Delaying the Inevitable

I'm a big fan of Stephen Levitt's research on the correlation between lower crime rates and Roe v. Wade and highly recommend his book Freakonomics to anyone. He had a bit more information on his blog earlier this week that is well worth checking out, but if you don't here's an excerpt:

When a woman gets an abortion, for the most part it is not changing the total number of children she has; rather, it is shifting the timing so those births come later in life. This is an important fact to remember. One in four pregnancies ends in abortion and this has been true for 30 years in the U.S. But the impact of abortion on the overall birth rate has been quite small.

I wonder about the impact of medically necessary abortions, which account for 6% of abortions in this country, on Levitt's numbers. Would women who have abortions by choice actually end up having more children. Perhaps women who have abortions actually have more children to make up for women who have medically necessary abortions and then aren't able to have more children.