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

Kelly Evans at the WSJ presents a chart comparing pending home sales to existing home sales. The chart uses a 1.5 month lag between pending and actual sales. See WSJ: Where Is the Bottom?


The WSJ credits: "Source: Lehman Brothers; Note: latest existing home sales reading is a forecast"

The 1.5 month lag suggests that the sharp drop in contracts signed during July (the July pending home sales index was revised to a 10.7% decline), only partially impacted August existing home sales - and will also impact existing home sales in September. The same is true for contracts signed in August; we will see a portion of the impact of fewer contracts in the September existing home sales report, and the remainder in the October report.

Evans at the WSJ adds:

On the bright side, economists are hopeful that the number can only go up from here, as mortgage lenders relax a bit from August’s panic.

Maybe not. We have to remember that existing home sale activity is still above the normal level.

Click on graph for larger image.

This graph shows annual existing home sales since 1969 (and inventory levels since 1982). (Note: 2007 is the August seasonally adjusted annual rate of sales).

The current sales rate is still above normal historical levels - even if sales fall 6% to 10% in September (as suggested by the pending home sales index).


The final graph shows sales and inventory as a percent of owner occupied units. This normalizes sales for increases in population and changes in household size.

Sales would have to fall to about 4.6 million units (SAAR) to reach the median level of sales as a percent of owner occupied units for the last 35 years. So even if sales fall to 5 million or 5.2 million units in September, sales could fall further. So the hopeful view that "the number can only go up from here" is probably overly optimistic.

by Calculated Risk on 10/02/2007 12:23:00 PM

The Auto companies report September sales today.

From MarketWatch: Ford September U.S. sales fall 20.5% (hat tip REBear)

Ford Motor Co. on Tuesday posted a 20.5% drop in September U.S. sales to 189,863 cars and trucks. ... The flagship Ford F-Series truck, long the nation's best-selling vehicle, saw its sales drop 20.8% amid stiffer competition in the segment and a steep downturn in the U.S. housing market.

by Anonymous on 10/02/2007 12:17:00 PM

Because everything has gotten so boring and predictable.

Here's the quote. Your job is to identify the speaker (without Google):

``The Wall Street firms were under real pressure to supply asset-backed securities, and the Wall Street firms were pressing the lenders to give them more raw material,'' [mystery speaker] said today. ``Credit standards just went straight down, and applications for subprime mortgages soared. The consequences of that are evident.''

I'll give you a hint: it wasn't me, and it wasn't CR.

by Calculated Risk on 10/02/2007 11:08:00 AM

From the NAR: Mortgage Problems Continue to Hamper Pending Home Sales

The Pending Home Sales Index*, a forward-looking indicator, fell 6.5 percent to a reading of 85.5 from an upwardly revised 91.4 in July, based on contracts signed in August. It was 21.5 percent below the August 2006 index of 108.9.

This follows the 12% decline in the index for July.

by Anonymous on 10/02/2007 11:06:00 AM

No longer stunned and surpised, we are now overwhelmed with ennui:

The September remittance reports revealed that conditions in the mortgage market are worsening, although the rate of credit performance dislocation is mostly slowing, according to a UBS report.

Delinquency rates in the buckets - 30, 60+, FC+REO - have increased, according to UBS's data. However, while the delinquency increases for FC+REO continued unabated, the rate of delinquency increases for all 30DQ and three out of four 60+DQ indices have eased. UBS reported a "somewhat surprising" 1.83% increase in FC+REO rates among ABX07-1 deals.

Meanwhile, ABX06-1 deals saw an increase of 1.43%, which UBS analysts credited to the impact of resets and reduction in pool factors. Cumulative losses jumped to 13 basis points from 10 basis points on all indices except ABX07-02.

The September data, with their increased delinquencies and foreclosures, do not come as a surprise, as the subprime market remains unstable.

"Our whole stance on this entire debacle right now is that it's so perfectly predictable that it's boring," said Michael Bykhovsky, CEO of Applied Financial Technology. "If we feed current [Home Price Indices] and projected HPIs into the model, the resulting delinquency output is very much consistent with what we are observing."

So, um, the models really work after all? Now that is boring.

by Anonymous on 10/02/2007 10:53:00 AM

Forgive me. I slept until nearly 8:00 am today, and then my PC got abducted by aliens. I am a drive-by victim of this "anti-virus software" scam foisted on me by software developers who will not allow me to take enough risk. You know.

Anyway, USA Today is always good for a laugh:

David Brannan, 44, of Monroe, N.C., is co-founder of a software company that's been in business since 1989. He and his wife have owned their home for 18 years and are in the process of buying a new, custom-built home. Brannan has an excellent credit score.
So he was stunned last week when CitiMortgage (C), which just a week earlier had said everything was in good shape, sent a letter saying his mortgage application had been rejected. It suggested he consider credit counseling.

Brannan called Citi, which told him his income for the past two years wasn't enough for the size of the loan. He says Citi refused to include profit distributions from his company that account for more than half his income. (CitiMortgage declined to discuss Brannan's application. But spokesman Mark Rodgers says the company will restructure or decline a preapproved loan if it can't sufficiently verify information from a borrower.)

Brannan belongs to a group that's become a kind of drive-by victim of the mortgage industry crisis: the millions of Americans who are self-employed.

Yes, those people who custom-build homes and don't have permanent financing lined up yet. We see a lot of that.

This part is nice, though:

McNamee says lenders that are still catering to the self-employed have been flooded with business, which means it might take longer to process a loan. That can be difficult for some successful self-employed borrowers to accept. "When you're talking about CEOs of companies, they want an answer, and they want it now," McNamee says.

But hurrying the process, he says, could increase the cost of the loan. Even for well-off business owners, qualifying for a mortgage is "not that smooth, easy no-brainer like it used to be. If you want it to be quick, you're paying a higher price."

Anyway, feel free to discuss pending home sales until we hear what happened to CR this morning.

by Anonymous on 10/01/2007 05:06:00 PM

So there I was, minding my own business, when a press release with the title "MBA Study Examines Fraud Committed Against Mortgage Lenders" shows up in my inbox. How very interesting, right?

Well, the MBA is lucky that the terms "study" and "examine" aren't defined in any law I know of, or I'd be tempted to sue. It's no less than a 20-page lobbying effort that could have been a three-pager, given the redundancies, repetitions, white space and giant margins. Its burden of wisdom is that mortgage fraud is on the rise, mortgage lenders are victims, not perpetrators, of mortgage fraud (insider fraud being a case of "a lending institution [being] deliberately deceived by . . . one of its own employees"), that fraud is clearly conceptually and practically distinct from predatory lending, and that no new legislation battling mortgage fraud is necessary, as current laws are more than sufficient. Its recommendation for action: to increase enforcement and information-sharing, all at the taxpayer's expense.

There is no "examination" of fraud--how it happens, how it goes undetected, what industry practices might enable it. There is, therefore, no examination of what the industry could do, on its own dime, to prevent fraud. Even better, the "study" keeps hammering the point that fraud and predatory lending are separate, for the purpose of making sure that no new anti-fraud legislation that also includes anti-predation components gets passed. That's the whole agenda in this little piece of special pleading, which uses the term "so-called predatory lending" several times:

[S. 1222] inappropriately conflates mortgage fraud with predatory lending. Indeed, several of the provisions have little, if anything, to do with mortgage fraud as that term is understood by law enforcement officials and the mortgage industry. For example, S. 1222 would:

• Impose heightened foreclosure requirements on subprime loans containing a variety of terms;
• Create assignee liability in vague and undefined cases of “deceptive practices” — a term that, in context, appears to mean something different than “fraud”; and
• Require the provision of housing counseling services to borrowers regarding “any other activities or practices that… are likely to increase the risk of foreclosure by such individuals” without providing any guidance as to what such “other activities or practices” may be.54

Whether one believes such provisions have merit as a matter of public policy, they are not directly related to mortgage fraud. Instead, these provisions clearly are intended to address concerns related to “predatory” lending.

Mortgage fraud and predatory lending differ in many important respects in terms of the actions, methods and targets involved. As discussed above, mortgage fraud, as the term is understood by federal law enforcement officials and the mortgage industry, is the intentional enticement of a financial entity to make, buy or insure a mortgage loan when it would not otherwise have done so, had it possessed correct information. In contrast, predatory lending is an undefined term that generally describes negative practices that are harmful to consumers. Clear definitional boundaries around the term predatory lending have yet to be drawn. Because the actions and targets of mortgage fraud and predatory lending differ, actions taken to remedy one rarely, if ever, will remedy the other. Conflating the two creates the danger that solutions appropriate only to one will be applied to both. While there are actions federal law makers can take to address each, the numerous and essential differences between them make their conflation, as well as their simultaneous treatment, inappropriate.

In other words, the MBA wants you to believe that fraud is only a matter of the industry being defrauded, that mortgage fraud is conceptually, practically, and legally distinguishable from predation, and that no bill that tries to clamp down on both things at the same time is acceptable. You get the impression that the industry is so frightened by anti-predation laws that it simply will not accept them even if they provide for more penalties or easier enforcement of fraud against lenders.

There's a lot of hand-wringing over "unintended consequences" of legislation in here--which you may or may not find compelling. There is no attempt to address the opposite problem, of refusing to regulate at all because regulation might not be perfect. What's truly amazing is that the MBA argues that mortgage fraud is actually on the rise--not just reports of fraud, and not, apparently, just proportionally to the increase in mortgages originated in recent years--but then argues that existing laws are sufficient, requiring only more taxpayer dollars poured into the Justice Department for task forces and databases. There isn't even a perfunctory recognition that industry practices, like "no docs," third-party originations, appraisal ordering practices, stripped-down internal controls, could have anything to do with any of this.

That might have something to do with the fact that the "study" keeps insisting that there is no widely-agreed definition of predatory lending, and that predation has nothing to do with fraud. I have argued before that during the bubble, lenders were happy to accept fraud as the "cost of doing business" with practices that were reckless but that threw off tons of money, like no-doc no-down loans, brokered and outsourced processes, skimpy due diligence, and incestuous and conflict-riddled "affiliated business arrangements." In fact, I gather from this MBA effort that the industry is still happy to pay the fraud toll; what has it worried is that anti-predation legislation will chip away at those money-spinners.

The MBA quotes the FBI as estimating that industry-wide fraud costs for 2006 ranged from $946 million to $4.2 billion. That's quite a range, and I frankly am deeply suspcious of those numbers. For one thing, they are based on Suspicious Activity Reports (SARS), which have the problem of multiple-counting (several categories of suspicious activity can be reported for a single loan application) and also that the dollar amount reported is generally the transaction amount, not an actual, after-the-investigation-trial-and-conviction assessment of the actual fraud loss. But even taking that $4.2 billion number seriously means that fraud costs for 2006 were a whopping 18 bps on 2006 gross mortgage production (~2.37 trillion). Could it be that the MBA doesn't want to define predatory lending because it doesn't want to address how revenues on grossly overpriced mortgage loans and reliance on fly-by-night brokers might function to offset fraud costs?

Shorter MBA: We're willing to invite fraudulent behavior and pay for it as long as you let us continue to prey. But we'll help out by asking the taxpayers to fund some "task forces" if you need us to appear to be doing something about it.

by Calculated Risk on 10/01/2007 03:56:00 PM

From Mathew Padilla at the O.C. Register: O.C. office rents, vacancies spike

A surge in office construction combined with sweeping layoffs at mortgage companies lead to both higher office vacancy and higher rents in O.C. in the third quarter ...

The county's office vacancy rate hit 10.53 percent at the end of last month, the highest in nearly three years ... its previous peak was 17.2 percent in March 2002, amid the last construction wave.

In all, developers have finished work on 3.5 million square feet of offices so far this year with another 1 million coming in Q4 ...
...
As for mortgage companies, they occupy about 4 million square feet of office space in the county, or roughly 4 percent of larger buildings. The total is down 30 percent from 6 million in mid 2005 ...

Orange County is home to many of the mortgage companies that have closed (like New Century). So the rising vacancy rate is no surprise. However I also expect to see rising vacancy rates nationwide, especially for office and retail space.

by Calculated Risk on 10/01/2007 11:40:00 AM

Homebuilders are once again seeing rising cancellation rates. KB Home reported Thursday:

The company posted a cancellation rate of 50 percent during the most recent quarter, down from 60 percent in the year-ago quarter but well above the 34 percent rate in the second quarter of this year.

This is well above the normal range. According to D.R. Horton, the normal range for cancellations (for Horton) is between 16% and 20%. For those trying to analyze the housing market, this means that the inventory levels reported by the Census Bureau are probably too low right now.

The Census Bureau, during periods of rising cancellation rates, overstates New Home sales and understates the increase in inventory. Conversely, during periods of declining cancellation rates, the Census Bureau understates sales. Here is discussion from the the Census Bureau on cancellations. Note: this shouldn't be confused with revisions that are unrelated to cancellations.

Using cancellation rates from several of the publicly traded home builders, we can estimate the actual new home inventory (as opposed to the inventory reported by the Census Bureau). Note: The Census Bureau breaks down the inventory as Completed, Under Construction, and Not Started. The following chart show the reported and cancellation adjusted inventory levels for the hard inventory (excluding the "Not Started" category).

Click on graph for larger image.

At the end of Q2, this analysis shows the Census Bureau is currently understating the hard inventory of new home sales by about 77,000 units. Even though the Census Bureau has shown a slight decline in new home inventory in the third quarter (through August), I expect the adjusted inventory to increase because of rising cancellation rates.

This will be important to follow later in the housing cycle. However, it isn't just the inventory of new homes for sale that will impact the homebuilders. Existing homes are a competing product for new homes, and the record inventory of existing homes for sale will also pressure home-building activity.


This graph shows the year end inventory levels since 1982 for new and existing homes. (2007 numbers are for August).

Inventory levels are at an all time record of 5.1 million units.

The third graph shows the total inventory normalized by the number of owner occupied units (this adjusts inventory for increases in population and household size).


Total inventory is currently 6.8% of the total owner occupied units in the U.S. This is far above the previous peak of 4.7% in the early '80s.

Finally, it's not just the level of inventory that matters, but also the level of distressed inventory. We are already seeing record levels of foreclosures in some states, and IMO it is about to get much worse. I spoke with one of the top agents in San Diego this weekend, and she was analyzing one neighborhood for a client in the $375K to $450K price range. There were 70 listings (very high for that neighborhood and price range), but she was shocked to find that approximately 75% of the listings were short sales, and a similar percentage were vacant. This suggests a flood of REOs in the coming months.

by Calculated Risk on 10/01/2007 10:16:00 AM

From MarketWatch: Big write-downs to slash Citi's quarterly net 60% (hat tip Lyle)

Citigroup Inc said Monday it expects ... huge write-downs for unsold debt it issued to finance corporate takeovers and for big losses on the value of subprime mortgage-backed securities.
...
The decline "was driven primarily by weak performance in fixed-income credit-market activities, write-downs in leveraged loan commitments, and increases in consumer-credit costs," Chairman and Chief Executive Charles Prince said in a statement.

Earlier Monday, Swiss banking giant UBS said it will take a hit of 4 billion Swiss francs ($3.4 billion) in the third quarter from its subprime mortgage exposure and plans sweeping management changes and job cuts at its investment-banking division.

Citi sees a write-down of $1.4 billion pretax, net of underwriting fees, on funded and unfunded loans for clients doing leveraged buyouts.
...
Citi also cut the value of its mortgage-related positions, as rival Wall Street investment banks did last month.

It said it expects losses of $1.3 billion pretax, net of hedges, on the value of subprime mortgage-backed securities warehoused for certain securitizations, and $600 million pretax in fixed-income credit trading due to significant market volatility and the disruption of historical pricing relationships.

I guess the music has stopped.

“As long as the music is playing, you’ve got to get up and dance. When the music stops, in terms of liquidity, things will be complicated.”
Chairman and Chief Executive Charles Prince, July 2007