Remember the debate that legalized abortion decreased the crime rate back 2001 from @Freakonomics economist Steven Levitt and coauthor Jon Dononohue? Because of an error on one of their tables the hypothesis was dismissed but it turns out it's pretty well evidenced- thread🧵
@Freakonomics 2/ The reduction in crime rate as a function of abortion legalization was backed up in the original 2001 paper but their most recent paper finds a strong effect of legalized abortion on crime rate with 17 more years of crime data bfi.uchicago.edu/wp-content/upl…
@Freakonomics 3/ Donohue and Levitt (2019) looked at the legalization of abortion in specific states and how much the crime has fallen. They estimate that 45% of the 50–55% decline in the peak of crime in the early 1990s is due to legalized abortion.
@Freakonomics 4/ Five states were early legalizers of abortion- Alaska, California, Hawaii, New York, and Washington- Violent crime fell by an additional 21% and homicide fell by 30% in early legalizing states relative to the rest of the nation between 1997 and 2014.
5/ "Aggregating over the entire time period 1985 to 2014 high abortion states have experienced a reduction in crime relative to low abortion states of 64% for violent crime 50% for property crime and 45% for homicide”. Donohue & Levitt (2019)
6/ Here you can see how much the violent crime rate has decreased in states with higher abortion "the higher violent crime rate in the high abortion states began shrinking almost to nothing by 2014" Donohue & Levitt (2019) 
7/ Figure VI of Donohue & Levitt (2019) shows that incarceration rates rose more sharply and are now substantially higher in the low abortion states.
7/ In the podcast Levitt says the findings aren't policy relevant "I don’t think anyone who is sensible should use our hypothesis to change their mind about how they feel about legalized abortion" freakonomics.com/podcast/aborti…
8/ In the podcast Levitt maintains that these findings aren't very policy relevant " I don’t think anyone who is sensible should use our hypothesis to change their mind about how they feel about legalized abortion. So it really isn’t very policy-relevant" freakonomics.com/podcast/aborti…
9/ There are about 600,000 abortions a year in the USA and according to Levitt and Donohue’s calculations there are 5,000 to 10,000 fewer homicides per year because of these abortions.
10/ As you'll see in the report the word “genetic” is conspicuously absent. The authors state that the children being “unwanted” and related environmental effects make them more prone to criminality- but this isn't examined in detail.
11/ They make the case that if the state makes coercive efforts to reduce the abortion rate, without also reducing the number of unwanted pregnancy in other ways you'll see the same increase in criminality in that birth cohort. Again, no mention of genetic confound.
12/ In 2001 Levitt got death threats "The number of death threats I got from the left was actually greater than the number of death threats I got from the right. Because the media coverage quickly became a question of race, even though really our paper wasn’t about race at all."
13/ There is a lot for the left and the right to hate about the evidence here- and how much more effective the reduction in unwanted pregnancies has been than interventions each side likes (eg policing, education, incarceration). So it's no wonder we never hear about it.
14/ When Levitt says we could decrease the number of "unwanted" pregnancies, one way would be to give women and men who don't want children cash incentives to take contraception. But this idea is incredibly controversial- see Project Prevention en.wikipedia.org/wiki/Project_P…
15/ To me, it seems much more ethical to give poor people money to take contraception than for them to spend huge amounts of money on abortions along with all the suffering that choosing an abortion entails- the spectre of eugenics here prevents us from making humane policy.
16/ Reduction in lead exposure has a complementary effect on crime but does not much reduce the abortion effect according to Reyes (2007) (i.e. when you control for one you still get the other and vice versa).
