Bryan Caplan  

I've Got a Little List

Henderson Speech at Hillsdale ... The Narrative of the Financial...
Ryan Avent's not happy with my exploratory list of anti-suburban regulations:

This is truly a remarkable list. One thing to note is that Caplan doesn't seem to grasp that criticisms of pro-suburban policies are largely about the forms that are encouraged rather than the locations. To the extent that regulations against developing empty land and government land ownership are actually major issues constraining metropolitan growth, they're barriers to development of any kind -- "urban" or "suburban" in form...

Legally, that's true.  But as a practical matter, there's a lot more empty and government land  in suburbia (or potential suburbia) than there is in major cities.  This is the mirror image of, say, a law capping buildings at ten stories: While the law applies equally to cities and suburbs, it matters a lot more for cities, because that's where people are eager to build tall buildings.

Avent continues:

I must admit that my mind is a little blown by Caplan's citation of "regulations against mixed use" as an anti-suburban policy. It's like saying that government is anti-suburb, because it forces people in suburbs to build using suburban development forms which are inconvenient and unpleasant.

Maybe my use of language is idiosyncratic.  But to me, the essence of suburbia is low-to-moderate density and widespread use of cars, not rigid separation of residential and commercial buildings.  Replacing one of the houses on my block with a small shopping center wouldn't make it any less suburban.  In fact, a lot of the suburbs in the San Fernando Valley where I grew up have such shopping corners.

And then Caplan says that gas taxes in congested areas, which means most of metropolitan America, should actually maybe be increased.

For the record, I only said that taxes in congested areas might make suburbia "more desirable," not that they "should actually maybe be increased."

He seems unwilling to wrestle with the fact that increasing gas taxes to reduce congestion would make suburbia more desirable and more efficient, and also (and necessarily) more expensive and smaller. That is, if you used higher gas taxes to get rid of the negative externalities produced by suburban growth, you'd probably wind up with less of it (so long as demand curves slope downward).

This is more complicated than it seems.  If a tax genuinely corrects for a negative congestion externality, it reduces driving (especially during peak times), but could actually on net increase demand to live in the area with congestion taxes.  Think about it this way: Suppose I-66 got so congested that it took 4 hours to get into DC from Fairfax.  Real estate prices in Fairfax would probably be pretty low due to the incredible inconvenience of getting into the city.  If you imposed a gas tax - or better yet, peak-load pricing - fewer people would drive.  But demand for Fairfax real estate could easily go up, increasing both the price and quantity of homes in this newly convenient location.

Or consider this hypothetical: Suppose a firm built a private highway from Winchester to DC.  The speed limit is 90 mph, and the tolls vary electronically to keep traffic moving at the speed limit throughout the day.  How would real estate development respond?  Consider: On my hypothetical road, commuters could cover the 74 miles from Winchester to DC in 49 minutes.  I predict a lot of McMansions shooting up around Winchester, even if the monthly toll were $600/month.  Am I wrong?

I admit that urban economics isn't my specialty.  If someone like Ed Glaeser told me that the government policies on my list weren't empirically important, he'd have my attention.  But for now, I'll stick to my guns.

P.S. Tyler chimes in here and here.

Comments and Sharing

COMMENTS (6 to date)
John Thacker writes:
If someone like Ed Glaeser told me that the government policies on my list weren't empirically important, he'd have my attention.

Ed believes that the balance of policies are pro-suburb. Randal O'Toole believes the opposite. Both are libertarians, so they're focused on means, and believe that whatever ends come from just means are basically okay, so they have the same policy recommendations (less regulation) and have no problem with each other.

Liberals seem more likely to focus on the ends of being anti-sprawl, so those like Ryan Avent will generally like Ed Glaeser (except when he says something they don't like, like high speed rail won't work) and hate Randal O'Toole.

aaron writes:

Ryan is a smart guy, but not very reasonable. He's more concerned with what he wants than the reality of what people want and what is good for them. He has some good points occasionaly and inklings of understanding of complex systems, but he's too damned religious. He can't accept that things aren't so cut and dry.

Brian Shelley writes:


One of the basic errors that you and Ryan seem to be making is that with notable exceptions few cities have a high percentage of jobs in the city center. Newer cities such as Los Angeles and Dallas have less than 10% of jobs in the city center. You have to think of them as multi-nodal webs, not as spoke and wheel. When you understand that jobs are more spread out you begin to understand that gas taxes won't necessarily kill the suburbs, they may kill the cities. At the very least gas taxes limit the efficient allocation of labor by increasing travel costs. This diminishes wealth until the glacial change of development patterns move to accomodate the new reality.

El Presidente writes:

Short answer:

No, you're probably not wrong about that particular example, but you're probably very wrong about many other places. Since the fight is over policy with regard to suburbia generally, yours is probably not the best example.

Long Answer:

The first thing to take into account is that there is a competitive equilibrium between communities for firms and workers. That said, the question really is one of whether the tax would crowd fixed capital investment in or out. That will determine whether you get stagnation or a cleaving and therefore increasing density and localization/urbanization economies of scale in one or both. The bid rent curves dictate the spread of growth and they are driven by the output and real wages in each area. Basically, Fairfax would become a more attractive place for retirees and a less attractive place for professionals while DC would probably increase its housing stock. If the areas had strong predilections for growth, like Los Angeles, then this policy would likely shape future growth more than thwarting it.

Your example of a congestion tax in Fairfax is dependent upon the income elasticity and price elasticity of demand for automobile transportation. $600/month might not be so harsh for those who can afford to live in Fairfax, but in other places it would be quite a hefty chunk of an individual's income ($7200/yr = almost half of full-time income at federal minimum wage). Whether it crowds fixed capital investment in or out, whether it attracts greater population density or causes population to plateau, it stands to reduce investment in the in-between infrastructure because it directly penalizes commuting and thereby improves the cost-benefit ratio of infrastructure spending in the region over time. It raises the question of whether the congestion, the tax, or increased infrastructure spending would have a greater cost-benefit ratio. That also requires knowing what the tax revenue would be spent on. That's an empirical question, not a theoretical one. In many places, it could go either way. Between Fairfax and DC, it probably would simply slow growth in population and reduce congestion, but it would also reduce the shared labor pool and that could have negative effects on certain employers and reduce the availability and/or increase the price of services to households.

Peter Gordon writes:

It is surprising how little effect policy differences have on the extent of suburbanization (auto-oriented development).

Patrick McCann writes:

A recently very important pro-suburb rule is the body of regulations surrounding financing FHA condo buying. Since FHA loans are becoming increasingly dominant as a part of the marketplace, the extra difficulty of securing an FHA loan for a condo is rather pro-suburb. The most stringent requirement is that 80% of the entire building be owner-occupied (rarely happens in a condo project) to even apply to the approved buildings list. Then buyers counting on FHA financing must search the approved buildings list and cross-reference that with real estate listings on the MLS.

Here is some background:

Here is an article:

Comments for this entry have been closed
Return to top