LIKE A RICKETY house that was already falling down, the federal flood insurance program has been destroyed by Hurricane Katrina. The theory of the program is that people who choose to live in areas prone to flooding should pay for that risk by buying insurance; they should not expect taxpayers around the country to rescue them from their own recklessness. But the truth is that, after a disaster like Katrina, the federal government will bail everybody out whether they are insured or not; it's humanly and politically unthinkable to do otherwise. Because the likelihood of a federal rescue is so strong, there never was much incentive to buy insurance. The huge federal effort after Katrina will undermine the program further.
The first response must be to make flood insurance compulsory, as auto insurance is. Currently, homeowners have to buy flood insurance if they live in a flood-prone area and they have a mortgage through a federally regulated or insured lender. But other mortgage providers escape this requirement; besides, people who own their homes outright are free to go uninsured, and many who buy insurance as a condition of their mortgage let the coverage lapse later. As a result, fewer than half of the homes in New Orleans had coverage, and the rate in Mississippi was much lower. To improve this situation, Congress needs to mandate coverage for everyone and then come up with a mechanism to enforce compliance. Regulators could oblige insurers that offer homeowners' policies to bundle flood insurance into their protection against fire, theft and so on. Or public authorities could collect flood insurance premiums at the same time they collect property taxes.
The second response must be to price flood insurance properly. In a vain attempt to persuade homeowners to buy coverage, the current system is subsidized; this allows people to build houses in exposed locations and then collect a federal check when the inevitable occurs, sometimes repeatedly. By pricing flood insurance accurately, the government would create price signals that would drive housing development to higher, drier land. Those who remain determined to live below sea level or on the beach -- or developers who put up low-income rental units in these places -- would pay the full cost of their preference. In some areas that have acute shortages of safe land -- New Orleans being one -- this may inflict hardship on the poor. But the best way to address that is to offer means-tested subsidies in a few jurisdictions, not to offer subsidies to rich and poor alike all over the nation.
Compulsory, fully priced insurance would in theory achieve the best of all worlds: It would ensure flood victims of compensation, it would protect taxpayers and it would deter risky construction. But in the real world the purchase of insurance policies will be less than complete, and accurate pricing of insurance risks would require inspecting every beach house to determine the quality of its flood precautions -- which is not likely to be practical. It may therefore be necessary to reinforce the federal construction standards in areas prone to flooding. If the feds are expected to come to the rescue in the event of a flood, it's reasonable to ask that zoning rules limit this liability.