The Roe Effect

Since the publication of Freakonomics, a lot of readers have written to comment upon or ask about “the Roe Effect,” the theory put forth by James Taranto of The Wall Street Journal. The Roe Effect posits that among the many unintended consequences of Roe v. Wade was a significant change in voting patterns. Although Taranto himself acknowledges many nuances and wrinkles in his theory, it is well worth reading about. Here is a link to his latest article.

Anonymous

Less crime and more republicans, or more crime and fewer republicans. Hmmm. Tough Choice.

Anonymous

Taranto points out that "40 million abortions" have occurred since Roe. This is indisputable. But Taranto obviously does not understand his basic Freakonomics.

Most economists estimate that while over 1 million abortions occur annually, only 100 thousand children are not born as a result.
How is this possible?

Well, consider that abortion serves consumers by many different angles. True, abortion reduces the number of births. It is also used as birth control, as a health procedure, even as an insurance plan (insurance in the event the condom breaks, etc). In other words, abortion CAUSES more conceptions than would occur without any abortion. In the process, the "Roe" effect is realistically only 4 million unrealized citizens, not the 40 million that Taanto claims. 36 million of these 'missing' voters simply never would have existed, Roe or No.

No wonder "the results are mixed."

Read more...

Ken D.

Abortions occurred before Roe and would have continued to occur, legally and illegally, had Roe been decided the other way. How laws would have changed absent Roe is unknown. Calculating how birth rates would have been different, much less the demographics involved, can probably never be more than wild speculation.

Anonymous

True, abortion reduces the number of births. It is also used as birth control, as a health procedure, even as an insurance plan (insurance in the event the condom breaks, etc). In other words, abortion CAUSES more conceptions than would occur without any abortion. In the process, the "Roe" effect is realistically only 4 million unrealized citizens, not the 40 million that Taanto claims.

So 90% of the aborted fetuses would not have been conceived before Roe? And 90% of abortions are the result of people choosing not to use protection or choosing to have sex with protection when they normally wouldn't because they have the insurance of an abortion?

Anonymous

This is the flipside to abortions resulting in less crime.

seamus

It is a statement of fact, not a moral judgment, to observe that every pregnancy aborted today results in one fewer eligible voter 18 years from now.

This assumes that the parent would have had just as many children after the abortion as they would have had if the abortion were not available.

Imagine two childless women. Both eventually want a child, and both get pregnant unexpectedly. One has an abortion, the other doesn't.

The one who has an abortion may still want to become a mom, and they may have a child or more children in the future. The one who had a baby instead of an abortion is now a mother, and may not feel the need to have any more kids. If this is the case, both women will eventually have one child, so an equal number of voters will be created, whether abortion is available or not.

You must also weigh the emotional impact of abortion, as well as how the accessibility of abortion impacts the sexual and social behavior of women.

Not quite a "statement of fact" that one abortion = one fewer voter, is it?

Read more...

Anonymous

You could call it time shifting, in that a significant percentage of the potential mothers were able to time shift the potential birth to a point in their lives where they felt they were able to provide the kind of home life they had always wished for their child.

Unfortunately, the efforts of the "pro-life" movement has an unfortunate and unrecognized reality - banning abortion would not produce the desired outcome - a good chance at a constructive home for every child.

Carl Johnson

9:21 Anon: you're ignoring the fact that the "desired outcome" for most abortion opponents isn't merely that children have constructive homes, but that fetuses, which they regard as equivalent to people, are not 'murdered.'

At the risk of invoking Godwin's Law, I submit that though we all would like a good economy, few of us are willing to follow the economic path of Germany in the 1930s. Why? Because their good economy was built on the back of slave labor and mass murder. Similarly, to those who consider abortion to be murder, no good result can justify using the means of 'murdering the innocent.'

Now, I'm not going to stake a claim one way or the other in the debate, but it seems to me that if you want debate, you have to understand your opponent's position.

Laura

The statistically probable criminals who Levitt and Dubner argue weren't born because of Roe v. Wade were also statistically unlikely to vote, and while Taranto's census argument deals with representation by population, there's a lot of noise in that equation.

Anonymous

It is a statement of fact, not a moral judgment, to observe that every pregnancy aborted today results in one fewer eligible voter 18 years from now.

Actually, that is true, regardless of whether the woman goes on to have a child later. Taranto has addressed that issue before. The deliberate "18 years from now" is what makes the statement true.

Time shifting children dampens the overall effect in the long run. By how much is debatable. For example, consider a teenage mother who has an abortion today. Perhaps she will go on to get married ten years later and have three kids. Maybe if she hadn't had the abortion she would have become a teenage single mom and never have any more kids.

Craven Moorehead

Yes, anti-abortionists seem to forget the days of the coathanger and the backyard abortion. There were many unwanted pregnancies caused by unprotected sex, despite the enormous risks involved in getting a backyard abortion. Sex, particularly unprotected skin-on-skin sex, is a very powerful human motivation. Look how many gay men in Western countries, and heterosexuals in the third world, got infected with AIDS long after the risks of unprotected sex were known, where the risk is not just unwanted pregnancies but a lingering and highly unpleasant death from cancer or pneumonia. How would you get accurate figures for pre-Roe unwanted pregnancies terminated by backyard abortions?

A while ago I read a book about a bohemian intellectual clique in Australia called the Sydney Push, who disdained condoms simply because they were considered too proletarian. They still had lots of unprotected sex. The men, who were often impecunious, were then obliged by the unwritten social rules of their clique to raise the substantial funds necessary to procure a backyard abortion. This is only anecdotal evidence of a highly non-representative group, but I wonder if black American youths in the 60's & 70's had much more sexual self-control than Sydney bohemian intellectuals.

Read more...

Anonymous

Again, abortion can be thought of in terms of supply and demand. The price of abortion decreases, both financially and morally, as a result of Roe. When prices decreases, quantity demanded increased. Thus it is not unreasonable to say that 40 million is an overstatement. Most economists estimate around 100K pre-Roe abortions; 1 million post-Roe. Therefore, 4 million children. Consider that most of these abortions are concentrated in "blue states," over the course of 30 years, and the political effects are probably negligible.

Anonymous

Roe did not reduce voters. Levitt says that many of the unborn children would have been criminals. Most of these criminals would have been felons, who do not have the right to vote. So the Roe effect eliminated voters without voting rights. In other words, the Roe effect was most likely not political.

Anonymous

"Levitt says that many of the unborn children would have been criminals"

and many wouldn't

Anonymous

"Since the parties split on abortion, the GOP has won five of seven presidential elections, and no Democrat has had a majority of the popular vote."

I know this seems awfully nit-picky, but if my memory serves me correctly, Al Gore did indeed have a majority of the popular vote in the 2000 election.

And as for my two cents, I very much agree that I would rather have (slightly) more Republicans around if it also means that there are fewer botched/unsafe abortions in a pro-life environment.

Elton

Alas, while Gore did win the most votes in 2000, because he still had less than half the total, he did not have a majority. 48.38%, to be more precise.

Andrew

The article claims that the proportion of Republican voters is growing because the Roe effect has reduced the Democrats by suppressing their fertility rates. This ignores the fact that even after Roe v Wade, traditional Democratics (minorities and the poor) have higher fertility rates than the general population.

Anonymous

This ignores the fact that even after Roe v Wade, traditional Democratics (minorities and the poor) have higher fertility rates than the general population.

Taranto does not ignore this fact. For example, "The Roe effect does not predict that younger voters will be more apt to vote Republican than older ones, only than they otherwise would be."

The question to investigate is whether the Roe effect exists and is large enough to be noticed next to other demographic trends pushing the political leanings of the population one way or the other.

Anonymous

If you want to know how many women die from backstreet and abortions, then do a study of the hospitals in the Philippines, where abortion is illegal. When I was an intern rotating in OB-Gyn in the Philippine General Hospita in 1984l, it seemed like 20% of the admissions to the OB-Gyn wards were women in sepsis who had undergone criminal abortions.

Anonymous

I just finished Freakonomics and I have been perplexed / fascinated with the Roe Effect. I have been reading the counterattacks to get a better sense of the data and what has been most fascinating is that no one is really looking at that the abortion rate data. The critics usually point to youth (age 14-17) violence during the late 80s / early 90s as inconsistent with Levitt's hypothesis. However, cursory research reveals that the abortion rate did not reach it's peak level (hence - peak future impact) until around 1980 - therefore the expected significant drop in age 14-17 violent crime would begin (and it does) in the 1994 - 1997 window of time. You check the data and prove it to yourself.