Malcolm Gladwell on the Freakonomics Paradox

Malcolm Gladwell, author of The Tipping Point, Blink, and, over the years, a collection of startlingly good New Yorker articles, has addressed on his blog the question of why he endorsed Freakonomics (by writing a blurb before it was published) even though its explanation of the 1990’s crime drop dismissed as a cause the “broken windows” theory of law enforcement put forth by James Q. Wilson and George Kelling, put into practice in New York by Rudy Giuliani and Bill Bratton, and put into the public’s eye by Malcolm himself first in a New Yorker article and then in The Tipping Point. (Malcolm and Steve Levitt held a friendly debate on this very issue many months ago.) As usual, Malcolm’s writing is well-considered and entertaining. One thing to consider, however: the theory put forth in Freakonomics examined why crime had fallen all over the country, not just in New York, and one of the many arguments against “broken windows” as a major cause was the fact that such innovative policing wasn’t being practiced elsewhere — and yet crime was falling in those places as well. A smaller point to also consider: Gladwell left out one other major reason that, according to Levitt’s research, crime did begin to fall in the 1990’s: the waning of the violent crack trade. (Thanks to Darren Rovell for pointing out the Gladwell blog.)

?? » Blog Archive » Gladwell vs. Freako guys on Broken Windows

[...] Gladwell Freakos respond Gladwell’s response [...]


I've been in law enforcement for over 11 years and worked through the highs of crime and now the lows of crime. I have always been taught the broken windows theory, but have always had doubts. One of the premises of the theory is the Kansas City experiment which was flawed from the outset. I am working on a theory that compliments yours, and includes the concept that having more police officers and smaller patrol areas reduces crime. In a nutshell, compare the size of small cities and their police forces to large cities and their forces. Break the larger cities down to equalize the smaller ones and so far I see a huge reduction in crime between the two. So my theory, if correct, means that if you divide cities into smaller beats, but no larger that 40,000 citizens and place an average of 60 street cops (not including 12 additional officers in specialized assignments), crime will drop.



An economist who hates math and a journalist who strives after the eccentric and unknown, Steven D. Levitt and Stephen J. Dubner formed a surprisingly efficient team, despite their seemingly unrelated backgrounds, and composed the intriguing work known as Freakonomics. Within the pages of this refreshingly controversial book, the co-authors' stated goal was the simple application of the principles of economics to our everyday lives. However, their ideas and their attention grabbing methods of presentation fundamentally revolutionize one's concepts of economics, transforming it from an arguably dull subject to one that is capable of mesmerizing the most ardent critics.

The charm of Freakonomics is that after completing the book, despite its array of statistics presented to illustrate interesting facts about a variety of events, there is an invaluable underlying premise to inspire a similar interest in the readers to actively seek and understand the little things in life, instead of taking them for granted. I am supportive of this point, because the book fosters a sense of curiosity which, in the end, gives birth to a heightened sense of appreciation of life and the vastness of the secrets of the world. By propagating such intense desires to know and to understand, I believe that the success of Freakonomics lies not in its ability to convince, as much as its tendency to produce inquiring young minds for the future.



I've left a comment on Gladwell's blog relating to that point.

My question is- what is the Socio-economic background of those who used the Pill from the mid 60s, compared to those who began using abortion after Roe v Wade?

The argument in Freakonomics is not that abortion itself for the sake of contraception reduced crime, but that the ability of those who required this emergency, after-the-fact contraception, to prevent an unwanted pregnancy reduced the amount of unwanted/unplanned children who would be born to parent(s) from lower socio-economic groups.

If this group were not using the Pill, then their point remains strong. If this group were using the Pill (or other forms of contraception), they would not have required abortion.

But I'd leave that conclusion for the statistics (I base it on nothing; I know nothing particular about this subject, and speak only as an interested reader; I'm not in the US).



"My question is- what is the Socio-economic background of those who used the Pill from the mid 60s, compared to those who began using abortion after Roe v Wade?"

I'd say the differences are fairly self-evident. Sexually active women of all backgrounds who didn't want to become pregnant used/use the Pill, while abortions are sought, for the most part, by women who accidentally became pregnant. It's not difficult to translate this into socioeconomic differences.


How does Freakonomics respond to this salvo:
"the biggest drop in fertility in the U.S. came with the advent of the Pill in the mid-1960's. The Pill allowed lots of women who would otherwise have become pregnant not to become pregnant because they were poor, or didn't want a child, or lived in an environment where it was hard to raise children. But the fertility drop caused by the Pill didn't lead to a decrease in crime eighteen years later. In fact, that generation saw a massive increase in crime. The advent of abortion in the early 1970's, meanwhile, caused a far, far smaller drop in U.S. fertility but-Levitt argues-that drop is consistent with a fall in crime."


In "Wealth of Nations" Adam Smith pointed out that poor people tended to have more children than wealthier people due to higher mortality rates. Is there a negative correlation between income and the number of children in an era of much better health care along with contraception?

The part of "The Pill allowed lots of women who would otherwise have become pregnant not to become pregnant." is true. The problem of making an argument always begin with "because..."

I would also disagree with
"I'd say the differences are fairly self-evident. Sexually active women of all backgrounds who didn't want to become pregnant used/use the Pill, while abortions are sought, for the most part, by women who accidentally became pregnant. It's not difficult to translate this into socioeconomic differences."
Many women find themselves pregnant while on the pill due to the use of antibiotics; a drug interation that they are only informed of once it is too late.
No manner of contraception is 100%. By "socioeconomic differences" I believe you are saying "poor and uneducated women are not bright enough to understand reproduction or to avail themselves of the low and no cost contraception available" I wonder if age is a variable that plays a more significant role. My only point being that there are many variables that determine a phenomenon and I question how much of anything is "self evident"



What I'm saying is, the chapter in Freakonomics is based on the idea that the kids who would have gone on to become criminals were among those who were being aborted. For this to be true, it must be accepted that it is possible to predict or derive a model of person likely to become a criminal. It should also be possible, on the same basis, to test Gladwell's point about the Pill be looking at the data of the peope who were using it pre-Roe v Wade.

I, in no way, would equate any reference of socieconomic difference to implying difference in intelligence or natural abilty. Any points I make here are made in the context of Freakonomics, Gladwell's The Tipping Point and the discussion currently ongoing.


My response was to prosa. But, I do think it may be muddled a bit to talk of women who used the pill prior to Roe v. Wade compared to women who have abortions after the ruling, as if these are distinct groups. I question the categorization of abortion as a choice of contraception over time, like the pill, versus a unique event.

I wonder if it might be the case that Planned Parenthood, by providing low cost or no cost contraception, including contraceptive counseling after abortion, may have contributed to
the decline in the "at risk" children and thus the impact on crime is not only due to aborted potential criminals, but criminals that weren't born due to expanded access to contraception after abortion became legal nationally.

Isn't there going to be data to test the crime/abortion issue soon in Romania, with the legalization 15 years ago or so?


I think it fairly obvious that women who get abortions tend to be of a lower socioeconomic class than those who use the Pill (note that I say *tend,* there certainly are exceptions). Even if it can be obtained free of charge, using the Pill requires some advance planning, it being prescription-only, and it then has to be taken according to a schedule rather than being popped indiscriminately. I would submit that women in lower socioeconomic classes tend to be less diligent in terms of planning things in advance and then following schedules. Again, please note my use of *tend.*

John S.

I find this argument to be very interesting from a statistical standpoint. There are a lot of people who claim that Roe v. Wade did not do a lot to change the rate of abortion. See this book, for example:
Figure 4 on page 210 claims that in 1962, Cook County Hospital treated 5000 women (mostly poor and black) for abortion-related consequences. That would mean the abortion rate there in 1962 was close to what it is now among black women (55 per 100 live births). Of course, that book is advocacy, so I take the figures with a large grain of salt.


"I think it fairly obvious that women who get abortions tend to be of a lower socioeconomic class than those who use the Pill."

Again I think it a mistake to identify these as distinct groups. I do not believe that abortion is used as a method of contraception over time.

Poor women can't remember to take a pill? The hardship and expense of an abortion would seem much more costly to a woman than to keep on a schedule.

John S.

"In 2000, 21 out of every 1,000 women of reproductive age had an abortion. Women who are aged 18-29, unmarried, black or Hispanic, or economically disadvantaged-including those on Medicaid-have higher abortion rates."

"Almost all women are at risk for unintended pregnancy throughout their reproductive years. However, adolescents, formerly married women, and women of low socioeconomic status are at greater risk for contraceptive nonuse and for contraceptive failure. ... The level of nonuse is greatest among younger women (21% of women under 20 years old are not using contraception) and poor women (19%)."


I'm still a little baffled as to how anything in Freakonomics shows that the main cause of developing a criminal mind is being unwanted. It may be true that these counterfactual children would have been unwanted. But that does nothing to show that being unwanted is the cause that leads people to commit crimes. Secondly, Dubner's reply misses the point of Gladwell's post. Gladwell claims that an increase in the number of police on the streets is just what "Broken Windows" policing is. Simply because other cities than NYC added police officers on the beat but didn't call the increase "broken windows" policing does not mean these other cities didn't impliment broken windows style improvements. Content matters more than labels.

nick davis

Tipping Point cs. Freakonomics

Malcolm Gladwell has been going back and forth with The Freakonomics Guys about the integration of the Broken Windows theory from The Tipping Point with the Abortion-reduced crime theory in Freakonomics.
Based on Gladwell's argument, I see no discong...


Does that NIH cited article say what the difference between poor women (19% are non users of contraception) versus higher income? What percent of higher income are non users?

I couldn't understand the numbers from the summary as it seemed to mush different descriptive statistics together. I took it to mean that 19% of poor women don't use contraception. But then it says "The remaining 8% of current nonusers tend to be either between methods or between relationships." So do the numbers reprensent a percent of the group of nonusers? The key word that makes me unsure of the meaning of the stats is the "remaining..." indicating these are percents of the total of nonusers. The actual article would clarify this but I'm not a subscriber.

Why would it be the case that " Separated, divorced, or widowed women have the highest contraceptive failure rates (26%) across all age groups."
And, "21% of women under 200% of poverty have contraceptive failures compared with 10% of higher income women." Would it perhaps be the case that higher income women are older and less fertile and less likely to be impregnated regrdless of contraceptive status, or is misuse of contraception higher among poor women? Or, do higher income women have less sex than poorer women? Controlling for age/fertility and number of sexual encounters would seem to be needed to have faith in those numbers (and again, may be included in the full article.)

It seems to me that there is a lot that goes into descriptive statistics that is vitally important before moving to the inferential.



Malcom Gladwell and Freakanomics

The more you look at blogs, the more you begin to realize how they are changing the way that people are finding information, as well as interacting in ways never before seen. Take for instance Malcom Gladwell. He's written two

Plugs » Blog Archive » The Freakin’ Point

[...] If you haven’t checked out our Blog of the Week — make that Blog of the Month — yet, what’s the problem, bub?! Currently there’s an intellectual “influencer” feud going on between Malcolm and the good fellas behind Freakonomics about the roots of the urban crime decreases of the 1990s. Do we fully understand the sociological ramifications of Gladwell’s “Broken Windows” theory or Freakonomics’ abortion-heavy theory? Do we care? The important thing is that we just feel smarter for having read both sites and cooler for linking to them here. [...]

e pur si muove » Blog Archive »

[...] Malcolm Gladwell enters into a discussion with Freakonomics authors Levitt and Dubner on the anomaly of New York City crime statistics. (Freakonomics made headlines for their economic, rather than primarily sociological, analysis for the general decline in crime in the United States over the last several decades.) Read Levitt’s and Dubner’s reply, and Gladwell’s counterpoint. [...]

Wordpress » Blog Archive » Levitt and Dubner respond

[...] [...]