Jump to:Page Content
The seasonally adjusted S&P/Case- Shiller index of housing prices in 20 cities has now fallen for nine consecutive months, but before we get all hot and bothered about another collapse, we should recognize the relatively modest nature of the current double dip.
The seasonally adjusted decline since last June has been only about 4 percent, whereas the decline from April 2006 to May 2009 was 32 percent. The current drop seems less like a second big bust, and more like just another phase of a long period of the housing blahs.
The Case-Shiller 20-city series began in January 2000. From then until April 2006, the peak of the bubble, prices rose 105 percent, before beginning their spectacular 32 percent decline in nominal terms, or 36 percent in real terms.
Even the Great Depression didn’t see such drops in housing prices. From 1925 to 1933, nominal prices fell 30 percent, according to the data compiled by Robert Shiller, who created the index with Karl Case, a fellow economist. Correcting for the depression’s deflation, that price fall ended by 1932 and was only 13 percent.
The Case-Shiller series doesn’t pretend to be nationally representative -- it includes just 20 big, and particularly volatile, metropolitan areas. Nationwide, the price shifts appear to have been more modest. The Federal Housing Finance Agency, or FHFA, produces a price series with greater geographic range. Yet its underlying sales data include only homes "whose mortgages have been purchased or securitized by Fannie Mae or Freddie Mac," so it misses housing with non-conforming mortgages, which can be at both the top and the bottom of the market.
That index is typically less volatile, and the boom, which lasted until 2007, according to the FHFA, registered at "only" 65 percent nominal price growth, or 38 percent in real terms. Moreover, the FHFA bust has pretty much continued unabated, and we are now down 21 percent nominally, or 26 percent in real terms.
Still, Case-Shiller gave us hope for a rebound in 2009, and until July 2010 their price index rose by 4.7 percent, or 2.7 percent in real terms. Those hopes have now vanished as prices have fallen 4.3 percent since the post-crash peak, or about 7 percent in real terms.
The U.S. has experienced more than a few housing bubbles and busts, and our post-bust history is remarkably uniform. There are generally modest gyrations, though there has never been either a quick turnaround or a second huge drop. The typical pattern is that nominal prices stay flat for years.
For example, research by the scholars Tom Nicholas and Anna Scherbina on prices in New York City during the Great Depression shows stability after 1931. There was a brief boomlet during 1933, which soon vanished, and prices in 1939 were pretty much the same as in 1931.
The housing boom of the late 1980s peaked in April 1990. There was a brief comeback in 1991; that mini-surge soon vanished and prices stayed even through 1996. Nominal prices in May 1996 and May 1991 were almost exactly the same.
One explanation for these long blah periods is that housing markets are far stickier than stock markets. There are plenty of owners who would like to move but they also would rather stick with their house rather than realize a big loss on their largest investment. Down payments have been wiped out, so people couldn’t buy a new home if they did sell. As homeowners sit and wait until they get a decent offer, prices appear to stay flat.
One common explanation for the speedy price drop in areas such as Phoenix and Las Vegas is that a wave of foreclosures packed the market with institutional owners willing to sell at any price. Once these sales are done, the blahs begin.
Last month’s Case-Shiller data really show more of the same stasis, rather than some big downturn. The overall seasonally adjusted price drop from February to March was .23 percent, which is pretty darn close to zero. Seven out of the 20 metropolitan areas experienced seasonally adjusted price increases.
Economic logic also suggests stability ahead. Prices in the growing parts of the U.S., where land is cheap and permitting is easy, have typically been weighed down by the costs of new construction. During the boom period, prices in some of those cities, including Phoenix and Las Vegas, lost touch with that reality but they are now back down to their historical norms. In other places, such as Dallas and Houston, prices never boomed much and they remain stable, close to the long-run cost of supplying housing.
Rather than expecting huge future price swings -- one way or the other -- in the near future, it makes more sense to plan for the doldrums, and there’s good and bad in that. On the negative side, people do seem to spend more when their houses are worth more, so low housing prices keep consumption down and that probably hurts the macro-economy.
On the plus side, the housing bust made a basic necessity far more affordable. We’re all short housing at some point in our lives and cheaper homes are a blessing for consumers, just like cheaper cars and computers.
The federal government certainly shouldn’t be in the business of artificially boosting housing prices. Why should taxpayers subsidize that particular asset class or try to raise the cost of a core expense?
At the same time, future policy should do more to reflect the fact that booms have always ended in busts. Bank examiners should appraise real estate-linked assets with tools that anticipate the tendency of prices to revert to their historical means. A wiser appraisal policy would counteract the tendency to lend big during booms, which might eliminate some of the suffering that follows.
(Edward Glaeser is a Bloomberg View columnist. The opinions expressed are his own.)