While commercial damage models do well, my empirical model provides better predictions of aggregate losses immediately after landfall and when there is a large proportion of uninsured and flood losses. There is also room for improvement with high frequency measures of rainfall and storm surge.

The chapter has alot in it and is worth reading in its entirety, but something I am particularly proud of is that we use a model to project hurricane damages under alternative scenarios. Bottom line: higher exposures (income and population) and a changing climate interact to make losses much worse.