The Macroeconomic Consequences of Mixing Sunnis and Shias: A Bayesian Errors-in-Variables Approach

How damaging is the Sunni/Shia split to the economies of Islamic countries? Is it better to be one or the other? Or is it better to have an even balance between the two? Answering these questions is complicated by the fact that the data are often missing or imprecisely measured. We employ the technique of Bayesian data augmentation to circumvent these two problems, and find that properly controlling for these features in the data leads to drastically different conclusions than what is found using ordinary least squares. We find that there is nothing in the differential nature of Sunni or Shia Islam to make one more economically prosperous than the other. Nor do we find any support for the popular hypothesis that Sunnis and Shias cannot mix.