Application of the Cantor-Land model in the Czech Republic

This paper is based on the methodology of the Cantor-Land model. The main idea of this model is that there exists a double effect of unemployment on property criminality. According to the assumptions of the model, unemployment affects criminality in two ways: in the short term, an increase in unemployment reduces crime through the opportunity effect; in the long term, an increase in unemployment leads to an increase in crime via the motivation effect. In the original version of the Cantor-Land model (Cantor and Land, 1985), the short run effect is expressed by changes in the levels of unemployment, and the long run effect through changes in the differences of unemployment. The original form of the model has been developed by many researchers, with the summary of the research presented in Box (1987) or Michálek (2015). The Cantor-Land model is criticized due to limited validity (for example in Britt, 2001) which is the reason for so much methodology innovation with respect to original idea of the model. An alternative approach that deals with the relationship between property crimes and unemployment is based on the identification of a significant cointegration vector. In our paper, we first estimate the original Cantor–Land model for data from the Czech Republic. Second, we try to identify significant cointegration relationships between the unemployment rate and the property criminality rate. We then construct an error correction model, which is applied to the Czech Republic data. We use time series data of crimes and unemployment that come from the Czech Statistical Bureau.