A Spatial Analysis of Property Crime Rates in South Africa

Abstract The main objective of this paper is to examine the relationship between two types of property-related crime and some socio-economic, demographic and deterrence factors. It employs the conditional autoregressive specification to account for spatial autocorrelation that characterises property-related crime data, which are aggregated up to the level of municipality. First, the analysis confirms the presence of spatial autocorrelation in the data, which means that neighbouring municipalities exhibit similar levels of property crime rates. In addition, and most importantly, empirical findings show that internal migration, youth and education are important predictors of property-related crime across municipalities in South Africa for the period under consideration