"The Roe Effect"

For a long time, James Taranto of the WSJ web page has been pushing the idea that legalized abortion has had a huge effect on American demographics, with Democrats killing themselves off. It's his hammer that works with every nail.

The Washington Post reports on an interesting new analysis by the National Campaign to Prevent Teen Pregnancy. The campaign, noting that U.S. teen birthrates fell 30% between 1991 and 2002, calculates that if those rates had instead remained constant, there would be some 406,000 additional children living below the federally defined poverty line and some 428,000 living in households with single mothers.

Since 1991 was exactly 18 years after Roe v. Wade, we got to wondering if the Roe effect might have something to do with all this. The Roe effect would predict that the effect of a reduction in birthrates would be greatest in liberal states, where pregnant teenagers would be more likely to exercise their "right to privacy" and thus less likely to carry their babies to term. The campaign's numbers seem to bear this out.

Here, in order, are the 10 states with the biggest percentage decline in teen birthrates (links for tables in PDF): California, Maine, Michigan, Alaska, New Hampshire, Washington, Vermont, Massachusetts, Connecticut, Hawaii.

I doubt it. If you'll look at his list of ten states, they are primarily northern states with mostly white populations. (The main exception is California, which is on the list because in 1991 it was in the throes of a huge Hispanic baby boom caused by the 1986 amnesty for illegal aliens. That has since died down ... but of course Bush wants to start another illegal immigrant baby boom with his amnesty plan. The other exception is Hawaii, which has a unique population whose social dynamics I don't pretend to understand well.)

Yet, the abortion rate has fallen since 1991, especially for whites: from about 19 per 1,000 15-44 year old white woman in 1991 to only about 11 by 1999. (See page 8 of this Alan Guttmacher Institute report). Black and Hispanic are also down too, although not by as large a percentage. So, I strongly doubt that increased abortion in these mostly white states accounted for the decline in teen births. I think the main cause was decreased teen pregnancies.

I haven't studied the general question of the validity of the "Roe Effect" in detail, but I suspect Taranto badly exaggerates its magnitude. What he doesn't grasp is that legalized abortion led to a big increase in pregnancies and only a small decline in birthrate -- in other words, most conceptions ended by abortions would never have happened in the first place without legalized abortion. According to Steven Levitt, in the 1970s conceptions went up by 30% but births declines by only 6%. So, the demographic impact of legal abortion is significantly smaller than the enormous raw numbers (almost 1.6 million abortions per year in the 1980s and about 1.2 million per year in the 1990s -- page 2 of the Guttmacher report) would suggest.

Legalized abortion, however might have had some noticeable effect on the size of the black population -- by 1999, the black abortion rate was close to five times the non-Hispanic white abortion rate, up from about three times earlier -- but Taranto is too coy to come out and say that.

My published articles are archived at iSteve.com -- Steve Sailer