How might South Africa's concentrated income inequality affect the incidence of burglary? Theoretically, burglary increases with income and inequality, but the linearity of these relationships can be interrupted by investment in private security. This idea has not been thoroughly tested before, and South Africa presents an interesting context in which to do so, given its high level of income inequality and fear of crime. We find both income and inequality have inverse U-shaped relationships with burglary, rationalized as the outcome of elites investing in private security. Our results also suggest that income and inequality interact: when precincts are richer, more inequality leads to more burglary, but when precincts are poorer, more inequality leads to less burglary. We suggest that when areas are already poor and unequal, more inequality could mean the widening of an already insurmountable gap between the ability of elites to protect themselves and the ability of low-income individuals to break and enter.