by Calculated Risk on 7/28/2011 10:36:00 PM

From the NY Times DealBook: Debt Ceiling Impasse Rattles Short-Term Credit Markets

Over the last week, big banks and companies have withdrawn $37.5 billion from money market funds that invest in Treasury debt and other ultra-safe securities, the biggest weekly drop this year. Meanwhile, in the vast market for repurchase agreements, in which many financial firms make short-term loans to one another, borrowers are beginning to demand higher yields.

From the WSJ: Default Worries Dry Up Lending

Banks ... are scrambling to design emergency plans to avoid a trading logjam in the huge markets for Treasurys and short-term funding facilities if Congress fails to raise the U.S. borrowing limits by next Tuesday's deadline.
...
Trading executives from the largest Wall Street dealers agreed on a Wednesday conference call, conducted by the industry trade group the Securities Industry Financial Markets Association, to a number of procedures to trade Treasury bonds if the U.S. misses a payment on its debt.

From CNBC: Will Debt Feud Clip Future Economic Growth?

Washington's political feuding over the deficit has damaged business and consumer sentiment in an already weak economy ...

I've heard comments from several executives this week that business has slowed sharply over the last week. People are getting nervous.

I've been trying to ignore the charade - obviously Congress will agree to raise the debt ceiling and pay the bills - but it is now impacting the economy.

by Calculated Risk on 7/28/2011 06:26:00 PM

I've seen several people compare total housing starts with previous decades and ask: "Why is there still excess supply?"

Below is the long term graph of both total housing starts and single unit starts. If we look at the graph, we notice that there were more starts at the peak in the '70s than during the recent housing bubble.

Obviously there were many more multi-unit housing starts in the '70s - and that is a clue.

Click on graph for larger image in new window.

The key to the number of housing starts is household formation.

Household formation is a function of changes in population, and also of changes in household size. During the '70s, the baby boomers started moving out of their parents' homes, and there was a dramatic decrease in the number of persons per household. And that led to a huge demand for apartments (the surge in total starts).

The table below shows the number of persons per household for every decade from 1950 through 2010 (based on the decennial census data). Also using the decennial census data we can calculate the number of households needed because of 1) population growth, and 2) changes in household size:

Decennial Census, Population and Households in Millions
Census	Population	Households	Persons per household	Increase in Households over decade	Increase in Households due to Population Growth	Increase in Households due to change in Household Size
1950	150.7	42.8	3.52	---	---	---
1960	179.3	53.0	3.38	10.2	8.1	2.1
1970	203.2	63.4	3.21	10.4	7.1	3.3
1980	226.5	80.4	2.82	17.0	7.3	9.7
1990	248.7	91.9	2.71	11.5	7.9	3.6
2000	281.4	105.5	2.67	13.6	12.1	1.5
2010	308.7	116.7	2.65	11.2	10.2	1.0

Because of the changes in household size, the U.S. needed far more additional housing units in the '70s than in the '00s. In the decade ending in 1980, there were 17 million households added. A majority of those households were added because of the decrease in the number of persons per household (boomers moving out!).

Unfortunately it is difficult to estimate the number of housing units needed in a given time period, even if we know the number of new households being formed (and we don't have timely data on household formation!). We also have to account for scrappage (demolitions), mobile homes and second homes. And this assume no excess supply - and right now there is a significant excess supply.

A simple formula would be:

Housing Starts + mobile homes needed = Households formed + scrappage + second homes added.

So if 1 million households are formed in a year, 200 thousand homes demolished (probably close), and say 100 thousand 2nd homes added, then the total housing starts plus mobile homes added would be 1.3 million.

Note: this doesn't account for location (most homes are not transportable), and the desires of each household (a mobile home isn't a substitute for a 4,000 square foot home).

So we can't just compare housing starts in different decades without looking at household formation. I'll have more on this ...

by Calculated Risk on 7/28/2011 02:44:00 PM

From San Francisco Fed President John Williams: The Outlook for the Economy and Monetary Policy

Some excerpts on housing:

One of the most important currents holding back recovery has been housing. The collapse of the housing market touched off the financial crisis and recession. In most recessions, housing construction falls sharply, but then leads the economy back when growth resumes. As you well know, that snapback hasn’t occurred this time. Before the crisis, residential investment as a share of the economy was at its highest level since the Korean War. Today, housing construction remains moribund and residential investment as a share of the economy has fallen to its lowest level since World War II.

On one level, that’s not surprising. We simply built too many—in fact, millions too many—houses during the boom and we are still feeling the effects of this overhang. Consider housing prices. From their peak in 2006 until early 2009, home prices nationwide fell by nearly a third. When you exclude distressed sales, prices appeared to bottom out in 2009 and early 2010. New housing starts also appeared to stabilize in 2009, after plummeting some 75 percent during the housing crash. ...

The $64,000 question is when will the housing market finally recover? One daunting challenge for such a recovery is the huge number of homes in foreclosure. Almost 7 million homes have entered into foreclosure since the first quarter of 2008 and some 2 million are still in the foreclosure process. In addition, there is a shadow inventory of homes currently owned by delinquent borrowers. When you add up unsold new houses left over from the boom, homes for sale by owners, foreclosed residences for sale by lenders, and the shadow inventory of houses at risk of distressed sale, you come up with a massive supply overhang.

Over time, more reasonable prices and an improving economy ought to bring buyers off the sidelines and set the stage for recovery. But high unemployment and anemic wage gains are leaving people worried about their income prospects and cautious about buying homes. Also, the dramatic plunge in home valuations since 2006 has made some first-time homebuyers wary about entering the market because of worries that prices might fall further.

These are key points: Usually housing is a key engine of recovery, but not this time because of the massive supply overhang. And looking forward:

It’s only a matter of time before we work off the inventory overhang and construction picks up. How much time it takes will depend in part on what happens with foreclosed properties. If we begin making progress on working down the foreclosure inventory, then single-family housing starts could plausibly rise from their current level of about 400,000 per year to an average level of perhaps 1.1 million per year in three or four years, according to research at the San Francisco Fed.⁴ To put this in perspective, such an increase would boost real gross domestic product, or GDP, by at least 1 percent.

⁴ By contrast, if we can't work down the foreclosure inventory, then a return to normal construction levels could be delayed several more years. See Hedberg and Krainer (2011).

This is why I continue to focus on the excess supply. This is a key number for housing and the U.S. economy. See The “Excess Supply of Housing” War

by Calculated Risk on 7/28/2011 11:00:00 AM

From the Kansas City Fed: Manufacturing Sector Slows After Solid Rebound in June

The Federal Reserve Bank of Kansas City released the July Manufacturing Survey today. According to Chad Wilkerson, vice president and economist at the Federal Reserve Bank of Kansas City, the survey revealed that growth in Tenth District manufacturing slowed in July after a solid rebound in June, but producers remain generally upbeat about future activity.

“Factory activity in our region grew at a slower pace in July after rebounding solidly in June,” said Wilkerson. “Several firms blamed the slowdown on customers being cautious until the national debt ceiling debate is resolved. However, expectations for orders, hiring and capital spending, later in the year, generally remained as solid as in recent months.”
...
The month-over-month composite index was 3 in July, down from 14 in June but up from 1 in May. ... Most month-over-month indexes fell in July. The production index plunged from 25 to 0, and the shipments, new orders, and order backlog indexes also decreased. The employment index dropped from 17 to 4, and the new orders for exports index posted a negative reading for the first time since mid-2009.

This is the last of the regional Fed surveys for July. The regional surveys provide a hint about the ISM manufacturing index - and the regional surveys were fairly weak again this month, although slightly stronger than in June (in aggregate).

Click on graph for larger image in graph gallery.

The New York and Philly Fed surveys are averaged together (dashed green, through July), and five Fed surveys are averaged (blue, through July) including New York, Philly, Richmond, Dallas and Kansas City. The Institute for Supply Management (ISM) PMI (red) is through June (right axis).

The regional surveys were slightly better in July than in June. The ISM index for July will be released Monday, August 1st.

by Calculated Risk on 7/28/2011 10:00:00 AM

From the NAR: Pending Home Sales Rise in June

The Pending Home Sales Index,* a forward-looking indicator based on contract signings, rose 2.4 percent to 90.9 in June from 88.8 in May and is 19.8 percent above the 75.9 reading in June 2010, which was the low point immediately following expiration of the home buyer tax credit. The data reflects contracts but not closings.
...
The PHSI in the Northeast slipped 0.4 percent to 68.9 in June but is 19.4 percent higher than June 2010. In the Midwest the index fell 3.7 percent to 79.7 in June but is 26.4 percent above a year ago. Pending home sales in the South increased 4.4 percent to an index of 99.2 and are 19.1 percent higher than June 2010. In the West the index rose 6.4 percent to 107.0 in June and is 16.4 percent above a year ago.

This was very close to Tom Lawler's forecast of a 2.6% increase. This suggests an increase in reported existing home sales in July and August (depending on the number of cancellations).

by Calculated Risk on 7/28/2011 08:42:00 AM

The DOL reports:

In the week ending July 23, the advance figure for seasonally adjusted initial claims was 398,000, a decrease of 24,000 from the previous week's revised figure of 422,000. The 4-week moving average was 413,750, a decrease of 8,500 from the previous week's revised average of 422,250.

This is the first week with initial claims below 400,000 since early April.

The following graph shows the 4-week moving average of weekly claims since January 2000 (longer term graph in graph gallery).

Click on graph for larger image in graph gallery.

The dashed line on the graph is the current 4-week average. The four-week average of weekly unemployment claims decreased this week to 413,750.

by Calculated Risk on 7/27/2011 08:29:00 PM

Last month, Tom Lawler posted on how the NAR estimates existing home inventory. The NAR does NOT aggregate data from the local boards (see Tom's post for how the NAR estimates inventory). Sometime this fall, the NAR will revise down their estimates of inventory and sales for the last few years. Also the NAR methodology for estimating sales and inventory will likely (hopefully) be changed.

While we wait for the NAR revisions, I think the HousingTracker / DeptofNumbers data that Tom mentioned might be a better estimate of changes in inventory (and always more timely). Ben at deptofnumbers.com is tracking the aggregate monthly inventory for 54 metro areas.

Click on graph for larger image in graph gallery.

This graph shows the NAR estimate of existing home inventory through June (left axis) and the HousingTracker data for the 54 metro areas through July. The HousingTracker data shows a steeper decline in inventory over the last few years (as mentioned above, the NAR will probably revise down their inventory estimates this fall).

Lawler wrote today:

The area covered by DON/HT does not necessarily track the nation as a whole. However, it’s listings have shown significantly larger YOY declines than has the NAR in its estimate of US existing homes for sale. Moreover, when I include areas I track that are not covered by DON/HT, I find that listings in June were down significantly more than the NAR shows. This may indicate that home sales this June were down more from a year ago than the NAR estimates suggest.

The second graph shows the year-over-year change in inventory for both the NAR and HousingTracker.

HousingTracker reported that the July listings - for the 54 metro areas - declined 11.1% from last year.

Of course there is a large percentage of distressed inventory, and various categories of "shadow inventory" too. But the decline in listed inventory will put less downward pressure on house prices and is something to watch carefully all year.

Here are the posts this month on June Home Sales and Prices:
• New Home Sales in June at 312,000 Annual Rate
• Existing Home Sales in June: 4.77 million SAAR, 9.5 months of supply
• Home Sales: Distressing Gap
• Graph Galleries: New Home Sales and Existing Home Sales

On House Prices:
• Case Shiller: Home Prices increase in May
• Real House Prices and Price-to-Rent
• Graph Galleries: Home Prices

by Calculated Risk on 7/27/2011 04:51:00 PM

From economist Tom Lawler:

[T]he rumor mill has it that the NAR is considering developing a “repeat transactions” price index, presumably based on property level data from various MLS across the country. NAR analysts have noted that the impact of the “mix of homes” on its median sales price had become even more dramatic over the past few years than was the case in the past, and some apparently have become resigned to the fact that the median is “no longer the message” when in comes to tracking home price trends.

The median price is useful for tracking prices when the mix of homes sold is stable. But the mix hasn't been stable for some time, and now most people follow Case-Shiller, CoreLogic and a few other price indexes.

by Calculated Risk on 7/27/2011 02:00:00 PM

Fed's Beige Book:

Reports from the twelve Federal Reserve Districts indicated that economic activity continued to grow; however, the pace has moderated in many Districts. The six Districts nearest the Atlantic seaboard reported a slowdown in activity since the previous Beige Book report; activity was little changed in the Atlanta District and unchanged or slightly improved in the Richmond District. Of the other six Districts, the Minneapolis District reported political and weather-related disruptions that temporarily slowed growth, and the Dallas District slowed to a moderate pace of growth. The remaining four Districts continued to grow modestly.
...
Consumer spending increased overall, with modest growth of nonauto retail sales in a majority of Districts. Falling gasoline prices throughout most of this reporting period may have encouraged a pickup in shopping trips and some additional spending since the previous Beige Book.
...
Manufacturing activity was reported as continuing to increase since the last report in all but two districts, although many noted that the pace of growth had slowed.
...
Labor market conditions remained soft in most Federal Reserve Districts. Employment, especially among temporary hiring agencies, improved in the Richmond District in recent weeks. Modest hiring increases, often within specific sectors such as advertising in the Boston District and manufacturing in the Cleveland District, contributed to modest overall employment gains.

And on real estate:

Residential real estate sales in almost all Districts were little changed from the last Beige Book. Activity edged up in the Richmond, Atlanta, and Minneapolis Districts. ... Increasing inventories of unsold homes in the Boston, New York, and Kansas City Districts have restrained building in the single-family housing sector. ... Since the previous Beige Book, construction and activity in the residential rental market have continued to improve in the New York, Chicago, Dallas, and San Francisco Districts.
...
Nonresidential real estate activity improved somewhat in the Boston, Philadelphia, Cleveland, Chicago, St. Louis, and Dallas Districts. The Chicago District reported strong demand for industrial facilities, particularly from the automotive sector. The Philadelphia District reported improvements in terms of lower vacancy rates for office space, industrial space, and apartments; the Chicago District reported generally lower vacancy rates. The New York, Richmond, Atlanta, Minneapolis, Kansas City, and San Francisco Districts all reported generally weak activity in nonresidential real estate.

This was based on data gathered before July 15th, and I've heard reports of further slowing since the middle of the month.

by Calculated Risk on 7/27/2011 11:06:00 AM

It looks like they're going to need a bigger bailout ...

From Reuters: Italian banks fall as Italy/Bund spread widens

The Italian BTP spread over German Bunds expanded by 15 basis points to 305 basis points early on Wednesday. The BTP/Bund yield gap was at around 290 basis points late on Tuesday

Here is a graph of the 10 year spread (Italy to Germany) from Bloomberg. This is probably the key graph to watch right now.

And from CNBC: S&P Expects Second Greek Haircut, New Downgrade

A new and bigger restructuring of Greek debt is likely within the next two years, an official from credit ratings agency Standard & Poor's said on Tuesday, adding a further downgrade of Greece's sovereign debt rating was "pretty certain."

Here are the links for bond yields for several countries (source: Bloomberg):

Greece	2 Year	5 Year	10 Year
Portugal	2 Year	5 Year	10 Year
Ireland	2 Year	5 Year	10 Year
Spain	2 Year	5 Year	10 Year
Italy	2 Year	5 Year	10 Year
Belgium	2 Year	5 Year	10 Year
France	2 Year	5 Year	10 Year
Germany	2 Year	5 Year	10 Year