Housing, Another Drag on US Growth

Several economists believe house prices have bottomed and will now begin a slow rise.
 I think they are nuts.

Here is a chart of their current forecast.

___________________________
The difficulty in a general forecast for the country, of course, is that the fall in house prices across the country varies enormously.  The Case-Schiller 20 data below h/t Calculated Risk (CR) shows more than a 10 to 1 variation in house price deterioration from state to state.

____________________________

I like to base my forecasts on nation wide average and median data, so I tend to look at data such as the NYT updated chart below from the Ritholtz blog which shows about another 20%-30% fall in average house prices before reverting to the pre 2000 "norms"

_______________________________________________

The rest of this post will focus on more chart data backing up mybelief that housing prices will continue to fall for some time -- probably for a few years.  The factors causing the continued drop in prices include:

• Excessive household and bank debt
• High and rising unemployment
• Large inventories of unsold houses

Below are a few charts presenting some of the data that I find convincing.

The first chart below is a long term macro view of housing vs income (Ritholtz.)  This is a variation of the "mortgage should not be more than x times income" data looked at historically.  Clearly, the ratio became badly skewed in the early 2000's.  

There is usually a tendency for overshoot in all "mean reverting" phenomena, so I suspect the curve will move towards about 3.8 or so from the current ratio of about 4.4, for a further drop of about 15%-20%. 

_________Data from Ned Davis Research_____

Interest rates, sales and inventory

The most recent sales and inventory data is not very encouraging:   30 year mortgage rates are at their lowest rate since pre WW2

Nevertheless, houses are not selling  

Below are existing home sales (CR) -- back to 1994 levels.  Sales of about 4 million units/year.

And new home sales -- at pre 1964 levels of about 300,000 units/year

And "Months of Supply" of existing housing inventories are at near record highs

As are existing home unit inventories of about 4 million units

Building contractors are nearly an extinct species -- witness new housing starts  

And they show no sign of life -- witness applications for building permits

 

With the sampling of data above and the delinquent mortgage data below --  (Note that 14% of mortgages is about 7 million units)

And bearing in mind that this "shadow inventory" of 14% of delinquent mortgages (approximately 7 million houses) are not yet on the market, I cannot imagine how house prices will rise unless the federal government steps in with some massive, yet to be identified program.

In summary the housing sales/inventory picture looks something like this:

1. Sales rate (existing houses) about 4 million houses/year
2. Current housing inventory for sale (existing houses) about 4 million houses
3. "Shadow inventory" (delinquent mortgages) not yet on market in the region of 7 million houses.  Most of these mortgages will default -- about 60% (4 million houses) are already more than 60 days delinquent.
4. Putting it all together, there is in the region of two years worth of housing inventory across America.  The "norm" is about four to six months.

In future posts I will relate attempt to connect house prices to household debt unemployment.