What Do Declining Abortion Rates Mean for Crime in the Future?

The abortion rate in the United States is at a thirty year low — though even with the decline, we are still talking about a large number of abortions in absolute terms, or 1.2 million per year. To put this number into perspective, there are about 4 million births per year in the U.S.

John Donohue and I have argued that the legalization of abortion in the 1970s reduced crime in the 1990s. The logic is simple: unwanted children have an increased risk of growing up to be criminals, and legalized abortion reduces the number of unwanted children. Consequently, legalized abortion lowers crime in the future.

So what does the steady decline in abortions performed in recent years predict for future crime patterns? The answer is not obvious, because it depends on why abortion rates are falling, and I’m not sure we know the answer to that question.

If abortion rates are falling because it has become harder or more costly to get an abortion, then a falling abortion rate is bad news for crime. As the “price” (whether in monetary terms, social stigma, having to travel a long distance, etc.) rises, women who otherwise would have sought an abortion will not get one. This suggests that more unwanted children are being born, and thus crime rates may rise in the future.

On the other hand, there are other reasons why the number of abortions might fall, and none of these have dire crime implications. For instance, because abortion has been legalized since the 1970s, there may be fewer women today who are seeking abortions — the women who might have been at highest risk for unwanted pregnancies today may never have been born.

A second scenario in which low abortion rates don’t lead to high crime is an increase in reliable birth control. For instance, following the increased spread of AIDS and other sexually transmitted diseases in the 1990s, condom use may have risen. More condoms would lead to fewer conceptions, including fewer unwanted conceptions. The result would be both a lower abortion rate and a lower number of unwanted births.

A third possibility is that the demographic most likely to obtain abortions is conceiving less for other reasons — including, possibly, less sex. My student Amee Kamdar has a fascinating paper in which she shows that incarceration of twenty-something males greatly reduces the number of teenage births. Indirectly, the continuing rise in incarceration may be contributing to reduced rates of both teen births and teen abortions.

As this discussion points out, the relationship between abortion rates and future crime is subtle. Subtle enough, in fact, that I believe that many researchers studying this issue do not understand it themselves.

Donohue and I present evidence that the rise in abortions in the years immediately following legalization was due to the fact that abortions got “cheaper.” As such, during that period, more abortions implied less crime. Whether that is still true today is quite questionable. My hunch is that the recent declines in abortion are not really a result of abortions becoming more expensive. There are fewer abortion providers today, so maybe that makes it harder to obtain an abortion. On the other hand, when demand for a good (in this case, abortions) falls, you expect the number of providers to shrink. It is unclear which is causing which.

Perhaps more importantly, one might think that the introduction of RU-486 represents an important technological shock that lowers the cost of getting an abortion. Yet there are fewer abortions.

So, ultimately, although this is a bit of a guess, I would surmise that the low abortion rate today is being driven by a decrease in unwanted conceptions. If that is true, then these low abortion rate statistics are good news for future crime rates.

Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. scott cunningham says:

    First, I agree that Amee’s paper is fascinating. Particularly since one can imagine that the effect of incarceration could go the other way. High rates of incarceration among males offset sex ratios, and theory also predicts that this imbalance could lead to an increase in sexual risk-taking among men due to the surplus of unattached women. Posner, Becker and William Julius Wilson have all looked at this possibility. What has not been heavily studied, though, is the incapacitation effect. Amee’s paper and Stephane Mechoulan’s papers on this are the first that I’ve seen that carefully examine this separate mechanism by which incarceration could have the reverse effect of driving down teen fertility. Amee and Stephane’s work on this both find evidence of such incapacitative effects.

    Also, just to quibble, it’s not necessarily the case that more expensive abortions would have the effect you note. If abortions are becoming marginally more expensive, then it might also have the effect of reducing crime (given your selection effect), and not the other way around. Studies have found some evidence for reductions in abortions with no change in birth rates when parental involvement laws are adopted, for instance. See Philip Levine’s excellent book Sex and Consequences for the literature review on that. Sexual behavior and fertility decisions are endogeneous, after all, to the price, and so if at the margin it is slightly more expensive, then women may adjust their pregnancy risk downward, resulting in fewer abortions and no increase in births.

    Also, the AIDS epidemic did have a significant effect on condom use. Ahitivus, Hotz and PHilipson have a 1996 JHR paper entitled “The Responsiveness of the Demand for Condoms to the Local Prevalence of AIDS”, and in it they find very large elasticities of condom use with respect to AIDS incidence locally. Thomas Dee and Harrell Chesson find that the AIDS mortality rate appears to be driving down syphilis rates, and in a working paper, Chris Corwnell and me find that it also is driving down gonorrhea (which is suggestive more of deterrence than incapacitation). Stephane Mechoulan has a new paper in the Forum for Health Economics and Policy entitled “Risky Sexual Behavior, Testing, and HIV Treatment” in which he finds the availability of highly active antiretroviral treatments (HAART) starting in 1996 resulted in increased risky sexual behavior among a sample of gay males in San Francisco. Other papers, such as a recent 2006 QJE paper entitled “HIV Breakthroughs and Risky Sexual Behavior” by Goldman, Lakdawalla, and Sood, also find such things. If the AIDS epidemic really is responsible for declines in abortions, and ultimately crime rates themselves, then the increased availability of HAART may be a way of testing the theory, since HAART has dramatically reduced mortality rates, and increased risky sexual behavior in HIV-positive and negative gay males, at least found in a few studies using observational data.

    Thumb up 0 Thumb down 0
  2. hammertime says:

    There’s also the morning after pill… talk about a life saver. (Kind of an unintentional pun)

    Thumb up 1 Thumb down 0
  3. Jeffrey says:

    What about recent studies which suggest environmental lead was a large cause of crime rate surges? I’ll bet that grossly outweighs abortion availability.

    Thumb up 0 Thumb down 0
  4. tmfa says:

    “What Do Declining Abortion Rates Mean for Crime in the Future?”

    Nothing. This myth of causation between the 2 was debunked years ago.

    Thumb up 2 Thumb down 0
  5. Angela says:

    Could fertility rates be dropping in females much like it is in the male population?

    Perhaps there is an environmental factor that is reducing fertility overall.

    Thumb up 0 Thumb down 0
  6. Eyal says:

    If the decrease in abortions was due to a decrease in the number of unwanted female babies, we should have seen the decrease ten years ago, when crime rates started falling (supposedly due the the decrease in unwanted born men). I could not clearly from the statistics I found when the downward trend in abortions started.

    In any case, the less-unwanted-females causes less-abortions hypothesis should be amenable to the same type of analysis of the abortion-crime link.

    Thumb up 0 Thumb down 0
  7. Casper Hauser says:

    Since abortion rates are highest amongst african-americans, this suggests that the birthrate for african-americans is much lower now. Since african-americans are arrested at much higher rates than non-african-americans, this suggests a decreasing crime trend in the future. Another way to put it: The genocide is working.

    Thumb up 1 Thumb down 0
  8. MRB says:

    “The genocide is working”? Ignorance is bliss, isn’t it, Casper Hauser? What a tired argument.

    Thumb up 0 Thumb down 1