MENU
Your location:

Wreaking revenge or just bad maths?

    “Cuts to officer numbers are Tory revenge for their ‘unfinished’ police business from their last stint in office,” claimed the chairman of the Police Federation of England and Wales, Paul McKeevor, this morning on the Today programme. He might have been better to focus on the Coalition’s maths rather than the Conservative motive.

    Attributing motive is never helpful (at least that’s the maxim here at the Centre for Policy Studies, as anyone who has had Tim Knox’s guiding editorial hand on their policy papers knows).  It deflects from the numbers – the analysis of which are essential to making a case.

    The Police Federation could find some very useful ones in a recent and very neat paper from our neighbouring think tank, Civitas, by Siddhartha Bandyopadhyay (a Lecturer in Economics at Birmingham University). This provides the ammunition Paul McKeevor needs to make Nick Herbert and the Home Secretary rethink the costs and consequences of their cuts.

    Cutting to the chase, through the current policy debates on social and economic cause of crime, custody versus community re-offending stats, the prison works debate and elected police commissioners furore, the author asks: what do we actually know about what deters crime? 

    His detailed ‘panel dataset’, across the 43 police forces from 1992 – 2008, allows him inter alia to answer the simple question:  does a higher crime detection rate in a neighbourhood lower the crime rate? The answer was a resounding yes. This, and not unemployment, or age demographics, was the key variable.

    The Police Federation might ask the Home Sec and her Minister if they are aware of this. For the strong and negative relationship between detection rates and crime is a certainty.  Just a few weeks ago, at a ‘Making Justice Work’ hearing, Sir Iain Blair remarked how little we could say about crime when our current detection rate is only 1 in 4/5 crimes. Here, in the Civitas analysis, is the evidence of the impact of a 1% rise in the detection rate.  

    It leads to a 0.11 per cent decrease in burglary, a 0.20 per cent decrease in theft and handling and a 0.14 per cent decrease in fraud and forgery. Even for violent crime, a 1% increase in the detection rates had a significant deterrent effect, leading to a 0.21 per cent decrease in violence against the person, a 0.34 per cent decrease in robbery and a 0.12 per cent decrease in sexual offences.

    At first glance these percentage changes sound insignificant. But translated into numbers this represents a significant number of crimes deterred for fear of detection. For burglaries for example, nationwide, it represented a total of 1000-odd fewer burglaries than in an average year. And so it went on for each crime category.  The maths of what would happen with further improvements in detection rates is not difficult.

    These findings may surprise those who maintain that policing is an ineffective tool against crime, or those who believe police would be better transformed into social workers, or offender ‘managers’. They should though, as the author says, be reassuring to those who believe that criminals respond to incentives. They should certainly give the Home Office and the MoJ pause for thought – not least about the reverse scenario of potentially considerable drops in detection rates which 20% cuts in police numbers could lead to.  The question takes on a new slant. Can the Coalition afford to cut? The inevitable increase in crime that this drop would lead to suggests not. This is what I would focus on if I were at the Police Federation.

    Kathy Gyngell has a first class honours degree in social anthropology from Cambridge and an Oxford M.Phil. in sociology. She has worked for the former ITV companies, LWT and TV-am as a producer and senior programme executive. A full time mother after the birth of her second son, she founded the voluntary organization Full Time Mothers.

    Be the first to make a comment

    Centre for Policy Studies will not publish your email address or share it with anyone.

    Please note, for security reasons we read all comments before publishing.