In the season three premiere of Orange Is the New Black, the inmate known as Big Boo comes across Pennsatucky weeping and ritualistically dripping Mountain Dew on six makeshift crosses, each of which represents one of the abortions she had before converting to Christianity. Big Boo comforts Pennsatucky with a theory she learned from Freakonomics, the bestseller by journalist Stephen Dubner and economist Steven Levitt. “By terminating those pregnancies, you spared society the scourge of your offspring,” she says kindly to the former meth addict. According to the idea Big Boo cites, the legalization of abortion in 1973 was partially responsible for a drop in the U.S.’s crime rate 20 years later. “These were children who weren’t wanted,” Big Boo says. “Children who, if their mothers had been forced to have them, would have grown up poor and neglected and abused– the three most important ingredients when one is making a felon.”
John Donohue, an economist and law professor at Stanford, and Levitt, who is currently based at the University of Chicago, first put forth this theory in a 2001 paper titled “The Impact of Legalized Abortion on Crime.” They observed that states that had legalized abortion earlier–New York, California, Washington, Alaska and Hawaii–also enjoyed a decrease in crime comparatively sooner, and that states with higher abortion rates experienced greater reductions in the crime rate. Their idea generated controversy before it was even published. In 1999, The New York Times reported that their theory–then still a work in progress–was already “drawing both outrage and intense debate.”
That debate, which has dragged on for years, is ideologically charged, of course. The theory has always been unpopular with pro-life groups, and the whole idea carries with it a whiff of eugenics. Within a year of its publication, another team found fault with the statistical model used in Levitt and Donohue’s paper; Levitt and Donohue thought they’d controlled for unobservable factors that influence crime rates, but when Christopher Foote and Christopher Goetz, economists at the Federal Reserve Bank of Boston, attempted to reanalyze the data, they discovered a problem in Levitt and Donohue’s code.
“The common view is that it’s been debunked,” Jeffrey Miron, an economist at Harvard and a senior fellow at the Cato Institute, a libertarian think tank, said of the Donohue-Levitt hypothesis. “There are two propositions that are being put forth,” he explained over the phone. “One is that the legalization of abortion led to a reduction in the number of births relative to the number of women.” That part, according to Miron, does hold up–and could partially explain a reduction in the number of crimes. (A smaller population can commit fewer crimes.)
The more interesting and provocative claim, though, is that a disproportionate number of aborted pregnancies took place in a “criminal-producing environment”; only then would the rate of crime, in addition to the number, go down. “That one, I don’t think the evidence supports,” Miron said. According to Miron, the drop in the crime rate can be better explained by diverse factors that vary from state to state. “If the early legalizers happened to be unusual in certain respects, you could potentially get a misleading conclusion,” he said–especially if those unusual factors existed in the more populous states, like California and New York. In New York, for instance, the decrease in violent crime during the 1990s might have more to do with law enforcement policies under Mayors David Dinkins and Rudy Giuliani and the decline of the crack wars than with an increase in abortion 20 years earlier.
Christopher Foote, one of the early skeptics, expounded his issues with the paper in a phone interview. “If abortion was having a big effect on the crime rate, you would expect to see it in terms of how likely people of different ages were to commit crimes,” he said.
One year, we should see a reduction of crime among 16-year-olds; the following year, we should see a decline in crime rate among 17-year-olds, and so on. “If you try to look at things at the state level”–as Levitt and Donohue did–”rather than looking by age, I think there’s more of a probability that you’re going to have confounding factors.”
John Donohue–who hadn’t seen the discussion of his paper on TV, but said he’d heard about it from many people–defended his original paper in an email to Women in the World (and in a follow-up paper he published with Levitt while a law professor at Yale). “Any paper as controversial as the Donohue-Levitt thesis will always draw some attempts at refutation,” he wrote, “but I think it is now widely accepted that children raised in adverse situations are at greater risk of subsequent bad life outcomes.”
“Of course, there can be a dispute over how large the effect is,” he conceded, “but there should be no dispute over the direction of the effect. At its peak, though, 1/3 of pregnancies were ending in abortion so it would not be surprising if a sudden change of that magnitude would have an impact.”
Whether or not Big Boo’s pep talk is economically sound, it does have a conciliatory effect on Pennsatucky, who, it’s safe to say probably would not have been an ideal mother. “I never thought about it that way,” she says. “That makes me feel a little better.” There are worse salves than shaky economic theories.