by Calculated Risk on 4/16/2008 12:53:00 PM

From the Q&A: (hat tip Brian)

No stabilization in home equity portfolio:

Analyst Question: Just a question on whether you're seeing anything that would give you some signs of stabilization of the loss and delinquency rates in the home equity portfolio?

JPM: No, it's exactly what we saw, higher, more houses are going negative equity, roll rates are high, home prices we expect to still go down. We have not seen it.

And on the deteriorating prime mortgage portfolio:

Analyst: You talked a little bit about prime mortgage and that it's definitely getting worse. But I look at a 48 basis point charge-off ratio on and say that's pretty bad. From that as a base, how much worse do you envision it getting or could it get?

JPM: You know, first of all, the risk factors in prime mortgage are exactly the same as in elsewhere, which is negative home prices, high LTV, things like that, and you know, I think it will probably get a little bit worse and we probably owe you a better answer on that. We did not -- but it's hard in almost all these mortgage areas to say exactly what's going to happen to behavior. You can guess as well as we can what's going to happen to home prices. We expect it will go down another 7, 8, 9% in '08.

And on Credit Card losses:

“Outlook, 4.5 to 5% full year losses trending higher as we get through the rest of the year and probably a little bit of an effect of slowing card spend which is what we've seen in the past couple of weeks.”
all emphasis added

How many companies - from GE to JPM - have said they saw a slowdown in March?

by Calculated Risk on 4/16/2008 11:50:00 AM

San Francisco Fed President Janet Yellen repeated an earlier speech today: The Economy: Where Are We and What Will Happen Next?.

This section is worth repeating:

[R]esearch ... reveals that the single best predictor of subprime delinquency rates is the pace of house price changes. ...

The link between house prices and delinquency rates is not surprising. When house prices have been stagnant or declining, a borrower with a recent mortgage secured with a very small or no down payment has little, if any, equity in the house and, therefore, can’t rely on it to help weather income and wealth stresses like job loss, illness, or divorce. Moreover, though some borrowers may be able to afford their loans, they may decide just to walk away, if, for example, their house is worth less than their mortgage.

I should also note that, while default rates for prime loans are lower than for subprime loans, delinquency rates among all categories are highly correlated with house price declines across the country, whether borrowers are prime or nonprime, or whether loans have fixed or variable rates.
emphasis added

Now that house prices are falling rapidly in many areas, delinquency rates can be expected to increase sharply for all loan categories; prime or subprime, fixed or variable.

by Calculated Risk on 4/16/2008 11:00:00 AM

There is an article this morning in the WSJ suggesting that the LIBOR rate might be too low, possibly because banks don't want to reveal their actual borrowing costs. See: Bankers Cast Doubt On Key Rate Amid Crisis

Here is a description of the LIBOR from the BBA:

Libor stands for the London Interbank Offered Rate and is the rate of interest at which banks borrow funds from each other, in marketable size, in the London interbank market.

The WSJ article suggests that banks are actually borrowing at a slightly higher rate (one estimate is 0.3% higher), but reporting a lower rate to the BBA. This doesn't make much sense to me, since understating their borrowing costs would reduce the banks net interest margin.

The article offers two possible explanations for a manipulated rate: reputation to the bank (banks reporting higher borrowing costs might see a bank run), and banks might be earning more profits from derivatives transactions. Neither seems convincing, but it is possible.

Other measures of the liquidity crisis are definitely showing the third wave is not over.

The TED Spread from Bloomberg: The TED spread has increased to 1.62%.

Note: the TED spread is the difference between the three month T-bill and the LIBOR interest rate. Usually the TED spread is less than 0.5%. The higher the spread, the greater the perceived credit risks (compared to "risk free" treasuries).

by Calculated Risk on 4/16/2008 08:44:00 AM

The Census Bureau reports on housing Permits, Starts and Completions.

Some key points:

Housing permits in March fell sharply to 927 thousand at a seasonally adjusted annual rate (SAAR). This is the lowest since 1991.

Single family housing starts were at 680 thousand SAAR. This is also the lowest since Jan 1991.

Completions are still very high. Privately-owned housing completions in March were at 1.216 million (SAAR). Completions will probably fall to the level of starts - and this will impact construction employment.

Click on graph for larger image.

Here is a long term graph of starts and completions. Completions follow starts by about 6 to 7 months.

Privately-owned housing starts in March were at a seasonally adjusted annual rate of 947,000. This is 11.9 percent below
the revised February estimate of 1,075,000 and is 36.5 percent below the revised March 2007 rate of 1,491,000.

Single-family housing starts in March were at a rate of 680,000; this is 5.7 percent below the February figure of 721,000.

The second graph shows Single family housing starts vs. New Home sales. Single family starts also include homes built directly by owners, in addition to homes built for sale (and some other minor differences).

This graph indicates the difference between single family starts and new home sales has narrowed recently, possibly indicating: 1) that fewer homes are being built by owners, and 2) that single family starts are now low enough to begin to reduce the inventory of new homes for sales.

Another very weak report.

by Anonymous on 4/16/2008 07:44:00 AM

Kind reader AK (bless you) sent me the link to this awful Slate piece on "walking away." It's a fact-free rehashing of the increasingly popular "walkaways are the new subprime" meme, worthwhile only as a kind of crystallization of everything that is wrong with this "story." Its burden of wisdom is the simple assumption that while those subprimers couldn't afford their mortgages, the Option ARM borrowers just don't want to be underwater after their loans recast and, um, they either can or can't afford their mortgages depending on how you do or do not look at it.

What kind of logic do you expect from an essay that begins:

California is to mortgage lending what Chicago is to pork bellies.

California (the whole state) has a famous exchange on which future prices of mortgage lending are determined? Well, no:

For years, that meant it was a place with soaring house values; today, the foreclosure rate across the state is twice the national average and going up fast.

"That meant"? Home prices in California bubbled and then busted because homes were traded like commodities futures? We're two sentences into the piece here and the wheels have already come off the logical wagon. The tone is set for the lede:

California should be the poster child for a mortgage-loan bailout. In few other places have so many taken on such onerous debts with so little equity. Unfortunately, the crisis in California is going to get much worse, and there is no bailout that will solve it. Why? Because if the first stage of the foreclosure crisis was about people who could not afford their mortgages, the next stage will be about people who have every reason not even to try to pay their mortgages.

As usual, I'm amazed at how short memories are, since the "subprime crisis" was only just last year. And why just last year, when we had been originating large volumes of subprime loans for years prior to that? Well, 2007 was when those people who couldn't afford their mortgage payments suddenly lost the ability to refinance or to sell the home ahead of foreclosure. And why was that? Because home values were sliding in high-subprime-concentration markets. So maybe the price declines as of a year ago were "only" single-digits, as opposed to the "40%-50%" our intrepid Slate reporter forecasts for California. So? Underwater is underwater: if you cannot or do not wish to bring cash to the closing in order to sell your home, and you cannot maintain the payments, you end up in foreclosure, whether you're 5% under or 50% under.

Insofar as there's any sort of point here, it seems to be the unsurprising one that so-called "prime" borrowers last longer in an RE bust than subprime borrowers do. By the time prime borrowers get to the end of their ability to handle crushing mortgage payments, RE values have dropped further than they had for subprime borrowers. What else are we to make of this:

The most common subprime loans were known as "2/28" in the industry: 30 years, including a two-year teaser rate before the interest rate rose. Now these loans have reset, and we're seeing the fallout.

But prime borrowers, too, got loans that started out with low payments; if you bought or refinanced your house in the last few years, it's not unlikely that you have one. With an "option ARM" loan you have the "option" (which most borrowers happily take) of paying less than the interest; the magic of "negative amortization." The loan grows until you hit a specified point—the exact point varies with the lender; with Countrywide, it'll come after about four and a half years—when the payment resets to close to twice where it was on Day 1. . . .

The really amazing thing is that the meltdown in California is already happening and virtually none of these loans have yet reset.

Option ARM loans were heavily marketed to upper-tier home buyers in California. It's hard to know how bad the option ARM crisis will be before it actually happens, but Moe Bedard, an advocate in Southern California who advises homeowners on foreclosure and blogs about the crisis at Loansafe.org says that the difference in the time until the rate rises is the main reason that upper-middle-class Orange County (now facing foreclosures at a rate merely twice the national average) hasn't yet been hit as badly as places like Riverside.

The 2/28 "subprime" foreclosure crisis also started well before most of those loans had reset, as the failed flippers and borrowers with crushingly high DTIs even before reset were trapped by even modest drops in value. It shouldn't be surprising that Option ARM failures are beginning to occur before the payment recast.

But, as with the CBOT analogy, what's the point here? That well-heeled Orange County OA borrowers will not be able to afford doubled mortgage payments when their OAs reset, and that by then they will be underwater by 40%? This makes them "walkaways," whereas the subprime borrowers were just plain old "foreclosures"? The way I see it, we have two choices here. We can take the perspective that "we're all subprime now," or we can insist that in fact "we" aren't subprime, "we" are walkaways, and that's different.

Such "rebranding" leads inexorably to this sort of fantasy:

Consider, too, that, yes, going through a foreclosure kills your credit rating and makes it a lot harder to buy a new house—but as more and more prime borrowers go into foreclosure, it's perfectly possible that buying a new home a year later will in the near future be as routine and unsurprising as the once inconceivable idea that you can get a whole batch of new credit cards two years after a bankruptcy.

What a foreclosure and a "killing" of your credit rating does to you is make you "subprime." "Prime" is not a birthright; it is not an immutable characteristic like having blue eyes. The confident assertion that credit will be easily and quickly available to these borrowers formerly known as prime rests on a hidden assumption that they are unlike any other "subprime" borrower, and therefore will get preferential treatment in a year or two.

Mystification aside, this is a prediction that the subprime mortgage lending industry--and the investors therein--will have recovered sufficiently in just a year that this new large crop of subprime borrowers with a year-old FC on their records will be deluged with mortgage offers. Perhaps that will happen, but what makes anyone think it will happen just because these were once "prime" borrowers? Most subprime borrowers were once prime. With the exception of borrowers who have never had any credit, which is a fairly small group, subprime borrowers once had prime credit, and did not manage it well, and therefore now have cruddy credit records and FICOs. How, exactly, will these "walkaways" be any different from any other subprime borrower?

The whole thing is so nonsensical that I am forced to the conclusion that for this (and many other writers), "subprime" is code for "poor people" and "prime" is code for "middle and upper class people," hence the need for distinguishing terms for loan failure: "foreclosure" for the poor, "walkaway" for the non-poor. Foreclosure is something that happens to you against your will; "walkaway" is something you do to the bank as an exercise of control over your finances. If we can maintain these illusory distinctions, we can maintain "our" distance from "them."

In the realm of rhetoric, that is. I for one suspect that the economics of mortgage lending in a year or two will be somewhat less fantasy-driven than that.

by Calculated Risk on 4/16/2008 01:20:00 AM

From the WSJ: Merrill Upped Ante as Boom In Mortgage Bonds Fizzled On Thursday

Merrill will report $6 billion to $8 billion in new write-downs, according to a person familiar with the matter. The latest would bring its total since October to more than $30 billion ...

$30 billion here, $30 billion there ... it really is starting to add up.

by Calculated Risk on 4/15/2008 06:36:00 PM

From some stories today:

“With the traditional home buying season now well underway, we have not seen the bump in sales activity that we normally would this time of year.”
Sandy Dunn, NAHB president, April 15, 2008

The seasonal boost in sales between February and March was less than half its normal level and a record low. The weak start to the home buying season also saw another record dive in the median sales price ...
DataQuick on Southern California, April 15, 2008

"[T]here are cases where people as early as 18 to 24 months ago had one value on that property, and as they started to sell it or refinance it, they realize that valuation was 40% below what it was 18 to 24 months ago, and they're walking away from those homes in those markets."
Dowd Ritter, CEO Regions Financial Corp., April 15, 2008

March is a key month for both new and existing home sales.

Another year, another lost selling season.

by Calculated Risk on 4/15/2008 04:40:00 PM

From the WSJ: WaMu Reports $1.14 Billion Loss

Washington Mutual Inc., which has been rocked by the meltdown in the housing market, swung to a first-quarter loss of $1.14 billion amid increased provisions for credit losses.
...
WaMu on Tuesday also announced the closing of the $7 billion cash injection it announced last week.
...
WaMu's provision for credit losses soared to $3.51 billion from $234 million a year earlier. Net charge-offs, loans it doesn't think are collectable, rose to $1.37 billion from $737 million a year earlier. Non-performing assets rose to 2.87% of total assets from 1.02%.

A billion here, a billion there.

by Calculated Risk on 4/15/2008 04:20:00 PM

From the Regions Financial conference call:

Steven Alexopoulos, JP Morgan - Analyst: In terms of home equity, the second lien that went delinquent, or into default in the quarter, what were you typically doing there? Where were you writing them down, holding the judgment, or are you going through with buying out the first and going through foreclosure?

Dowd Ritter, Regions Financial Corp. - CEO & President: No, I would say basically the jump there that you see, is when we talked about valuation decline and, you know, probably anybody you talk to at any bank that lived through the late 80's and the early 90's, to me, the biggest difference in what's happening right now is, I'll use the word velocity, and it's how fast things have changed, and the biggest change that we've seen, quarter over quarter, and even if you go back and look at third to fourth and comparing them, is property valuations in certain markets. I would tell you that in a few of those that you saw us basically write off, we did not write them down, and because some of those -- they did have firsts, but there are cases where people as early as 18 to 24 months ago had one value on that property, and as they started to sell it or refinance it, they realize that valuation was 40% below what it was 18 to 24 months ago, and they're walking away from those homes in those markets.
emphasis added

Wow. The CEO of Regions is seeing 40% price declines in some markets over the last 1 1/2 to 2 years.

by Calculated Risk on 4/15/2008 02:02:00 PM

From DataQuick: Southland home sales log tepid gain; record price drop

The onset of spring did little to thaw Southern California's semi-frozen housing market: The seasonal boost in sales between February and March was less than half its normal level and a record low. The weak start to the home buying season also saw another record dive in the median sales price, the result of depreciation, slow sales for higher-priced abodes and growing sales for discounted homes fresh out of foreclosure.

A total of 12,808 new and resale houses and condos sold in Los Angeles, Riverside, San Diego, Ventura, San Bernardino and Orange counties in March. That was up 18.8 percent from 10,777 the previous month but down 41.4 percent from 21,856 in March 2007, according to DataQuick Information Systems.

Over the past 20 years Southland sales have risen by an average of 38 percent between February and March. Last month's 18.1 percent increase from February was the lowest in DataQuick's statistics, which go back to 1988.

March was the seventh consecutive month in which sales have fallen to the lowest level on record for that particular month. On average, March sales have been about twice as high - 25,407 - as last month.

Foreclosure resales - houses sold after being foreclosed on - continue to dominate many inland neighborhoods. More than one out of three Southland homes that resold last month, nearly 38 percent, had been foreclosed on at some point in the prior year. This time last year such sales were only 8 percent of the market. At the county level, foreclosure resales ranged from 28.8 percent in Los Angeles County to 56.4 percent in Riverside County.
...
The median price paid for a Southland home was $385,000 last month, the lowest since $380,000 in April 2004. Last month's median was down 5.6 percent from February's $408,000, and down a record 23.8 percent from $505,000 in February 2007. That peak median of $505,000 was reached several times last spring and summer.
...
Foreclosure activity is at record levels ...
emphasis added

Grim.