The Union Bank of Switzerland today confirmed sub-prime losses to date of $12.5 billion, pushing it into its first ever annual loss. This is a bigger amount than those announced in January by Citigroup and Merrill Lynch, and it takes the total write-downs declared so far to $145 billion.
UBS’ Chief Executive Officer Marcel Rohner, on a conference call with journalists, described the results as “unacceptable” and declined to predict whether the bank will return to profit in the first quarter. Rising U.S. subprime-mortgage defaults led to more than $145 billion in writedowns and loan losses at the world’s biggest financial companies. The Group of Seven nations estimates the markdowns may swell to $400 billion, German Finance Minister Peer Steinbrueck said on Feb. 9. Bloomberg
On that estimate, it means that there is not even half of the total losses yet declared, maybe only one third of them.
That’s hearing the story from the big end of the telescope. How do things looks from the other end?
zillow.com has a lot of updated information on the US real estate market, and from there you get a picture of how things look at ground level. Take the case of a single property in Ramona California, near San Diego. A buyer bid $410,000 to a bank for this property which they had re-possessed from a defaulting mortgagee. The bank in question declined his offer saying it was too low. But he merely had a revaluation done by a local agent, and resubmitted the same offer. The second time around, the bank accepted it. The debt they were still owed on the property was however a tad higher than $410,000. It was no less than $580,000.
What one wonders, is how many more of these losses there are having to be dealt with. In 2007 there were 27,000 repossessions in the UK. In the USA, in 2007 there were no less than 2,000,000 individual repossessions. It is now estimated by Lehman Brothers that there will be another 1,000,000 in 2008, and another 1,000,000 in 2009.
To get to the $400 billion total loss figure, given by the German Finance Minister, these 4 million losses would have to be valued at exactly $100,000 each.
Accepting that there will be 4,000,000 home repossessed and not a bigger number for the time being, how likely is it that the losses will comply with the estimated $100,000 per unit? In 2006 when the property peak was reached, property prices in California were at no less than 13.3 times average earnings. Today, they stand at just over 11 times. The problem mentioned on Zillow.com is that property prices historically have been more like 6/7 times average earnings so they probably still need to fall anther 40% in California, and about one third in Arizona and other states, for the market to find a floor.
In effect prices must finally halve from their peak level to get back to a point that will allow the market to rebuild. Taking the property that had the $580,000 mortgage, recently sold at $410,000, that might well have been valued at $800,000 at the peak of the market. It would suggest to me that, based on just this one example, and then on the back of an envelope calculation, the world’s banks will be very lucky indeed if the losses they experience as a result of the sub-prime crisis are limited to $100,000 a house.
The market has as yet only fallen by 8-15% from the 2006 peak, and it could well tumble another 30-40%. There might well in these circumstances be quite a few more properties that get caught up with mortgage defaults, once property owners start to feel the impact of negative equity, which they are bound to do in such circumstances. Why keep paying a mortgage on a house where you are 90% borrowed and have already lost all your equity? You might as well cut your losses and run, and start again somewhere else.
If the number of repossessed houses, instead of the Lehman Bros’ estimated 4,000,000 houses, becomes 6,000,000, and the value of the loss to banks per house is not $100,000 as implied by Peer Steinbrueck’s estimated figures, but $150,000, the sub-prime losses to the world’s banks will turn out to be not $400 billion, but possibly as much as $900 billion.
If that sum is right or about right, it means that so far only 15% or so of the losses in the pipeline are as yet declared.
If one of the Lehman Bros/Steinbrueck estimated figures turns out to be right on the other hand, and the other wrong by around 50%, then maybe 25% of the losses to come have already been declared.
And that’s only the losses in the USA.
The UK government has already committed $200 billion dollars to bailing out the Northern Rock Building Society in January. Is that the last problem of this kind to come to the surface in the UK? The Bradford & Bingley share price fell 16% today, as if traders are aware that there might well be others with trouble similar to the Northern Rock.
Maybe combined with coming European distress from its long-running property boom, the losses between Europe and the USA in total could be nearer $2 trillion, when the last cent is counted.
This crisis has a while to run yet, it seems.
UPDATE – Bloomberg 14th February 2008
Prices have fallen more than 10 percent since their July 2006 peak in the worst U.S. housing slump in 26 years as the number of unsold homes has grown and prospective homeowners had a tougher time getting home loans. As many as 15 million U.S. households may owe more on their mortgages by the end of this year than their homes are worth, according to an estimate by Jan Hatzius, chief U.S. economist at New York-based Goldman Sachs Group Inc.
If Jan is right, the numbers defaulting will very likely substantially exceed the 1 million a year estimate of Lehman Bros. So far the defaulters are happening simply because people can’t meet their monthly payments. This category will grow as it is, as many bought mortgages with short term discounted rates, and the periods of the discounted rates are going to start running out. But once the people struggling to pay the monthly mortgages they’ve taken on, find that their property is tumbling in price as well, many more will decide to quit the struggle and find somewhere cheaper to live.