Once a sex offender, always a sex offender. Everyone knows that, right? Sex offenders are incurable monsters. They will commit new outrages as soon as they think they can get away with it, which is why the federal government recently passed legislation creating a national sex offender registry. There isn't a robber registry or a thief registry because robbers and thieves can go straight. But not sex offenders. Forcing them to register their home addresses at least allows the police to pounce at the first sign of trouble. And there will be trouble because sex offenders can't change. Everyone knows that. Everyone, it seems, except sex offenders. Late last year, the United States Department of Justice released a massive study that looked at all male sex offenders released in 1994 from prisons in 15 states, a total of nearly 10,000 men. Other research has shown that the period when ex-cons are most likely to commit new crimes is the first few years after release. Research has also shown that whether an ex-con is arrested is a pretty good indicator of whether he has committed new crimes or not. So the justice department researchers checked to see how many of the sex offenders had been arrested again within three years. And they compared that result to other offenders released from the same prisons in the same year. If it were true that sex offenders are far more likely to commit new crimes than other criminals, the sex offenders would have been re-arrested at a much higher rate. But they weren't. Far from it. The researchers found that 43 per cent of sex offenders had been re-arrested for any offence, compared to 68 per cent of other criminals. So sex offenders were actually far less likely to commit a new crime. And that's just the data for arrests involving any new crime, even something as trivial as pot possession. What people are really worried about is released sex offenders committing new sex crimes. Once a sex offender, always a sex offender, right? Not even close. The study found just 5.3 per cent of sex offenders were arrested again for a sex crime. In other words, 95 per cent of sex offenders were not re-arrested for a sex crime in the three years after their release, and 57 per cent kept out of trouble entirely. Of course, this doesn't in any way deny that some sex offenders are likely to commit new crimes after release. Nor does it deny that there are some sex offenders who are simply too dangerous to ever release. But it does show that not all sex offenders are like that. In fact, it shows that most sex offenders are not like that. That conclusion may be starkly at odds with conventional wisdom, but the justice department is no aberration. Five other American studies that compared recidivism of sex offenders with other criminals all came to the same conclusion. And a wealth of research shows that few released sex offenders go on to commit new sex crimes: One survey of 61 studies found just 13 per cent of sex offenders committed another sex offence within five years of release. So the basic idea behind the sex offender registry -- that sex offenders are more dangerous than others so they have to be monitored more closely -- is simply wrong. There's another flaw in the registry concept. Most people think sex crimes are committed by strangers lurking in bushes, but according to StatsCan, 80 per cent of adult victims and 91 per cent of children knew the person who victimized them. So a registry of addresses will be redundant in the vast majority of cases. Combined, this evidence indicates the new national sex offender registry will do little to keep people safe. It will, however, cost a great deal of money -- money that could have been spent on effective crime-prevention programs. Appalling as that waste is, more appalling is the fact that the government knows it is a waste. Four years ago, a federal-provincial working group examined the research and advised the government that a national sex offender registry "would not significantly improve" public safety or the protection of children. Top experts told the solicitor general the same. But the sex offender registry was, and is, wildly popular among people who know little about criminology -- which is most people -- and the Conservatives bashed the Liberals for not setting one up. For a while, the Liberals resisted the pressure, but eventually they caved. Politics trumped research -- which is the story behind far too many justice policies.