Poverty and Witch Killing

Poverty and Witch Killing, by Edward Miguel, published in The Review of Economic Studies in 2005, is a peer reviewed economics paper using survey data from Tanzania to argue that extreme income shocks, particularly droughts, statistically predict spikes in the murder of women accused of witchcraft, treating the killings as an economic response to scarcity within households under strain. Miguel is a development economist and the paper applies econometric method, regression analysis against rainfall and income data, to a phenomenon usually studied anthropologically. Its strength is a quantitative, testable causal claim linking economic shock to violence, the kind of evidence an economics journal's peer review specifically scrutinizes. Its limit is that an economic model of witch killing as a response to scarcity says relatively little about the belief system itself, why witchcraft accusation specifically rather than another social response, so a claim about the meaning or logic of witchcraft belief in Tanzania needs the anthropological literature on the practice, with this paper cited for its economic correlation rather than for a full explanation.