September 30, 2005

William Bennett blasted for citing Steven D. Levitt's Freakonomics theory

Ever since New Orleans, the hysteria among the political and media elite has been building: Who among us bigshots will crack first and allude to the elephant in the living room?

That trumpeting pachyderm that they've all been trying to ignore is the higher crime rate among African-Americans. According to the official Bureau of Justice Statistics, "Blacks were 7 times more likely than whites to commit homicide in 2002," but You Can't Talk About That.

Today, the mounting pressure finally burst over merely an abstract musing on the radio.

House Democratic Leader Nancy Pelosi immediately spoke on the floor of the House:

"Mr. Speaker, I rise this evening to express my deep disdain and disgust for comments made yesterday by former Reagan Secretary of Education William Bennett on his radio call-in show. ... These are shameful words. I am appalled to have to say them on the floor of the House of Representatives. Secretary Bennett's comments reflect a narrow-minded spirit that has no place within American discourse."

Democratic National Committee Chairman Howard Dean issued the following statement:

"Bill Bennett's hateful, inflammatory remarks regarding African Americans are simply inexcusable. They are particularly unacceptable from a leader in the conservative movement and former Secretary of Education, once charged with the well being of every American school child. He should apologize immediately. This kind of statement is hardly compassionate conservatism; rather, Bennett's comments demonstrate a reprehensible racial insensitivity and ignorance. Are these the values of the Republican Party and its conservative allies? If not, President Bush, Ken Mehlman and the Republican Leadership should denounce them immediately as hateful, divisive and worthy only of scorn.

So, what horror of horrors did Bennett blurt out? The Washington Post reports:

Bennett Under Fire for Remark on Crime and Black Abortions

Democratic lawmakers and civil rights leaders denounced conservative commentator William J. Bennett yesterday for suggesting on his syndicated radio show that aborting black children would reduce the U.S. crime rate.

The former U.S. education secretary-turned-talk show host said Wednesday that "if you wanted to reduce crime, you could -- if that were your sole purpose -- you could abort every black baby in this country, and your crime rate would go down." Bennett quickly added that such an idea would be "an impossible, ridiculous and morally reprehensible thing to do." But, he said, "your crime rate would go down." ...

Bennett's comments came Wednesday, during a discussion on his talk show "Morning in America." A caller had suggested that Social Security would be better funded if abortion had not been legalized in 1973 because the nation would have more workers paying into the system.

Bennett said "maybe," before referring to a book he said argued that the legalization of abortion is one of the reasons the crime rate has declined in recent decades. Bennett said he did not agree with that thesis.

"But I do know that it's true that if you wanted to reduce crime, you could -- if that were your sole purpose -- you could abort every black baby in this country, and your crime rate would go down," Bennett said, according to an audio clip posted on Media Matters for America's Web site. "That would be an impossible, ridiculous and morally reprehensible thing to do, but your crime rate would go down. So these far-out, these far-reaching, you know, extensive extrapolations are, I think, tricky."

That's it? It's a symptom of the September Sickness that has afflicted our elites all month that the repressed frenzy finally broke over that.

There are two somewhat separate logical issues to which Bennett is referring. The first is the utterly theoretical question of the impact on crime of aborting every black baby (since African-Americans made up 50.8% of homicide offenders in 2002 according to federal statistics, it would obviously be large), while the second is the much-admired Freakonomics theory that legalizing abortion in the 1970s reduced crime in the 1990s. Bennett, obviously, opposes both ideas on moral grounds.

As for the first, democrat Brad DeLong sensibly points out:

His caller said: "Abortion is bad because it has worsened the financing of Social Security." Bennett says: "Stay focused. We're anti-abortion not because we think that abortion is a means that leads to bad ends like a higher Social Security deficit; we're anti-abortion because abortion is bad; make arguments like 'abortion is bad because it increases the Social Security deficit' and other people will make arguments like 'abortion is good because it lowers the crime rate' and we'll lose sight of the main point."

Bennett is attempting a reductio ad absurdum argument.

As for the second, ABC reported:

In an interview with ABC News, Bennett said that anyone who knows him knows he isn't racist. He said he was merely extrapolating from the best-selling book "Freakonomics," which posits the hypothesis that falling crimes rates are related to increased abortion rates decades ago. "It would have worked for, you know, single-parent moms; it would have worked for male babies, black babies," Bennett said.

Bill, Bill, Bill, that's what you get for reading the softball reviews of Freakonomics in the NYT, the WSJ, the WP, and the LAT instead of reading, where you would have learned that economist Steven D. Levitt's ultrapopular but slapdash abortion-cut-crime theory disastrously failed to predict even the past.

Bill, if you'd gone to the source for statistical social analysis instead of all those credulous, innumerate mainstream sources, you would have known that when abortion was legalized over 1970-1973, the homicide rate of 14-17 year old black males, rather than declining, more than quadrupled in the decade from 1983 (when all living 14-17 year olds were born in the last prelegalization years of 1965-1969) to 1993 (when they were born in the high abortion years of 1975-1979, when the nonwhite abortion rate peaked in 1977 -- see page 8 of this report for abortion trends).

You can go look for yourself at the homicide graphs that Levitt was too slipshod in his research methods to look at when he came up with his theory in 1999. Go to the federal Bureau of Justice Statistics page here and page down to the second set of four graphs, which show homicide offending rates by age by race by sex.

The main reason Levitt's theory didn't work in reality was because the larger impact of legalizing abortion was to drive up the number of unplanned pregnancies. Levitt himself wrote in Freakonomics that following Roe, “Conceptions rose by nearly 30 percent, but births actually fell by 6 percent …” The most unremarked but remarkable historic fact about legalizing abortion was how pointless it turned out to be: mostly it just caused the very problem -- unwanted pregnancies -- it was purported to cure.

I pointed this out to Levitt in 1999 in our debate in Slate, but he went ahead and left all these inconvenient facts out of Freakonomics six years later. Misleading the public has made him a rich man, but he has to live with his conscience.

I laid all this out in even more monomaniacal detail last Spring, and I apologize to my long-term readers for taking up their time then and now.

One thing I've noticed is that the pro-lifers have shown almost zero interest in the fact that Levitt's theory isn't empirically valid. Strikingly, many of them want it to be true in order to prove the purity of their moral intentions: Even though legal abortion would lessen the chance of me being murdered or mugged, I'm still against abortion on principle.

Well, swell, but that's just moral vanity. Whatever happened to "the truth shall set you free"?

David Brock's Media Matters, which mostly broke the story, claimed:

Bennett's remark was apparently inspired by the claim that legalized abortion has reduced crime rates, which was posited in the book Freakonomics (William Morrow, May 2005) by Steven D. Levitt and Stephen J. Dubner. But Levitt and Dubner argued that aborted fetuses would have been more likely to grow up poor and in single-parent or teenage-parent households and therefore more likely to commit crimes; they did not put forth Bennett's race-based argument.

That's disingenuous, almost as disingenuous as Dr. Levitt.

Levitt's original draft paper with John J. Donohue in 1999 specifically referred to the higher rates among blacks of both abortion (which currently is about five times higher than among non-Hispanic whites) and crime (blacks are currently incarcerated at a rate 7.2 times the non-Hispanic white per capita rate) as one of the reasons why legalizing abortion should have cut crime. The New York Times reported in 1999:

"Most of the reduction," Dr. Levitt and Dr. Donohue write, "appears to be attributable to higher rates of abortion by mothers whose children are most likely to be at risk for future crime." Teen-agers, unmarried women and black women, for example, have higher rates of abortion, the researchers note, and children born to mothers in these groups are statistically at higher risk for crime in adulthood.

Levitt took out the reference to the much higher abortion and crime rates of blacks when he published Freakonomics. Instead, it was all supposed to work by getting rid of "unwanted" fetuses, even though he admitted that legalization vastly increased the number of unwanted fetuses.

But let's get real. Last Spring, when Levitt was the toast of American intellectual life, everybody who was proclaiming his wonderfulness knew deep down that his abortion-crime theory was still based in large measure on aborting black fetuses, but nobody would come out and say it.

I was the only one who kept pointing out the new emperor of the bestseller lists had no empirical clothes, but nobody cared, because the unwritten message of Freakonomics -- no black, no crime, as Stalin might have said -- seemed so convincing.

But since you aren't supposed to discuss the higher black crime rate in public, our national immune defenses against bogus ideas couldn't resist Levitt's lie.

Have you noticed lately how, ever since New Orleans, America is knee-deep in lies? See, once you start denouncing people for telling the truth, you just have to lie and lie and lie some more. Every truth leads to more truths, but once you start down the path to lying, every lie means you need to lie again.

God, I am sick of lies.

My published articles are archived at -- Steve Sailer


Anonymous said...

william bennett?? from whitehouse?

better than al jourgenson anyway

Anonymous said...

go here to buy william bennett records