In this post, I’m going to sketch out some ideas about suburban sprawl, and then hopefully flesh them out in subsequent blog posts. I don’t really have the time to do the research right now, so forgive me if the concepts are vague.
I’ve written a couple things about rural areas and subsidies, but lately there has been a lot of commentary about John Stossel’s recent report about suburban sprawl. In it, Stossel makes a fairly ridiculous claim that suburban sprawl should not be hated because it is a reasonable market outcome.
Now I generally like John Stossel, even if his analysis is often overly-simple…and I don’t exactly dislike suburban sprawl. However, as has been pointed out quite a few times, suburban sprawl is definitely a product of state planning. Furthermore, sprawl is the direct product of implicit subsidies that allow for the cheap operation of motor vehicles…which increases the distance people live from various destinations.
One obvious subsidy is roads. When you drive, you are creating all kinds of externalities that you are not held responsible for. Most people implicitly identify the pollution cost — but a much more pertinent cost is congestion. As should be obvious to anyone who has tried to run through a door, two objects cannot occupy the same physical space at the same time. The fact that you car is on the road at any given time means that another car cannot be there…even though there may be a driver that is willing to pay a higher price to occupy your spot. Because travelling on roads is ostensibly free, you are imposing a cost upon them that you do not have to pay for. Furthermore, the very existence of free roads creates a congestion price — remember, supply is demand, however, this price is very low (simple inconvenience). Civil engineers have often been puzzled by the removal of roads not creating congestion elsewhere. This is because expectations change. Furthermore, increasing the availability and convenience of public transportation increases automobile traffic, because of changes to congestion expectations.
How do we square this circle? Congestion pricing? Pigouvian taxes? As long as roads remain “public goods”, planning is a necessity. However, as long as the planning is done by government, it is unlikely that an optimal plan will be presented.
A more implicit subsidy is housing. Since the post-war boom, Federal and state governments have seen it as “their job” to promote homeownership. Remember Bush’s “ownership society”? The housing market is veritably owned and directed by the government — and as long as this is the case, we will end up with sprawl. Remember, two single-family units cannot occupy the same physical space. Also remember, if you subsidize something, you get more of it. The government heavily subsidizes homeownership…so we get more homes, and less muti-dwelling units.
But the question is, how much planning would walkable urbanism require? I’m not as much of a fan of the concept as someone like Matt Yglesias…but it definitely does have benefits, and the truth of the matter is that our current fascination for single-dwelling units in suburbs is a result of preference training for the last half-century…not something that is inherent in “Americanism”. The US has, in fact, planned for suburban sprawl, but would we really need to plan for walkable urbanism? I think it would require a less of a degree of planning than our current situation, and even much less than planning proponents, like Yglesias and Avent, recognize. Without the current subsidies and regulations, the market bias would be toward density. Suburban, exurban, and rural living would be much more costly.
I’m hoping that lays out a good enough foundation for a right-wing liberal argument that it gives you an idea about the position I’m arguing from. I think it’s deceptively easy for leftists to argue that we just need a different kind of planning, and pass that off as a realistic goal…but I don’t buy it. Regulation doesn’t happen in a vacuum, and the incentives of regulators are not on the side of urbanism.
As always, any input would be appreciated =].