CR,
I love your blog (which also includes the sound contributions of many regulars). Many thanks for the free education.
If your hypothesis is true, then we should see a proportional drop in recorded sales in three months, since the purchase index is coming down to previous levels.
We are in the midst of a unique Credit cycle. The ability for originators to sell loans and immediately book profits; for investment bankers to buy, securitize and immediately book profits; and for leveraged speculators to acquire various securitizations and other derivatives and book easy profits from various spreads and mark-to-model - have all made this Credit boom unlike any other. ..
Today, Subprime mortgage originator profits are collapsing lending volumes are sinking; gain on sale is reversing to loss; Credit losses (especially from returned early defaults) are surging; and the liquidity necessary to operate is disappearing overnight. This has initiated the ugly downside of operating as a Ponzi Finance Unit. The issue of early payment defaults where investment bankers/securitization pool operators return problem mortgages in droves back to the originator is rapidly bankrupting this thinly-capitalized industry. And the more acute the risk of insolvency, the greater the incentive for investment banks to rush to dump problem loans while the originator still retains some liquidity (think bank run).
This has enormous ramifications for the industry. Subprime Credit conditions are in the process of tightening how tight only time will tell. Subprime borrowers this year facing payment resets will confront a changed industry. Desperate borrowers stretched the truth in 2006 to get new mortgages approved. The weakest originators are disappearing
But like everything else associated with this most extraordinary Credit Cycle, the analysis is infinitely more complex than what meets the eye. We are undoubtedly in the midst of a major liquidity event for the subprime originators. Additionally, the riskiest CDO and securitization tranches (and related derivatives) will suffer heavy losses. This is a decisive Credit event for the subprime industry Thus far, however, there is little indication that the (paramount) market for prime mortgages is being impacted much at all. And when it comes to the subject of overall system liquidity, the current level of tumult could prove less than significant.
In a different or perhaps typical - environment, recent subprime developments would prompt a nervous reaction from the markets. But today we operate in a most extraordinary backdrop of rampant global liquidity excess, evidenced by ongoing international booms in M&A, securities markets, real estate, energy, and most asset markets. Here at home, continued strong growth in (non-subprime) real estate lending combines with robust corporate and government Credit growth to ensure - for now - continued sufficient non-financial debt growth. More ambiguous but equally important, indications point to continued robust expansion in securities finance, at this stage of the Credit Cycle a liquidity-creating be
How long does it take to be rejected by a lender? Wouldn't somebody denied multiple times show up as multiple applications? A credit tightening could increase the number of people who applied multiple times for a loan based on the same property, even if NO lenders went Tango Uniform.
So MBA covers 50% of the mortgage market, the respectable side. And when the other side caves in, those in the sub-prime arena are forced to cough up the extra nickel or delay that mortgage application until they can afford the prime rates. Since New or Existing home sales didn't reflect this increase as CR notes, it's hard to think of any other explanation.
Was there a bulge in refis? no.
Was there an increase in the cheapest variety (ARMs) of mortgage application? probably.
Would CR provide his thoughts/analytics on the possibility that the MBA applications have increased as the result of borrowers needing to submit multiple applications?
"DJ C-BASS Named As Lead Bidder In Auction Of Ownit Mortgages
2007-02-21 12:37 (New York)
New York mortgage lender C-BASS LLC has bid $4 million for 54 mortgages being
auctioned off by bankrupt subprime lender Ownit Mortgage Solutions Inc.
Ownit Mortgage and the committee representing its unsecured creditors named
C-BASS as the stalking horse, or lead, bidder in the sale. Mortgage insurers
MGIC Investment Corp. (MTG) and Radian Group Inc. (RDN) are the majority owners
of C-BASS.
Judge Kathleen Thompson of the U.S. Bankruptcy Court in Woodland Hills,
Calif., has scheduled a hearing for Thursday to consider approval of the sale."
We don't know what the average mortgage size is, but if its $200K, their opening bid is less than 40 cents on the dollar (and they get the servicing rights in the deal). We'll see what happens at the auction, but 60% is quite a haircut.
I can't begin to assess all the factors that might affect the MBA stat, but I do want to mention that first-time home buyers are a particular subject of tightening, as they have unique risk factors. Therefore, I would expect there to be a detectable increase in the average ratio of apps to approvals for the poorer credit quality FTHBs in the next two months and this has clearly been going on for a couple of months.
I have read multiple threads by brokers bitching about the pile of apps on their desks that they can't get approved. So....
He mentioned something to the effect that he picked it up from posted mortgage brokers' comments as he was looking for info for his ML implode o meter site.
So probably anecdotal but consistent with other info.
At the end of 2006 there appeared to be a surge in purchase applications. This might have been due to increased activity, or possibly favorable weather.
Goddammit CR. That's just too funny. You owe me another keyboard.
Might I suggest other factors? Tightening credit. With closer scrutiny would not applicants either prophylactically or reactively make multiple applications?
Second musing, much like the first; With 24 name brand mortgage brokerages down in the last 9 weeks wouldn't every one of the loans applied for but not funded be counted a second time in their new morgage brokerage house?
I guess the problem I have imputing this to the "sub-prime bust" as opposed to the weather is that you have to conclude that the sub-prime market has been picking up in recent weeks.
I think both things are a factor, but weather moreso. As you get further up north it just makes a huge difference having 1 month of winter instead of 3.
Brian - short answer - crooks. It seems to be concentrated around the metro area, which of course had favorable conditions for such activity.
Judging from my personal experience, I think we are catching more of them RIGHT NOW than other states. I think there is more fraud in FL than in GA overall.
I notice that whenever incomes stagnate and foreclosures rise a bunch of these schemes come to light - there's been a lot of reports of fraud in the CO foreclosure epicenter, for example. My guess is that it is not detected as readily if the market is good.
I've been reading these comment threads for several weeks, and haven't heard anyone complain about the rating agencies.
At the end of the day, it is these guys that look at what these originators are doing and slap credit ratings on the senior tranches of RMBS, which brings in 90 percent of the funding for the deal.
How could all of this fraud (cahs back at closing, appraisal fraud, stated-income) have gone on undetected by the rating agencies? At the end of the day, isn't that their job? Investors certainly deserve blame for not looking, but shouldn't these guys deserve a big part of the blame for not nipping this in the bud a long time ago?
IMHO this may be reasonable since this is a thin tranche not a bond. If losses are bad it is possible for the BBB's to have hardly any return before losses wipe it out.
IMO would take a lot of time, good data, and a good model to get even a reasonable value for these things.
Sadly many buyers probably failed to do this analysis.
Someone please correct me if I've missed something in this analysis.
IMHO this may be reasonable since this is a thin tranche not a bond. If losses are bad it is possible for the BBB's to have hardly any return before losses wipe it out.
Minor disagreement.
My understanding is that the BBB ABX is a portfolio of mezzanine ABS tranches. Each tranche has a 5 percent weight. The investor has the same leverage to the asset class as buying one mezz tranche, but gets a diversification benefit through a portfolio of issuers.
The BBB refers to the quality of collateral in the index. When the index itself is tranched, then the investor in the BBB-tranche of the index would have even more leverage wrt the asset class.
Just curious: how do they actually do the calculation that since the ABX-HE-BBB- 06-2 is now at 75.81 that it takes $1.1Million to insure $10million?
Also, how do the starting prices for these instruments (usually just above 100) get set? I'm a newbie and I look at a lot of these ABX graphs and think that they're 'kinda-sorta' like bonds and thus they shouldn't fluctuate much, but when you see those graphs tanking, well, I can tell that something is up, but I'm not quite sure how to explain it to people.
how do they actually do the calculation that since the ABX-HE-BBB- 06-2 is now at 75.81 that it takes $1.1Million to insure $10million?
Excellent question -- I'd like to know too.
And here's another one: what exactly does the paragraph below mean? Are hedgefunds using unregulated derivative contracts to manipulate outcomes in other markets?
Stock investors betting on a decline in the shares of companies that make subprime loans have used derivatives contracts tied to ABX indexes as an alternative, said Howard Hill, a managing director in Springfield, Massachusetts at Babson Capital Management LLC. The firm is a unit of MassMutual Financial Group, which manages about $30 billion of mortgage- and asset-backed bonds.
While the pricing on the BBB tranche is very dramatic, take a look at the pricing on the A tranche of the two most recent series - it's down to 95...while the ABX trades wide of the cash bonds, I suspect these levels make it hard to get any securitizations done....the higher up the rating scale you go, the bigger the amounts involved and the bigger the hit to the economics for the issuer.
Another leg of the Novastar stool starts to wobble.
"NEW YORK, Feb 21 (Reuters) - Moody's Investors Service on Wednesday said it may cut its servicer quality ratings on NovaStar Mortgage, Inc. due to the mortgage servicer's exposure to weakness in the subprime mortgage market.
"NovaStar, like a number of other independent subprime mortgage finance companies, is facing lower profitability as well as potentially an increased level of liquidity risk given current market conditions," Moody's said in a statement.
"Although the company has currently maintained servicing performance, Moody's believes that such financial pressure could impact the company's ability and willingness to continue to invest and maintain current resource levels in its servicing platform in the future," Moody's said."
Adam, the 20 issuers does make it much more likely the whole thing won't go to 0. However, if I interpret correctly, and 2006 is a bad year, values less than 60 are possible.6
My limited understanding of the product is that in decent years you get 100% back. However once every 20 years you get a year like 1984 and you may get very little back.
IMO the prices are very weak because someinvestors are afraid 2006 may end up more like 1984 than recent years.
I requested the access to the ABX calculator on the Markit website but as a mere mortal I got no reply. But you can try.
My understanding is that 100 looks just like the starting value and in this case, you need to pay the coupon (e.g 389 basis points for the last BBB- tranche).
The whole index looks like a price of bonds and 75.81 is probably 75.81 cents on the dollar. It should indeed behave like for bonds but these are BBB- only in theory. The tranches are thin and it is quite likely that the investors would see nothing. These CDOs are yielding currently probably upper teens of percent so these are complete junk bonds.
I'm not an expert, just a curious observer so I may be wrong.
Brian: do you mean that it does not bode well for Alt-A because all of those brokers jumped in and were willing to do shady "Alt-A" deals, or are you talking about that one comment from ndebt that said "No more lenders"?
IMO would take a lot of time, good data, and a good model to get even a reasonable value for these things.
Here is where I have a problem - where are you going to get the data? The data is all garbage - data about the borrower and data about the asset valuation.
Preponderance of fraud and stated income loans means data about the borrower is suspect.
Collateral values have no relation to reality (supported by buyer's incomes or rents) Few reckless buyers with no skin took on fraudulent loans and bid the values to sky-high levels. That raised comps on every house in the locality, ensuring that even Alt-A loans may have over-valued collateral.
In what way is interpreting the data different from reading goat entrails?
From Greenspan's speech on April 8, 2005:
"A brief look back at the evolution of the consumer finance market reveals that the financial services industry has long been competitive, innovative, and resilient. Especially in the past decade, technological advances have resulted in increased efficiency and scale within the financial services industry. Innovation has brought about a multitude of new products, such as subprime loans and niche credit programs for immigrants. Such developments are representative of the market responses that have driven the financial services industry throughout the history of our country.
"With these advances in technology, lenders have taken advantage of credit-scoring models and other techniques for efficiently extending credit to a broader spectrum of consumers. The widespread adoption of these models has reduced the costs of evaluating the creditworthiness of borrowers, and in competitive markets cost reductions tend to be passed through to borrowers. Where once more-marginal applicants would simply have been denied credit, lenders are now able to quite efficiently judge the risk posed by individual applicants and to price that risk appropriately. These improvements have led to rapid growth in subprime mortgage lending; indeed, today subprime mortgages account for roughly 10 percent of the number of all mortgages outstanding, up from just 1 or 2 percent in the early 1990s.
"As we reflect on the evolution of consumer credit in the United States, we must conclude that innovation and structural change in the financial services industry have been critical in providing expanded access to credit for the vast majority of consumers, including those of limited means. Without these forces, it would have been impossible for lower-income consumers to have the degree of access to credit markets that they now have. This fact underscores the importance of our roles as policymakers, researchers, bankers, and consumer advocates in fostering constructive innovation that is both responsive to market demand and beneficial to consumers."
Like rubyfan and 4shzl I'm puzzled about how to interpret the ABX values. How do you do the computation referenced in the Bloomberg article where a drop from 100 to 78.6, with a coupon of 389 implies that "an investor would pay more than $1.1 million a year to protect $10 million of bonds against default, up from $389,000 last month"? I'm sure there's a simple answer, but I don't see it.
tj, Don't you mean "...ensuring that even prime loans may have over-valued collateral"?
True, but at least the borrower data is right.
df; With goat entrails you can still eat the goat.
Reminds me of stew. I like goat . But sheep stinks. Trying to pass sheep meat as goat meat is a standard butcher tactic. Sheep and goat w.r.t. MBS - the puns are endless.
IMO the prices are very weak because someinvestors are afraid 2006 may end up more like 1984 than recent years.
I don't disagree with your overall take.
I was just pointing out that there is just a big difference in the leverage you have to sub-prime credit risk between a portfolio of BBB-rated RMBS tranches (like you have when selling protection on the BBB ABX index) and a BBB-rated tranche of a portfolio of BBB-rated RMBS tranches (a tranching of the index).
The ABX BBB-index has been tranched and trading started in the last week, but I can't find any prices on Markit.
My sense of what is important is that RMBS issuers have to place these lower rated tranches first before than can get the real money (the 90 percent of the funding rated AAA). Since the spreads they have to pay investors on these lower-rated tranches are going through the roof, either they pass that along to borrowers in the form of higher rates or the economics of securitization breaks down and you see issuance grind to a halt.
You'd think we'd start hearing 'reports' of issuances grinding to a halt if they really were, wouldn't you?
I realize this market isn't as visible as say the DOW with good looking CNBC babes strutting the trading floor, mic in hand... but I doubt this would stay a secret for long.
dryfly,
The "reports" were headline news. My gold futures spread just topped out. I cannot think of a better indicator to sell the contracts and move liquidity into a stable currency. If the gold isn't enough of a hint what is? The BOJ going to 0.5%? The ABX?
Regarding home price tracking, I really like the old <a href="http://www.housingtracker.net/old_housingtracker/>Housing Tracker. This covers list prices of homes for sale instead of sold prices, so it avoids the problem of stalled out markets shifting to the upside in reported sales. It lags actual events - it takes a while for a stall-out in sales to penetrate through to expectations.
It shows figures for 25%, median and 75% which is very useful. When we get to the two year mark of data this fall people will start to p-ss their britches.
For a preview of what's going to be happening in hot areas toward the end of the selling season in 2007, look at Cape Coral, Tampa, Miami and Orlando. The Florida prime time season in many areas is the reverse of the nationwide pattern - it's hot during snowbird season which runs from November through April.
I wonder how many outstanding contracts there are in those ABX.HE swaps? $100 billion?
The total losses would be, for now, in $10 billion area? It's enough money to be visible on the books when Q1 reports start in April. Some hedge funds are probably suffering as well.
poszi just read your link. It confused me a bit but has completely convinced me there is no way to ballpark the value of these BBB's without a massive effort. Several people who are fairly well connected have stated nobody knows who has all the ugly paper.
My previous posts relied on a simplified analysis of the way they worked I read some time ago.(And don't have)
amamitphd's post about what is important makes complete sense to me.
For those of you who are interested in how the ABX is used, here is a presentation on how the money changes hands between the buyers and sellers of protection.
I believe the ABX pricing is actually simpler than it looks.
First you have to understand that what is being traded is an MBS insurance policy, not a bond. The buyer agrees at the beginning of the contract to pay a "premium" consisting of part of the MBS proceeds to the seller every month.
A buyer is (hypothetically) a person who owns MBS, although it can be a speculator who wants to bet against MBS. A seller is a person willing to make up any shortfalls in interest, principle, or any writedown taken on the MBS during the year. This is conceptually pretty similar to your car insurance or house insurance - the insurance seller collects premiums from the buyer every month and only pays out if something bad happens.
The premium is initially expressed in basis points on the total loan amount. The premium for each index is set when the index is created. As long as people are still willing to buy and sell these contracts for the same level of premium payments, the index stays at 100.
The confusing part is that changes in market conditions are expressed by payments at the start of a new contract, not by changes in the monthly premium. No matter what happens, the buyer's monthly premium for a contract stays the same. As long as the index stays at 100 the initial start-up payment for a contract is zero, and the insurance seller is compensated only by receiving the regular monthly premiums.
When the index moves off 100, then somebody has to pay something at the start of the contract. Each whole number above or below 100 reflects a percentage point adjustment in price. So if the index goes up to 101, the SELLER will actually pay the BUYER 1% of the amount of bonds being insured in exchange for the right to receive the established monthly premiums. This would happen if credit conditions are bad at the start of the index and then later things get better, so people are willing to pay up front for the right to receive those big monthly premium payments.
If the index drops to 98, the insurance BUYER will pay 2% of the amount of the bonds being insured to the SELLER, plus the regular monthly premium. If the index drops to 80, the buyer will pay 20% of the amount of the bonds being insured to the seller, plus the regular monthly premium. So the index only represents the up front cost of a contract and is separate from the premiums.
The Nomura paper linked above from Russ Winter's site sort of explains this. There is also a powerpoint sales presentation from 2006 that comes up on google if you type in "ABX pricing."
albrt, Thanks for the explanation of the way the ABX's work. The mechanics are different than I expected but will result in prices nearly identical to my simpler concept.
I also believe that the "blip" up in the MBA purchase apps index Dec and Jan was related to lenders outside of the MBA survey either closing shop or dramatically curtailing certain types of loans. The MBA survey, while covering "50% of the retail" originations market, is actually based on data from less than 15 big lenders!
From albrt's description, its sounds like there's some counterparty risk in the ABX market. What if the side committed to paying the monthly premiums can't pay?
I think the element in the current financial system structure most likely to bring the system down is misjudgment of counterparty risk.
We're now seeing various subprime mortgage originators who sold off loans with commitments to buy them back in the event of early payment default going out of business and being unable to honor those commitments.
If the other party in an ABX deal is a hedge fund operating with very high leverage, will they be able to make the premium payments if their bets go wrong?
Another interesting aspect of this situation is that, since hedge funds are so little regulated, and are allowed to be so secretive about their trading strategies, how long will it take before large losses on bad trades in such derivatives become visible to the fund investors?
Agree JM. They call this insurance, but these are not insurance companies or anything close. Real insurance companies generally don't write policies for speculation to people who don't have an insurable interest in whatever is being insured.
Albrt, thanks for the explanation on the ABX. You did a better job than the Nomura and Markit pdfs. The Bloomberg reporter in the article being discussed must have had access to the mysterious ABX calculator which makes some assumptions as to the life of the security, and came up with the $1.1 million per year, amortizing the massive up front cost over the estimated lifetime of the security. That is a remarkably punitive price and the BBB-06-2 is even lower. But then these low tranches must be genuinely nasty stuff.
"The BBB- and BBB tranches are clearly making no sense anymore."
They ARE making sense and CAN trade a lot lower.
These are not BBB- or BBB bonds we are talking about. Whereas if 10-15% of the bonds defaulted across a portfolio of BBB- or BBB bonds would entail a loss of 10% or less on the portfolio, here is a default rate of 10-15% hits the underlying for these things, these things can become WORTHLESS, ALL of them.
What doesn't make sense anymore is for anyone to accept these tranches in a new securitization. If they wanted risk, they'd get risk in the market at a much higher yield.
Which COULD lead to a complete halt on CDO issuance, at least on CDOs having this kind of sheet in them.
I think they deliberately made these indexes look like bonds in order to confuse the public. For a long time they were presented in the media as reflecting minimal risk as long as they appeared to be trading near "par." A bond dropping from 100 to 95 is not actually that big of a change in the risk premium, although it is a big loss to a leveraged bond speculator. ABX index dropping by a couple of points was huge.
I think the ABX conspiracy theories are misplaced. The things have not been around that long, they are not available to the general public, and if you tried to buy one you would get an IDSA contract 20 pages long that you would be asked to sign. (If you are not convinced, follow the link to the Nomura document that explains them - the instruction manual is over 10 pages long) The people trading these things are all institutions who are assumed to have adult supervision.
The issue of counterparty risk is real and has yet to be tested.
The other thing to keep in mind when looking at the ABX is that it trades at a wider spread to the underlying credit default swaps which in turn trade at a wider spread than the cash tranches that they reference. There are hundreds of basis points difference in spread between the ABX and the BBB- tranches. So the ABX overstates the hit to the securitization economics. That said, the math in the pools has to be getting pretty challenging.
Brian - I agree the people actually trading ABX understand it much better than I do, but for a while the indices were being pointed out in the mainstream financial media as evidence that everything was great in subprime land. I can't think of any other reason why the price would be quoted this way except to try and confuse the uninitiated. It would be so much simpler to just say "A pays B 2% when the contract is initiated."
CR,
I love your blog (which also includes the sound contributions of many regulars). Many thanks for the free education.
If your hypothesis is true, then we should see a proportional drop in recorded sales in three months, since the purchase index is coming down to previous levels.
When did these small lenders join the party? Does this mean the numbers were distorted on the way up, too?
Doug Noland from over at
prudentbear.com
We are in the midst of a unique Credit cycle. The ability for originators to sell loans and immediately book profits; for investment bankers to buy, securitize and immediately book profits; and for leveraged speculators to acquire various securitizations and other derivatives and book easy profits from various spreads and mark-to-model - have all made this Credit boom unlike any other. ..
Today, Subprime mortgage originator profits are collapsing lending volumes are sinking; gain on sale is reversing to loss; Credit losses (especially from returned early defaults) are surging; and the liquidity necessary to operate is disappearing overnight. This has initiated the ugly downside of operating as a Ponzi Finance Unit. The issue of early payment defaults where investment bankers/securitization pool operators return problem mortgages in droves back to the originator is rapidly bankrupting this thinly-capitalized industry. And the more acute the risk of insolvency, the greater the incentive for investment banks to rush to dump problem loans while the originator still retains some liquidity (think bank run).
This has enormous ramifications for the industry. Subprime Credit conditions are in the process of tightening how tight only time will tell. Subprime borrowers this year facing payment resets will confront a changed industry. Desperate borrowers stretched the truth in 2006 to get new mortgages approved. The weakest originators are disappearing
But like everything else associated with this most extraordinary Credit Cycle, the analysis is infinitely more complex than what meets the eye. We are undoubtedly in the midst of a major liquidity event for the subprime originators. Additionally, the riskiest CDO and securitization tranches (and related derivatives) will suffer heavy losses. This is a decisive Credit event for the subprime industry Thus far, however, there is little indication that the (paramount) market for prime mortgages is being impacted much at all. And when it comes to the subject of overall system liquidity, the current level of tumult could prove less than significant.
In a different or perhaps typical - environment, recent subprime developments would prompt a nervous reaction from the markets. But today we operate in a most extraordinary backdrop of rampant global liquidity excess, evidenced by ongoing international booms in M&A, securities markets, real estate, energy, and most asset markets. Here at home, continued strong growth in (non-subprime) real estate lending combines with robust corporate and government Credit growth to ensure - for now - continued sufficient non-financial debt growth. More ambiguous but equally important, indications point to continued robust expansion in securities finance, at this stage of the Credit Cycle a liquidity-creating be
Another great chart and analysis. It's hard to see through some of the noise, and CR is good at that.
How long does it take to be rejected by a lender? Wouldn't somebody denied multiple times show up as multiple applications? A credit tightening could increase the number of people who applied multiple times for a loan based on the same property, even if NO lenders went Tango Uniform.
So MBA covers 50% of the mortgage market, the respectable side. And when the other side caves in, those in the sub-prime arena are forced to cough up the extra nickel or delay that mortgage application until they can afford the prime rates. Since New or Existing home sales didn't reflect this increase as CR notes, it's hard to think of any other explanation.
Was there a bulge in refis? no.
Was there an increase in the cheapest variety (ARMs) of mortgage application? probably.
Aaron Krowne, the guy who runs the mortgage lenders implode o meter site, mentioned that 4 of 5 subprimes are now being rejected.
Podcast link below, its a bit long with some interesting stuff.
StreetIQ.com - The Leader in Financial and Business Podcasts
My thought was the same as Jim a's.
Would CR provide his thoughts/analytics on the possibility that the MBA applications have increased as the result of borrowers needing to submit multiple applications?
Thanks.
C-BASS looking more and more like a vulture fund:
"DJ C-BASS Named As Lead Bidder In Auction Of Ownit Mortgages
2007-02-21 12:37 (New York)
New York mortgage lender C-BASS LLC has bid $4 million for 54 mortgages being
auctioned off by bankrupt subprime lender Ownit Mortgage Solutions Inc.
Ownit Mortgage and the committee representing its unsecured creditors named
C-BASS as the stalking horse, or lead, bidder in the sale. Mortgage insurers
MGIC Investment Corp. (MTG) and Radian Group Inc. (RDN) are the majority owners
of C-BASS.
Judge Kathleen Thompson of the U.S. Bankruptcy Court in Woodland Hills,
Calif., has scheduled a hearing for Thursday to consider approval of the sale."
We don't know what the average mortgage size is, but if its $200K, their opening bid is less than 40 cents on the dollar (and they get the servicing rights in the deal). We'll see what happens at the auction, but 60% is quite a haircut.
Sorry, I don't have a public link for this.
I can't begin to assess all the factors that might affect the MBA stat, but I do want to mention that first-time home buyers are a particular subject of tightening, as they have unique risk factors. Therefore, I would expect there to be a detectable increase in the average ratio of apps to approvals for the poorer credit quality FTHBs in the next two months and this has clearly been going on for a couple of months.
I have read multiple threads by brokers bitching about the pile of apps on their desks that they can't get approved. So....
sts, Did Aaron cite source for that stat on subprime rejection rate?
Brian,
He mentioned something to the effect that he picked it up from posted mortgage brokers' comments as he was looking for info for his ML implode o meter site.
So probably anecdotal but consistent with other info.
At the end of 2006 there appeared to be a surge in purchase applications. This might have been due to increased activity, or possibly favorable weather.
Goddammit CR. That's just too funny. You owe me another keyboard.
Might I suggest other factors? Tightening credit. With closer scrutiny would not applicants either prophylactically or reactively make multiple applications?
Second musing, much like the first; With 24 name brand mortgage brokerages down in the last 9 weeks wouldn't every one of the loans applied for but not funded be counted a second time in their new morgage brokerage house?
I guess the problem I have imputing this to the "sub-prime bust" as opposed to the weather is that you have to conclude that the sub-prime market has been picking up in recent weeks.
I think both things are a factor, but weather moreso. As you get further up north it just makes a huge difference having 1 month of winter instead of 3.
MaxedOut,
This highlights the fraud in GA. Any particular reason fraud is so bad down there?
http://www.mari-inc.com/pdfs/mba/MBA8thCaseRpt.pdf
Steve, yes, there could have been an impact the other way too - we have to remember the MBA Index isn't perfect.
All, yes, multiple applications is another possible source for the recent spike. Going forward I expect the index to trend.
Best to all.
"This highlights the fraud in GA. Any particular reason fraud is so bad down there?"
Brian, I've wondered the same.
Somebody has to sit at the top of the heap, but why Georgia? I would have guessed CA, or FL, or AZ or NV.
Brian - short answer - crooks. It seems to be concentrated around the metro area, which of course had favorable conditions for such activity.
Judging from my personal experience, I think we are catching more of them RIGHT NOW than other states. I think there is more fraud in FL than in GA overall.
I notice that whenever incomes stagnate and foreclosures rise a bunch of these schemes come to light - there's been a lot of reports of fraud in the CO foreclosure epicenter, for example. My guess is that it is not detected as readily if the market is good.
I've been reading these comment threads for several weeks, and haven't heard anyone complain about the rating agencies.
At the end of the day, it is these guys that look at what these originators are doing and slap credit ratings on the senior tranches of RMBS, which brings in 90 percent of the funding for the deal.
How could all of this fraud (cahs back at closing, appraisal fraud, stated-income) have gone on undetected by the rating agencies? At the end of the day, isn't that their job? Investors certainly deserve blame for not looking, but shouldn't these guys deserve a big part of the blame for not nipping this in the bud a long time ago?
"Fed take over CU"
Bakersfield Bubble
re the MBA report, please note the date was May 24, 2006, and references events through "early 2006".
Who is the bozo who think he can get such loan:
Mortgage Grapevine: CA, $1,400,000, 728 FICO, 100% LTV, Purchase, SF, nina, OO
No income, No assets needs $1.4M for 100% of a new home......
Yowie! Zowie! Default insurance for the 07H1 BBB- tranche: it now costs $1.1 million to insure $10 million for one year!!?? Can this be correct?
Subprime Mortgage Bond Derivatives Fall After NovaStar's Loss - Bloomberg.com
IMHO this may be reasonable since this is a thin tranche not a bond. If losses are bad it is possible for the BBB's to have hardly any return before losses wipe it out.
IMO would take a lot of time, good data, and a good model to get even a reasonable value for these things.
Sadly many buyers probably failed to do this analysis.
Someone please correct me if I've missed something in this analysis.
"Someone please correct me if I've missed something in this analysis."
You haven't. It will be a BIG surprise for a lot of people who think they bought "BBB- paper".
It really makes one thinm how is it possible for the rating agencies to get away with this, as well.
IMHO this may be reasonable since this is a thin tranche not a bond. If losses are bad it is possible for the BBB's to have hardly any return before losses wipe it out.
Minor disagreement.
My understanding is that the BBB ABX is a portfolio of mezzanine ABS tranches. Each tranche has a 5 percent weight. The investor has the same leverage to the asset class as buying one mezz tranche, but gets a diversification benefit through a portfolio of issuers.
The BBB refers to the quality of collateral in the index. When the index itself is tranched, then the investor in the BBB-tranche of the index would have even more leverage wrt the asset class.
Just curious: how do they actually do the calculation that since the ABX-HE-BBB- 06-2 is now at 75.81 that it takes $1.1Million to insure $10million?
Also, how do the starting prices for these instruments (usually just above 100) get set? I'm a newbie and I look at a lot of these ABX graphs and think that they're 'kinda-sorta' like bonds and thus they shouldn't fluctuate much, but when you see those graphs tanking, well, I can tell that something is up, but I'm not quite sure how to explain it to people.
how do they actually do the calculation that since the ABX-HE-BBB- 06-2 is now at 75.81 that it takes $1.1Million to insure $10million?
Excellent question -- I'd like to know too.
And here's another one: what exactly does the paragraph below mean? Are hedgefunds using unregulated derivative contracts to manipulate outcomes in other markets?
Stock investors betting on a decline in the shares of companies that make subprime loans have used derivatives contracts tied to ABX indexes as an alternative, said Howard Hill, a managing director in Springfield, Massachusetts at Babson Capital Management LLC. The firm is a unit of MassMutual Financial Group, which manages about $30 billion of mortgage- and asset-backed bonds.
While the pricing on the BBB tranche is very dramatic, take a look at the pricing on the A tranche of the two most recent series - it's down to 95...while the ABX trades wide of the cash bonds, I suspect these levels make it hard to get any securitizations done....the higher up the rating scale you go, the bigger the amounts involved and the bigger the hit to the economics for the issuer.
Another leg of the Novastar stool starts to wobble.
"NEW YORK, Feb 21 (Reuters) - Moody's Investors Service on Wednesday said it may cut its servicer quality ratings on NovaStar Mortgage, Inc. due to the mortgage servicer's exposure to weakness in the subprime mortgage market.
"NovaStar, like a number of other independent subprime mortgage finance companies, is facing lower profitability as well as potentially an increased level of liquidity risk given current market conditions," Moody's said in a statement.
"Although the company has currently maintained servicing performance, Moody's believes that such financial pressure could impact the company's ability and willingness to continue to invest and maintain current resource levels in its servicing platform in the future," Moody's said."
Business & Financial News, Breaking US & International News | Reuters.com
This does not bode well for the Alt-A mkt.
Alt-A Lenders - Contact information only
Adam, the 20 issuers does make it much more likely the whole thing won't go to 0. However, if I interpret correctly, and 2006 is a bad year, values less than 60 are possible.6
My limited understanding of the product is that in decent years you get 100% back. However once every 20 years you get a year like 1984 and you may get very little back.
IMO the prices are very weak because someinvestors are afraid 2006 may end up more like 1984 than recent years.
OT - but I wanted to make sure folks here saw it: another CR forecast being validated in the fullness time. Props to our host.
Housing `Hangover' Kills Jobs as Spending Wanes; More Cuts Loom - Bloomberg.com
rubyfan,
I requested the access to the ABX calculator on the Markit website but as a mere mortal I got no reply. But you can try.
My understanding is that 100 looks just like the starting value and in this case, you need to pay the coupon (e.g 389 basis points for the last BBB- tranche).
The whole index looks like a price of bonds and 75.81 is probably 75.81 cents on the dollar. It should indeed behave like for bonds but these are BBB- only in theory. The tranches are thin and it is quite likely that the investors would see nothing. These CDOs are yielding currently probably upper teens of percent so these are complete junk bonds.
I'm not an expert, just a curious observer so I may be wrong.
Brian: do you mean that it does not bode well for Alt-A because all of those brokers jumped in and were willing to do shady "Alt-A" deals, or are you talking about that one comment from ndebt that said "No more lenders"?
Talk about falling off a cliff:
ABX-HE-BBB- 06-2
(Not certain this link will work for other people.)
IMO would take a lot of time, good data, and a good model to get even a reasonable value for these things.
Here is where I have a problem - where are you going to get the data? The data is all garbage - data about the borrower and data about the asset valuation.
Preponderance of fraud and stated income loans means data about the borrower is suspect.
Collateral values have no relation to reality (supported by buyer's incomes or rents) Few reckless buyers with no skin took on fraudulent loans and bid the values to sky-high levels. That raised comps on every house in the locality, ensuring that even Alt-A loans may have over-valued collateral.
In what way is interpreting the data different from reading goat entrails?
From Greenspan's speech on April 8, 2005:
"A brief look back at the evolution of the consumer finance market reveals that the financial services industry has long been competitive, innovative, and resilient. Especially in the past decade, technological advances have resulted in increased efficiency and scale within the financial services industry. Innovation has brought about a multitude of new products, such as subprime loans and niche credit programs for immigrants. Such developments are representative of the market responses that have driven the financial services industry throughout the history of our country.
"With these advances in technology, lenders have taken advantage of credit-scoring models and other techniques for efficiently extending credit to a broader spectrum of consumers. The widespread adoption of these models has reduced the costs of evaluating the creditworthiness of borrowers, and in competitive markets cost reductions tend to be passed through to borrowers. Where once more-marginal applicants would simply have been denied credit, lenders are now able to quite efficiently judge the risk posed by individual applicants and to price that risk appropriately. These improvements have led to rapid growth in subprime mortgage lending; indeed, today subprime mortgages account for roughly 10 percent of the number of all mortgages outstanding, up from just 1 or 2 percent in the early 1990s.
"As we reflect on the evolution of consumer credit in the United States, we must conclude that innovation and structural change in the financial services industry have been critical in providing expanded access to credit for the vast majority of consumers, including those of limited means. Without these forces, it would have been impossible for lower-income consumers to have the degree of access to credit markets that they now have. This fact underscores the importance of our roles as policymakers, researchers, bankers, and consumer advocates in fostering constructive innovation that is both responsive to market demand and beneficial to consumers."
That raised comps on every house in the locality, ensuring that even Alt-A loans may have over-valued collateral.
Don't you mean "...ensuring that even prime loans may have over-valued collateral"?
In what way is interpreting the data different from reading goat entrails?
With goat entrails you can still eat the goat.
This is big;
Bakersfield Bubble
S&P launches national home price index
Like rubyfan and 4shzl I'm puzzled about how to interpret the ABX values. How do you do the computation referenced in the Bloomberg article where a drop from 100 to 78.6, with a coupon of 389 implies that "an investor would pay more than $1.1 million a year to protect $10 million of bonds against default, up from $389,000 last month"? I'm sure there's a simple answer, but I don't see it.
tj,
Don't you mean "...ensuring that even prime loans may have over-valued collateral"?
True, but at least the borrower data is right.
df;
. But sheep stinks. Trying to pass sheep meat as goat meat is a standard butcher tactic. Sheep and goat w.r.t. MBS - the puns are endless.
With goat entrails you can still eat the goat.
Reminds me of stew. I like goat
IMO the prices are very weak because someinvestors are afraid 2006 may end up more like 1984 than recent years.
I don't disagree with your overall take.
I was just pointing out that there is just a big difference in the leverage you have to sub-prime credit risk between a portfolio of BBB-rated RMBS tranches (like you have when selling protection on the BBB ABX index) and a BBB-rated tranche of a portfolio of BBB-rated RMBS tranches (a tranching of the index).
The ABX BBB-index has been tranched and trading started in the last week, but I can't find any prices on Markit.
My sense of what is important is that RMBS issuers have to place these lower rated tranches first before than can get the real money (the 90 percent of the funding rated AAA). Since the spreads they have to pay investors on these lower-rated tranches are going through the roof, either they pass that along to borrowers in the form of higher rates or the economics of securitization breaks down and you see issuance grind to a halt.
You'd think we'd start hearing 'reports' of issuances grinding to a halt if they really were, wouldn't you?
I realize this market isn't as visible as say the DOW with good looking CNBC babes strutting the trading floor, mic in hand... but I doubt this would stay a secret for long.
Anyone?
dryfly,
The "reports" were headline news. My gold futures spread just topped out. I cannot think of a better indicator to sell the contracts and move liquidity into a stable currency. If the gold isn't enough of a hint what is? The BOJ going to 0.5%? The ABX?
On Russ Winter's blog, a reader linked this paper about ABX
http://www.securitization.net/pdf/Nomura/SyntheticABS_7Mar06.pdf
Alas, I hardly understand it better after reading it.
I'm also curious who is currently getting margin calls on this market.
I think securitization does not happen every week. Only 2 weeks passed sinse HSBC & NEW bomb.
Maybe it is already halted, it just takes another week or two before we know?
The BBB- and BBB tranches are clearly making no sense anymore.
Regarding home price tracking, I really like the old <a href="http://www.housingtracker.net/old_housingtracker/>Housing Tracker. This covers list prices of homes for sale instead of sold prices, so it avoids the problem of stalled out markets shifting to the upside in reported sales. It lags actual events - it takes a while for a stall-out in sales to penetrate through to expectations.
It shows figures for 25%, median and 75% which is very useful. When we get to the two year mark of data this fall people will start to p-ss their britches.
For a preview of what's going to be happening in hot areas toward the end of the selling season in 2007, look at Cape Coral, Tampa, Miami and Orlando. The Florida prime time season in many areas is the reverse of the nationwide pattern - it's hot during snowbird season which runs from November through April.
I wonder how many outstanding contracts there are in those ABX.HE swaps? $100 billion?
The total losses would be, for now, in $10 billion area? It's enough money to be visible on the books when Q1 reports start in April. Some hedge funds are probably suffering as well.
poszi just read your link. It confused me a bit but has completely convinced me there is no way to ballpark the value of these BBB's without a massive effort. Several people who are fairly well connected have stated nobody knows who has all the ugly paper.
My previous posts relied on a simplified analysis of the way they worked I read some time ago.(And don't have)
amamitphd's post about what is important makes complete sense to me.
For those of you who are interested in how the ABX is used, here is a presentation on how the money changes hands between the buyers and sellers of protection.
http://www.markit.com/information/affiliations/tabx/content2Paragraphs/02/document/TABX.HE.Marketing.Launch.pdf
That raised comps on every house in the locality, ensuring that even Alt-A loans may have over-valued collateral.
Don't you mean "...ensuring that even prime loans may have over-valued collateral?
Confidence! is always the 800lb gorilla in any financial market.
I believe the ABX pricing is actually simpler than it looks.
First you have to understand that what is being traded is an MBS insurance policy, not a bond. The buyer agrees at the beginning of the contract to pay a "premium" consisting of part of the MBS proceeds to the seller every month.
A buyer is (hypothetically) a person who owns MBS, although it can be a speculator who wants to bet against MBS. A seller is a person willing to make up any shortfalls in interest, principle, or any writedown taken on the MBS during the year. This is conceptually pretty similar to your car insurance or house insurance - the insurance seller collects premiums from the buyer every month and only pays out if something bad happens.
The premium is initially expressed in basis points on the total loan amount. The premium for each index is set when the index is created. As long as people are still willing to buy and sell these contracts for the same level of premium payments, the index stays at 100.
The confusing part is that changes in market conditions are expressed by payments at the start of a new contract, not by changes in the monthly premium. No matter what happens, the buyer's monthly premium for a contract stays the same. As long as the index stays at 100 the initial start-up payment for a contract is zero, and the insurance seller is compensated only by receiving the regular monthly premiums.
When the index moves off 100, then somebody has to pay something at the start of the contract. Each whole number above or below 100 reflects a percentage point adjustment in price. So if the index goes up to 101, the SELLER will actually pay the BUYER 1% of the amount of bonds being insured in exchange for the right to receive the established monthly premiums. This would happen if credit conditions are bad at the start of the index and then later things get better, so people are willing to pay up front for the right to receive those big monthly premium payments.
If the index drops to 98, the insurance BUYER will pay 2% of the amount of the bonds being insured to the SELLER, plus the regular monthly premium. If the index drops to 80, the buyer will pay 20% of the amount of the bonds being insured to the seller, plus the regular monthly premium. So the index only represents the up front cost of a contract and is separate from the premiums.
The Nomura paper linked above from Russ Winter's site sort of explains this. There is also a powerpoint sales presentation from 2006 that comes up on google if you type in "ABX pricing."
The powerpoint I was talking about is the same one Brian linked.
albrt, Thanks for the explanation of the way the ABX's work. The mechanics are different than I expected but will result in prices nearly identical to my simpler concept.
I also believe that the "blip" up in the MBA purchase apps index Dec and Jan was related to lenders outside of the MBA survey either closing shop or dramatically curtailing certain types of loans. The MBA survey, while covering "50% of the retail" originations market, is actually based on data from less than 15 big lenders!
From albrt's description, its sounds like there's some counterparty risk in the ABX market. What if the side committed to paying the monthly premiums can't pay?
I think the element in the current financial system structure most likely to bring the system down is misjudgment of counterparty risk.
We're now seeing various subprime mortgage originators who sold off loans with commitments to buy them back in the event of early payment default going out of business and being unable to honor those commitments.
If the other party in an ABX deal is a hedge fund operating with very high leverage, will they be able to make the premium payments if their bets go wrong?
Another interesting aspect of this situation is that, since hedge funds are so little regulated, and are allowed to be so secretive about their trading strategies, how long will it take before large losses on bad trades in such derivatives become visible to the fund investors?
Agree JM. They call this insurance, but these are not insurance companies or anything close. Real insurance companies generally don't write policies for speculation to people who don't have an insurable interest in whatever is being insured.
Albrt, thanks for the explanation on the ABX. You did a better job than the Nomura and Markit pdfs. The Bloomberg reporter in the article being discussed must have had access to the mysterious ABX calculator which makes some assumptions as to the life of the security, and came up with the $1.1 million per year, amortizing the massive up front cost over the estimated lifetime of the security. That is a remarkably punitive price and the BBB-06-2 is even lower. But then these low tranches must be genuinely nasty stuff.
"The BBB- and BBB tranches are clearly making no sense anymore."
They ARE making sense and CAN trade a lot lower.
These are not BBB- or BBB bonds we are talking about. Whereas if 10-15% of the bonds defaulted across a portfolio of BBB- or BBB bonds would entail a loss of 10% or less on the portfolio, here is a default rate of 10-15% hits the underlying for these things, these things can become WORTHLESS, ALL of them.
What doesn't make sense anymore is for anyone to accept these tranches in a new securitization. If they wanted risk, they'd get risk in the market at a much higher yield.
Which COULD lead to a complete halt on CDO issuance, at least on CDOs having this kind of sheet in them.
vicjim:
I think they deliberately made these indexes look like bonds in order to confuse the public. For a long time they were presented in the media as reflecting minimal risk as long as they appeared to be trading near "par." A bond dropping from 100 to 95 is not actually that big of a change in the risk premium, although it is a big loss to a leveraged bond speculator. ABX index dropping by a couple of points was huge.
I'd go farther. Not only are these indexes made to look like bonds, the CDO tranches themselves are made to look like bonds and rated like bonds.
But one thing is sure, they aren't bonds.
I think the ABX conspiracy theories are misplaced. The things have not been around that long, they are not available to the general public, and if you tried to buy one you would get an IDSA contract 20 pages long that you would be asked to sign. (If you are not convinced, follow the link to the Nomura document that explains them - the instruction manual is over 10 pages long) The people trading these things are all institutions who are assumed to have adult supervision.
The issue of counterparty risk is real and has yet to be tested.
The other thing to keep in mind when looking at the ABX is that it trades at a wider spread to the underlying credit default swaps which in turn trade at a wider spread than the cash tranches that they reference. There are hundreds of basis points difference in spread between the ABX and the BBB- tranches. So the ABX overstates the hit to the securitization economics. That said, the math in the pools has to be getting pretty challenging.
Brian - I agree the people actually trading ABX understand it much better than I do, but for a while the indices were being pointed out in the mainstream financial media as evidence that everything was great in subprime land. I can't think of any other reason why the price would be quoted this way except to try and confuse the uninitiated. It would be so much simpler to just say "A pays B 2% when the contract is initiated."