by Calculated Risk on 12/28/2007 06:57:00 PM

From Bloomberg: Legg Mason Shores Up Cash Funds With $1.12 Billion

Legg Mason Inc. pumped $1.12 billion into two non-U.S. cash funds to prevent losses, the biggest bailout by a money manager tied to asset-backed debt sold by structured investment vehicles.

The move, along with an earlier cash infusion, will reduce earnings per share by 15 cents in the quarter ending Dec. 31, the Baltimore-based company said today in a statement.

The Confessional is still open.

by Calculated Risk on 12/28/2007 03:40:00 PM

Update: added comments from Tanta at bottom.

Several recent articles have referenced my post on the possibility of a change in attitude towards foreclosure. I wrote:

One of the greatest fears for lenders (and investors in mortgage backed securities) is that it will become socially acceptable for upside down middle class Americans to walk away from their homes.

Although the purpose of the post was to discuss changing social norms, the primary reason the post was mentioned was because of some pretty big numbers, like the possibility of $1 trillion in mortgage losses based on a few rough assumptions. A couple of quotes:

Australia theage.com: Americans 'walk' from loans

Late last week ... Calculated Risk figured that housing losses might reach $US1 trillion and even $US2 trillion.

WSJ: How High Will Subprime Losses Go?

Back in the U.S., the Calculated Risk blog sidestepped the colorful language and went straight for the big number: “The losses for the lenders and investors might well be over $1 trillion.”

The $1 trillion number is a simple calculation: If prices fall 30%, then approximately 20 million U.S. homeowners will be upside down on their mortgages. If half of these homeowners walk away from their homes, with an average 50% loss to lenders and investors, that is $1 trillion in losses (the average mortgage is just over $200K).

This wasn't a forecast, just a simple exercise to show why changing social norms is very scary for the lenders.

And this isn't just a subprime problem, and these aren't just potential "subprime losses". In the original post I referred to the potential of foreclosures becoming "socially acceptable" for the middle class.

But even though the problems have spread far beyond suprime, some reporters are stuck on the subprime meme. As an example, Bloomberg columnist John M. Berry wrote: Subprime Losses Are Big, Exaggerated by Some

As the U.S. savings and loan crisis worsened in the 1980s, analysts tried to top each other's estimates of the debacle's cost to the federal government.

Much the same thing is happening now with losses linked to subprime mortgages, with figures of $300 billion to $400 billion being bandied about.

A more realistic amount is probably half or less than those exaggerated projections -- say $150 billion. That's hardly chicken feed, though not nearly enough to sink the U.S. economy.

A loss of $150 billion would be less than 12 percent of the approximately $1.3 trillion in subprime mortgages outstanding. About $800 billion of those are adjustable-rate mortgages, the remainder fixed rate.

This is subprime reporting. It's absurd to think that mortgage losses will only come from subprime loans; in fact most of the upside down homeowners will be Alt-A and prime borrowers. By focusing solely on subprime, Berry misses the larger mortgage problem.

And the credit problems extend well beyond mortgages. As an example, from Bloomberg this morning: Citigroup, Goldman Cut LBO Backlog With 10% Discounts

Citigroup Inc., Goldman Sachs Group Inc., Morgan Stanley and JPMorgan Chase & Co. are offering discounts of as much as 10 cents on the dollar to clear a $231 billion backlog of high-yield bonds and loans.

We don't know the exact haircut on each LBO deal, but an average 5% haircut would be over $10 billion in losses. And there are losses coming from corporate debt, CRE loans, and other consumer debt too. As Floyd Norris asked at the NY Times this morning: Credit Crisis? Just Wait for a Replay

"... just how different was subprime lending from other lending in the days of easy money that prevailed until this summer?"

Short answer: not very.

Back to mortgage losses, both Felix Salmon: Are Subprime Losses Being Exaggerated? and Paul Krugman, Jingle mail, jingle mail, jingle mail — eek! bring up a key issue: recourse vs. non-recourse loans. Felix writes:

... no one is going to have a real handle on mortgage losses unless and until someone manages to get a handle on the percentage of mortgage loans which are non-recourse. If your house falls in value and you have a non-recourse mortgage, then it makes perfect economic sense for someone in a negative-equity situation to simply walk away – something known as "jingle mail". But given the amount of refinancing going on during the last few years of the mortgage boom, I suspect that the vast majority of mortgages are not non-recourse. (Refis are never non-recourse.)

This is an important issue. In California, purchase money is non-recourse. If the borrower walks away and mails in the keys (Fleckenstein's "jingle mail"), the lender is stuck with the collateral. However, if the California borrower refinanced, then the lender has recourse, and can pursue a judicial foreclosure (as opposed to a trustee's sale), and seek a deficiency judgment.

The lender can enforce that deficiency judgment by attaching other assets, or by garnishing the borrower's wages. Historically lenders rarely pursued (or enforced) deficiency judgments, but that could change if many middle class borrowers, with solid jobs and assets, resort to jingle mail.

For purchase money, state law determines the recourse vs. non-recourse issue. As Felix noted, refis are always recourse, and there was significant refi activity in recent years.

But before we think most loans are recourse, we have to remember that about 22.3 million homes were purchased during the last 3 years (2005 through 2007), and in a period of rising rates, many of these homeowners probably did not refi. It's difficult to estimate the exact losses - since we don't know the percent of recourse loans, we don't know how far prices will fall, and we don't know how homeowners will react to being upside down - but we know the losses will be significant.

Note: I'm trying to find a state by state list of recourse vs. non-recourse for purchase money.

On recourse loans, Tanta adds (via email):

I suspect that you will have some very aggressive lenders and some not very aggressive lenders in that respect.

Which will tell you who thinks it doesn't have a fraud problem and who does.

Back in my day working for a servicer, we never went after a borrower unless we thought the borrower defrauded us, willfully junked the property, or something like that. If it was just a nasty RE downturn, it rarely even made economic sense to do judicial FCs just to get a judgment the borrower was unlikely to able to pay. You could save so much time and money doing a non-judicial FC (if the state allowed it) that it was worth skipping the deficiency. Plus we had a soft spot, I guess.

At any rate, the absolute all time last possible thing you could get me to do is send an attorney barging into court demanding a deficiency judgment if I had any reason whatsoever to fear that my own effing loan officer was implicated in fraud on the original loan application. Any borrower with half a brain will raise that as a defense, and any judge even slightly awake will not only deny the deficiency but probably make the lender pay all costs, or worse. And I'd call that justice.

This goes double if the loan was made as a "stated asset" deal. I can just hear a judge asking me why I deserve to get other assets now when I never bothered to make sure the borrower had any in the beginning.

So yes, some people will get hit with a deficiency. But I'm not sure it will be as many as you might think. T

by Calculated Risk on 12/28/2007 01:57:00 PM

Dr. Christopher Thornberg writes in the LA Times: Realty reality: Housing prices are headed way down

In 2002, the median price of a single-family home in Los Angeles was $270,000 and the median homeowner's income was $65,000. With a $50,000 down payment, the annual cost of that house (taxes, insurance and payment on a 30-year fixed-rate conventional mortgage) would add up to about 33% of the median household's income -- just under the 35% mark that the Federal Housing Administration calls the upper limit of "affordable."

By 2006, the cost of that same house doubled, to $540,000 -- pushed by unbridled speculation fueled by unparalleled access to mortgage capital. But median income rose a paltry 15%. So today that same set of costs come to 60% of gross income.

... Prices must and will fall. Everywhere. Probably 25% to 30% from their peak.

I believe Thornberg's price forecast is for Los Angeles. It appears his "everywhere" comment is referring to all the neighborhoods in LA.

by Calculated Risk on 12/28/2007 11:01:00 AM

First, a couple of key points to consider on housing.

Note: For more graphs, please see my earlier post: November New Home Sales

Let's start with revisions. This month (November) is one of the few months were the initial report wasn't higher than the previous month. Usually the small reported gain in sales is then revised away in subsequent releases.

Look at the report today (for November), the Census Bureau revised down sales for August, September and October. This has been the pattern for most of the housing bust; almost all the revisions have been down. I believe the Census Bureau is doing a good job, but the users of the data need to understand what is happening (during down trends, the Census Bureau initially overestimates sales).

For an analysis on Census Bureau revisions, see the bottom of this post.

Next up, inventory. The Census Bureau reported that inventory was 505 thousand units. But this excludes the impact of cancellations. Currently the inventory of new homes is understated by about 100K (See this post for an analysis of the impact of cancellations on inventory).

This also means that the months of supply is understated. The Census Bureau reported the months of supply as 9.3 months. After adjusting for the impact of cancellations, the actual months of supply is probably closer to 11.3 months.

Click on graph for larger image.

This graph shows New Home Sales vs. Recession for the last 35 years. New Home sales were falling prior to every recession, with the exception of the business investment led recession of 2001.

This is what we call Cliff Diving!

And this shows why so many economists are concerned about a possible consumer led recession - possibly starting right now.

The second graph compares annual New Home Sales vs. Not Seasonally Adjusted (NSA) New Home Sales through November.


Typically, for an average year, about 93% of all new home sales happen before the end of November. Therefore the scale on the right is set to 93% of the left scale.

Through November, there have been 732 thousand New Home sales, and, with one month to go (and a few revisions) it looks like New Home sales for 2007 will be around 775 thousand - the lowest level since 1996 (758K in '96).

by Calculated Risk on 12/28/2007 10:00:00 AM

According to the Census Bureau report, New Home Sales in November were at a seasonally adjusted annual rate of 647 thousand. Sales for October were revised down to 711 thousand, from 728 thousand. Numbers for August and September were also revised down.

	Click on Graph for larger image. Sales of new one-family houses in November 2007 were at a seasonally adjusted annual rate of 647,000 ... This is 9.0 percent below the revised October rate of 711,000 and is 34.4 percent below the November 2006 estimate of 987,000.
	The Not Seasonally Adjusted monthly rate was 46,000 New Homes sold. There were 71,000 New Homes sold in November 2006. November '07 sales were the lowest November since 1995 (46,000).
	The median and average sales prices are generally declining. Caution should be used when analyzing monthly price changes since prices are heavily revised and do not include builder incentives. The median sales price of new houses sold in November 2007 was $239,100; the average sales price was $293,300.
	The seasonally adjusted estimate of new houses for sale at the end of November was 505,000. The 505,000 units of inventory is slightly below the levels of the last year. Inventory numbers from the Census Bureau do not include cancellations - and cancellations are once again at record levels. Actual New Home inventories are probably much higher than reported - my estimate is about 100K higher.
	This represents a supply of 9.3 months at the current sales rate.

This is another VERY weak report for New Home sales. All revisions continue to be down. This is the fourth report after the start of the credit turmoil, and, as expected, the sales numbers are very poor.

I expect these numbers to be revised down too. More later today on New Home Sales.

by Tanta on 12/28/2007 08:45:00 AM

A quite decent piece on Option ARMs in the LA Times. I liked this part:

Despite such risks, the initial low payments on option ARMs have kept a lid on serious delinquencies -- 3.7% of all option ARMs, Standard & Poor's analysts said in a report last week. That's higher than before, but still low compared with the 6.3% delinquency rate on loans to good-credit borrowers with so-called hybrid ARMs, which have a low fixed rate for two to 10 years before becoming adjustable-rate loans.

At Calabasas-based Countrywide Financial Corp., which S&P said made about a quarter of all option ARMs last year, 3% of such loans held by the lender as investments were delinquent at least 90 days, up tenfold from 0.3% a year earlier. Delinquency rates were even higher on option ARMs from other lenders, including Pasadena-based IndyMac Bancorp and Seattle's Washington Mutual Inc., S&P said.

Countrywide and other lenders tightened their lending standards last summer to ensure borrowers could afford loans after the interest rates adjusted upward.

Had those guidelines been in effect previously, Countrywide recently said, it would have rejected 89% of the option ARM loans it made in 2006, amounting to $64 billion, and $74 billion, or 83%, of those it made in 2005.

I made an argument a while ago that focussing regulator attention exclusively on disclosure documents can be just a touch beside the point if lenders are no longer offering the product in question. You have to wonder, if we just cut off 80-90% of the OA borrower pool, whether the remaining 10% really need those new and improved disclosures, or can muddle along with the ones already in use. If you take the OA out of the mass market and put it back into the high-net worth, high-income crowd it was originally designed for, you might find that your borrowers are already selected to be people who either read and understand disclosures, or who hire an attorney or financial planner to read them. I can certainly think of better uses for regulators' time and energy than fooling around with disclosure documents that would be clear to borrowers who are now in a rather different kind of trouble than not understanding the teaser rate on their OA.

The other thing to notice is that the obligatory example borrower supplied in the article is having trouble with her first payment increase (the typical 7.5% annual increase in the minimum payment), which is still not enough to cover all the interest due. As that sort of situation increases (as more and more 2005-2006 vintage OAs get to their second or third payment increase), we'll start seeing defaults long before the recast date.

Speaking of which, when I am not making cartoon pigs I have been creating some spreadsheets to show examples of how to project the recast date on Option ARMs. That's total and compleat Nerd territory, but if anyone is interested I'll post them (as spreadsheets or as images thereof). You tell me whether that's more detail than you can stand or not.

by Tanta on 12/28/2007 08:14:00 AM

Floyd Norris, New York Times, December 28, 2007:

What if it’s not just subprime?

Gee Willikers. If it's not just subprime, there might be enough material in it to, oh, fill a whole blog. Or several dozen of them.

by Calculated Risk on 12/27/2007 02:56:00 PM

Update: from Reuters: Moody's may cut Technical Olympic deeper into junk (hat tip Mike)

From Bloomberg (no link yet, hat tip Brian): TOUSA, Florida Homebuilder, May See Rating Reduction

TOUSA ... will see its debt rating slashed to default status if it fails to make Jan. 1 interest payments on senior notes, Standard & Poor's said today.
...
TOUSA is considering several restructuring options, including a potential Chapter 11 filing, S&P said.

TOUSA was the #13 largest builder in 2006 according to BuilderOnline. I believe the largest home builder to go bankrupt so far was #50 Levitt & Sons. Here is the BuilderOnline top 100.

by Calculated Risk on 12/27/2007 01:02:00 PM

From the Fed weekly report on commercial paper this morning, here is the discount rate spread:

Click on graph for larger image.

According to the Fed, the discount rate spread is 145 bps. This graph was released this morning.

Here is a simple explanation of this chart: This is the spread between high and low quality 30 day nonfinancial commercial paper.

What is commercial paper (CP)? This is short term paper - less than 9 months, but usually much shorter duration like 30 days - that is issued by companies to finance short term needs. Many companies issue CP, and for most of these companies the risk of default is close to zero (think companies like GE or Coke). This is the high quality CP. Here is a good description.

Lower rated companies also issues CP and this is the A2/P2 rating. This doesn't include the Asset Backed CP.

The spread between the A2/P2 and AA paper shows the concern of default for the A2/P2 paper. Right now the spread is indicating that the fear of default is very high. Higher than in August. And higher than after 9/11.

by Calculated Risk on 12/27/2007 10:45:00 AM

It's been some time since I graphed the weekly claims numbers.

From the Department of Labor:

In the week ending Dec. 22, the advance figure for seasonally adjusted initial claims was 349,000, an increase of 1,000 from the previous week's revised figure of 348,000. The 4-week moving average was 342,500, a decrease of 1,000 from the previous week's revised average of 343,500.

Click on graph for larger image.

This graph shows the weekly claims and the four week moving average of weekly unemployment claims since 1989. The four week moving average has been trending upwards for the last few months, and the level is now approaching the possible recession level (approximately 350K).

Labor related gauges are at best coincident indicators, and this indicator suggests the economy is close to recession.