Yeah, I know, you’re sick of hearing about Matt Yglesias. But I want to address something that kept coming up in his posts and the comments thereto, as well as Ross Douthat’s defense of abortion criminalization. Namely, the idea that the Guttmacher study is flawed and those who say that criminalization does not affect abortion rates are wrong because “everyone knows” abortions in the US skyrocketed after Roe.
What Matt said:
In the United States, when abortion was legalized in the 1970s, the number of abortions went up.
What Ross Douthat said:
Whereas we know that when abortion was legalized in America in the early 1970s, the abortion rate went up dramatically;
Actually, we know no such thing. According to the Guttmacher Institute, the number of LEGAL abortions rose dramatically after abortion was legalized; there’s no data on the absolute number of abortions performed pre-Roe. From the note on page 4 of the report:
Prior to the nationwide legalization of abortion, information on the number and rate of abortions was not gathered, and estimates of illegal and self-induced abortions varied widely. In the years immediately following the Roe v. Wade decision, the number of LEGAL abortions grew rapidly for several reasons. The number of physicians trained and experienced in the procedure increased, and a nationwide network of outpatient abortion clinics developed that enabled women who would previously have had an illegal abortion, or would not have been able to obtain one at all, to do so legally in a medical facility.
Now, there was data on legal abortions prior to Roe because 15 states had legalized abortion or reformed their abortion laws by then. Naturally, when you go from abortion being legal in 15 states to abortion being legal in all 50, you’re going to see a dramatic rise in the number of legal abortions. But the data doesn’t support the conclusion that Matt and Ross have both drawn from it — that criminalizing abortion in the US would necessarily result in a dramatic drop in the absolute rate of abortion, rather than just a drop in the rate of legal or safe abortion.
So, what we have here is a case of a statistic that “everybody knows” is true being used to support a conclusion that really, when you go to the source of that statistic instead of relying on what you “know,” isn’t supported by that statistic at all.