Abstract:
The abstract outlines the strategy to be used in the analysis and evaluation of the effects of police numbers on crime rates amid problems arising from causality since many issues that affect crime and police numbers such as social changes or economic cycles exist. The report will focus on critically analysing whether more crime causes more policing or whether more policing leads to less crime. However, on a different view, high police presence may also tend to cause more crime therefore the need to examine temporal sequences arises.
Introduction:
Classical criminology studies assume that criminals are rational beings who weigh the costs and benefits of their actions (Becker 2008). It is this research that raised interest in the economics of criminal behaviours. A key prediction of Becker’s theory was an increase in police presence would lead to a corresponding decrease in crime rates. However, the theory fails greatly since it lacks empirical evidence. Other theorists suggest that increased policing does not change or negatively impacts on the crime rates.
Background:
Becker’s theory was quantified by Corman and Mocan (2000) who exploited 30 years of high frequency data on New York City to show that increased police presence causes a reduction in one out of five categories of crimes especially burglary. Monthly data was used since recruiting and training activities cause delays in police activity and responses to increased crime reduce the simultaneity bias evident in low frequency data. To address the simultaneity bias, the identification strategies heavily rely on the assumptions that police authorities lack the ability to forecast crime fighting needs.
After the analysis, they observed a 10% increase in growth rate of arrests led to a consequent 9.4% decrease in the long term growth rate of robberies, burglary growth rates decreased by 2.89% while motor vehicle thefts reduced by 2.72%. They also found strong evidence to support the deterrent impacts of police arrests on various categories of serious felonies leading to reduction in crime rate. However, even then, the evidence remains unsatisfactory thus the need for further research.
Methodology:
The report will look at how crime rate is determined and the relationships between police numbers and/or arrests rates as well as the crime rates by discussing some criminology studies conducted.
Simple determination of crime rates:
The determination of crime rates varies from country to another mainly depending on the amount of political stability, geo-location, social behaviours etc. However, a rate is simply a comparison of a value to another and thus in the determination of crime rates, several basic measures are considered: Crime incidence (i), the number of crimes per unit at risk; crime prevalence rate (v) which is the number of targets or victims per unit at risk (a unit may be vehicle in the case of motor vehicle theft and when a unit refers to persons, it is per capita measure). Incidence and prevalence can be defined as shown below:
Crime concentration C refers to the number of victims per victimised target and is also the ratio of crime incidence (i) to crime prevalence (v) as shown below:
(Fienberg and Reiss, 1980)
Crimes in high crime areas are mainly composed of cases of repeat victimisation and thus both prevalence and concentration contribute to the crime incidences in an area, while concentration differences contribute disproportionately to high crime levels in crime ridden areas sampled (Trickett 2009). However, various other methods of crime rate determination exist but with varying levels of complexity; the simple method shown above is universal and all other complex methods derive from the three basic quantities i.e. incidence, prevalence and concentration.
Discussion:
As stated earlier in the report, none of the studies bear rigid proof that higher numbers of police leads to less crime since most of them suffer significant conceptual and methodological flaws. A good example is the study using lagged effects to estimate the effect of police numbers on crime rates e.g. Corman and Mocan vary the length of lag used to openly maximise the significance of the association to their study without giving much consideration as to why the time spans involved in the lag should vary with the crime type.
The Klick and Tabarrok study shows the reliance of natural or quasi-experiments on spontaneous and rather unsustainable deployment patterns resulting from emergency events such as terrorist attacks which are not frequent. Such events may initiate circumstances for policing that deviate from day to day activities and police responses to such activities may take a symbolic meaning and may require a unique approach to handle them. The general claims made by the paper about the potential implications of police numbers on crime are undermined by the special context of the events which they describe which are indeed emergencies requiring special attention. While natural experiments allow the drawing of firmer conclusions as opposed to observational studies in conduction of research, most of these experiments indicate steep rises in police presence but do not consider the converse as shown below.
While observational studies show better evidence in linking police numbers to crime rates as compared to quasi-experiments, they still suffer from common issues involving crime recording and activities of officers. Most of them use recorded crime rates as opposed to real rates. The absence of firm evidence to back up claims that more police lead to less crime is an issue to be taken care of. However, a case such as the “Aggressive Driving Initiative” in the Harford county of Maryland in September 2009 was a successful operation not only in the enforcement view but also the educational perspective as well. Similar operations were conducted on roads with frequent accident reports or citizen complaints in the county and in a week were able to offer 35 citations and 102 citations with violations ranging from speeding, seatbelt violations etc. This led to consecutive decline in the number of accidents reported and citizen complaints filed.
However, observational studies suffer from problems of excessive trust in instrumental variables where authors believe that they have solved problems arising from possibilities that police presence affects the crime rate and vice versa as shown in the case of Maryland police above, this may not actually be the case. Another problem is that of papers relying simplistic and rational models of crimes. Their stipulations are that crimes committed are influenced significantly by considerations of risk sanctions weighed against potential rewards where the number of police represents one element of calculation of risks. These causal claims that higher police lead to less crime and based on observational data are weakened by lack of mechanisms to explain the associations uncovered.
Conclusion:
Despite the consistency of research in trying to provide a causal link to higher police numbers and lower crime rates, more work need be done before drawing such strong conclusions. The reverse causality issue makes it difficult to empirically test the effect of additional presence on crime rates. While some may argue that massive police re-deployment after the July 2005 London bombings was a natural experiment to prove the theory that more police reduce crime, it should be considered that large scale alterations in police deployment due to shock attacks such as terrorism are just unique scenarios where such response is ultimately called for. However, it is fair to say that relatively strong evidence exists to back the potential effect of increased police presence on crime.
Works Cited:
Corman, Hope, and Nancy. Mocan. A Time-Series Analysis of Crime, Deterrence, and Drug Abuse in New York City. American Economic Review. 90.3 (2000): 584-604. Print.
Klick, Jonathan, and Alexander Tabarrok. "Using Terror Alert Levels to Estimate the Effect of Police on Crime." Journal of Law and Economics. 48.1 (2005): 267-279. Print.
Becker, Gary The Economic Approach to Human Behavior. Chicago [u.a.: Univ. of Chicago Press, 2008. Print. Trickett, Shirley. Coping Successfully with Panic Attacks. London: Sheldon Press, 2009. Print.
Code of Maryland Regulations. Annapolis, MD: Division of State Documents, Office of the Secretary of State, 2011. Continually updated resource.
Draca, Mirko, Stephen Machin, and Robert Witt. Panic on the Streets of London: Police, Crime and the July 2005 Terror Attacks. Bonn: IZA, 2008. Internet resource.
Fienberg, Stephen E, and Albert J. Reiss. Indicators of Crime and Criminal Justice: Quantitative Studies. Washington, D.C: U.S. Dept. of Justice, Bureau of Justice Statistics, 1980. Print. http://www.daggerpress.com/2009/09/22/maryland-state-police-aggressive-driving-initiative-yields-results-on-harford-roads
