DO WE NEED a version of Megan’s Law? This is the American law that bears the name of seven-year-old Megan Kanka who was raped and strangled by a convicted paedophile in New Jersey in 1994. Unbeknown to Megan’s parents, he had moved into a house across the street and lured the inquisitive young girl inside by offering to show her a puppy.

Megan’s Law allows citizens to receive or actively seek information from sex-offender registers about those in their area deemed to be at high risk of reoffending. Its implementation was hotly debated here after the murder of Sarah Payne in 2000, and is back into the media spotlight after the sentencing last week of the paedophile Craig Sweeney. The main aim of this “public notification” law is to promote community safety. But has it worked there, and would it work here?

There is in fact no single Megan’s Law. All 50 states in the US have now enacted their version of it and there is a wide variety in how the law is applied, how the risk of reoffending is assessed or how neighbours are notified. Given this, it is hard to measure its effectiveness but studies that have tried to do so suggest that Megan’s Law has not reduced assaults against children or the rate of reoffending.

Megan’s Law addresses the horror of random abuse or attack by a stranger. In fact, children are most likely to be assaulted by someone they know: a relative or family friend. In such cases, making the attacker’s details public is likely to reveal the identity of the victim. The people who really need to know the identity of the abuser, his family, will already be aware. Further publicising the abuser’s identity risks causing even greater damage to his family. His child/victim’s school life and that of his other children is hardly likely to be made easier by publication.

Such a law also presupposes that paedophiles will reoffend within the area in which they live. Trains, planes and automobiles give the lie to that. Public notification can only hope to increase safety within a limited area. Organised hostility to sex offenders in a particular vicinity could tempt determined offenders to move to less organised communities, where they can re-offend unnoticed.

And what effect will public notification have on a local community? Fears of widespread vigilantism have proved unfounded in the US, but there are enough isolated cases of harassment and violence to cause concern. The disturbances in Paulsgrove in Portsmouth that followed Sarah Payne’s murder showed us how volatile things can become. There the home of a paediatrician was attacked instead of a paedophile.

In Britain 97 per cent of offenders comply with the requirements of the sex offenders register. That means that their whereabouts are known and they can be monitored by the police and other agencies. By contrast it is thought that in America Megan’s Law has stopped some offenders from registering, with only 80 per cent of paedophiles complying. Although offenders are required to give their address to the police, many bogus ones have been supplied. There is, after all, a great incentive to avoid registration. Sex offenders in Oregon can be forced to display a sign in their windows. It is perhaps no coincidence that in that state a sex offender’s house was burnt down, and another’s dog was beheaded and the head set on his doorstep.

And what happens to those that do evade the register? They can move around unmonitored and untreated.

There are two other assumptions that bear heavily on the debate over Megan’s Law. First, that rates of re-offending among sex offenders generally, and paedophiles in particular, are high. Secondly, it is commonly thought that treatment won’t work. In fact the picture that emerges from a raft of international studies puts re- offending rates far lower than is widely believed. One large US study of 7,753 sex offenders found reoffending rates of 10.9 per cent among treated offenders and 18.5 per cent among the untreated (the recidivism rate for other offences run at between 50 and 60 per cent). Other studies put the percentages higher, but not dramatically so. There are some important caveats here. Studying recidivism among sex offenders is an inexact science. Much depends on how “sex offenders” are defined, over what period they are monitored and whether they are offered confidentiality by researchers. It is also the case that reoffending rates among serious paedophiles are higher. One Canadian study recorded rates in excess of 50 per cent.

While those kinds of figures give serious cause for concern, there is some cautious optimism that monitoring and treatment can be effective. Here the reoffending rates for those who are being supervised or resident in probation hostels are generally low. That is why organisations such as the NSPCC recommend an effective prevention strategy such as that used by Stop It Now, a body that provides challenging treatment, help and advice to sex abusers, those concerned by their own thoughts and behaviour, and to people worried about inappropriate sexualised behaviour in others.

There is no more emotive subject on our political agenda than the creation of a Megan’s Law. Revulsion at the grotesque crimes of a Craig Sweeney will always raise the temperature at the Home Office. And what comes across from the US experience is that people like Megan’s Law. It makes them feel safe even if it doesn’t actually make them safer.

Clive Coleman is a barrister and the presenter of Law in Action on Radio 4

Times Recommends

• Fantasy world fuelled by sex, drink and drugs
• In short, the story is growing on us again
• A literary feast that I can’t wolf down