages 20–27 among Younger Generation X and Older Millennials plus two additional cohorts: Older Millennials (born 1989–1996) and Generation Z (born 1997–2004). This analysis documents a rapid decline in the probability of recent childbirth among the younger cohorts, even after controlling for generational differences in factors like educational attainment. Although these younger cohorts are still in their childbearing years and could “catch up” later, fully closing the fertility gap would require a substantial increase in childbearing among women in their 30s. Taken together, these findings highlight the rapidly changing components of fertility in the United States and add to the literature on the evolving roles of sociodemographic differences in the odds of childbirth.

Document Type

Dissertation

First Advisor

Kenneth Johnson

Second Advisor

Rebecca Glauber

Third Advisor

Cliff Brown

Department or Program

Sociology

Degree Name

Doctor of Philosophy

Download

Share

COinS