The claim
“Crime is now coming down… And this is the first government ever, since the war, which has seen a sustained fall in crime, year-on-year.”
Jack Straw, Justice Secretary, 28 April 2010
The background
Addressing an audience at the Honeywell Community Centre in Oldham, Justice Secretary Jack Straw made a bold claim on crime. But in the murky world of crime statistics, there are only two accepted sets of data, and they yield rather different results – so which one was Straw relying upon?
The claim from Straw is drawn from the British Crime Survey (BCS). First conducted in 1981 this poll is commissioned by the Home Office. It is drawn from a sample of the population of the UK asking about their experiences of victimisation by certain crimes over the course of the previous 12 months.
It’s widely regarded as the most robust dataset, but there are questions over sampling methods. The figures are extrapolated up, so they can only give a statistical projection based on the experience of the 40 – 50,000 people surveyed.
In contrast the Police Recorded Crime Statistics rely on a totally different set of data – those of crimes actually reported to the police. While these statistics show trends in well-reported crimes, indicate police workload, and can be used for local crime-pattern analysis, it’s important to remember that not all offences recorded by or reported to the police will end up in the Police Recorded Crime data.
The analysis
Even using the British Crime Survey as a proxy measure of ‘total crime’ the fall has not been “sustained, year-on-year” as claimed by the Justice Secretary.
In fact crime has risen in at least three separate yearly periods during Labour’s 13 year period in office – 2005/6, 2006/7 and 2008/9. Although the BCS does show a significant overall fall in crime levels under Labour – from 16.7 million incidents in 1997 to 10.7 million in 2008/9 with a peak of 19.3 million in 1995.
But Straw makes the claim by harking back to the end of WWII. In the period between 1945 and 1981 the only statistics available came in the form of the Recorded Crime figures, so his claims over rises and falls are not like for like comparisons by any stretch of the imagination.
Recorded Crime figures show a steady increase in violence after the war, to a peak in the early 1990s, after which rates started to drop. In fact the figures show a downward trend from 1991 until 1997 – but have been climbing steadily since 1997.
The two data sets also differ markedly in how they measure when the fall in crime levels began. According to the BCS the decline began in 1995. In contrast the recorded figures show it began in 1992. So, while it’s true this government has presided over a fall, the decline began under the previous government led by John Major’s Tories.
Neither if these methods can provide a comprehensive view of crime in the country. Used together they probably give a very good indication of what is happening. The problem is each party chooses the set of data that best reflect’s their government’s performances, which can only be misleading to the public.
As stressed by the Centre for Crime and Justice Studies: “Neither the BCS or police recorded crime produce a comprehensive picture of all criminal activity… Both measures systematically ignore or underestimate certain categories of crime”
A Labour Spokesperson said: “This Labour Government is the first since the war to preside over a fall in crime, according to the most widely accepted measure of overall crime that existed at the time in each period.
“The most widely accepted measure that exists now and that has existed over the course of this government is the British Crime Survey.”
The verdict
Neither of these sets of statistics can provide a comprehensive view of crime in the country since the Second World War. Indeed, the most respected set of figures – and the ones relied upon by the government – go back less than three decades and show some year-on-year increases during the current Labour government.
Had Mr. Straw not used the phrase “year-on-year” we would have chastised him for not comparing like with like but, as he did, it’s quite easy for FactCheck to award Jack Straw a fiction rating for this statement.