What Economics Can (and Can't) Tell Us About the Legacy of Legal Abortion

Research suggests Roe v. Wade left the U.S. with fewer births, fewer poor children, and maybe less crime. But economists still have yet to crack some the biggest questions about the decision's impact, explains Wellesley College Professor Phillip Levine.
(Wikimedia Commons)

America's abortion debate won't ever be resolved by an economics paper. But that doesn't change the fact that child-bearing and fertility are deeply entwined with not just our demographic growth, but also women's  professional lives and families' financial health. So with the 40th anniversary of Roe v. Wade falling this week, I called up Wellesley College Professor Phillip Levine to get a sense of what we do and don't know today about the economic impacts of legalized abortion.

A leading researcher on the topic, Levine explained that while there are couple of settled facts, there's still an amazing amount that, decades after the Supreme Court's landmark decision, is still murky. Our conversation is below, edited for length and clarity.

I'd like to start with the broad question: What do we know for sure about how the legalization of abortion impacted the economy?

I think the first thing that we can learn from the research that's been conducted is that abortion legalization clearly had a dramatic impact on women's fertility. It wasn't necessarily clear a priori that was the case. People had talked about illegal abortion being so prevalent that once it was legalized it wasn't obvious it would have much of an impact. But that's just not true. It had a very large effect. A reasonable estimate is that American women's fertility fell on the order of about 5 percent in direct response to abortion being legalized. That's the average effect. For certain population subgroups, it was much larger than that. So you're talking about a 10, 12 percent reduction in fertility for teenagers, for African American women. In certain pockets, certain segments of the population, there was just a very dramatic impact on women's fertility. And that is something which I think we certainly know.

A second thing I think we could say we've learned -- and this may not come as a surprise either -- but if you're having changes in fertility of that magnitude, the characteristics of the children who are born are different than they otherwise would have been. It's very difficult to believe that the women who chose to have an abortion and didn't have a child because of that were randomly selected in the population. And so, because of that, the lack of random selection, the children who were born were different than the ones who would have been born otherwise.

Living standards of children growing up were very different as a result. Fewer children grew up living in poverty, fewer children grew up in single parent households, fewer children grew up in households headed by welfare recipients. In some sense, you can think about following that cohort's path through life into things like educational attainment, labor market outcomes. You observe increases in college graduation, lower rates of welfare use for the children themselves, reduced likelihood of becoming a single parent themselves. These are outcomes for the children who were born in the early 70s that we observe 20 years later, that we observe for the cohort as a whole. Because it's a different group of children born relative to those who would have been otherwise. That's not to say that's a good thing, that's just what happened.

Is it fair to call those economic benefits?

I think this is an incredibly sensitive topic, so I want to avoid using terms like ... "benefit." Different children were born. An empirical fact is that the ones who were born are the ones who were more likely to have better economic outcomes. The value judgment on whether that's a good thing or a bad thing is something I don't want to get into.

What about women in particular? How did legalizing abortion affect their economic standing?

It turns out getting direct evidence of that is not such an easy thing to do. While there certainly are indications that women's economic outcomes were directly improved by abortion legalization, I wouldn't necessarily say that's the sort of thing that research has conclusively proven. We are clearly able to find support for the notion that their fertility was affected. And you certainly could imply from that if you're better able to time giving birth, you're better able to make other sorts of decisions that would improve your economic well being. Direct evidence of that is a little bit limited. There is some work that shows educational attainment and labor market outcomes have improved, but it's a little bit limited.

Why is that research so limited?

Statistically, it's a very hard thing to find support for. It's a statistical issue as opposed to an economic issue.

In some sense, what economists are always trying to do when we find evidence for something -- you're trying to look for something that looks like an experiment. Where one group is exposed to something and another group wasn't, and what was the difference in the outcomes. That works really well when thinking about a fertility decision when the law changes and not so well when you're thinking about these longer term decision-making processes.

So what do you think is the most concrete thing you can say for sure about what abortion did for women's economic lives?

It clearly affected their ability to regulate their fertility. One-hundred percent that happened. It certainly is a direct implication from there that they are then able to make different economic decisions.

But it's hard to confirm that's what happened for sure using data?

Unfortunately, we're not at that stage.

What do you think are the most controversial claims you've seen about the economics of abortion?

I think the abortion and crime debate pretty much overshadowed all other research on the topic, which I think is a shame because there are some other interesting issues that have been raised, like the sorts of things we've discussed.

Can you explain what happened in that debate?

The notion about abortion and crime is that some of the children who were not born were children who would have grown up to be individuals who were more likely to commit crime. At the time that this literature was coming out, this was in the mid 90s, and we were right in the middle of experiencing this big decline in crime rates and nobody had any good explanations as to why. And the argument in some sense was that the criminals were not born because abortion was legalized 20 years ago, or exactly 20 years ago, and that's when criminals are committing the bulk of their crime. That led to "is that a good thing or a bad thing?" and "does that mean we should favor legal abortion because the criminals aren't being born?" It got sidetracked into a discussion about abortion rights. And then the economists were all focusing on the use of the data and the methods being used. It was controversial in the public sphere and it was controversial in a statistical sense. There's some good work that's been done that seems to indicate at least that their effect was overstated, if not completely invalidated.

Why was that a distraction?

Crime is an important outcome, but not the only important outcome in the world. The literature tended to be a distraction in the sense that people weren't really focusing on the actual question of what did abortion legalization do in a sort of empirical sense. Instead it became a philosophical question about abortion rights.