Presentation (CTI)

OBJECTIVES: Despite steadily declining over the past few decades, the teen birth rate in the US is still one of the highest among developed countries. Amid discussions about the consequences of state abortion bans following the overturning of Roe v. Wade in 2022, the context of teen childbearing in the United States makes it important to study the impacts of teen birth and decreased abortion access on the long-run outcomes for women. Estimating this causal relationship remains challenging due to the limitations of data. This study proposes a novel method to assuage this issue.
METHODS: First, I examine the short-term impacts of poor abortion access. My main data come from the Myers Abortion Facility Database (MAFD) and the American Community Survey (ACS) 2014-2023. The Myers Abortion Facility Database provides me with a measure of abortion access by county. I match the distance data between 2009 and 2015 with the women born between 1991 and 1996 of the ACS to evaluate the short-term impacts of changes in abortion access and teen motherhood. Since the length of the datasets did not allow me to examine longer-term outcomes, I develop a random forest model to predict the county-level availability of abortion providers in the US between 2000 and 2009, using the data from the MAFD as my training data.
RESULTS: I find a highly statistically significant negative impact of 1-1.6 percentage points on high school graduation. While the distance, which proxies for the teen childbearing risk, affects education and marriage outcomes, I find weakly significant impacts on Supplemental Nutrition Assistance Program recipiency, monthly earnings, and family-size adjusted household income.
CONCLUSIONS: Overall, my results reveal that the impacts of abortion access are adverse but rather modest. A possible driving mechanism is that women in low access areas pre-emptively make "defensive" investments in more effective contraceptive methods.