by Calculated Risk on 11/27/2007 05:34:00 PM

Goldman Sachs Chief Economist Jan Hatzius released a new report on housing today: Housing (Still) Holds the Key to Fed Policy

There is no link available.

Note: The following excerpts are used with permission.

In this new report, Goldman revised down their housing outlook significantly (here is their August forecast). Goldman now sees housing starts falling to 750K (their earlier forecast was for starts to fall to 1.1 million units).

On housing prices:

Home prices are also likely to decline substantially. If the economy narrowly escapes a full-blown recession—as we continue to expect in our baseline forecast—a peak-to-trough decline of 15% in house prices is the most likely outcome. This would imply price declines in states such as Florida of up to 30%. If the economy does enter a recession, prices could decline as much as 30% nationwide.

On the impact of less Mortgage Equity Withdrawal (MEW) on consumer spending:

Consumer spending growth has remained stable over the last 1-2 years as rising equity prices and sturdy income growth have offset the drag from falling mortgage equity withdrawal (MEW) and slowing home prices. Nevertheless, consumption has underperformed income growth, as predicted by our MEW-augmented consumption model. Going forward, our model points to a more substantial drag of housing on real consumer spending growth, with a slowdown from the recent 3% pace to a 1% annualized rate in early 2008.

Click on graph for larger image.

On negative equity:

The basic problem is that house price declines create large amounts of negative equity. Homeowners with negative equity lose their ability to respond to adverse financial events such as job loss or mortgage reset by refinancing or selling their home, and they therefore become much more likely to default. The importance of this problem is illustrated in Exhibit 16, which shows the distribution of home equity among US mortgage holders at the end of 2006 according to an analysis by First American CoreLogic, Inc. About 7% of US mortgage holders had negative equity at that point, and another 14% had equity of less than 15%. Thus, 21% of all mortgage holders—holding about $3 trillion in aggregate mortgage debt given the average mortgage debt held by the vulnerable borrowers—would be put into a negative-equity position if home prices fell by 15%.

There is much more in the report. Goldman now puts the odds of a recession in 2008 at around 40%, and they see the unemployment rate rising to 5.5% by the end of 2008.

In the past, such a rise in the unemployment rate has invariably come in the context of an overall recession ... The risk that it would do so again is high ... however, our analysis does not imply a recession when taken at face value. Instead, it points to a long period of below-trend growth.

I'll post more on the details of their analysis - and offer my view - later this week. Goldman has definitely turned significantly more bearish on housing and the economy.

by Calculated Risk on 11/27/2007 02:59:00 PM

The S&P Case-Shiller National home price index was released this morning. (See note at bottom).

The reported Case-Shiller numbers are nominal; not adjusted for inflation. Most people think in nominal terms, but it's also important to look at real house prices.

Click on graph for larger image. The first graph shows the nominal Case-Shiller index. The index peaked in Q2 2006, and nominal prices have declined about 5% from the peak.

The second graphs shows real prices according to the Case-Shiller index (adjusted using CPI less Shelter). Real Prices peaked in Q1 2006, and are off about 8% from the peak.

The second graph probably provides a better first estimate of how far prices still need to fall (for the Case-Shiller universe). If prices fall to 120 (in real terms) that is about another 25% from the current level.

This could happen with falling nominal prices, or from several years of inflation, or a combination of both. Say nominal prices fall 15% over the next three years, with a 2% per year inflation rate, then real prices would fall to about 130 on the Case-Shiller index.

This suggests to me that price declines have just started, and that the process will last several years. It's important to remember that different areas will see different percentage price declines - the bubble areas will see the largest declines - and the time frames for each location will be different.

NOTE: There are significant differences between the OFHEO HPI and the Case-Shiller National index. This is an excellent summary by OFHEO economist Andrew Leventis: A Note on the Differences between the OFHEO and S&P/Case-Shiller House Price Indexes. The OFHEO note suggests that the primary reason for the difference between Case-Shiller and OFHEO price indices is geographical coverage (not the loan limitations for OFHEO).

by Calculated Risk on 11/27/2007 01:35:00 PM

Peter Y. Hong at the LA Times writes: Homeowners' big question: How low will prices go?

Click on graph for larger image.

The LA Times article focuses on California. This graph shows the Case-Shiller home price index for several selected cities. If SoCal prices fall 25%, then prices in other areas - like Miami and Las Vegas - will probably decline a similar amount. These are nominal prices, I'll have more on Real vs. Nominal prices later today.

From the LA Times:

No one knows how severe the slump will be, but economists and real estate experts interviewed by The Times, and who were willing to make predictions, said prices could fall 15% to 25% before turning back up.

Most said values would continue falling through at least next year, and some thought the market wouldn't reverse course until 2010.

That could translate to big declines for home buyers who bought at the peak of the market, which various measures place in late 2006 or early 2007.
...
Some analysts, including UC Berkeley professor Kenneth Rosen, believe the severity of the downturn will vary by region.

Areas such as the Central Valley and the Inland Empire will be the hardest hit, he said, because these attracted a higher percentage of new buyers with shaky credit, and many of them are now defaulting on their loans. He believes values in these communities could fall by 15%.

But "in areas where there is very little new housing, where it's hard to build and a lot of wealthy people live, there will be little decline or maybe none at all."
...
But others call this wishful thinking, saying low prices eventually work their way to even the most affluent areas.

"Every place takes the hit in the long run," said Christopher Thornberg of Beacon Economics, a consulting firm in L.A.
...
[Edward E. Leamer of UCLA's Anderson Forecast] and Thornberg are among the most bearish of analysts, saying the recently ended housing boom pushed prices out of sync with incomes.

Los Angeles County median home prices are about 40% to 50% higher than the median income justifies, Thornberg said. He said the market would settle when prices and incomes became more closely aligned.

"Southern California prices will fall 25% from their peak and won't find their bottom until the end of 2009," Thornberg said.

Leamer also sees a drop-off at the high end of the range -- 20% to 25% -- and sees the downturn lasting into 2010.

by Calculated Risk on 11/27/2007 11:08:00 AM

From OFHEO: 2008 Conforming Loan Limit $417,000

Office of Federal Housing Enterprise Oversight Director James B. Lockhart today announced the maximum 2008 conforming loan limit for single-family mortgages purchased by Fannie Mae and Freddie Mac (the Enterprises) will remain at the 2007 level of $417,000 for one-unit properties for most of the U.S. Higher limits apply to Alaska, Hawaii, Guam and the U.S. Virgin Islands as well as to properties with more than one unit.

The conforming loan limit determines the maximum size of a mortgage that an Enterprise can buy or guarantee. By law the maximum conforming loan limit is based on the October-to-October change in the average house price in the Monthly Interest Rate Survey (MIRS) of the Federal Housing Finance Board (FHFB). The FHFB reported the decline in the average price was $10,685 or 3.49 percent, from $306,258 in October 2006 to $295,573 in October 2007. The combined two-year decline is now 3.65 percent.

“While the house price survey data used in determining the conforming loan limit show a decline over the past year, as previously announced and consistent with the proposed new conforming loan limit guidance, the level will remain at $417,000 for the third straight year,” said Lockhart.

The FHFB is reporting the year over year national price decline was 3.49%.

by Calculated Risk on 11/27/2007 09:58:00 AM

From MarketWatch: Home prices falling everywhere: S&P

U.S. home prices were falling in every region of the country in September, according to a closely watched index of home prices released Tuesday.

Home prices fell in September in all 20 major cities covered by the Case-Shiller price index, even in cities that had been holding up before the August freeze in mortgage markets, Standard & Poor's reported.

"There is no real positive news in today's data," said Robert Shiller, chief economist at MacroMarkets LLC, and the co-developer of the index.

For the national Case-Shiller home price index, prices fell 1.7% in the third quarter compared with the second quarter, and were down a record 4.5% in the past year. It was the largest quarter-to-quarter price decline in the 20 years covered by the index.

by Calculated Risk on 11/27/2007 12:35:00 AM

From the WSJ: Abu Dhabi to Bolster Citigroup With $7.5 Billion Capital Infusion

Citigroup Inc. ... is receiving a $7.5 billion capital infusion from the investment arm of the Abu Dhabi government.
...
As a result of the deal, the investment authority known as ADIA will become one of Citigroup's largest shareholders, with a stake of no more than 4.9%.
...
In exchange for its investment, ADIA will receive convertible stock in Citigroup yielding 11% annually. The shares are required to be converted into common stock at a conversion price of between $31.83 and $37.24 a share over a period of time between March 2010 and September 2011.

by Calculated Risk on 11/26/2007 08:52:00 PM

From the Cleveland Fed: Fed Funds Rate Predictions

Click on graph for larger image.

A number of Fed presidents will be speaking this week, including Cramer's favorite Fed President William Poole on Friday.

The Fed has been saying don't expect a rate cut in December; meanwhile the market is debating the size of the cut: 25bps or 50bps.

Since the credit crunch is worsening again, I've linked to Cramer's August meltdown at the bottom of the posts.

by Calculated Risk on 11/26/2007 04:07:00 PM

The Irvine Housing blog brings us these details (hat tip Atrios):

Asking Price: $1,249,000 Purchase Price: $1,157,000 Purchase Date: 1/6/2005

According to the Irvine Housing blog:

The property was purchased in January 2005 for $1,157,000. The combined first and second mortgages totalled $1,156,730 leaving a downpayment of $270. Let’s just call it 100% financing.

By April, they owners were able to find refinancing through Countrywide with a $999,999 first mortgage. This mortgage was an Option ARM with a 1% teaser rate. The minimum payment would be $3,216 per month.

Also in April of 2005, they took out a simultaneous second mortgage for $215,000 pulling out their first $58,000.

So look at their situation: They are living in a million dollar plus home in Turtle Ridge making payments less than those renting, and they “made” $58,000 in their first 4 months of ownership.

Apparently, these owners liked how hard their house was working for them, so they opened a revolving line of credit (HELOC) in August 2005 for $293,000. Did they spend it all? I can’t be sure, but the following certainly suggests they did.

In December of 2005, they extended their HELOC to $397,990.

In June of 2006, they extended their HELOC to $485,000.

In April of 2007, the well ran dry as they did their final HELOC of $491,000. I bet they were pissed when they couldn’t get more money.

So by April 2007, they have a first mortgage (Option ARM with a 1% teaser rate) for $999,999, and a HELOC for $491,000. These owners pulled $333,000 in HELOC money to fuel consumer spending.

Assuming they spent the entire HELOC (does anyone think they didn’t?), and assuming the negative amortization on the first mortgage has increased the loan balance, the total debt on the property exceeds $1,500,000. The asking price of $1,249,000 does not look like a rollback, but if the property actually sells at this price, the lender on the HELOC (Washington Mutual) will lose over $300,000.

These owners will probably just walk away. I doubt they have any assets. They never put any money into the deal, they pulled out $333,000 in cash, and they got to live in Turtle Ridge for 3 years. Not a bad deal — for them.

This story has been repeated all across America (usually on a smaller scale). This was not a subprime loan when the home was first purchased, but the collateral is now less than the total loan amount. The house hasn't sold yet, so perhaps the $999,999 Option ARM first is also impaired.

And look at the Mortgage Equity Withdrawal (MEW). One third of a million dollars, or over $100K per year. Perhaps the money was invested. Perhaps it was spent on new cars, flat screen TVs, vacations, or more - but this Home ATM appears out of money, and I suspect that will impact the homeowners' lifestyle.

This illustrates two important points: We are all subprime now, and, with falling house prices, the Home ATM is running dry.

by Calculated Risk on 11/26/2007 02:02:00 PM

From MarketWatch: Citigroup leads slide on report of 'massive' layoffs

Citigroup shares fell Monday after CNBC reported the firm could lay off up to 45,000 staffers ...

CNBC reported early Monday that the bank is planning a large number of layoffs as part of a response to recent huge write-offs for bad mortgage investments.

CNBC described the layoffs as "massive" and said they would not be restricted to the fixed income and mortgage divisions.

In April, Citi set layoffs of 17,000 people, or about 5% of its more than 300,000 employees.

UPDATE: a comment from Citi via the WSJ :

"We are engaged in a planning process in anticipation of our new CEO, and our business heads are planning ways in which we can be more efficient and cost effective to position our businesses in line with economic realities," [Citi] spokeswoman Christina Pretto said in a statement ... "any reports on specific numbers (of layoffs) are not factual."

by Calculated Risk on 11/26/2007 12:10:00 PM

What does it mean that HSBC is moving their SIVs to their balance sheet?

Let's start with the structure of an SIV (Structured Investment Vehicle). First an SIV has investors - like hedge funds or wealthy individuals - who invest say $1 Billion in the SIV (the equity). Then the SIV issues commercial paper (CP) and medium-term notes (MTN) that pay slightly higher rates than similar duration paper. The typical SIV, according to Fitch, uses 14 times leverage, so in our example the SIV would sell CP and MTN for $14 Billion.

Now the SIV invests this $15 Billion ($1 Billion equity and $14 Billion borrowed) in longer term notes. The idea is simple: borrow short, lend long, hedge the interest rate and credit risks - and the profits flow to the investors in the SIV.

So what does a bank like HSBC have to do with this? Usually the bank sets up the SIV, attracts the investors, manages the SIV for a fee - and there was always the appearance that the SIV CP was backed by the bank - perhaps allowing the CP and MTN to pay lower interest rates.

So what is the problem? Some SIVs invested in asset backed paper, backed by home mortgages. Even though the SIVs almost always invested in the highest tranches (with no losses to date), the market value of these assets has fallen recently (not a news flash). This means that the investors in the SIV (the equity) have taken paper losses on their $1 Billion investment.

UPDATE: Note the following NAVs are for the equity portion. A NAV of 71% means the $1 Billion equity in the example is now worth $710 million.

In fact many of the SIV NAVs have fallen substantially. From Moody's: Moody's says some SIV NAVs have fallen below 50%

Moody's [on Nov 8th] said that the average NAV across the SIV sector has fallen from 101% at the beginning of July to 71% at the beginning of November, and the shut-down of the CP market has led to realised losses in some cases.

However, the rating agency pointed out that there was significant variation between the NAVs of different SIVs, with some declining only to 90% and others falling below 50%.

Once the value of the equity falls enough (usually 50%) there is usually a trigger event forcing the SIV to liquidate the longer term investments. A forced liquidation might not only wipe out all the remaining SIV equity, but the holders of the CP and MTN might take some losses too.

This has made potential investors in CP and MTN (not to be confused with the investors in the equity of the SIV) to refuse to buy any more CP. Since there is a duration mismatch - the investments are in longer term notes, CP is less than 9 months - the SIV is stuck with a liquidity problem when the CP comes due.

To solve this problem, a bank like HSBC could explicity guarantee the CP and MTN, and this would attract investors in CP and MTN again. But under accounting rules, this guarantee means the SIV belongs on the bank's balance sheet. The structure stays the same - the SIV equity investors still take the losses - but there is no liquidation event. If the losses exceed the equity investment ($1 Billion in our example), then the bank would start taking losses.

From the HSBC article this morning:

[HSBC] insists earnings won't be materially impacted, because existing investors will continue to bear all economic risk from actual losses.

Clearly HSBC think these is adequate equity in these SIVs to cushion the bank from any losses.

Finally, to the balance sheet!

The balance sheet lists the assets and liabilities of the company. Moving the SIV to the balance sheet simply means adding the $15 Billion in assets (those longer term notes) to the Asset portion of the balance sheet, and moving the $15 Billion in CP, MTN and SIV equity to Liabilities. The new assets balance with the new liabilities, and there is no income or loss for the bank. Since the equity will take the losses first, any mark down in the $15 Billion in assets will be matched by a mark down in the liabilities - up to $1 Billion.

So what is the problem if there are no losses for the bank? There is an impact on the ratios of the bank - the reason the SIVs were off the balance sheet in the first place - and this limits other lending activities of the bank, contributing to the credit contraction.