Pittsburgh, phoenix

by Ryan Avent

One of the things I often have occasion to write about at my personal blog is the curious punditry of Ed Glaeser. Glaeser is a very good economist who has done work on urban growth and land-use, upon which I rely very heavily. But when he writes for laypeople, he frequently ends up saying things that are peculiar, or simply wrong.

Let me give you an example. At Economix today, Glaeser writes that the Superbowl was an upset, kind of, because the city of Pittsburgh has spent the last half-century falling apart while Phoenix was one of the past decade’s fastest growing metropolitan areas. And he’s basically right about those two trends. But he omits some important details about the story and says some other strange stuff to boot. He writes:

There is no variable that predicts urban population growth in the 20th century better than January temperature. The figure below illustrates the connection between metropolitan area population growth from 1980 to 2000 and the January temperature. While 19th-century cities formed in places where companies had a productive edge, generally because of access to water ways or coal mines, 20th-century cities formed in pleasant places where people wanted to live.

The transport cost story for manufacturing cities is indisputably true, and the “sunshine is pleasant” story sounds plausible. But then I scoot over to a list of Glaeser’s recent academic work and find a paper from April of 2007 titled, “The Rise of the Sunbelt, and I read this:

We infer that new construction in warm regions represents a growth in supply, rather than demand, from the fact that prices are generally falling relative to the rest of the country.   The relatively slow pace of housing price growth in the Sunbelt, relative to the rest of the country and relative to income growth, also implies that there has been no increase in the willingness to pay for sun-related amenities.  As such, it seems that the growth of the Sunbelt has little to do with the sun.

Which, you know, is the opposite of what he says in the blog post.

Or consider this bit:

The fact that Phoenix has experienced a 42 percent housing price drop since its June 2006 peak is a sign of the area’s strength, not weakness. The high housing prices were always unsustainable, because of Phoenix’s capacity to build. Unrestricted supply meant the price boom was always a mirage. The decline in prices reflect the ability of Phoenix’s great growth machine to create inexpensive housing.

This passage, at least, is consistent with his research — Phoenix, like other Sunbelt cities, has grown largely because of its willingness to build massive amounts of new housing. And he’s right that the high prices in Phoenix were unsustainable. Phoenix basically caught the bubble from Los Angeles and Las Vegas. Price increases didn’t reflect inadequate supply relative to demand.

But this is more troubling for Phoenix than Glaeser lets on. Because supply was built based on phantom demand, the metropolitan area has an enormous inventory overhang. This has led to ghost streets and neighborhoods which will have difficulty recovering, and millions of dollars worth of infrastructure that’s unsupported by residential taxpayers. Worse still, the metropolitan economy of Phoenix relied extraordinarily heavily on home construction. Construction came to employ about one in ten workers in the area during the long housing boom. Given the massive housing overhang, it’s unlikely that most of those jobs will come back, even after the national economy recovers. It’s not surprising, then, that there are signs of population decline in the Phoenix area. I believe the city will ultimately recover, but there is little in this tale that reflects strength, and little worth emulating.

Meanwhile, Pittsburgh isn’t doing all that badly. The New York Times opened a recent piece on the city’s transition from steel to knowledge economy with this:

Unemployment is 5.5 percent, far below the national average. While housing prices sank nearly everywhere in the last year, they rose here. Wages are also up. Foreclosures are comparatively uncommon.

Pittsburgh has focused heavily on using local universities to catalyse growth outside the old industries — the collapse of which did nearly kill the city. Population is still falling in the city, but the fundamentals are far stronger than they were a decade ago, and the lack of a substantial housing boom has enhanced Pittsburgh’s economic position relative to many peer cities.

In the end, Glaeser’s post seems oddly out of sync with other things he’s said and with actual conditions. It’s like part of his brain shuts off when he starts writing something for a non-academic audience.