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.

中文翻译：

---

南非财产犯罪率的空间分析

本文的主要目的是研究两类与财产有关的犯罪与一些社会经济、人口统计和威慑因素之间的关系。它采用条件自回归规范来解释表征与财产相关的犯罪数据的空间自相关，这些数据汇总到市政当局的水平。首先，分析证实了数据中存在空间自相关，这意味着邻近城市的财产犯罪率水平相似。此外，最重要的是，实证结果表明，在所考虑的时期内，国内移民、青年和教育是南非各市与财产相关犯罪的重要预测因素。