by Calculated Risk on 8/17/2010 07:08:00 PM

An update on a theme ...

Click on graph for larger image in new window.

This graph shows single family housing starts and the unemployment rate through July (inverted).

You can see both the correlation and the lag. The lag is usually about 12 to 18 months, with peak correlation at a lag of 16 months for single unit starts. The 2001 recession was a business investment led recession, and the pattern didn't hold.

Usually housing starts and residential construction employment lead the economy out of a recession, but not this time because of the huge overhang of existing housing units. After rebounding a little - mostly because of the home buyers tax credit - housing starts (blue) have moved mostly sideways (and down some recently).

This is what I expected when I first posted the above graph a year ago. I wrote:

[T]here is still far too much existing home inventory, a sharp bounce back in housing starts is unlikely, so I think ... a rapid decline in unemployment is also unlikely.

Usually near the end of a recession, residential investment (RI) picks up as the Fed lowers interest rates. This leads to job creation and also household formation - and that leads to even more demand for housing units - and more jobs, and more households - a virtuous cycle that usually helps the economy recovery.

Note: RI is mostly new home sales and home improvement.

However this time, with the huge overhang of existing housing units, this key sector isn't participating. So in this recovery there is less job creation, less household formation, and less demand for housing units than a normal recovery. This is sort of a circular trap for both GDP growth and employment.

This is one of the reasons I expect the unemployment rate to tick up over the next several months.

by Calculated Risk on 8/17/2010 03:30:00 PM

From DataQuick:

A total of 18,946 new and resale homes were sold in Los Angeles, Riverside, San Diego, Ventura, San Bernardino and Orange counties in July. That was down 20.6 percent from 23,871 in June, and down 21.4 percent from 24,104 for July 2009, according to MDA DataQuick of San Diego.

This was the slowest July since 2007, when 17,867 homes were sold, and the second-slowest since July 1995, when 16,225 sold.

Other reports show similar declines as home sales fell sharply all across the country in July. We are now seeing double-digit months of supply nationwide - and we might even see the months of supply metric reach 1 year. The previous months-of-supply high for this downturn was 11.2 months in 2008.

Housing economist Thomas Lawler's preliminary estimate for existing home sales in July is 3.95 million SAAR. If so, this would be fewest sales since 1996. Lawler's estimate for inventory in July was 4.04 million (although it is a bit of a mystery how the NAR calculates inventory). That would mean 12.3 months of supply!

A normal housing market usually has under 6 months of supply. The following graph shows the relationship between supply and house prices (using Case-Shiller).

Click on graph for larger image in new window.

This graph show months of supply and the annualized change in the Case-Shiller Composite 20 house price index.

Below 6 months of supply (blue line) house prices are typically rising (black line).

Above 6 or 7 months of supply, house prices are usually falling. This isn't perfect - it is just a guideline. Over the last year, there have been many programs aimed at supporting house prices, and house prices increased slightly even with higher than normal supply. However those programs have mostly ended.

The dashed red line is the estimate for months of supply in July. Through the roof! And I expect we will see double-digit months-of-supply for a number of months.

This is a key reason why I expect house prices to fall further later this year as measured by the Case-Shiller and CoreLogic repeat sales house price indexes, although I don't expect huge declines like in 2008.

by Calculated Risk on 8/17/2010 02:15:00 PM

From Minneapolis Fed President Narayana Kocherlakota: Inside the FOMC

The FOMC’s decision has had a larger impact on financial markets than I would have anticipated. My own interpretation is that the FOMC action led investors to believe that the economic situation in the United States was worse than they, the investors, had imagined. In my view, this reaction is unwarranted. The FOMC’s decisions were largely predicated on publicly available data about real GDP, its various components, unemployment, and inflation. I would say that there is no new information about the current state of the economy to be learned from the FOMC’s actions or its statement.

Kocherlakota points out that the Fed's balance sheet was falling quicker than anticipated because of the high level of refinancing as mortgage rates have declined.

But Kocherlakota fails to note that the mortgage rates have declined because of the weaker economy - and the Fed appears to be behind the curve in adjusting their views lower.

Kocherlakota is forecasting that real GDP growth in the 2nd half of 2010 will be about the same as in the first half:

Based on estimates from our Minneapolis forecasting model, I expect GDP growth to be around 2.5 percent in the second half of 2010 and close to 3.0 percent in 2011. There is a recovery under way in the United States, and I expect it to continue.

Although Kocherlakota forecast is possible - and is a weak recovery - I think the economy will slow in the 2nd half.

And I think the growing view isn't that the economy is worse than investors had imagined, but that the Fed is once again behind the curve on the economic outlook.

by Calculated Risk on 8/17/2010 11:49:00 AM

This morning the Census Bureau released the "Quarterly Starts and Completions by Purpose and Design" report for Q2 2010.

Click on graph for larger image in new window.

This graph shows the NSA quarterly starts intent for four categories since 1975: single family built for sale, owner built (includes contractor built for owner), starts built for rent, and condos built for sale.

Condo starts in Q2 were just above the all time record low (4,000 vs 3,000 in Q4 2009).

Units built for rent set an all time record low in Q1 (19,000 units in Q1 2010) and rebounded to 31,000 units in Q2. Some of this increase is seasonal, but it does appear that many large apartment owners think the vacancy rate has peaked - and some builders and owners are starting to build new apartments (probably for delivery in 2011).

Starts for owner built units increased too. In Q2, there were 44,000 owner built units started - up from 38,000 in Q2 2009.

And the largest category - starts of single family units, built for sale - increased to 93,000 in Q2 from 86,000 in Q1. Some of this was seasonal, and some was related to the tax credit (although most of the tax credit starts were probably in Q1).

Of course single family starts, built for sale, will decline sharply in Q3.

Comparing Housing Starts and New Home Sales

Monthly housing starts (even single family starts) cannot be compared directly to new home sales, because the monthly housing starts report from the Census Bureau includes apartments, owner built units and condos that are not included in the new home sales report.

However it is possible to compare "Single Family Starts, Built for Sale" to New Home sales on a quarterly basis. This is not perfect because of reporting differences and changes in cancellation rate - but it is close. The quarterly report shows that there were 93,000 single family starts, built for sale, in Q2 2010, and that is just below the 97,000 new homes sold for the same period. This data is Not Seasonally Adjusted (NSA).

This suggests that home builders are starting about the same number of homes that they are selling (unlike in 2005 and 2006 when builders built far too many spec homes).

by Calculated Risk on 8/17/2010 09:15:00 AM

From the Fed: Industrial production and Capacity Utilization

Industrial production rose 1.0 percent in July after having edged down 0.1 percent in June, and manufacturing output moved up 1.1 percent in July after having fallen 0.5 percent in June. A large contributor to the jump in manufacturing output in July was an increase of nearly 10 percent in the production of motor vehicles and parts; even so, manufacturing production excluding motor vehicles and parts advanced 0.6 percent. The output of mines rose 0.9 percent, and the output of utilities increased 0.1 percent. At 93.4 percent of its 2007 average, total industrial production in July was 7.7 percent above its year-earlier level. The capacity utilization rate for total industry moved up to 74.8 percent, a rate 5.7 percentage points above the rate from a year earlier but 5.8 percentage points below its average from 1972 to 2009.

Click on graph for larger image in new window.

This graph shows Capacity Utilization. This series is up 9.8% from the record low set in June 2009 (the series starts in 1967).

Capacity utilization at 74.8% is still far below normal - and well below the the pre-recession levels of 81.2% in November 2007.

Note: y-axis doesn't start at zero to better show the change.

The second graph shows industrial production since 1967.

This is the highest level for industrial production since Oct 2008, but production is still 7.3% below the pre-recession levels at the end of 2007.

The increase in July was above the consensus of a 0.5% increase in Industrial Production, and an increase to 74.5% for Capacity Utilization.