Housing Bubble Bursting?

[Scott Kibler]

The thing you have to ask first is just what does 620 signify.

 

[Brad Ellis]

Ready for a Reality Check?

OK- so we have all this press and all the hype, but we are appraisers and we are supposed to deal with facts and not suppositions, right? We shall see.

However, before I get into that, a couple of comments.

Scott- wrongo, bucko! If I hold the loan and season it I get more- not less when I sell it.

Moh,

The highlighted portion of the comments from Schumer (even though he may be misguided in other ways) are right on. Our CEO has already told us to get ready for loan suitability testing. It is in development already and will be implemented well before Congress or anyone else can get off their duffs. And as to some regulation of loan producers in non-bank firms- bravo. Long overdue. I support both of these.

Randolph,

Perhaps part of the problem is who is measuring what and how. NAR uses surveys. Schiller uses resale data only. OFHEO blends in other data but it is limited to conforming loan levels, etc.

Schiller says home prices declined for the first time year over year just last month!. NAR said they have been declining for some months. The median we have been using may well be the wrong measure- perhaps the mean is better- and that shows a year over year decline of 4/10 of 1% for February.

Now- hype vs. fact:

You all heard the horror stories of the delinquency rates on sub-prime and many of you have assumed or even stated that Alt-A is a problem (read carefully , Moh). Here are the actual facts from FALP- First American Loan Performance-

For the period 2002-2006 (yes, it includes all those risky, pesky, loans no one should have made last year too) the actual credit losses are as follows as a percentage of unpaid balance:

Alt-A:

All Alt-A Issuers 4.70 bps (or 0.470%)
#1 Alt-A Issuer 3.60 bps (or 0.360%)
Indymac (just for bragging!) .81 bps (or 0.0081%)

SUBPRIME:

All Subprime Issuers 55.9 bps (or .559%)
#1 Subprime Issuer 22.9 bps (or .229%)
Indymac (just for bragging) 16.4 bps (or .164%)

These are actual results- not by us- not by a biased firm acting on our behalf- just the facts. Pretty clear (I hope) from the above that 1) Alt-A is not subprime; subprime actual losses of UPB (unpaid balance) are 12 times higher +/- than are Alt-A losses and 2) even the industry as a whole the actual losses on subprime are barely above 1/2 of 1%.

Of course I am truly proud of our performance. But more important, I think this independent data shows clearly that all the talk of the subprime meltdown, its impact on the economy, etc. is pretty much overblown hype. And, assumptions that Alt-A has any sort of true relationship to subprime in performance would be an unsupported assumption.

Of course, perception can end up being reality but I mush prefer facts.

Alright, gentlemen, start your keyboards- I AM ready for the spin.

Brad

 

[Brad Ellis]

Another Reality Check

And this just hot off the presses for the New York Times:

http://www.nytimes.com/2007/03/29/business/29scene.html

Brad

 

[Scott Kibler]

Scott- wrongo, bucko! If I hold the loan and season it I get more- not less when I sell it.

Of course you do! You assumed the risk of the FPD/1st year default spike and won.

Watch what happens when your portfolios age without recent attrition rates due to refinancings.

 

[Randolph Kinney]

The thing you have to ask first is just what does 620 signify.

I suspect it signifies that you are going to be paying a higher interest rate than someone who has, say 720, all other things be equal.

It may not be fair, it may not signify your willingness to pay come hell or high water, but that is the way it is.

 

[Randolph Kinney]

Brad, your data for credit losses as measured by or as a percentage of unpaid balance is very low at 0.559% of all subprime loans.

I just don't know why Congress is that upset about it or the federal regulators for that matter. Investors seem to be punishing the entire financial sector for such a small small problem.

Are they looking at it the way you are?

On the declining values, I believe no matter which source you want to look at, they are all telling you the same. The only argument or dispute becomes what is the magnitude of the decline.

 

[Randolph Kinney]

IndyMac: subprime contagion concern overblown

http://www.marketwatch.com/news/story/indymac-says-subprime-contagion-concern/story.aspx?guid=%7B3A6AE3A9%2DD8C1%2D495E%2DAEDE%2D56BD68CC8634%7D&dist=TQP_Mod_mktwN

SAN FRANCISCO (MarketWatch) -- IndyMac Bancorp said on Thursday that concerns about subprime mortgage problems spreading into so-called Alt-A home loans are "overblown."


The company, which was the largest Alt-A originator in 2006, disclosed loss and delinquency data on its loans to show that its business is withstanding the shakeout in the lower end of the mortgage market.

The rate of losses on Alt-A loans IndyMac originated from 2002 through 2006 is less than one basis point. The industry's rate was 4.7 basis points, the company said. (A basis point is one hundredth of a percentage point.)

"There's nothing really new in the data," Paul Miller, an analyst at Friedman, Billings, Ramsey, said. "The data is somewhat misleading on the losses too. They used a five-year time frame. That includes 2002 to 2004 when there were very few losses. People are really worried about 2006."

"The company seems more focused on hyping the stock."

IndyMac is still facing other challenges. Loans that the company sold on to other investors last year are experiencing so-called early-payment defaults, when borrowers miss payments during the initial months of the mortgage, he explained. That's forcing IndyMac to buy back some of the loans, he added.

An exception to this is the company's Alt-A loans originated in 2006. These loans have 60+ day delinquencies of 1.75%, higher than the industry's 1.67% level.

At the end of 2006, the 30+ day delinquency rate for Alt-A loans in the industry was 5%, versus 21.7% for subprime loans, he added.

 

[Scott Kibler]

I suspect it signifies that you are going to be paying a higher interest rate than someone who has, say 720, all other things be equal.

It may not be fair, it may not signify your willingness to pay come hell or high water, but that is the way it is.

Which pretty much sums up the problem with FICO scoring. You know a 720 is better than a 620 but what are the underlying attributes that makes a 620 a 620?

People that write credit policy are privy to the default predictions each scoring benchmark indicates but the people on the front lines in the QC areas often do not. I wonder if the various talking heads in the media do?

I was watching MSNBC one day and they had luscious Wall Street queen Maria Bartiromo on talking about mortgage woes. She made the most comical gesture when queried about the concept of 80/20 loans. She obviously had no clue at that time that some people had such loans, or more importantly, needed such loans to buy a house.

Symptom of the Wall St/Main St disconnect that is part of the sub-prime issue.

 

[Randolph Kinney]

Which pretty much sums up the problem with FICO scoring. You know a 720 is better than a 620 but what are the underlying attributes that makes a 620 a 620?

What the lower FICO score is indicating is the past history and the likelihood that the person will pay his or her bills. You are right: No public information is available to determine what the scores mean in terms of statistics. The statistical models that generate credit scores are subject to federal regulations. If an individual is denied credit, then specific reasons for the denial must be provided to the individual. A statement that the individual "failed to score high enough" is insufficient; the reasons must be specific ("too many delinquencies 60 days or greater").

Fair Isaac Corporation has disclosed the following components and the approximate weighted contribution of each:

• 35% - punctuality of payment in the past (only includes payments later than 30 days past due)
• 30% - the amount of debt, expressed as the ratio of current revolving debt (credit card balances, etc.) to total available revolving credit (credit limits)
• 15% - length of credit history
• 10% - types of credit used (installment, revolving, consumer finance)
• 10% - recent search for credit and/or amount of credit obtained recently

 

[Randolph Kinney]

Subprime Mortgage Market Woes May Curb Spending at Home Depot, Wal-Mart

http://www.bloomberg.com/apps/news?pid=20601014&sid=aKtd6Q3pyzsk&refer=funds

March 29 (Bloomberg) -- Al Ynigues bought his first house in 2004. Since October, his monthly mortgage payment has climbed 16 percent, to $2,417. It will rise again April 1.


Ynigues, 65, makes $2,800 a month as a self-employed music teacher. He says he eats once a day, has stopped paying his utility bills, and is late on payments for his home in Apple Valley, Minnesota, 20 miles south of St. Paul.


``My mortgage has changed everything,'' said Ynigues. ``It's really demoralizing.''


Ynigues is one of about 800,000 U.S. homeowners who took out so-called subprime mortgages and now are struggling to make monthly payments. As these consumers spend less on products including home furnishings and clothing, sales of retailers such as Home Depot Inc. and Wal-Mart Stores Inc. may suffer.