banner supportlinkscontactaboutarticlesmain


It may seem pretty obvious that the more police one has, the less crime there will be. Well, to some of us it is also pretty obvious that more use of prison equates to less crime, but there is a vociferous body of prison reformers who disagree which demonstrates that nothing in the law and order debate can be taken for granted. I daresay there are individuals who will argue that to reduce crime, money spent on the police could be better spent on social services or community initiatives.

Alan’s post (“America: Number of Police.”, Aug 9th 2006) presented the evidence from the USA in favour of increased policing levels as an effective way to combat crime. I couldn’t resist carrying out a similar analysis of my European data to see if Alan’s results would be confirmed by the data collected on this side of the Atlantic.

As outlined in my earlier post, “Prison Works – A Review of the Evidence.” posted on Aug 3rd 2006 I used the data presented by the WODC (the Dutch abbreviation for Wetenschappelijk Onderzoek- en Documentatiecentrum, in English: Research and Documentation Centre), an international criminal justice knowledge centre which publishes the European Sourcebook of Crime and Criminal Justice Statistics (link).

For 24 European countries, police numbers per 100,000 population in 1995 were compared with the crime rates (offences per 100,000 population) in 1999 for each country. The same statistical comparison employed in my earlier post was used.

The time lag between 1995 for police numbers and 1999 for crime rates was employed to overcome the problem highlighted by Alan that knee jerk responses by governments to increase police numbers at the same time crime rises, or to cut police numbers at the same time crime falls, tends to mask the underlying longer term relationship between crime and police numbers. An examination of the data for the 24 European countries only revealed 3 which had significantly increased policing numbers between 1995 and 1999, on average police numbers within any country had only increased by 3% over this period. In contrast, absolute policing levels between the countries varied by up to +/-100% so any distortions introduced by changing police numbers as a government response to changing crime within any given country were small.

A statistically significant negative correlation was found between the numbers of police and the rate of crime in any given country, i.e. the more police, the lower the crime.

The correlation coefficient between the two data sets was -0.53, indicating that 28% of the variation in crime levels between the different countries could be accounted for by variations in their policing levels.

This result for Europe in the late 1990’s (28%) is not far off the conclusion presented in Alan’s post that increasing police numbers might account for up to 20% of the decline in crime recorded in the USA over the 1990’s.

We are often told by those on the “left” of the law and order debate that crime data doesn’t support the solutions presented by those of us on the “right”. That more use of prison, harsher penalties, more policing etc., are not supported by the evidence. I have to say that I am extremely grateful to this forum for giving me the incentive to actually look at the evidence for myself. My conclusions from this study of police numbers, and my earlier study of prison use, are that the data speaks volumes for those prepared to listen, and that the message is clear.

More policing, and more prison works.

Post a comment