Housing Bottom -- NOT!!!!!


For economic commentary and analysis, go to the Bonddad blog

There has been a great deal of discussion among economists about whether or not the housing market has bottomed out. News from the last two days indicates we aren't near a bottom -- and may now be near the bottom for awhile.

New Home Sales Plunge:

Sales of new homes plunged 16.6% in January to a seasonally adjusted annual rate of 937,000, the Commerce Department reported Wednesday.

It was the lowest sales pace in four years and represented the biggest percentage decline in 13 years.

Sales were down 20.1% compared with January 2006.

As if that weren't had enough, prices are decreasing and inventory is increasing. The months of available inventory increased to 6.8 months and prices dropped 2.1% on a year-over year basis.

And there is troubling news on the mortgage front. One of the largest subprime lenders reported:

Countrywide Financial Corp., the biggest U.S. mortgage lender, said payments were late at the end of last year on almost 20 percent of the subprime loans it tracks for other companies and investors who own them.

Delinquencies of at least 30 days on ``nonprime'' loans, those made to borrowers whose credit rating fell short of the highest criteria, widened to 19 percent as of Dec. 31 from 15 percent a year earlier, the Calabasas, California-based lender said in an annual regulatory filing with the U.S. Securities and Exchange Commission. The rate stood at 17 percent at the end of September, according to the company's last quarterly filing.

As is that weren't enough,

The mortgage market has been roiled by a sharp increase in bad loans made to borrowers with weak credit. Now there are signs that the pain is spreading upward.

At issue are mortgages made to people who fall in the gray area between "prime" (borrowers considered the best credit risks) and "subprime" (borrowers considered the greatest credit risks). A record $400 billion of these midlevel loans -- which are known in the industry as "Alt-A" mortgages -- were originated last year, up from $85 billion in 2003, according to Inside Mortgage Finance, a trade publication. Alt-A loans accounted for roughly 16% of mortgage originations last year and subprime loans an additional 24%.

The catch-all Alt-A category includes many of the innovative products that helped fuel the housing boom, such as mortgages that carry little, if any, documentation of income or assets, and so-called option adjustable-rate mortgages, which give borrowers multiple payment choices but can lead to a rising loan balance. Loans taken by investors buying homes they don't plan to occupy themselves can also fall into the Alt-A category.

....

To be sure, defaults have remained very low in the prime market -- and despite the uptick in bad loans, the problems in the Alt-A sector aren't as severe as those that have roiled the subprime market. Some 2.4% of Alt-A loans are at least 60 days past due, according to UBS, which looked at mortgages that were packaged into securities and sold to investors. That is well below the 10.5% delinquency rate for subprime mortgages. (During the housing boom, delinquencies were low for all types of loans because borrowers who wound up in trouble could refinance or sell.)

Note that about 40% of loans written on 2006 were either sub-prime or alt-A. That's a lot of risk in the system. Also note the 4-fold increase in the amount of alternate debt. That's also a big increase in the amount of risk.

What all of this tells us is this could be a very long year in housing.


Bonddad March 1, 2007 - 1:15pm

Wall Street must be quietly but furiously rethinking the models it used to structure and price these mortgage-backed securities. Were the historical data underlying these models flawed in some way? Maybe not enough data points far enough back in time? Maybe too many discrete but useless categories? How does "subprime" differ from "nonprime", and do they all fit into the "Alt-A" category? And by Wall Street, I include Standard & Poors and Moody's, who rated these issues one after another as AAA but failed to predict the toxicity of these categories (I prefer to call these subprime borrowers the Not Ready for Prime Time players - maybe not original, but apt).

Here's what I think happened. We have a long history of home mortgages behaving in a certain way, just as house valuations used to be predictable. It took a personal calamity like a lost job or major illness for a borrower to fall behind on payments. Similarly, housing values were predictable - they would increase by the rate of inflation year after year, never straying too far from serving as a basic inflation hedge.

That all changed around 1995. Mortgages became much more exotic under the illusion that "tailor made" and "flexible" products were necessary to distinguish one mortgage broker from the rest of the competition. Home owners were given HELOCS as a tool to "take some cash" out of their home (no mention of the exchange for debt in the process). Home equity withdrawals became tantalizing because right around this time house valuations began to soar well beyond the rate of inflation.

The miracle of rapid price appreciation opened the door for previously-outlawed products like negative amortization mortgages. All risks would float away unto all those investors buying mortgage-backed securities, and they didn't have to worry because rapidly rising home values protected themselves and the borrower.

It's easy for management to loosen up modeling and analytic standards in such marvelous market conditions. If the model can't really say how millions of borrowers new to mortgages would react when they were stuck in the loan, with no chance to refinance at a higher valuation, but with higher monthly payments staring them in the face once the teaser features expired, then who cares? This is a hypothetical situation that can be ignored, because as we all know, house values never go down. In fact, they've been going up at double-digit rates for ten years, so let's extrapolate that experience into our modeling while we are at it.

Even if the modeling never went quite that far, it seems very obvious that senior management everywhere in the mortgage business extrapolated existing conditions to justify the increased risk. This is what bankers do - they rarely have the sense to buck the business cycle or the herd mentality. They just join the competition and rationalize the increased risk as the business cycle lengthens. Then, when things blow up, the reaction always is "nobody could have predicted the depth of the problem" When you start hearing this out of the mouths of the mortgage industry experts and players, you'll know we will be getting close to a bottom for the housing market.

Numerian March 1, 2007 - 1:59pm

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.