Recidivism: The Great Lie of “Frightening and High”
Derek W. Logue of
Created 15 January 2018, Last Update 13 October 2018

In so many instances these individuals should never ever be allowed out for a second chance, they’re ticking time bombs, its not a questions if they a re- offend, it’s a question of when they re-offend.” – Lauren Book, Current FL State Senator and victim industry advocate [1]

There is a 90 percent likelihood of recidivism for sexual crimes against children. Ninety percent. That is the standard. That is their record. That is the likelihood. Ninety percent.” – Disgraced former Florida senator Mark Foley [2]

It is a common misconception even today that those who commit crimes are arrogantly breaking the law and deriving a fiendish pleasure from getting away with it. Many people in discussing an offender will assume a hostility not too far removed from the original hostility of the person committing the crime. Thus the old idea of revenge is continued and the possibility of understanding the criminal is kept at a minimum.” — David Abrahamsen [3]


Laws targeting people convicted of sexually-based offenses were justified primarily by a myth that “once a sex offender, always a sex offender,” and that a vast majority of those listed on the public registry will inevitably reoffend.  Media personalities, politicians, and victim advocates will repeat this myth. Even the US Supreme Court has propagated this myth. The myth of high offense rates fuel draconian legislation.

What is recidivism? People demand an easy answer to a complex question. Anti-registry advocates recognize the importance of answering this question:  “So it is important that these two issues related to re-offense rate that must be dealt with. The first one is; of the people that are on the registry, what is the percentage that are involved in new sexually related crimes, and the second question is, what is the percentage of the people on the registry that are involved in a new sexually related crime in comparison to the ordinary citizen who have never been convicted of a sexually related crime? If in either case, there is a high re-offense rate for people on the registry, than the justification for the law could exist, if not than the justification evaporates.” [4] 

This paper is intended to cover this myth and the latest studies that debunk this most dangerous of myths.


While a lot of emphasis on recidivism rates in recent months have centered around the “Frightening and High” claims of Justice Kennedy in Smith v Doe, it is important to realize that views on perceived recidivism rates have fluctuated wildly over the past century and a half.

In his book, “Moral Panic: Changing Concepts of the Child Molester in Modern America,” Historian Phillip Jenkins notes that the word “recidivism” first entered the English vernacular in the 1880s. [5]  It was not until 1894 that child sexual abuse was considered a primary type of sex crime. [6] The modern era reflects the moral panic of the 1930s to early 1950s, commonly known to researchers as the “Sexual Psychopath era.” (The origin of this phrase can be traced back to the 1886 German book “Psychopathia Sexualis” by Richard von Krafft-Ebing.)

Even during the height of the “Sexual Psychopath era,” there was no evidence of high re-offense rates among those convicted of sex crimes. The (New York City) Mayor’s Committee Reports on the Study of Sex Offenses, published in 1944, found that only 40 out of 555 (7%) of people convicted of sex crimes in 1930 were rearrested between 1930-1941.[7]  Paul Tappan’s NJ commission found that “sex offenders have one of the lowest rates as ‘repeaters’ of all types of crime…Among serious crimes homicide alone has a lower rate of recidivism.”[8]

During the Treatment-oriented era of the late 1950s to the mid-1970s, the viewpoint of the “sex offender changed from an incurable threat. The term “child molester” was coined as a term to denote a petty offender, one who did not use violence or force; the term “pedophile” fell out of favor, and the system saw most offenders as amenable to treatment, and recidivism studies continued to fid extremely low recidivism rates.[9]  However, by the 1980s, the modern era of sex abuse panic began; renewed interest in sexual abuse coincided with the rise of Moral Conservatism as well as the Feminist Movement. It became the one issue feminists and conservatives could agree upon.[10]  It is here where our modern narrative on recidivism picks up.


The myth of “frightening and high” recidivism has gained a lot of attention in recent months following the oral arguments for Packingham v North Carolina, 582 US __ (2017). The NY Times reported the State’s lawyer, Robert C. Montgomery, stated, “This court has recognized that they have a high rate of recidivism and are very likely to do this again.” The Times NY stated the term “frightening and high” has been “exceptionally influential. It has appeared in more than 100 lower-court opinions, and it has helped justify laws that effectively banish registered sex offenders from many aspects of everyday life.”[11]  

This term “frightening and high” was cited by Justice Anthony M. Kennedy as he was writing for the Majority in Smith v Doe, 538 US 84 (2003), which in turn was a citation in McKune v Lile, 536 US 24 (2002). McKune, in turn cited the 1988 US Dept. of Justice report “A Practitioner’s Guide to Treating the Incarcerated Male Sex Offender”[12] , which in turn cited a 1986 Psychology Today article[13].  Before addressing the origin of this catchphrase, we should examine the actual statements posted in the two US Supreme Court decisions.

Justice Kennedy’s words from McKune v. Lile:

Therapists and correctional officers widely agree that clinical rehabilitative programs can enable sex offenders to manage their impulses and in this way reduce recidivism. See U.S. Dept. of Justice, Nat. Institute of Corrections, A Practitioner’s Guide to Treating the Incarcerated Male Sex Offender xiii (1988) (“[T]he rate of recidivism of treated sex offenders is fairly consistently estimated to be around 15%,” whereas the rate of recidivism of untreated offenders has been estimated to be as high as 80%. “Even if both of these figures are exaggerated, there would still be a significant difference between treated and untreated individuals”)… The critical first step in the Kansas Sexual Abuse Treatment Program (SATP), therefore, is acceptance of responsibility for past offenses. This gives inmates a basis to understand why they are being punished and to identify the traits that cause such a frightening and high risk of recidivism.

Here is the passage recited in Smith v Doe mere months after the McKune decision:

Alaska could conclude that a conviction for a sex offense provides evidence of substantial risk of recidivism. The legislature’s findings are consistent with grave concerns over the high rate of recidivism among convicted sex offenders and their dangerousness as a class. The risk of recidivism posed by sex offenders is “frightening and high.” McKune v. Lile, 536 U.S. 24, 34 (2002); see also id., at 33 (“When convicted sex offenders reenter society, they are much more likely than any other type of offender to be rearrested for a new rape or sexual assault” (citing U.S. Dept. of Justice, Bureau of Justice Statistics, Sex Offenses and Offenders 27 (1997); U.S. Dept. of Justice, Bureau of Justice Statistics, Recidivism of Prisoners Released in 1983, p. 6 (1997)).”[14]

Returning to the original source of the frightening and high claim, Ira Ellman, professor at Arizona State University’s Sandra Day O’Connor College of Law, wrote a scathing critique of this dubious claim, including breaking down the origin of the frightening and high claim: 

McKune provides just one citation for its much-quoted statement: a 1988 Justice Department ‘Practitioner’s Manual’. That reference likely came from the amicus brief supporting Kansas filed by the Solicitor General, then Ted Olson, which also cites it. This Practitioner’s Guide itself provides but one source for the claim, but it’s no scientific study. It’s a 1986 article from Psychology Today, a mass market magazine aimed at a lay audience, which had this sentence:

“Most untreated sex offenders released from prison go on to commit more offenses–indeed, as many as 80% do.” Freeman-Longo, R., & Wall, R, Changing a lifetime of sexual crime, Psychology Today (1986). 

“That sentence is a bare assertion with no supporting reference. Nor did its author have the scientific credentials needed to qualify at trial as an expert on recidivism.  He was a counselor, not a scholar, and the article containing the sentence isn’t about recidivism statistics. It’s about a counseling program for sex offenders he then ran in an Oregon prison. His unsupported assertion about the recidivism rate for untreated sex offenders was offered to contrast with his equally unsupported assertion about the lower recidivism rate for those who complete his program.”[15]

In response to Professor Ellman’s study, the David Post of the Washington Post wrote:

It appears to be an unfortunate, but crystal-clear, illustration of the cognitive bias that leads people — even Supreme Court justices — to believe statements that confirm preconceived beliefs without demanding any actual evidence in support. What the court really should have said — the only statement that would have been consistent with the facts before it — was:

‘The author of some article published in Psychology Today thinks that the recidivism rate is frightening and high, possibly as high as 80%.’

Not quite as impressive that way, and an awfully thin reed on which to rest the crushing burdens that these prohibitions and restrictions impose.”[16]

Yes, it is entirely possible that Justice Kennedy engaged in confirmation bias. He chose to ignore much of the research cited in the 1988 study. The NY Times describes this 1988 Practitioner’s guide as, “a compendium of papers from outside experts, is 231 pages long, and it contains lots of statistics on sex offender recidivism rates. Many of them were in the single digits, some a little higher. Only one source claimed an 80 percent rate, and the guide itself said that number might be exaggerated.”[17]  On page 65 of the Practitioner’s guide, numerous studies written by various treatment programs show long term sex crime re-offense rates for treated offenders between 0%-30% and untreated offenders between 10%-37%, though it should be noted that these were all studies from hospitals with intensive treatment programs. In other words, they were programs for offenders deemed high risk and not indicative of the re-offense rates of the entire class of people convicted of sex crimes.

The general public believes that roughly three-fourths of those convicted of sex crimes will reoffend.[18]  The belief system has become cyclical—the public believes recidivism is high because of “experts” like media personalities and courts, and these personalities in turn believe recidivism is high because the general public believes it. It is imperative to address this most dangerous of myth due to the impact it has on public policy. As Reason magazine reports, “These numbers matter because fear of recidivism is at the heart of the harsh and sweeping policies aimed at sex offenders, including mandatory minimum sentences, indefinite civil confinement after prison, lifelong registration, residence and presence restrictions, occupational bans, and stigmatizing passports. The risk that sex offenders will commit new crimes is relevant not only in justifying these measures on policy grounds but in resolving constitutional issues such as whether registration and all the burdens associated with it should be viewed as punishment or as regulations promoting public safety.”[19]  


The problem with recidivism numbers is that it is the application is that there really is no universal standard for defining it. Every study varies greatly, which influences numbers. Time frame, definition of actions that are considered reoffending, and other factors play key roles in recidivism. However, society demands a simplistic answer to a complicated question. As time has gone on, even those within the sex offender legal reform movement has evolved their understanding of recidivism.

For example, the largest study and the most cited study is the 2003 US Dept. of Justice study what cited a rearrest rate of 5.3% and a reconviction rate of 3.5%after 3 years.[20]   Over the years, many legal reformists recite the rearrest rate as the official recidivism rate. Many have even simply stated recidivism rate is just 5%. However, this is one study, and few individuals have even mentioned the large discrepancy (40%) between the rearrest rate and reconviction rate, nor is an explanation offered in the study itself.

The simplest definition of recidivism is the rate at which an offender will commit another crime. However, studies vary greatly on what constitutes
“recidivism.” Harris and Hanson’s 2004 survey of available recidivism studies, found a variety of factors that greatly influence the final numbers of each study:

  1. How the word “recidivism” is defined; re-conviction rates, re-arrest rates, informal reports to child agencies, self-reporting, violations of conditional release, or simply being questioned by police. The broader the definition, the higher the number, obviously.
  2. Follow-up period: The longer the follow-up period, obviously the higher the number. However, the longer the individual remains offense free, the less likely the offender will be to re-offend
  3. Diversity among sex offenders: Certain classes of offenders have differing re-offense rates
  4. Determining the percentage of crimes that go unreported to the police. This is usually achieved by victimization surveys. One major problem, however, is the definition of “sexual assault.” The researchers noted a study by Besserer and Trainor (2000) study, noting that study used “a very broad definition of sexual assault,” including all unwanted forms of sexual touching and threats, and possibly “behaviors not conforming to the popular image of a sexual offense.” The researchers noted that 59% of the respondents of this study stated the reason they did not report was because they felt the “incident was not important enough” to report. “Consequently, readers may wonder what counts as a sexual assault.”[21]  

The 1988 Practitioner’s guide recognized the problem with defining recidivism. To the researchers in the report, the term “re-offense” was suggested as the term for rearrest or reconviction for a sexually based offense. They suggested that rearrest is the more liberal of the two rates but could compensate for possible underreporting.[22]  Most studies today use either rearrest rate, reconviction rates, or both. To reiterate, recidivism is generally used to define rearrest or reconviction for any offense, which could include supervision violations, whereas re-offense is typically used to measure the rate at which a previous convicted offender commits the same type of offense.

Some earlier reports referred to “sex specific” recidivism and “general” recidivism,[23]  but it is perhaps less confusing to use a separate term for re-offense rates, since people imply the rates at which someone convicted of a sex crime commits a subsequent sex crime as opposed to the rate at which someone convicted of a sex crime breaks any law or supervision rule.

Having a universal definition will help accurately reflect re-offense rates in America; using numbers from foreign nations or from old studies that largely lacked peer review and accurate record-keeping will not reflect current rates among registered citizens in the US. The reason for settling on a universal standard finds its basis in scientific research. Ellman recognized the differences between American re-offense rates and the re-offense rates in foreign countries, noting foreign recidivism rates would be different “without the distinctly harsh American system of long sentences and post-release restrictions.” [24]  A 1989 meta-analysis of numerous studies also limited recidivism to North America, noting there are “myriad factors affecting the reporting of sex offenses, arrest and conviction rates, and re-offense records, which are likely to cause recidivism rates to be different in other countries.” This study also noted the “methodological weaknesses and lack of uniformity,” noting many previous studies on the era focused only on “a specific type of offender (e.g., pedophiles) or a specific treatment approach (e.g., hormonal treatments).” But even this 1989 study admitted to including unpublished studies, studies that only used as little as 10 subjects, and studies that were “without screening for scientific merit.”[25]

Thus, for the sake of uniformity, this report will largely limit relevant studies written within the past 20 years in America and only those using rearrest and/or reconviction rates for determining re-offense (i.e., sex offense-specific recidivism). However, it is my belief that reconviction rates should be the gold standard, citing the discrepancy between rearrest and reconviction and the fallacy of assuming high rates of underreporting.

It is important to note there are indeed many ways to measure recidivism, and each form of measurement has drawbacks. Below are many of the types of measures you might find in various studies and the potential fallacy of reliance on particular measure:

  1. Self-Reporting: Studies that have relied on self-reporting might find high numbers of reoffending, but self-reports often come from programs where inmates are rewarded by giving the desired answers. A recent example of this came from the debunked Butner study, which made the bold claim that 85% of inmates who were arrested for CP charges had also committed hands-on offenses that have gone undetected. One US District Court rejected that study, finding it “highly coercive,” “not peer reviewed,” and suffering from “methodological flaws.”[26]  
  2. Rearrest Rates: The drawback to relying on rearrest rates is the probability that a person is arrested but not convicted of an actual re-offense. The landmark 2003 US Department of Justice study reported a 5.3% rearrest rate after 3 years, but only a 3.5% reconviction rate in that same period. That is a very large discrepancy the study failed to address.
  3. Reconviction rates: In my personal opinion, reconviction rates offer the best standard for measuring re-offense because of the added filter of a court process. However, since the majority of cases are decided by pleas (even innocent people who plead guilty), that fact can still overestimate the results.
  4. Re-incarceration rates: Since many people are placed on parole/ supervision upon release, many also return to prison for technical violations that may be sexual in nature (such as viewing adult porn) but not necessarily a sexual offense. The Ohio 10 year recidivism study released in 2001 broke down recidivism by type and found only 8% were recommitted for a re-offense while 3% were recommitted to prison on sex-related technical violations.[27]

While my personal viewpoint is that reconviction rates give the best indication of re-offense rates, most studies rely upon re-offense rates, so both rates are shared here.

One solution to clear up confusion with seeming fluctuating numbers would be to create an annual rate from these numbers. For example, we use annual rates to determine the average number of fatalities caused by lightning strikes [28] or the murder rate in a metropolitan area like Cincinnati.[29]  It makes sense from a practical standpoint, as many agencies are prone to cherry-picking the numbers that best align with their philosophy and goals.


Now that we have limited our look at re-offense to American studies from the past 20 years and only considering rearrest and reconviction for subsequent sex offenses, it is time to look at a number of studies meeting this criteria. (Since some studies show both rates, some studies will be repeated.)[30]  

Reconviction Rates

  1. New York (1996): 34 of 556 SOs (6%) were reconvicted within 9 years of release.[31]  
  2. Iowa (2000): 14 of 434 (3.2%) were reconvicted within 4.3 years of release [32]
  3. Ohio (2001): 70 of 879 (8%) were reconvicted after 10 years [33]
  4. US Dept. of Justice/ Federal (2003): 339 of 9641 (3.5%) were reconvicted after 3 years [34]
  5. Washington State (2005): 111 of 4091 (2.7%) reconvicted after 5 years [35]
  6. Minnesota (2007): 10% of 3166 reconvicted after 8.4 years [36]
  7. California (2008): 121 of 3577 (3.4%) reconvicted after 10 years [37]
  8. Arizona (2009): 2 of 290 (0.7%) reconvicted after 3 years [38]
  9. Connecticut (2012): 20 of 746 (2.7%) reconvicted after 5 years [39]
  10. Nebraska Pre-AWA (2013): 48 of 2816 (1.7%) reconvicted after 2 years [40]
  11. Nebraska post-AWA (2013): 6 of 209 (2.6%) reconvicted after 2 years [41]
  12. Michigan (2014): 32 of 4109 (0.8%) reconvicted after 3 years [42]

Rearrest Rates

  1. US Dept. of Justice (2003): 517 of 9641 (5.3%) rearrested after 3 years [43]
  2. Rutgers U. (2005): 4.5% of 917 rearrested after 3.5 years [44]
  3. Alaska (2007): 3% rearrested after 3 years [45]
  4. Multistate Study (2009): After 3 years, each state in the study varied in rearrest rates:
    • Delaware: 3.8%
    • Illinois: 2.4%
    • Iowa: 3.9%
    • New Mexico: 1.8%
    • South Carolina: 4%
    • Utah: 9% [47]
  5. New Jersey (2011): 13% of 274 pre-SOR SOs rearrested within 8 years; 9.7% of 248 post-SOR SOs rearrested within 8 years [47]
  6. Florida/ New Jersey/ Minnesota/ South Carolina (2012): 153 of 1469 (13.7%) rearrested after 10 years [48]
  7. US Department of Justice/ Federal (2014): 5.6% rearrested after 5 years [49]
  8. California (2016): 78 of 1626 (4.8%) rearrested after 5 years [50]

Based on the numbers from this list, it is reasonable to assert that dozens of state, federal and University studies have found re-offense rates to be extremely low, with nearly all studies showing long term re-offense rates in the single digits. However, some continue to be unsatisfied with the results of the numerous studies. There is one major problem with citing the numbers of these studies, namely each study varies in the number of years. (The US Dept. of Justice study was limited to child sex offenders and rapists, i.e., “hands on offenders,” and only of those released from prison.) Again, this stresses the importance of a universal annual rate standard.

Below are the aforementioned studies adjusted into an Average Annual Rate:

Average Annual Rate: Reconviction

  1. New York (1996): 0.66%
  2. Iowa (2000): 0.74%
  3. Ohio (2001): 0.8%
  4. US Dept. of Justice/ Federal (2003): 1.16%
  5. Washington State (2005): 0.54%
  6. Minnesota (2007): 1.19%
  7. California (2008): 0.34%
  8. Arizona (2009): 0.23%
  9. Connecticut (2012): 0.54%
  10. Nebraska Pre-AWA (2013): 0.85% [51]
  11. Nebraska post-AWA (2013): 1.3% [52]
  12. Michigan (2014): 0.26%

Average Annual Rate: Rearrest Rates

  1. US Dept. of Justice (2003): 1.77%
  2. Rutgers U. (2005): 1.29%
  3. Alaska (2007): 1%
  4. Multistate Study (2009):
    • Delaware: 1.27%
    • Illinois: 0.8%
    • Iowa: 1.3%
    • New Mexico: 0.6%
    • South Carolina: 1.34%
    • Utah: 3%
  5. New Jersey (2011): 1.63% pre-SOR SOs; 1.21% post-SOR SOs
  6. Florida/ New Jersey/ Minnesota/ South Carolina (2012): 1.37%
  7. US Department of Justice/ Federal (2014): 1.12%
  8. California (2016): 0.96%

When adjusted into a yearly average, the annual re-arrest/reconviction rates hover at a mere 1% rate for the majority of studies. It is no small wonder many organizations and possibly some researchers show reluctance to adopt this universal standard.


There is one key  caveat in discussing recidivism studies—each study assigns overall risk to an individual whose risk factors may be higher or lower. A number of factors influence the likelihood of reoffending, such as the length of time since an offender’s release, the type of offense committed, whether or not the offender received treatment, and whether or not the offender has multiple convictions on his record. It is also important to address the underreporting myth and a misleading statement that comes from the US Department of Justice study.

The Recidivism Curve: Reoffense risk lowers the longer one has been free from incarceration

Recidivism studies function the same way—you take a pool of offenders and follow them for a set amount of time (generally a number of years), and count how many are accused or convicted of another sex crime after sentencing or release from prison. It sounds simple enough, but the problem with determining averages is a higher number raises the average. Cumulative totals break the law of gravity—what goes up never goes down. Cumulative totals give imperfect numbers that assign a universal number that may be too high for low-risk individuals.

The Ohio DRC found that “Sex offenders who returned for a new sex related offense did so within a few years of release.  Of all the sex offenders who
came back to an Ohio prison for a new sex offense, one half did so within two years, and two-thirds within three years.” [53]  

A 2009 NY Recidivism study found that sex offence recidivism rates after 1 year were 2%, 3% after five years, 6% after 5 years, and 8% after 8 years. [54] A 2008 California study found a 2.21% recidivism after 1 year of release, 2.94% recidivism after 2 years of release: 3.3% recidivism after 5 years of release, and 3.38% after 10 years of release. The study concluded, “The total of sexual recidivists is lower than some might have believed. Most re-offenses and parole violations occur in the initial period of reentry after release. Sex offenders are more likely to commit some other type of offense than to commit a new sex offense.” [55]

Researchers Karl Hanson and Andrew Harris noted, “Another factor to consider is the length of the follow-up period.  As the follow-up period increases, the cumulative number of recidivists can only increase.  It is important to remember, however, that an increase in the number of recidivists is not the same as an increase in the yearly rate of recidivism.  For all crimes (and almost all behaviours) the likelihood that the behaviour will reappear decreases the longer the person has abstained from that behaviour.  The recidivism rate within the first two years after release from prison is much higher than the recidivism rate between years 10 and 12 after release from prison.”[56]  Harris and Hanson found that the 5 year recidivism rate for offenders who have been out  of prison five years was 7%; among offenders out 10 years, 5%; and among offenders out 15 years, 4%. [57]  

As previously noted, the Harris and Hanson study’s results for overall recidivism study was not used to determine American re-offense rates because the Harris and Hanson study included foreign recidivism data. However, many advocate groups quote this study, at least seemingly in part because it gives a higher number than American studies. The Harris and Hanson study offered recidivism rates for 5 year increments, with an overall cumulative recidivism rate after 14% after 5 years, 20% after 10 years, and 24% after 15 years. Put another way, in the first five years, overall re-offense rates were 14%, but between years 5-10, 6% of the cumulative total reoffended, and between 10-15 years, only 5% reoffended, which validates the study’s proclamation that “an increase in the number of recidivists is not the same as an increase in the yearly rate of recidivism.

Harris and Hanson conducted a follow-up study in 2014, categorizing subjects by perceived risk levels based upon actuarial risk-assessment test (specifically, the Static-99R). “As can be seen from Figure 1, the risk of reoffending was highest in the first few years following release, and declined thereafter. This pattern was particularly strong for the high-risk offenders. During the first year after release, 7% reoffended, and during the first five years after release, a total of 22% reoffended… No high-risk sexual offender in this sample reoffended after 16 years offense-free (126 high-risk cases started year 17, of which 61 were followed for 5 years or more). … Whereas the 10-year sexual recidivism rate of the high-risk offenders from time of release was 28.8%, the rate declined to 12.5% for those who remained offense-free for 5 years, then 6.2% for those who remained offense free for 10 years… A 10-year sexual recidivism rate of 6.2% for the high-risk group (10 years offense-free) was less than the expected rate of moderate risk offenders from time-at-release (10.4%)… Inspection of Table 2 indicates that the expected recidivism rates were approximately cut in half for each 5 years that the offender was sexual offense–free in the community. For example, the 5-year sexual recidivism rate of the high-risk groups was 22.0% at release, 8.6% after 5 years, and 4.2% after 10 years offense-free. The same pattern applied to the moderate risk offenders (and the full sample). In contrast, the recidivism rates for the low-risk offenders were consistently low (1%-5%), and did not change meaningfully based on years offense-free. For example, the 10-year sexual recidivism rate for the low-risk offenders was 3.1% from time of release and 3.4% for those who remained offense-free in the community for 10 years.” [58]

The 2014 Hanson et al. study concluded, “The purpose of this study was to examine the extent to which high-risk sexual offenders remain high risk over time. As has been found for general offenders and violent offenders, the risk of sexual recidivism was highest in the first few years after release, and then decreased the longer they remained offense-free in the community. The decline in hazard rates was greatest for sexual offenders who had been identified as high risk at time of release. For low-risk offenders, time free had little influence: their risk was consistently low (1%-5%). The same relative risk reductions were observed for subgroups categorized by age at release, treatment involvement, country, and victim type. The current findings indicate static risk factors (e.g., prior offenses, victim characteristics) are valid, but time-dependent, markers for risk-relevant propensities. If high-risk sexual offenders do not reoffend when given the opportunity to do so, then there is clear evidence that they are not as high risk as initially perceived. The current study found that, on average, their recidivism risk was cut in half for each 5 years that they remained offense-free in the community.” [59]

The 2014 Hanson et al. study found that in time, the recidivism risk posed by most offenders becomes equal with that of-non-sexual offenders released from prison, and recidivism risk is generally cut in half for each five-year increment an offender is free in the community: “Although the moral consequences of sexual offending may last forever, the current results suggest that sexual offenders who remain offense-free could eventually cross a ‘redemption’ threshold in terms of recidivism risk, such that their current risk for a sexual crime becomes indistinguishable from the risk presented by nonsexual offenders. Previous large sample studies have found that the likelihood of an “out of the blue” sexual offense committed by offenders with no history of sexual crime is 1% to 3%: 1.1% after 4 years (Duwe, 2012); 1.3% after 3 years (Langan, Schmitt, & Durose, 2003); 3.2% after 4.5 years (Wormith, Hogg, & Guzzo, 2012). In comparison, only 2 of 100 moderate-risk sexual offenders in the current study committed a new sexual offense during a 5-year followup period if they were able to remain 10 years offense-free in the community. The high-risk offenders in the current sample, however, never fully resembled nonsexual offenders. Although their recidivism rates declined substantially when they were 10 years offense-free, the 5-year recidivism rate of the initially high-risk offenders (4.2%) was still higher than the expected rate for nonsexual offenders (1%-3%).” [60]

The bottom line is that even for those registrants labeled a “high risk,” recidivism dramatically decreases, and that recidivism decreases dramatically within the first three to five years of release. Many researchers agree that emphasis for rehabilitation should focus on the first three years of release.

Offense Types do play a role in recidivism, but beware of stat manipulation

While the Harris and Hanson study is not useful to determine American reoffense rates, it is still useful for showing recidivism differences between specific offense types, implying a need to divide recidivism rates by offender type. In order from, least likely to re-offend to most likely to re-offend (in parentheses, recidivism rates after 5 years, 10 years, and 15 years):

1.        Extended Incest Child Molesters (6%, 9%, 13%)
2.        “Girl Victim” Child Molesters (9%, 13%, 16%)
3.        Rapists (14%, 21%, 24%)
4.        “Boy Victim” Child Molesters (23%, 28%, 35%)

Also worth noting, the Harris and Hanson study also found that one time offenders (10%, 15%, 19%) reoffended lower than those who had two or more prior convictions (25%, 32%, 37%). [61]  

A 2005 study by Seto and Eke found that those convicted of child pornography offenses alone reoffended at a lower rate than an offender arrested for both CP offenses and other sex crimes. They also found a 6% CP recidivism rate with a 4% rate for a contact sexual offense among 201 subjects of varying periods with an average follow-up period of about 2.6 years. “Child pornography offenders who had ever committed a contact sexual offense were the most likely to reoffend. These group differences could be detected even though the overall rate of sexual recidivism was low (4%). Only one of the offenders with only child pornography offenses committed a contact sexual offense in the follow-up period.” [62]  

A Marshall and Barbaree 1990 study found an even greater discrepancy between those sex offenders with the lowest recidivism rates (Incest offenders, 4% -10%) and those with the highest (Exhibitionists, 41%-71%). [63]  

Breaking the general “sex offender” group into specific offense categories could be seen as integral for risk management.

The Myth and Misrepresentation of “Unique Risk” and the “Degree of Specialization”

“Sex offenders are four times more likely to re-offend sexually than non-sex offenders” is a mantra that proponents of sex offender legislation love to quote often, even influencing the outcome of the McKune v Lile case. The 2003 US Dept. of Justice study notes 5.3% of sex offenders released in 1994 were arrested for a new sex offense in 3 years, while 1.3% of all non-sex offenders were arrested for a sex crime. There are two issues with this line of logic.

First of all, specialization is no more common for sex offenders than any other type of offender; thieves are more likely to steal, drug offenders are more likely to commit drug crimes, et cetera. Furthermore, strong evidence suggests people convicted of sexual offenses actually specialize at a lower rate than most other offenses. The 2000 Michigan Parole Board study shows people convicted of sexual offenses committed subsequent sexual offenses (2.46%) at a lower rate than forgers committing forgery (6.86%), burglars committing burglary (10.56%), robbers committing robbery (5.17%), drug offenders committing drug crimes (6.42%),  and larcenists committing larceny (12.65%). [64]

“Langan and Levin found that 2.5% of  rapists were rearrested for rape within three years of release from prison and the DOJ found that 3.3% of child molesters were arrested for another sex crime against a child during that same period. In contrast, during that same three year period, Langan and Levin found that 13.4% of robbers were rearrested for robbery; twenty-two percent of assailants were rearrested for assault; 23.4% of burglars were rearrested for burglary; 33.9% of thieves were rearrested for larceny/theft; 11.5% of car thieves were rearrested for the same; and 41.2% of drug offenders were rearrested for a drug crime. The only released offenders who had a lower specialized recidivism rate than rapists and child molesters were those who had been convicted of homicide. Just 1.2% of offenders were rearrested for homicide within three years after release from prison”[65]

Secondly, the statement takes places emphasis on people convicted of recidivist sexual offenses, while failing to address the fact people convicted of sexual offenses commit only a small percentage of sex crimes that occur every year. Going by number of offenders rather than percentages, The 2003 US Dept. of Justice study found that 517 sex offenders committed a new sex crime, while 3,228 non-sex offenders committed sex crimes, roughly six times the number of sex offenders committing sex crimes. Consider also the following statistic from the US Dept. of Justice (1997) study which found 86% (about 6 of 7) of inmates in prison on sex crimes were first time offenders; only 14% had previous sex crimes.[66]  Thus, this misleading statement convolutes the bigger picture and is a nonsensical stat promoted simply to increase fear of people convicted of sexual offenses.

The Underreporting Myth

One of the most prevailing myths tied to recidivism as a way to downgrade the dozens of studies concluding re-offense rates among those convicted of sexual offenses as low is the opinion that large amount of recidivism goes undetected. This myth persistently pops up in conversations about recidivism despite a lack of evidence of any connection between recidivism rates and reporting. The statement is a red herring designed to connect the two assumptions together to make the receiver of the message assume a high number of unreported sex crimes must be the result of registered sex offenders, in turn justifying stricter sanctions. Because there is no true way of proving or disproving the amount of unreported sex crimes, victim industry advocates can proclaim astronomical claims of 90% or more undetected crim4es without worrying about expert rebuttals.

The 2004 Hanson and Harris study discussed the difficulty in addressing underreporting long before the explosion of sexual misconduct claims in the latter half of the 2010s:

“The Besserer and Trainor (2000) study showed that sexual assault had the highest percentage of incidents that were not reported to police (78%).  When respondents were asked why they did not report sexual victimization to the police, 59% of the respondents stated that the “incident was not important enough” to report.  Consequently, readers may wonder what counts as a sexual assault. The Besserer and Trainor (2000) victimization study used a very broad definition of sexual assault.  They counted all attempts at forced sexual activity, all unwanted sexual touching, grabbing, kissing, and fondling, as well as threats of sexual assault (Jennifer Tuffs, personal communication, January 15, 2003). Their broad definition undoubtedly included some behaviours that do not conform to the popular image of a sexual offence. All unwanted sexual advances are wrong, possibly criminal, and have the potential to do psychological harm to the victim.  As a society, however, we need to decide whether we wish to count an unwanted touch on the buttocks as an unreported sexual crime.   Coming to an agreement on what constitutes a sexual crime will be a difficult task.”[67]

The US Department of Justice conducts the “National Crime Victimization Survey” (NCVS) biennially, which measures various crime statistics, whether reported or not. The NCVS studies estimates of non-reporting and the reasons for non-reporting. The 2003 NCVS found that 67.3% of rapes and attempted rapes (attempts included verbal threats of sexual violence) and 53.2% of sexual assaults go unreported (Table 91). The NCVS found numerous reasons for lack of reporting; the reasons are as follows: Private/ Personal matter (17.1%), Offender unsuccessful (11.4%), Fear of reprisal (11.2%), Lack of proof (10.6%), Police would not want to be bothered (7.4%), Reported to another official (7.1%), Not aware crime occurred until later (4.9%), Not important enough (3.7%), Police inefficient, ineffective, biased (3.7%), Too inconvenient, time consuming (3.7%), No way to ID (2.8%), and “Other reasons” (18%). [68]

One major influence in the decision to report a crime is victim-offender relationship.  The 2005 NCVS [69] Table 104 illustrates this major difference between non-reporting with a stranger offender and an offender known to him. Among the top reasons for non-reporting when a stranger is involved: Reported to another official (49.6%), Police don’t want to  be bothered  (19.9%), Offender unsuccessful (12.7%), Fear of reprisal (11%), and Police inefficient, ineffective, or biased (6.9%). Among the top reasons for non-reporting when a non-stranger is involved: Other Reasons (47%), Private or personal matter (31.1%), Police don’t want to be bothered (10.8%), Report to another official (5.1%), Police ineffective, inefficient, biased (2.8%), and Too incontinent, time consuming (2.4%).

The NCVS defines sexual assault as “A wide range of victimizations, separate from rape or attempted rape. These crimes include attacks or attempted attacks generally involving unwanted sexual contact between victim and offender. Sexual assaults may or may not involve force and include such things as grabbing or fondling. Sexual assault also includes verbal threats.” Rape is defined as “Forced sexual intercourse including both psychological coercion as well as physical force. Forced sexual intercourse means vaginal, anal or oral penetration by the offender(s). This category also includes incidents where the penetration is from a foreign object such as a bottle. Includes attempted rapes, male as well as female victims and both heterosexual and homosexual rape. Attempted rape includes verbal threats of rape.”

The NCVS is very thorough, but there are a few problems. Firstly, it combines actual crimes with threats of crimes in many places; Table 1 of the NCVS divides reported rapes (72,240) and attempted rape (44,650), but combines sexual assault with attempt (81,950). Secondly, the reports are limited to crimes involving people age 12 and older, no children.  Thirdly, the phrase “verbal threats” is a rather vague term. Still, the NCVS is the most likely source for under-reporting estimates. That being said, there are many reasons why certain crimes go unreported.

In an attempt to rectify certain issues  with the NCVS, the DOJ conducted a study called “Sexual Assault of Young Children as Reported to Law Enforcement: Victim, Incident, and Offender Characteristics” in 2000. [70]  Since the NCVS only relates to victims 12+ years of age, the study relied on the FBI’s National Incident-Based Reporting System (NIBRS).  While there is no feasible way to survey children on unreported sex crimes, this study offers up another X factor to further complicate the issue: the number of sexual assaults committed by juveniles who are likely not a registered sex offender. The study found that juveniles made up 23.2% of total sex crimes against children: 40% of offenders of children under age 6 and 39% of offenders against children ages 7-11 were juveniles; and the single age with the greatest number of offenders is 14 (p. 8).

The unreported numbers reported in the NCVS had been steadily declining over the first decade of the new millennium, yet is rising in the 2010s. No adequate explanation or assumption is offered in the NCVS reports. It could be speculated that increased exposure to awareness campaigns such as the campus sexual assault stories, #MeToo and sex trafficking stories that have dominated the airwaves during the 2010s have influenced these numbers, but there are no studies that have studied this phenomenon, so the impact of these campaigns on the number of unreported reports cannot be full determined (remember: correlation is not causation). The numbers from the NCVS below should be taken only with extreme caution given the nature of our ever-changing definitions of sexual assault.

  • 2003 NCVS: 67.3% of rapes and attempted rapes (attempts included verbal threats of sexual violence) and 53.2% of sexual assaults go unreported
  • Between 2003-2005, sex assaults/ rapes were consolidated
  • 2005 NCVS: 61.7% of rapes/ sexual assaults go unreported
  • 2010 NCVS: 50% of rapes/ sexual assaults go unreported [71]
  • 2014 NCVS: 66.4% of rapes/ sexual assaults go unreported [72]
  • 2016 NCVS: 77% of rapes/ sexual assaults go unreported [73]

Researcher Debra Patkin states that under-reporting for crime in general is under-reported; citing the 1999 NCVS, Patkin notes 64% of overall crime goes unreported [74]. Thus, the idea that sex crimes go reported at a uniquely higher rate than other crimes is another myth.

In short, it is rather difficult to sufficiently determine just how many factors impact the ability to estimate exactly how many crimes go unreported. On one hand, there are a number of sex crimes going unreported. On the other hand, what are the reasons for the under-reporting? Also, how do we even determine a base number to use to determine the total number of actual sex crimes being committed by people old enough to be sexual recidivists?  Do we count verbal threats as bona fide sex crimes or not? In fairness, what about the number of false allegations? A number of recent exonerations of convicted rapists as a result of DNA evidence immediately come to mind. And sadly, sex crimes are typically he-said-she-said events. Even if we could somehow come up with a formula that accurately portrays the amount of unreported crimes in general, how can we truly determine the number of registered citizens that have committed these acts? In reality, this line of thinking merely serves to attempt to downplay an undesired result, namely, people convicted of sexual offenses have a lower recidivism rate than other offenders.

The Impact of Megan’s Law on Recidivism

Proponents of Megan’s Law argue the low re-offense rates are the result of the impact of Megan’s Law, but no studies have reached such a conclusion. A 2008 study found no significant changes in sex crime rates in the 10 years prior to and after the enforcement of Megan’s Law in New York State in 1997. “[R]esults of the analyses indicate that the 1996 enactment of SORA (and thus the beginning of the registry) had no significant impact on rates of total sexual offending, rape, or child molestation, whether viewed as a whole or in terms of offenses committed by first-time sex offenders or those committed by previously convicted sex offenders (i.e., repeat offenders).”[75]  It is important to note that a number of researchers warn Megan’s Law might actually contribute to recidivism through the stress and ostracism it creates. Psychologist John Q. LaFond points out a Washington state study that found notification laws do not prevent crime; instead, it leads to quicker  arrest times, either by the constant scrutiny, or by disrupting employment, housing, and support for the sex offender, causing stress and increasing the likelihood of recidivism.[76]

“As former offenders are denied opportunities to reintegrate into society and stigmatized, they lose hope. Stigmatized offenders are more likely to recidivate than reintegrated offenders, as the resistance to recidivate diminishes among offenders who are ostracized. On the other hand, a ‘pro-social identity,’ including concrete recognition of their reform, is integral to reducing recidivism.”[77]

“Employment problems experienced by the RSO, and subsequent financial hardships, emerged as the most pressing issue identified by family  members. The likelihood of housing disruption was correlated with residential restriction laws; larger buffer distances led to increased frequencies of housing crisis. Family members living with an RSO were more likely to experience threats and harassment by neighbors. Children of RSOs reportedly experienced adverse consequences including stigmatization and differential treatment by teachers and classmates. More than half had experienced ridicule, teasing, depression, anxiety, fear, or anger. Unintended consequences can impact family members’ ability to support RSOs in their efforts to avoid recidivism and successfully reintegrate.”[78]

There is one other important issue to consider in discussions about disruptions to stable lives of former offenders– Failure To Register (FTR). Often, registrants arrested for failure to register can exacerbate the myth of recidivism since news reports often highlight the relatively few arrested during a “compliance check” operation. [79] A 2018 report published on The Appeal found that more than three times as many registrants were arrested for an FTR than for suspicion of committing a sex crime in Franklin Co., PA in 2016. [80] . A 2010 study found FTR had not influence on sex crime rates whatsoever:

“With respect to failure to register (FTR) as a sex offender, no significant differences were found between the sexual recidivism rates of registered offenders with FTR charges and those without FTR charges (11% vs. 9%, respectively). There was no significant difference in the proportion of sexual recidivists and nonrecidivists with registration violations (12% and 10%, respectively). Failure to register did not predict sexual recidivism, and survival  analyses revealed no significant difference in time to recidivism when comparing those who failed to register (M = 2.9 years) with compliant registrants (M = 2.8 years).” [81]

OUTRAGEOUS CLAIMS: Addressing the controversial Langevin and Prentky Studies

Two influential studies stand out as having the most outlandish recidivism claims, the 1997 Pentky study, which claims child molesters have a recidivism rate of 52%, and the 2004 Langevin study, which makes a bold claim that “sex offenders” have a 90%-94% recidivism rate. Oftentimes these studies are used by victim industry advocates and lawmakers when promoting draconian sex crime legislation.

The Prentky Study: Outdated, Limited Data

The 1997 Prentky study made the controversial claim that after 25 years sex offenders’ recidivism is 52% for child molesters and 39% for rapists.[82]  
Prentky’s findings were cited in amici briefs filed in favor of the registry during the Smith v Doe hearings. However, there were a number of issues with this study. First, eAdvocate points out Prentky himself offered a warning against applying his study to determine long-term recidivism in a subsequent study: “We would like to conclude with two important caveats. The obvious, marked heterogeneity of sexual offenders precludes automatic generalization of the rates reported here to other samples.” Prentky’s warning, however, was not in the study that first mentions this study, but in a later study in the series.[83]

The study involved recidivists who were civilly committed between 1959 and 1985. “It was also the case that Prentky’s sample of molesters were highly likely to recidivate because they were convicted of many more sex offenses (N=4.6) than a more typical sample of incarcerated sex offenders, about 90% of whom have been convicted of only one sex offense.”[84]  

One thing that critics have failed to point out is that the 52% number is not the actual re-offense rate, but an estimate called the survival or failure rate, “i.e., the estimated probability that child molesters would ‘survive’ in the community without being charged, convicted, or imprisoned for a sexual offense over the 25-year study period.”[85]  The actual re-offense rates of the previously-committed repeat offenders were actually similar to the findings by the later Harris and Hanson study. It should be noted the actual length of the study period varied by subject, since each subject was released over a 25-year span. Overall, 32% of those in the study were rearrested with an average length of 3.64 years before re-arrest, and a 25% reconviction rate with an average time frame of 3.98 years before reconviction.[86]  In terms of actual numbers, 78 out 115 subjects were never charged with any offense, 8 were charged but not convicted, and 29 were convicted of a subsequent sex crime.[87]  Like many other recidivism studies, most offenses occurred within the first few years of release.

Even the SMART Office recognizes the Prentky study’s flaws: “Prentky and colleagues acknowledged that generalizing the recidivism rates found in the study to other samples of sex offenders was problematic due to the ‘marked heterogeneity of sex offenders,’ but they also suggested that the ‘crucial point to be gleaned from this study is the potential variability of the rates’ and not the specific rates themselves.”[88]

In short, the Prentky Study is merely an estimated rate of success of a group of recidivists based upon a complex but convoluted formula. The actual rates of failure by the hard numbers don’t support the 52% estimate for the small number of repeat offenders, much less people labeled “sex offenders” as a collective unit. Even Prentky warns the results are not indicative of those on the registry in general.

The Langevin Study: The Study Purged Non-Recidivists from the Results

There is one major similarity between the Prentky Study and the Langevin study,[89] particularly the reliance on data from a single civil commitment center as well as the 25-year follow-up period. However, Langevin’s study differs in two ways—it purged people without rearrests after 15 years, and it expanded the definition of recidivist to include crimes committed before subjects were to be included in the study.

Writing for, Chris Dornin found that Langevin’s unusual methodology led to unusual results: “Langevin reported a 61.1 percent sex crime recidivism rate, including 51.1 percent for incest. The researchers also tabulated confessions the offenders made during counseling and new arrests that failed to bring convictions. Adding those presumed crimes to actual convictions increased the overall sexual recidivism rate to 88.3 percent, including 84.2 percent for incest. Measured this way, molesters of young children outside their own family had an even higher rate, 94.1 sex crime recidivism over 25 years. To this writer’s knowledge, that is the highest reported rate in any of the hundreds of existing recidivism studies. It underlies much of the widespread belief that all sex offenders are incurable and unrepentant.” [90]  

Non-recidivist records were purged from the Langevin study’s final tally. Karl Hanson, noting his understanding of the difficulty of maintaining records after 15 years, added, “It appears that the way in which Langevin, Curnoe, Fedoroff, Bennett, Langevin, Peever, Pettica, and Sandhu (2004) dealt with missing data inflated the observed recidivism rates of their sample. In an earlier report of the same data set, Langevin and Fedoroff (2000) reported that they were able to obtain follow-up criminal history records for 378 (54%) of the first 700 cases assessed at Dr Langevin’s clinic between 1969 and 1974. In the 2004 report, the offenders lacking criminal history records in 1994 and 1999 were eliminated from the sample. Such a decision would retain recidivists and eliminate non-recidivists. The RCMP policies for retaining criminal history records have changed through the years, but they have always been more inclined to keep the records of active offenders (recidivists) than to keep those of inactive offenders (non-recidivists). In 1993, for example, the purge policy was that all convictions would be removed from the RCMP database if the offender (a) died, (b) was pardoned, or (c) attained age 70 and had no criminal activity for 10 years, or if (d) there was no recorded criminal activity for 15 years, provided that the offender was not the subject of a current investigation (Purge Unit, RCMP 1993). Given that the follow-up period in Langevin’s study was more than 15 years, all of the inactive (non-recidivist) offenders should have been deleted from the RCMP database. By selecting offenders from the 1960s with active criminal history records in the 1990s, researchers should expect to find recidivism rates close to 100%” [91]  

Hanson also noted that Langevin “used an unusual definition of recidivism that includes both prior offences and…future offences (the usual definition of recidivism)…. In the 378 cases with follow-up information, 139 (36.8%) had subsequent convictions for sexual offences. This figure is very similar to the long-term sexual recidivism rate (35.1%) found for another group of sexual offenders (child molesters) from Ontario in the 1960s and 1970s followed using RCMP records (Hanson, Scott, and Steffy 1995).” [92]

Indeed, Langevin’s definition to recidivism included crimes committed BEFORE the study began, which is different than every study before or after this study was published. “In a rebuttal entitled “Results by Design: The Artefactual Construction of High Recidivism Rates for Sex Offenders,” Webster said more than half the individuals in the sample were already recidivists by Langevin’s definition at the time of their evaluations, thus ensuring at least a 50 percent recidivism rate. In the rest of the literature on criminology and in the popular press, recidivism generally means a new crime committed after release from prison.” [93]

A 2006 study also condemned the Langevin study as fatally flawed: “A detailed analysis of the study demonstrates that this unusually high level is uninterpretable because the offenders whose criminal careers were followed are unlikely to be representative of sex offenders in general. Furthermore, the measure of recidivism used in the study not only distorts the normal meaning of recidivism but also artefactually creates an inflated – and consequently meaningless – recidivism rate.” [94]  

The Langevin study relied on a flawed definition of recidivism and purging non-recidivists (or at least a lack of any evidence of recidivism) to come up with a meaningless study.


A 2018 study by Tamara Rice Lave and Franklin E. Zimring found the California Department of Mental Health suppressed a study by Dr. Jesus Padilla showing just 6.5% of untreated sexually violent predators were arrested for a new contact sex crime within 4.8 years of release from a locked mental facility (1.35% average annual rate). “A person with a score of six on the Static-99 was estimated as having a 36% chance of being convicted of a new sexually violent offense within five years of release, a 44% chance of being convicted of a new sexually violent offense within ten years of release, and a 53% chance of being convicted of a new sexually violent offense within fifteen years of release.136 That means that the released SVPs performed much better than expected based on their Static-99 score. The difference is that much more striking considering that Padilla used arrests to measure recidivism, and the creators of the Static-99 used convictions. Since many arrests do not end up in a conviction, the disparity would have been even greater if they had both used arrests as their basis of measurement.” [95]

Lave and Zimring found that efforts to censor Padilla began when  Jon de Morales, the new head of the Sex Offender Commitment Program, accused Padilla and Russell of illegally accessing criminal history from a particular database called CELTS. The doctor was cleared of any wrongdoing after a 6 month investigation in 2007. After Morales took over the Atascadero civil commitment program, Morales shut down Padilla’s study, forced Padilla to turn over all files, then had the files destroyed.

“Padilla and Russell attempted to continue the study. They were told that because they were doing basic research and not program evaluation, they would have to reapply. They submitted a new proposal to Atascadero, but were told to go through DMH. They sent the proposal to DMH, but DMH said they could not evaluate the proposal because they did not have a human subjects committee. They then sent the proposal to the Health and Human Services committee which oversaw DMH. Padilla received a call from the Head of Human Subjects who told him that it was program evaluation and not basic research. They then went back to DMH, but DMH said they had to go through Atascadero. They were now eighteen months into the process, and it took Atascadero a few more months to respond.Finally, on March 13, 2008, Morales sent them a memorandum informing them that they could not conduct the recidivism study because they would need “legislation or approval from the Department of Mental Health” to access the CLETS, and “[n]either ASH nor DMH would permit ‘volunteers’ to conduct this research.” At first, the Department of State Hospitals (DSH) denied having any info on Padilla’s study; only after Lave and Zimring shared Padilla’s documents did DSH make a serious attempt to find the study files. Upon receiving them, Padilla confirmed someone tampered with the files. [96]

Lave and Zimring concluded they have no definitive answer as to why the study was suppressed. “The only explanation we have comes from Jon De Morales’s June 8, 2007 memorandum in which he writes, ‘to conduct research of this type, one would need to follow Special Order 288 and gain separate approval from the DOJ to have access to criminal Offender records.’ We are dubious of this explanation for two reasons. First, we reviewed Special Order No. 228, and it does not say anything about the special approval Morales claimed was necessary. Second, just one month before, the independent investigation had determined that Padilla and Russell did nothing wrong in accessing these records, even without the special approval supposedly necessary.” However, one legitimate reason why the study was suppressed  was because recidivism rates were far lower than expected (6.5% after 5 years, far lower than the 36% estimated by the STATIC-99), a fact that surprised even Padilla. “Perhaps higher-ups at DMH had not initially paid attention to the study because they did not expect the results,” Lave and Zimring added. “Once Padilla testified, DMH may have realized the study had to be stopped because it threatened the legitimacy of the entire SVP program. As explained earlier, the only constitutionally acceptable rationale for SVP commitment is that offenders are so dangerous that they must be locked away, and this study showed otherwise. If the SVP law were to be declared unconstitutional, it would threaten the $147.3 million annual budget DMH (and now Department of State Hospitals) receives for the civil commitment program. People have done far worse than bury a study for a hundred million dollars.” [97]

Lave and Zimring argues the study is important because if it can be proven that most civilly committed sexual offenders do not re-offend, it may dispel the belief that civil commitment is a necessary institution. “[I]f SVP laws were to be declared unconstitutional, it would have a tremendous financial impact on the institutions used to house and treat SVPs.” Indeed, $288.8 million in taxpayer funds were at stake in this industry. [98]


For most of this article, the focus has been on sex-specific re-offenses rather than for overall recidivism rates. Unfortunately, the general public does not understand the difference, thanks in part to the ignorance of mainstream media and legislatures. When people think “recidivism,” they think about the number of those convicted of sex offenses committing another sex crime, not being returned to prison for any offense type. They will often confuse rates for any crime type (recidivism) and sex-specific (reoffense) rates and assume people convicted of sexual offenses are a unique threat to society.

In reality, sex crime recidivism is the least likely reason a person convicted of a sex offense returns to incarceration.

According to the 2014 California Dept. of Corrections and Rehabilitation Outcome Evaluation Report, 5522 (65.2%) of the 8471 persons convicted of sex offenses released from the CDRC in the 2009-2010 Fiscal Year were returned to prison within 3 years of release. But of those 5522 returns:

  • 5074 (91.9%) returned on a parole violation;
  • 294 (5.3%) returned on a new non-sex crime;
  • 109 (2%) returned for a “Failure To Register” Offense; and
  • 45 (0.8%) returned for a new sex offense [99]

Taking those not arrested into account, only 0.5% of persons convicted of sex offenses (45 of 8471) were returned to prison for a new sex offense within 3 years of release.

Later recidivism rates published by the California CDRC removes “parole violations” in the data and uses re-conviction rates instead of re-incarceration rates, but the end result still shows sex crime convictions are lower than arrests for Failure To Register cases. Of the 3,298 offenders convicted for sex offenses released in the 2014-2015 Fiscal Year, 1040 (31.5%) were re-convicted of a crime. Of those 1040 re-convictions:

  • 399 (38.4%) were convicted of a misdemeanor non-sex crime (38.4 percent or 399 offenders);
  • 379 (36.4%) were convicted of a felony non-sex offense;
  • 198 (19.0%) were convicted for a Failure To Register offense;
  • 44 (4.2%) were convicted of a felony sex offense; and
  • 20 (1.9%) were convicted of a misdemeanor sex offense. [100]

Taking those not arrested into account, only 1.9% of persons convicted of sex offenses (64 of 3298) were re-convicted a new sex offense within 3 years of release.

In August 2018, a review of the Franklin County PA arrest records by Joshua Vaughn of The Appeal found:

  • “An analysis of all charged criminal sexual offenses in the county in 2016 found only three cases where a defendant was on the registry when a new offense was committed. And more than 96 percent of defendants charged with a sexual offense in 2016 had no criminal history of sexual violence.”
  • “In Franklin County, more than three times as many people on the sex offender registry were charged failing to comply with registry requirements in 2016 than were charged with a new sexual offense.” [101]

A June 2021 study specific to non-production CP offense convictions found:

  • When tracking 1,093 nonproduction child pornography offenders released from incarceration or placed on probation in 2015, 27.6 percent were rearrested within three years. (This includes revokation of parole.)
  • Of the 1,093 offenders, 4.3 percent (47 offenders) were rearrested for a sex
    offense within three years.
  • Eighty-eight offenders (8.1% of the 1,093) failed to register during the three-year period.
  • The most common arrest (179 oe 16.4%) were for “Administration of Justice” charges. Administration of justice offenses include contempt of court, failure to appear, obstruction of justice, escape, prison contraband, and probation and parole violations (excluding FTRs).
  • Of those rearrested for a new sex offense, the median time to arrest (i.e., the point in timr where half of the total number of recidivists were arrested) was 12 months. 
  • Also worth noting: Of those who were were released from incarceration in 2015, 86.1% had little or no prior criminal history and placed in the lowest category, “Criminal History Category I” (CHC I). Less than 2% were category IV or higher. [102]


Trying to come up with a simple, yet conclusive, hard number for such a complex subject as re-offense rates for those convicted of sexual offenses is difficult. There should be a universal standard, however, and the evidence presented here suggests a superior method of presenting a true recidivism rate should be an average annual percentage rate based upon reconviction rates. Dividing the recidivism rate by the number of years in the numerous, recent, and existing studies that focused on reconviction rates show an average recidivism rate of roughly 1% per annum. One caveat in settling on a single number for determining recidivism is that it does not take into account the factors that influence an individual recidivism rate, such as length of time since the last offense/ release from corrections, the offense type, and the number of prior offenses may influence the overall rates. In addition, it is important to remember many recidivists will commit their offenses within the first three years or so after release.

Criticism of this method will undoubtedly come from agencies that rely on Predator Panic for funding; many rely on a combination of myths (more likely to reoffend, underreporting) and studies that promote their beliefs that all people who are convicted of sexual offenses reoffend (such as the Prentky and Langevin studies). However, each approach is the result of skewed numbers arising primarily from assumptions. The hard numbers simply do not support the assumption that those convicted of sex crimes pose a unique threat to society.

It is reasonable to conclude that dozens of state, federal, university, and even international studies have consistently concluded that recidivism among those convicted of sex crimes are extremely low. This conclusion invalidates the belief that a unique and draconian series of laws are necessary in our society.


  1. Jones, Brittany. “Petition gains support for ‘Cherish Law.’” ABC 27 WTXL. Raycom Media. 3 Sept. 2013. Web. <>
  2. Sullum, Jacob. “Perverted Justice.” 14 June 2011. Web. <>
  3. Abrahamsen, David. “Family Tension, Basic Cause of Criminal Behavior.” Basic Cause of Criminal Behavior, 40 J. Crim. L. & Criminology 330 (1949-1950). Journal.
  4. Bassler, Will. “So Why are the Reconviction Rates So Important?” 26 June 2017. Web. <>
  5. Jenkins, Philip. Moral Panic: Changing Concepts of the Child Molester in Modern America. Vail-Ballou Press. 1998. p39-40
  6. Ibid., p.15
  7. “The (New York City) Mayor’s Committee Reports on the Study of Sex Offenses.” Journal of Criminal Law and Criminology (1931-1951) Vol. 34,
    No. 5 (Jan. – Feb., 1944), pp. 324-327. Journal.
  8. Tappan, Paul. “New Jersey Commission on the Habitual Sex Offender.” Tenton Publishing. 1950. Book.
  9. Supra., Jenkins, p.98-100
  10. Ibid., p.120-126
  11. Liptak, Adam. “Did the Supreme Court Base a Ruling on a Myth?” NY Times. 6 Mar 2017. Web. <>
  12. Brown, Raymond C. and Moore, John E. “A Practitioner’s Guide to Treating the Incarcerated Male Sex Offender:Breaking the Cycle of Sexual
    Abuse..” US Dept. of Justice. 1988. Web. <>
  13. Freeman-Longo, R., & Wall, R, Changing a lifetime of sexual crime, Psychology Today (1986). Magazine. 
  14. There is a second myth here that must be addressed, the “much more likely to reoffend” myth. This statement will also be addressed in this report.
  15. Ira Ellman. “The Supreme Court’s Crucial Mistake About Sex Crime Statistics.” Casetext. July 25, 2015. Web. <
  16. Post, David. “The Volokh Conspiracy: More fuel for the movement to reform sex offender laws.” Washington Post. 18 Aug. 2015. Web. <>
  17. Supra., Liptak, “Based on Myth”
  18. Levenson, J.S., Branon, Y. N., Fortney, T., and Baker, J. “ Public Perceptions About Sex Offenders and Community Protection Policies.” Analyses of
    Social Issues and Public Policy, Vol. 7, No. 1., 1-25. 2007. Web. <>
  19. Sullum, Jacob. “The Lingering Impact of Justice Kennedy’s Trumpesque Claim About Sex Offenders.” 8 March 2017. Web. <>
  20. US Department of Justice, “Recidivism of Sex Offenders Released into the Community in 1994.” Nov. 2003.
  21. Harris, Andrew JR and Hanson, Karl. “Sex Offender Recidivism: A Simple Question.” Public Safety and Emergency Preparedness Canada. 2004. Web. <>
  22. Ibid., Brown & Moore 1988, p. 66
  23. See Logue, Derek. “The Complexities of Sex Offender Recidivism.” Once Fallen. 2008. Web <http://www.>
  24. Ibid., Ellman.
  25. Furby, Lita, Weinrott, Mark, and Blackshaw, Lyn. “Sex Offender Recidivism: A Review.” Psyhological Bulletin, Vol. 105, No. 1,3-30. 1989.
  26. United States v. Johnson, 588 F. Supp. 2d 997 (S.D. Iowa 2008)
  27. “Ten Year Recidivism Follow-up of 1989 Sex Offender Releases.” Ohio Department of Rehabilitation and Correction. April 2001.
  28. “How Dangerous is Lightning?” National Weather Service. 2017. Web. <>. For the record, between 1987-2016, there was an average of 47 annual fatalities, though between 2007-2016, that number is reduced to 30.
  29. Knight, Cameron. “Cincinnati homicide rate on roller coaster pattern.” USA Today Network. 3 Jan. 2017. Web. <>. Note the murder rate stays consistently in the 60-70 range though fluctuates in small numbers.
  30. Note: A more complete recidivism study summary can be found at; for the sake of brevity, not all studies listed on the Once Fallen recidivism chart are listed here.
  31. “Profile and Follow-up of Sex Offenders released in 1986.” New York State Department of Correctional Services, Division of Program Planning,
    Research and Evaluation.  July 1996
  32. “THE IOWA SEX OFFENDER REGISTRY AND RECIDIVISM.” Iowa Department of Human Rights, Division of Criminal and Juvenile Justice
    Planning and Statistical Analysis Center.  December 2000. Web.<>
  33. Supra, Ohio DRC 2001
  34. Supra. US DoJ
  35. eAdvocate, “CHART: Michigan Recidivism Rates: All released sex offenders -vs- non-sex offenders.” May 5, 2009. Information was extrapolated from Annual Michigan Department of Corrections, Statistical Report, Parole Board Charts D2 and D2a, years 1990 through 2000. Technical violations not included. The Statistical reports for years 1998 to current can be found under the “Publications and information” section of the Michigan Dept. of Corrections website,,4551,7-119-1441—,00.html.
  36. “Sex Offender Recidivism in Minnesota.” Minnesota Dept. of Corrections. April 2007
  37. “RECIDIVISM OF PAROLED SEX OFFENDERS—TEN (10) YEAR STUDY. California Sex Offender Management Board.  June 2008
  38. “Recidivism of Sex Offenders Released from the Arizona Department of Corrections in 2001.” Arizona Criminal Justice Commission. Feb. 2009
  39. “Recidivism among sex offenders in Connecticut.” State of Connecticut, Office of Policy and Management, Criminal Justice Policy & Planning
    Division. February 15, 2012
  40. “Nebraska Sex Offender Registry Study.” Consortium for Crime and Justice Research, U. of Nebraska – Omaha. July 31, 2013. Table 5, p.20. It is
    important to note that the study also found a 1 year pre-AWA re-offense rate of 18 of 2809 (0.6%), as well as a 5 year average of 162 of 2832 (5.7%) in a 5 year study of 4 states
  41. Ibid., Table 6; It is important to note the study also found a 1 year AWA re-offense rate of 4 of 230 (1.7%)
  42. Barbara Levine & Elsie Kettunen. “Paroling people who committed serious crimes: What is the actual risk?”  Citizens Alliance on Prisons & Public
    Spending, Dec. 1, 2014
  43. Supra, US DoJ
  44. Michelle L. Meloy, “The Sex Offender Next Door: An Analysis of Recidivism, Risk Factors, and Deterrence of Sex Offenders on Probation.” Criminal Justice Policy Review, Volume 16, Number 2, June 2005.  p. 211-236
  45. Alaska Judicial Council. “Criminal Recidivism in Alaska.” January 2007. This study does not break down the number or registrants nor offer the exact number rearrested in the study.
  46. Stan Orchowsky and Janice Iwama. “Improving State Criminal History Records: Recidivism of Sex Offenders Released in 2001.” Justice Research and Statistics Association, November 2009. Table 5, p. 17.
  47. Richard Tewksbury, Wesley G. Jennings and Kristen M. Zgoba. “A longitudinal examination of sex offender recidivism prior to and following the
    implementation of SORN.” Behav. Sci. Law 30: 308–328 (2012)
  48. Kristen M. Zgoba, Michael Miner, Raymond Knight, Elizabeth Letourneau, Jill Levenson, David Thornton. “A Multi-State Recidivism Study Using
    Static-99R and Static-2002 Risk Scores and Tier Guidelines from the Adam Walsh Act.” November 2012.  Table 7, p.20
  49. “Recidivism of Prisoners Released in 30 States in 2005: Patterns from 2005 to 2010: Supplemental Tables: Most serious commitment offense and types of post-release arrest charges of prisoners released in 30 states in 2005.” US Dept. of Justice. Dec. 2016.
  50. Seung C. Lee, Alejandro Restrepo, Annie Satariano, & R. Karl Hanson. “The Predictive Validity of Static-99R for Sexual Offenders in California: 2016 Update.” State Authorized Risk Assessment Tool for Sex Offenders (CA) Committee. Gov’t Report. July 13, 2016.
  51. Notice a discrepancy between the annual average rate and the one year rate in the UNO study.  It is important to note that the study also found a 1 year pre-Adam Walsh Act re-offense rate of 18 of 2809 (0.6%) and a 1 year AWA re-offense rate of 4 of 230 (1.7%) in the study. The second year numbers deviate somewhat from the first year numbers. though the minor differences in the 1 year rate and the average annual rate are insignificant. The average annual rate should be different from a 1 year study result since the average annual rate is an average from a number of years. Thus, the one year recidivism rate for Nebraska was 0.6% for pre-AWA registrants, while the average annual rate from 2 years of data is 0.85%. Both numbers are statistically correct.
  52. See note 50 as this result was from the same study
  53. Supra., Ohio DRC 2001, Executive Summary
  54. New York State Division of Probation and Correctional Alternatives. “Research Bulletin: Sex Offender Populations, Recidivism and Actuarial
    Assessment.” 2009. <>
  55. California Dept. of Corrections. “Recidivism of Paroled Sex Offenders- A 10 year study [Chart].” 2009. Found online at <>
  56. Hanson, Karl and Harris, Andrew. “Sex Offender Recidivism: A Simple Question.” Public Safety and Emergency Preparedness Canada. March  2004. Found online at <>, p.1
  57. Ibid., p.8, Table 2
  58. R. Karl Hanson, Andrew J. R. Harris, Leslie Helmus and David Thornton. “High-Risk Sex Offenders May Not Be High Risk Forever.” J Interpers
    Violence, 2014 29: 2792 originally published online 24 March 2014. pgs.2796-2801
  59. Ibid., pgs.2801, 2805
  60. Ibid., pgs. 2806-2807
  61. Supra., Harris & Hanson 2004, p.8 Table 2
  62. Michael C. Seto and Angela W. Eke. “The Criminal Histories and Later Offending of Child Pornography Offenders.” Sexual Abuse: A Journal of Research and Treatment, Vol. 17, No. 2, April 2005.
  63. Marshall, W.L. & Barbaree, H.E. (1990). Outcomes of comprehensive cognitive-behavioral treatment programs. In  W.L. Marshall, D.R. Laws, and H.E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 363-385). New York: Plenum.
  64. Michigan Dept. of Corrections (2000). Recidivism Statistics: from Annually Published Michigan Dept. of  Corrections, Statistical Reports (Parole
    Board) Cumulative Parole Board Statistics for Parolees 1990 through 2000.
  65. Lave, Tamara Rice and Zimring, Frank E. “Assessing the Real Risk of Sexually Violent Predators: Doctor Padilla’s Dangerous Data.” American
    Criminal Law Review, Vol. 55, 2018., Retrieved 13 Aug 2018. p.719
  66. US Department of Justice, “Recidivism of Sex Offenders Released into the Community in 1994.”
  67. Supra, Hanson & Harris 2004, p.2
  68. “Criminal Victimization Statistical Tables, 2003. “US Sept. of Justice. 2004. Web. <>
  69. “Criminal Victimization Statistical Tables, 2005. “US Sept. of Justice. 2006. Web. <>
  70. “Sexual Assault of Young Children as Reported to Law Enforcement: Victim, Incident, and Offender Characteristics.” US Dept. of Justice. 2000. Web. <>
  71. “Criminal Victimization, 2010. “US Sept. of Justice. 2011. Web. <>
  72. “Criminal Victimization, 2014. “US Sept. of Justice. 2015. Web. <>
  73. “Criminal Victimization, 2016. “US Sept. of Justice. 2017. Web. <>
  74. Patkin, Debra. “Megan’s law and the misconception of sex offender recidivism.” 2008. <
  75. Sandler, Jeffrey C., Freeman, Naomi J., and Socia, Kelly M. “DOES A WATCHED POT BOIL? A Time-Series Analysis of New York State’s Sex
    Offender Registration and Notification Law.” Psychology Public Policy and Law · November 2008. Web.
  76. John Q. LaFond, “Preventing Sexual Violence.” APA. Book.  2005
  77. Hollida Wakefield, “The Vilification of Sex Offenders: Do Laws Targeting Sex Offenders Increase Recidivism and Sexual Violence?” Journal of Sex
    Offender Civil Commitment: Science and the Law, 2006, p. 141-149
  78. Levenson, J. S., & Tewksbury, R. (2009). “Collateral damage: Family members of registered sex offenders.” American Journal of Criminal Justice
  79. See, for example, Tana Weingartner, “Twenty arrested during sex offender compliance sweep,” WVXU, May 28, 203, http://www.wvxu.
    org/post/twenty-arrested-during-sex-offender-compliance-sweep#stream/0, Retrieved Aug. 5, 2018; and “Deputies check addresses on hundreds of Hamilton Co. sex offenders,” WLWT, May 28, 2013,, Retrieved Aug. 5, 2018. In both articles, the spotlight was on 20 arrests, in particular, one arrest involving a gun collection by a parolee. Pictures of the gun collection were prominently displayed. The officers referred to the operation as “spring cleaning.” In reality, none were arrested for a new sex crime.
  80. Joshua Vaughn, “FAILURE-TO-COMPLY ARRESTS REVEAL FLAWS IN SEX OFFENDER REGISTRIES.” The Appeal,  Aug. 1, 2018.,  Retrieved Aug. 5, 2018
  81. Elizabeth J. Letourneau, Ph.D., Jill S. Levenson, Ph.D., Dipankar Bandyopadhyay, Ph.D., Debajyoti Sinha, Ph.D., Kevin S. Armstrong. “Evaluating the
    Effectiveness of Sex Offender Registration and Notification Policies for Reducing Sexual Violence against Women.” Sept. 2010., Retrieved Aug. 5, 2018.
  82. Prentky, R. A., Knight, R. A., & Lee, A. F. S. (1997, June). Child sexual molestation: Research issues. Research Report, National Institute of Justice,
    Washington, D.C. Web. <>
  83. eAdvocate. “Special Report: Misquoting of Prentky’s 1997 Long Term Recidivism Study: Affecting a MAJOR US Supreme Court Decision.” Truths,
    Authority and Factoids. June 2005. Web. <>; See also Prentky, R. A., Lee, A. F. S., Knight, R. A., & Cerce, D. (1997 Dec). Recidivism rates among child molesters and rapists: A methodological
    analysis. Law and Human Behavior, 21, 656-657
  84. Richard Wollert, Ph.D. “An Analysis of the Argument That Clinicians Under-predict Sexual Violence in Civil Commitment Cases.” Behav Sci  Law.
    2001;19(1):171-84. Found online at <>
  85. Supra, Prentky 1997 “Research Issues.” p.11
  86. Ibid., p.12
  87. Ibid., p.14, Exhibit 5
  88. Roger Przybylski. “Chapter 5: Adult Sex Offender Recidivism.” SMART Office. SOMAPI. 2018. Web. <>
  89. Ron Langevin, Suzanne Curnoe, Paul Fedoroff, Renee, Mara Langevin, Cheryl Peever, Rick Pettica, and Shameen Sandhu. “Lifetime Sex Offender
    Recidivism: A 25-Year Follow-Up Study.” Canadian Journal of Criminology and Criminal Justice, Vol. 46, No. 5. DOI: 10.3138/cjccj.46.5.531 (2004)
  90. Dornin, Chris. “Facts and Fiction about Sex Offenders.” 22 May 2010.   Web. <
  91. Hanson, Karl. “Long-Term Follow-Up Studies Are Difficult: Comment on Langevin et al. (2004).” Canadian Journal of Criminology and Criminal
    Justice, Jan 2006. DOI: 10.3138/cjccj.48.1.103 <
  92. Ibid.
  93. Supra., Dornin 2010
  94. Webster, Cheryl Marie, Gartner, Rosemary, and Doob, Anthony. “Results by Design: The Artefactual Construction of High Recidivism Rates for Sex
    Offenders.”  CAN J CRIMINOL CRIM JUSTICE. 48. 79-93. 10.1353/ccj    2006.0013 Web.  <https://www.researchgate.
    net/publication/240747161_Results_by_Design_The_Artefactual_Construction of_High_Recidivism_Rates_for_Sex_Offenders>
  95. Lave and Zimrng, “Padilla”, p.723
  96. Ibid., p.724-725
  97. Ibid., p.727
  98. Ibid., p.737-738
  99. “2014 Outcome Evaluation Report.” California Dept. of Correction and Rehabilitation, Office of Research. July 2015. Pages 29-30. Accessed 8 July 2020 at
    AND REHABILITATION IN FISCAL YEAR 2014-15.” California Dept. of Correction and Rehabilitation, Office of Research. January 2020. Pages 34-
    35. Accessed 8 July 2020 at
  101. Joshua Vaughn. “FAILURE-TO-COMPLY ARRESTS REVEAL FLAWS IN SEX OFFENDER REGISTRIES,” The Appeal, 1 Aug 2018. Accessed 25
    Aug 2020 at
  102. “Federal Sentencing of Child Pornography: Non-Production Offenses.” US Sentencing Commission. June 2021. Accessed 13 Oct 2021 at