Kleck, Gary and Barnes, J.C. (2014). Do more police lead to more crime deterrence? Crime & Delinquency, 60(5), 716-738.
With the beginning of the Economics of Policing initiative lead by Public Safety Canada, and the various efforts that have been inaugurated around this project, many politicians, police leaders and policy-makers have been focussed on the right number of police officers required in any given jurisdictions for the delivery of policing services.
We have seen a number of efforts aimed at making these determinations in a, more or less, scientific (or evidenced-based) manner. Research has been done on calls for service and the average time it takes to 'clear' those calls. The calculations may take into account a number of relevant factors including the seriousness of the crime occurrence, time of day, day in the week, and other elements that may provide insights into this area of human resource deployment. Complex mathematical models have been developed to guide this process, including an approach known as data envelopment analysis (DEA) which is currently being utilized within the Toronto Police Service to assist with personnel deployment.
However, beyond the complexities and adequacy of these approaches, common practice tends to equate the number of police officers with crime deterrence. It is really a simple formula: the more police you are able to put on patrol, the lower your crime rate will be in any given jurisdiction. More police equates to less crime. It's common sense that is readily accepted by the public, it is used by police executives to defend increased budgets and it is turned to by politicians (especially at election time) to convince the public (and the police) that they are tough on crime and willing to pay the price for public safety.
Unfortunately, this is one of those instances where common sense, as well as, the available research moves in an entirely different direction. Remarkably, we have had research on the actual impact of police activities upon crime rates for nearly forty years. In 1974 ground-breaking field research was conducted by George Kelling and his colleagues in Kansas City, Missouri. Well-known as the Kansas City Preventive Patrol Experiment this project examined 15 precincts (chosen at random). Several precincts would receive no routine patrols while others would receive four times the normal frequency of routine patrols. Across all measures, it was determined that there was no significant impact on crime rates as a result of any of these approaches. This was measured by police-based statistics and victim survey data. Of course, these results were remarkable and flew in the face of expectation.
George Kelling
Then, in 1981, research was completed through the Newark Foot Patrol Experiment which served to confirm and consolidate the findings in Kansas City. In Newark, foot patrols were stopped entirely in four areas, new patrols were started in four areas, and existing patrols were maintained in another four areas within the city. Again, there was found to be no difference in crime rates in the experimental or the control areas. Once, again, police activity (regardless of the number of officers) appeared to have no significant impact on crime rates.
It should be noted that there was some skepticism among police researchers. However, these remain fairly compelling experimental results that policy-makers need to consider. The police community needs to remind themselves about these evidence-based findings and the academic community in Canada should seriously consider the research value of replicating this kind of field study in the 21st century.
Kleck and Barnes speak directly to the common sense view that more police should equate to reduced crime rates:
Despite the intuitive theoretical linkage between police activity, individual-level perceptions of punishment risks, and crime rates, no studies have assessed these associations empircally. Purely macro-level research cannot directly assess the deterrent effects of police activity because it cannot directly measure perceptions of risk among potential offenders. (p. 720)
What this current article summarizes is the work completed by Kleck and Barnes in 54 US counties. A total of 1500 adult respondents were randomly selected and interviews were completed with participants in April and May of 1998. The dependent variable related to perceptions of arrest risk and the key independent variables included: police strength, jail ratio, and arrest ratio (i.e., actual arrest rates).Drawing upon their research results, the authors make the following observations:
People who live in places with more police perceive slightly lower risks of being arrested. The negative correlations, however, are quite weak; none of the four correlations between the police strength variable and the perceptions of arrest likelihood were more than .10 in magnitude. (pp. 725-26)
And further:
...police strength appears to have no more effect on perceived risk of arrest among offenders as approximated by self-reported arrestees, than among non-offenders. (p. 730)
These are useful findings that serve to support some of the conclusions generated from the work completed in Kansas City and Newark years ago. And, for those who proceed to argue that police strength would have an impact on crime rates through an incapacitation effect, Kleck and Barnes offer the following statement:
...it is unlikely that increases or decreases in police strength -- even a doubling or halving of manpower [sic] -- could have any measurable effect on the number of criminals incarcerated. Because nearly all prison systems are full nearly all the time, the only factor that substantially affects the number of criminals "taken off the streets," in the sense of being incarcerated, is prison capacity. Police ability to arrest criminals guilty of prison-worthy offenses is never a limiting factor on the number imprisoned, so it is implausible that variations in police manpower [sic] affect the collective incapacitative effect of the criminal justice system. (pp. 731-32)
So, neither increases nor decreases in the number of police officers deployed to operational duties will likely have any dramatic impact on crime rates or the number of individuals arrested for crimes. Again, it would be valuable to have Canadian research to validate the findings of American police academics. However, we can certainly extrapolate from this study to suggest that any results would likely be similar in a Canadian context.
Of major importance to Canadian policy-makers and police leaders is what the authors assert near the conclusion of their article:
Increasing police manpower [sic] levels may have various benefits other than crime control, such as traffic control or the delivery of various public services, or it may affect crime rates in ways that do not involve either increased general deterrent effects or increased incapacitating effects. At this point however, the evidence indicates that police manpower [sic] levels do not affect crime rates by affecting perceptions of risk of actual arrest and punishment.
In light of these findings, policy makers may want to reconsider whether increases in police manpower [sic] bring sufficient crime reduction benefits to justify their costs. Conversely, in time of fiscal crisis, it is worth considering the possibility that cuts in police strength may be implemented without causing crime increases. Even in time of fiscal plenty, it may be worth thinking about alternative investments that are more likely to reduce crime. (pp. 735-36)
These are quite clear statements that should be taken to heart by everyone interested in the adequacy and effectiveness of modern Canadian policing. We can begin to see that on the basis of this kind of research, combined with the historical record of earlier systematic study, that police may provide valuable and necessary public safety services. However, there is no axiomatic relationship between police strength (in numbers) and crime rates. Here again, we could benefit from home-grown research that tests the findings of studies like the one completed by Kleck and Barnes. In the meantime, it is essential that these kinds of insights become an important part of the dialogue regarding the future of policing in Canada.
Police executives are well-schooled in the art of pressing for increased resources; human, financial, technological, and material. This is part of what being an police leader is all about. Politicians are easily convinced of the simple equation: more cops equals safer streets. We often see that during an election that political parties will vie with one another to increase the complement of police officers to combat crime and increase public safety and security. In Nova Scotia a popular "Boots on the Street" campaign was introduced by the Conservatives and sustained by the NDP when they took hold of the reins of power. In 1994, President Clinton made a commitment to hiring 100,000 additional police officers to bring greater safety to American states. This was consolidated in the Violent Crime Control and Law Enforcement Act. A similar initiative appeared on the scene in Canada and was seen as a path to improvement for police services. No research was offered to support or substantiate these new funding initiatives. They were seen to be self-evident and by all means politically expedient.
No comments:
Post a Comment