August 20, 1999   NY Times   http://www.nytimes.com/library/national/science/082099hth-abort-crime.html

          Roe v. Wade Resulted in Unborn
          Criminals, Economists Theorize

          By ERICA GOODE

            No one is certain why crime rates have plummeted in the United
               States over the last decade. But that fact has not prevented
          politicians from gleefully taking credit for the downturn, or academics
          from ruminating endlessly on its causes.

          The newest theory about why crime is down, however, put forward in a
          report by two highly regarded economists, is drawing both outrage and
          intense debate -- even before the full report has been published or
          subjected to peer review.

          In the report, Dr. John J. Donohue 3d of Stanford Law School and Dr.
          Steven D. Levitt of the University of Chicago contend that a large share
          of the drop in crime in the 1990's -- perhaps as much as half -- can be
          attributed to the sharp increase in abortions after the Supreme Court
          ruling in Roe v. Wade in 1973.

          Fewer crimes are being committed now, the researchers say, because
          many of the children who might have grown up to commit those crimes
          were never born. Within a few years of the Roe v. Wade decision, which
          established a constitutional right to abortion, up to a quarter of
          pregnancies ended in abortion, statistics show.

          Dr. Donohue and Dr. Levitt base their thesis on economic analyses of
          crime rates from 1985 to 1997, examined as a function of abortion rates
          two decades before.

          The timing of the decline in crime, they found, coincided with the period
          when children born shortly after the Roe v. Wade decision would be
          reaching the late teen-age years -- the peak ages for criminal activity.

          States that were the first to legalize abortion, Dr. Donohue and Dr. Levitt
          found, including New York, Washington, Alaska and Hawaii, were also
          the first to experience a decrease in crime.

          For example, in states that legalized abortion in 1969 or 1970, the
          researchers found, the cumulative decrease in crime from 1982 to 1997
          was greater than for the rest of the nation. The decrease in murder was
          16.2 percent greater, the decrease in violent crime over all was 34.4
          percent greater, and the decrease in property crime was 35.3 percent
          greater.

          Also, states with the highest abortion rates, the researchers found, had
          larger reductions in crime than states with low abortion rates.

          The most likely explanation for these findings, the researchers assert, is
          that abortion has occurred selectively, decreasing the number of children
          likely to commit crimes as adults.

          "Most of the reduction," Dr. Levitt and Dr. Donohue write, "appears to
          be attributable to higher rates of abortion by mothers whose children are
          most likely to be at risk for future crime." Teen-agers, unmarried women
          and black women, for example, have higher rates of abortion, the
          researchers note, and children born to mothers in these groups are
          statistically at higher risk for crime in adulthood.

          The economic benefit to society of abortion in reducing crime, the
          researchers suggest, "may be on the order of $30 billion annually."

          The conclusion of the report, a draft of which was posted on a Web site
          of the Social Science Research Network, is not a popular one. When it
          was reported in The Chicago Tribune on Aug. 8, it provoked angry
          op-ed columns, tirades on radio talk shows and expressions of
          indignation by groups on both sides of the abortion divide.

          And the economists have been accused of everything from promoting
          eugenics to recommending abortion as a means to reduce crime.

          "It takes great skill to simultaneously infuriate the right and the left," said
          Dr. Alfred Blumstein, an expert on crime rates and University Professor
          at the Heinz School of Carnegie Mellon University, observing the ferocity
          of the response.

          In a news release, Joseph Scheidler, executive director of the Pro-Life
          Action League, called the study "so fraught with stupidity that I hardly
          know where to start refuting it."

          "Naturally, if you kill off a million and a half people a year," Mr. Scheidler
          said, "a few criminals will be in that number. So will doctors,
          philosophers, musicians and artists."

          Frances Kissling, president of Catholics for a Free Choice, an
          organization based in Washington, commented that even if the report's
          findings were right, "the question in a certain sense is, 'So what?'

          "I don't think it has any policy implications whatsoever," Ms. Kissling
          said. "Abortion is a profoundly private decision" that women make based
          on "their own lives and circumstances," and not public benefit.

          And in a recent column, Carl Rowan wrote, "I've seen a lot of
          far-fetched and dangerous ideas passed off as 'social research,' but none
          more shallow and potentially malicious than the claim that the drop in
          crime in the United States can be attributed to legalized abortions."

          Dr. Blumstein and other criminal justice experts acknowledged that the
          economists may well have demonstrated that abortion rates have had an
          effect on crime, though until the paper is published and subjected to
          academic scrutiny, they said, it is difficult to assess the findings.

          But they expressed skepticism that abortion's effect on crime was as
          great as Dr. Donohue and Dr. Levitt posited. Declining crime rates, Dr.
          Blumstein said, appear to be the result of many complicated factors,
          including the dwindling of the crack cocaine epidemic, an improved
          economy, greater job opportunities for low-income youth and the steady
          growth of the prison population.

          Changing attitudes among teen-agers and innovative policing strategies
          may also be contributors, other experts said.

          "These are very able guys," Dr. Blumstein said of Dr. Donohue and Dr.
          Levitt, "and I'm prepared to believe that they've discerned an effect. But I
          think they've gone too far in claiming that it can account for half of the
          decline, when there are a multitude of effects going on that are much
          more proximate to the situation."

          Dr. David J. Garrow, a historian at Emory University and author of
          "Liberty and Sexuality" (University of California Press, 1998), a history
          of the abortion debate, called the economists' theory "interesting and
          original." But he questioned the researchers' knowledge of abortion
          history and was skeptical of the usefulness of the report.

          "The policy implication of this paper is that if you renewed Medicaid
          funding for abortion at the Federal level, you'd dramatically reduce crime
          17 years from now," Dr. Garrow said. "But are we going to find any
          interest group in America that is going to make that argument? No.
          Neither side in the debate wants to touch it with a 10-foot pole."

          For his part, Dr. Levitt said in an interview that the research was "by no
          means a complete explanation" and that he and Dr. Donohue were aware
          that "the world is complicated." He added that they did not intend their
          work to influence public policy.

          "Our paper should have little to no impact on any policy regarding
          abortion," Dr. Levitt said.

          "There's nothing in our paper that either indirectly or directly suggests that
          we condone denying anyone the right to have children if they want to
          have children," Dr. Levitt added. "We've been accused of having a
          eugenic agenda and it just is not an accurate appraisal of what we're
          doing at all. If anything, what our paper says is that when you remove a
          government prohibition against a woman choosing, the woman makes
          choices that lead to better outcomes for her children."

-------------------------------------------  ABSTRACT OF ORIGINAL ARTICLE ---------------
               Legalized Abortion and Crime

                        JOHN J. DONOHUE
                         Stanford Law School
                         STEVEN D. LEVITT
               University of Chicago; American Bar Foundation

                             June 1999

         Stanford Law School, Public Law and Legal Theory Working Paper No. 1
 

Abstract:

Crime has fallen dramatically in the 1990s. While many explanations for this decline have been
offered, each of them has difficulty explaining the timing, large magnitude, persistence, and
widespread nature of the drop. In this paper we propose a new explanation for falling crime: the
legalization of abortion roughly twenty years earlier. The empirical evidence we present is
consistent with abortion playing an important role. First, the timing of the crime drop corresponds
to the period in which the first cohorts affected by abortion are reaching the peak ages of criminal
activity. Second, states that legalized abortion before the rest of the nation were the first to
experience decreasing crime. Third, states with high abortion rates have seen a greater fall in crime
since 1985. The estimated elasticity of crime with respect to abortion rates is roughly -.10. The
abortion-related reduction in crime is predominantly attributable to a decrease in crime per capita
among the young, rather than smaller cohort sizes. Declining crime rates could result from two
mechanisms: selective abortion on the part of women most at risk to have children who would
engage in criminal activity, and improved child rearing or environmental circumstances caused by
better maternal, familial, or fetal circumstances. Extrapolating our estimates out of sample to a
counterfactual in which there were no abortions, crime rates might be 10-20 percent higher than
they currently are with abortion. If these estimates are correct, legalized abortion can explain
about half of the recent fall in crime. All else equal, we predict that crime rates will continue to fall
slowly for an additional 15-20 years as the full effects of legalized abortion are gradually felt.

Note:  The entire report in Adobe Acrobat format can be downloaded from the source.

A SLATE dialogue on Abortion and Crime with one of the authors provides an interesting discussion and self-criticism of the paper.

Commentary in Newsweek by Robert J Samuelson, Do We Care About Truth?
 

But herewe have another paper by an equally sophisticated econometrician arguing exactly the opposite.  This is typical in research which applies highly complex statistical modeling techniques to data sets that do not meet the conditions of the method.

Abortion and Crime: Unwanted Children and  Out-of-Wedlock Births

                                       JOHN R. LOTT, JR.
                                        Yale Law School
                                       JOHN E. WHITLEY
                                       University of Adelaide
 

                                          April 30, 2001

                                 Yale Law & Economics Research Paper No. 254

                      Abstract:
                      Abortion may prevent the birth of "unwanted" children, who would
                      have relatively small investments in human capital and a higher
                      probability of crime. On the other hand, some research suggests that
                      legalizing abortion increases out-of-wedlock births and single parent
                      families, which implies the opposite impact on investments in human
                      capital and thus crime. The question is: what is the net impact? We find
                      evidence that legalizing abortion increased murder rates by around about
                      0.5 to 7 percent. Previous estimates are shown to suffer from not
                      directly linking the cohorts who are committing crime with whether they
                      had been born before or after abortion was legal.