by Calculated Risk on 5/20/2007 11:34:00 PM

Another day, another deal. The WSJ reports: TPG, Goldman Acquire Alltel

TPG Capital LLP and the private-equity arm of Goldman Sachs Group Inc. Sunday night agreed to purchase wireless operator Alltel Corp. for about $27.5 billion, in the largest foray yet of private-equity money into the wireless business.

Research has shown that "merger waves", especially M&A activity involving stock transactions, happen during periods of market overvaluation (see Rhodes-Kropf and Viswanatan, Market Valuation and Merger Waves). The NY Times quoted Professor Rhodes-Kropf last December:

“Periods of high relative valuation are nearly always associated with high M.& A. activity, and the stock market has fallen after each major merger wave.”

Of course much of the current M&A activity is private equity LBOs using OPM (Other People's Money). Greg Ip writes in Friday's WSJ: Fed, Other Regulators Turn Attention to Risk In Banks' LBO Lending

Banks have been major players in the surge of takeovers, both as lenders to and investors in buyout targets. ... Regulators see signs of that in LBO lending, particularly in what is known as "bridge" financing, or the temporary credit that serves as a stopgap between the buyout and longer-term financing.

LBO loan volume hit $121 billion last year, compared with $31 billion in 1998, the peak of the previous cycle, according to Standard & Poor's Leveraged Commentary & Data. Volume this year has reached $88 billion, more than double the year-earlier period. Meanwhile, interest-rate spreads have fallen to their lowest levels ever, and loan restrictions have been loosened.

"There are some significant risks associated with the financing of private equity, including bridge loans, [and] we are looking at that," Federal Reserve Chairman Ben Bernanke said in response to questions at a Chicago conference [last Thursday].

Rhodes-Kropf addressed this issue in the NY Times article:

Professor Rhodes-Kropf cautions stock-market investors not to take solace in that difference [between stock and debt financing]. “To the extent the current merger wave reflects an overvalued debt market, it stands to reason that it will eventually correct — just as overvalued stock markets eventually correct,” he said. “And it can’t be good news for the stock market if money is destined to become much tighter in coming years.”

And Ip describes a prior case of "bridge" financing gone bad in the WSJ:

In a famous event dubbed the "Burning Bed," First Boston Corp. in 1989 made a $457 million bridge loan to the purchasers of Ohio Mattress. When the junk-bond market collapsed soon afterward, First Boston couldn't refinance the loan and ended up owning most of Ohio Mattress. Credit Suisse had to inject additional capital into First Boston, culminating in a full takeover.

I try not to comment on the stock market on this blog (just the economy), and this isn't to say I expect an imminent market correction in stocks or bonds. In fact the "merger wave" might still have legs. China just took a $3 Billion position in private equity firm Blackstone, so maybe there will be plenty more money for M&A investments. See the Financial Times: Beijing to buy stake in Blackstone

Yet, for some reason, all this activity reminds me of the ill-fated Time Warner-AOL merger that happened in January 2000. That merger never made sense to me. In this case, if debt is so "cheap", why are these LBOs using so much bridge financing?

by Calculated Risk on 5/20/2007 06:55:00 PM

From the O.C. Register: Laguna digs get top bid at foreclosure auction

Winning bids on the 15 properties in Orange County ranged from $225,000 on a Garden Grove condo previously valued at $299,000 – a 25 percent discount – to [$1.95 million on the Laguna home, a 30% discount from the previous sales price].

Many were 15 percent to 20 percent below the previous value, but the bid on a Santa Ana condo valued at $255,000 came in at $250,000, only a 2 percent savings.

This doesn't seem like much of a price discount for a foreclosed property. But this is just the beginning:

As recently as a year ago in April, there were only 22 foreclosures in Orange County, according to DataQuick Information Systems. That rose to 234 this April, but is nowhere close to the peak of 674 foreclosures in October 1996.

... lenders are beginning to turn over their properties to companies like Irvine-based Real Estate Disposition Corp., which oversaw Saturday's foreclosure auction. The three auctions the company is holding this month in San Diego, Los Angeles and Riverside are the first since the recession of the mid-1990s ...

As these auctions become more common, the price discounts, as compared to previous sale prices, will probably increase.

by Anonymous on 5/20/2007 08:22:00 AM

Bloomberg reports (thanks, risk capital!) on subpoenas arriving at the office of a large New York appraisal firm:

May 18 (Bloomberg) -- New York State is investigating Manhattan real estate practices, seeking information about whether brokers pressured appraisers to inflate property values as prices doubled in the last five years.

Attorney General Andrew Cuomo issued a subpoena to Manhattan appraiser Mitchell, Maxwell & Jackson Inc., the company said. Manhattan Mortgage Co., a broker, also received a subpoena, Chief Executive Officer Melissa Cohn said. . . .

Y. David Scharf, an attorney at New York law firm Morrison Cohen LLP, who is representing Mitchell, Maxwell & Jackson, said his client has been told it's not a target of the investigation.

``The information that is being requested is whether or not pressure has been brought to bear on appraisers to change their appraisals,'' Scharf said. The firm is ``continuing to gather information'' in response to the subpoena, he said.

``We did not change appraisals in any circumstances,'' he said.

Oh, bummer. Those appraisers who didn't cooperate when we needed a better value are now cooperating with the Attorney General. I have exactly zero inside dope about this, but personally I'd be shocked that any appraisal firm anywhere at any time couldn't cough up at least a few e-mails and billing records that provide evidence of pressure to inflate the value. I have somewhat less confidence in the number of firms who can back up the claim "we did not change appraisals in any circumstances," but that's not the point, is it? The point is that he who gets on the witness list gets to make certain claims that he who is on the suspect list might live to regret having made. This ought to get interesting.

In other news, Steve Jakubowski of the Bankruptcy Litigation Blog kindly directed my attention to this follow-up to my unsubtle request back in April for a look at the complaint in the matter of Bankers Life Insurance Co. v. Credit Suisse First Boston Corp., et. al. It's a doozy. Steve is a bit skeptical of Bankers Life's legal grounds--and he's a Real Lawyer™. I did a fair amount of eye-rolling over the nitties and gritties of the alleged wrongdoing, although I am a mere Mortgage Punk™. (Note that the suit involves a purchase in 2004 of a couple subordinated tranches of a security backed by loans originated in 2001. Until the 2005-2006 vintage came around, 2001 was shaping up to be the Worst. Vintage. Ever. Also, Bankers paid 104.75 for one of them.) Steve promises to keep us all updated on the matter.

Until then, we are left with the strangled cries of lawyers in love.

UPDATE: Oh for Peat's sake. I come up with an excuse to provide you with a bonus rock video, and then forget to include it in the post. You can thank me later.

by Anonymous on 5/19/2007 02:50:00 PM

Because it's always a good time to pick on brokers . . .

I am just a broker and my story’s seldom told
I have squandered your down payment for a pocketful of mumbles, such are yield spreads
All lies and jest, still the bank hears what it wants to hear
And disregards the rest, hmmmm

When I hocked your home and your cash back deal,
I was no more than a pimp in the company of bankers
In the blowing of the lending bubble, runnin’ nude
Feeding low, seeking out the toxic products, where the naive people go
Looking for the spiff disclosures wouldn’t show.

Lie la lie, lie la lie lie liar loan,
Lie la lie, lie la lie lie lie lie lie lie liar loan...

Asking only five points back end, I come lookin’ for a mark, but I get no offers
Just a warehouse from the whores on 7th Avenue
I’m tellin’ you, there were times when I was so greedy
I took some front fees, too, la la la la la la la

And I’m laying out my guideline changes, wishing I was gone, goin’ home
Where the New York City traders aren’t bleedin’ me, leadin’ me to take your home

In the hearing stands a broker, and a sleazeball by his trade
And he carries the recaptures of every loan that put back or haircut him
Til he cried out in his anger and his shame:
“I am tightening, I am tightening, but the funder still remains,” hmmmmm . . .

Lie la lie, lie la lie lie liar loan,
Lie la lie, lie la lie lie lie lie lie lie liar loan...

by Anonymous on 5/19/2007 09:57:00 AM

“Borrowing to Get Ahead, and Behind: The Credit Boom and Bust in Lower-Income Markets” is a new paper by Matt Fellowes and Mia Mabanta of the Brookings Institution. It uses Federal Reserve and credit repository data to look at credit usage, total debt levels, and credit delinquency among the poorest quartile of households in 50 metropolitan areas. Although in some cases the paper raises more questions than it answers, it does provide some base data for questioning the wisdom of “ownership society” initiatives justified by the claim that owning is always and everywhere a better deal for the poor than renting.

I was struck by the data regarding debt management:

Still, while the vast majority of lower-income households are managing credit just fine, there is at the same time a large, relative share of lower-income borrowers who are struggling to manage their debt. This is indicated by at least two credit trends that set lower-income borrowers apart from other borrowers. First, lower-income households are highly leveraged. In fact, about 27 percent of lower-income families are now paying more than 40 percent of their income on debt payments—a dramatically higher proportion of households compared to those with a higher income. In fact, among the second income quartile, 15 percent of families pay these high debt-to-income ratios;
about 10 percent within the third quartile; and just 3 percent of the top income quartile. With such high debt service obligations, these 27 percent of lower-income borrowers face greater difficulty saving for additional investments and paying bills on time.

Second, lower-income borrowers are much more likely to fall behind on payments compared to higher-income borrowers. In fact, about one out of every three lower-income borrowers (33 percent) reported in 2004 that they have trouble making payments on time. In contrast, between 22 and 25 percent of borrowers in the second and third income quartiles, fell behind on payments; and just 10 percent of borrowers in the top income quartile fell behind on credit payments in 2004. . . .

To put that into some perspective, 55% of bottom quartile households and 89% of top quartile households had debt in 2004. I have to say the figures on “falling behind” in the middle quartiles—in 2004, while credit costs were still quite low and MEW was running at records—are at least as startling as the figures for low incomes.

This study also throws some cold water on the assumption that increased supply of mortgage credit to lower-income households is primarily a phenomenon of the bubbliest markets:

Differences in the costs of living across areas are a second major influence on the amount of debt held by the typical borrower. Where there are higher costs of living, borrowers in lower-income markets tend to borrow less compared to borrowers living in more affordable areas of the country. This is an important indication of how markets, not just the behavior or characteristics of borrowers, can regulate borrowing behavior. Places that are cheaper to live also afford more opportunities for people to borrow, because goods and services are more affordable, and because borrowers in these areas likely have more disposable income to spend on down-payments for credit-backed goods. Variable housing affordability is one sign of how this is so. Houses in relatively low-cost areas like Jacksonville and Indianapolis, for instance, are much more affordable than in expensive places like New York and San Jose, leading to sharp differences in homeownership rates. That is reflected by the systematically higher median debt held by borrowers from lower-income neighborhoods in low cost places like Jacksonville and Indianapolis compared to their higher cost peers.

But, it is not just mortgage debt that drives up the amount of debt held in more affordable areas of the country: median, non-mortgage debt is also higher in these low cost areas. What does cost of living have to do with those differences? For one, greater home buying rates in the lower cost areas of the country also likely produces higher relative demand for installment loans to buy appliances—costs that are less likely to be directly incurred by renters. Similarly, savings for downpayments to buy other credit-backed goods—like cars, consumer electronics, and furniture—are easier to accumulate when costs of living are low. This suggests that market differences can be nearly as an important influence on credit behavior as the decisions made by borrowers and lenders.

Fellowes and Mabanta conclude with a serious challenge to the idea that lower-income families are winning the bet via highly-leveraged and expensive debt:

While increased lending expanded the spending power and asset ownership in lower income markets, over one-third of lower-income borrowers now struggle to manage debt. Similarly, over one-fourth of lower-income borrowers now devote at least $4 out of every $10 earned for debt payments, pointing to the highly leveraged position of a wide number of lower income borrowers.

Lower-income households are faced with that relatively heavy debt burden mostly because of increased borrowing for mortgages, and installment trades tied to homeownership, like loans for furniture and appliances. In fact, homeownership-related debt accounts for about $7 out of every $10 owed by lower-income families, and is the fastest growing type of debt held by lower-income families.

Other trades, particularly credit cards, represent a very small share of the overall debt held by lower-income families. In fact, this paper finds that credit card debt represents just 6 percent of all debt held by lower-income households. While that is a higher proportion of credit card debt to all debt than exists at higher-income brackets, home equity borrowing among lower-income households—a source of debt widely used for purposes similar to credit cards—represents a much lower share of debt owed by lower-income households compared to all other households.

To meaningfully bring down the amount of debt owed by lower-income households, the unusually high debt service payments they are now burdened with, and the extremely high delinquency rates in some of these markets, policymakers will thus have to put an emphasis on homeownership-related debt—a type of debt that is heavily promoted by government policies. To be sure, this goal of expanding homeownership in lower-income markets should be reexamined, and not just because of evidence that debt has become such a dominating, and too often unsustainable, share of household expenditures among lower-income consumers. Evidence cited earlier also suggests that homeownership may not be a wise decision for every person that qualifies for credit, which suggests a more measured, even cautious, approach to homeownership-boosting initiatives than often exists. Mortgages do substitute for rent, but transaction costs, short holding periods, market downturns, home upkeep costs (i.e., repairing and replacing appliances) and interest-only and other exotic mortgages all can make homeownership a more expensive form of renting.

by Calculated Risk on 5/18/2007 06:08:00 PM

Via Mathew Padilla at the O.C. Register: Subprime delinquencies higher than reported. Padilla writes:

Forget that 13% subprime delinquency number you heard about so much in the press ... I quizzed the MBA and got this in response from Jay Brinkmann, vice president of research and economics:

... our latest subprime numbers are 14.4% delinquent by at least one payment, plus another 4.5% in foreclosure, for a total of 18.9% either delinquent or in foreclosure. For just subprime ARMs that number is 21.1%...

by Calculated Risk on 5/18/2007 12:30:00 PM

From Jon Lansner at the O.C. Register: Wells Fargo sees SoCal home prices down 6%-plus through '08

Scott Anderson, senior economist at Wells Fargo Bank, sees SoCal median home sales prices falling 2.3% this year and 4.1% in '08.

And Anderson on the impact of the housing bust on the SoCal economy:

... it appears the “direct” impacts from the drop in housing demand has yet to be fully realized in the economic and payroll data, and we will still have to deal with the “indirect” impacts on household wealth and consumer spending. Credit quality has taken a major hit over the past year in Southern California. Expect a period of below average and perhaps disappointing economic performance as the residential housing adjustment continues.

by Anonymous on 5/18/2007 12:17:00 PM

The following question was posed to Marketwatch's personal financial columnist yesterday. You can click here to see Lew Sichelman's answer. Or, being the Calculated Riskers that you are, you could offer alternative responses in the comments.

Note: "Statisticians should only talk to other statisticians" is not an acceptable answer, because we used that to great effect yesterday. ("We" in this case is Sippn.)

Question: For people who can't scrape together a 20% down payment, private mortgage insurance helps them get a home of their own. But what about the large segment of the population who can afford 20% down but don't want to pay it?

I'm in contract on a vacation home. Both I and my wife work white-collar managerial jobs in New York, so we have more than enough to buy the place outright. But we're lumped together with those that can't make 20% down and are forced to pay some 1%-2% more for PMI. That just doesn't make sense, not even for the lender, does it?

So rather than borrow 90% from a bank, I'll borrow only 80%. Then they make less money. Or maybe I'll borrow 90% but then pay some useless middleman his 1%-2% extra. Where's the market pressure to satisfy borrowers like me for whom mortgage insurance doesn't add any value?

While researching PMI, the key insight for me was reading the phrase "studies show that people who pay less than 20% are more likely to default." That exact phrase comes up all the time. In fact, I'd like to see those studies! When was the last one prepared, 1975? Isn't it odd that in our advanced world of actuarial analysis, no one breaks down those numbers to find that those low-down-payment defaulters also have lousy credit, don't have a job, are younger than 25 or whatever.

And isn't it odd that no bank seems to be interested in helping older, perfect-credit people with money get the property they desire without the wasted cost of PMI? Richard Campbell.

by Calculated Risk on 5/18/2007 01:47:00 AM

The BLS released the Business Employment Dynamics statistics for Q3 2006 yesterday. This data is released with a substantial lag (Q3 2006 was just released), and it gives a fairly accurate estimate of the annual benchmarking for the payroll survey. In 2006, the benchmark revisions were substantial. On Feb 2, 2007, the BLS reported:

The total nonfarm employment level for March 2006 was revised upward by 752,000 (754,000 on a seasonally adjusted basis). The previously published level for December 2006 was revised upward by 981,000 (933,000 on a seasonally adjusted basis).

This BED report suggests that the revisions will be down this year, especially for - you guessed it - construction employment! Based on the BED data, construction employment was overstated by 111K during Q3 2006. Since non-residential construction was still strong in '06, most of this downward revision will probably come from residential construction employment.

For those interested, the BED data is from the "Quarterly Census of Employment and Wages (QCEW), or ES-202, program." The sources "include all establishments subject to State unemployment insurance (UI) laws and Federal agencies subject to the Unemployment Compensation for Federal Employees program."

Click on graph for larger image.

This graph shows housing completions vs. residential construction employment. The green line (Q3 only) with the arrows shows the impact of the BED revision on residential construction employment.

My guess is Q4 (and Q1 2007) will also show substantial downward revisions in residential construction employment. This probably means the expected job losses have been occurring, but simply haven't been picked up in the initial BLS reports.

Unfortunately, we will not know the size of the revisions until the advanced estimate is released in October.

by Calculated Risk on 5/17/2007 01:10:00 PM

For those that enjoy statistics, Professor Menzie Chinn writes: Follow up on Housing Permits and Housing Starts: Do Permits "Predict"

In general I've ignored permits, and focused on starts and completions for housing. Professor Chinn argues there is some predictive value for permits:

Update: NOTE, the following graph is from Professor Chinn (see link). It is a log scale, and the gray area is future (not recession).

"These results lead me to the conclusion that -- while permits might not be incredibly informative on their own for future housing starts -- they are useful when taken in conjunction with the gap between log levels of housing starts and permits, as well as lags of first differenced log housing starts and permits.
...
The model predicts continued decline in housing starts of 5.1% (in log terms), calculated as changes in predicted values (as opposed to using the actually observed value for 2007M04). Of course, with a standard error of regression (SER) of 0.055, a zero change lies within the 67% prediction interval."

by Calculated Risk on 5/17/2007 12:35:00 PM

From the Department of Labor:

In the week ending May 12, the advance figure for seasonally adjusted initial claims was 293,000, a decrease of 5,000 from the previous week's revised figure of 298,000. The 4-week moving average was 305,500, a decrease of 12,000 from the previous week's revised average of 317,500.

Click on graph for larger image.

This graph shows the four moving average weekly unemployment claims since 1968. The four week moving average has been trending sideways, and the level is low and not much of a concern.

A word of caution: weekly claims is a weak leading indicator.

Note that the Conference Board uses weekly claims as one of their ten leading indicators. From the AP: Economy may slow this summer

The Conference Board said its index of leading economic indicators dropped 0.5 percent, higher than the 0.1 decline analysts were expecting. The reading is designed to forecast economic activity over the next three to six months.
...
"The data may be pointing to slower economic conditions this summer. With the industrial core of the economy already slow, and housing mired in a continued slump, there are some signs that these weaknesses may be beginning to soften both consumer spending and hiring this summer," said Ken Goldstein, labor economist for the Conference Board.

Another word of caution: many economists, including former Fed Chairman Greenspan, have little confidence in the Conference Board leading indicators. In 2000, Greenspan commented that he thought the Conference Board leading indicators were useless in real time:

"As an aside, the probability distribution based on the leading indicators looks remarkably good, but my recollection is that about every three years the Conference Board revises back a series that did not work during a particular time period, so the index is accurate only retrospectively. I’m curious to know whether these are the currently officially published data or the data that were available at the time. I know the answer to the question and it is not good!" [Laughter]

Update: And Northern Trust's Paul Kasriel (hat tip ac) argues that the LEI has value: When The Facts Change, I Change My Model – What Do You Do?

by Calculated Risk on 5/17/2007 10:50:00 AM

Remarks by Fed Chairman Ben Bernanke: The Subprime Mortgage Market

The recent sharp increases in subprime mortgage loan delinquencies and in the number of homes entering foreclosure raise important economic, social, and regulatory issues. Today I will address a series of questions related to these developments. Why have delinquencies and initiations of foreclosure proceedings risen so sharply? How have subprime mortgage markets adjusted? How have Federal Reserve and other policymakers responded, and what additional actions might be considered? How might the problems in the market for subprime mortgages affect housing markets and the economy more broadly?

Short answer according to Bernanke: everything will be fine. See Bernanke's speech for his longer answers.

Tanta provides the translation (from the comments):

CR quoted the first paragraph. Here's how the rest of it goes:

2. Technology lets us find more subprime borrowers faster.
3. Securitization lets us find more bagholders faster.
4. We made a lot more loans this way.
5. Ownershipsocietyminoritypoorpeoplehelpedfeelgood.
6. Somehow, subprime borrowers still default more than prime borrowers do.
7. Number 6 is a recent problem.
8. ARMs have rates that go up, and home prices don't always rise. This is new, too.
9. Some of that loan underwriting was also kind of sucky.
10. It paid to make junk loans if you didn't have to own them.
11. Apparently some borrowers didn't get the memo.
12. Hedgies are bailing out the CDOs, so there's still some party left in the punchbowl. So far the banks can pass a breath test.
13. Even though this isn't a bank problem, the Fed is working with banks to help solve it.
14. We think someone should buy out those crappy securities and start modifyin', baby.
15. For some reason these borrowers think the lenders want to foreclose. Sure, it looked that way when the loan was made, but it's different now. Please call your lender, it's ready to make nice.
16. We're snorting a fine line here.
17-22. The best solution is to disclose to the borrower that these loans rarely make sense. It would be bad to ban them entirely, because they often do make sense.
23-24. We are also guiding the underwriting of the banks that aren't the ones making these loans.
25-26. None of this will affect the home market unduly, because jobs and wages will go up.
27-28. The market will correct any problem we can't mop up with disclosure requirements.

by Anonymous on 5/17/2007 07:19:00 AM

On the way to my Yahoo! mailbox, I caught this headline, "Subprime Shakeout: Where Do We Go From Here?" That demanded to be read.

New Century had a 5-star Morningstar Rating for stocks during its collapse. Why? Good question. Admittedly, the company was risky, hence our above-average risk rating. Indeed, we even recognized that delinquencies would rise. In fact, we had assumptions of increasing charge-offs built into our New Century valuation model. But we underestimated two things.

First, we missed the risk that early payment defaults posed. As we stated, early payment defaults had never been an issue before, and we just did not see that risk on the horizon until it was too late. The risk we saw was loans defaulting as they reset from the "teaser" rate to a fully adjusted rate--the interest rate borrowers will ultimately have to pay after their low teaser rate expires--after two years.

Second, we didn't recognize how quickly the company's financing would disappear. New Century had been a darling of Wall Street, supplying banks with loans to package into new collateralized mortgage obligations. However, once the first problems arose, the very firms that had benefited from the flow of loans were the first to turn off New Century's financing spigot, demanding their money back and effectively killing New Century. The firm was out of cash, could not fund many loans already being processed, and was eventually forced to file for bankruptcy.

Arguably, if New Century had relied upon long-term debt instead of short-term financing, the company would probably still be alive today. No doubt, it would be struggling, as early payment defaults, high delinquency rates, and a reduced number of buyers of subprime mortgages would be taking its toll. But, we believe that with access to cash, New Century might have lived to see another day and maybe even stabilization in the market.

The implication that if New Century had been borrowing operating cash instead of lending capital, it wouldn't have gone bankrupt may not be the funniest thing you read today, but it has its charm. Yes, I'm aware that this is a reductio of the point really being made, which is that if New Century hadn't been using a warehouse line of credit to fund and carry its new held-for-sale loans (borrow short and lend long while you produce enough loans to securitize), and had instead used long-term investors' funds to sell those loans right out of the factory door (borrow long and lend long, really really fast), things would have been different. They surely would have.

What does startle me is the failure to connect the two dots: the EPD problem and the financing problem. My contention is that we have seen unprecedented numbers of EPDs because we were not, actually, lending long, we were pretending to lend long. (This is my Bridge Loan theory about Alt-A and subprime: a loan structure that forces you to refinance in two years or face financial ruin is not a long-term loan, regardless of what the technical final maturity date is).

Traditionally, the "early" in "early payment default" is in reference to a pretty long loan life; the "traditional" average life of a 30-year mortgage just before the boom got underway after 2001 was about 7-10 years, so defaults in the first 90 days were weird and rare. There is, however, something odd about understanding a third-payment delinquency on a 2/28 "exploding ARM" made to a speculator as particularly "early." I mean, how many payments did we expect to take?

Something starts to suggest that we were just borrowing short and lending short, until of course those EPD rates hosed up the liquidity of the loans in the warehouse and we were suddenly borrowing long in real dollars and lending short in Monopoly money. You could see that as an issue of a lack of access to cash, even if like me you think the missing cash is repayments from borrowers rather than loans from investors. In any case I'm still struggling with the idea of how you live to fight another day by soldiering on with a neutral or negative carry. Presumably that would have been solved by the fall-off in loans originated: you can always make it up on lack of volume. So where do we go from here? Those of us who assumed the point was to make money making loans, rather than just make loans, can go back to bed. Everybody else should buy stock.

In other news, I received an email containing some color on current Alt-A and subprime trades, that contained this gem: "Sellers continue to look to product development to figure out ways to originate higher LTV product without subordinate financing."

What this means, for you civilians, is that since we replaced mortgage insurance with borrowed down payments until the down-payment lenders got burnt to a crisp right at the time the MIs decided they wouldn't play with us even if we tied a pork chop around our necks--those things are probably connected--we're back to "product development," which is going to be a bit tricky after that "nontraditional mortgage guidance" thingy told us to quit developing products that fake their way through a lack of borrower equity. But we're willing to try something tricky again, given that the alternative, limiting high LTV loans to people who can afford them, still sucks. What we need is a way for Joe Homebuyer to borrow long and lend short . . .

by Calculated Risk on 5/16/2007 08:59:00 PM

"We believe the home-improvement market will remain soft throughout 2007."
Frank Blake, Home Depot Chairman and CEO, May 15, 2007

Soft? Actually real spending on home improvement is holding up pretty well. If this housing bust is similar to the early '80s or '90s, real home improvement investment will slump 15% to 20%.

Click on graph for larger image.
This graph shows real home improvement investment (2000 dollars) since 1959. Recessions are in gray.

Although real spending was flat in Q1 2007, home improvement spending has held up pretty well compared to the other components of Residential Investment. With declining MEW, it is very possible that home improvement spending will slump like in the early '80s and '90s.

by Calculated Risk on 5/16/2007 08:05:00 PM

From DataQuick: California April 2007 Home Sales

A total of 34,949 new and resale houses and condos were sold statewide last month. That's down 12.2 percent from 39,811 for March, and down 28.5 percent from 48,894 for April 2006. Last month's sales made for the slowest April since 1995 when 27,625 homes were sold. April sales from 1988 to 2007 range from the 27,625 in 1995 to 66,938 in 2005. The average is 46,141. On a year-over-year basis, sales have declined the last 19 months.

by Calculated Risk on 5/16/2007 01:13:00 PM

By popular demand, here is a graph of Starts and Completions with Recessions.

Click on graph for larger image.

An optimist might argue that we should already be in a recession based on previous housing slumps - so if the worst is over for this housing slump, a soft landing is very possible. Of course I think starts will fall further - perhaps to the 1.1 million SAAR level.

The WSJ housing story led with this sentence:

Home construction unexpectedly rose during April, making a surprise increase despite bloated inventories and tighter credit for subprime borrowers ...

And yes, fundamentals like "bloated inventories" and less demand, do matter.

by Calculated Risk on 5/16/2007 10:50:00 AM

The Census Bureau reports on housing Permits, Starts and Completions. Seasonally adjusted permits declined:

Privately-owned housing units authorized by building permits in April were at a seasonally adjusted annual rate of 1,429,000. This is 8.9 percent below the revised March rate of 1,569,000 and is 28.1 percent below the revised April 2006 estimate of 1,987,000.

Starts rebounded:

Privately-owned housing starts in April were at a seasonally adjusted annual rate of 1,528,000. This is 2.5 percent above the revised March estimate of 1,491,000, but is 16.1 percent below the revised April 2006 rate of 1,821,000.

And Completions declined to the level of starts:

Privately-owned housing completions in April were at a seasonally adjusted annual rate of 1,523,000. This is 5.8 percent below the revised March estimate of 1,616,000 and is 26.0 percent below the revised April 2006 rate of
2,058,000.

Click on graph for larger image.
The first graph shows Starts vs. Completions.

As expected, Completions have followed Starts "off the cliff". Completions are now at the level of starts. Starts will probably fall further (based on permits and housing fundamentals) and completions will most likely decline further too.



This graph shows starts, completions and residential construction employment. (starts are shifted 6 months into the future). Completions and residential construction employment were highly correlated, and Completions used to lag Starts by about 6 months.

Both of these relationships have broken down somewhat (although completions have fallen to the level of starts). Why residential construction employment hasn't fallen further is a puzzle. Also the time between start and completion has increased recently.

This report shows builders are still starting too many projects, and that residential construction employment is still too high.

by Calculated Risk on 5/15/2007 08:44:00 PM

With the release of the Fed Loan Survey, we can connect another dot in the possible Commercial Real Estate (CRE) slump later this year.

Click on graph for larger image

The first graph shows the YoY change in nonresidential structure investment vs. loan demand data from the Fed Loan survey: Net Percentage of Domestic Respondents Reporting Stronger Demand for Commercial Real Estate Loans

Unfortunately the demand survey data is only available since 1995, but the correlation is clear: falling demand leads lower investment by about a year. The causation is obvious, loans taken out today impact investment over the next couple of years.

This is more evidence that the normal pattern will hold: nonresidential structure investment will probably follow residential investment "over the cliff".

Here are some earlier dots:

1) Non-residential structure investment has been booming for the last few years. Commercial rents have been rising and office vacancy rates have been falling. But signs of a trend change are emerging, from the IHT on May 2nd: Alarm raised over U.S. commercial real estate lending

... signs are emerging that the office market is slowing down in the United States. Though rents continued to rise in the first quarter of this year, the average vacancy rate for 58 U.S. metropolitan markets rose to 12.6 percent from 12.5 percent, the first increase for any quarter since 2004, according to Torto Wheaton Research, a division of CB Richard Ellis.

2) This graph shows the YoY change in Residential Investment (shifted 5 quarters into the future) and investment in Non-residential Structures. In the typical cycle, non-residential investment follows residential investment, with a lag of about 5 quarters. Residential investment has fallen significantly for four straight quarters (following two minor declines). So, if this cycle follows the typical pattern, non-residential investment will start declining later this year.

3) The WSJ noted earlier this month that demand was "sluggish" for office space. The current office space absorption rate is about 8 to 10 million square feet per quarter, but "... developers will open 76 million square feet of new office space by the end of this year." Since supply will grow significantly faster than the absorption rate, vacancy rates will rise.

4) Many banks are over-exposed to CRE lending. With rising vacancy rates, defaults will probably start to rise for CRE loans.

And remember, non-residential investment is the great hope of the soft landing view. That is why many economists are watching non-residential investment closely.

by Calculated Risk on 5/15/2007 02:54:00 PM

From the LA Times: SoCal home sales hit 12-year low

Southern California home sales plunged to a 12-year low for the month of April, dragged down by a dearth of transactions at the lower end of the market even as prices held steady ... DataQuick Information Systems reported.

... nearly a third fewer homes in Los Angeles, Orange, Riverside, San Bernardino, San Diego and Ventura counties closed escrow last month compared to a year earlier for the worst April showing since 1995, DataQuick found.
...
Most of the erosion in sales appeared in the lower-priced markets of the Inland Empire that only a year ago still seemed to be soaring. In Riverside County, sales dropped 45.1% to 2,987 year over year, while in neighboring San Bernardino County, sales plunged 46.7% to 2,049 ...

"The falloff in starter home sales has the effect of pushing median prices up a bit, although it's still somewhat surprising prices haven't declined more," said DataQuick president Marshall Prentice.

The changing market poses a problem with the "median price" method of tracking prices. As an example, say 100 homes sold in a certain month, 20 homes each at $300K, $400K, $500K, $600K, and $700K. The median price would be $500K.

However, during the next month, say half the homes in the two lower brackets didn't sell (only 10 each for $300K and $400K), but all the other homes sold at the same price as the month before. Sales would decrease by 20% (similar to the declines reported by DataQuick). However the median price would increase 10% to $550K! Clearly this is misleading.

The OFHEO house price index uses same home sales, so it doesn't suffer from this problem (although there are other problems with the OFHEO Index, especially for more expensive areas). The Q1 OFHEO HPI will be released on Thursday, May 31st.

by Calculated Risk on 5/15/2007 01:56:00 PM

Click on graph for larger image.

NAHB Press Release: Builder Confidence Slips Again In May

Ongoing concerns about subprime-related problems in the mortgage market caused builder confidence about the state of housing demand to decline three more points in May, according to the National Association of Home Builders/Wells Fargo Housing Market Index (HMI), released today. With a current reading of 30, the HMI has now returned to the lowest level in its current cycle, which was previously hit in September of 2006.

“Builders are feeling the impacts of tighter lending standards on current home sales as well as cancellations, and they are bracing for continued challenges ahead,” said NAHB President Brian Catalde, a home builder from El Segundo, Calif.

“The crisis in the subprime sector has infected other parts of the mortgage market as well as consumer psychology, and as a result the housing outlook has deteriorated,” added NAHB Chief Economist David Seiders. “We’re now projecting that home sales and housing production will not begin improving until late this year, and we’re expecting the early stages of the subsequent recovery to be quite sluggish. There still are tremendous uncertainties regarding our baseline forecast going forward, owing largely to the subprime crisis that is having widespread effects throughout the mortgage market.”

Derived from a monthly survey that NAHB has been conducting for more than 20 years, the NAHB/Wells Fargo HMI gauges builder perceptions of current single-family home sales and sales expectations for the next six months as either “good,” “fair” or “poor.” The survey also asks builders to rate traffic of prospective buyers as either “high to very high,” “average” or “low to very low.” Scores for each component are then used to calculate a seasonally adjusted index where any number over 50 indicates that more builders view sales conditions as good than poor.

All three component indexes declined in May. The index gauging current single-family sales slipped two points to 31, while the index gauging sales expectations for the next six months fell three points to 41 and the index gauging traffic of prospective buyers fell four points to 23.

Three out of four regions posted declines in the May HMI. The Northeast posted a six-point decline to 32, while the South posted a four-point decline to 33 and the West posted a three-point decline to 32. The Midwest eked out a one-point gain, to 23.