Police fail to report 1.4m crimes

An estimated 1.4m crimes are going unrecorded by the police every year partly because officers bend the rules to exaggerate their success, government inspectors have discovered.

An estimated 1.4m crimes are going unrecorded by the police every year partly because officers bend the rules to exaggerate their success, government inspectors have discovered.

Police officers have been found grossly to misrepresent and massage crime statistics to improve their detection rates while downplaying the number of offences committed.

Her Majesty's Inspectorate of Constabulary found that 24 per cent of crimes reported to the police in 11 forces examined were not recorded as offences. Some forces required hard evidence that there had been an offence before recording it as such. One force was found to have recorded only about half the crimes reported to it.

The Home Office announced proposals yesterday for a radical overhaul of how the police collect, record and publish crime statistics. One of the expected changes will be the future publication of detailed neighbourhood crime statistics to allow citizens to find out the level of lawlessness in their local streets or villages.

The changes will certainly result in a huge rise in the published crime rate. Criminologists have always known that the official number of offences recorded by the police is an underestimation of the real rate, but the report details widespread rule bending.

The report, On the Record, published yesterday, uncovered evidence of hugely differing recording practices among the 43 police forces in England and Wales. Of the 11 forces examined in detail, inspectors found errors in between 15 per cent and 65 per cent of the crime records they scrutinised.

Offences were wrongly classified, many offences were incorrectly labelled "no crime" - not recorded after officers ruled that nothing had happened - and multiple crimes were often recorded as single offences. On average the proportion of reported crimes that were not recorded by the police was 24 per cent, equivalent to about 1.4m offences of the current official figure of 5.3m crimes recorded in England and Wales in the past year.

Among the example of incorrect or dubious recording found was the frequent failure to record cheque and credit-card fraud. In several forces, offences that had supposedly not been recorded previously were suddenly classified as crimes while the inspectors were examing the police logs.

Inspectors also found that officers give the impression of greater success - of less crime and more detection - by using much tougher criteria to record an offence than to classify one as "detected" or solved, or as "no crime".

A second HMIC report, Review of Crime Statistics, found that the Kent police "no crimed" 8.36 per cent of reported offences compared with the lowest rate of 0.41 per cent in Cleveland, where the force has championed a zero-tolerance approach to policing, and only false allegations were allowed to be classified as "no crime".

Instances of the "inappropriate" use of the classification included several reported assaults which were officially wiped off the records because officers could not prove the offence or obtain a confession. Two cases involving motorists driving off without paying for their petrol were "no crimed" because there was no registration number for the vehicles.

Her Majesty's Inspector of Constabulary, Keith Povey, said that he although he would not use the term "fiddling" to describe police counting practice, he admitted officers were "interpreting" the rules to their benefit.

The Home Office's statistics directorate wants to tighten up the police recording procedures and publish two sets of figures, one for crimes where there is evidence of illegality, and another for all allegations of an offence.

The Home Office is setting up a working group to consider the proposed changes and has invited both the Liberal Democrats, who have accepted, and the Tories to join the discussions.

The Home Office Minister Charles Clarke said: "Reliable statistics are essential to inform policy making, the police and the public about the true picture of crime. This review of crime statistics will enable the public to know what they should be worried about and what they shouldn't."