Features
Person

UK Crime

The aim of this project was to analyse the effectiveness of apprehension and punishment in influencing the UK’s level of crime by using a supply of labour model. The model allowed for an individual’s labour supply and consumption of leisure to both consist of different intensities of legal and illegal activities. It was intended to relate criminal behaviour to economic behaviour so that any individual with a given taste for crime will alter their incentives to changes in the costs and benefits associated with the type of activity. The model is based on a study by Matti Virén which uses Finnish aggregate time series data from the period 1951 – 1995 and compares the results to pooled international cross country data. The study also takes into account the role of demographic and socio-economic factors but the results showed that they were of little importance compared to apprehension and punishment. These factors were therefore included in the study, as I was interested in seeing how significant they are when taking UK data into account.

image
Trends in the probability of crime detection and the level of crime during this time period

Overall I found that the results from my regression only partially agreed with my theory and hypothesis and tests showed that that were some faults with this regression. There are certain measures that can be taken to avoid these errors from happening and they are mainly concerned with maintaining the precise structure of the model, rather than using proxies to estimate the variables.

For example the definition of the parameter α was changed considerably and was assumed that taste for crime is represented by demographic and socio-economic factors. Factors I chose were net migration, population and unemployment but then there are many other factors such as the distribution of wealth and population age structure that would help to determine people’s level of crime. However some of these datasets would require more time and effort to locate which is why they could not be used in representing this parameter. Of course it may be argued that some of these factors might not play any role in determining α, in which case this parameter has been dropped from my regressions, further invalidating my final results. This problem can only be solved by perhaps setting up a survey for each individual of the UK population, asking them to rate their preferences for crime, making α very accurate. This method is very laborious however and cannot possible for me to undertake, but results would be very interesting if this sort of extensive study was carried out.

The method I used for estimating the probability of detecting a crime was criticised earlier on. The probability variable was determined by calculating the ratio of individuals captured to the total number of offences but this is only an estimate based on past data. Accuracy can be increased substantially if a probability reference value could be calculated instead, which might be measured by the strength of the police force and the ability of individuals to avoid being captured. The level of expenditure on training and equipment could measure the former but the latter might require another survey, asking individuals to estimate their probabilities of being captured. This type of survey would be ridiculous for many obvious reasons but honest answers may provide a true estimate of p.

The severity index of punishment is equally as flawed because the average level of time spent in prison is likely to be determined by the type and number of offences, rather than a level set by the authorities as the theory suggests. One possible way I think this could be solved is if the authorities recorded were to construct their own general punishment index that would apply to all types of crime and would be predetermined. This type of data is more likely to exist but was not available for me to use in this study.
I mentioned earlier that the use of national income as a proxy for criminal rate of return in very inaccurate due to conflicts between earnings in the legal sector. The true rate of return the illegal sector would be extremely difficult to measure as many criminals themselves do not know this information until they have committed the crime. It is possible that a significant proportion of criminals in the UK would associate their returns with non-monetary rewards, such as murders in which case the cost of human life would be of great concern to us and would further complicate this measure.

The full text of this project can be downloaded in PDF format by clicking here.

 

Text only version : About the site / Cookies : © 2002-2017

This page has been printed from Revado, http://www.revado.com/
Go online now for pictures and reviews.
© Sam Watling 2002-2017