By Ed Cara
It’s been called everything from one of the greatest achievements in the name of freedom to one of the largest instances of legalized slaughter since the Soviet purges. Whatever your personal feelings, there’s no denying the tremendous impact Roe vs Wade, and the subsequent legalization of abortion in the United States in 1973, has had on the generations of Americans born before and since the landmark trial.
As far as the US is concerned though, the battle over women’s reproductive rights is long from settled. Setting aside the ruckus over the administration’s attempt to provide female employees with a guaranteed birth control option in their health insurance plans, no matter what personal beliefs their employers may hold, it hardly seems like a day can pass by without some new state bill being proposed to limit access to the abortion rights of women across the country. From attempts to marginalize and shut down the few clinics in a state that can even provide abortions (Mississippi) to introducing draconian restrictions and deceptive language that seek to only confuse, shame or mislead those who manage to find a brave enough doctor to perform the procedure (Minnesota). There is a wave of anti-abortion and frankly anti-women sentiment running through the conservative segments of the country; sentiment disguised as heated talks over religious freedom, or hypocritically as compassionate pleas over unborn human lives. Strangely enough, the freedoms and rights of the presently alive and certainly very human mothers to be are rarely given as much credence.
However, relatively few among us have attempted to quantify the impact those from the other side of the equation, the generations of those who never saw the light of day, have had on our world. That is until economist Steven Levitt and Professor John Donohue of Yale Law School came out with a study intended to do just that in 2001. What they found would fascinate and anger people from all walks of life, whether they were fellow economists, politicians or your average every-day citizen.
That’s because Levitt’s study concluded that the legalization of abortion in the 1970’s was one of the most profound factors in the significant and unexpected crime rate drop seen across the country in the 1990’s.
As America exited the post-war prosperity of the 50’s, the national crime rate began to steadily grow with each decade; urbanization bringing us larger cities and crowding more people together. Once cheap crack-cocaine entered the U.S. in the 80’s though, the resulting turf wars between crack-dealers and gangs sparked an epidemic of crime as murder, rape, assault and robbery rates reached a fevered pitch.  After years of exponentially higher crime rates and with a poor economy to boot, the country was at the brink of chaos. Many in the media predicted the U.S. would buckle under the never-ending waves of violence and collapse onto itself as we approached the millennium.  But we didn’t. As the last decade of the 20th century reached its end, violent crime took a sharp inexplicable dive across the board right as the crack epidemic burned out and the economy began booming once again. The chaos simply stopped. 
Once the drop became apparent, those same doomsayers rushed to analyze(or at least appear to) what had happened. Reasons varied from everything to the bustling economy to innovative policing strategies recently implemented in New York and other large cities at the time; but as Levitt pointed out in his book Freakonomics, his best-selling novel in 2005, co-written by journalist Stephen Dubner, and which laid out the abortion-crime theory(among other ideas) to the general public, most of these explanations proved to have little evidence behind them. No, according to Levitt, the data showed it came down to factors like longer jail sentences for violent offenders, the increased hiring of police officers, and the stabilization of the crack supply. But he also found that those factors didn’t come close to explaining away the massive crime drop. There was still a large piece of the puzzle missing, and that’s where his seemingly out of left field theory came into the picture.
Levitt’s reasoning was simple enough. When abortion became legal nationwide in the 70’s, this led to fewer unwanted and/or poverty-stricken children being born into the world, the same sort of children statistically more likely to become violent and unlawful as they reached the criminal peak of their late teen years. As the 90’s approached and the first generation of Americans born after Roe reached those late teen years, there were that much less potential criminals in the population, thus crime as a whole dropped and leveled off. The less criminals existed in the first place, the less crime in the country. A simple, if bold, claim and one Levitt and Donohue backed up with multiple strands of evidence.
In their study, The Impact of Legalized Abortion on Crime, they showed that in states where abortion became legal earlier than 1973, such as New York, Hawaii and California, the crime rate dropped earlier than in those later states. In addition, those states with higher abortion rates showed a higher crime rate drop relative to those states with lower abortion rates, a correlation that doesn’t begin to show up until after the first generation of post Roe births reached young adulthood and one that doesn’t appear among the older population of criminals. They also relied upon studies of Australian and Canadian populations showing a similar relationship between abortion and crime.  Their conclusion out of all this data? Even when controlling for as many variables as possible, the legalization of abortion might have been responsible for up to as much as 50% of the 90’s crime drop (The original study looked at data from 1985 to 1997, though crime has continued its downward spiral to this day ).
It should hardly come as a surprise that with as a dramatic a theory as theirs, there came some strong reactions from all sorts of circles. From praises of his best-selling Freakonomics to accusations that Levitt’s study was advocating the use of abortion as a part of an eugenics campaign against minorities. And of course, the requisite death threats. 
Some pointed out that African American women have been shown to be up to three times more likely to abort than white women, and from there they made the argument(if poorly) that Levitt, through his study, was championing the use of abortion as a blunt but effective tool for culling the criminal minority population of the US; and then there were those from the religious pulpit offended at the thought that Levitt was providing a convenient justification for the millions of unnecessary murders since 1973 to pro-choice proponents.  From the very start though, Levitt has brushed aside any talk of eugenics or advocacy in the conclusions he and Donahue came to, pointing out in the study itself that their results “not be misinterpreted as either an endorsement of abortion or a call for intervention.” Later in Freakonomics, Levitt made the case that he was only attempting to show the link between abortion and crime his research had illuminated, not make any sort of value judgement as to its findings. Though it’s not only pro-life or racially sensitive groups that Levitt has had to defend his study from, but even other economists and journalists who, since before its official release in 2001, have challenged the study’s conclusions outright.
Steve Sailer, a journalist who has sparked controversy of his own  with several of his articles on race relations in the US, debated Levitt in 1999 over the study’s statistics and assumptions.  He made the case that Levitt hadn’t taken into account the fact numbers showed that murder rates among 13-14 year olds in 1993 and 1994 were the highest they had ever been, the exact opposite effect Levitt had predicted abortion would have on younger populations. Sailer went on to argue that the influence of crack cocaine was understated by Levitt, and in there laid the true answer to the end of the crime epidemic. Levitt, to his credit, acknowledged that Sailer’s numbers were right, but that he and Donohue had taken the crack cocaine effect into account, and that states where abortion rates were high but untouched by the drug still showed lower crime rates. More significantly, the end of the crack epidemic didn’t reduce crime numbers back to their pre-epidemic levels as might be expected if Sailer’s theory was right. The years where teen murder had escalated could certainly be seen as being heavily brought on by the rising crack use, but overall the data showed that crack alone was not responsible for its subsequent drop later on, Levitt argued. Once controlled for statistically, there was still a missing gap in the story, one that explained by abortion.
A similar but more scientific challenge came in the form of Ted Joyce, a Professor of Economics and Finance at Baruch College, who published a paper in 2004 looking at some of the numbers of Levitt and Donohue’s original study.  He concluded, among other arguments, that a “Roe effect” couldn’t be seen when you looked at populations directly born before and after 1973 during the years between 1985-1990. Levitt and Donohue responded in kind with their own paper,  refuting Joyce, arguing that by only looking at those specific years, the same years in which the crack epidemic was at its peak, Joyce buried the effect of abortion in statistic noise. Noise that was removed once you took a look past the ’85-’90 years. They also pointed out that the same states which had legalized abortion earlier were also the same states hit more severely by the influx of crack between those years, and that Joyce hadn’t adequately looked at the entire picture of crime before, during and after the drug’s rise. They went on to use more recent, stronger data to reinforce the original’s conclusions. Joyce has since published his refutation to their refutation.
Most significantly, in 2005, economists Chris Foote and Chris Goetz came out with their own analysis and provided one of the harshest blows to the theory.  They first correctly found that there was missing data from the original study, data which they then went on to say, when introduced to the equation, lowered the effect of abortion on crime to a statistically insignificant level. In essence they found Levitt and Donohue had made a grave technical error which invalidated a good portion of the original’s results. True to form though, Levitt and Donohue came back with a paper that acknowledged the original’s mistake in accidentally omitting data while presenting a re-analysis using the same standards of Foote and Goetz and under an apparently more stringent set of data points.  Their conclusion was that while the effect of abortion on crime might have been less significant than they believed, the recent science was still behind their original findings.
If there’s a feeling of dizziness after having read all that, it would be well understood; the statistical back and forth between Levitt and fellow economists over his research being as challenging a read as you would expect about the merits of separating one singular factor from the annals of history to explain so large a change twenty years later. Levitt attempted to do so, and while there have been reasonable challenges made as to his and Donohue’s conclusions, there’s also been as much support for his theory from others. 
Far from hiding though, Levitt has continued to invite criticism from all corners, constructive or otherwise. There’s been no bullying for others to accept his results, no heavy handed sermon as to what to even do with those results, simply a plea to look at what he has uncovered and decide for ourselves of its veracity. While many have lauded Levitt for doing so, even as they disagree with him, others have attacked him for daring to remind us about the unsightliness of abortion, let alone the implication that it might have actually done us a public good. That fewer unwanted children leads to less crime and violence down the road is not a controversial thought in of itself, but the idea that what many still see as an unnecessary and brutal act of death can save many a life is one not easily reconciled. In the wake of the continued fervor over the moralities and legalities of abortion, Levitt illuminates the simple fact that the world is not black and white, but rather shades of gray.
But one needn’t be a staunch pro-life advocate to feel discomfort about the reality Levitt has laid out for us. As science is often apt to do, the implications of Levitt and Donohue’s research force us to stare darkly at our past, present, and future and examine the hidden consequences a private decision can have on an increasingly public world. Is it any surprise that there are those who would rather take the path more traveled instead?
- Donohue, John and Steven Levitt, “The Impact of Legalized Abortion on Crime,”Quarterly Journal of Economics, CXVI, no. 2 (2001): 379-420, http://pricetheory.uchicago.edu/levitt/Papers/DonohueLevittTheImpactOfLegalized2001.pdf
- Drug Enforcement Adminstration, “DEA History Book 1985-1990.” http://www.justice.gov/dea/pubs/history/1985-1990.html.
- Dilulio Jr., John. “The Coming of the Super Predators.” The Weekly Standard, November 27, 1995. http://cooley.libarts.wsu.edu/schwartj/criminology/dilulio.pdf
- Eurekalert, “New research reveals historic 1990s US crime decline.” http://www.eurekalert.org/pub_releases/2007-02/uoc–nrr021207.php.
- Levitt, Steven. “Understanding Why Crime Fell in the 1990′: Four Factors That Explain the Decline and Six That Do Not.” Journal of Economic Perspectives. 18. no. 1 (2004): 163-190. http://pricetheory.uchicago.edu/levitt/Papers/LevittUnderstandingWhyCrime2004.pdf
- Levitt, Steven, and Stephen Dubner. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York, NY: HarperCollins, 2005.
- Loeber, Rolf, and Magda Stouthamer-Loeber. JSTOR, “Family factors as correlates and predictors of juvenile conduct problems and delinquency.” Last modified 1986. http://www.jstor.org/pss/1147516.
- Sen, Anindya. “Does Increased Abortion Lead to Lower Crime? Evaluating the Relationship between Crime, Abortion, and Fertility.” The B.E. Journal of Economic Analysis & Policy. 7. no. 1 (2007). http://www.bepress.com/bejeap/vol7/iss1/art48/
- Leigh, Andrew, and Justin Wolfers. “Abortion and Crime.” AQ: Journal of Contemporary Analysis. 72. no. 4 (2000): 28-30. http://people.anu.edu.au/andrew.leigh/pdf/Abortion&Crime(AQ).pdf
- Benfield, Kaid. Switchboard:Natural Resources Defense Council Blog, “City crime continues to drop, especially where diversity is strong.” Last modified July 8, 2011. http://switchboard.nrdc.org/blogs/kbenfield/city_crime_continues_to_drop_e.html.
- Deem, Rich. Evidence for God from Science , “Modern Eugenics: How Abortion is Getting Rid of “Undesirables”.” Last modified July 12, 2009. http://www.godandscience.org/doctrine/moderneugenics.html.
- Levitt, Steven. “Levitt on Abortion/Crime Feedback.” Posted Sept 19, 2007. The New York Times. Web, http://video.nytimes.com/video/2007/09/19/opinion/1194817093143/levitt-on-abortion-crime-feedback.html
- Barber, La Shawn. “Steven Levitt Says Child Killing Reduces Crime.” La Shawn Barber’s Corner (blog), April 20, 2005. http://lashawnbarber.com/archives/2005/04/20/steven/.
- “Crime-Abortion Study Continues to Draw Pro-life Backlash .” Ohio Roundtable Online Library, , sec. Life and Health, August 11, 1999. http://www.aproundtable.org/library/articles/life/crimeabortion.html
- Media Matters, “American Conservative reportedly to publish far-right columnist’s baseless, racially charged claims about “wigger” Obama.” Last modified March 14, 2007. http://mediamatters.org/research/200703140004.
- Levitt, Steven, and Steve Sailer. “Does Abortion Prevent Crime?.” Slate Magazine , August 23, 1999. http://www.slate.com/articles/news_and_politics/dialogues/features/1999/does_abortion_prevent_crime/_2.html
- Joyce, Ted. “Did Legalized Abortion Lower Crime?.” The Journal of Human Resources. 39. no. 1 (2004): 1-28. http://users.ipfw.edu/dilts/E 306 Readings/Did Legalized Abortion Lower Crime.pdf.
- Donohue, John, and Steven Levitt. “Further Evidence that Legalized Abortion Lowered Crime:A Reply to Joyce.” The Journal of Human Resources. 39. no. 1 (2004): 29-40. http://pricetheory.uchicago.edu/levitt/Papers/DonohueLevittReply2004.pdf
- Joyce, Ted. “A Simple Test of Abortion and Crime.” The Review of Economics and Statistics. 91. no. 1 (2009): 112-123. http://www.mitpressjournals.org/doi/abs/10.1162/rest.91.1.112.
- 20. Foote, Christopher, and Christopher Goetz. “The Impact of Legalized Abortion on Crime: Comment.” Quarterly Journal of Economics. 123. no. 1 (2008): 407-423. http://qje.oxfordjournals.org/content/123/1/407.abstract.
- Donohue, John, and Steven Levitt. “Measurement Error, Legalized Abortion, the Decline in Crime: A Response to Foote and Goetz (2005).” http://pricetheory.uchicago.edu/levitt/Papers/measurement error.pdf.
- Martin, Spencer. University of Missouri-Columbia, “A Further Analysis of the Causal Link between Abortion and Crime.” Last modified May, 2007. https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/4904/research.pdf?sequence=3