Legalizing Abortion Has Positive Consequences Too – Reducing Crime, For Instance

1. You Can't Mean That: Positive Consequences of What?
Imagine a country where crime drops substantially. With crime, I mean violent crime, property crime and murder: serious business. And substantially is substantially. Take a look at the graph. We see on the horizontal axis the years 1970 to 1999 and on the vertical axis an index number. The lines represent the development of violent crime, property crime and murder over the years. The numbers are index numbers where all three sorts of crime have been set to 100 in the base year 1973. Notice two important things. First, all three crimes have a peak in the early nineties: violent crime has risen with 80% 1990, property crime with 30% and murder has risen back at the same level. And then, all three crimes start to fall, substantially. Within nine years, violent crime drops with 33%, property crime with 36% and murder with 47%. Murder and property crime fall below their original 1973-level and both have never been so low.

Source: figure 2 in Donohue and Levitt (2001), p. 392

This country does exist, it is the United States and this development in crime actually happened. All sorts of explanations have been offered to explain this drop in crime: more police, a strong economy and gun control. All of those fail to explain the huge and persistent decline in crime rates in the 90s. Two economists offered a different explanation. It was the legalization of abortion by the US Supreme Court in 1973. With a large dataset and a battery of econometric work, they show that the legalization in 1973 not only explains the starting point of the drop around 1991, but also explains a major part of the persistent decline thereafter. (“Explains a major part” means that things like economic growth are additional minor explanations for crime reduction, and that more police also helps somewhat. See the website of Steve Levitt for more studies about this).

2. Think as (Some) Economists: Legalization Lowers the Price
First, we have to think as an economist. Put your moral sentiments aside for a moment and consider what happens when something illegal becomes legal. No, don't think in statements as 'good' or 'bad', but think in market terms. In the case of abortion, there is a demand for abortion. When it is illegal, the supply of abortion is low, because there are no hospitals offering it, a small number of doctors will practice it hidden in the evening hours and so forth. That means that the 'price' of an abortion is high, either in monetary terms or in other expressions. Women have to travel abroad to countries where it is legal, have to bribe doctors or law enforcers or put their lives at stake in bad treatments. Even when abortion is illegal, it happens, but at a higher 'price'. When abortion becomes legal, supply of the treatment will rise, it will be offered in clinics or hospitals, the treatment will be paid for by medical insurance. In other words, the 'price' of an abortion falls. In usual markets, when the price drops, the quantity of the traded good increases. We see both things happening in the US. The price of an illegal abortion was between $400 and $500 and thirteen years later, the price of a legal abortion was around $80. Secondly, the number of abortions rose from 750,000 before legalization to a number of 1,600,000 seven years after, at which it stabilized (Donohue & Levitt, p. 385). Consistent with these findings is that the number of children put up for adoption declined after legalization. This is a story how an economist would explain both the existence of illegal practices and the rise after legalization. Now back to the crime-and-abortion story.

3. Legalizing Abortus lowers Crime… Some 20 Years Later
How should we assess the impact of abortion on crime? This is mainly through two channels. The first one is a simple: less people is less crime. Abortion lowers the population growth, which means that there are less people to commit crime. Here you have to note that the number of living births stays more or less equal over the years. In the US, abortion became legal by a ruling of the Supreme Court in 1973. After the ruling, we see a steady increase in the number of abortions from 1973 up to 1980, where the number stabilizes at double the number or 1973. Think of the aborted in 1973 as if they had not been aborted, but born and raised. When would some of them have committed their first crime? The high-crime years are between age 17 and, let's say, age 25. Some of the aborted in 1973 would have committed their first crime around 1990. But since they have been aborted, there is no crime for them in the 90s. Now in reverse: think five years later in 1995. We miss here the generations aborted between 1973 and 1978. In 1995, there are even less people to commit crime, neither from the 1973-generation who would be 22 years by then, nor from the 1978-generation, who would have been 17 then. By a pure population reducing-effect, we would expect crime falling after 1990. By the same argument, we would expect to see crime stabilizing after 1998, but at the time of this study, those data were not available yet.

The effect of a shrinking population on a reduction in crime would already been true if abortions are done randomly over the population, and more precise, over all social classes. But abortions in the US are more performed on black, ill-educated teenage girls. So here is another argument on top: the quality of living births increases. Replace yourself in the decision process of a pregnant woman who is considering an abortion. Most people would agree that this is an abnormal situation for a pregnant woman. Most pregnant women are happy with their pregnancy and have a desire to give birth to a child and have a family. But not this woman. So something is wrong here and departs from a normal pregnancy. Rape could be one possibility. But more likely is that the woman is too young, has no husband or no support from her relatives. Maybe she is not ready herself to become a mother. This woman has no means or the support to raise the child in a decent life. She perceives the price as too high, and perhaps the price is too high. Now imagine this woman actually having the baby, instead of aborting. What is the social environment in which this child grows up? Most likely, it is a poor family without a father, a low-educated mother, who is likely to be unemployed. In other words, these are one for one conditions to foster criminal behavior.

There is evidence that abortions are more performed in underclass than in upper-class families. If this is true, then there is not only a number effect (less people is less crime), but also a quality effect (less disadvantaged people is less crime). Both count for as arguments why legalizing abortion lowered substantially crime in the US in the 90s.

4. Let Us Therefore Legalize Abortion! No. Let Us Think More Carefully
What can we learn from this study? Can we use this finding as an argument for the legalization of abortion? Can we say that the good consequences outweigh the bad ones, and that therefore abortion should be legal? No, definitely not. First a scientific argument: this study is based on the situation in the United States. The US history differs from the Italian one, as does the composition of the population, as do the institutions. You cannot take the empirical findings of the US, transplant them in an Italian setting and expect the same outcomes. The societies are too different. Now a moral argument against this good versus bad consequences. It may sound obvious, but not every aborted baby is a would-be criminal. The aborted could also have become a new Beethoven or Einstein. However, we will never know when we perform abortions. And even if it would be the case that the now-aborted run a very high risk of turning into a criminal, then we would violate a basic right by performing abortion: we would preventively punish someone before he has committed the crime.

One thing we can learn for sure. Implementing a law like the legalization of abortion can have unintended consequences. The ruling Supreme Court in the United States of Wade vs Roe was never meant to reduce crime, it was about the right of a woman to rule over her own body. I think that some people never have expected the number of abortions to grow like this in the US. This study provides a new perspective on abortion. Maybe the economic view makes sense, that a lot of women perform abortion because the price of childbearing and child raising is too high. Does it therefore not make more sense to create a world in which mothers can choose to deliver babies they want to keep, a world in which they really can take care of her child, a world in which children can grow up in favorable conditions to good people. I believe that in such a world it ultimately does not matter whether abortus is legal or illegal.

This opinion is based on the original study: Donohue, J.J. and S.D. Levitt, “The Impact of Legalized Abortion on Crime”, Quarterly Journal of Economics, May 2001, vol. 116(2), pp. 379-420. This is rather technical article. The study is popularized S.D. Levitt and S.J. Dubner Freakonomics: A Rogue Economist Explores the Hidden Side of Everything, chapter 4, Where Have All the Criminals Gone? pp. 117-146.