April 28, 2005

Levitt on the waste caused by legalizing abortion:

The impact of legalizing abortion on the crime rate appears beyond the ability of contemporary social science reliably to tease out from the maelstrom of currents roiling American social life over the last 40 years. If you had started looking at the question in 1996, using crime statistics through 1994, the most likely conclusion would have been that legalizing abortion increased crime. Starting in 1999, using statistics from 1997 compared to 1985 (but ignoring the intervening years), Levitt was able to make the case for the opposite conclusion. Judging from data through 2002, however, Levitt's case is once again eroding as the murder rate for 25-34 year olds goes up as the post-legalization cohort enters that group.

My best guess is that legalization worsened crime, but that this effect faded after a number of years as society adjusted. But, that's just a guess.

Instead, the key fact, the under-appreciated take home lesson from this controversy, the observation that Levitt and I agree upon, is that legalizing abortion greatly increased the number of unwanted pregnancies.

Tim Harford writes in The Financial Times of London:

In fact, [Steven D.] Levitt now says that the research made him more pro-life. “I grew up in Minnesota. Very liberal,” he says. “I was just from birth taught to be pro-choice.” But when he discovered while writing the paper that after Roe v Wade the number of abortions rose to nearly 1.5 million a year, and that while the number of births fell, the number of conceptions rose, he thought again. “One in four of the pregnancies which took place were just because people were lazy,” he says. “That’s a lot. That’s a lot of abortions.”

Of course, that fact undermines the persuasiveness of Levitt's assumption that abortion cut crime by reducing the number of unwanted births. Because legalizing abortion caused tens of millions of conceptions that wouldn't have happened otherwise, the overall impact on who actually got born becomes extremely hard to model. Levitt's simplistic assumption that legalization improved the quality of parents and children, which is the key to the popularity of his argument, thus drops in plausibility from a sure-thing to a nobody-knows.

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

No comments: