WIREDGOV

Detection key to deterring criminals

Government claims that police cuts will be made without endangering the public are dealt another blow this week. A new Civitas report finds that sudden police cuts could potentially trigger a vicious cycle of crime and disorder. In An Analysis of Crime and Crime Policy, Birmingham University economist Siddhartha Bandyopadhyay finds "a strong and negative relationship between police detection rates and crime". (p. 4)

This means that the more crime police detect, the more crime is prevented. The Government's announced 20% cut to police resources, which will be achieved by a predicted loss of 12,000 police officers, puts this check on criminality at risk. Dr. Bandyopadhyay warns that potential offenders might notice the thinning out of police officers and become more confident that they can avoid sanctions:
"The perception that police are stretched may encourage criminals to commit more crime which would indeed stretch the police".
(p. 8)

Policing works

The report is based on new research by Dr. Bandyopadhyay, and by his colleagues Lu Han and Samrat Bhattacharya. They measure the impact of several factors determining crime rates in 43 Police Force Areas in England and Wales between 1992 and 2008.

The most robust and powerful factors determining criminal behaviour were found to be police detection rates. Criminals were more affected by the threat of detection than any other factor. Even relatively small improvements in the detection rate were associated with visible reductions in criminal activity across all crime types, including property and violent crimes.

A modest one percentage point increase nationwide in detection rates for each type of crime in a typical year leads to an estimated average fall:

• in recorded burglary of 6,000 offences
  
• in recorded theft and handling of 22,500 offences
  
• in recorded fraud and forgery of 940 offences
  
• in recorded violence against the person of 2,300 offences
  
• in recorded robbery of 1,400 offences
  
• in recorded sexual offences of 100
  (p. 6)

This is in addition to the straightforward benefit of detecting more crime and, therefore, bringing more offenders to justice.

Catch up, Ken!

Ken Clarke, Secretary of State for Justice, argues that evidence that criminal justice interventions tackle crime is equivocal, and that economic prosperity has just as much chance of reducing crime. In last year's Mansion House speech, he claimed:
"No-one can prove cause and effect. The crime rate fell but was this the consequence of the policies of my successors as home secretary or, dare I gently hint, mine as chancellor of the exchequer at the beginning of a period of growth and strong employment? We will never know."

Ken Clarke is out of step with the evidence. Although measures of causality always leave some room for uncertainty, the Birmingham University researchers use statistical techniques to isolate highly probable causal effects:
"Our panel data methods allow us to separate… correlation from causation and once we apply appropriate techniques we get a strong significant effect of law enforcement variables." (p. 7)

The study also included a number of key socio-economic variables: unemployment, Gini coefficient (a measure of income inequality) and real income. In contrast to law enforcement variables, these were not consistently associated with changes in crime rates, suggesting that their effects on crime are limited:
"Economic variables which are thought to be important… are not significant across all crime types". (p. 4)

This means that, whilst socio-economic changes may well influence criminality, the latest evidence points much more strongly towards policing interventions. Clarke is misreading the stats when he implies that recent crime reductions can just as easily be attributed to economic growth as robust policing.


Nick Herbert shouldn't bet other people's houses on efficiency savings
Nick Herbert, Policing Minister, defending swingeing cuts to the police, claims, "There is no simple link between police numbers and crime." Citing the Audit Commission and Her Majesty's Inspectorate of Constabulary, he has suggested:

"… it is possible for forces to make savings of more than £1 billion a year-12% of the annual budget-through things such as improving productivity, cutting costs, sharing services and addressing savings in the back and middle offices of police forces".

But Dr. Bandyopadhyay warns that there is no simple measure of efficiency either.

He warns:
"The suggestion is that if the worst performing forces could reach the standard of the best performing ones, one could easily cut 12 per cent of the budget without losing out on services. However, questions remain on how one measures efficiency: how does one develop a sensible measure of efficiency taking into account that different police forces are not comparable in that they face very different populations with different crime propensities?" (p. 8)

Without adjusting for the different challenges that police authorities face, it is difficult to establish where genuine efficiency gains can be made. The Government is taking a great risk with public safety by assuming that police cuts ultimately can be made painlessly alongside dramatic productivity gains.

For more information contact:

Siddhartha Bandyopadhyay by calling 020 77996677 and s.bandyopadhyay@bham.ac.uk

Nick Cowen (Civitas crime researcher) on 020 7799 6677 and info@civitas.org.uk

i. Dr. Siddhartha Bandyopadhyay is Lecturer in Economics at the University of Birmingham and an internationally recognised scholar in political economy.

ii. The report, An Analysis of Crime and Crime Policy, is available here. It includes research by Bandyopadhyay, Lu Han and Samrat Bhattacharya. Their key discussion paper, Understanding the Determinants of Property and Violent Crime in England and Wales: A Panel Data Analysis, was released this month. It is available here.

iii. Civitas is an independent social policy think tank. It has no links to any political party and its research programme receives no state funding.