Published on Development Impact

Power of the Pill or Power of Abortion?

This page in:

I am a dual citizen of two countries, both of which legalized safe abortions when I was little or young, meaning that I grew up taking a woman’s right to a safe abortion as granted. Usually, when I hear family planning policy, I think of men and women planning the number, the timing, and the spacing of their children with the aid of modern contraceptives. Even when accidents happen, I think that emergency birth control is mostly available these days (although see here on the recent Plan B decision in the U.S.), so thinking abortions and abortion policy come well after all of these.

Goldin and Katz (2002), and Bailey (2006, 2009) are seminal papers on the role of the introduction of the birth control pill in the U.S. in delaying marriage and pregnancy, and paving the way to increased school attainment and labor market participation. A new discussion paper by Caitlin Knowles Myers at Middlebury College challenges these findings by proposing that the pill may not have been responsible for these changes – at least not solely. She suggests that legalization of abortion may have played a crucial role in delaying marriage and motherhood.

Myers did two things, which look painstakingly difficult. First, pointing out that the variation in coding the timing and de facto status of women’s access to the pill across states, she has “…revised, reconciled, and corrected the codings of these laws after conducting and extensively documenting the results of a new review of the legal environment affecting adults’ and minors’ access to reproductive control (Myers, 2012).” These are documented very clearly in long tables in appendices to the 77-page paper.

Second, she carefully distinguishes between legal accesses of adults and minors to the pill and to abortion. While the previous papers did control for abortion laws in examining the effects of the pill, she notes that, for each of abortion and the pill, there were states in which no one had access, there were others where only adults and married women had access, and yet others where all women of reproductive age had access regardless of their age or marital status. She carefully details the timings of these as well, makes distinctions between de jure and de facto access and codes them clearly to include in her analysis.

After these changes to the coding, she uses the Current Population Survey (CPS) June fertility supplements to show that once she controls separately for the pill being legalized, pill being accessible to minors, abortion being legalized, and minors being able to consent to abortion, it is the abortion variables that are associated with decreases in the likelihood of marriage, childbirth, and shotgun marriages (marriages followed by a birth within eight months).

As there are many moving parts here, the author also tries to show us that it is the changes she has done to the coding of access to abortion and the pill for adults and minors that’s causing the change, rather than moving from Census data to the CPS. In Table 5, she uses the data and sample from Goldin and Katz (2002) and shows that once she moves to her legal coding from theirs, the effect of the pill becomes smaller and statistically insignificant. Interestingly, when she moves to her own data, the legal coding of Goldin and Katz no longer show an effect for the pill while access to abortion reduces the likelihood of being married by age 23.

When discussing the plausibility of the finding that the pill may not have decreased the likelihood of giving birth by age X, the author provides some evidence that while the pill likely decreased the pregnancy rates conditional on being sexually active (some women substituted away from condoms but the pill is more effective in preventing unwanted births), this effect may have been offset by the increased likelihood of having sex as a teenager among those cohorts who had access to the pill. While not completely persuasive, I find this demonstration plausible. Interestingly, but not surprisingly, the pill led to a large reduction in marital fertility (where this offsetting effect was absent). The author also discusses the plausibility of legal access to abortions causing these same changes, but this part remains more speculative.

Should these results hold up to scrutiny from reviewers, do they change how we approach family planning policy in developing countries? On the one hand, the introduction of the pill 50 years ago in the U.S. is certainly different than the landscape that we face now. Today, we have Plan B, injectables, and implants. We also have HIV/AIDS. The opportunities to work with school-age girls, young women, and older adults with respect to family planning with modern contraceptives are more abundant. On the other hand, women in many countries still don’t have access to safe abortions. In Malawi, where it is a crime punishable by prison to receive or provide abortions, traditional authorities have recently been agitating the government to review the anti abortion laws in the country, because they realize that many women are resorting to unsafe abortions.

Markus recently blogged about two papers that examine the negative effects of the global gag rule imposed by the U.S. (now rescinded by Obama). Evidence is mounting, suggesting that a comprehensive family planning policy may not only ensure unfettered access to modern contraceptives but also access to safe abortions. Given the status and bargaining power of women in many developing countries, that will not be enough and we need to be thinking more creatively about increasing women’s, especially young women’s and adolescents’, access to family planning.


Berk Özler

Lead Economist, Development Research Group, World Bank

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000