Automobiles and Hurricanes
After Hurricane Katrina, Tyler Cowen posted a link to a post on Peter Gordon’s weblog, which quoted extensively a Randal O’Toole article, titled “Lack of Automobility Key to New Orleans Tragedy”. The thrust of this post was that the automobile is key to evacuations, and that Katrina proved that car-free cities are a bad idea. Then, after Rita and the debacle that was the exodus from Houston, Tyler posted a suggestion that we pay people who own cars to stay behind.
It’s pretty much a truism that any time there’s a major event, two things will occur immediately afterwards: people with an agenda will find some way, however logically flawed, to argue that the event supports their view; and people (some of them being in the former group, some not) will offer up some pretty ludicrous suggestions. Apparently, Tyler enjoys being a conduit for both phenomena.The problems with the argument posited by Mr. O’Toole are numerous, not least of which being that the conclusion is a complete non sequitur.
Let’s consider a few of his statements, starting with this:
“Economists commonly attribute such declines to increasing wealth. Wealth differences are also credited with the large number of disaster-related deaths in developing nations vs. developed nations. But what makes wealthier societies less vulnerable to natural disaster? There are several factors, but the most important is mobility.Number of Deaths Caused by Hurricanes in the U.S.
1900-1919 10,000
1920-1939 3,751
1940-1959 1,119
1960-1979 453
1980-1999 57″
Of course, improvements in dwellings, levees and other protective structures, and most importantly, weather forecasting and communications couldn’t be more important than mobility. If we weren’t totally stupid, we could test this assertion by comparing the data to other weather phenomena that aren’t affected by mobility– like tornadoes. We’d throw out the 1900-1919 data, because those 10,000 deaths were based on a single catastrophe that was unanticipated, with no attempt at evacuation being considered. Then, given the downward trend from the 1920s until today, we’d compare that to the trend in tornado deaths [second graph]. We’d then reject Mr. O’Toole’s analysis.
He makes this further argument:
“In the late 1980s and 1990s, New Orleans spent at least $15 million converting an abandoned rail line into the 1.5-mile Riverfront Streetcar line. In 2004, New Orleans opened the 3.6-mile Canal Street streetcar line at a cost of nearly $150 million. New Orleans was planning to spend another $120 million on a Desire Street streetcar line.These tourist lines do nothing to help any local residents except for those who happen to own property along the line. The city was not deterred by its own analysis of the Desire line showing that each new rider on this line would cost taxpayers more than $20. About 26,000 low-income families in New Orleans don’t own a car. If all the money spent on New Orleans streetcars from 1985 to the present had been spent instead on helping autoless low-income families achieve mobility, the city would have had more than $6,000 for each such family, enough to buy good used cars for all of them. Add the money the city wanted to spend on the Desire Street streetcar and you have enough to buy a brand-new car for every single autoless low-income family — not a Lexus or BMW, certainly, but a functional source of transportation that would have allowed them to escape the current disaster.” [references deleted]
This argument is flawed on two counts.
First, it undermines Mr. O’Toole’s own assertion that the experience of Katrina provides evidence against car-free cities. He acknowledges that New Orleans didn’t have an effective mass transit system, it had a tourist novelty. If an effective transit network had actually existed, including rail links between cities, it would have been as effective as the extra 26,000 automobiles he calls for. In fact, considering how awful the recent automotive evacuation from Houston was, it may in fact have been more effective– trains can move large numbers of people very, very efficiently. So what, exactly, is his argument? That we shouldn’t rely on government planners? That’s fair, especially given that FEMA turned down an offer from AMTRAK to transport thousands of people out by train. Fine– we won’t ban the automobile, no argument here. But that doesn’t mean that developing good alternatives in the form of mass transit isn’t something we should pursue. Had such a system existed, it would have served the same purpose as, and potentially been more effective than, Mr. O’Toole’s suggestion. And once such alternatives exist, people will be able to freely and rationally choose those alternatives. Mr. O’Toole hasn’t advanced his argument at all yet.
Second, he casually throws out the number of a $6,000 automobile for each of the 26,000 low-income families who lack one. He does this without considering the total costs of automobile ownership– gas, maintenance, insurance, et cetera (assuming we waive vehicle taxes for these families). Good discussions of these costs can be found here and elsewhere– suffice it to say, most poor families can’t afford to maintain and drive a vehicle, even if it is provided free of charge. The per-mile costs of mass transit are significantly lower than those associated with automobiles, so in comparison to $6,000 cars given away once, a real mass transit system rather than an assortment of tourist toys would have been a gift that kept on giving. Add to this the fact that the automotive infrastructure probably cannot support another 26,000 vehicles on the road in New Orleans and that significant additional expenditures would have to be made to accomodate them, and suddenly the Great Automobile Giveaway doesn’t seem so cheap compared to the streetcars anymore. This is, by the way, before considering the negative externalities, like increased smog, associated with every additional car on the road.
The problem in New Orleans wasn’t a lack of cars, it was an abundance of poverty that cars alone would have done little to solve– and that an effective mass transit system would help address far more cheaply than a car giveaway could. It should be said, though, that the solution to such widespread poverty goes far beyond transit alone.
Of course, Mr. O’Toole isn’t interested in addressing poverty. When he states that bad transit networks offer fewer earnings opportunities to workers than autos, he offers a good reason to build better transit networks, which would be accessible at lower cost to the poor than an automobile. Of course, Mr. O’Toole doesn’t acknowledge this, because when someone’s mind is already made up, anything he encounters either doesn’t exist or is a confirmation of his beliefs. It’s a little disappointing, though, to read arguments that are so disingenuous– to argue for cars on behalf of the poor is a step away from “build mass transit and the terrorists win”.
Finally, it’s worth going all the way back to Tyler’s posts for a moment. His second post, which was a little ill-conceived, in that classic economist-ignoring-moral-sentiment sort of way, suggested paying car owners to remain behind, in the path of an oncoming hurricane. This, of course, is exactly what we’re trying to avoid when we order evacuations, but hey, whatever. We’ll ignore the implementation problems with the idea, because it’s bunk to begin with (and I think, or at least hope, that he wasn’t all that serious about it).
If we followed the advice of both posts, there would be every bit as many dead poor people in New Orleans. Why, you ask? If we gave the poor cars, but then paid anyone owning a car who was willing to stay behind (which would represent a larger relative incentive to the poor than anyone else), the poor would have stayed behind (with their cars). We’d be in the same situation as before we spent all that money, except we’d have less money and there would be about 26,000 additional flooded cars in the streets. Genius!