April 22, 2005

Isn't it a logical inevitability that abortion reduces the crime rate?

While the historical evidence raises strong doubts about this popular theory, many people assume it must be true on simple logical grounds. A reader writes:

You began your "Pre-emptive Executions?" article by asking:

Did legalizing abortion in the early 70s reduce crime in the late 90s by allowing "pre-emptive capital punishment" of potential troublemakers?

Steve, the answer to the above question is obviously yes. If you abort a disproportionate number of the fetuses that would grow up to be criminals, you must reduce the crime rate. Of course there may be many other factors that effect the crime rate, as you point out, but these factors don't change the basic fact that elective abortion has reduced the crime rate. To argue otherwise is to make you come off as a doctrinaire conservative, rather than as a scientist.

This seems tautological, but keep in mind that in our country, educated people have a notorious history of misreading how not-so-educated people would react to changes in family structure incentives. For example, all the smart people in 1961 favored raising welfare payments to, say, $300 per month and giving it to unmarried mothers. Nobody they knew would have a baby out of wedlock just to get $300 per month.

Levitt assumes that legalizing abortion reduces the "unwantedness" of the babies who do get born. A close reading of Steven Levitt's book suggests that the reality, however, is not clear at all.

F
irst, we certainly didn't see an increase in wantedness by the fathers of the unborn babies that managed to get born. Legalizing abortion reduced the moral pressure on impregnating boyfriends to marry their girlfriends.

The illegitimacy rate grew steadily from 1964 (which, counterintuitively, was the year The Pill was introduced, yet was also the inflection point in the great illegitimacy upswing), until it suddenly somewhat pleateaued in 1995, the year after the violence rate began dropping, and a few years after the abortion rate began dropping, perhaps not coincidentally.

Lots of people assume that illegitimacy and abortion must be inversely correlated, but the historical record in America shows that they are both high at the same time and low at the same time.

The simplest model appears to be that the Crack Era of the early 1990s was when a lot of the offshoots of the Liberal Ascendancy of 1964-1980 -- crime, illegitimacy, abortion, and venereal diseases such as AIDS -- were seen by many people as all coming home to roost, and a broad turn toward more traditional morality began in reaction to the horrors on the streets.


fter the legalization of abortion, there was not a major drop in unwanted births as Levitt assumed when he concocted his theory, and he still implies even though he knows the facts are otherwise. Instead, there was a major rise in unwanted pregnancies. According to Levitt's own words, "
"Conceptions rose by nearly 30 percent, but births actually fell by 6 percent …" I know I reiterate this, but it's a stunning fact that you never hear in the abortion debate from either side, and it's a key to grasping what the impact of legalizing abortion was in reality, not theory.

Nor is it clear that this small decline in birthrate improved the quality of upbringing of the survivors.

Imagine a woman who started having unprotected sex because abortion was legalized. She gets pregnant, but then, for one reason or another, doesn't have an abortion.

Perhaps she hopes that having the baby will persuade the father to marry her. Perhaps when the father refuses to marry her she decides that if no man loves her, well, at least a baby would love her and cheer her up. Maybe all her girlfriends are having babies and it seems like the fashionable thing to do in her circle. Maybe it gets her out of having to go to high school and take a lot of boring classes she doesn't understand. Perhaps she finds she can get her own public housing project apartment and move out of her nagging mother's house if she becomes a mother herself, and then she can have sex with all the men she wants. Perhaps she keeps forgetting her appointment at the clinic because she's not too bright. Perhaps every time she gets the cash together for an abortion, she spends it on drugs first.

It's a statistical certainty that millions of babies were conceived because abortion was legalized but then were born for these kind of reasons. How many? I don't know.

But it's not at all impossible that legalizing abortion could have, on the whole, lowered the quality of parents and the upbringing they give their kids. In fact, it seems pretty likely that out of the tens of millions of women who had unwanted pregnancies due to legalizing abortion (tens of millions according to Levitt's own numbers), the ones who went ahead and had abortions tended to be the more ambitious, better organized women, while some the the ones didn't get around to having abortions were the more scatter-brained women.

This model fits what we all saw on the streets a lot better than Levitt's model. Urban black women had huge numbers of legal abortions from 1971 onward, far more than any other group. According to Levitt's logic, that should have improved the black male teenagers of the late 1980s through early 1990s.Yet, what evidence is there from, say, 1990 to 1994 that black males born in 1971-1979 were better behaved than the previous generation? The better behaved generation of black teens actually were the ones born in the early 1980s, yet the nonwhite abortion rate peaked back in 1977.

A reader writes:

Like Steven D Levitt, I am a published economist... I am so heart warmed to begin to see empirical data graphically presented, albeit in a critique of economic theory by your art critic, as to be tempted to forgive his sophomoric understanding of empirical social science. In the discipline's lingo, Levitt has relied upon something called ceteris paribus (all other things equal), an inference presumptively valid unless shown otherwise, to conclude abortion lowers crime.

In his critique, Steve Sailer shows that today's youth are more violent and depraved. But his post hoc ergo propter hoc reasoning fails to prove is that abortion has made them so. Rather, in the absence of such proof, ceteris paribus tells us that crime would be worse had it not been "culled" [of] 'the children who stood the greatest chance of becoming criminals.'"

Leaving aside the condescension dripping from this, there are two logical issues here: ceteris paribus and upon whom the burden of proof rests.

I addressed ceteris paribus in my debate in 1999 with Levitt:

Admittedly, it's still theoretically possible that without abortion the black youth murder rate would have, say, sextupled instead of merely quintupling [from 1984 to 1993].

Logically, this is what Levitt must be arguing over these last six years. But you can instantly see why he never makes clear his case. There's two problems: the first is that saying this instantly raises the question of why Levitt refuses to investigate the at least equally interesting question of whether legalizing abortion first drove crime up. As I wrote then:

Still, there's a more interesting question: Why did the places with the highest abortion rates in the '70s (e.g., NYC and Washington D.C.) tend to suffer the worst crack-driven crime waves in the early '90s?

The other reason is the obvious dubiousness of what Levitt is claiming: He is implying that: Although my theory fails its single best test case in catastrophic fashion, I can still separate out the very subtle breeze of the effects of legalizing abortion from the hurricane of other simultaneous events, such as the rise and fall of the crack wars, vast increases in imprisonment, changes in police tactics, the decline in the abortion rate from 1992 onward, changes in the economy, increased sales of guns to law-abiding citizens, increased number of cops, the rise of rent-a-cops, the spread of alarms and video cameras, the rise of marijuana among the urban underclass, the spread of Depo-Provera contraceptive shots, etcetera etcetera...

Well, good luck...

And that brings us to the question of the burden of proof. Upon whom should it rest: Levitt or me?

Levitt is a sympathetic figure, perhaps a heroic one, considering the difficulty of the analytical burden he has undertaken.

I am a less appealing figure: the scoffer, the sniper, the naysayer. I do not offer a complete model of the causes of crime trends as Levitt claims to do. Nor do I feel competent to undertake one. I am merely poking holes in his big theory.

Yet, it's a wise maxim in the sciences that large assertions require large evidence. Levitt's abortion-cut-crime theory is one of the bigger social science assertions of recent times. The weight of the evidence, however, falls far short of the weight of the importance of his claim. So, by all traditions of science, the burden of proof lies upon him, and he has failed to meet it.


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

No comments: