As you read this article about what economists call “housing elasticity,” picture Flagstaff as separate from Arizona. Flagstaff’s housing situation is more like the California coastal areas described here, where prices inflated during the boom but there was not excessive building, than it is like Phoenix where prices inflated and too many homes were built because there were no physical (topographical) barriers to building. Here is this interesting piece from today's New York Times based upon work by Edward L. Glaeser, Harvard economics professor:
“The Census Bureau recently released estimates of state-level population changes, which showed a decline in the growth of some Sun Belt areas. The growth of much of the Sun Belt has been achieved through vast amounts of construction providing low-cost, high-quality homes for ordinary Americans. New construction has declined by more than two-thirds during this recession, and this fall in new building has led to a fall in the population growth of the places that had been building most.
“This shift should remind us that America’s future is being determined, in no small part, by its housing industry, and that events that affect the supply of housing ripple throughout all of America.
“The chart [on the right] shows the relationship between housing units and metropolitan-area population growth from 1970 to 2000. It shows that the supply of new homes in an area is almost perfectly correlated with the growth of population in that area. This makes sense, since people need homes to live in.
"But this tight relationship doesn’t automatically imply that anything about housing determines population flows. After all, it could be that population growth is driven by only other desirable, demand-side factors — like wages, unemployment and sunshine — and that new construction just follows along. Perhaps people just move to attractive areas and homes are built for them. (Ann: Did we need economists to tell us this?)
"But if every place had an identical housing supply and housing demand was the only thing that mattered, then high-demand places should have high prices and high levels of construction. Low-demand places should have low prices and low levels of construction. *The chart below shows the connection between prices in 2005, and housing permits between 2000 and 2005 (as a fraction of housing stock in 2000). As you can see, price and permitting don’t move together:
Edward L. Glaeser
"Places that are really expensive don’t build a lot. (Ann: Think Flagstaff.) Places that build a lot aren’t really expensive. That fact is incompatible with the view that housing demand is the only thing that matters and that housing supply is more or less irrelevant.
"Indeed, there is a growing literature emphasizing the power of housing supply in explaining regional growth patterns.
"Joseph Gyourko and I have argued that the long, slow declines of cities like Detroit are best understood as the result of a particular feature of housing supply: Homes are durable and last for decades or centuries. As a result, places that are hit by economic shocks don’t disappear overnight. Their homes remain, and remain occupied, because they are still valuable assets.
"A recent paper by a star M.I.T. graduate student, Matt Notowigdo, comes to a similar conclusion. Kristina Tobio and I have found that much of the recent growth of the Sun Belt can be explained by that region’s elastic housing supply. Albert Saiz at Wharton has pushed the literature further by showing how natural barriers to building — like hills — can explain patterns of prices and local growth. (Ann: Flagstaff's barriers are the abundance of National Forest and National Parks land -- great barriers to have, but barriers nonetheless.)
"So what does the significance of housing supply mean for Sun Belt growth and the recession?
"America vastly overbuilt during the 2003-6 period, but that overbuilding wasn’t uniform across our country. Booming Sun Belt cities — Atlanta, Phoenix, Miami, Las Vegas, Tampa — had the most extreme combinations of high prices and oversupply. Houston and Dallas also built plenty, but their prices stayed earthbound and as a result, they don’t seem to have overbuilt. Coastal America had booming prices, but regulatory and natural limits on building restricted construction. And now we see the sharpest declines in the rate of population growth in places that had the most overbuilding.
"The four states where population growth fell most in 2009, relative to statewide trends in 2000-8, are Arizona, Georgia, Nevada and — above all — Florida. These are not the states with the worst economic conditions today. (California and Michigan have that honor.) They are, however, places that had particularly extreme housing booms — measured by both quantities and prices. No one should be surprised that their population growth rates have fallen furthest.
"Last year, Professor Gyourko and I wrote a paper about the interaction of housing supply and bubbles. We argued that while bubbles should be rarer in places where housing supply is more elastic — that is, where builders are more sensitive to price changes — overbuilding will actually be more severe in places where housing supply is more elastic.
"The inevitable consequence of that overbuilding is a housing hangover that depresses future population growth. That’s what Arizona, Florida and Nevada are experiencing now.
"The housing bubble distorted the American economy—not just portfolios but the very contours of our country. (Ann: Think Anthem, Queen Creek, Surprise.) Boomtowns grew spectacularly as people overpaid and developers overbuilt. Right now a few Sun Belt places are growing less, as the market undoes some of that distortion. But no one should count these Arizonas, Floridas and Las Vegases out.
"For decades, growth has gone to places with sunshine and unfettered housing supply. A cyclical downturn is unlikely to undo that long-term trend."
Ann: Not to be too pessimistic, but in the “long-term,” we’re all dead. More seriously, the question is not whether Flagstaff’s homes market will recover its 2006-07 price peak, but when. | 
-->
Email to a Friend
