Hurricanes and Moral Hazard

Yevgeny Shrago

This weekend’s highly hyped hurricane didn’t materialize with the expected vengeance, but it did manage to do significant damage to property across the Mid-Atlantic and Northeast.  Flooded out homeowners find themselves in two very different categories: those with flood insurance and those without.  Since most private insurers won’t provide flood insurance, deeming it too risky, the government steps into the breach and provides the insurance.

Flood insurance is an odd creature. In places where flooding is likely to happen, the government subsidizes the insurance in order to spur development.  Suzy Khimm argues that subsidized insurance creates an incentive for overdevelopment of flood-prone regions, since owners in those areas do not have to pay the full cost of the insurance.  Khimm has several good points about problems with the way this insurance is handled, but the subsidy regime is even more perverse than Khimm recognizes.

First, flood-prone property is often beach-front property, which makes it quite expensive. Subsidized flood insurance not only encourages development, but it props up the values of the already developed homes, since the owners don’t have to worry about losing their investment.   When floods happen, the insurance program payout includes the bump in house value from the subsidized insurance.  Therefore, the government is paying for the benefit it bestows on the homeowners.  Second, because the insurance is only subsidized in flood-prone areas, residents of other areas see something closer to market rates—and given the low chances that they will be flooded out, probably won’t buy flood insurance.  When an extreme weather event like Irene causes substantial inland flooding, the people living outside of flood-prone areas lose their houses and their savings, while the ones who were living dangerously get a check for their troubles.

The problem with Khimm’s argument for eliminating the flood insurance subsidy is that it won’t end the subsidy to people living in the flood-prone areas.  Edward Prescott and Finn Kydland won a Nobel Prize in Economics in 2004 in part for their discussion of this problem in their 1977 paper, “Rules Rather than Discretion: The inconsistency of optimal planning.” In the paper, Prescott and Kydland point out that governments are not credible actors and use the example of homeowners building houses in a flood plain despite the government’s stated refusal to provide flood-control measures.  The authors point out that homeowners, as rational agents, will realize that once they have build the houses, the government will have no choice but to take the flood control measures, thereby subsidizing development in flood prone areas.  Without subsidized flood insurance, we will only see increased pressure to provide costly flood protection in advance of every major storm or some other expensive flood countermeasure.

Kydland and Prescott have an answer for this problem, and it’s a blunt one.  If an area is particularly flood prone, then the government should ban development.  If that’s not politically feasible, then a tax on development in the region can force homeowners to at least bear a larger share of the costs of development.

Old Paper by