Here's a paper by Nejat Anbarci, Monica Escaleras, and Charles Alan Register:
In our theoretical model, we show that as per capita income decreases and the level of inequality increases, different segments of society are less likely to agree on the distribution of the burden of the necessary collective action, causing the relatively-wealthy simply to self-insure against the disaster while leaving the relatively-poor to its mercy. We then evaluate 269 large earthquakes occurring worldwide (1960-2002), taking into account other factors such as an earthquake's magnitude, depth and proximity to population centers. Using a Negative Binomial estimation strategy with both random and fixed estimators, we find strong evidence of the theoretical model’s predictions.
I haven't read it yet but wanted to pass it along; here are varying drafts, go through JSTOR (gated) and Journal of Public Economics for the final version. By the way, the model predicts bad news for Haiti.
For the pointer I thank the excellent Daniel Sutter, who has done a good deal of work on the economics of natural disasters.