• The state of Mississippi is challenging the landmark 1973 US Supreme Court ruling that currently protects women's right to abortion before fetal viability.
  • The state believes abortion availability doesn't affect women's ability to participate in society.
  • Here, two experts examine the role of economic research in helping answer objective questions.
  • They establish how powerful statistical methodologies can be used in the social sciences to measure the causal effects of abortion access on reproductive, social and economic outcomes for women and their families.

Dobbs v. Jackson Women’s Health Organization considers the constitutionality of a 2018 Mississippi law that prohibits women from accessing abortions after 15 weeks of pregnancy. This case is widely expected to determine the fate of Roe v. Wade as Mississippi is directly challenging the precedent set by the Supreme Court’s decisions in Roe, which protects abortion access before fetal viability (typically between 24 and 28 weeks of pregnancy). On December 1, 2021, the Supreme Court will hear oral arguments in Dobbs v. Jackson. In asking the Court to overturn Roe, the state of Mississippi offers reassurances that “there is simply no causal link between the availability of abortion and the capacity of women to act in society”[1] and hence no reason to believe that abortion access has shaped “the ability of women to participate equally in the economic and social life of the Nation”[2] as the Court had previously held.

While the debate over abortion often centers on largely intractable subjective questions of ethics and morality, in this instance the Court is being asked to consider an objective question about the causal effects of abortion access on the lives of women and their families. The field of economics affords insights into these objective questions through the application of sophisticated methodological approaches that can be used to isolate and measure the causal effects of abortion access on reproductive, social, and economic outcomes for women and their families.

Separating Correlation from Causation: The “Credibility Revolution” in Economics

To measure the causal effect of abortion on women’s lives, one must differentiate its effects from those of other forces, such as economic opportunity, social mores, the availability of contraception. Powerful statistical methodologies in the causal inference toolbox have made it possible for economists to do just that, moving beyond the maxim “correlation isn’t necessarily causation” and applying the scientific method to figure out when it is.

This year’s decision by the Economic Sciences Prize Committee recognized the contributions[3] of economists David Card, Joshua Angrist, and Guido Imbens, awarding them the Nobel Prize for their pathbreaking work developing and applying the tools of causal inference in a movement dubbed “the credibility revolution” (Angrist and Pischke, 2010). The gold standard for establishing such credibility is a well-executed randomized controlled trial – an experiment conducted in the lab or field in which treatment is randomly assigned. When economists can feasibly and ethically implement such experiments, they do. However, in the social world, this opportunity is often not available. For instance, one cannot feasibly or ethically randomly assign abortion access to some individuals but not others. Faced with this obstacle, economists turn to “natural” or “quasi” experimental methods, ones in which they are able to credibly argue that treatment is as good as randomly assigned.

Pioneering applications of this approach include work by Angrist and Krueger (1991) leveraging variation in compulsory school attendance laws to measure the effects of schooling on earnings and work by Card and Krueger (1994) leveraging minimum wage variation across state borders to measure the effects of the minimum wages on employment outcomes. The use of these methods is now widespread, not just in economics, but in other social sciences as well. Fueled by advances in computing technology and the availability of data, quasi-experimental methodologies have become as ubiquitous as they are powerful, applied to answer questions ranging from the effects of economic shocks on civil conflict (Miguel, Sayanath, and Sergenti, 2004), to the effects of the Clean Water Act on water pollution levels (Keiser and Shapiro, 2019), and effects of access to food stamps in childhood on later life outcomes (Hoynes, Schanzenbach, Almond 2016; Bailey et al., 2020).

Economists also have applied these tools to study the causal effects of abortion access. Research drawing on methods from the “credibility revolution” disentangles the effects of abortion policy from other societal and economic forces. This research demonstrates that abortion access does, in fact, profoundly affect women’s lives by determining whether, when, and under what circumstances they become mothers, outcomes which then reverberate through their lives, affecting marriage patterns, educational attainment, labor force participation, and earnings.

The Effects of Abortion Access on Women’s Reproductive, Economic, and Social Lives

Evidence of the effects of abortion legalization

The history of abortion legalization in the United States affords both a canonical and salient example of a natural experiment. While Roe v. Wade legalized abortion in most of the country in 1973, five states—Alaska, California, Hawaii, New York, and Washington—and the District of Columbia repealed their abortion bans several years in advance of Roe. Using a methodology known as “difference-in-difference estimation,” researchers compared changes in outcomes in these “repeal states” when they lifted abortion bans to changes in outcomes in the rest of the country. They also compared changes in outcomes in the rest of the country in 1973 when Roe legalized abortion to changes in outcomes in the repeal states where abortion already was legal. This difference-in-differences methodology allows the states where abortion access is not changing to serve as a counterfactual or “control” group that accounts for other forces that were impacting fertility and women’s lives in the Roe era.

Among the first to employ this approach was a team of economists (Levine, Staiger, Kane, and Zimmerman, 1999) who estimated that the legalization of abortion in repeal states led to a 4% to 11% decline in births in those states relative to the rest of the country. Levine and his co-authors found that these fertility effects were particularly large for teens and women of color, who experienced birth rate reductions that were nearly three times greater than the overall population as a result of abortion legalization. Multiple research teams have replicated the essential finding that abortion legalization substantially impacted American fertility while extending the analysis to consider other outcomes.[4] For example, Myers (2017) found that abortion legalization reduced the number of women who became teen mothers by 34% and the number who became teen brides by 20%, and again observed effects that were even larger for Black teens. Farin, Hoehn-Velasco, and Pesko (2021) found that abortion legalization reduced maternal mortality among Black women by 30-40%, with little impact on white women, offering the explanation that where abortion was illegal, Black women were less likely to be able to access safe abortions by traveling to other states or countries or by obtaining a clandestine abortion from a trusted health care provider.

The ripple effects of abortion access on the lives of women and their families

This research, which clearly demonstrates the causal relationship between abortion access and first-order demographic and health outcomes, laid the foundation for researchers ­to measure further ripple effects through the lives of women and their families. Multiple teams of authors have extended the difference-in-differences research designs to study educational and labor market outcomes, finding that abortion legalization increased women’s education, labor force participation, occupational prestige, and earnings and that all these effects were particularly large for Black women (Angrist and Evans, 1996; Kalist, 2004; Lindo, Pineda-Torres, Pritchard, and Tajali, 2020; Jones, 2021).

Additionally, research shows that abortion access has not only had profound effects on women’s economic and social lives but has also impacted the circumstances into which children are born. Researchers using difference-in-differences research designs have found that abortion legalization reduced the number of children who were unwanted (Bitler and Zavodny, 2002a), reduced cases of child neglect and abuse (Bitler and Zavodny, 2002b; 2004), reduced the number of children who lived in poverty (Gruber, Levine, and Staiger, 1999), and improved long-run outcomes of an entire generation of children by increasing the likelihood of attending college and reducing the likelihood of living in poverty and receiving public assistance (Ananat, Gruber, Levine, and Staiger, 2009).

Access to abortion continues to be important to women’s lives

The research cited above relies on variation in abortion access from the 1970s, and much has changed in terms of both reproductive technologies and women’s lives. Recent research shows, however, that even with the social, economic, and legal shifts that have occurred over the last few decades and even with expanded access to contraception, abortion access remains relevant to women’s reproductive lives. Today, nearly half of pregnancies are unintended (Finer and Zolna, 2016). About 6% of young women (ages 15-34) experience an unintended pregnancy each year (Finer, Lindberg, and Desai, 2018), and about 1.4% of women of childbearing age obtain an abortion each year (Jones, Witwer, and Jerman, 2019). At these rates, approximately one in four women will receive an abortion in their reproductive lifetimes. The fact is clear: women continue to rely on abortion access to determine their reproductive lives.

But what about their economic and social lives? While women have made great progress in terms of their educational attainment, career trajectories, and role in society, mothers face a variety of challenges and penalties that are not adequately addressed by public policy. Following the birth of a child, it’s well documented that working mothers face a “motherhood wage penalty,” which entails lower wages than women who did not have a child (Waldfogel, 1998; Anderson, Binder, and Krause, 2002; Kelven et al., 2019). Maternity leave may combat this penalty as it allows women to return to their jobs following the birth of a child – encouraging them to remain attached to the labor force (Rossin-Slater, 2017). However, as of this writing, the U.S. only offers up to 12 weeks of unpaid leave through the FMLA, which extends coverage to less than 60% of all workers.[5] And even if a mother is able to return to work, childcare in the U.S. is costly and often inaccessible for many. Families with infants can be expected to pay around $11,000 a year for childcare and subsidies are only available for 1 in 6 children that are eligible under the federal program.[6] Without a federal paid leave policy and access to affordable childcare, the U.S. lacks the infrastructure to adequately support mothers, and especially working mothers – making the prospect of motherhood financially unworkable for some.

This is relevant when considering that the women who seek abortions tend to be low-income mothers experiencing disruptive life events. In the most recent survey of abortion patients conducted by the Guttmacher Institute, 97% are adults, 49% are living below the poverty line, 59% already have children, and 55% are experiencing a disruptive life event such as losing a job, breaking up with a partner, or falling behind on rent (Jones and Jerman, 2017a and 2017b). It is not a stretch to imagine that access to abortion could be pivotal to these women’s financial lives, and recent evidence from “The Turnaway Study”[7] provides empirical support for this supposition. In this study, an interdisciplinary team of researchers follows two groups of women who were typically seeking abortions in the second trimester: one group that arrived at abortion clinics and learned they were just over the gestational age threshold for abortions and were “turned away” and a second that was just under the threshold and were provided an abortion. Miller, Wherry, and Foster (2020) match individuals in both groups to their Experian credit reports and observe that in the months leading up to the moment they sought an abortion, financial outcomes for both groups were trending similarly. At the moment one group is turned away from a wanted abortion, however, they began to experience substantial financial distress, exhibiting a 78% increase in past-due debt and an 81% increase in public records related to bankruptcies, evictions, and court judgments.

If Roe were overturned, the number of women experiencing substantial obstacles to obtaining an abortion would dramatically increase. Twelve states have enacted “trigger bans” designed to outlaw abortion in the immediate aftermath of a Roe reversal, while an additional 10 are considered highly likely to quickly enact new bans.[8] These bans would shutter abortion facilities across a wide swath of the American south and midwest, dramatically increasing travel distances and the logistical costs of obtaining an abortion. Economics research predicts what is likely to happen next. Multiple teams of economists have exploited natural experiments arising from mandatory waiting periods (Joyce and Kaestner, 2001; Lindo and Pineda-Torres, 2021; Myers, 2021) and provider closures (Quast, Gonzalez, and Ziemba, 2017; Fischer, Royer, and White, 2018; Lindo, Myers, Schlosser, and Cunningham, 2020; Venator and Fletcher, 2021; Myers, 2021). All have found that increases in travel distances prevent large numbers of women seeking abortions from reaching a provider and that most of these women give birth as a result. For instance, Lindo and co-authors (2020) exploit a natural experiment arising from the sudden closure of half of Texas’s abortion clinics in 2013 and find that an increase in travel distance from 0 to 100 miles results in a 25.8% decrease in abortions. Myers, Jones, and Upadhyay (2019) use these results to envision a post-Roe United States, forecasting that if Roe is overturned and the expected states begin to ban abortions, approximately 1/3 of women living in affected regions would be unable to reach an abortion provider, amounting to roughly 100,000 women in the first year alone.

Conclusion

Whether one’s stance on abortion access is driven by deeply held views on women’s bodily autonomy or when life begins, the decades of research using rigorous methods is clear: there is a causal link between access to abortion and whether, when, and under what circumstances women become mothers, with ripple effects throughout their lives. Access affects their education, earnings, careers, and the subsequent life outcomes for their children. In the state’s argument, Mississippi rejects the causal link between access to abortion and societal outcomes established by economists and states that the availability of abortion isn’t relevant to women’s full participation in society. Economists provide clear evidence that overturning Roe would prevent large numbers of women experiencing unintended pregnancies—many of whom are low-income and financially vulnerable mothers—from obtaining desired abortions. Restricting, or outright eliminating, that access by overturning Roe v. Wade would diminish women’s personal and economic lives, as well as the lives of their families.