Correlation Between Unemployment and Crime Rates in Europe, Research Paper Example

According to macroeconomic theory, increase or decrease of the wages in the labour market can be caused by a various number of reasons; economic and social. However, the prevalence of low wages in any given country has a number of effects on the socio-economic aspect of the country. One of the effects of a decrease in wages is the increase of the crime rate. This paper delves into the mathematical procedures employed in the analysis of the correlation between crime rates and the levels of unemployment. A number of articles shall be reviewed for the correlation and regression analysis employed.

Literature Review

Bahrom A.H and Habibullah M. S. employ a linear representation of a model that describes the problem in question in a general manner. The following linear model is used;

Where: represents the population of a country represents the number of periods.

The model, denoted by the equation 1, implies infeasibility and generality. However, Bahrom and Habibullah restrict it to ensure heterogeneity across the different panel units. The slope coefficients are restricted to allow the intercept to either vary by time or unit, while at the same time the slope coefficients being constant over time and units.

Horst Entorf and Philip Sieger employ two alternative hypotheses to the approach in determining the correlation between unemployment and crime. The first alternative is an opportunity-based behavior, while the second is stigma-based behavior analysis. Both analyses are crucial to determine the relationship between crime and unemployment. The authors employ quantile regressions to examine the conjectured nonlinearities between opportunity-based and stigma-based variables. The dependence between unemployment and crime is mapped out in the following model, quite similar to equation 1:

Where: is the coefficient of interest and it depicts the effect unemployment in a country,, in year has on the levels of crime for the same country in the year in question. Other explanatory variables, such as economic and demographic variables are captured by the vector of the parameters of. Fixed time effects represented by, capture the effects of shocks on crime rates within a country. These sshocks usually affect every country in similar ways. The error in the model is represented by.

Discussion

In the model adopted by Bahrom and Habibullah, an alternative pooled ordinary least squares technique can be applied as opposed to taking into consideration the nature of the panel data. Such an assumption can be represented by;. However, such an assumption would develop a number of complications within the model. This includes serial correlation within panel units and heteroskedasticity across panel units. This model, just like the fixed effects model and random effects model, applies restrictions of thus limiting the constant to differ over.

In Horst Entorf and Philip Sieger’s study, the model used is characterized by unobserved heterogeneity as a result of a number of factors. These include the unreported crimes that are region-specific, which usually result in OLS estimates that are biased and inconsistent. This is solved by adopting a structured panel of the data. This has the advantage of including the individual fixed effects, a feature that is not available when using pooled cross sections of data. Panel data structure allows the researchers to record all the factors that are invariant with time.

Data Description and Sources

A simple regression analysis will be employed in testing the correlation between crime and unemployment. The regression analysis will use total crime as the main testing variable. However, crime contains three categories. They include; 1) intentional homicide, 2) Robbery, 3) burglary of private residential premises. Take into consideration that other crime have not been include, such as grand theft auto and rape.

Data was sourced from Eurostat website and was grouped according to region and year. Unemployment was provided for on a monthly basis. The yearly average for every country was obtained and used for the correlation analysis

The following depicts the correlation analysis:

Regression

R Square (0.03927) reveals a poor fit as only around 4% of the variation in these elements of total crime are explained by the state of unemployment.

Assumptions

Before analysis of the data, it is important to outline the assumptions applied in analysis. An increase in unemployment leads to a corresponding increase in crime rates.

The correlation analysis revealed a positive relationship between unemployment and crime, including all three classifications of crime. However, most surprisingly, unemployment had the least correlation to burglary of private residential homes. Intentional homicide was recorded to have the highest level of correlation to unemployment. Robbery was second most correlated to unemployment.

Owing to the nature of the significance value F, it is important to consider that there may be another variable that may have to be included in the analysis to provide a clearer picture. While most of the classifications of crime have a positive correlation, this correlation is not significantly strong. It may be important to consider the fact that these

Conclusion

The results depict the initial opinion that there would be a positive correlation between unemployment and crime. The regression analysis depicts that the variables had a significant positive effect on the level of crime recorded within a given country. The multiple regression model depicted a negative relationship between unemployment and property crime, this is attributed to the prevailing poverty rates. The robustness testing reveled that poverty rates are jointly significant. The hypothesis, “Unemployment triggers and influences crime rates within Europe”, is validated by the results collected from the study.

The results from this study has numerous applications. One of the most useful applications would be in the economic sector. The development of economic strategy and policy requires statistical analysis such as the one provided. This can also be useful in law enforcement, in the area of profiling suspected criminals.

Real average earnings within any country directly influences the number of reported offences. Crime is motivated by income and wealth, when individuals feel there is need to accumulate more wealth or the simple need to make a living due to various economic factors that create shortages like inflation, some resort to criminal acts to fill that gap. However, it is important to consider that the levels of crime (i.e. the total number of recorded crimes) and rates of crime, (i.e. crimes committed as an average of the population) varies across the board.

References