by Calculated Risk on 10/21/2007 11:45:00 AM

From MarketWatch: More firms expected to join SIV fund: officials (hat tip Bob)

More financial firms are expected to join the "Super SIV" special fund to help guarantee liquidity in the commercial paper market, officials said Saturday.

"Participation is expected to broaden in the weeks ahead," said Robert Steel, the U.S. Treasury undersecretary ...

Bank of Italy Governor Mario Draghi told reporters on Friday that Treasury Secretary Henry Paulson had informed his G7 colleagues that Pacific Investment Management Co., the world's largest bond fund, and Fidelity Investment, the Boston-based mutual fund giant, have decided to join the SIV fund.

But a spokesman for PIMCO said Draghi was incorrect.

"PIMCO is not participating," PIMCO spokesman Mark Porterfield said in an email on Saturday.

PIMCO is smart to stay away from this mess.

by Calculated Risk on 10/20/2007 03:42:00 PM

by Calculated Risk on 10/20/2007 02:40:00 PM

From the WSJ: SIVs Pose Risks for Money-Market Funds

Complex investments known as SIVs are roiling Wall Street and the world of high finance. But the investment vehicles also are threatening trouble in a seemingly unlikely place: money-market funds, the choice for many individual investors seeking safety.

In recent years, the short-term debt issued by such structured investment vehicles, or SIVs, had become a favorite for many money-market funds, thanks to their attractive yields, high credit ratings and added diversification.

As a result, many money-market mutual funds were holding 10% to 20% of their portfolios in debt issued by SIVs. Funds overseen by Bank of America Corp.'s Columbia Management Group, Credit Suisse Group's Credit Suisse Asset Management, and Federated Investors Inc. recently held big stakes in SIVs, including some of the most troubled names.
...
Most important for money-fund investors, fund companies would almost certainly take steps to prevent losses from reaching shareholders -- such as absorbing the losses themselves by purchasing the money-losing securities from the fund at their full price.

SIVs really aren't that complicated. They borrow short (via commercial paper less than 9 months duration so the don't have to file with the SEC) and lend long. Money market funds buy the commercial paper with deposits from their customers. Here is a good description:

... bankers hatched the idea of setting up a fund that would issue short-term commercial paper and medium-term notes to investors, then use the money to buy higher-yielding assets, typically longer-term ones. The bank would profit by collecting fees for operating the fund. The fund's assets would belong to its investors, so they would stay off the bank's balance sheet. SIVs had an advantage over conduits, a similar structure that was already gaining popularity: They didn't require banks to cover fully the fund's debts if the commercial-paper market dried up.

The funds can be off balance sheet because - at least theoretically - the investors (like the money market funds) will take the losses, not the banks. What is complicated (really opaque to the investors) is the quality of the SIVs investments.

Usually the main concern with borrowing short and lending long is interest rate risk. In this case, the problem is more credit risk with poor performing longer term investments.

by Calculated Risk on 10/19/2007 09:31:00 PM

From Reuters: Pimco, Fidelity to join SIV rescue fund - Draghi

Investment fund giants PIMCO and Fidelity have joined the so-called super SIV fund set up by three big U.S. banks, boosting confidence in the plan, Bank of Italy Governor Mario Draghi said on Friday.

Draghi said U.S. Treasury Secretary Henry Paulson had discussed the fund with officials attending the G7 meeting of central bankers and finance ministers.

"Paulson has done a short briefing on the SIV fund," Draghi told journalists at the close of the G7 meeting. "PIMCO and Fidelity have joined."

From the WSJ: Banks May Pony Up $60 Billion for SIVs

Banks and other financial firms have expressed interest in putting up more than $60 billion toward a super-size investment fund...

If the expressions of interest turn into firm commitments in the next few weeks or months, the three U.S. banks organizing the fund would come close to their goal of raising a fund of $80 billion to $100 billion.

The banks also are targeting several big institutions in Europe, such as HSBC Holdings PLC in London and Dresdner Bank AG in Germany. Both rank among the largest managers of the kind of structured investment vehicles, or SIVs, that the fund is intended to support. HSBC and Dresdner declined to comment.

by Calculated Risk on 10/19/2007 04:24:00 PM

More downgrades (on what I think will be busy afternoon and weekend):

S&P Lowers Ratings on 1,413 U.S. RMBS Classes Backed by Subprime Mortgage Loans from the 4Q 2005 - 4Q 2006 (hat tip rs)

Standard & Poor's Ratings Services announced today that it has downgraded 1,413 of U.S. residential mortgage-backed securities (RMBS) backed by first-lien subprime mortgage loans that were issued from the beginning of the fourth quarter of 2005 through the fourth quarter of 2006. These downgraded securities had an original par value of $22.02 billion, which represents 4% of the $554.4 billion of U.S. RMBS backed by first-lien subprime mortgage loans rated by S&P during this period. These actions, combined with downgrades previously announced by S&P, impact a total of 1,671 securities of U.S. RMBS backed by first-lien subprime mortgage loans issued during this period, representing $24.8 billion, or 4.5% of the $554.4 billion mentioned above. S&P also affirmed its ratings on securities representing $531.6 billion original par value of U.S. RMBS backed by first- lien subprime mortgage loans from this same period.

Of the 1,413 securities downgraded today, approximately 47% were rated in the 'BBB' category and below. Fifteen 'AAA' rated securities were downgraded, accounting for roughly 0.01% of all downgraded securities and 1.1% of the total dollar amount downgraded. No 'AAA' rating was lowered below 'AA'.

We took these rating actions at this time because, based on the most recent data, we expect further delinquencies and losses on the underlying mortgage loans; the consequent reduction of credit support from current and projected losses; and continued declines in home values.

While cumulative losses to date remain low, they have increased since our July 2007 review and we expect them to increase further. Based on the most recent data from September 2007, cumulative losses for the period have increased from 29 bps to 69 bps -- a 138% increase since our July 2007 review.

The September 2007 data shows increasing levels of overall delinquencies and serious delinquencies. Seriously delinquent loans include loans that are either: delinquent by more than 90 days, in foreclosure, or for which the real estate is possessed by the servicer. For all U.S. RMBS backed by first-lien subprime mortgage loans issued during this period, overall delinquencies averaged 21.43%, and serious delinquencies averaged 14.17%. This is in contrast with the downgraded transactions, for which overall delinquencies averaged 15.73% and serious delinquencies averaged
23.33%.

We expect that the downgraded securities will be particularly vulnerable to increased losses because, on average, 60%-70% of the loans backing them are subject to some type of payment adjustment in the near future. Most of these are 2/1 adjustable-rate mortgages already in their adjustable-rate stage and already past their first, and typically largest payment reset. Despite some industry claims of increased accommodations to subprime borrowers, we expect losses to increase for borrowers who have experienced (1) rising loan payments due to resetting terms of their adjustable-rate loans, and (2) principal amortization that occurs after the
interest-only period ends for adjustable- and fixed-rate loans.

Standard & Poor's expects that the U.S. housing market will continue to experience price decreases. We project that property values will decline 11% on average from peak to trough and will begin to recover in late 2008, with the peak having occurred in the spring of 2006. This continued decline in home prices will apply additional stress to these securities.

As part of this review, we assumed losses for defaulted loans that closed during the second half of 2005 at a level of 40%, and for those that closed during 2006 at a level of 45%. During our July 2007 review we assumed losses for defaulted loans that closed in 2006 of 40%. We have now increased this assumption based on the most recent data and projected declines in home values.

by Calculated Risk on 10/19/2007 04:04:00 PM

From Bloomberg: Cheyne, IKB SIVs Default on Commercial Paper as Assets Fall (hat tip FFDIC)

Cheyne Finance Plc and IKB Deutsche Industriebank AG's Rhinebridge Plc, two structured investment vehicles that bought securities backed by home loans, defaulted on more than $7 billion of debt as the value of their holdings fell.
...
``The fallout from the credit crisis is far from over,'' said Jim Reid, head of fundamental credit strategy at Deutsche Bank AG in London. ``There are probably more skeletons in the closet. The problem is knowing when and where they are going to emerge.''

Rumors were flying during the last hour of trading of problems at Merrill and Bear. The Merrill rumor was of a special board meeting this weekend with possible additional writedowns - we will see.

by Calculated Risk on 10/19/2007 04:02:00 PM

From Fitch: Affirms $1.27B & Downgrades $265.5MM from 4 IndyMac Subprime Transactions (hat tip sr)

Fitch Ratings has taken the following rating actions on IndyMac Banks INABS certificates. Affirmations total $1.27 billion and downgrades total $265.5 million.

(see press release for details)

The rating actions are based on changes that Fitch has made to its subprime loss forecasting assumptions. The updated assumptions better capture the deteriorating performance of pools from 2006 and late 2005 with regard to continued poor loan performance and home price weakness.

by Calculated Risk on 10/19/2007 11:26:00 AM

Wachovia conference call comments (from Brian):

"Much of the increase in non-performing loans and the losses are on loans in certain California markets that have experienced fairly steep declines in prices. Our delinquency call centers report that the primary reasons for borrowers struggling to pay are three fold. First is reduction of income or underemployment. Second is the assumption of additional debt from lenders other than Wachovia and thereby changing the credit profile from the origination of the loan. And third unemployment. We have seen some uptick in unemployment in some of these markets. Let me also point out that while the average current estimate at the appraised value of non-performing loans is 77%, there is $380 million in balances out of the total $1.7 billion where the current estimate of value is over 90%. Actually on that pool, averages in the high 90s, again reflecting the dramatic decline of house prices in certain markets. These particular loans have a low loan to value of just under 80% at origination. It's interesting to note here that problems in these markets, really for all lenders seem to be across the board without originating FICO, the type of loan or the property. Given our outlook for continued weakness in the housing market and possibility for slow income consumer sector, we anticipate loans on consumer mortgage book continue to increase over the next few quarters and that losses will be up albeit fairly modest charge operates. To manage the increase in loans in foreclosure, we have significantly increased our staff responsible for handling Oreo properties and working with delinquent borrowers. Prepare the property to sell and sometimes choosing to maybe take a somewhat higher loss on that sale rather than risk holding out for a top dollar opportunity that may or may not come down the road.” emphasis added

Note that Wachovia is seeing rising unemployment - already - as a factor in delinquencies. But the most important comment is that the problem loans are: "across the board without originating FICO, the type of loan or the property".

Across the board!

On Commercial RE portfolio:

“The commercial real estate portfolio continues to perform very well overall. Loan secured with income producing property continues to enjoy solid underlying fundamentals with favorable vacancy rates and cash flows. As we have noted, probably on the last couple of calls, the portion of the commercial real estate portfolio connected to housing is experiencing an upward trend in criticized assets higher charge-offs and non-performs. While these loans have generally performed well overall and the charge-offs and total real estate financial services portfolio, we're only $3 million in the quarter, we do anticipate further softening and have recently undertaken a thorough review of commercial real estate loans. Nearly all of these loans are on a with-recourse basis. We are and will be moving aggressively to work with borrowers to shore up as best we can. We believe credit costs will be rising, we believe the deterioration will be manageable”

CRE sounds OK so far.

Here’s what they had to say about the marks in their CDO portfolio:

“Next line addresses other structured products [Total of $438MM]. Here we have the marks on warehouse positions and trading inventory, both of which we hold in trading portfolios. This includes the positioning Ken referred to in reference to sub prime mortgage exposure and AAA rated securities. $308 million is associated with sub prime securities [Their slides say $347 of the mark was related to subprime of which $308 was AAA subprime]. Basically there, we never would have expected that you see AAA securities trade so far so quickly from par.”emphasis added

They were later asked if they were happy with their risk management:

“And you know, we'll change the way we do some things. I would say that as we look at results, I think the biggest disappointment for me is that of those $1.3 billion end marks, we had about $300 million, in losses on AAA sub prime paper in trading desks and inventory. And the thing that disappoints me about that, we avoided it in our origination efforts and avoided it in, for the most part in our securitization efforts. So frankly, I think we had a little bit of a break down in having AAA sub prime in some of our portfolios we took losses on. I think it is amazing that we could take $300 million of losses on AAA paper. We didn't expect that that paper could degenerate that fast, with that kind of swiftness”emphasis added

Comments about the general credit environment:

Analyst:

As we talked to companies, September, I think you mentioned this too, was a particularly weak month for credit and is that trend, as you see it in October, about the same? Would you say it's slowed or accelerated?

Wachovia:

"Still kind of early in the month, but I would say that the trends we saw late August and September, you know, halfway through this month are about the same. I wouldn't say they've accelerated, but they haven't backed off either."

by Calculated Risk on 10/19/2007 10:52:00 AM

Yeah, another house price index, but this one has a cool map.

From First American: LoanPerformance Releases August House Price Index

Click on graph for larger image.

SAN FRANCISCO, Oct. 18, 2007–First American LoanPerformance ... today announced the release of its August 2007 LoanPerformance Home Price Index (HPI).

The LoanPerformance HPI provides a comprehensive set of monthly home price indices and median sales prices covering 7,376 ZIP codes, 956 Core Based Statistical Areas (CBSA) and 655 counties in all 50 states and the District of Columbia....

"This latest home price index confirms that property values in key mortgage markets like California, Nevada, Florida, and Arizona continue to exhibit on-going declines,” said Damien Weldon, vice president of collateral and prepayment analytics for First American LoanPerformance.

“Within these States, cities like Los Angeles, Las Vegas, Miami and Phoenix are leading the market downwards. At the ZIP code level, the picture is often much bleaker because there are individual ZIP codes that are down nearly 20 percent compared to last year,” added Weldon.

I doubt any states have really seen price increases over the last 12 months.

by Anonymous on 10/19/2007 09:30:00 AM

Given the questions in the comments to yesterday's post on seller-funded down payment assistance (DAP), I thought I'd offer a very simplified example of what the issue is. Yes, this is simplified; FHA loan calculations are pretty complex, even though they aren't as complex these days as they used to be.

Currently, FHA requires a minimum cash investment from borrowers equal to 3.00% of the contract sales price. The effective LTV can still exceed 97% even with a 3.00% investment, because borrowers can finance a portion of allowable closing costs, including their up-front mortgage insurance premium (UFMIP), in the loan amount. (FHA borrowers with a base LTV of more than 90% also pay an additional mortgage insurance premium in the monthly payment of 0.50% annually.) The current UFMIP with 3.00% down is 1.50% of the loan amount.

The administration's proposed zero-down program would have UFMIP of at least 2.25% and a monthly premium of at least 0.55%.

FHA does allow the borrower's down payment to come from gift funds provided by relatives, employers, governmental agencies or true charitable organizations. The point here is that 1) these are supposed to be true gifts with no expectation of repayment, not disguised loans, and 2) they may come only from parties who do not have an interest in the transaction.

Property sellers may contribute up to 6.00% of the sales price to an FHA borrower without affecting the appraised value of the property, but this contribution may be applied only to closing costs and points, repairs, etc., not to the minimum investment (i.e., the down payment). If the seller contributions exceed 6.00%, the excess amount is subtracted from the sales price of the property (as a "sales inducement"), which lowers the maximum loan amount accordingly. HUD has never allowed property sellers to directly provide funds for the minimum 3.00% down payment.

The seller-funded DAP programs get around this problem by having the property seller contribute the down payment funds to a "nonprofit" company which then "gifts" the funds to the borrower. Sellers are generally charged a fee of at least $400 for "processing" their contributions. Every reputable study (non-industry-sponsored) of the resulting loans (like this one) shows that 1) the sales prices of the properties are inflated by the amount of the "assistance" and that 2) the loans default at least twice as often as those with bona-fide gifts from a disinterested party. Even worse, because they are processed with the standard UFMIP charged to loans with a 3.00% down payment, this additional risk is not offset by a higher premium.

Here's how it works. First, here's a "typical" FHA loan with a 3.00% down payment (we'll assume that the seller pays closing costs other than UFMIP in cash, just to keep things simple):

• Original list price: $100,000
• Contract sales price: $100,000
• Appraised value: $100,000
• Required borrower down payment: $3,000
• Base Loan Amount: $97,000
• UFMIP: 1.50% or $1,455
• Total loan amount: $98,455
• Effective LTV (based on original list price): 98%

Here's how it works with a typical seller-funded down payment:

• Original list price: $100,000
• Contract sales price: $103,505 (list price plus $400 processing fee, divided by 0.97)
• Appraised value: $103,505 (or any amount above that, as LTV is calculated on the lower of appraised value or contract sales price)
• Required borrower down payment: $3,105 (provided by the seller via the DAP)
• Base loan amount: $100,400
• UFMIP: 1.50% or $1,506
• Total loan amount: $101,906
• Effective LTV (based on original list price): 102%

If the DAP loan were treated as the same risk as the proposed zero down program, you would get UFMIP of 2.25% or $2,259, resulting in a total loan amount of $102,659 and effective LTV of 103%. That would actually produce a higher loan amount than a true zero down program would, because of that $400 "processing fee" to the "nonprofit" (zero down base loan amount of $100,000, UFMIP of $2,250, total loan amount of $102,250, or $409 less than the "assistance" loan).

What happens if the appraiser refuses to play along with this scheme? Well, that would create a problem: the maximum loan amount is calculated on the lesser of the sales price or appraised value, and so the borrower could not borrow enough to pay the inflated sales price if it were greater than the appraised value.

What if the seller simply reduced the contract price by $3,505 (the cost of assistance plus processing fee)?

• Original list price: $100,000
• Contract sales price: $96,495
• Appraised value: $100,000
• Required borrower down payment: $2,895
• Base Loan Amount: $93,600
• UFMIP: 1.50% or $1,404
• Total loan amount: $95,004
• Effective LTV (based on original list price): 95%

The problem with that last scenario, of course, is that the borrower still has to come up with a down payment. The whole purpose of seller-funded DAPs is to get borrowers with no funds into loans, not merely to facilitate legitimate seller concessions.

Does it really matter whether gift funds come from an interested party? Yes, it does. A party without an interest in the transaction has no incentive to induce or persuade the borrower to pay more than the fair market value of the property; in fact, a distinterested party has an incentive to assure otherwise, since the lower the appraised value and contract sales price, the less the third party has to contribute. Government agencies and true nonprofits who provide such assistance are known for being mean and skeptical reviewers of appraisals and sales contracts, you see. (They also generally have income limits and other rules designed to keep speculators and other non-needy folks out of their programs.) Seller-funded DAPs avoid all that "red tape" and "excessive processing time."

I personally have never had any enthusiasm for the proposed zero down FHA program. But even it is better than the DAP scam. Those who claim that DAP loans provide a benefit to borrowers without funds are making no sense even if you grant that making loans to people without even minimal skin in the game is a good idea: the DAP programs simply keep contract sales prices inflated, channel fees into the pockets of "nonprofits" who provide no other service than laundering money, and result in lower insurance premiums than FHA should be getting for loans with riskier profiles. If you care at all about the long-term survival of the FHA program, you would be doing everything you can to protect it from this kind of damage.

By their own logic, the Congressional defenders of DAPs should be pursuing the zero down program, and/or funding for true nonprofits and local governments who provide forms of down payment assistance that don't inflate sales prices and that offer real, useful homebuyer counseling services. One of the arguments for DAP is that it is available for borrowers who aren't lucky enough to have family, an employer, or a local agency or true nonprofit who can provide gift funds. That's right: if you aren't lucky enough to receive a true gift that enables you to buy a market-priced property, you can be thrown to a bunch of sharks who will provide you with a "gift" with a hidden price tag. This is a good thing, since owning an overpriced home and making the higher payments is, I guess, a major blessing.

Supporting DAPs means supporting property sellers--particularly but not limited to builders and developers--and the "entrepreneurs" who form "nonprofits" to extract fees from naive homebuyers, not to mention loan originators who pocket higher commissions, with the risk being carried by government insurance. It is, precisely, the kind of sleazy, conflict-ridden, self-serving "initiative," overtly "faith-based" or its sort-of secular equivalent "dream-based," that thrives in an environment where regulation is dismantled or unenforced and "government" is bashed with one hand and milked with the other. It is an "innovation" just like plainer, older-fashioned forms of money-laundering are "innovations." It takes a profound ideological blindness to march behind the DAP banner in the name of "helping first time homebuyers."