WHO would have thought the debate over abortion in America could become any more inflammatory? Now it has. John Donohue of Stanford University Law School and Steven Levitt, an economist at the University of Chicago, have released the results of an unpublished paper* that attributes as much as half of the sharp drop in American crime rates in the 1990s to the 1973 Roe v Wade decision by the Supreme Court that legalised abortion throughout the United States.
The authors argue that the connection between abortion and crime is chillingly simple: a steep rise in abortions after 1973 has meant that many individuals prone to criminal activity in the 1990s were never born. There are two reasons for this. First, abortion shrinks the number of people who reach the age where they are most prone to commit crimes. Second, and more important, abortion is not random. Teenagers, unmarried women and blacks are more likely than average to have abortions; they are also more likely to have children at risk of committing crimes later in life. Similarly, women with unwanted pregnancies are less likely to be good parents and may do things during pregnancy, such as take drugs, that make future criminality more likely.
The authors present three strands of evidence to support their conclusion. First, the precipitous drop in crime across the country coincides with the period in which the generation affected by Roe v Wade would have reached the peak of its criminal activity, at the age of 18-24. Second, the five states that legalised abortion in 1970, three years before Roe v Wade, were the first to experience the drop in criminal activity. And, last, states with high abortion rates from 1973 to 1976 have seen the largest fall in crime since 1985, even after checking for other factors such as incarceration rates, racial composition and income. (In contrast, there is no relationship between abortion rates in the 1970s and crime before 1985, when abortion-affected groups had not yet reached criminal age.)
The authors reckon that a 10% increase in the abortion rate is associated with a 1% decrease in crime; current crime rates would be 10-20% higher if abortion had not been legalised. Using current estimates for the social cost of crime, the authors estimate the concomitant social benefits of abortion to be in the order of $30 billion a year. They also predict that crime rates will continue to fall by 1-2% a year for 15-20 years as the full effects of legalised abortion are felt.
The premises of the paper all seem reasonably sound. It is by putting them together that the authors of the paper have managed to shock a lot of people, by no means all of them in America.