The recent move by the government and leading mortgage banks to freeze sub prime rates is doomed to failure. I believe the intent is admirable, but they are trying open the wrong door, so even if the succeed in getting in the door they would not be standing in the right room.
Like sub prime loans , Alt-A and prime loans will drop soon. I have been saying all along, this is not a sub-prime issue, but a lax underwriting issue - borne from the ability of banks to underwrite loans whose performance they would not be held responsible for (past a short risk retention period) due to securitization. As much as I jump all over the ratings agencies, even some of them are getting it. The government is going to have to give all of us a bail-out, not just the low FICO sub-prime guys. I will glean from other blogs who share my viewpoint on this. Notice how all of us have been in the real estate game, in lieu of simply just analyzing it from afar.
From the SoCalMtGuy:
"I hate to say it, but this thing is going to be ugly, and we are a good 2-3 years from even thinking about the bottom. We
finally started reaching critical mass with 'subprime' foreclosures,
but we haven't even begun to hit the alt-a and a-paper sham loans. They
are starting to pop up, but they will take longer. Most alt-a and
a-paper borrowers were doing 5, 7, and even 10 year ARM loans.
5 years seemed like an eternity when things were good. Appreciation was
guaranteed! Well, I know several people with 5 year ARMs and all of a
sudden 2-3 years doesn't seem so far away...especially the way things
are headed. Sadly, I think many people are going to be in for a nasty
surprise when their 5 year interest-only ARM starts to adjust, and they
realize that their property is worth the same or less than it was 5
years ago, and their mortgage balance hasn't changed one bit!"
"Here are the ALT-A delinquency numbers from CNBC:
2004 - 0.9%
2005 - 1.59%
2006 - 4.21%
To compare, subprime used to be in the 2-4% range, but is now over 14%. That said, I was talking to a contact who deals with capital markets and they were telling me about some pools of 'recent' alt-a loans where the delinquency rate had 18-20% of the loans 60 days late or more.
Alt-a loans were famous for the low and no-doc loans, stated income, stated assets, etc. Just because these people had 'good credit' doesn't mean they made the money to afford the homes they were buying. These borrowers are probably a bit smarter, and will be able to 'hang-on' a bit longer than the typical subprime borrower, but don't think there isn't going to be double digit delinquencies in the ALT-A markets in the near future.
Investors and speculators made up a large part of the alt-a market. As these 'investments' become huge financial burdens, watch them cut their losses by cutting prices to sell, or just foreclosing. These properties will add even more downward pressure on the market."
I have been crooning about this for months now, it is finally getting some media coverage as well - Subprime Woes Hit Credit-Worthy.
On a related topic, I am taking the liberty of annotating a post from the Bubbletracking blog: Prices approaching 2002 levels, on a nominal basis unadjusted for inflation. This is actually more pessimistic than my housing value forecasts, which I have not updated recently.
Dolphine Cove in Otay Mesa is a 121 home community developed by local builder Reynolds Communities in 2002. When the project opened in mid-2002, 1,800 sqft homes went for $350k while 3,000 sqft homes went for mid-$400k.
As the prices took off for outer space in 2003, these same homes went for between $450k to $550k. With the prices spiral up and up some more, prices reached north of $800k with prices per sqft closing in on $300/sqft.
4609 Pacific Riviera Way, San Diego, CA 92154--4 beds, 3 baths, 2,855 sqft--07/2002: purchased from builder for $385,900 ($135/sqft).--10/2006: purchased for $805,000 ($282/sqft).Looks like Mr. 4609 Pacific Riviera is our village idiot here in Dolphin Cove. Unfortunately for this village idiot, a year was all this neighborhood needed to drop his equity by 44%.
4537 Pacific Riviera Way, San Diego, CA 92154
--4 beds, 3 baths, 3,092 sqft
--11/2002: purchased for $473,758 ($153/sqft).
--07/2007: REO back to lender at $725,000 ($234/sqft).
--08/2007: listed for $670,000 ($217/sqft).
--Price Reduced: 08/18/07 --- $670,000 to $599,000 ($194/sqft)
--Price Increased: 08/23/07 -- $599,000 to $650,000 ($210/sqft)
--Price Increased: 09/02/07 -- $650,000 to $670,000 ($217/sqft)
--Price Reduced: 09/21/07 --- $670,000 to $640,000 ($207/sqft)
--Price Reduced: 10/10/07 --- $640,000 to $599,000 ($193/sqft)
--Price Increased: 11/04/07 -- $599,000 to $640,000 ($207/sqft)
--Price Reduced: 11/27/07 --- $640,000 to $490,000 ($158/sqft).It is now December (year end and a reason for the bank to want to get this non-performing asset off of its books), is the bank desperate enough to take an offer at $450,000? If so, then we are solidly in 2002 territory after prices peaked in 2005 to 2006.
--- End Annotation ---
Now, notice how it was REOd back to the bank at $725,000. Did the owner stack a bunch of debt on it through cash out refi's? Probably. Let's supposed they did the cash out refi with Countrywide's (with many CA loans outstanding) or WaMu's no doc option ARM product that I warned about some time ago, at 90% LTV - quite the plausible scenario. Now we have a loan valued at about $765,000 (about what the bank took it back for, which makes sense) that would have been 90% of an (inflated) appraised value of $850,000. Negative amortization combined with falling home prices puts this property's MAXIMUM recoverable asset value below (not at, but below) 62%, gross, before legal fees and transaction expenses, which should be no less than 7%. Factor in carrying costs such as taxes, utilities, repairs... This can add up and will be netted out before the mortgage holder gets paid. Then realize that we still do not have a bottom line price to discover the recovery rate on this house, we just know it will not be $490,000 (62%) or more. I think that many are significantly overestimating recovery value in some areas. The top tranche holders of structured products holding mortgage assets backing communities like this will probably get burned much more than anticipated on this one, and the lower tranches will get close to nothing. This will put heavy pressure on valuations of the builder's inventory as well - see my post on Lennar's haircut.
If you follow my blog, I have warned about this several months ago in the Bubbles, Banks and Builders (and this one, too) series of posts. The banks will compete aggressively with the builders to drive down prices to levels that wouldn't be imagined by many equity and credit analysts today. This will filter through the credit markets. Just ask E-trade why they sold an MBS portfolio valued on their books for $3 billion last quarter for $800 million. Many say this was a distressed sale and did not reflect normal economic activity. I say this is normal economic activity for these assets in this macro environment. There are too many attempts by corporations and government to try and circumvent market pricing mechanisms - from the treasure department to Citibank and the Super SIV to the sub prime bailout measures. These "tricks" cannot be maintained for any significant period of time and simply worsens the eventual downfall.
Hi Reg,
Well I don't agree with your pessimism about all things credit, I do admire the serious work that you are doing. I must say I am having some problems following your Ambac model and I wonder if you would know the following from your model.
1) What cumulative net losses are you projecting in your base case for 2006 and 2007 subprime RMBS collateral?
2) What losses are you projecting under your base case for Ambac's $18 billion HELOC exposure?
3) What losses are you projecting in your base case for Ambac's $1.7 billion in BBB auto securitizations?
4) What losses are you projecting for Ambac's $12 billion of insured credit card securitizations?
5) On a more general note, given your very negative credit outlook, what are you projecting for US GDP in 2008?
With this information, I think many of your readers would be in a better position to compare your assumptions with others.
Thanks and great site!
Posted by: Scott Baker | January 06, 2008 at 12:11 PM