Crime rates don't matter. Fear of being a victim does

Crime is not like the weather, Jack Straw said somewhat shirtily on the Today programme yesterday, as the annual crime figures were published. No minister likes to admit that he is responsible for things he cannot control.

But then in the same breath he said that while crime figures are still on a downward trend, they may rise because of demographics: the number of boys entering the crime-prone age group is rising. In truth, crime is like the weather. Home secretaries are storm-tossed little boats bobbing about on the great waves of social change: they can trim their sails better or worse, but they can't do much about the winds and tides.

This is a lucky government and Jack Straw has had his fair share of it. This is the sixth year in which crime (both property and for the first time violence) has decreased. Yesterday home office ministers were at great pains not to offer reasons why, but it is unlikely to have very much to do with them.

Look at long-term trends and it is demographics, economics, employment, population upheaval and wars that count. Across Europe, you can track the rise and fall in crime against boom and bust in each country's economy. The more jobs, the less crime. Some suggest crime may fall as the poor acquire more of the moveable goods of the rich - videos, mobiles, even cars - because there is less they want to steal. (You can't steal a foreign holiday, suburban house or an au pair).

But if you ask Michael Howard or this week's Daily Telegraph leader writer why crime is falling, they say it is because prison works. See how crime falls when prisons fill! Howard inherited 42,000 prisoners and swelled their ranks by 50%. Numbers have risen under Labour to 65,674, with 10,000 more expected in the next six years, and still crime falls.

Home secretaries can influence prison numbers, but they bear no particular relation to crime rates. In the US crime has been falling sharply while prison numbers quadrupled in 20 years. QED, said Howard, and used the US experience as his model. What he didn't use was any of the research evidence: some US states imprison many more criminals than others, but the justice department reports no correlation between states with the steepest drop in crime and those with the highest imprisonment rates. Zero tolerance in New York produced exactly the same crime drop as Houston, which has no such policy. What does make a difference is catching criminals: criminals assess their risk of being caught, not the sentence they will get.

Jack Straw knows all this but still presses ahead with "three strikes and you're out" for burglars and automatic prison for probation defaulters, both of which will add to prison numbers. He talks tough on prison but only whispers to the experts about the government's many measures that really might reduce the causes of crime. He has done nothing to create a better-informed electorate. Research shows how abysmally little the public knows about sentencing policy. They are easily whipped up to shout for harsher sentences, but when asked what they think the sentencing tariff is already, they underestimate prison terms by years. Asked to suggest an appropriate sentence for a specified crime, the great majority suggest a lower sentence while assuming it to be tougher than the current tariff. Now why didn't Straw seize on that research to start a public debate?

Straw complains about the damage done to people by fear of crime, diminishing and restricting the lives of children and old people, but he does nothing himself to assuage it.

Indeed politicians are as much to blame as the lurid press for inflaming irrational fear. (Hague and Widdecombe both boldly lied and said crime was rising at last week's Tory conference, claiming criminals are roaming free on electronic tags, committing more crimes: virtually none have reoffended.) As a result people wildly over-estimate the danger of crime, especially of violence. Ask about children being murdered by a stranger and they will guess at 20 times the actual number of victims, (an average of seven a year). People think the old most at risk because pensioner muggings attract more publicity, but the young are 13 times more likely to be victims. A third of people in the British Crime Survey say they think they are "certain or very likely" to be burgled in the next year, when in fact out of 21m households, there are some 1m burglaries, many of them repeat burglaries in high crime areas while huge swathes of the country have very few.

The truth is that Jack Straw knows a great deal about how to reduce crime and is doing the right things. He looks at the research, he follows the evidence and he knows what works. The £400m for crime reduction in high-crime areas will reduce burglaries by at least 5% with "target-hardening" locks and protection for 2m high-risk households. Car crime may hit his target of a 30% reduction in four years. The Germans 20 years ago decreed all cars, new and old, must have steering-column locks and car theft plummeted. However, Straw caved in to the car-owning lobby and made it mandatory only for new cars, which may mean he misses his target.

But most impressive are the government's attacks on the real causes of crime - the new deal for the unemployed, better treatment of excluded children, new programmes for juvenile offenders, better (but not yet good enough) prison education and drug treatment and a plethora of other human rescue plans. There is genuine admiration and enthusiasm for all these throughout the criminal justice system. In the very long run, there is optimism that Sure Start will bear fruit. Much hope is invested in this still small programme for supporting children aged three and under in the poorest areas, who may otherwise be past helping by the time they reach school. It's admirable that the government takes such a long-term view, as it will be years before Sure Start's success emerges in anything as tangible as crime figures.

Yesterday home office ministers were refusing to predict future crime. This is probably wise, since a Wall Street crash or the economy of Brazil might make a greater impact than anything they can do. But given an economy as good as the treasury promises, the home office's excellent programmes really might hold down any natural demographic rise in crime and even keep on bringing the figures down.

However Jack Straw should ask himself this: does it matter if crime goes up or down if people live with quite unrealistic fears of the true risk? How are their fears appeased by jailing more and more criminals at £25,000 a year each, when no one knows what the sentences are anyway? Explaining what really works would make people feel better about crime and punishment. How the home secretary talks matters as much as what he does.