House prices are down (again) and will continue to go down!

The Case Shiller housing data came out last week. As I have said before, there are many house prices indices but Case Shiller gives the most accurate picture of the housing market, since it measures repeat sales in an apples-to-apples comparison. Hence you should pay attention when the data comes out (the last Tuesday of every month).

The data now covers the month of March 2009 and it is not pretty. In fact, the price decline set new month-to-month records in a number of cities, including New York (-2.5%) and, spectacularly, Minneapolis (-6.1%), apparently the worst monthly decline for any city in the index' history.

Curiously, after Armageddon has crushed many US metro areas for months on end, it is only now that New York is starting to fall off the cliff. After having been declared "immune" to the housing crash by numerous real estate gurus, reality is finally setting in. The 2.5% monthly drop that New York clocked from February to March 2009 may not seem as a lot in light of the disastrous declines other cities have posted but it was in fact the worst monthly decline for New York in the index' 22-year history.

San Francisco bay area, Las Vegas and Phoenix also continued their spectacular declines. San Francisco is now down 46% from the peak in 2006 and the index is back to mid-2000 levels. Las Vegas officially became the second major metro area in the US to decline by more than 50% from the peak (Phoenix passed that milestone in February and is now down 53% from peak level). Detroit continued its slide into the abyss and prices there are now back to 1995 levels (adjusting for inflation they are back to 1980'ies levels).

A couple of cities held up well and actually managed to eek out minor price increases. Most notably, Dallas, Denver and Charlotte were up by a few tenths of a percent. These were cities, however, that never had an exorbitant housing boom to begin with. Charlotte and Houston never flew very high (prices there "only" went up some 25-35% from 2000 to 2006) so there is less room to fall. By contrast, prices in Phoenix went up a good 125% over 6 years. That's why Phoenix is today a smoking crater of sub-prime and option-ARM wealth destruction.

Somewhat remarkably, the price declines in Los Angeles, Washington DC and San Diego slowed down to a crawl, even though these cities are far from bottom. I have no idea why that is (one can speculate that Washington gets a fair amount of giddy business from lobbyists swarming the new administration). However it is a safe bet that these cities are not done falling, and that almost anyone who buys at the current price levels will be sitting on a loss a year or two from now. In LA, for example, prices are still up 60% from the baseline year 2000, so there is a long way to fall.

I have previously predicted that the 20-city Case Shiller index will decline at least 40% from peak to trough before this correction has run its course We are currently at a 32% decline from the peak, so there is another 12% still to go (not 8% because of compounding effects!).

For those who have read my columns on the stock market you may wonder how I can comfortably make these predictions. After all, I maintain that although the stock market may not always be efficient, history has shown that it is exceedingly hard for any individual to predict and profitably speculate on the future direction of prices.

What makes me comfortable about making general statements about the direction of the housing market, by contrast, is the fact that house prices exhibit what economists call "serial correlation". This basically means that if prices fell last month, they are biased towards falling again this month (and vice versa for gains). This is not true for the stock market, but there are a number of reasons why it is correct for the real estate market.

One key difference is that stock prices ideally reflect the future stream of earnings of a company (something that is very hard to predict correctly, but that the stock market nevertheless tries to do every day). Unlike a factory or machine, a house is not a productive asset - it does not throw off products or services and is generally just a shelter.

When housing supply is very high (e.g. due to a poor economy which forces foreclosures) and demand is low (due to tight financing, high degree of uncertainty and high unemployment etc) house prices will fall. And once they start falling for these macro economic reasons, they tend to fall for a long time (typically several years).

When you add to this that we just had the most spectacular real estate bubble in the country's history, and that bubbles always pop and prices come down to pre-bubble levels, it is not hard to predict that house prices will continue to decline probably well into 2010 (if not 2011).