Relationship of unemployment and crime: An application on Turkey (1991-2018)

The high unemployment rates in the face of increasing crime cases in recent years has strengthened the idea that there may be a link between variables. Accordingly, the debate what are the factors that cause the phenomenon of crime has become frequent; the idea that the existence of unemployment has a significant impact on crime rates has come to the fore. At this point, it is very important to examine unemployment and crime together. When the national and international literature is examined, it has been investigated whether there is a meaningful relationship from unemployment to crime in the vast majority of studies. In this study, it was aimed to examine the relationship thought to exist between unemployment and the phenomenon of crime in Turkey sample. For this purpose, in the empirical part of the study, the annual unemployment rates and annual prison population per 100 000 of general population were analyzed as of the period of 1991-2018. The related crime data was obtained from the justice statistics document prepared by the Turkish Statistical Institute, while the IMF database was used for unemployment rates. In the first stage, the stationarity of the series were tested before examining the long-term relationship between the variables subject to the analysis. First, preliminary information was obtained with graphical and correlogram methods, then it was found that the variables were first-order stationary with the help of Augmented Dickey-Fuller and Philips-Perron Unit Root Tests containing statistical results. In the second stage, the long-term relationship mentioned was investigated with the Johansen Co-integration Test, while the short-term causality relationship was examined with the Granger Causality Test. The findings from the analysis show that there is no co-integrated vector between unemployment and crime variables in the period in question. This result is valid for both the trace test and the maximum eigenvalue test. In addition, as a result of the Granger Causality Test conducted to determine whether there is a causality relationship between these two variables, it was found that there was no causality from unemployment to crime or from crime to unemployment. The results, contrary to popular belief, do not support the view that there is a relationship between variables in the period under study.