Last week, Republican House Majority Leader Eric Cantor insisted that disaster relief money for victims of Hurricane Irene has to be offset by spending cuts in other areas. MoveOn.org responded promptly with an ad accusing Cantor of ”abandoning families who have lost everything”. President Obama quickly promised that the government would provide “all the resources that are necessary” thus meeting its “federal obligations”.
Should disaster aid be a government “obligation”? If your home burns down in an electric fire or you get into a car accident, there is no federal program you can apply to for aid. If a tornado destroys your house and leaves everyone else’s untouched, chances are that the government doesn’t declare your area a disaster area, which in turn means that you cannot get any cash from the government (except a tax write-off). Is it reasonable that you should get more money when you lose your house if someone else also lost theirs?
According to the Consolidated Federal Funds Report (CFFR), which tracks federal spending, the US has spent $90.7 billion on disaster aid between 1983 and 2008 for an average of $3.5 billion per year. While this number pales in magnitude to some other spending, such as Medicare, disaster aid is notable in that states pay nothing whatsoever for it. At the very least, disaster aid is redistributive: according to the CFFR, the total amount of disaster aid received between 1983 and 2008 ranged from a low of $74 per person in Washington DC to a high of $41,977 per person in Florida (in 2008 dollars). Not only is this a large difference in funding; it is also not clear whether this redistribution is progressive or regressive (last I checked, people living on Florida coasts weren’t exactly struggling). At worst, the implicit promise of federal aid may encourage more people to live in disaster-prone areas and discourage individuals and cities from implementing protective measures that can reduce damage if nature does strike.
Of course, there is no credible way for the government to commit to not bailing victims of disasters out (as demonstrated by the links above). In addition, we might think it’s unethical to leave victims to fend for themselves, especially if whole towns are left devastated and local employment is affected, making it hard for people to earn money to rebuild their homes. However, why not make people pay disaster aid premiums before they’re affected?
One way I can think of to implement real disaster “insurance” is to mandate that every county or city pay a premium to the federal government based on its risk level and number of people. The city is then free to decide how to raise that money. Presumably, cities are better at setting up proper incentives for their residents than the federal government.
Particularly poor areas could receive premium subsidies based on their population and risk characteristics in the year before the scheme is implemented. This is similar to “grandfathering”, which is implemented in many government regulations. However, in this case “grandfathering” ensures that cities still pay the marginal cost of having extra residents in disaster-prone neighborhoods.
Why not mandate insurance on an individual level? Contrary to what you might think, a large fraction of the aid does not flow into individuals’ pockets. According to FEMA, $5.8 billion was given to individual victims of Katrina. An additional $8.9 billion was given in Public Assistance, of which $5.2 billion was for repairing and rebuilding affected areas. Of course, $5.8 billion is not a small chunk of change, but it makes up only 39% of the total.
One could then do what insurance companies do with individuals – lower premiums for cities that implement measures that reduce their risk. This kind of scheme would largely eliminate concerns about moral hazard, at least on the city level. Of course, it would first have to pass Congress, and I’m guessing the residents of Florida and California will not be thrilled about having to pay for the risks they’re facing.