Nigel Hawkes: How crime stats cast police as Hercule Poirots

Behind The Numbers
Click to follow
The Independent Online

Anyone who has been burgled, had their car stolen or been the victim of criminal damage knows from bitter experience that the criminal is very unlikely to be caught. The detection rate for these crimes runs at a piffling 11-15 per cent.

Yet every year police forces across the country publish statistics for crime as a whole that claim a much higher detection rate. It ranges from around 25 per cent to more than 40 per cent. When these documents drop through letterboxes, belief in crime statistics sinks a little lower. Are they lying to us?

They are not. The collection of crime statistics is tightly governed by the Home Office and there is no reason to disbelieve them. But that is not to say that they present a complete or wholly accurate picture.

Take Northumbria Police. In 2008-09 they recorded 105,458 crimes, and a 39.3 per cent detection rate. For burglary, vehicle crime and criminal damage, which represent almost half the crimes recorded in the area, the detection rate was 19.2 per cent – better than the national average but a long way below the overall claim.

So how does 19.2 per cent become inflated to 39.3 per cent? One reason is that some crimes are virtually invisible unless they are detected, so event and detection are almost identical. Take drug crime. Because there is no victim, drug crime is not reported at all unless it is detected. Northumbria Police claims a detection rate of 97.5 per cent for drug crime.

The same applies to shoplifting. People who steal from shops do not usually enter the crime statistics unless they are caught in the act. Northumbria claims a 78.6 per cent detection rate for shoplifting. And for harassment, Northumbria records a detection rate of 82.4 per cent.

These figures are largely meaningless, because the huge majority of crimes in these categories are never reported or detected. To claim such high detection rates is a sleight of hand, perfectly acceptable under the rules, but creating a misleading impression. If drug offences, shoplifting, and harassment are subtracted from Northumbria's figures, the detection rate falls from 39.3 to 29.6 per cent.

In fact, Northumbria recorded more than 143,000 incidents of antisocial behaviour in 2008-09, a larger number than all the crimes it reported. Nationally, Denis O'Connor, Chief Inspector of Constabulary, says the police received 3.6 million calls about antisocial behaviour, against 4.6 million about crime. "Members of the public on the receiving end of antisocial behaviour find it hard to distinguish from crime" he said.

Very few of these incidents even make it on to the crime statistics – 4,005 cases of harassment were the only examples recorded in Northumbria. But it's all above board. Crimes that everybody knows are commonplace are only recorded if detected, creating the illusion that the police outperform Hercule Poirot in tracking down drug-takers and shoplifters. If only they could.

Nigel Hawkes is director of Straight Statistics (