by Calculated Risk on 9/08/2010 10:00:00 AM

Note: The temporary decennial Census hiring and layoffs has distorted this series over the last few months. The total separations has increased, but that includes the temporary Census workers that were let go.

From the BLS: Job Openings and Labor Turnover Summary

There were 3.0 million job openings on the last business day of July 2010, the U.S. Bureau of Labor Statistics reported today. The job openings rate increased over the month to 2.3 percent. The hires rate (3.3 percent) and the separations rate (3.4 percent) were unchanged....

Note: The difference between JOLTS hires and separations is similar to the CES (payroll survey) net jobs headline numbers. The CES (Current Employment Statistics, payroll survey) is for positions, the CPS (Current Population Survey, commonly called the household survey) is for people.

The following graph shows job openings (purple), hires (blue), Total separations (include layoffs, discharges and quits) (red) and Layoff, Discharges and other (yellow) from the JOLTS.

Unfortunately this is a new series and only started in December 2000.

Click on graph for larger image in new window.

Notice that hires (blue) and separations (red) are pretty close each month. In July, about 4.4 million people lost (or left) their jobs, and 4.23 million were hired (this is the labor turnover in the economy) for a loss of 168,000 jobs in July (this includes Census jobs lost).

The employment report (revised) showed a loss of only 54,000 jobs in July, and usually these numbers are pretty close, so this is a little puzzling. I expect some revisions to one or both reports.

When the hires (blue line) is above total separations, the economy is adding net jobs, when the blue line is below total separations (as in July), the economy is losing net jobs.

It appears job openings are still trending up, however labor turnover is still fairly low.

by Calculated Risk on 9/08/2010 07:32:00 AM

The MBA reports: Mortgage Purchase Applications Up, Refinance Applications Fall Slightly in Latest MBA Weekly Survey

The Refinance Index decreased 3.1 percent from the previous week. The seasonally adjusted Purchase Index increased 6.3 percent from one week earlier.
...
“Purchase applications increased last week, reaching the highest level since the end of May. However, purchase activity remains well below levels seen prior to the expiration of the homebuyer tax credit, and is almost 40 percent below the level recorded one year ago,” said Michael Fratantoni, MBA’s Vice President of Research and Economics. “On the other hand, refinance volume dropped last week for the first time in six weeks, but the level of applications to refinance remains close to recent highs, as historically low mortgage rates continue to draw borrowers into the market.”
...
The average contract interest rate for 30-year fixed-rate mortgages increased to 4.50 percent from 4.43 percent, with points decreasing to 0.96 from 1.34 (including the origination fee) for 80 percent loan-to-value (LTV) ratio loans.

Click on graph for larger image in new window.

This graph shows the MBA Purchase Index and four week moving average since 1990.

As the MBA's Fratantoni noted, "purchase activity remains well below levels seen prior to the expiration of the homebuyer tax credit" and this suggests existing home sales in August, September and even October, will only be slightly higher than in July.

by Calculated Risk on 9/07/2010 10:21:00 PM

David Leonhardt writes at the NY Times Economix: Mortgage Rates and Home Prices

Interest rates are historically low right now. They will surely rise at some point. All else equal, higher rates should push home prices down.
...
[see graph]
...
It’s not easy to see much of a relationship.
...
My best guess for why the two don’t correlate more closely is the role that psychology plays in housing markets. Prices just don’t move as quickly as economic theory suggests they should.

Nope.

I've tried to explain this several times in several different ways. Price is what you pay for something. Interest rates are related to how the item is financed. Some people pay cash for a house. Would they pay more because interest rates are low? Nope.

Imagine just one buyer gets a special interest rate. Would that lucky buyer be willing to pay more than all other buyers for the same property? Nope.

It is true that low rates make buying more attractive as compared to renting. And that can increase the demand for buying - and more demand might mean slightly higher prices. But if rates are low, a rational buyer will expect mortgages rates to rise when they sell the property, and under the theory that mortgage rates impact price, the price will then fall in the future. That makes the property less attractive, and the buyer in the low interest rate environment will not want to overpay for the house.

So the buyer needs to consider both current interest rates and future interest rates, and by the time they are done doing all the calculations, you get the graph that Leonhardt shows. And that is exactly what I'd expect - there is little relationship between house prices and mortgage rates. That doesn't surprise me at all.

Note: When I wrote about this in 2005, I used a car example. Would people pay more for a car if interests rates are low?

by Calculated Risk on 9/07/2010 06:17:00 PM

The following graph shows total housing starts and the percent vacant housing units (owner and rental) in the U.S.

Note: this is a combined vacancy rate based on the Census Bureau vacancy rates for owner occupied and rental housing through Q2 2010.

Click on graph for larger image in new window.

The vacancy rate continued to climb even after housing starts fell off a cliff. Initially this was because of a significant number of completions. Then some hidden inventory (like some 2nd homes) probably became available for sale or for rent, and also some households have doubled up because of tough economic times.

It appears the total vacancy rate might have peaked and started to decline. This suggests more households are now being formed than net housing units added to the housing stock. But there is still a long way to go.

I know I'm a broken record - but it is very unlikely that there will be a strong rebound in housing starts with a near record number of vacant housing units.

by Calculated Risk on 9/07/2010 02:51:00 PM

The Obama administration is proposing that businesses be able to expense new investment in plants and equipment, through 2011, instead of writing off the investment over several years.

Catherine Rampell at the NY Times Economix provides a nice summary: Reactions to Obama’s Business Tax Write-Off Proposals

A few excerpts:

From Goldman Sachs on the "small effect":

To the extent it does have an effect, it is likely to pull forward demand into the quarter just before expiration (in this case Q4 2011) so the near-term effect should be even more modest ...

From Professor Greg Mankiw:

"[T]he impact will be relatively modest. Notice that expensing merely accelerates deductions. Thus, the value to the firm depends on interest rates. With interest rates near zero, the impetus to investment is small. Put another way, this policy can be seen as giving firms a zero-interest loan if they invest in equipment. But with interest rates near zero anyway, the value of the loan is not that great.”

This is basically a large sounding proposal ($200 billion) with little impact. With excess capacity in most sectors, why do we want to incentivize companies to invest anyway? And as Goldman notes, most of the modest impact will probably be in Q4 2011.

by Calculated Risk on 9/07/2010 12:04:00 PM

One of the key questions for housing, jobs, and the economy is: When will the excess housing supply be absorbed?

The answers depends on:
1) The current number of excess housing units,

2) how many net units are added to the housing stock, and

3) how many net households are being formed.

There is no timely data for net household formation, and estimates of the excess housing supply vary widely. So the answer involves some guesses (I'll get back to these questions).

The best data is for completions of housing units - although the number of demolitions is unclear. So the limited purpose of this post to to provide an estimate of the net units added to the housing stock in 2010.

Housing units include single family homes (included as 1 to 4 units), apartments (5+ units), and mobile homes. Demolitions are subtracted from the stock.

NOTE: Table is based on Completions. Housing units added to stock:

	2009	20101
1 to 4 units	534.7	480
5+ units	259.8	135
Mobile Homes	52.2	55
Sub-Total	846.7	670.0
Demolitions2	200	200
Total	646.7	470.0

¹ Estimates for 2010 based on completions through July.
² estimated.

Notice for 2010 that the estimate is for 5+ unit completions to collapse. This is already in the works as shown in the following diagram:

Click on graph for larger image in new window.

The blue line is for multifamily starts and the red line is for multifamily completions. Probably all the multifamily units that will be delivered in 2010 have already been started since, according to the Census Bureau, it takes on average over 1 year to complete these projects.

Since multifamily starts collapsed in 2009, completions will collapse in 2010. This finally showed up in the data for July as completions fell to 8.1 thousand from 17.1 thousand in June. Completions averaged over 14 thousand per month during the first 6 months of 2010, and will average close to 8 thousand per month during the 2nd half of 2010.

Note: this decline in completions will impact construction employment in the 2nd half.

Similar logic applies to single family units, although these only take around 7 months to complete. Almost all of the single family units that will be completed this year have already been started. There were 249.2 thousand completions during the first 6 months of the year, and completions will probably slow in the 2nd half.

The manufactured homes data is from the Census Bureau (and demolitions are estimated).

This means a record low number of housing units will be added to the housing stock in 2010. But unfortunately household formation is probably very low too - so the excess inventory might not be reduced substantially (I'll get back to this).

UPDATE: To be clear - the record low completions is bad news in the short term, especially for employment, but overall it is GOOD news for the economy and housing since it helps reduce the excess supply.

Special thanks to housing economist Tom Lawler who shared with me some of his thoughts on completions.

by Calculated Risk on 9/07/2010 09:03:00 AM

After the WSJ story last night on the European stress tests, here is an update on a few European bond spreads:

The 10-year Ireland-to-Germany bond spread has risen to 376 bps. This spread is larger than during the financial crisis in May when the spread peaked at 306 bps.

The 10-year Greece-to-Germany bond spread is now 946 bps, just below the peak level of 963 bps in May.

The 10-year Portugal-to-Germany bond spread is now 351 bps, just above the peak in May of 349 bps.

by Calculated Risk on 9/06/2010 09:30:00 PM

From the WSJ tonight: Europe's Bank Stress Tests Minimized Debt Risk

We've discussed this several times - as an example, in Part 5D of the sovereign debt series, "some investor guy" wrote:

Q1. Was there much sovereign stress in the European bank stress tests?

NO. The most glaring oversight, in the opinion of the author and many other analysts, is assuming there would be no sovereign defaults, and thus not showing any losses on the bank’s long term holdings (in the banking category vs the “trading book”). According to the Committee of European Banking Supervisors, the sovereign stress scenario results in “39 billion euro associated with valuation losses of sovereign exposures in the trading book “.

“The haircuts are applied to the trading book portfolios only, as no default assumption was considered, which would be required to apply haircuts to the held to maturity sovereign debt in the banking book.”

Note: Here are links for the entire sovereign debt series.

Here is the weekly summary for last week (a busy week).

And the Economic Schedule for this week (a light week).

by Calculated Risk on 9/06/2010 04:10:00 PM

Every month the BLS puts out a report that discusses the difference between the household and establishment surveys: Employment from the BLS household and payroll surveys: summary of recent trends

The Unemployment Rate comes from the Current Population Survey (CPS: commonly called the household survey), a monthly survey of about 60,000 households.

The jobs number comes from Current Employment Statistics (CES: payroll survey), a sample of approximately 390,000 business establishments nationwide.

These are very different surveys: the CPS gives the total number of employed (and unemployed including the alternative measures), and the CES gives the total number of positions (excluding some categories like the self-employed, and a person working two jobs counts as two positions).

The linked monthly report from the BLS discusses the differences, and adjusts the household survey to "an employment concept more similar to the payroll survey’s".

Click on graph for larger image in new window.

This graph from the BLS shows the household survey, the payroll survey and the adjusted household survey.

I was inspired to post this graph by Professor Nancy Folbre's post at Economix: Taking the ‘Un’ Out of Unemployment

A focus on employment, rather than unemployment, provides additional perspective. ...

The employment measure is unaffected by assumptions regarding the character, motives or incentives facing the unemployed.

And trends in this measure, as shown above, could discourage even the most optimistic among us, if they would just pay attention to it.

As Steven Hipple, a Bureau of Labor Statistics economist, puts it in a more detailed analysis of trends through the end of 2009, “Economic decision-makers might not understand the depth of the economic hole in the labor market.”

Little employment growth for a decade is quite a "hole".

Note: Over the same decade, according to the Census Bureau, the U.S. population, has increased from around 285 million to 310 million.

by Calculated Risk on 9/06/2010 12:06:00 PM

From the NY Times: Obama to Call for $50 Billion Spending on Public Works

President Obama on Monday is to call for as much as $50 billion in government spending to start up a long-term public works plan emphasizing transportation projects – roads, rail and airport runways – over the next six years.
...
The spending is one part of a broader economic recovery package that Mr. Obama is to unveil during a speech in Cleveland on Wednesday.
...
While Mr. Obama’s plan would call for investment over six years, the White House says it would be front-loaded with an initial investment of $50 billion in taxpayer money, to help create jobs in the shorter term.

Based on some media leaks, I'm not optimistic that the recovery package to be announced on Wednesday will help significantly with (un)employment. However this investment seems to make sense.