On the optimal design of disaster insurance in a federation

Recent experience with disasters and terrorist attacks in the US indicates that state and local governments rely on the federal sector for support after disasters occur. But these same governments invest in infrastructure designed to reduce vulnerability to natural and man-made hazards. We show that when the federal government is committed to full insurance against disasters, regions will have incentives to under-invest in ex-ante protective measures. We derive the structure of the optimal second-best insurance system when regional governments choose investment levels non-cooperatively and the central government cannot verify regional investment choices. For low probability disasters this will result in lower ex-post intergovernmental transfers (and hence less ex-post redistribution) and greater ex-ante investment. However, the second-best transfer scheme suffers from a time-inconsistency problem. Ex-post, the central government will be driven towards full insurance rather than the second-best grants, which results in a type of soft budget constraint problem. Sub-national governments will anticipate this and reduce their investment in protective infrastructure even further. The result is that the central government may be better off suffering the underinvestment that results with first-best transfers because investment is even lower under second-best transfers when the central government is unable to commit.