Questionable Statistic Pervades High Court Sex Offender Cases

Bloomberg Law’s® extensive network of reporters and editors helps subscribers to stay ahead of legal

By Jessica DaSilva

Statistics about the likelihood of sex offenders to re-offend took center stage during U.S. Supreme Court oral arguments Feb. 27 ( Packingham v. North Carolina , U.S., No. 15-1194, argued 2/27/17 ).

The case concerned a North Carolina law prohibiting registered sex offenders from visiting social media and other websites like Facebook, YouTube and

But much of the debate focused on a hotly contested statistic highlighted in two Supreme Court opinions, Carissa Hessick—a law professor at the University of North Carolina, Chapel Hill, N.C., whose research focuses on criminal sentencing, criminal law and child pornography—told Bloomberg BNA.

Multiple studies from legal and psychology scholars undermined the “conventional wisdom” that sex offender re-offense rates are high, Hessick said Feb. 27. In fact, many studies have found sex offender re-offense rates are lower than those of offenders who commit non-sexual crimes, she said.

But a Justice Department report cited by North Carolina’s brief suggests those studies may fail to take into account under-reporting by victims of sex crimes and other factors, and that the actual re-offense rate is higher. But it has little reliable data to support that suggestion.

Cited by SCOTUS

North Carolina Senior Deputy Attorney General Robert C. Montgomery repeatedly stated at oral argument that sex offenders have higher rates of recidivism than other offenders. Montgomery specifically referenced statistics cited in the 2003 Supreme Court case Smith v. Doe, which upheld the constitutionality of sex offender registries.

The statistic mentioned in Smith v. Doe—that 80 percent of convicted sex offenders will offend again—is off by an average of 65 percent, depending on offenders’ risk levels, according to research from Arizona State University law professor Ira Ellman.

If the Supreme Court relied on the faulty figures again, the ruling would unjustly ban sex offenders who have already served their sentences from maintaining accounts on social media websites like Facebook, YouTube or newspaper websites that allow minors to become members or create profiles, Ellman told Bloomberg BNA Feb. 27.

That makes the First Amendment claim much more powerful than it might appear, because the people affected by the law pose “no clear and present danger,” Ellman said. “They’re no more likely to offend than most of us.”

‘Up For Dispute.’

Ellman told Bloomberg BNA that his research has found re-offense figures are closer to 5 percent for low-risk offenders, 10 percent for medium-risk offenders, and 30 percent for high-risk offenders. Risk factors vary according to different evaluations, but seek to analyze propensity for re-offense based on behavior and psychology, rather than just on the offense itself.

Despite variations in the exact numbers, all the research indicates recidivism rates are lower than those of offenders who commit non-sexual crimes, Ellman and Hessick agreed.

For example, the Association for the Treatment of Sexual Abusers cited in its amicus brief to a study on sex offender re-offense after prison, which found that about 5.3 percent of convicted sex offenders committed another sex offense.

ATSA is an organization of people who work with sex offenders, including parole officers, psychologists, victims’ advocates and academics.

“The problem that we’re running into is that the Supreme Court has rested its blessing for sex offender registries at least in part on that fact” that sex offenders have high recidivism rates, Hessick said. “But I actually think that’s a fact that’s very much up for dispute.”

Hessick said many studies and data collection efforts—especially those conducted by government institutions like prison administrations—that cite high re-offense rates are fraught with less reputable research methods. Data from civil commitment programs, for example, often offer skewed data, because inmates have fabricated past crimes to avoid serving their sentences with general prison populations, according to The New Yorker.

Dubious Origins

The 80 percent figure appears not to be based on any sort of research or data collection, Ellman said. Smith v. Doe got the number from an opinion from the year before.

Justice Anthony M. Kennedy wrote that opinion, McKune v. Lile, and called the figure “frightening and high.”

It cited to a Justice Department memorandum, which in turn pointed to a brief from the U.S. Solicitor General that was ultimately traced back to a 1986 Psychology Today article, Ellman said. That piece was written by an Oregon prison psychologist with a master’s degree and no research background, who claimed that his sex offender treatment program worked better than anyone else’s, Ellman said.

Ellman discredited the figure in an article published by Constitutional Commentary, a faculty-edited journal at the University of Minnesota law school.

At the very least, a single number shouldn’t be used to indicate re-offense because the range of dangerousness and different types of sexual crimes is so wide, both Ellman and Hessick said. The most dangerous offenders usually don’t suffer under restrictive laws like North Carolina’s because they’re usually still in prison, Ellman said.

It’s also important to keep in mind that many restrictive laws apply to underage kids themselves, Hessick said. Juvenile defendants 16 years old or older are automatically tried as adults in North Carolina, she said. Kids who are prosecuted for statutory rape of a younger romantic partner are banned from interacting with minors, who are their peers, on social networking websites.

These types of restrictive laws can sometimes encourage re-offense by alienating former offenders, Maia Christopher, executive director of ATSA, said. Cutting them off from websites that can offer communication with friends and family, offer job seeking opportunities or just allow them to feel part of a community can increase the risk of recidivism, Christopher said.

“Oftentimes we’re very fearful of people who commit sexual offenses,” she said. “Unfortunately the policies put in place don’t often evaluate what keeps people from re-offending and what we can put in place that will make it less likely that they will offend.”

Failure to Report

In its brief, North Carolina cites to a Justice Department report that mentions the low recidivism rate in the context of studies about under-reporting of sex crimes and studies quoting higher re-offense rates.

Because sex crimes are under-reported, recidivism is likely higher than lower re-offense rates suggest, the brief argues. But neither the brief nor the report contain any data directly connecting low reporting to low re-offense rates.

“Petitioner is on even weaker ground when he suggests that registered sex offenders pose no greater risk than members of the general public,” the brief states. “Not surprisingly, no lower court in this case made a finding of fact with respect to that unsupported and astonishingly counter-intuitive claim. And the claim merits little credence, in large part because recidivism rates do not nearly reflect the actual rate of re-offense by sex offenders.”

In an amicus brief supporting North Carolina’s law, two organizations—Stop Child Predators and Shared Hope International, which advocate on behalf of victims of child sexual abuse—say that research and data support the need for strict laws, but also cited Smith v. Doe and the Justice Department report without acknowledging any other studies calling that data into doubt.

“The risk that these offenders will access the online personalities of children through social-networking platforms and use those platforms to sexually exploit children is quite real,” the brief states. “And limiting social-networking access for a class of persons who pose a heightened statistical risk of abusing the information generated on those platforms is reasonable and compelling.”

“It is reasonable for legislatures and courts to consider whether—relative to other members of society, or other convicted criminals—registered sex offenders present an elevated risk of committing sex crimes in the future,” Samantha Vardaman, senior director at Shared Hope International wrote Bloomberg BNA in an email March 7. She cited the Justice Department report, Smith v. Doe, and a 2003 Bureau of Justice Statistics report examining recidivism rates for sex offenders released in 1994.

“Legislatures have responded reasonably with laws protecting children from online exploitation and recruitment by restricting access from those who present a heightened risk of committing sex crimes,” she wrote.

The North Carolina Attorney General didn’t respond to repeated requests for comment. A request for comment to the law firm Latham & Watkins LLP, Washington, who wrote the brief for Stop Child Predators and Shared Hope International, was not returned.

To contact the reporter on this story: Jessica DaSilva in Washington at

To contact the editor responsible for this story: C. Reilly Larson at

Copyright © 2017 The Bureau of National Affairs, Inc. All Rights Reserved.

Request White Collar & Criminal Law News