The Abortion Debate Gets Heated

I guess that it would have been helpful if Ms. Marjorie Signer had actually explained why my research showing that the liberalization of abortion rules increased crime contained "many unfounded, fallacious, racist, and confused assumptions."

"Lott offers so many unfounded, fallacious, racist, and confused assumptions that his overall opinions are useless," said Marjorie Signer, spokeswoman for the Religious Coalition for Reproductive Rights, in a statement sent to Cybercast News Service.

"To speculate that abortion is responsible for various and sundry social trends is irresponsible and misses critical points," she said.

"Abortion has always been an option that women at all socioeconomic levels have considered, and many abortions are performed for medical reasons. Trying to correlate abortion with social trends smacks of pseudo-science and shows little understanding of women's lives and decisions," Signer added. . . . .

John Donohue also refused to comment on the actual research, but was reduced to saying this: "I am a social scientist, however, so Lott's behavior has in my mind, put him outside the bounds of scientific discourse." That will certainly settle the debate, though I can understand why Donohue is reluctant to actually discuss the facts on this issue. Since John Whitley co-authored this paper with me, does Donohue's statement tar Whitley also?

Donohue also doesn't comment on recent issues regarding his co-author Levitt.

New Research that Abortion Increases Violent Crime Gets Attention

From today's Chicago Sun Times:

A high-profile economist is challenging the conclusion in the best-selling book Freakonomics by University of Chicago professor Steven D. Levitt that the legalization of abortion in the early 1970s led to a major drop in murder and other violent crimes a generation later.

John R. Lott Jr., a former U. of C. economist now teaching in New York, says the Supreme Court's 1973 Roe v. Wade decision actually caused violent crime to rise.

Lott and fellow researcher John Whitley plan to publish a paper in October in Economic Inquiry that questions Levitt's research on abortion and crime.

Lott and Levitt already were feuding over Lott's charge that Levitt had defamed him in Freakonomics.

In their new paper, Lott and Whitley say that legalization of abortion prompted a cultural change that increased the number of children born out of wedlock. Those children of unwed mothers caused murders to rise by more than 700 cases in 1998 alone, saddling the public with more than $3.3 billion in "victimization costs," the paper says.

On the other hand, Levitt's research found that Roe v. Wade resulted in a savings of $30 billion a year that crime would have cost the public.


More unwed mothers

His Freakonomics, co-authored by Stephen Dubner and published last year, says legalized abortion led to a large drop in murder and other violent crime in the late 1980s and early '90s, and continues to reduce crime.

The book suggests that if the aborted fetuses had instead been born, they would have become adults more likely to commit crimes because they were unwanted by their mothers.

To illustrate the point, the book says the five states that allowed abortion three years before Roe vs. Wade saw major declines in violent crime between 1988 and 1994 -- earlier than the other states.

But Lott says the Levitt study did not fully consider the increase of children born out of wedlock. His theory is that with the option of abortion, women became more likely to have premarital sex, but then had their babies and raised them as single parents.

Children born out of wedlock have had smaller investments in "human capital" by their parents and are more likely to get into trouble when they grow older, Lott says.

On average, his paper says, about 5 percent of whites were born out of wedlock from 1965 to 1969, rising two decades later to 16 percent. For blacks, the figure rose from about 35 percent to about 62 percent, the paper says.

Before legalized abortion, more than 70 percent of children born out of wedlock ended up in families with a father, but the fraction fell to 44 percent in 1984, according to the paper. . . . .

If you want to read the research, you can find it here. THe newspaper article says that I concede that abortion through its effect on "unwanted" births slightly reduces violent crime, but what I believe that I said is that it is possible. The net effect however is abortion increases violent crime.

Will Donohue and Levitt predict that a massive new crime wave is coming?

Guttmacher Institute, associated with Planned Parenthood, claims that: "The rate of unintended births — unintended pregnancies carried to term — rose by 44 percent among poor women from 1994 to 2001, but declined by 8 percent for wealthier women." If true, I assume that those few who still believe that abortion massively reduces violent crime rates after the errors were discovered in the original research will be predicting that we will soon see a big increase in violent crime rates. If up to 80 percent of the changes in murder rates can be explained by abortion as has been argued, those who believe this relationship must be worried we could be in for big increases in murder. (Of course, the abortion is more likely to slightly increase crime.) People born in 1994 will become teenagers starting next year. Presumably, proponents of this theory will start warning people very soon.

The Economist Magazine on Steve Levitt and Abortion

Of course, lots of people have always thought Mr Levitt was in the wrong. Even if abortion cuts crime, it is still immoral, they fulminate. But this is largely beside the point: Mr Levitt's research does not take a position on abortion's social virtues, but aims merely to uncover its societal effects. Besides, for someone of Mr Levitt's iconoclasm and ingenuity, technical ineptitude is a much graver charge than moral turpitude. To be politically incorrect is one thing; to be simply incorrect quite another.

But what is most amazing is that despite all the statistical significance being eliminated from their panel data set being eliminated when the results are done Donohue and Levitt's way with the the data they want to use, John Donohue now says what they wanted to do:

It may be asking too much of the numbers to convince everybody. “The debate over abortion and crime will not be resolved within the parameters of our paper,” says Mr Donohue. He thinks the arrest figures are “muddy” and the state population data “sloppy”. Combining the two generates so much noise, it is hard for the statistical tests to hear anything. Ted Joyce, a professor at Baruch College (part of the City University of New York), who has had his own methodological disagreements with Messrs Donohue and Levitt, also thinks the debate is stretching the data too far. He points out that if you add controls for 50 states and 12 years—as Messrs Foote and Goetz do, and as Messrs Donohue and Levitt meant to do—you are, in effect, holding another 600 things constant. This robs the data of most of their variety, and of much of their ability to explain anything.

There was no warning that these were the wrong tests at all in either of their previous papers or in Levitt's book. Levitt now refers to the “collage of evidence," but the panel data was the only test that really amounted to anything worthwhile. The rest of the data was merely cross sectional or time series.

Incorrect claims about the data in More Guns, Less Crime

I have received an email comparing my work to that of Steve Levitt's regarding claimed coding errors in More Guns, Less Crime. Of course, this discussion is not very accurate.

1) There are no coding errors in the data used in “More Guns, Less Crime.” The book used crime data from 1977 to 1996, and as far as I know, no academics have claimed that there were any coding errors in that data. An interesting and useful Stanford Law Review article by Plassmann and Whitley added an additional four years to the data and the problem arose in this additional data. Overall, less than 200 cells out of 7.5 million cells were accidentally left blank. More important, the results that Plassmann and Whitley noted were the results which they thought were the correct ones were not affected at all by this minor change in the data.

2) I had put the data set on my website as a favor to Plassmann because it was very large and he did not have the ability to put it up for people to download. The data was corrected as soon as the missing cells were discovered and it was available for people to download. A note was added to the site to alert people to the correction. While I had helped them out on their paper, I let Plassmann and Whitley make their own statements about their results. My website reported the regressions in the same way that I had done them in all my previous estimates.

3) In statements here and here, I have discussed what was involved in Levitt's errors. The regressions where they left out the fixed effects were essentially the only ones that were at all useful in testing their hypothesis because they were the only ones that didn't require large degrees of aggregation to get to their "effective abortion rate" that were related to total murder rates in a state. The work that I did with Whitley directly related the age of the murder to the number of abortions when that murderer was born.

Thanks to Bob Thomas for his email.

Levitt's Abortion "Disappeared" When "Programming" Error Fixed

Levitt's Abortion results disappeared when programming Error Fixed

"The Boston Fed's Mr. Foote says he spotted a missing formula in the programming oversight of Mr. Levitt's original research. He argues that the programming oversight made it difficult to pick up other factors that might have influenced crime rates during the 1980s and 1990s . . . . He also argues that Levitt should have counted arrests on a per-capita basis. Instead, he coundted overall arrests. After he adjusted for both factors, Mr. Foote says, the abortion effect disappeared.

Christopher L. Foote and Christopher F. Goetz's paper can be found here. Personally, I think calling this a "programming oversight" is being much too nice (See my post here to see an example of what you would have to miss to not notice whether you have used fixed effects). More importantly everyone who works with panel data knows that you use fixed effects.

My own work concentrated on murder rates, but I also included fixed effects. Donohue and Levitt never provided us with all their data or their regressions and would never answer any questions that we had so I just assumed that they had included fixed effects from the beginning. It would have been nice if they had provided us with this same information years ago.

My paper with John Whitley is available here.
A letter that I had in the WSJ is available here.
James Q. Wilson's review of this debate is available here.

UPDATE:

See also Steve Sailer's past work on this topic.
See also Steve Sailer's compliation of what others are saying here.

UPDATE II (I have been asked about inaccurate claims regarding my own research in More Guns, Less Crime):

See this for a discussion about claimed errors in More Guns, Less Crime

Revisiting John Donohue's cancelling of our gun debate last fall

1) It was very disappointing that John Donohue backed out of our debate on guns last December with just two days to go. I had looked forward to the debate, and obviously had my plane tickets well in advance. It was a date that Donohue had picked (I had said that I could do all but one day over the end of November through December and I would have gone beyond December if necessary), and the first email that I had confirming the event was received in September. I ended up speaking at Chicago on another topic so that a later debate could be attempted to be set up again with Donohue.

2) Unfortunately, the second planned debate was also cancelled. The debate was rescheduled for April 13th this year, and I made sure to reconfirm it because of the previous cancelation. The student at Chicago who set up the debate said that even though he had confirmed the debate with me multiple times and even though we had taken a date that I was told that Donohue wanted, the claim is that the debate somehow hadn't been completely confirmed with Donohue. (You would think that if the previous cancelation was an accident, Donohue would have been careful to double check this time just in case there had been any possible misunderstandings, but no one is claiming that he checked anything.) Attempts were made to get Donohue's coauthor Steve Levitt to substitute for Donohue and debate either guns or abortion claims were unsuccessful. From an email that the organizer sent me after the second debate was cancelled: "You have been the most patient and flexible speaker I could imagine . . . ."

3) In conversations with Joe Cascio (the speaker coordinator for the UC Fed Soc Chapter), I had said several times if we were going to schedule a third try at a debate, we should do it on the abortion and crime claims, especially given the amount of attention that this poorly done empirical claim had gotten. Despite Donohue claiming that the debate would actually take place this time, he was unwilling to debate this hot topic. Steve Levitt was unable to debate (he was debating the topic but only with people who didn't seem to know much about the issue) and Dubner, Levitt's coauthor, was unwilling to debate the abortion claims. The University of Chicago Federalist Society was again willing to have me give another talk that wasn't in a debate format, but they insisted that I remove the statements regarding the canceled debates from my website. (They simply viewed my posts about the April 13th event being critical of the Federalist Society (I disagree), and the earlier post that they wanted removed because they didn't want to be involved in a debate about a debate.) I told them that I would correct anything that they told me was wrong, but I wasn't going to remove the postings. If people set up debates and back out at the last moment, they should be held accountable.


On the substance of Donohue's work on guns, I would direct people to this paper by Plassmann and Whitley. There are two straightforward points that they make. 1) That the graphs at the beginning of the Ayres and Donohue work are very misleading because they don't make it clear that the sample of states is changing. The crime rates clear fall for a decade and a half and the sudden increase isn't real for the remaining states. It just looks like an increase because the couple of rural states that remain did not experience the decline in murder rates that more urban states experienced after right-to-carry laws were passed. 2) The second claim has to do with them fitting an intercept and line to the data and then limiting their reported impact on crime to only five years. Often there is no problem with this, but as Plassmann and Whitley clearly show it doesn't fit in this case. The Ayres and Donohue approach implies that crime initially increase (despite the fact that the year by year data discussed in point (1) doesn't, and it is just an artifact of their estimates over predicting in the early years because they are fitting this straight line with an intercept shift to crime that is falling at an increasing rate. But to compound the problem, they only discuss what the estimates mean for the first five years. Even with this approach if they had picked the sixth year it would have reversed their claims.

I also think that it is pretty clear why they find it difficult to debate either guns or abortion. Steve Landsburg was very nice to suggest in April that Levitt and I have a discussion at Rochester this fall, but I haven't heard anything back. I immediately wrote back saying that I was interested. Landsburg wrote in April that:

"Our department would like to arrange, sometime in the fall, to have you and Levitt visit simultaneously and give back-to-back workshops on guns, abortion or both. We want to hear your criticisms of each others' work and your responses to those criticisms. We *don't* want this to turn into a debate; we want it to be an academic seminar with all the usual rules of logic and evidence. The presenter is presumed to have the floor; there is an expectation of periodic interruptions from the audience; but there's also someone in charge to cut off questioning if it starts to take over the seminar. We would most assuredly *not* advertise this event to the general public, because we don't want an audience of advocates for any point of view. Ideally, we'd have you both here for two days, with you presenting on one day and Levitt presenting on the other. We are open to alternative suggestions. We'd pay you both an honorarium of some size to be determined, though surely far less than you deserve. We'll do what we can to make it attractive, though."

I am posting this because of an email that I received asking questions about Donohue's cancelation last fall.

John Donohue's inaccurate claims

The fact that John Donohue's own research does not show any statistically significant increase in violent crime after right-to-carry laws are passed apparently doesn't stop him from claiming that to the media. A nice discussion of Ayres and Donohue's results can be found here. The estimates range from not finding any affect of the law to those finding large drops in violent crime rates. That doesn't stop Donohue, though. Of course, it would be nice if he decided to show up for scheduled debates (see the posts for 11-30-04 and 4-7-05, an additional attempt to set up a debate is discussed here: 5-7-05).

Thanks to Alphecca for pointing to this article.

Abortion and Crime "debate" at the University of Chicago Scheduled for May 25th

The University of Chicago Federalist Society has tried for a third time to set up a debate between myself and John Donohue. Since the last two debates on the issue of guns at the University of Chicago were canceled at the last moment with Donohue withdrawing from one debate with just 2 days to go, I thought that we might have more luck scheduling a debate on another topic that is getting a lot of attention these days: abortion and crime. Donohue and Steve Levitt were the coauthors on a paper that got a lot of attention on this issue and Levitt as also recently coauthored a book with Steve Dubner that again goes over the issue. All three were asked to pick a time to debate the issue, but even though Donohue is free on the 25th and despite all the attention currently being given to the abortion research, none of them were willing to debate their work on abortion with me. (I think that I know why.) I will still be presenting on the 25th with the hope that Donohue will change his mind and defend his research.

Further discussions about John Donohue, including his backing out of one of the two previously scheduled debates on guns, are here.

Debate with John Donohue is again canceled, this time with only six days to go

It seems like we have been through this before. In December, John Donohue cancelled our debate that was scheduled at the University of Chicago for December 2 on November 30th. After what had happened in the Fall, I double checked to make sure everything was confirmed before I turned down another talk that I had the chance to give, and I was assured that the event was set and that I should get my plane ticket. This time our debate that was scheduled at the University of Chicago on April 13th and canceled with just six days to go.

Further discussion of Ian Ayres & John Donohue are here.

Update: Here is an email that I received from Joe Cascio:


On Apr 7, 2005, at 3:08 PM, wrote:

Dear Dr. Lott,

I know that you are not going to be happy about this, but unfortunately, Professor Donohue cannot make it on the 13th of April. I am currently trying to work out the 21st (a Thursday with him). This time it really is more my fault. If you think that you could make it on the 21st (if that works out--I will let you know when Professor Donohue has his plane reservations in his hands), then we, of course, will pay for your current ticket and/or any changes that you have to make (or for both your current ticket and a new one if it is unalterable). If, on the other hand, you don't think you might be able to make it on that date or any other later in the quarter, I completely understand (obviously, we will still take care of your current reservation).

You have been the most patient and flexible speaker I could imagine and I feel as though you have been treated rather shabbily. I am really sorry about that. I still really hope that this event works out, if it does, I really feel like it might the best of our entire year. However, if it does not, I have nevertheless enjoyed working with you and I still really appreciate everything that you did for us in the fall at the last minute as well. Thank you very much, and I regret that I, once again, must convey my apologies.

Joe

Notes from me: Bold markings added. Before I bought my plane ticket, I had asked Joe in an email well before this one whether everything had been set up and he confirmed that they had. Unlike the first event being cancelled by Donohue with just two days to go, I have my suspicions about what happened here and I suspect that Joe is falling on his sword to provide cover. I was so mad that Donohue would cancel the first event the way he did, I have my suspicions that Joe is trying to keep things undercontrol by taking the blame "this time."

Donohue withdraws from Thursday's University of Chicago debate on Guns

Disappointingly, John Donohue has at the last minute withdrawn from our scheduled debate on Thursday (see note for 11/29 below). I will still give a talk, though I will instead discuss the changing judicial confirmation process. Here was the advertising of the debate in the Chicago Daily Law Bulletin:


Chicago Daily Law Bulletin

November 26, 2004, Friday

SECTION: Pg. 3

LENGTH: 921 words

HEADLINE: A Web of plans at Illinois schools

BYLINE: JERRY CRIMMINS

. . .

University of Chicago Law School

- At 12:15 p.m. Thursday, Dec. 2, at the law school, the Federalist Society will present a debate entitled, "Do More Guns Result in Less Crime?" John Lott, resident scholar at the American Enterprise Institute for Public Policy Research and author of " More Guns, Less Crime: Understanding Crime and Gun Control Laws," will debate Yale Law School Professor John Donohue, author of "Shooting Down the ' More Guns, Less Crime' Hypothesis."

UPDATE: When I gave my talk at Chicago, I asked the students if they knew what had happened, but no one apparently understood why Donohue really backed out of the debate just a couple of days before the event. The debate had been set up months in advance. I had received the same emails that Donohue and Bernard Harcourt, the moderator, received confirming the event more than a month in advance. The emails were cc'd to my address, Donohue's, and Harcourt's at the same time. If there had been some misunderstanding, Donohue should have notified everyone much earlier than two days before the event.

UPDATE 2: Here are some of my very early email correspondance with Joe Cascio, with the University of Chicago Federalist Society, from the just the first half of September. The claim has been made that the event wasn't confirmed until November 29th, but I have it first confirmed in September and then several emails after that checking on the event (including just asking if people had any questions, confirming flights, and setting up the ground rules for the debate).

From: jcascio@uchicago.edu
Date: September 13, 2004 9:29:10 PM EDT
To: John Lott
Subject: Re: Speaking to the Federalist Society w/John Donohue


Dear Dr. Lott,

Great. The topic is "Do more guns (i.e. less restrictive gun laws) mean less crime?" As far as format, anything mutually agreeable to you and Professor Donohue would be fine, but we had in mind a debate with fifteen minutes for you to speak, then fifteen minutes for Professor Donohue, then ten minutes for you, then ten minutes for him, followed by ten minutes of student questions. Please let me know if that's acceptable to you. In any case, thank you very much for accepting our invitation and I'm really looking forward to this event. Please let me know if you will need a hotel room the night of the first.

Joe

Quoting John Lott :



That should be fine. What exactly is the topic? What are the ground
rules?

On Sep 9, 2004, at 6:40 PM, jcascio@uchicago.edu wrote:


Dear Dr. Lott,

I'm sorry I didn't respond sooner to your message. I was out of town for a bit, then I checked with Donohue and Harcourt to see what day that you had said you would be available would work for them. How about December 2? We could have the debate at lunch and if you need to stay the night before we could put you up at the Omni downtown. In any case, please let me know if this can work out and thanks for your patience.

Sincerely,

Joe Cascio



Debate with John Donohue is again canceled, this time with only six days to go.

Further discussion of Ian Ayres & John Donohue