August 13, 2008
It's A Good Thing It's Simply A Subprime (And District 10) Problem...
"Yields on mortgage securities guaranteed by Fannie Mae rose this week to about their highest relative to Treasuries since March amid concern that defaults are spreading to prime and Alt-A mortgages from subprime loans.
Fannie's current-coupon 30-year fixed-rate bonds currently yield 6.07 percent, 213 basis points more than 10-year Treasuries, according to data compiled by Bloomberg. That's 25 basis points from the 22-year high of 238 reached March 6, a week before the Federal Reserve helped bail out Bear Stearns Cos.
The worst housing slump since the Great Depression has blotted out much of the wealth Americans accumulated in their homes, hurting their ability to pay bills and boosting spreads on auto-loan and credit-card backed bonds as well as mortgage securities. Fannie Mae, the largest U.S. mortgage-finance company, last week slashed its dividend 86 percent after posting a worse-than-expected loss and said it will stop buying and guaranteeing Alt-A loans."
[Editor's Note: And no, this shouldn't catch any plugged-in people by surprise.]
∙ Agency Mortgage Bond Yield Spreads Widen as Loan Losses Expand [Bloomberg]
∙ Subprime And Alt-A Statistics By County: The Feds Mortgage Map [SocketSite]
∙ Fannie Follows Freddie (And Makes It Easy For The Copywriters) [SocketSite]
∙ Fannie Mae, Battling Losses, to End Alt-A Mortgages [Bloomberg]
∙ JustQuotes: Is The Subprime Sickness Spreading? [SocketSite 7/07]
First Published: August 13, 2008 2:30 PM
Comments from "Plugged In" Readers
God, I hate the way the media describes this, like these poor "prime" borrower's can't pay their mortgage because a bunch of subprime mortgages defaulted.
Defaults aren't "spreading into" Alt-A and Prime. The lender's did a crappy job of qualifying the borrower's and approved them for loans they couldn't possibly, or were unlikely, to ever repay.
These borrower's aren't defaulting because a bunch of subprime loans defaulted, the borrower's are defaulting because they should have never been lent the money.
Posted by: badlydrawnbear at August 13, 2008 3:30 PM
It makes you wonder if "prime" credit really meant anything in 2005-2007 anyway, much like many AAA rated mortgage securities were really just junk.
Wasn't there a chart on this site a while ago showing that S.F. didn't have much Alt-A activity?
It's amazing that treauries are still at 4% and the 30 year Fannie bonds are at 6%. The economy is in trouble and the dollar is weak. At some point don't these yields have to rise substantially?
Posted by: anono at August 13, 2008 3:45 PM
couldn't agree more badlydrawnbear. but here's my version:
Defaults aren't "spreading into" Alt-A and Prime. The borrowers did a crappy job of reading their loan docs and applying 3rd grade multiplication and addition, and signed for loans they couldn't possibly, or were unlikely, to ever repay.
These borrower's aren't defaulting because a bunch of subprime loans defaulted, the borrower's are defaulting because they should have never borrowed the money.
Posted by: resp at August 13, 2008 3:47 PM
Bdb is right. The subprime, prime and Alt-A issues are really quite different, even though they are all heading in the same direction.
Here is a pretty unbelievable set of numbers reported by CNNMoney showing how quickly the prime problems are beginning to snowball.
Perhaps of particular note to the SF market: "Delinquencies jumped even more for prime loans of more than $417,000, so-called jumbo loans. They rose to 4.03% of outstanding loans in May, compared with 1.11% a year earlier." And the re-sets in the Alt-A space (and prime to a lesser degree) are just really starting to hit, so who knows where things will go once that takes hold.
Can't say for sure what effect, if any, this will have on the SF market and prices, but it sure does not appear to indicate that the current declines will reverse any time soon.
Posted by: Trip at August 13, 2008 3:52 PM
"It's amazing that treauries are still at 4% and the 30 year Fannie bonds are at 6%. The economy is in trouble and the dollar is weak. At some point don't these yields have to rise substantially?'
Not necessarily IMO. Treasuries can continue to go lower in terms of rates. In fact, that is what I expect because treasuries - for better or worse - are the best credit quality USD debt instruments around.
It is actually common for yields on sovereign debt for "1st world" countries to plummet in recession/depression. This happened in Japan for instance throughout the 1990s into the 2000s, when yields on 10Y Japanese gov's fell from the 7% range in 1989/90 to below 1% (incredible, isn't it - this is ten year money!) in 2003/4 (if my memory is right). A relatively weak currency would not stop this fall in rates. In japan, the yen weakened substantially from its highs in the early 90s through the dollar spike in 1998/99.
In addition, this fall in sovereign rates happened in the US after 1933 or so, falling right into the war period, even though the dollar was devalued approximately 33% in terms of its peg to gold (from $20.50/oz to about $31/oz in 1934).
As for mortgage rates, again, it will depend on credit quality. With very tight lending standards, mortgage rates could conceivably follow treasuries down. This was also the experience of Japan. However, no one who actually needs the money will be able to borrow!
Posted by: Satchel at August 13, 2008 4:11 PM
it's not a surprise.
Mortgage rates hardly budged when the Fed changed the Fed Funds Rate from 1% to 5.25%. It was Greenspan's so called "conundrum" or "worldwide saving's glut"
And now they're hardly budging with the Fed Funds rate going from 5.25% to 2%. The reverse of the above.
absent further major governmental intervention, it is highly unlikely we will ever see mortgage rates as low as they were in 2003, possibly ever again in our lifetime. However, I never rule out governmental intervention so who knows what rediculous scheme they will come up with!
but many of us here (including me) discussed ad nauseum why mortgage rates wouldn't fall much and may go up despite lowered FFR and Fannie/Freddie intervention.
Posted by: ex SF-er at August 13, 2008 4:13 PM
Maybe the subprime HAS something to do with the defaults on prime & Alt-A, mainly home price and expectations.
Subprime loans default made the home prices go down, so now the expectation is that price will go down, so they continue.
If you borrow 100% to pay your house and that house loose $100k or more, you'll stop paying your mortgage (even if you can afford to pay it) live rent free for 9 months waiting for the default process to kick you out of the house, get a crappy credit report and then move to rent a place.
So in effect if you loose money on your property you have to choose whether to cut the loss or to continue paying.
Posted by: rno at August 13, 2008 4:15 PM
Sometimes I just love the quality of discourse on socketsite.
At what point does money (borrowing) become too cheap, and did we reach that in 2003? For my own opinion, a little strong medicine is necessary. People, borrowers and banks included, need to feel the pain of their mistakes. There needs to be a connection between risk and who's bearing the risk. That is why I am happy that in the long run, covered bonds will replace mortgage backed securities and that finally, finally, we are realizing the wisdom of these instruments.
Posted by: dede at August 13, 2008 4:59 PM
"There needs to be a connection between risk and who's bearing the risk."
Agreed... However, god forbid we get to the point where small investors no longer feel comfortable taking risk. No risk, no economic growth. I think it is good to encourage reason-based risk taking. Conforming loans tended to have it right. Subprime and Alt-A - not so much.
Who in the world thought it would be a good idea to encourage the offering of unregulated debt to unsophisticated borrowers? Commercial banks lent money to risky borrowers - and unloaded the risk through securitizations. Letting banks and wall street do this sort of thing on an unregulated basis is like shipping carrot by rabbit.
By the way - guess who bought the riskiest traunches of these securitizations? Pension plans in Australia and Europe. In addition to outsourcing jobs, we outsourced our nasty brew of financial alchemy. Many public pension plans in Australia are suffering horribly.
When investing in nonregistered hedge funds, at least the investors need to have $1.5 million in liquid assets (Reg D) or even more to qualify (3c1, 3c7). With debt, regulation was thrown out the door.
The invisible hand is called "invisible" for a reason - it doesn't exist. Capitalism can only work when it is enforced. The job of government is not business - but the ethnical and reasonable regulation of business practices.
Posted by: marketwatcher at August 13, 2008 5:23 PM
If this is the worst downturn ever, I think we can survive anything, b/c it doesn't feel bad at all in SF. Would you guys not agree?
Posted by: Wonder [aka Prime] at August 13, 2008 7:02 PM
This is how it worked last time. Inflation heated up, values maybe rose or fell a percentage point or two a year, and people didn't even realize they were shedding enough equity for a genuine correction until it was too late. We have years to go with this, and San Francisco has traditionally recovered slowly from recessions just like it slowly dips into them.
Posted by: Mole Man at August 13, 2008 7:33 PM
It's unclear what you were getting at with your 'yields have to rise comment', but for those who don't know:
The more treasuries are in demand, the higher the price becomes, thus the lower the yield. This is somewhat hard to figure out at first, but actually does make sense. Say I bought a $10,000 treasury at auction yielding 5%/yr, or $500. Now, say the stock market collapses and everyone wants to buy bonds, especially US govt bonds, so now I can sell my bond for $15000 (that's extreme, but makes the math easy). Now, the bond will still pay $500 a year (5% of face), but now $500 is only 3.33% of what I paid, so the yield has fallen substantially.
So, in my opinion, the worse the economy does, and, especially as the stock market continues to fall, the lower the yields on treasuries will get.
A 200 point basis spread between treasuries and Fannie Mae's means that people are willing to pay much less for FN securities, so the yield goes higher (higher risk, higher reward).
Posted by: Ryan at August 13, 2008 8:37 PM
"At what point does money (borrowing) become too cheap"
Nobody can truly answer that question, but I have a rule of thumb.
Borrowing money should for most times be at positive REAL interest rates (as measured using the Fed Funds Rate)
we did not have this in 2003, or now. Currently, inflation is higher than the Fed Funds Rate. Thus, the banks are getting paid by the Federal Reserve to Borrow. They are not passing these savings on. Instead they are taking money cheap and lending at higher rates, using the wider spread to repair their balance sheet. In the end this may be necessary, and it is what the Fed WANTS them to do (but NOT what Congress wants). But most people don't realize what is happening here (a massive bailout of the banks by the Fed that has real consequences on American Families).
bad things (i.e. bubbles) tend to occur when the Fed artificially drives interest rates negative and holds them there. (because there is no "penalty" for borrowing, so people borrow money to "invest" in stupid endeavors, and they borrow money to buy stupid non-productive "assets"... people call this "misallocation of resources"). it's what we did in 2003, and what we are doing now.
as for future Treasury Yields, nobody knows what they will be. Satchel and I disagree as example. I am not sure what they will do. It is possible that the Fed will hold or continue dropping the Fed Funds Rate, and that Japan will hold at 0% and Eurozone/Bank of England will start dropping their rates too due to worldwide recession. Worldwide recession would lower stock returns and also could lower the safety of bonds. If this occurs, then more people will seek relative safety of the US Treasury, which is still the safest investment on the planet. (more people buying Treasuries equals higher price and thus lower yield). This is what Satchel sees. It could happen (I'm not sure)
Or, we could run into a problem with our Federal Govt putting out too much debt. Due to all the extra debt (bailing out Bear Stearns, Fannie, Freddie, possible bailout of other entities like FDIC or the PBIG, not to mention continued Iraq war borrowing, Stimulus checks, etc etc etc) the Govt will have to create a lot more Treasuries. More Treasuries equals more supply of treasuries. More supply of Treasuries equals lower Treasury prices equals higher Treasury Yields.
I am not sure what will happen yet. my personal opinion is that we will likely see no major move in Treasuries in the near future, slightly to moderately higher mortgage rates going forward near term, and substantially higher Treasury Rates and Mortgage Rates in the DISTANT future (but have not thought about timing of "distant future"). but I don't feel strongly on this opinion and don't think of it as "tradable". Just some thoughts off the top of my head.
but the rates will be the least of people's problems. The real problem will be QUALIFICATION standards. The lenders will lend, but only to people who don't need it. Absent govt intervention (which will likely happen... only god knows what monstrosity they'll concoct after the latest "housing bill" that gave money to Cerebrus and Chrysler, and a blank check to Paulson). My guess is that Fannie/Freddie fail totally and are nationalized, and can then lend to more unqualified people since profit would no longer be a concern. either that or we create another fannie/fredde/ginnie-like creature to do that instead.
Posted by: ex SF-er at August 14, 2008 4:39 AM
If you think this downturn is over, or that SF is going to "escape" it. then I would say you don't understand the implications of the above quote.
The economic forces that have dragged down subprime mortgage owner's is now affecting "alt-a" and prime mortgage owners, which SF has plenty of.
Remember it took nearly 5 years, 54 months, for the 89/90 bubble to correct, similar to the late 70's bubble. We are only 2 years into the current correction of a bubble several times larger then the 89/90 bubble.
This is all far from over.
Welcome to the next leg down.
Posted by: badlydrawnbear at August 14, 2008 8:20 AM
The housing downturn will peak in 2010-11 when the 5-year pay option ARM's that were signed in 2005-2007 start to recast and peak. As you may know, those were given out to prime borrowers -- subprime and people with lower FICO scores had recasts at 2-years -- who stretched to buy homes that were really out of their price range (but could be bought because the initial rate for 5 years was barely "doable"). The buyers believed (hoped?) that values would rise and that after 5 years they'd either i) sell the house for much more they paid or ii) refinance into an affordable loan. Those options will be off the table when the loans recast and their monthly payments soar.
To make matters worse, the option part of the loan allows people to make de minimus payments on their loans, called negative amortization. That means their loan become bigger than the original amount. When the loans hit 115%-125% of the original amount, the loans recast -- could be sooner than 5 years, of course.
Anyway, you see the problem, right? Double whammy. Total owed is growing at the same time that the value of the house is falling. When the recast happens, it will be utter destruction. Many, many prime borrowers will simply walk away in a completely rationale decision, even if they could afford to keep making payments.
Oh, and to make matters worse, if there's even higher unemployment (there will be), there's even less chance that people will not walk away when the monthly payments go up.
Remember, the peak of the recasts on the 5-year loans won't come in 2011-12 -- 5 years after 2006-07 when the buble peaked and people most egregiously overpaid -- it will come sooner because borrowers will hit that 115%-125% level and the recast will happen.
And where are these homes? In the really expensive areas where housing prices really went through the roof. Areas where people really stretched for that expensive home, believing they were missing out on a "sure thing."
Posted by: Kondratieff at August 14, 2008 9:22 AM
sounds pretty gloomy. sounds inescapable.
but let's get a little perspective. the total number of households that will find themselves in this situ (in SF) is maybe a few hundred? and of those i would bet a small percentage will actually lose their homes.
how many households do YOU SS PERMABEARS think are in this situation??
Posted by: paco at August 14, 2008 9:40 AM
a question that i have not heard addressed:
what incentive will the banks/lenders have to forclose on a depreciating asset worth substantially less than the amount they loaned in the first place?
Posted by: gloomy at August 14, 2008 11:30 AM
The incentive is to minimize the loss to the extent possible.
If banks/lenders don't ultimately foreclose when the borrower stops payment they are forgiving the entire value of the loan (i.e., 100% loss).
If they foreclose, upon the sale of the house the lender/bank will receive some repayment on the principal of the loan (i.e., total loss will be less than 100%).
The other option is a workout with the borrower, which the bank/lender would only accept if it would have a smaller loss than in a foreclosure. Workouts will not be an option with all borrowers.
Posted by: Tom at August 14, 2008 12:03 PM
My analysis pertains to the national RE situation. As far as SF, I have no idea. But, here's what I do know. If a large number of the 5-year pay option ARM's balloon, recast, and lead to owners walking away in a couple of years, the carnage in the financial system will make people's heads spin. Bank b/s's simply cannot withstand that stress.
Absent massive government/taxpayer help, hundreds and hundreds of banks will be wiped out. That will surely make getting a loan more difficult, decreasing the number of qualified buyers (lower demand), and sending prices down further. Not to mention all of the foreclosed homes coming onto the market -- higher supply. This negatve feedback loop (fewer buyers, more inventory, lower prices, more walkaways, more bank losses...) will send prices down further, causing more foreclosures, and more losses for financial institutions.
I cannot imagine SF escaping this scenario in 2-3 years if it plays out the way I hypothesized.
I do not know what the Fed has up its sleeve and whether their machinations will ameliorate this situation. I do think the recasting of these 5-year pay option ARM's is going to be much more catastrophic than the subprime portion (which we are largely finished with) of the housing downturn.
Again, my basic point -- which people don't talk about much...I don't know why -- is that we're in the eye of the storm right now, and that foreclosures will start to ramp again in 2009 and peak in 2010 or so. The eye of the storm is the difference in 2-year resets (subprime) and 5-year resets (prime). The 2-years are done and the 5-years will start next year.
Posted by: Kondratieff at August 14, 2008 12:12 PM
Speak of the devil. The Economist (UK) just published an article today about this very topic.
Posted by: Kondratieff at August 14, 2008 12:32 PM
thanks. makes sense that the lender would like to minimize the loss to whatever extent possible.
it will be interesting to see how amenable banks will be to "working something out" with the borrowers as the RE/Credit recession continues.
does anyone have any idea what the current ratio of forclosures-to-"workouts" is? 1000-to-1? 10-to-1?
Posted by: gloomy at August 14, 2008 1:31 PM
The housing downturn will peak in 2010-11 when the 5-year pay option ARM's that were signed in 2005-2007 start to recast and peak
I disagree. I believe the trough of the downturn (I think saying "peak in the downturn" is odd) will occur sooner than 2010-2011.
I do agree that we will continue to see signicant pressure on RE through December 2011 (I've said as much on here many many times). I simply disagree that the peak of pain will occur in 2011-12.
Although there is a wave of option arms that recast in the 2011-12 time period, that is only but a part of the mortgage products that were used (Most of the bad subprime has recast, the Alt A products recast mostly end-2009, the 7 and 10 year ARMs won't recast until later than 2012, and so on.)
A few reasons why I don't worry about the Option ARM's peaking in 2011/12:
1) most of those Option ARMs won't even make it to their recast date. The owners will default long before this.
2) 72% of Option ARMers are paying their minimum payment (less than interest due). they will thus hit their maximum negative amortisation prior to 2011 and their payments will reset prior to their scheduled recast. (this peaks around Jan 2010 as opposed to 2011/12)
3) there will be much much more pressure on RE in 2009 as the subprime mortgages continue to default, and we get the wave of Alt-A defaults. this may cause Federal action (nationalization of mortgages?) which may be forced PRIOR to 2011-12. the option ARMs will likely be part of whatever bailout is crafted in 2009-2010
this isn't to say that RE has a rosy future. it doesn't. it likely will be many years before we see any meaningful appreciation. it is only to say that I disagree with 2011/12 as the RE trough or the "peak in pain".
Posted by: ex SF-er at August 14, 2008 3:32 PM
these maps were posted on socketsite a few months ago. they have been updated.
Posted by: waiting for the bottom at August 14, 2008 5:12 PM
Haha. I think we agree ex-SF'er. If you re-read my post, I say (repeatedly) that the peak will happen long before the 2011-12 timeframe of the 5-year payment option loans due to borrowers hitting the neg am limit of 115%-125%. In fact, my guess is that the recasts peak in very early 2010, causing significantly higher monthly payments for huge numbers of buyers, numerous LTV's higher than 100%, numerous walkaways (though, only aftermany months of non-payment), a flood of foreclusures, destruction of bank balance sheets, inability to borrow for huge swaths of the US public, massive build-ups of housing inventories, steep drops in home values. Again, this accelerating downturn will be precipated by the 5-year pay option ARM's, which will peak in early 2010, and will climax later in 2010-2011. There's a lag time between the recasts, and the max pain. There just is.
Posted by: Kondratieff at August 18, 2008 12:16 PM
I've been told that some banks will be capable of policy shifts and or loan modifications to stem this potential tide. Rather than lose a great deal of borrowers to foreclosure, extensions can and will be granted. We'll see.
Posted by: fluj at August 18, 2008 12:26 PM
Likely just delays the true price discovery mechanism in housing. My guess is that we're going to go the Japanese route of devising "fixes" that allow banks to keep more bad loans on their books (defer their losses), rather than excise the bad stuff from their balance sheets. You know, all of this in the name of "protecting" the banking system. Unfortunately, the "fixes" in Japan only served to elongate the downturn. The idea, I think, should be to allow prices to adjust (which means accepting losses), forming a bottom and then the inevitable recovery. Ah well. I think we all see that the "accepting losses" part is, well, unacceptable to the powers that be. After all, you are talking about bankers who would like to protect banks. I understand their desire, but I remain unconvinced that having an elongated trough is all that great for the country.
As far as modifications, we have yet to see an agreed solution there. The best we've seen is states enacting unilateral moratoriums on foreclosure which, of course, does absolutely nothing to help housing values and delays true price discovery. They're just creating more future foreclosures vs. current foreclosures.
Posted by: Kondratieff at August 18, 2008 5:12 PM