A friend of mine introduced me to the Freakonomics podcast that you can download from iTunes. Two Steve’s (Stephen Dubner and Steven Levitt) put together a book called Freakonomics. They seek to uncover “the hidden side of everything.” Netflix has the Freakonomics movie, and you can watch it here (if you have a streaming Netflix account.) The movie discusses several topic: potty training, cheating teachers, bizarre baby names, self-dealing Realtors, crack-selling mama’s boys, sumo wrestling, and many other topics.
If you watch the movie, at about the 50 minute mark is a segment titled “It’s Not Always a Wonderful Life.” The crime rate in the late 1980’s was growing. Crack cocaine was an epidemic, fueling the crime rate; conventional wisdom said that crime would continue to increase.Yet a funny thing happened in the early 1990s: the crime rate dropped.
Many theories came out to explain why the crime rate dropped. Levitt’s data showed that Community Policing strategies didn’t make any difference. The following items explained a little more than half of the drop in crime:
- harsher sentences: 30%
- decrease in crack cocaine use and associated violence: 15%
- gun control, better economy, and more police officers: 10%
This still leaves about 45% of the decrease in crime unexplained. Levitt had an unusual explanation: abortion.
In 1966, Romania outlawed abortion in order to create massive population growth. The birth rate of the country doubled, as did unwanted children and associated crime. On the other hand, in 1973 the Supreme Court legalized abortion in the U.S. after the Rowe vs Wade decision. 20 years after Rowe vs. Wade would have been the early 1990’s. Levitt believes that these unwanted children weren’t born, lowering the crime rate here in the U.S. He views the U.S. and Romania as a natural experiment to show the effect of abortion on crime rates.
As further evidence in support of his theory, he gives the following pieces of data.
- 5 states (NY, CA, HI, AK, and WA) legalized abortion 3 years before Rowe vs Wade. Their crime rates dropped 3 years before the other states.
- After Rowe vs. Wade, states that made it easier to get abortions had better improvement in crime rates than states with hard to get abortions.
Levitt views this as an “unintended consequence” of abortion, and says his study doesn’t have a position on whether abortion should be legal or not, but he does think this should be part of the discussion. What do you think of Levitt’s theory?
His math is pretty good, all in all. Classic micro-econ, which is really applied statistics, always an interesting nugget here or there.
Here are a couple of old rebuttals to Levitt’s theory in case you’re interested:
Yes, I studied Econometrics in college and analyzed Leavitt’s method. Of course, he had to disingenuously tweak his regression model in order to support his “unorthodox” theory. See John Lott’s rebuttal as well.
interesting rebuttals. I liked the lew rockwell rebuttal, because he does seem to look at more factors than just crime rate, such as the rise in single-parent families. it does seem that levitt didn’t really account for those statistics.
John Lott. http://www.amazon.com/review/RZZLKHGVWWIIW/ref=cm_cr_pr_cmt?ie=UTF8&ASIN=1596985062&nodeID=&tag=&linkCode=#wasThisHelpful and a fame for sock puppets.
Citing to Lott is not going to make very many people happy. At least not me, given the way I had a chance to look at a specific citation of his and the negative tone of the interaction we had on it.
Studies have also shown that cigarettes lower long term health care costs by lowering life expectancy and the attendant costs of old age.
I love it when morality and pragmatism come into conflict.
Nate, you seem to be leaving out the fact of all the costs incurred by states for treating lung cancer patients. Yes life expectancy does go down, but costs of trying to keep a smoker alive are quite similar to the “attendant costs of” treating heart attacks, strokes, lung cancer, mouth cancer, etc.
I don’t know what to think of Levitt’s study, or any of the rebuttals. I don’t think Levitt’s study is the final word on the subject, but I found the idea interesting.
My daughter gave me the Freakonomics book as a present before my cataracts started developing. Now that I’ve got an eye back in shape, I’ll have to put it back on the reading list.
I guess the debate proves that all acts have effects and side-effects, and whether they’re good or bad depends on the weighting assigned by the value system behind them.
All family patriarchs named Laban take note.
Mormon Heretic — the cost/social benefits equation of cigarettes depends on what social health benefits you have. A number of Eastern European countries don’t pay out for the “attendant costs.”
Or why in the United States cigarettes are a net loss for the state, some places a net plus.
Leavitt’s study was only intended to be a beginning observation.
My takeaway is that pretty much all issues are more complex than we like to think they are, especially when morality leaks over into legality.
I wrote a post way back in June 2008 about abortion as a legal issue, entitled “When Moral Issues Become Legal Issues”. If anyone is interested: