by Calculated Risk on 12/03/2007 02:54:00 PM

From Bloomberg: WestLB, HSH Nordbank Bail Out $15 Billion of SIVs (hat tip Brian)

WestLB AG ... and Hamburg-based HSH Nordbank AG provided financing to more than $15 billion of troubled investment funds to prevent a fire sale of their assets.

WestLB provided a credit line for its $11 billion structured investment vehicle called Harrier Finance to repay commercial paper, the Dusseldorf-based bank said in an e-mailed statement today. HSH Nordbank said it will provide backup funding to cover all commercial paper issued by its 3.3 billion- euro ($4.8 billion) Carrera Capital SIV, spokesman Reinhard Schmid said in an interview.

To understand these stories, it helps to understand the structure of an SIV (Structured Investment Vehicle). (see SIV Accounting for more)

First an SIV has investors - like hedge funds or wealthy individuals - who invest say $1 Billion in the SIV (the equity). Then the SIV issues commercial paper (CP) and medium-term notes (MTN) that pay slightly higher rates than similar duration paper. The typical SIV, according to Fitch, uses 14 times leverage, so in our example the SIV would sell CP and MTN for $14 Billion.

Now the SIV invests this $15 Billion ($1 Billion equity and $14 Billion borrowed) in longer term notes. The idea is simple: borrow short, lend long, hedge the interest rate and credit risks - and the profits flow to the investors in the SIV.

Back to the story: what happens when the CP comes due and no one wants to buy any more? To cover the CP, the SIV might have to sell the longer term assets at a steep discount, and this would trigger a liquidation of the entire SIV. To prevent this "fire sale", the sponsoring banks stepped up and provided the financing to cover the expiring CP.

Of course this limits the banks ability to make other loans (aka Credit Crunch). Perhaps this story is related: Banks Urge UK Clients To Stop Borrowing (hat tip FFIDC)

The banks are urging some of their biggest clients not to draw on standby credit facilities as the sub-prime crisis and squeeze on interbank lending have affected banks' ability to fund themselves.

by Calculated Risk on 12/03/2007 02:07:00 PM

Ahhh, I'm reminded of Tanta's post in early June: Reelin' In the Suckers

Once again, from David Evans at Bloomberg: Montana, Connecticut Hold SIVs Downgraded, Reviewed by Moody's (hat tip energyecon)

Montana and Connecticut state-run investment funds hold debt tainted by the subprime mortgage collapse that was cut or put under review by Moody's Investors Service, leaving local governments vulnerable to losses.

... Montana owns $50 million of the paper. Moody's put another $105 billion of SIVs on review for a possible downgrade, of which Montana holds $80 million and Connecticut holds $300 million, records show.

``This just reinforces the fact that we have a serious issue,'' said State Senator Dave Lewis, of Helena, Montana, a member of the Legislative Audit Committee.
...
The Montana pool, managed by the Montana Board of Investments, has 25 percent, or $550 million, invested in SIVs, all of which carried top investment ratings when purchased.
...
Connecticut's Short-Term Investment Fund, which invests cash for state agencies and municipalities, is holding $300 million in debt issued by SIVs that may be downgraded by Moody's. The state's $5.8 billion fund held notes issued by SIVs affiliated with Citigroup ...

Connecticut also holds $100 million in defaulted SIV notes issued by Cheyne Finance.

More bagholders found.

by Calculated Risk on 12/03/2007 12:39:00 PM

From the Boston Fed: Subprime Outcomes: Risky Mortgages, Homeownership Experiences, and Foreclosures

... house price appreciation plays a dominant role in generating foreclosures. In fact, we attribute most of the dramatic rise in Massachusetts foreclosures during 2006 and 2007 to the decline in house prices that began in the summer of 2005.

Click on graph for larger image.

From the linked Fed paper, this figure compares the foreclosure rate in Massachusetts with changes in house prices. As prices rise, the foreclosure rate falls, since homeowners in trouble can either sell or refinance their homes. As prices fall, there is no way out - except foreclosure - for homeowners facing difficulties.

Last week I graphed the historical relationship for California: House Prices and Foreclosures. The pattern was the same.

Forget resets (although they are important). As prices fall over the next couple of years (or longer), foreclosures will rise, with or without resets. And a real concern is that it will become socially acceptable for underwater prime borrowers to just mail their keys into their lender (what Fleck calls "jingle mail").

by Calculated Risk on 12/03/2007 10:48:00 AM

Remarks by Secretary Paulson on Actions Taken and Actions Needed in U.S. Mortgage Markets at the Office of Thrift Supervision National Housing Forum

As we are all aware, the housing and mortgage markets are working through a period of turmoil, as are other credit markets, as risk is being reassessed and re-priced. We expect that this turbulence will take some time to work through, and we expect some penalty on our short-term economic growth. ...

And as I have said before, the housing market downturn is the biggest challenge to our economy. When home foreclosures spike, the damage is not limited only to those who lose their homes. Homes in foreclosure can pose costs for whole neighborhoods, as crime goes up and property values decline.

... foreclosure is expensive for all participants - lenders and investors – and this expense is an incentive to avoid foreclosure when a homeowner has the financial wherewithal to own a home. ...

And so, Treasury is aggressively pursuing a comprehensive plan to help as many able homeowners as possible keep their homes.

And on the modification plan:

... our plan involves a pragmatic response to the reality that the number of homeowners struggling with their resetting subprime mortgage will increase throughout 2008. As volume increases, we will need an aggressive, systematic approach to fast-track able borrowers into a refinance or mortgage modification. This third element does not, and will not, include spending taxpayer money on funding or subsidies for industry participants or homeowners.

While the reality is a bit more complex, in the interest of simplicity, there are four categories of subprime borrowers. There are those who can afford their adjusted interest rate; these homeowners need no assistance. There are also a substantial number of homeowners who haven't been making payments at the starter rate on their subprime loan and may not have the financial wherewithal to sustain home ownership; some of these homeowners will become renters again. A third category of homeowners might choose to refinance their mortgage - putting them in a sustainable mortgage while keeping investors whole. This is the first, best option. Servicers should move quickly to assist those who can refinance.

And the fourth category is those with steady incomes and relatively clean payment histories who could afford the lower introductory mortgage rate but cannot afford the higher adjusted rate. We are focusing on this group, determining who they are and what steps may appropriately assist them.
...
We are determined to ... develop a set of standards that will be implemented across the industry, from the largest mortgage servicers to the smaller specialty servicers. An industry-wide approach is critical to the effectiveness of this effort.

To speed up the modification process, Treasury is working through the HOPE NOW alliance with the American Securitization Forum to convene servicers and investors so they can develop categories of borrowers eligible for appropriate modifications and refinancings, and an industry-wide solution. This work takes time, as all parties seek to define categories of borrowers for streamlined refinance and modification where that is in the best interest of both the borrower and the mortgage investor. I am confident they will finalize these standards soon. And I expect all servicers will implement them quickly, and create benchmarks to measure their progress along the way. As a result, what was a fragmented, cumbersome process can be a coordinated effort which more quickly helps able homeowners.

Paulson clearly defined the group of borrowers that are being targeted for modifications: Homeowners with "steady incomes and relatively clean payment histories who could afford the lower introductory mortgage rate but cannot afford the higher adjusted rate".

Whenever the freeze ends, most of the homeowners in the defined group will still face foreclosure. So the purpose of this plan is clear - since the industry lacks the infrastructure to handle the work load, this guideline helps decide which loans to foreclose on now, and which loans to foreclose on later.

by Tanta on 12/03/2007 09:19:00 AM

From the San Diego Union-Tribune, a fabulous distillation of bubble-think in the story of Michael and Suzanne, who got Countrywide to modify their ARM.

Details: In around mid-September 2004, Michael and Suzanne borrowed $437,750 to buy a $440,000 condo. The $352,000 first mortgage was an interest-only 3/27 ARM with a start rate of 4.97%, a 3.00% first adjustment cap and 2.00% annual (1.00% every six months) periodic cap after that, with a maximum lifetime rate of 11.97%. It is presumably indexed to the 6-month LIBOR. The $85,750 second mortgage was a fixed rate (of unspecified term) at 8.00%.

The first scheduled adjustment on the first mortgage would have taken the monthly interest payment up by $880. Michael and Suzanne cannot, apparently, afford another $880. Nor is sale or refi a great option, since the value of the condo is apparently now $400,000. Michael and Suzanne did not have $40,000 for a down payment in 2004 and they still don't have $40,000 for a down payment.

They feel a touch let down by the world:

“We understood the situation with loan adjustments to be that after our first three years, our low rate would increase to the rate that everyone else is buying at right now,” said Suzanne, 38. “We didn't realize that we would see an increase of our monthly mortgage payments by several hundred dollars or that we'd now be facing this uphill interest rate climb that we're not going to be able to afford.”

That's an interesting way of thinking of an ARM: it won't hurt you because the rate will only go up to the rate buyers will buy at. This will make that rate adjustment affordable to you because nobody will ever buy in the future at a rate you cannot afford, even though your plan is that everyone will buy in the future at a higher price than you did.

A note to Countrywide: You get the borrowers you deserve in this business.

by Calculated Risk on 12/03/2007 12:47:00 AM

In Paul Krugman's dismissal of Ben Stein's NY Times piece, Krugman writes:

For what it’s worth, Goldman’s forecast of a 15 percent decline in home prices seems implausible to me, too — but on the low side. A 15 percent decline would bring prices back to their level in early 2005 — when the bubble was already well inflated. If prices fall back to their level in early 2003, that’s a 30 percent decline.

Click on graph for larger image.

This graph shows 15% and 30% nominal price declines for the S&P/Case-Shiller U.S. National Home Price Index and the OFHEO, Purchase Only, SA index.

A 15% nominal price decline would take prices back to late 2004 for both indices. A 30% price decline for Case-Shiller would take prices back to mid-2003; 30% for OFHEO would take prices back to late 2002.

If we look at price declines in real terms (inflation adjusted), and assume the price declines will occur over several years, a 15% price decline is almost guaranteed. The Case-Shiller index is already off 8% in real terms from the peak.

BTW, in a debate between Jan Hatzius and Ben Stein, the smart money will be on Hatzius. That said, here are Ben Stein's housing predictions from 2006 (along with a couple of guys that be would perfect for the Southwest "Want to get away?" ad campaign):

This show aired at the end of 2006. Note that LongIslandBubble.com overlaid the graphics and text on the video.

by Calculated Risk on 12/02/2007 07:06:00 PM

From the Financial Times: US property risks

Banks have significantly tightened their lending standards this year, and commercial real estate has felt the effects ... Commercial mortgage-backed security issues, which finance about half of deals and were a key driver of the recent market boom, dropped 84 per cent in October from a record high of $38.5bn in March. At one point, some loans actually exceeded property values. Now, typical loan-to-value ratios have retreated to about 70 per cent – when deals are completed at all.

... US banks could see $11bn to $78bn of commercial real estate losses if the lending crisis spreads, according to Goldman Sachs. ...

The commercial property sector is not likely to suffer the huge falls experienced by the worst-hit residential markets ... Supply is near its tightest point in decades.

That CRE investment would slow - and prices decline - was pretty obvious, but I'm not sure how severe the downturn will be. In the '01 recession, CRE investment was hit pretty hard (unlike residential investment), so there probably isn't the significant excess supply that exists for residential real estate.

Still, there were plenty of silly loans made in the CRE market. And there is a large amount of supply in the pipeline (to be completed in the next year). With the economy slowing, demand for office and commercial space will probably slow (or even decline). It appears the CRE slowdown is here, but how bad it will get is uncertain (more research required!).

by Calculated Risk on 12/01/2007 11:23:00 PM

Update: here is an article from Aftenposten in Norway (sent to me two weeks ago, hat tip Impy): Townships caught up in international credit crisis

Officials in four northern Norwegian townships (Narvik, Rana, Hemnes and Hattfjelldal) went along with an alleged recommendation by Terra Securities to invest a total of NOK 451 million in what they're now calling "high-risk structured products" offered by Citibank and sold for Citibank by Terra.

The American commercial paper was also tied to bonds issued by local governments in the US, and Norwegian Broadcasting (NRK) reported that hedge funds were involved. To boost returns, the Norwegian townships also borrowed NOK 3.5 billion to invest in Citibank's products, which later lost as much as 50 percent of their value because of the US credit crunch.

News started leaking out about the troubled investments when the townships were ordered to pay in millions more, to satisfy guarantee requirements. Mayor Asgeir Almås in Hattfjelldal feels cheated.

From the NY Times: U.S. Credit Crisis Adds to Gloom in Arctic Norway

What is keeping [Karen Margrethe Kuvaas] awake are the far-reaching ripple effects of the troubled housing market in sunny Florida, California and other parts of the United States.

Ms. Kuvaas is the mayor of Narvik, a remote seaport where the season’s perpetual gloom deepened even further in recent days after news that the town — along with three other Norwegian municipalities — had lost about $64 million, and potentially much more, in complex securities investments that went sour.

Tanta and I (and many others) have been wondering for a couple of years who the bagholders would be. Add Narvik, Norway to the list.

by Calculated Risk on 12/01/2007 02:35:00 PM

On Thursday, Brian provided a spreadsheet of the assets included in the E*Trade portfolio sale. He estimated the deal was for 27 cents on the dollar, and we were definitely surprised by the implied size of the haircuts for the prime first lien portion of the portfolio.

Here is more from Reuters: E*Trade firesale seen hurting Wall St portfolios (hat tip Alan)

UPDATE: I've received several emails pointing out that Bhatia is probably wrong.
Analysts are suggesting the sale valued the portfolio ranging from 11 cents to 27 cents on the dollar:

Citigroup investment bank analyst Prashant Bhatia said E*Trade actually received 11 cents on the dollar for its portfolio, if you factor in that the brokerage received $800 million in cash minus 85 million shares it issued. He said that implies Citadel's received stock compensation worth about $450 million, leaving E*Trade with only $350 million for its $3.1 billion portfolio.

And everyone was surprised by the price considering the assets in the portfolio:

Goldman Sachs analysts said they were surprised by the size of the discount on the E*Trade portfolio because 73 percent of the assets were backed by prime mortgages, or loans to people with solid credit.

It is worth emphasizing that a large portion of the assets were backed by prime - not subprime - mortgages. And many of the prime loans were first liens with decent average FICO scores (average 725) and LTV (71%).

by Tanta on 12/01/2007 11:30:00 AM

Another item sure to get some attention--or maybe not, since it kind of complicates the narrative of "predatory servicers." From the Wall Street Journal:

Law firms handling thousands of foreclosure cases on behalf of mortgage lenders and servicers are drawing criticism from judges, who say roughshod filing practices are trampling borrowers' rights.

Lawyers operating so-called foreclosure mills often are paid based on the volume of cases they complete. Banks and mortgage servicers often contract with such firms to handle foreclosures; the pay in Ohio, for example, is around $1,000 a case.

Um, is the pay based on volume, or is it just $1,000 a pop regardless of how many you do? My impression here is that it's the latter, and the problem is that this is a flat fee, not depending on how complex an individual suit is or--to the current point--how much time and effort might be needed just to assemble the documents and verify the liens in the land records to produce the original filing. It therefore becomes a matter of firms relying on volume because margins are skinny, and of treating every filing as a "no brainer" from the beginning. It's annoying when regular old newspapers don't get basic business practices. It's appalling when the WSJ doesn't.

Anyway, to continue:

The firms are typically small but may handle thousands of cases a year. Using computer software, they plug in variables such as a borrower's name, address and mortgage amount to generate a suit. Firms compete for business in part based on how quickly they can foreclose.

Um, no. They compete for business based on how quickly they can begin foreclosure proceedings. That's the problem here: a sloppy filing up front gets you onto the court's docket faster, but as we've seen, it tends to drag out the process and make foreclosures longer at the end of the day. And this description of the process ignores what's wrong with just "plugging in variables": we skipped the step of doing a search of the land records to verify that the last recorded assignment puts the foreclosing entity into current first-lien-holder status. And the step of going back to the servicer or trustee or whoever requested the foreclosure in the first place and reporting that an assignment needs to be recorded before the FC complaint can be filed.

"In general, most of the firms that practice this kind of law do a very good job," said Peter Mehler, a Cleveland-area lawyer who handles foreclosures on behalf of mortgage servicers. But in the "gold rush" to get a piece of the growing business, some firms "have cut corners."

Lately, judges are faulting law firms for what has become a common practice: filing a foreclosure suit, in states that require them, without showing proof that the plaintiff actually holds the mortgage and has the right to foreclose. (Such plaintiffs are often banks that act as trustees for investors of securities backed by mortgages.) The situation occurs in part because mortgage documents and the contracts between borrowers and lenders may change hands multiple times and may not be assigned to the plaintiffs at the time the suits are filed.

What this has really got to do with loans changing hands "multiple times" isn't very clear. If a loan changed hands exactly once, and no assignment was recorded in the land records exactly once, you'd have exactly the same problem. Of course the odds of having a missing assignment or a gap in the assignment chain go up when there are multiple assignments. But in that case, the problem is often not that the plaintiff--the last party in the chain--doesn't have an assignment. It's that the party who assigned to the plaintiff didn't have an assignment from the party who assigned to it, or something like that.

Why be so obsessed with the details here? Because way too many people have taken that unfortunate phrasing of the problem to mean that securities are purchasing delinquent loans just for the purpose of foreclosing. The WSJ, intentionally or not, falls into this kind of language:

This month, a state judge in Cincinnati dismissed a foreclosure lawsuit brought by Wells Fargo Bank because the bank filed the suit before it had acquired the mortgage. In dismissing the case, the judge sent a warning letter to the bank's law firm, John D. Clunk Co. LPA, in Hudson, Ohio. Judge Steven E. Martin wrote that it was "troubling" that the plaintiff "and its counsel filed the lawsuit with no basis whatsoever" and that firm must not do so again.

The law firm didn't respond to requests for comment. Wells Fargo declined to comment.

"Before it had aquired the mortgage" makes people think that Wells filed to foreclose a loan before it ever owned that loan, as if Wells saw, say, a bad loan at Podunk National and decided to buy it just for the pleasure of foreclosing it, but somehow managed to file first and buy later.

There is exactly zero reason to believe that this is what happened. Wells "acquired" the loan (or some security acquired the loan and Wells became the master servicer or trustee or something) back when the loan was fresh and new. What someone failed to do was to record the evidence of transfer of the beneficial interest in the collateral (known as an "assignment of mortgage") in the land records before the day the FC was filed.

It is quite common practice in the industry, as I have explained before, to execute assignments in "recordable form" when a loan is sold, and for the buyer or the buyer's custodian to take physical custody of that assignment, but to refrain from actually sending it to the county recorder of deeds for recordation in the land records unless and until it becomes necessary to foreclose. I know of no judicial opinion yet that has ever implied that the failure to record a document voids the loan sale; in fact, Judge Kathleen O'Malley's Order of November 14,* one of the several dismissals for inadequate documentation (along with Boyko's and Rose's) making the rounds, explicitly states that

The Court is only concerned with the date on which the documents were executed, not the dates on which they were recorded (if recorded) with the county recorder’s office.

The trouble with valid, executed, but unrecorded assignments is that even if a foreclosure attorney ran a records search before filing, in order to verify current lienholder, the assignment would not appear in the land records. It really is incumbent on the trustee or servicer to provide the original assignment for recordation, since the trustee or servicer is the one who has custody of it (or can get it from the custodian) and therefore the only one who can reliably vouch for its existence.

There are exellent reasons to record that old assignment first, then file your FC complaint. But as far as I can tell, judges aren't even asking for recorded assignments; they're just asking for valid assignments. What seems to have happened in at least one case--the Deutsche Bank case that Boyko went ballistic over--was that plaintiff's attorney, not having the real original assignments handy, simply executed new ones, after the fact. That's pretty amazing practice for an officer of the court, and His Honor reacted exactly the way one ought to. But it does not mean that the original assignments do not exist. Absence of evidence is not evidence of absence. Forging a new assignment because you can do that in twenty minutes, while just breaking down and requesting the originals from the custodian might take several days, is bad lawyering. It is not evidence that anyone is buying deliquent loans in order to foreclose them.

What reputable banks like Wells Fargo are learning here, I think, is a painful lesson in reputation risk. Wells hired some cheap corner-cutting law firm to handle its foreclosures (as did Deutsche Bank), and as a result, its name is now all over the press in association with practices that can be made to sound exceptionally sinister. Remember Boyko's "priceless" comment? Well, I'm here to suggest that Wells Fargo's good name is worth a whole lot more to it than $1,000. Legally, plaintiff is responsible for the actions of plaintiff's counsel.

Here, by the way, is the relevant part of Judge Thomas Rose's order** involving a number of foreclosure filings by several different trustees:

To date, twenty-six (26) of the twenty-seven (27) foreclosure actions based upon diversity jurisdiction pending before this Court were filed by the same attorney. One of the twenty-six (26) foreclosure actions was filed in compliance with General Order 07-03. The remainder were not.2 Also, many of these foreclosure complaints are notated on the docket to indicate that they are not in compliance. Finally, the attorney who has filed the twenty-six (26) foreclosure complaints has informed the Court on the record that he knows and can comply with the filing requirements found in General Order 07-03.

Therefore, since the attorney who has filed twenty-six (26) of the twenty-seven (27) foreclosure actions based upon diversity jurisdiction that are currently before this Court is well aware of the requirements of General Order 07-03 and can comply with the General Order’s filing requirements, failure in the future by this attorney to comply with the filing requirements of General Order 07-03 may only be considered to be willful. Also, due to the extensive discussions and argument that has taken place, failure to comply with the requirements of the General Order beyond the filing requirements by this attorney may also be considered to be willful.

A willful failure to comply with General Order 07-03 in the future by the attorney who filed the twenty-six foreclosure actions now pending may result in immediate dismissal of the foreclosure action. Further, the attorney who filed the twenty-seventh foreclosure action is hereby put on notice that failure to comply with General Order 07-03 in the future may result in immediate dismissal of the foreclosure action.

My boldface, there: it seems clear to me that this is an admission that the assignments really do exist, and can, in fact, really be produced. But what, we ask, has relying on this attorney done for the reputation of the lienholders? The story wasn't reported as "some worthless lawyer screwed up"; it was reported as Deutsche Bank and Wells Fargo and HSBC et al. screwed up. If you don't want your name in the headlines like that, hire a better lawyer. And pay for it. Oh, wait . . .

Allow me to close by observing that Curly and Larry (if not Moe) have lent some weight to a proposal that would basically mean servicers shoving through across-the-board modifications to "freeze" interest rates. I'm not here to argue the wisdom of rate freezes in this post. I am here to point out that a modification of mortgage is a legal document that has to be recorded in the land records in which the original mortgage was filed. If the modification is being executed by a servicer or trustee on behalf of the noteholder, then any intervening assignments up to the one to the modifying party need to be recorded first, so that the recordation of the modification is valid. Also, modification agreements are complicated documents; you want to be very careful with their wording, so that you are sure you are modifying only certain specified terms of the original mortgage and note. More than a few sloppy servicers have been haunted by a bad modification agreement that inadvertently waived rights or terms that servicer needed to keep.

So it really just sounds like a fantastic idea to push through a major effort to execute modifications really fast and cheaply, doesn't it? Frankly, the whole idea gives me goosebumps.

-------------------

*UNITED STATES DISTRICT COURT, NORTHERN DISTRICT OF OHIO, EASTERN DIVISION, In Re Foreclosure Actions 1:07cv1007 et al., November 14, 2007. No, I didn't go to law school and learn how to cite court orders in proper format. So sue me if you can find a decent lawyer.

**UNITED STATES DISTRICT COURT, SOUTHERN DISTRICT OF OHIO, WESTERN DIVISION AT DAYTON, IN RE FORECLOSURE CASES 3:07CV043 et al., November 15, 2007.