Massachusetts State House.
Boston Bar Journal

Risky Policies: How Effective Are Restrictions on Sex Offenders in Reducing Reoffending?

October 07, 2014
| Fall 2014 Vol. 58 #4

Tennen_Ericby Eric Tennen

Legal Analysis

Sex offenders are one of the most reviled groups of felons.  In the past few decades, there has been a dramatic increase in laws and regulations governing virtually every facet of their lives.  The policy decisions about how to deal with sex offenders flow largely in one direction: restrict them as much as possible.  This one-sided approach has created a complex web of restrictions that ultimately causes more harm than good.

Differentiating sex offenses from other types of offenses is perfectly rational.  Further, government should try to address those harms by creating policies that will reduce offenses.  In light of what we know today about sex offenses and sex offenders, it is possible to reduce the rate of recidivism, but the policies currently in place are not the answer.  They are reactive, based often on outdated misconceptions, and far too broad.

Evidence-based practices have become common in almost all areas of public policy, because it is now well established that conventional wisdom is often inaccurate. But though there is an abundance of empirical data about sex offenders and reoffending, our policies are created without reference to the data.  Rather, our policies are based on anecdotal information and inaccurate beliefs about sex offenders and reoffending.  As a result, current policies do not address the problems they are seeking to correct.  In most instances, the policies are ineffective; in many instances, they also create conditions that may increase the risk of reoffense.

Sex Offender Registry

Perhaps the most common restriction placed on sex offenders is the obligation to register.  Every state has a registry that requires sex offenders to register with some frequency and dictates what information will be made available to the public.  Though there are many other restrictions placed on sex offenders, the registry provides a good lens through which to evaluate the effectiveness of these policies.

In 1999, the Massachusetts Legislature created the current version of the Sex Offender Registry Board [“Board”]. A sex offender is defined as anyone convicted of an enumerated sex offense. G. L. c. 6 § 178C. The Board decides whether someone qualifies as a “sex offender” and then classifies them.  The Board also disseminates information to the public about offenders.  See G. L. c. 6 §§ 178C-178Q; 803 C.M.R. 1.00.

There are four possible outcomes to the classification process.  The Board may determine that an offender poses no risk to reoffend, in which event the offender is not required to register.  If the Board determines that an offender poses some risk of reoffense, he will be classified as either a Level 1, 2, or 3 offender. Level 1, 2 and 3 offenders are considered, respectively, to have a “low,” “medium,” and “high” risk of re-offense, and to pose an increasing danger to the public.  G. L. c. 6 § 178K (2)(a)-(c).  The Board is required by statute to promulgate regulations that govern how they conduct these evaluations; the regulations must weigh, but are not limited to, certain factors enumerated by the Legislature. G. L. c. 6 § 178K.

All sex offenders, regardless of levels, are regularly[1] required to update their residential address, any secondary address, and employment or attendance at a post-secondary institution of higher learning. Level 1 offenders are required to register by mail. Their information is not available to the public in any form. Level 2 and 3 offenders must register in person, at a police station. Information concerning Level 2 and 3 offenders is available to the public and is posted on the internet.[2] Additionally, Level 3 offenders are subject to “active” dissemination, which involves a “community notification plan” where police departments notify organizations and individuals in the community which are likely to encounter such a sex offender.  G. L. c. 6, § 178K(2)(c).  In addition, “[n]eighboring police districts [are required to] share sex offender registration information of level [three] offenders and may inform the residents of their municipality of a sex offender they are likely to encounter who resides in an adjacent city or town.” Id.  Regardless of one’s level, all sex offenders are subject to criminal penalties for failing to register, which can encompass anything from registering late or failing to update one’s information.  G. L. c. 6 § 178H.

Most offenders must register for life.  G. L. c. 6 § 178G.  However, an offender may move to lower his classification or be relieved of his obligation altogether. 803 CMR 1.37A-1.37B.  On the other hand, the Board may move to increase an offender’s level any time after classification upon receiving new information indicating an increased risk to reoffend. 803 CMR 1.37C(10).

Additionally, an offender’s level may affect other aspects of his life.  A growing number of residency restrictions prohibit Level 2 and 3 offenders from living within a certain distance from parks, schools, or other places.  See e.g. General Ordinances of the Town of Barnstable, Chapter 147, Article IV, § 147-12 (C); Ayer Town By-Law, Article XLIX.  A person classified as a Level 2 or 3 offender can never seal his criminal record.  G. L. c. 276 § 100A(6).  No Level 3 offender may live in a nursing home, G.L. c 6 § 178K(e), though the Supreme Judicial Court has said this provision is facially unconstitutional in certain, if not most, situations, Doe v. Police Cmmr. of Boston, 460 Mass. 342 (2011).  No sex offender, regardless of level, may operate an ice-cream truck. G.L. c. 265 § 48.  Any sex offender subject to lifetime registration is prohibited from receiving any federally assisted housing.  42 U.S.C. § 13663.

Do All Convicted Sex Offenders Pose a High Risk to Reoffend?

Registration and its collateral effects were designed to address a serious concern: the risk of sex offender recidivism. But since creation of the registry, we know more about sex offenders than when the law went into effect, and we also know more about the effectiveness of the registry in preventing reoffense.

Public perception is that sex offenders reoffend at high rates.  See e.g. McKune v. Lile, 536 U.S. 24, 32-33 (2002); Commonwealth v. Cory, 454 Mass 559, 574 (2009).  In reality, the most current research indicates that sex offenders, as a group, reoffend less than other criminal offenders as confirmed by federal, state, and academic studies.[3]  And because of the myth that sex offenders reoffend at very high rates, many people believe that most sex offenses are committed by repeat offenders.  In fact, up to 95% of all sexual offenses are committed by first time offenders.[4]  It is extremely difficult, if not impossible, to prevent first time offenders from offending; however, our current regulatory framework is directed exclusively at prior offenders.  Therefore, our policies can only be directed towards trying to prevent the relatively small proportion of all offenses committed by prior offenders. While that is a worthy goal, it must be considered in context.

Does Registration Work?[5]

Study after study shows that registries do not prevent reoffending.[6]  In fact, registration may actually increase the risk to reoffend in unintended ways.  Persons with stable, supportive lives, with steady employment and housing, reoffend at lower rates.[7]  On the other hand, those who, because of registration, are unemployed, homeless, and generally unstable, suffer psychosocial stressors that may increase their risk to reoffend. Id. These policies also put significant stress on offenders’ families, who in turn may abandon the offender; this aggravates an offender’s stress and also isolates him away from persons who might otherwise provide a watchful eye. Id.

The Registry is particularly troubling in its application to juveniles.  Juvenile sex offenders are less likely to reoffend than just about any other group of sex offenders.[8]  The juvenile justice system is designed to give all juveniles a second chance: proceedings are closed to the public, records are generally unavailable, and the goal of the juvenile justice system is rehabilitation rather than punishment.  However, the offender registry makes public the juvenile’s status as a sex offender and the crimes for which he was convicted.  For a juvenile who is seeking to put his past behind him, registration and dissemination is devastating.  Every other juvenile, no matter the conviction, has some assurance that if he avoids further trouble, his juvenile indiscretions will not haunt him.  The same is not true for juvenile sex offenders.  And, of course, registration destabilizes a juvenile’s support systems by, among other things, “adding to the [juvenile’s] family’s economic challenges, difficulty in securing or maintaining an approved residence, and straining or severing family relationships.”[9]

Characteristics Associated with Risk

The reason registries do not prevent reoffense is that they are predicated on outdated and inaccurate assumptions. Most sex offenses are committed by persons known to the victim (either a relative or acquaintance).[10]  Yet our policies are largely driven by the idea that sex offenders go after strangers.  Indeed, offending against strangers is correlated with a higher likelihood of reoffending.  But offenses against strangers are the exception.

Research has identified other specific factors that are correlated with a higher risk to reoffend.  Having a legitimate sexual disorder, such as pedophilia, is also correlated with a higher risk to reoffend.[11]   In certain situations, having a male victim (typically a male child victim) is correlated with increased risk.  So is having intimacy deficits—someone who has a hard time establishing lasting relationships.

On the other hand, certain groups of offenders pose much lower risks to reoffend than other sex offenders, regardless of the details of their offenses: people over 60, people whose offense(s) occurred when they were juveniles (under 18), females, people who have been in the community, without reoffending, for five years or more, and people convicted of possessing child pornography.

Then there are characteristics which are not correlated with an increased or decreased risk to reoffend. Thus, for purposes of whether someone is likely to reoffend, it is irrelevant whether they completed sex offender treatment, whether they deny their offenses, whether they use drugs or alcohol, or whether they were abused as a child.[12]

Lastly, if an offender does reoffend, it is often in a manner similar to his first offense.  It is very rare that a sex offender reoffends against different types of victims.  For example, if a person offends against pre-pubescent children, it is likely that any reoffense by that person will be against a pre-pubescent child.  In contrast, if a person assaults an adult, it is likely that any reoffense by that person will be against another adult.

While registration attempts to take some of these factors into account, it does not take them all into account.  And when the Legislature imposes collateral restrictions on all offenders who are required to register, e.g. residency restrictions or exclusion from nursing homes, it treats all offenders the same, despite their varying risks.


There are many restrictions that, if properly targeted, could help reduce the rate of sex offender recidivism.  Unfortunately, most current restrictions sweep too broadly.  Registration is but one example; there are more (e.g. residency restrictions[13] or GPS monitoring for all sex offenders on probation, regardless of risk to reoffend).  If public policy were based on the best available, current evidence about sex offenders, and less on the public’s perception of them, we could monitor, more efficiently, the true high risk offenders.  But that requires acceptance that sex offenders are a heterogeneous population and not all of them pose a grave risk to the public.  Only then could our policies better reflect the true risk that only a small percentage of convicted sex offenders pose to the public.


Eric Tennen is a partner at Swomley & Tennen, LLP. His practice focuses on criminal defense, specializing in sex offenses and related matters.


[1]  Sex offenders must register annually unless they are homeless, wherein they must register every 30 days. G.L. c. 6 § 178F ½.

[2]  Prior to July 12, 2013, only Level 3 offenders’ information was available on the internet. The law was amended to include Level 2 offenders. However, the Supreme Judicial Court ruled that only Level 2 offenders classified after July 12, 2013 can be subjected to on-line dissemination. Thus, there are currently two classes of Level 2 offenders: those whose information is on-line and those whose information is not. See generally Moe v. SORB, 467 Mass. 598 (2014). For those whose information is not on-line, their information may still be disclosed if a member of the public actively requests it. Id. at 601-602.

[3]  See e.g. U.S. Department of Justice, Bureau of Justice Statistics, Recidivism of Sex Offenders Released from Prison in 1994 (2002) (after three years, 5.3% of sex offenders were rearrested for a new sex crime and 3.5% were convicted of a new sex crime); Comprehensive Recidivism Study, Massachusetts Sentencing Commission, June 1, 2002, pg. 38 (“Of the major offense categories, recidivism rates were lowest for sex offenders (20.8%) and highest for property offenders (56.5%)”); Hanson, K.R.; Bussière, M.T., Predicting relapse: A meta-analysis of sexual offender recidivism studies, Journal of Consulting and Clinical Psychology, Vol 66(2), Apr 348-362,(1998)(13.4% sexual recidivism for all offenders in meta-analysis of 61 studies and 23,400 offenders); Hanson, K.R.; Morton-Bourgon, K., The Characteristics of Persistent Sexual Offenders: A Meta-Analysis of Recidivism Studies, Journal of Consulting and Clinical Psychology, Vol 73(6), 1154-1163 (Dec 2005)(13.7% sexual recidivism in meta-analysis of 95 studies with over 31,000 sexual offenders over five year follow-up period

[4]  Fact Sheet: What you need to know about Sex Offenders, Center for Sex Offender Management, December 2008 (estimating about 12-24% of all offenses are repeat offenders); A Better Path to Community Safety, California Sex Offender Management Board, available at (estimating about 95% of all offenses are first time offenders)

[5]  For links to various articles on this subject, go to < >

[6]  Prescott, JJ and Rockoff, J., Do Sex Offender Registration and Notification Laws Affect Criminal Behavior?, (2011); Tabachnick, J and Klein, A., A Reasoned Approach: Reshaping Sex Offender Policy to Prevent Child Sexual Abuse, (2011), pg 26.

[7]  A Reasoned Approach, supra, note 6.

[8]  Raised on the Registry: The Irreparable Harm of Placing Children on Sex Offender Registries in the U.S., Human Rights Watch, (2013);  Justice Policy Institute, Registering Harm: How Sex Offender Registries Fail Youth Communities (2008).

[9]  Irreparable Harm, supra note 8.

[10]  What you need to know about Sex Offenders, supra, note 4; A Better Path to Community Safety, supra, note 4.

[11] Not everyone who commits an offense against a pre-pubescent child can be considered a pedophile, and persons who commit an offense against post-pubescent children, i.e. teenagers, are likewise not pedophiles.

[12]  See e.g. Hanson, K.R.; Bussière, M.T., Predicting relapse: A meta-analysis of sexual offender recidivism studies, Journal of Consulting and Clinical Psychology, Vol. 66(2), Apr 348-362 (1998); Hanson, K.R.; Morton-Bourgon, K., The Characteristics of Persistent Sexual Offenders: A Meta-Analysis of Recidivism Studies, Journal of Consulting and Clinical Psychology, Vol. 73(6), 1154-1163 (Dec 2005).

[13]  Like registration, residency restrictions are not likely to reduce sexually reoffending, largely because they apply to all offenders, regardless of risk. See Residential Proximity and Sex Offense Recidivism in Minnesota (2007) < >