Q4 2006 Foreclosure Data Alert

[FRED]

Editor's Note: The following is a report by iTulip contributor Sean O'Toole of proprietary timely and accurate foreclosure data only available in graphical form on iTulip.com. Sean's company will launch the data service shortly, available to investors, hedge funds, and government agencies on a subscription basis.

Q4 2006 Foreclosure Data Alert - January 12, 2007 (1PM EST)

Number of Foreclosed Homes Returned to Lenders at Auction up more than a factor of 10 since July 2006

by Sean O'Toole

Coming 4th Quarter Foreclosure Activity Reports and their implications for housing, lending institutions, and the economy.

The last credible data that was widely published on foreclosure activity in the US was by DataQuick on October 18, 2006 for the 3rd quarter. Most of the reports you have read in the press since then about foreclosure and its implications for housing, sub-prime lenders, asset backed securities, and the economy in general were based on that report.

On one hand, the report was alarming. For Q3’06, CA foreclosures were up 111.8% from Q3’05 and 28.3% from the prior quarter. Despite this strong upward trend, most articles invariably followed up this data with a statement such as: Foreclosures are still at historically low levels, and the effect on today’s market is negligible.

When DataQuick says "historically low levels," they also disclose that they’ve only been tracking foreclosures since 1992, a key fact that many folks who use DataQuick's data don't know is very important. The reason that the start date of comparable foreclosure data for the last cycle is so important is that about two years after the last housing cycle bottom, foreclosures were already declining. If housing were indeed in the process of "bottoming out" now, as many forecast, then foreclosures should be starting to decline now. Instead, they are rising rapidly. Note the DataQuick data also show that foreclosures peaked in Q1’96, which correlates reasonable well with the bottom for the housing market. Foreclosures do not appear to be peaking, so it is unlikely that the housing market is near a bottom, yet.

As such it appears that DataQuick’s data spans a period of time that began during a period of increased foreclosure levels in the last cycle. As a result, we don’t have a clear picture of comparable foreclosure rates for the housing cycle period we are in currently compared to the last cycle, so no one can say whether current levels are "historically low" or not. Even if they are, that may well be a sign of a market peak rather than a bottom, implying further market volatility rather than market stability as usually inferred from the data.

In the coming weeks, as we see the Q4’06 foreclosure report released, it will be interesting to see how easily apparently alarming data can again be brushed aside as insignificant for the housing market, lending institutions, hedge funds and the economy. Unfortunately, due to a lack of sufficient historical data, and lags in current foreclosure data, I suspect this will the case.

Here for example is some data for California that you won't see soon anywhere but here on iTulip.

The number of foreclosures that lenders are taking back in California has increased from an average of 32 a day in August 2006, to 300 a day in December 2006. In dollars that’s an increase from $13.3M per day to $45.9M in four months. So far, for the first week of 2007, the numbers are 161 homes and $63.8M–per day.

If we extrapolate the numbers from the first week of January for the month, we estimate $1.3 billion in loans will return to lenders in January. If the pattern of previous months followed, this number is likely conservative; the rate of increase has tended to rise week by week during previous months.

That’s right, lenders are on track to take back over a billion dollars worth of loans a month in January in California alone. A billion here, a billion there. That may soon add up to real money.

You heard it here first.

__________________________________________________

Special iTulip discounted subscription and pay services flogged here:
For a book that explains iTulip concepts in simple terms, see americasbubbleeconomy
For a macro-economic and geopolitical View from Europe see Europe LEAP/2020
For macro-economic and geopolitical currency ETF advisory services see Crooks on Currencies
For the safest, lowest cost way to buy and trade gold, see The Bullionvault
To receive the iTulip Newsletter or iTulip Alerts, Join our FREE Email Mailing List

Copyright © iTulip, Inc. 1998 - 2007 All Rights Reserved

All information provided "as is" for informational purposes only, not intended for trading purposes or advice. Nothing appearing on this website should be considered a recommendation to buy or to sell any security or related financial instrument. iTulip, Inc. is not liable for any informational errors, incompleteness, or delays, or for any actions taken in reliance on information contained herein. Full Disclaimer

[DemonD]

Re: Q4 2006 Foreclosure Data Alert

Sean - not that you need any positive reinforcement, but this is a bangup job you've done, and I thank you.

One question I have, and I don't know if you can answer it, is that do you know what lenders are being affected the most? Is it still mostly small lenders or are there some big boys like Countrywide and Washington Mutual that are taking hard hits?

Also I'm wondering if these foreclosures will spur an even faster reversion to the mean of housing in california. (Self-disclosure: I really want to buy a house here already!)

[bart]

Re: Q4 2006 Foreclosure Data Alert

Sean - not that you need any positive reinforcement, but this is a bangup job you've done, and I thank you.

One question I have, and I don't know if you can answer it, is that do you know what lenders are being affected the most? Is it still mostly small lenders or are there some big boys like Countrywide and Washington Mutual that are taking hard hits?

Also I'm wondering if these foreclosures will spur an even faster reversion to the mean of housing in california. (Self-disclosure: I really want to buy a house here already!)

Thanks from here too, Sean - very nice job.

DemonD, Russ Winter has some interesting data on his blog in the areas. Here's an extract:

I have been tracking Countrywide Financial’s REOs In California. Thought I check this one a month, but decided to to it now for this post. On Dec. 27th CFC had 360 REOs in California. Just 16 days later they now have 483. At $400k per pop they go from $144 million in inventory to $193 million.

22 6.23%
23 7.18%
24 8.10%
25 10.01%
26 11.92%
27 14.36%

http://wallstreetexaminer.com/blogs/winter/?p=314

[SeanO]

Re: Q4 2006 Foreclosure Data Alert

Sean - not that you need any positive reinforcement, but this is a bangup job you've done, and I thank you.

One question I have, and I don't know if you can answer it, is that do you know what lenders are being affected the most? Is it still mostly small lenders or are there some big boys like Countrywide and Washington Mutual that are taking hard hits?

Also I'm wondering if these foreclosures will spur an even faster reversion to the mean of housing in california. (Self-disclosure: I really want to buy a house here already!)

Thank you, I definitely appreciate the feedback. I can't wait to launch the service in March.

Your question on lenders is one that we are working hard on now. At the moment we know the identity of the originator on about 50% of the loans. We should be far better by launch. Here are some names we see often:

Long Beach Mortgage (subprime arm of WaMu)
New Century Mortgage
Countrywide
Washington Mutual
Well Fargo
Chase Home Financial
Litton Loan Servicing
Option One

[SeanO]

Re: Q4 2006 Foreclosure Data Alert

Thanks from here too, Sean - very nice job.

DemonD, Russ Winter has some interesting data on his blog in the areas. Here's an extract:

http://wallstreetexaminer.com/blogs/winter/?p=314

Thanks Bart! I'm just glad to be able to give a little back. I've learned a lot from you, Eric, Finster, JK, Jim, Ann and everyone else here at iTulip.

One thing to note about Russ Winter's data is that it appears to be based on Countrywide's for sale listings. I don't believe that Countrywide typically posts those properties for sale until they are vacant and have been cleaned up. Since the majority of foreclosures in CA require an eviction after the sale, those listings likely don't appear for 3-5 months after the actual foreclosure sale. As you can see from the charts above, things have changed dramatically in that time frame.

Best,
Sean!

[bart]

Re: Q4 2006 Foreclosure Data Alert

Thanks Bart! I'm just glad to be able to give a little back. I've learned a lot from you, Eric, Finster, JK, Jim, Ann and everyone else here at iTulip.

Same here Sean, on both points... well, maybe except that Finster guy... ;)

One thing to note about Russ Winter's data is that it appears to be based on Countrywide's for sale listings. I don't believe that Countrywide typically posts those properties for sale until they are vacant and have been cleaned up. Since the majority of foreclosures in CA require an eviction after the sale, those listings likely don't appear for 3-5 months after the actual foreclosure sale. As you can see from the charts above, things have changed dramatically in that time frame.

Very true on Russ Winters data, he's very much on the bleeding edge of data and views and he's also quite bearish overall (and also quite "colorful" in his writing). But to see the large jumps in both potential Countrywide REOs, and the possibilities he brought up about subprimes and the quantity and trend of late payments were things of which I was unaware.

[EJ]

Re: Q4 2006 Foreclosure Data Alert

Thank you, I definitely appreciate the feedback. I can't wait to launch the service in March.

Your question on lenders is one that we are working hard on now. At the moment we know the identity of the originator on about 50% of the loans. We should be far better by launch. Here are some names we see often:

Long Beach Mortgage (subprime arm of WaMu)
New Century Mortgage
Countrywide
Washington Mutual
Well Fargo
Chase Home Financial
Litton Loan Servicing
Option One

Just spoke with a friend who works for a lender. Says that in a declining market, nearly all properties return to lenders at auction. Why? Because if they sell at a loss, they have to book a loss. Book too many losses and they may not meet regulatory standards. They'd rather hold the property on their books as assets at inflated prices and wait it out. So, as long as the market is declining, you will see this "properties returned to lender" number rise; it's a function of a declining market. When this happened in the early 1990s, my friend went on, unemployment increased, home price drops became self-reinforcing and many lenders ran out of runway. So, the wild card in this is employment. If it holds up, we may get a soft crash. If it doesn't we go down hard.

[SeanO]

Re: Q4 2006 Foreclosure Data Alert

Just spoke with a friend who works for a lender. Says that in a declining market, nearly all properties return to lenders at auction. Why? Because if they sell at a loss, they have to book a loss. Book too many losses and they may not meet regulatory standards. They'd rather hold the property on their books as assets at inflated prices and wait it out. So, as long as the market is declining, you will see this "properties returned to lender" number rise; it's a function of a declining market.

That's exactly my understanding, and absolutely what the data suggests. I'd add that I also understand that there are regulatory standards around how much property lenders can hold relative to loans outstanding, no matter how inflated the price on the books. So the key is that they can only play this game for so long...

When this happened in the early 1990s, my friend went on, unemployment increased, home price drops became self-reinforcing and many lenders ran out of runway. So, the wild card in this is employment. If it holds up, we may get a soft crash. If it doesn't we go down hard.

While unemployment is probably the right correllation for the 90's, I sincerely doubt it matters this time. Unless you believe the housing market will dramatically improve in 2007, lenders will likely run out of runway before this market corrects regardless of recession or employment.

[Jim Nickerson]

Re: Q4 2006 Foreclosure Data Alert

Sean,

There is a lot on iTulip that lies beyond either my appreciation or comprehension.

In your initial article in this thread you wrote

Here for example is some data for California that you won't see soon anywhere but here on iTulip.

I am not trying to be extraordinarily picky, but does the "some" in that quote mean "a bit of the data for the state of California" or does it mean "the data for the state of California"?

Both the numbers "an average of 32 a day in August 2006" and "300 a day in December 2006" strike me as large numbers. However, 32 might be miniscule or 300 per day might be average and relatively small if from a large database, which I consider the entire state of California to be--despite there being a 10 fold increase in the numbers. The 10 fold is seemingly very impressive.

If you have familiarity with whenever was the top of sales in California, do you know how many sales per day there may have been?

Are there similar data for how many loans were being returned to the lenders for some periods in the state of California when the market was going up?

I look forward to your on-going work.

[jk]

Re: Q4 2006 Foreclosure Data Alert

Just spoke with a friend who works for a lender. Says that in a declining market, nearly all properties return to lenders at auction. Why? Because if they sell at a loss, they have to book a loss. Book too many losses and they may not meet regulatory standards. They'd rather hold the property on their books as assets at inflated prices and wait it out. So, as long as the market is declining, you will see this "properties returned to lender" number rise; it's a function of a declining market. When this happened in the early 1990s, my friend went on, unemployment increased, home price drops became self-reinforcing and many lenders ran out of runway. So, the wild card in this is employment. If it holds up, we may get a soft crash. If it doesn't we go down hard.

is there some way for the lender to write off the inventory as a "one time" adjustment? that's what e.g. intel does. they overproduce to keep their average cost of goods down, then they write off the inventory as a "one time adjustment," then they sell the [now] zero basis goods to give [reported] earnings a boost.

[SeanO]

Re: Q4 2006 Foreclosure Data Alert

I am not trying to be extraordinarily picky, but does the "some" in that quote mean "a bit of the data for the state of California" or does it mean "the data for the state of California"?

The data is for the entire state.

Both the numbers "an average of 32 a day in August 2006" and "300 a day in December 2006" strike me as large numbers. However, 32 might be miniscule or 300 per day might be average and relatively small if from a large database, which I consider the entire state of California to be--despite there being a 10 fold increase in the numbers. The 10 fold is seemingly very impressive.

If you have familiarity with whenever was the top of sales in California, do you know how many sales per day there may have been?

Are there similar data for how many loans were being returned to the lenders for some periods in the state of California when the market was going up?

I look forward to your on-going work.

It's a great point Jim. Unfortunately I'm still building my historical database so I can't give you a definitive answer. But I think we can make a reasonable guess based on previously published reports, and a couple of observations.

First, as to when the market was going up, let's look at a DataQuick report from February 2005. That report said there were 56,125 notices of default filed in CA in 2004. It also states "only about ten percent of the homeowners in the default process actually lose their homes to foreclosure". From that we can estimate 5,612 homes were sold in 2004. Based on my experience buying foreclosures in 2004, I'd estimate that only 1 in 10 of those went back to the lender as investor competition was fierce at the auctions then. So maybe 500-600 properties went back to the lender for all of 2004, vs. 3,241 in December 2006.

That same report also mentions that the peak (at least since they started tracking data in 1992) was 217,410 notices of default in 1996. If you apply their same 10% statement, you'd get 21,741 lost at auction in 1996. Even if we assume they all went back to the lender that is only 1,811/mo.

Which leads me to ask whether or not $1B/mo in bad loans is very much vs. total loans outstanding. I looked around for what the total dollar volume of loan originations were in CA with no luck. My guess is that it isn't that significant, but that it will still be a shock to a system that has largely priced out risk after years of practically no default related losses.

Interesting stuff. Thanks for the great questions Jim!!!

[SeanO]

Re: Q4 2006 Foreclosure Data Alert

is there some way for the lender to write off the inventory as a "one time" adjustment? that's what e.g. intel does. they overproduce to keep their average cost of goods down, then they write off the inventory as a "one time adjustment," then they sell the [now] zero basis goods to give [reported] earnings a boost.

I think it depends on who holds the debt, and how it is held. For instance I wonder what percentage of loans are made by banks using fractional reserves vs. packaged as securities and sold. Do the same "regulatory standards" Eric mentioned above even apply to mortgage backed securities?

I have a suspicion that by packaging and selling the mortgages the banks get themselves out of the regulatory issues around how many non-performing loans they can hold... even if those securities are held by someone who purchased them with dollars borrowed using the same fractional reserve system those regulatory controls were designed to protect.

[EJ]

Re: Q4 2006 Foreclosure Data Alert

I think it depends on who holds the debt, and how it is held. For instance I wonder what percentage of loans are made by banks using fractional reserves vs. packaged as securities and sold. Do the same "regulatory standards" Eric mentioned above even apply to mortgage backed securities?

I have a suspicion that by packaging and selling the mortgages the banks get themselves out of the regulatory issues around how many non-performing loans they can hold... even if those securities are held by someone who purchased them with dollars borrowed using the same fractional reserve system those regulatory controls were designed to protect.

They may get out of the pan of banking regulators but into the fire of securities rating agencies. I spoke at length with hedge fund CEO Jim Finkel in NYC last week who runs a firm that specializes in Collateralized Debt Obligations (CDOs). He expects that since ratings agencies don't really know how to price many Asset Backed Securities (ABS) but haven't had to worry too much about it while the market's been on the way up, when they see anything changing rapidly, say, in spreads, they're likely to start slashing and burning ratings just to be on the safe side. That's what happened in the corp. (junk) bond market early 2000s. It was a bloodbath. In case anyone missed the reference on another thread, William J. Bernstein put it well back in 2000 in Credit Risk: How Much? When?:

There's a pretty clear-cut relationship here: As expected, the higher the spread, the greater the advantage of bearing credit risk.

For a believer in efficient markets, these conclusions are profoundly disturbing, but not unprofitable. Although most of the time, it does not pay to take credit risk, there are periods when expected returns are too high to ignore. Yes, the devout efficient marketeer will point out that there's a reason why one does not often find $10 bills lying on the sidewalk, and that if this junk-bond opportunity were really a free lunch, it would have been arbitraged out long ago. However, there are limits to arbitrage. I ran smack into this limit at a conference of institutional fixed-income managers recently. It was easy to pick out the "spread investors" they were the ones with the deer-in-the-headlights stare and the portfolios suffering from the bond equivalent of irresistible-force-meets-immovable-object. I'm talking about huge mutual-fund-redemption demands running smack into illiquid, impossible-to-price securities. If you're a small investor with modest portfolio exposure to junk, say 1% to 2%, you can afford to wait a few years for prices to recover. These folks looked like they didn't even have a few weeks.

Same thing's likely to happen with ABSs. Jim also mentioned that spreads on the BBB traunch of ABS had grown 200 to 500 basis point over the past two months–a huge jump. He was surprised by your data and is interested. BTW, he's expecting rate hikes this year.

[metalman]

Re: Q4 2006 Foreclosure Data Alert

Jitters Rock Subprime Mortgage Market

Published: January 05, 2007

By Anusha Shrivastava
Of DOW JONES NEWSWIRES

NEW YORK -(Dow Jones)- Investors worried about the slowing housing market and the shuttering of several subprime lenders Friday sent the home equity credit derivative index to its widest levels ever.

The index, the ABX.HE 06-2, has widened considerably in the past few days as investors buy credit protection on the bonds listed in the index.

Friday, the BBB- subindex, the riskiest of the ABX's subindexes, hit 450 basis points over swaps, according Derrick Wulf, portfolio manager at Dwight Asset Management. That compares with its level late Tuesday of 400 basis points and 431 basis points on Thursday.

"It's about time," said Mark Adelson, managing director and head of structured finance research at Nomura Securities. Risk premiums "were way, way too tight and now we have moved into stressful times."

Created to allow investors to take a broader view on the creditworthiness of bonds backed by subprime mortgages - loans to borrowers with low credit standing - the index, which is priced over swaps, widens when investors buy credit protection and narrows when they sell it.

The current index is based on loans originating in the first half of 2006, widely perceived to be risky. Analysts note that the widening is likely to continue throughout 2007 as the 2006 securities age.

The slowing of the housing market, declining home prices and rising delinquency rates as well as the closing of subprime lenders Ownit Mortgage Solutions in California and Sebring Capital Solutions in Texas have led to the expectation of higher losses on the pool of bonds that ultimately underlie the index.

"The fundamentals of the market are weighing on the index," said Alex Pritchartt, a trader at UBS. "People are trying to balance risk with reward."

One of the biggest undercurrents in the asset-backed securities market this week was the announcement by Mortgage Lenders Network in Middletown, Conn. that it has stopped funding loans and accepting new applications for loans.

"There was a slight lag between the news and the impact on the index," said Peter Nolan, senior vice president and product manager for structured finance at Smith Breeden Associates in Chapel Hill, N.C., which has $30 billion in assets.

Nolan said that with sellers outweighing buyers in the current environment, the index's response may be distorted.

Dwight Asset Management's Wulf agreed, noting that the movements of one trader could easily be amplified in a thinly-traded market.

Describing the ABS market as a "fairly responsive" market, Nolan said he expects the widening to continue this year because there does not seem to be any apparent reason for it to tighten.

Indeed, though recent data have shown the decline in the housing market may be bottoming, signals are still mixed overall - existing and new home sales rose in November, but building permits declined as did pending home sales.

"If home prices shot up by 10%, you might see a tightening," Nomura's Adelson said, adding that this is highly unlikely.

http://www.smartmoney.com/bn/ON/inde...05-000854-1554

[FRED]

Re: Q4 2006 Foreclosure Data Alert

The data is for the entire state.

It's a great point Jim. Unfortunately I'm still building my historical database so I can't give you a definitive answer. But I think we can make a reasonable guess based on previously published reports, and a couple of observations.

First, as to when the market was going up, let's look at a DataQuick report from February 2005. That report said there were 56,125 notices of default filed in CA in 2004. It also states "only about ten percent of the homeowners in the default process actually lose their homes to foreclosure". From that we can estimate 5,612 homes were sold in 2004. Based on my experience buying foreclosures in 2004, I'd estimate that only 1 in 10 of those went back to the lender as investor competition was fierce at the auctions then. So maybe 500-600 properties went back to the lender for all of 2004, vs. 3,241 in December 2006.

That same report also mentions that the peak (at least since they started tracking data in 1992) was 217,410 notices of default in 1996. If you apply their same 10% statement, you'd get 21,741 lost at auction in 1996. Even if we assume they all went back to the lender that is only 1,811/mo.

Which leads me to ask whether or not $1B/mo in bad loans is very much vs. total loans outstanding. I looked around for what the total dollar volume of loan originations were in CA with no luck. My guess is that it isn't that significant, but that it will still be a shock to a system that has largely priced out risk after years of practically no default related losses.

Interesting stuff. Thanks for the great questions Jim!!!

Maybe this helps. Using DataQuick data, Houses Sold in CA in the period vs returned to lender: