First came the hurricane, then came the torrent. We're awash in accusations that the government has done too little to help Katrina's victims. Is it impertinent to ask how much would be enough? What's the right amount of federal assistance for disaster victims?
The suffering that we see on our screens crowds out our instincts for coldblooded policy analysis. But coldblooded policy analysis is our best hope to relieve heartbreak and suffering in the future, so let's steel ourselves for the task. Let's also divorce the discussion from the particulars of Katrina. We're looking for general principles that will apply to a wide range of future disasters.
So, I'll use what economists call a model and humanists call a fable: a simple fiction that has enough in common with reality to focus our attention on some (but not all) of the key issues.
Here's my model: There are two cities; call them Gog and Magog. Gog is subject to violent earthquakes; Magog isn't. Otherwise, they're identical. Except, of course, that housing must be cheaper in Gog; otherwise, nobody would live there.
The people in this model have a choice: They can live cheaply in Gog, where they risk intermittent devastation, or they can pay higher rents in Magog, where they're relatively safe. Because different people have different risk tolerances, some prefer Gog and some prefer Magog.
Now, suppose the government adopts a policy of taxing Magogians to restore the losses of Gogians after any earthquake. The result? All disaster losses are shared equally. There's no longer any financial risk for living in Gog, and Magog absorbs the extra costs for Gog's safety. Gogian housing prices rise because of the government's promise to replace houses damaged by earthquake. Magogian prices drop, because a Magog address no longer insulates you from the financial consequences of future earthquakes. As housing price equalize, there's no longer any reason to prefer one city to the other. Instead of two cities appealing to different kinds of people, we have two cities that are interchangeable.
This is no clear improvement for anyone. Those who preferred to live cheaply and accept some risk are now forced to live more expensively; those who chose to live safely and pay for it are now forced to subsidize the risk-taking of others. That kind of homogenization is exactly what New Orleans has always stood against. It's good to have cities with different cultures, it's good to have cities with different musical heritages, and it's good to have cities with different risk characteristics. Without differences, how can we celebrate diversity?
There is no need to e-mail me with the observation that New Orleans and, say, St. Louis, differ in important ways from Gog and Magog. For one thing, the burden of the Gulf Coast tragedy is hardly being shared equally by the people of New Orleans and St. Louis. But the rest of us are taking on at least a part of the Gulf Coast's pain, and, in the process, making the country a little more homogeneous. For another thing, St. Louis and New Orleans are not identical like Magog and Gog. They differ in important ways that go beyond housing prices and the probability of floods. That's why New Orleans housing doesn't have to be cheaper than St. Louis housing, though it still has to be cheaper than it would be in a world without floods. The model is not 100 percent realistic; that's why it's a model. Models abstract from reality. Their offsetting advantage is that they clearly highlight important policy considerations that might otherwise be overlooked.
One important way this particular model fails to reflect reality is that it takes the locations of Gog and Magog as given. A better model would highlight the fact that a policy of federal disaster relief encourages cities—and the individuals who populate those cities—to locate in potentially disastrous regions.