Blog detail

Let's break both iron laws

One of the supposed iron laws of a recession is that crime goes up. Well that's what most people said until now.

Crime has fallen over the past year in the UK in the teeth of this recession. And while a lot of people are rightly still worried about crime the new crime stats are important. Rather than just publishing the Government figures the BBC reporter Mark Easton makes interesting reading on the subject. (see below)

Another iron rule has been that support for extremist’s parties has also risen. Having nicked the crime trend I hope we can also break that second societal consequence of the recession. After all this recession wasn't caused by immigrants or any of the usual right wing scape goats but by bankers many of them very close to home in Scotland’s two biggest banks.

Let me know your view. I have put below the full info from Mark Easton's BBC blog.

Conservative estimates on violence

Mark Easton | 12:12 UK time, Monday, 25 January 2010

Are the Tories being honest with their claims on violent crime? Last week, David Cameron told me that one reason he could justify the phrase "broken society" was because of "significant" increases in violent crime, notably gun and knife crime in Britain.

When I challenged him to produce the evidence, his party press office sent the BBC a list of statistics.

It emerges that the only way the Conservative leader can back up his claims is to ignore the klaxon warning attached to the statistics following changes in the way police record violent incidents in England and Wales.

Tory Central Office e-mailed this claim to me:

The document cited, however, includes this massive caveat:

And yet, that is exactly what Mr Cameron appears to do.

The NCRS wasn't some sneaky ministerial trick to massage violent crime figures. Quite the opposite: it made violent crime look much, much worse.

The Home Office had been accused of under-estimating violence for years because the decision as to whether an incident was a violent crime had been taken by police. So, for instance, a drunk with a fat lip staggers into a police station and claims that he has been the victim of assault. Under the old system, the desk sergeant might have offered a weary shrug and said that the police had better things to worry about.

After 2002, though, officers were obliged to record all incidents as violent crimes if the alleged victim said that is what it was. The aim was to stop police fiddling the figures and to get a better picture of violence. The obvious result was to send the statistics shooting up.

In 2002, there were 230,704 recorded incidents of "violence against the person with injury". The following year, the number recorded shot up to 372,124. Three years later, it had risen still further to 543,605.

Now, you might argue that the later figures are more reflective of the level of violent crime, but what you cannot do is compare one with the other to identify a trend. To do so may only reveal the effect of new counting rules, rather than any real change in the level of violence.

It might be frustrating for politicians who want to compare today's picture of violence with the situation when Labour came to power, but the independent officials who oversee and quality-check the stats are clear: you should not do it.

Fortunately, there is a measure of violent crime that has not changed in almost 30 years - the British Crime Survey. The BCS regularly asks 46,000 adults in England and Wales about their experience of crime in the previous year. This graph shows what people have been saying about violence over the past three decades. The story is of falling and then stable violence for over a decade.

The BCS is not perfect - until very recently it didn't interview 10-15 year-olds and, obviously, it cannot talk to victims of homicide. But 650 murders a year out of more than two million crimes would barely change the graph at all. And it is inconceivable that offences of violence against very young children are so numerous and increasing at such a rate that they change the story the figures tell.

Mr Cameron specifically mentioned increases in gun and knife crime - issues which he knows play directly into voters' fears that our country is becoming less safe. What is his evidence that either is getting worse?

Figures on knife crime in England and Wales have only been collected for a couple of years and the data so far suggests the number of offences is falling or stable. When we asked Tory Central Office to justify the claims on knives, a spokesperson sent us this:

The figures are plucked from the latest homicide figures for England and Wales [629Kb PDF] which contain a table dedicated to "apparent method of killing". The category the Conservatives select is "sharp instrument" (which includes glasses and bottles as well as knives) but does show an increase from 201 in 1998/9 to 255 in 2008/9.

Is this a story of escalating knife crime? Every murder is a tragedy, but homicide is at its lowest level for a decade and the figure for stabbings has been higher in six of the last 10 years. As the Conservatives concede, such murders actually fell 6% year-on-year.

Gun crime? Well, using the methodology they employ with knives, the homicide data reveals the fewest people were shot dead last year in the period since Labour came to power - 39 victims. It was 46 in 1998/99 and 96 in 2001/2.

On this occasion, however, the Conservatives select a different category. Here is what the Central Office e-mail argues:

Is this evidence of "significant" increases in gun crime? If one stands back to look at the whole picture, this is what one sees:

Even if one only focuses on the dark black line and ignores the crimes committed with air weapons, it is hard to justify a claim of a "significant" increase in gun crime.

The argument that "gun crime resulting in injury" has increased 104% again requires us to ignore the entreaties of the statisticians with regard to the changes in the counting system. Looking at the table below, one can see the numerical gymnastics the Conservatives have had to perform to reach their conclusion.

If one avoids the problem of the new recording system by comparing 2008/9 with 2002/3, the story of gun crime injuries is one of decline.

We are going to get a lot of this during the election - the use of what Winston Churchill called "political statistics".

It is a subject I talked about on Radio 4's More or Less programme last week (you can listen below or subscribe to the podcast) and, on your behalf, I intend to keep a gimlet eye on any numerical nonsense as we head for polling day.