Responding to recent discoveries related to the presidential race, Slate’s Matthew Yglesias argued yesterday that the U.S. generally doesn’t need any more roads going to suburbs:
There may be a some well-justified suburban arterial projects in the United States, but generally speaking this is a terrible use of federal money when what we should be focused on is better management of existing infrastructure. The basic model of suburbanization says that people balance land prices against commute times. If you move further from an employment center you can get a cheaper or bigger place (“drive ’till you qualify”) but you pay that back in financial (gasoline) and time costs of commuting. Adding to the metro area’s roadway capacity serves in the first instance to decrease congestion, speeding commuting times. But with commuting times sped up, the cost-benefit analysis switches and you get new development further out until the road is re-congested and you reach equilibrium again.
As an economic journalist, Yglesias’ argument is useful because it offers a clear financial explanation of the reasons to stop subsidizing suburban arterial roads. He goes on to argue that it would even make sense for many cities to remove freeways, and suggests that if nothing else the federal government shouldn’t subsidize suburban development:
Either way, rather than spend money on new arterials what you want to do is make money by pricing congestion properly. In some cities spending that money on new rail transit capacity would be a good idea, while in other places it would make more sense to just hand it to poor people directly to offset the regressive fiscal impact. Now none of that is to say that if there’s a jurisdiction that wants to take a rural area and build roads to attract suburban development that the federal government needs to stop them. But there’s no particular reason that encouraging this should be a federal priority.
Ultimately there are plenty of reasons not to build big suburban roads. But as Yglesias points out, chief among those reasons is that they don’t really work.