Research shows that abortion and crime rates are related. The question is whether they are directly linked. One study claimed that legalizing abortion increased murder rates by up to 7%. This is because legalizing abortion is a contributing factor to the great increase in out-of-wedlock births and single parent families, which in turn contribute to increased crime rates.
However, most people believe the results of a faulty study—which its authors recently updated and defended—that concluded that abortion decreases crime rates. Let’s consider the evidence.
Where It All Started: Donohue and Levitt’s 2001 Study
Throughout the twentieth century, eugenicists promoted abortion and birth control, claiming that if the “lower classes” would only have fewer children, crime would also decline. This was one of the primary themes of Margaret Sanger’s Birth Control Review, in which Montgomery Mulford wrote: “I am of the belief that the acceptance of birth control by society, and its frank teaching, can help diminish criminal activity!”1
This theme still resonates strongly with many people today. The best-known study of the abortion-crime connection was performed by John J. Donohue III and Steven D. Levitt in 2001. In Harvard University’s Quarterly Journal of Economics, the authors concluded: “Legalized abortion has contributed significantly to recent crime reductions…. Legalized abortion appears to account for as much as 50 percent of the recent drop in crime.” Donohue and Levitt noted that “crime began to fall roughly 18 years after abortion legalization,” and stated that the social benefit of this decrease in crime is about $30 billion annually.2
Donohue and Levitt wrote that from 1991 to the time of their first study—and 18 years after Roe v. Wade legalized abortion—murder rates fell faster than at any time since the end of Prohibition in 1933. They added that the five states that legalized abortion earlier than 1973 (New York, California, Washington, Hawaii, and Alaska) also experienced earlier declines in crime. Finally, they found that states with especially high abortion rates in the 1970s and 1980s had equally dramatic crime reductions2 in the 1990s.
Levitt went on to coauthor the 2005 bestseller Freakonomics, in which he reiterated his thesis that the legalization of abortion is responsible for half of the recent drop in violent crime.
Prominent pro-abortion groups and leaders immediately seized on the conclusion of the Donohue-Levitt study and used it as justification for promoting and funding the practice of abortion. For example, in a 2001 op-ed piece heartlessly entitled “It’s Better for Us That They Died,” Canadian abortionist Henry Morgentaler declared moral vindication and grumbled that he had made this claim for decades.3
Counterevidence from Lott and Whitley
Donohue and Levitt were certainly correct when they said that violent and property crimes had drastically decreased since 1991. At the time of their study:
- Murder rate had decreased 49%
- Rapes, 32%
- Robberies, 50%
- Aggravated assault, 39%
- Property crimes, 41%
Additionally, states with very high abortion rates in the 1970s and 1980s also had correspondingly dramatic crime reductions4 in the 1990s.
While Donohue and Levitt were doing their research, however, other scientists were arriving at opposite results, refuting the theory that abortion and crime rate directly affect one another.
Law professors John R. Lott, Jr. of Yale Law School and John E. Whitley of the University of Adelaide found that legalizing abortion increased murder rates by up to 7%. They concluded that legalizing abortion is a contributing factor to the great increase in out-of-wedlock births and single parent families, which in turn contribute to increased crime rates.
Since 1968, the percentage of children “…living with an unmarried parent [in the United States] has more than doubled…from 13% to 32%” (as of 2017).
The Brookings Institution has more sobering numbers for the rate of out-of-wedlock births. “Since 1970, out-of-wedlock birth rates have soared. In 1965, 24 percent of Black infants and 3.1 percent of white infants were born to single mothers. By 1990 the rates were 64 percent for Black infants, 18 percent for whites. Every year one million more children are born into fatherless families.”
The Results of the Donohue & Levitt 2001 Study Were Faulty
There were many fundamental problems with the conclusion that legalized abortion and a decrease in crime rate are related:
- Crime rates dropped first among older people. Statistician David Murray confirmed that young males between the ages of 17 and 25 commit the majority of crimes. If abortion reduces crime, the crime rates in the United States should have dropped first among young people. Instead, the number of crimes committed by older people dropped first. Nearly 60% of the decline in murder since 1990 involved killers aged 25 and older—who were born before Roe v. Wade.5
- Other nations with high abortion rates had a large increase in crime about 18 years after legalizing abortion, as Murray found. For example, in Great Britain, which legalized abortion in 1968, violent crime rose steeply from about 1985 to 2000―exactly when it should have been declining, according to the Donohue-Levitt thesis.
- Murder rates skyrocketed among black people. Since black women have abortions at a much higher rate than white women, we should have expected the murder rate among black youth to have declined beginning in about 1991. Instead, it increased more than 500% from 1984 to 1993.6
- Crime rates rose when abortion rates were constant. If abortion reduces crime, the crime rate should have been relatively stable during the time period 1980 to 1991. Instead, the crime rate rose during the time period 1984 to 1991, after a decline from 1980 to 1984. Further, the crime rate over the past 20 years has continued to decline.
Researchers believed that the huge increase in violent crime that peaked in 1991 and then began to decline was more closely related to the crack epidemic, not abortion. The Donohue-Levitt study confirms that the crime rate rose and fell exactly where crack cocaine was most easily available―in the large cities and among young black males.6
Donohue and Levitt’s 2019 Study
In May 2019, Donohue and Levitt reexamined data from their first paper and published a working paper entitled “The Impact of Legalized Abortion on Crime over the Last Two Decades.” The data in this paper covers the years 1998-2014 and purports to confirm the hypothesis of their first paper.
Now that nearly two decades have passed, we must look at what—if anything—has changed. How have crime rates changed since this initial study? According to data reported by the FBI in 2019: “The violent crime rate fell 49% between 1993 and 2019, with large decreases in the rates of robbery (-68%), murder/non-negligent manslaughter (–47%) and aggravated assault (–43%) … [while] property crime rate fell 55%, with big declines in the rates of burglary (–69%), motor vehicle theft (–64%) and larceny/theft (–49%).”
Obviously, crime rates vary from city to city and from state to state. The population density and economic conditions of individual cities play a huge role in crime rate. But can it be said that this decline is due to decriminalized abortion?
According to a report published by the Guttmacher Institute in 2019, abortion rates during the period of 2011-2017 have fallen. During this time period, abortions fell by 196,000. This decrease likely has multiple factors, including tougher abortion restrictions, clinic closures, and the expansion of pro-life pregnancy care resources.
Critics of both the 2001 Donohue-Levitt study and the 2019 paper remind people that correlation does not equal causation. They cite other reasons—such as the war on drugs, a stronger economy, a tougher police force, and gentrification—that could all contribute to the decrease in the national crime rate.
Further, since abortion rates and crime rates have both been dropping, opponents of the Donohue-Levitt theory claim that this fact proves them wrong.
A Moral Argument
The central thesis of both the 2001 Donohue-Levitt study and the 2019 working paper is a refinement of the pro-abortion slogan “Every child a wanted child.” The authors claim that because a difficult home environment leads to an increased risk of criminality, increased abortion reduces unwantedness, and therefore abortion and crime rates are connected.
However, although criminals may more likely come from a “difficult home environment,” that is not always the case. Many talented and gifted individuals have come from a difficult home environment, including Leonardo Dicaprio, Charlie Chaplin, Louis Armstrong, playwright Eugene O’Neill, Audrey Hepburn, James Dean, Merle Haggard, comedian Tim Allen, and, ironically, Barack Obama—a fanatical supporter of abortion.7
The point here is this: No matter how terrible a home environment is, no child is certain to become a criminal. Crime is not programmed into our DNA. Any program intended to help exterminate the preborn children of the poor is simply preemptive capital punishment—a punishment curiously supported by many of the same people who oppose the death penalty for adult criminals.
Correlating abortion and crime rate increase assumes that a child cannot choose whether or not he will become a criminal.
Negative Effects of Abortion on Society
Donohue and Levitt also claimed in their 2001 study that, because there is a correlation between abortion and crime rate increase, legalizing abortion has caused a social benefit that amounted to $30 billion annually.
This is a narrow and short-sighted view, completely neglecting the benefits each person contributes to society. The direct cost of each abortion to society in terms of lost consumption and taxes paid amounts to, on average, $3,720,000. According to the CDC, there were 619,591 legal induced abortions in the US in 2018. This means that the total direct cost of abortion in the United States every single year is nearly two and a half trillion dollars.
Related: Is Abortion a Constitutional Right?
Final Thoughts
From the 50 years of legalized abortion and the faulty study from Donohue and Levitt, we can see that the claim that abortion lowers crime rates is false and faulty.
In the last paragraph of their 2001 paper, Donohue and Levitt agree that “providing better environments for those children at greatest risk for future crime,” would cause an equivalent reduction in crime.
So, the question for all of us, pro- and anti-lifers and alike, is this: Do we want to attack the symptom or the cause? Do we clumsily and bloodily try to eliminate even more criminals through the mechanisms of eugenics, abortion, sterilization, and birth control (programs that have proved themselves unequal to the task), or do we embrace the proven remedies of strengthening family life, enforcing the law, and providing education, resources, and better living conditions for the poor?
Stephen Levitt believes that working on his controversial research actually moved him further toward a pro-life position. He agrees that one could conclude from the evidence he and Donohue compiled that the answer isn’t more abortions but better education and living conditions for the poor.8
Margaret Sanger advocated the elimination of “human weeds” many years ago. Wrong about so many things, she certainly erred in claiming a correlation between abortion and crime rate increase. And her eugenics programs did not improve the lot of the poor. All she did was turn large poor families into small poor families.
Current-day eugenicists are pushing the same failed program.
This article was most recently updated in May 2023.
+ Endnotes
[1] Montgomery Mulford. “Birth Control Lessens Crime.” Birth Control Review, Volume XV, Number 10 (October 1931), page 293. For 888 quotes from Margaret Sanger’s Birth Control Review, e-mail Brian Clowes at bclowes@hli.org.
[2] John J. Donohue III and Stephen D. Levitt. “The Impact of Legalized Abortion on Crime.” Harvard University’s Quarterly Journal of Economics, May 2001.
[3] Canadian abortionist Henry Morgentaler. “It’s Better for Us that They Died.” Guest op-ed piece in the May 18, 2001 Canadian National Post.
[4] Other statistics that seem to support Donohue’s and Levitt’s conclusions include:
- Unmarried women and teenage girls account for 84% of all abortions obtained in the United States, and half of all unmarried women’s pregnancies end in abortion, as compared to only 8% of married women’s pregnancies. Children born to such mothers have a higher than normal probability of committing crimes in the peak ages for crime, 18 to 24. Therefore, a higher abortion rate among unmarried women would apparently lead to a decrease in crime.
- The likelihood of future criminal behavior declines if children are born into better environments. Teenagers and unmarried and poor women are most likely to consider a pregnancy unwanted, and unintended pregnancies are associated with poor prenatal care, greater smoking and drinking during pregnancy, and lower birth weights, all associated with higher crime rates in children.
[5] “Statistician Says Abortion-Crime Study is Erroneous.” Fox News, May 16, 2001; Steven Ertelt’s Pro-Life Infonet, May 17, 2001. Also see the National Post online, May 17, 2001.
[6] Federal Bureau of Investigation, United States Department of Justice website, table entitled “Homicide Offenders by Age, Gender, and Race, 1976-99.”
[7] John W. Whitehead. “Check the Facts.” The Washington Times, June 28, 2001.
[8] John J. Donohue III and Stephen D. Levitt. “The Impact of Legalized Abortion on Crime.” Harvard University’s Quarterly Journal of Economics, May 2001.
Dr. Brian Clowes has been HLI’s director of research since 1995 and is one of the most accomplished and respected intellectuals in the international pro-life movement. Best known as author of the most exhaustive pro-life informational resource volume The Facts of Life, and for his Pro-Life Basic Training Course, Brian is the author of nine books and over 500 scholarly and popular articles, and has traveled to 70 countries on six continents as a pro-life speaker, educator and trainer.
Thank you so so much for speaking the truth.
I don’t agree with many things in this article. Sterilization and artificial birth control methods help prevent unintended pregnancies, this preventing abortion, so why is the author lumping those two methods with abortion or eugenics? Yes, the crime rate partly decreased because less people were born. These pro-life types seem to never see the potential negative impact of adding more human life to the planet. As more people are born, there’s increased potential for more problematic humans to exist. These include bullies, psychopaths, sociopaths, criminals and terrorists. We don’t need an additional 1-2 billion potentially problematic humans existing on this planet that contribute to using already dwindling natural resources, increased pollution, increased trash, increased environmental degradation, increased polluted air, increased polluted water, increased deforestation to support creating homes or agricultural land, increased potential rapists, increased potential child molesters, increased potentially abusive parents and list goes on. They always love to say one-sided nonsense like “If you abort that baby, you could be killing the next doctor yada yada”. I could also say “Those millions of unborn babies could grow up to be child molesters, rapists, bullies, terrorists, domestic abusers, etc.” They have this one-sided, emotional rhetoric that acts like all unborn humans will add good to society after their born and that’s not true in many cases.
If you condemn people based on what they may do, you forfeit any right to defense against being declared a threat and punished accordingly. As you judge, so shall you be judged.
I consider this irresponsible, illogical fake news, and I am a strong conservative.
[…] pointed out that statistician David Murray confirmed that males among ages 17 to 25 commit the bulk of violent crimes. Murray instructed Fox News when the 2001 investigation was […]
I love how this article attempts to make it seem as if legalized abortion SHOULD have had a significant impact on crime pre-1993, when the entire Freakonomics assertion is based on more recent years SINCE the early ‘90s. The only way this article can even get the figures they’re touting is to use non-applicable data **rolls eyes/face-palm**. I really hope most of you are smarter than to believe this nonsense without doing the actual math yourself.