Knife attacks: damned lies and crime figures (Daily Mail)

I have a piece in today's Mail on the crime stats. Here's an extract:

Well that's OK then. Home Secretary Jacqui Smith said yesterday that she is 'extremely pleased' that the Government has exceeded its target for reducing crime. What a surprise! Home Secretaries always claim that crime is falling. The Left-liberal consensus, typified by the BBC, pushes the same line and dismisses any claims that Britain is suffering from an epidemic of crime.

For example, when people call phone-in programmes and point out that in their childhood they could play safely in the streets, or leave their front door open when they left the house, they are chastised for talking nonsense. The mantra is repeated: Look at the figures! Crime is down!

Yet the truth is that most people have an instinctive understanding that crime is a lot worse than it used to be. Although it is difficult to sort the reliable evidence from the massaged statistics, the evidence is clear. Things are worse than the official statistics suggest.

It was claimed that the figures published yesterday showed that there were five million recorded crimes in England and Wales - a fall of 9 per cent in the 12 months to March.

The phrase 'lies, damned lies and statistics' is never more apposite than when dealing with crime figures. Indeed, there were two sets of 'official' figures: first, the number of crimes recorded by the police and, second, crimes counted by the British Crime Survey. The police figures say there were five million crimes last year while the British Crime Survey figures say there were 10.1 million.

And if you think that's confusing, it gets worse: both sets of figures are wrong.

The figures for crime recorded by the police are very misleading because they only take into account, as is clear from their name, those crimes which are actually reported to the police.

On the other hand, the British Crime Survey is based on 47,000 people who are asked about their experience of crime. So, in theory, it's a better guide to the true level of crime than the police's figures. But it, too, is still deeply flawed.

For one thing, the British Crime Survey is hamstrung because its researchers only interview people who are prepared to discuss the problem - and they are hard to find in high-crime, inner-city areas.

Worse, the survey doesn't count more than five repeat violent crimes against a victim. So, if someone is repeatedly mugged, not only might the crimes never be recorded because the victim was never included in the survey, but even if he was, the number of crimes would be under-counted.

Thus one university criminologist has calculated that the total number of violent crimes committed against adults is likely to be 80 per cent higher than the figure recorded by the British Crime Survey.

Ridiculously, the BCS figures also fail to include any crimes involving a victim aged less than 16. Whereas other independent research shows that one in four children between 12 and 16 has been a victim of crime.

Nor does it include crimes related to illegal drug use, sexual offences, crimes against commercial victims (such as vandalism and van thefts), murder or shoplifting.

...The think-tank Civitas has calculated, using these and other statistics which don't feature in the British Crime Survey, that there are around 11million extra crimes every year. If these are added to the British Crime Survey's estimate of 12.6million for 2003, this means a total of almost 24million crimes. (And, of course, this much higher figure doesn't include sex or drug-related crimes.)

...Even on the official figures, it is also clear that crime is much worse than it used to be. For more than a century, recorded crime remained more or less the same - sometimes dipping (to 79,000 incidents in 1886) or rising to a peak (105,000 in 1908).

By 1931, the figure had started to rise, to 159,000. But it was only in the 1960s that there was a dramatic take-off: By 1971, the number had risen to 1,166,000.

...Just using the woefully inadequate British Crime Survey figures, the breakdown for specific crimes shows increase over the past 30 years. For example, in 1979 there was a one in 11 chance of being a victim of car crime; today, three quarters of drivers are victims. Equally, in 1979, there was a one in 32 chance of suffering a burglary; today it is one in 11.

In 1957, 340 per 100,000 of the population were convicted of a criminal offence; by 1991 the figure had risen to 1,400, and this at a time when detection rates fell. So, even though clear-up rates were falling, ever greater numbers of the population were convicted of crime.

This pattern is also borne out by recorded police figures, which dramatically underplay the scale of crime. For example, in 2001 there were 6,500 robberies in one London borough (Lambeth) whereas, in 1972, there were 8,900 robberies recorded in the whole of England and Wales.

During the month of December 2002, police recorded 282 robberies in Lambeth - a larger figure than for the entire number of robberies recorded for all of England and Wales in every year bar two between 1918 and 1939. In fact, the national annual figure for robbery did not exceed 400 until 1941.

Nevertheless, the BBC reported yesterday's figures with its usual slavish adherence to the fiction that crime is a mirage. Indeed, one report was headlined 'Facts not fears', as if the public's fear of crime is irrational.

Next time you hear a politician or someone from the liberal-Left try to dismiss fear of crime as a media panic, remember this: statistics can be used to support almost any argument. But the truth will always out.