The idea that the United States does not have a “housing market” but rather a series of segmented regional housing markets became a source of false confidence during the growth of the bubble, leading many policymakers to assert that a nationwide price decline was impossible. That, obviously, was wrong. But it’s still true that we have a series of segmented regional housing markets. At the moment, everything is trending downward, but as Ryan Avent observes behavior diverged considerably across markets from 2001–2008 and there’s good reason to believe that divergence will happen again in the near future:
Generally speaking, there are four different kinds of markets out there. There are the bubble monsters, represented in the chart below by Phoenix, which rose dramatically and are now plunging just as (or perhaps even more) dramatically. There are the zoned zone metropolitan areas, represented below by New York, which rose substantially and then held up fairly well until last autumn, at which point they began falling more rapidly. There are the sunbelt juggernauts, represented below by Charlotte, which saw continued price growth until late last year. And there are the Rust Belt dead zones, represented below by Detroit, which began falling at or before the general market peak and haven’t stopped falling since.
The debunking of the old, wrong conventional wisdom that prices always go up has led to a bit of a counter-wisdom that houses always return to the long-run average. I think there’s good reason to think that’s true in the aggregate but the relative desirability of different places to live does change. A lot of neighborhoods in DC are much better places to live than they were 15 years ago, and regulatory issues prevent the supply of housing from expanding in a commensurate way. Conversely, economic opportunities are vanishing in the rust belt faster than the housing supply is.