by Calculated Risk on 10/23/2015 11:41:00 AM

From the Philly Fed:

The Federal Reserve Bank of Philadelphia has released the coincident indexes for the 50 states for September 2015. In the past month, the indexes increased in 41 states, decreased in six, and remained stable in three, for a one-month diffusion index of 70. Over the past three months, the indexes increased in 43 states, decreased in six, and remained stable in one, for a three-month diffusion index of 74.

Note: These are coincident indexes constructed from state employment data. An explanation from the Philly Fed:

The coincident indexes combine four state-level indicators to summarize current economic conditions in a single statistic. The four state-level variables in each coincident index are nonfarm payroll employment, average hours worked in manufacturing, the unemployment rate, and wage and salary disbursements deflated by the consumer price index (U.S. city average). The trend for each state’s index is set to the trend of its gross domestic product (GDP), so long-term growth in the state’s index matches long-term growth in its GDP.

Click on graph for larger image.

This is a graph is of the number of states with one month increasing activity according to the Philly Fed. This graph includes states with minor increases (the Philly Fed lists as unchanged).

In September, 42 states had increasing activity (including minor increases).

The worst performing states over the last 6 months are West Virginia (coal), North Dakota (oil), Alaska (oil), Wyoming, and Oklahoma (oil).


Here is a map of the three month change in the Philly Fed state coincident indicators. This map was all red during the worst of the recession, and is mostly green now.

Note: Blue added for Red/Green issues.

by Calculated Risk on 10/23/2015 10:05:00 AM

From Black Knight: Black Knight Financial Services' First Look at September Mortgage Data: Delinquency Rate Rises for Second Consecutive Month, Now 4.9 Percent

According to Black Knight's First Look report for September, the percent of loans delinquent increased 1.7% in September compared to August, and declined 13.9% year-over-year.

The percent of loans in the foreclosure process declined 1.5% in September and were down 23% over the last year.

Black Knight reported the U.S. mortgage delinquency rate (loans 30 or more days past due, but not in foreclosure) was 4.87% in September.

The percent of loans in the foreclosure process declined in September to 1.46%.

The number of delinquent properties, but not in foreclosure, is down 392,000 properties year-over-year, and the number of properties in the foreclosure process is down 214,000 properties year-over-year.

Black Knight will release the complete mortgage monitor for September in early November.

by Calculated Risk on 10/22/2015 06:11:00 PM

Following the release today of the National Multifamily Housing Council (NMHC) quarterly apartment survey, I had an email exchange with NMHC chief economist Dr. Mark Obrinsky. I've known Mark for several years, and he has helped me understand the apartment market (and other economic topics). His bio is here.

McBride: I’ve found the NMHC quarterly apartment survey very useful in analyzing the apartment market. I understand the indexes are standard diffusion indexes. Could you share with us the coverage of the survey?

Obrinsky: We survey NMHC members who are apartment owners, property managers, developers, brokers, investors, and lenders.

McBride: In the October 2015 survey released today, the tightness index was at 53 (Any reading above 50 indicates tighter conditions from the previous quarter). This would seem to suggest upward pressure on rents, and downward pressure on the vacancy rate. Is that a reasonable interpretation?

Obrinsky: Indeed, a market tightness index reading above 50 indicates higher rents, lower vacancies, or both.

Note that in the most recent survey, 60% of respondents indicated conditions are unchanged from 3 months earlier; 23% indicated tighter conditions, 16% indicated looser conditions. A reasonable interpretation would be that most markets saw little change – and that the number of markets where conditions tightened was a little higher than the number of markets where conditions loosened.

McBride: Other sources, such as the Reis apartment survey (large cities only) seem to suggest the vacancy rate has bottomed. Reis Senior Economist and Director of Research Ryan Severino recently wrote:

“It appears as if the market has finally reached its inflection point during the third quarter. Although the national vacancy increase of 10 basis points was slight, it was actually a slight acceleration of a trend that began during the second quarter of 2014. ... Importantly, this rise in vacancy has occurred without the deluge of new supply that is in the pipeline but has not yet hit the market. When that occurs, likely in the next few quarters, vacancy increases are sure to accelerate because the market will not be able to digest that much new product.”

Do you think the vacancy rate has bottomed or is near a bottom?

Obrinsky: By most measures, the national vacancy rate for apartments is quite low. For example, the Census Bureau’s estimate for rental units in buildings with at least 5 units is at its lowest point since 1984. Some private data providers estimate the vacancy rate for investment grade apartments at 4% or less. Thus, there isn’t much room for the vacancy rate to decline from here, especially with new construction finally getting close to the level needed to meet the increase in demand for apartments.

That said, the vacancy rate is only one indicator, and no single indicator is likely to capture the dynamics of the industry. Higher demand can show up as either a lower vacancy rate or higher rent growth or both. In today’s market, the impact of strong demand is evidenced not only in the low vacancy rate, but even more so in rent growth. Several data sources show rent growth at or above 5% annually, and by one measure rent growth is the strongest since days of the tech bubble (when national averages were skewed by the outsized rent growth on the West Coast, particularly in northern California). This is especially interesting considering that rent growth slowed in 2012-2013, but has picked up strongly since then.

McBride: An indicator I follow for new multifamily construction is the AIA Architecture Billings Index. The sector index for the multi-family residential market was negative for the eighth consecutive month in September - and this might be indicating a slowdown for new apartment construction. Multifamily has been a significant driver of the rebound in total housing starts over the last several years, so a slowdown would be important. Do you expect multifamily housing starts to level out, or do you expect further growth?

Obrinsky: I have long said we need 300-400 thousand new apartments annually for an extended period of time to meet the ongoing demand increase plus pentup demand from the recession and its aftermath. (This range also includes the estimated 100-125 thousand units lost each year due to destruction, deterioration, or conversion to other use.) I would expect to see starts in this range for some time.

McBride: Back in April 2010 the tightness index increased significantly, and this was one of the first signs of the coming rental boom. Back then you and I discussed the favorable demographics for apartments, and also the impact of the housing bust and foreclosure crisis on the rental market. The foreclosure crisis is mostly behind us. What is your current view on demographics and the apartment market?

Obrinsky: I continue to think demographic trends are bullish for apartments. In particular:
• overall population growth (good for rental and for-sale housing). The UN projections show that only 7 countries will add more people over the next 10 years (and only 6 countries will add more people in the next 20 years).
• continuing trend in household composition: more single-person households (now the most common household type, surpassing married couples without children), single parents, and roommates. All these household types have a higher propensity to rent than the national average.
• immigration has bounced back from its recession-induced lows, and immigrants are more likely to rent than native born. And: newly arrived immigrants (less than 10 years in the US) are more likely to rent than immigrants who arrived earlier (in the US 10 years or more), but even immigrants who arrived earlier are more likely to rent than native born householders.
• young people are entering the housing market in large numbers (and will continue to do so for years), and young people are the most likely group to rent.
• baby boomer downsizing, giving up the large suburban single-family home to move to multifamily housing (condos and apartments) in close-in suburbs or cities. There is mixed evidence on this, but even if the share of older households who do this is unchanged from previous generations, the fact that there are so many more boomers than previous generations mean this could be an important source of demand.

And just to clarify: renting vs. owning doesn’t have to be – and through most of our postwar history wasn’t – a zero-sum game. We can have more of both. In saying that demographic trends are favorable for apartments, I am not saying that the homeownership rate will continue to fall indefinitely.

McBride: Thanks Mark!

by Calculated Risk on 10/22/2015 02:30:00 PM

From the National Multifamily Housing Council (NMHC): NMHC Quarterly Survey of Apartment Conditions October 2015

The Market Tightness Index decreased by 8 points from last quarter (and increased by 1 point from a year earlier) to 53. Thirty-one percent of respondents reported tighter conditions than three months ago.

This is the seventh consecutive quarter where the index indicates tighter conditions.  And the index indicated tighter conditions in 21 of the 23 quarters.


Click on graph for larger image.

This graph shows the quarterly Apartment Tightness Index. Any reading above 50 indicates tighter conditions from the previous quarter. This indicates market conditions were tighter over the last quarter.

As I've mentioned before, this index helped me call the bottom for effective rents (and the top for the vacancy rate) early in 2010.

by Calculated Risk on 10/22/2015 12:17:00 PM

From the Kansas City Fed: Tenth District Manufacturing Activity Steadied

The Federal Reserve Bank of Kansas City released the October Manufacturing Survey today. According to Chad Wilkerson, vice president and economist at the Federal Reserve Bank of Kansas City, the survey revealed that Tenth District manufacturing activity steadied somewhat and was expected to remain largely unchanged heading forward.

“Following six months of composite index readings of worse than -6, this month’s reading of -1 was somewhat encouraging,” said Wilkerson. “Modest increases in new orders and production nearly offset declines in employment, supplier delivery time, and inventory indexes.”
...
Tenth District manufacturing activity steadied somewhat, and expectations for future activity were largely flat following last month’s more negative reading. Most price indexes edged higher for the first time in several months.

The month-over-month composite index was -1 in October, up from -8 in September and -9 in August ...
emphasis added

The earlier decline in the Kansas City region manufacturing was probably mostly due to lower oil prices, although respondents are also blame weaker exports on the strong dollar.

by Calculated Risk on 10/22/2015 10:11:00 AM

From the NAR: Existing-Home Sales Regain Momentum in September

Total existing–home sales, which are completed transactions that include single–family homes, townhomes, condominiums and co–ops, increased 4.7 percent to a seasonally adjusted annual rate of 5.55 million in September from a slightly downwardly revised 5.30 million in August, and are now 8.8 percent above a year ago (5.10 million). ...

Total housing inventory at the end of September decreased 2.6 percent to 2.21 million existing homes available for sale, and is now 3.1 percent lower than a year ago (2.28 million). Unsold inventory is at a 4.8–month supply at the current sales pace, down from 5.1 months in August.

Click on graph for larger image.

This graph shows existing home sales, on a Seasonally Adjusted Annual Rate (SAAR) basis since 1993.

Sales in September (5.55 million SAAR) were 4.7% higher than last month, and were 8.8% above the September 2014 rate.

The second graph shows nationwide inventory for existing homes.

According to the NAR, inventory decreased to 2.21 million in September from 2.27 million in August.   Headline inventory is not seasonally adjusted, and inventory usually decreases to the seasonal lows in December and January, and peaks in mid-to-late summer.

The third graph shows the year-over-year (YoY) change in reported existing home inventory and months-of-supply. Since inventory is not seasonally adjusted, it really helps to look at the YoY change. Note: Months-of-supply is based on the seasonally adjusted sales and not seasonally adjusted inventory.

Inventory decreased 3.1% year-over-year in September compared to September 2014.  

Months of supply was at 4.8 months in September.

This was above expectations of sales of 5.35 million. For existing home sales, a key number is inventory - and inventory is still low. I'll have more later ...

by Calculated Risk on 10/22/2015 08:34:00 AM

The DOL reported:

In the week ending October 17, the advance figure for seasonally adjusted initial claims was 259,000, an increase of 3,000 from the previous week's revised level. The previous week's level was revised up by 1,000 from 255,000 to 256,000. The 4-week moving average was 263,250, a decrease of 2,000 from the previous week's revised average. This is the lowest level for this average since December 15, 1973 when it was 256,750. The previous week's average was revised up by 250 from 265,000 to 265,250.

There were no special factors impacting this week's initial claims.

The previous week was revised up to 256,000.

The following graph shows the 4-week moving average of weekly claims since 1971.

Click on graph for larger image.


The dashed line on the graph is the current 4-week average. The four-week average of weekly unemployment claims decreased to 263,250.  This is the lowest level in over 40 years.

This was below the consensus forecast of 265,000, and the low level of the 4-week average suggests few layoffs.

by Calculated Risk on 10/21/2015 06:57:00 PM

Here is a hint ... take the "over" on existing home sales.

Thursday:
• At 8:30 AM ET, the initial weekly unemployment claims report will be released. The consensus is for 265 thousand initial claims, up from 255 thousand the previous week.

• Also at 8:30 AM, the Chicago Fed National Activity Index for September. This is a composite index of other data.

• At 9:00 AM, the FHFA House Price Index for August 2015. This was originally a GSE only repeat sales, however there is also an expanded index. The consensus is for a 0.5% month-to-month increase for this index.

• At 10:00 AM, Existing Home Sales for September from the National Association of Realtors (NAR). The consensus is for 5.35 million SAAR, up from 5.31 million in August. Economist Tom Lawler estimates the NAR will report sales of 5.56 million SAAR. A key will be the reported year-over-year change in inventory of homes for sale.

• At 11:00 AM, the Kansas City Fed manufacturing survey for September.

• During the day: Q3 NMHC Apartment Tightness Index.

by Calculated Risk on 10/21/2015 02:18:00 PM

I've been asked again about the relationship between initial unemployment claims and monthly payroll employment. Why are claims so low, yet employment gains have slowed?

Here is a repeat of a previous answer with updated graphs. There is definitely a general relationship between payroll employment and unemployment claims as shown in the first graph.  Note that unemployment claims are graphed inverted.

Note: For smoothing, this graph use a 3-month centered average of net payroll employment, and the 4-week average of initial unemployment claims.

Click on graph for larger image.

A few observations:
1) Even with a "low level" of initial weekly claims, there are a large number of claims per week (and per year). If there were 260,000 initial weekly claims per week, that would mean 13 million layoffs per year! However, some of these layoffs are regular - as an example when workers are furloughed (common in some industries) they are eligible for unemployment benefits.

2) Unemployment benefits have been trending down over time.  This is probably because of changes in hiring practices.

3) Following the recession, a number of analysts pointed out that when claims dropped below 400 thousand per week, the economy would probably start adding jobs.  That was pretty close, but a rough number.

4) Even though there is a general relationship, claims do not suggest a coming surge in employment. As the economy has improved, it is easier to find a new job - so some people who might have filed for unemployment don't because they find new employment.

Each month, when I post an "employment preview", I look at weekly claims (especially for the BLS reference week).  This seems to provide a hint - sometimes.

Talking about turnover, the second graph is from JOLTS that I post each month (Job Openings and Labor Turnover Survey).

Note that hires (dark blue) and total separations (red and light blue columns stacked) are pretty close each month. This is a measure of labor market turnover.  When the blue line is above the two stacked columns, the economy is adding net jobs - when it is below the columns, the economy is losing jobs.

In August there were almost 5.1 million workers hired, and about 4.9 million total separations (Layoffs, quits, and other). That is a significant amount of turnover each month.

by Calculated Risk on 10/21/2015 10:17:00 AM

Note: This index is a leading indicator primarily for new Commercial Real Estate (CRE) investment.

From the AIA: Strong Rebound for Architecture Billings Index

The Architecture Billings Index (ABI) returned to positive territory after a slight dip in August, and has seen growth in six of the nine months of the year so far. As a leading economic indicator of construction activity, the ABI reflects the approximate nine to twelve month lead time between architecture billings and construction spending. The American Institute of Architects (AIA) reported the September ABI score was 53.7, up from a mark of 49.1 in August. This score reflects an increase in design services (any score above 50 indicates an increase in billings). The new projects inquiry index was 61.0, down from a reading of 61.8 the previous month.

“Aside from uneven demand for design services in the Northeast, all regions are project sectors are in good shape,” said AIA Chief Economist Kermit Baker, Hon. AIA, PhD. “Areas of concern are shifting to supply issues for the industry, including volatility in building materials costs, a lack of a deep enough talent pool to keep up with demand, as well as a lack of contractors to execute design work.”
...
• Regional averages: South (54.5), Midwest (54.2), West (51.7) Northeast (43.7)

• Sector index breakdown: mixed practice (52.6), institutional (51.5), commercial / industrial (50.9) multi-family residential (49.5)

emphasis added

Click on graph for larger image.

This graph shows the Architecture Billings Index since 1996. The index was at 53.7 in September, up from 49.1 in August. Anything above 50 indicates expansion in demand for architects' services.

Note: This includes commercial and industrial facilities like hotels and office buildings, multi-family residential, as well as schools, hospitals and other institutions.

The multi-family residential market was negative for the eighth consecutive month - and this might be indicating a slowdown for apartments - or at least less growth.

According to the AIA, there is an "approximate nine to twelve month lag time between architecture billings and construction spending" on non-residential construction.  This index was positive in 9 of the last 12 months, suggesting a further increase in CRE investment over the next 12 months.