as is usually the case, the Bureau of Labor Statistics issued two employment reports under the guise of one Employment Situation Summary for October, leading to much confusion among media pundits who have come to think of them as one…the Current Employment Statistics (CES), which is a BLS survey of roughly 145,000 businesses and government agencies out of a total of roughly 557,000 US employers indicated that seasonally adjusted nonfarm payroll employment increased by 204,000 during October, while the data collected by the Current Population Survey (CPS) of roughly 60,000 US households conducted by the Census Bureau was extrapolated to indicate there were 735,000 less of us employed during the month…part of the divergence between the surveys was because the reference week for the surveys, as normally including the 12th of the month, coincided with the government shutdown and associated worker furloughs, which should have had the affected individuals classified in the household survey as unemployed on temporary layoff; however, there was not a corresponding increase in the count of the unemployed, which only rose by 17,000 over the month, as the civilian labor force fell by 720,000, and those of us not in the labor force rose by 932,000, dropping the labor force participation rate 0.4% to 62.8%, the lowest in 35 years, back to well before Reagan busted the unions, forcing women into the workforce to make ends meet…

Establishments Report Decent October Job Creation

FRED Graph

the unadjusted CES non-farm payroll data indicated payroll employment of 137,540,000 in October, up 940,000 from September’s unadjusted 136,600,000 payroll jobs…after running through a program to adjust for normal seasonal employment differences, BLS generated seasonally adjusted totals of 136,350,000 payroll jobs in September and 136,554,000 jobs in October, resulting in the headline net job increase of 204,000…in addition, the seasonally adjusted payroll increase for September was revised to show job creation of 163,000, rather than the 148,000 reported two weeks ago, and also revised the adjusted August payroll job gains upward from 193,000 to 238,000, netting an additional 60,000 jobs in the revisions, all shown by our adjacent FRED bar graph…thus, over the past three months, the seasonally adjusted payroll jobs gain was 605,000, while the actual unadjusted increase in non-farm payrolls totaled 1,963,000 over the same period…

most of the job creation in October was again in the same low paying sectors that have accounted for the lion’s share of the new jobs during the recovery…of the 204,000 seasonally adjusted jobs added in October, which was nearly twice the number expected, 53,000 were in leisure and hospitality, of which 29,300 were in food service, and 44,400 were in retail sales, with 11,500 of those working in food and beverage stores..another 44,000 jobs were generated in the broad professional and business services category, with 21,400 of those in professional and technical services and 15,400 in administrative and waste services…19,000 jobs were added in manufacturing, an area which has been weak over the past year despite logging decent job gains in 2011, and another 11,000 jobs were added in construction, split fairly evenly between residential construction and nonresidential specialty trade contractors; in addition, 17,500 more were also employed in health care and social assistance, with 11,000 of those in ambulatory care services…in other sectors, there was little or no job creation in resource extraction, wholesale trade, transportation and warehousing, information technology or financial industries, and the government sector showed a decrease of 8,000 jobs, as Federal agencies shed 12,000 while states added 7,000 and local governments cut 3,000…

our FRED bar graph below shows the monthly change in payroll employment in selected sectors over the past year and a half; jobs added are above the 0 line, with job cuts below it….in each monthly grouping of 8 bars, the monthly change in manufacturing employment for that month is indicated in blue, the change in monthly construction employment is in red, the monthly change in retail employment is in dark green, the monthly change in government jobs is in yellow, and the change in employment in professional and business services, which includes everything from systems design to garbage collection, is in grey; also included are the BLS employment subcategories of jobs in bars and restaurants in light green, and new health care jobs in orange, and non government jobs in education shown in violet…we can see strong job creation in retail (dark green), food service (light green) and health care (orange) with uneven job creation in manufacturing (blue) and construction (red), and more jobs cut than added in the government sector (yellow)

FRED Graph

data extracted from the CES also showed that the average workweek for all payroll employees in October was at 34.4 hours, unchanged from September; with the average workweek as high as 44.0 hours in mining and logging to a low average of 25.9 hours for those working at leisure and hospitality jobs…the manufacturing workweek was 40.9 hours and factory overtime was at 3.4 hours, both unchanged from September…the average workweek for nonsupervisory employees slipped a bit from 33.7 hours in September to 33.6 hours in October; again, those production workers in resource extraction logged the most average hours at 45.4, while grunt workers in leisure and hospitality averaged just 24.9 hours of work a week…average hourly pay for all workers was up 2 cents to $24.10 and ranged from a high of $35.08 an hour for the average utility worker to a low of $13.50 for the average worker in leisure and hospitality; meanwhile, average pay for nonsupervisory workers was also up by just 2 cents to $20.26; again, utility linemen garnered the highest average at $32.39 an hour, while the average non-supervisory worker in leisure and hospitality jobs made just 11.85 an hour…

Nearly a Million Drop Out of Labor Force; Participation Rate at 35 Year Low

it’s difficult to reconcile what appears to be modestly decent job creation indicated by the establishment survey with the data extrapolated from the household survey;  while the wide margin of error of +/- 300,000 in the monthly change in the number unemployed and the increase of 448,000 who were classified as unemployed on temporary layoff mostly due to the government shutdown might explain the fact that 735,000 less of us were classified as employed in October, it doesnt explain why the number of us “not in the labor force” rose by almost a million, as 932,000 of us apparently dropped out of the labor force altogether…the box under the heading “Partial Federal Government Shutdown” in the employment summary explains that some of the furloughed federal employees may have been even classified as employed but absent from work, an error which could have occurred if respondents misunderstood survey questions or interviewers recorded answers incorrectly, but whether they were classified as employed or unemployed, there is no way they should have been considered not in the labor force…nor should anyone temporary off work because of the shutdown, such as a restaurant employee servicing federal employees, have been classified as not in the work force…another speculation on what happened suggested that since the reference week of the survey was ran between October 6th and October 12th but the data collection for the survey didnt begin until the 20th, that some who were employed or who looked for work during that period may have forgotten their employment status during that week and hence been misclassified…in a survey of 60,000 meant to represent the employment status of a working age population of more than 246 million, the classification of each answer counts for the status of more than 4,000 individuals, so it’s easy to see how a few errors can throw the whole survey off…so it’s with a very large grain a salt that we’ll document what this Current Population Survey reveals…

FRED Graph

the seasonally adjusted extrapolation from this October survey of households indicated that 143,568,000 of us were employed, 735,000 less than were employed in September, while 11,272,000 of us were unemployed, an increase of 17,000 over September; as the working age population rose by 213,000 to 246,381,000 and the civilian labor force fell by 720,000 to 154,839,000…thus, those of us not in the labor force rose by 932,000 to a record 91,541,000, which was the third highest monthly increase of people dropping out of the labor force in US history…the results of this can be seen in the two major metrics computed from this survey that we watch; the labor force participation rate, which is the percentage of us working or actively looking for work, fell from 63.2% in September to a 35 year low of 62.8% in October, which is shown in red on our adjacent FRED graph, while the employment to population ratio, aka the employment rate, shown in blue, fell from 58.6% in September to a 2 year low of 58.3% in October; which in part reflects that the furloughed government workers were not counted as employed…

so, with 720,000 less of us counted than were included in September, the headline unemployment rate rose from 7.2% to 7.3%; if the 448,000 furloughed employees had been counted as unemployed, it should have added more than 0.2% to that ratethe number working part time for economic reasons, ie, those who reported they wanted full time work but had their hours cut or could only find part time employment, rose by 124,000 from 7,926,000 in September to 8,050,000 in October, while those working part time voluntarily fell by 181,000 to 18,786,000; the alternate measure of unemployment, U-6, which includes those who wanted full time work, rose from 13.6% in September to 13.8% in October…among those not officially in the labor force and hence not counted, an additional 5,683,000 reported that they still want a job; of those, 2,283 000 were categorized as “marginally attached to the labor force” because they’ve  looked for work sometime during the last year, but not during the 30 day period covered by the October household survey…815,000 of those were further characterized as “discouraged workers, because they say that they haven’t looked for work because they believe there are no jobs available to them…the number of “discouraged workers” was virtually unchanged from a year ago… 

Divergence Between BLS Employment Reports Widens

finally, we want to look at the raw unadjusted numbers from this survey and compare the seasonal adjustment thereon to the seasonal adjustment made on non-farm payrolls in the establishment survey…before adjustment, the extrapolated number who reported they were employed in October was 144,144,000, down 507,000 from the 144,651,000 who reported they were employed in September; thus the seasonal adjustment subtracted 228,000 from the change in the employed to arrive at a seasonally adjusted 735,000 decrease in those employed…as we documented last month, the seasonal adjustment in August household survey added 489,000 to the count of the employed, while the September household survey added 9,000, which, when combined with the August adjustment,  was a difference of 1,162,000 in the adjustments between the the two survey (ie, the CES seasonal adjustment subtracted 682,000 payroll jobs, while the CPS seasonal adjustment added 480,000 to the count of employed)…with October’s data, we now see a seasonal subtraction of 1,358,000 jobs from the establishment survey over 3 months, but still a seasonal addition of 268,000 to the household survey over the same three months…the point of this exercise is to illustrate how different the two unemployment reports we and others have been covering as one are; so different, in fact, that they shouldn’t even be released togetherour FRED graph below shows the unadjusted count of those employed from the household survey in blue and the unadjusted count of payroll jobs from the establishment survey in red since 2000….the scale for the household employed count in thousands is on the right sidebar, and the scale for the establishment job count is on the left sidebar, allowing the surveys, which actually count a different population, to overlap; it’s fairly clear in just looking at this graph that there are times when the lines match, while there are also times when each survey runs as many as 2 million jobs ahead of the other….you can even see the action we’ve described over the last three months, wherein the unadjusted payroll job count went up by 1,963,000, while the true count of the employed went down by 967,000, only to have them brought into approximate alignment by the seasonal adjustments..

FRED Graph 

Incomplete Advance GDP Data Better than Expected

the advance estimate of 3rd quarter Gross Domestic Product was surprisingly better than most forecasts, indicating that our seasonally adjusted output of goods and services grew at an annual rate of 2.8% over the three months through September vs. the previous three months…however, the internals were not as solid as the headline, as .83% of the 2.8% 3rd quarter growth increase came from increasing inventories, or gross production that’s still sitting on the shelf; take that out, and the real final sales of our domestic product just increased 2.0% in the third quarter, down from the real sales increase of 2.1% in the second…on the plus side, every major component of the economy – consumers, investment, net exports, and government – contributed to growth in the quarter for the first time in a year….in current dollars, the value of all our national output of goods and services — increased 4.8% or $196.6 billion, in the third quarter to an annualized level of $16,857.6 billion…in the inflation adjusted chained 2009 dollars, upon which the change in real GDP is calculated, third quarter GDP increased by $110.4 billion, from an annualized $15,679.7 billion in the second quarter to an annualized $15,790.1 at the end of the third…generally, the dollars amounts we’ll be looking at will be current dollars, while the percentage changes BEA gives us are adjusted for inflation based on chained 2009 dollars…

before we start with the specifics, we should also make it clear that all national accounts data is reported at a seasonally adjusted annual rate; thus, when it’s said a component of GDP grew at a 2.8% rate, it’s understood that it means that if the growth rate in the quarter were extrapolated over the entire year based on the rate the seasonally normalized 3rd quarter economy grew at, we’d see 2.8% growth over an entire year; that means that actual growth in the quarter was something close to a seasonally adjusted 0.7%…another important point the media and most analysts missed in their coverage of this report was the technical note accompanying this report (pdf), which notes that because of delayed reporting by government agencies, the source data was incomplete and subject to revision; namely that 3 months of source data were only available for consumer spending on goods; shipments of capital equipment; motor vehicle sales and inventories; durable and nondurable manufacturing inventories; federal government outlays; and consumer, producer, and international prices, while only two months of data were available for most other key data sources; the BEA’s “assumptions” for the third month are shown in a table included with the technical note (pdf); since the average change, either plus or minus, from the first to the second estimate of GDP is plus or minus 0.5% even under the best conditions, it’s a fair to say that this report is only a bit better than a wild guess at 3rd quarter GDP…

personal consumption expenditures, which accounts for approximately 70% of nominal GDP, grew at a 1.5% annual rate over the quarter, from $11,427.1 billion in the second quarter to $11,525.4 billion in the 3rd, which was the slowest growth in consumer spending since the 2nd quarter of 2011, and contributed just 1.04% to the 2.84% increase in GDP…our real, inflation adjusted spending for durable goods grew at a 7.8% rate and contributed 0.57% to the GDP change, our spending for nondurables rose 2.7% and added 0.42% to the growth rate, while our outlays for services grew at a 0.1% rate and contributed 0.05% to GDP..that was the slowest quarterly growth rate in services outlays of the recession.was precipitated by an inflation adjusted $11.2 billion decrease in outlays for housing and utilities coupled with much smaller increases in every other services spending subcategory…

gross private investment, which includes the aforementioned inventories, grew at a 9.5% annual rate to $2,689.8 billion and contributed 1.45% to the quarter’s growth rate; without inventories, fixed investment increased at an annual rate of 4.1% and contributed just 0.63% as residential investment grew at a 14.6% clip and added 0.43 while non-residential fixed investment increased at a 1.6% rate and added 0.20%…within non-residential fixed investment, investment in structures grew at a 12.3% annual rate and contributed 0.32% to GDP, investment in equipment fell at a 3.7% annual rate and subtracted 0.21% from the final GDP growth figure, and investment in intellectual property grew 2.2% and contributed 0.09% to the quarter’s growth rate…meanwhile, the nominal $110.3 billion increase in inventories was adjusted to an $86.0 billion quarterly change which added 0.83% to the 3rd quarter GDP; of that, $59.2 billion was an increase in business inventories, which added 0.71%, and $22.5 billion was the increase in farm inventories, which added 0.11%…

the next component of GDP, net exports, which are our exports minus our imports, have subtracted from GDP annually since 1976; however, what we’re looking at here is the change from the 2nd quarter to the 3rd quarter, and in that regard they’ve added, since exports increased more than imports over the 3 month period…real exports of goods and services increased 4.5% in the 3rd quarter and added 0.60% to GDP, while real imports increased 1.9%, and subtracted 0.30% from the final growth rate…

lastly, real net government consumption and investment increased $1.5 billion to $2,906.0 billion in chained 2009 dollars and added 0.04% to to the quarterly growth change; federal government expenditures and  investment decreased 1.7%, as defense spending fell 0.6% from quarter to quarter while non-defense spending fell 3.3% over the period; on net, federal contraction subtracted .13% from the quarter’s growth rate…picking up the slack and boosting the government contribution into the black were state and local governments, who saw consumption expenditures and gross investment increase at a 1.5% rate over the quarter, which added .17 to 3rd quarter growth…contributions from each of these components vis a vis the 2nd quarter contributions can be seen in the simple insert to the right above, which comes from ed dolan at economonitor; it shows the contributions to growth from each of the four major sectors of the economy; consumers, private investment, government and exports; note that they add up to a 2.84% growth rate, which is rounded to 2.8% because of the inexactness of these estimates…similarly, in the bar chart below from doug short, the contributions from each of those components for each quarter since the beginning of 2007 are color coded, with those that added to GDP above the ’0′ line, and those that subtracted for that quarter below it, with personal consumption spending in blue, investment in red, net exports (change in exports minus imports) in chartreuse, and government in violet…then the dashed blue line is the GDP reading for the quarter, which is the sum of those components…it’s clear that our recovery has been driven by consumers and investment, and that except for the stimulus, government cutbacks have been a drag..

Click to View

… 
a couple of private reports we also want to look at this week are the Mortgage Monitor for September (pdf) from LPS (Lender Processing Services) and the 3rd Quarter National Delinquency Survey from the MBA (Mortgage Bankers Association); both of these reports cover essentially the same data on home mortgages: those that are delinquent, or behind on their payments, and those that are in the process of being foreclosed; the LPS mortgage monitor is a monthly report that we’ve covered monthly for several years, while the MBA National Delinquency Survey is only released quarterly and is seasonally adjusted based on patterns of mortgage delinquency that have repeatedly appeared annually…since both of these reports are of mortgage conditions as of the last day of September, comparing them side by side should give us a better insight into the ongoing mortgage crisis than looking at one or the other in isolation…

MBA 3rd Quarter Delinquency Survey Shows Nearly 1 in 10 Homeowners Still Behind on Mortgage

according to the MBA, the seasonally adjusted national delinquency rate, which is the percentage of homeowners who were late at least one house payment late but not in foreclosure, fell to 6.41% of all mortgage loans outstanding at the end of the 3rd quarter, down from 6.96% at the end of the second quarter, and down from a delinquency rate of 7.40% at the end of the 3rd quarter a year ago, and the lowest level since lowest level since the second quarter of 2008…in addition, they report that the percentage of mortgage loans in the foreclosure process at the end of the third quarter was 3.08%, down from 3.33% at the end of the 2nd quarter and from 4.09% at the end of the 3rd quarter last year, and also the lowest percentage foreclosure inventory since 2008…new foreclosure actions were initiated on 0.61% of mortgages in the 3rd quarter, down from the 0.64% rate of new foreclosures in the 2nd quarter…the serious delinquency rate, which combines the percentage of mortgages in foreclosure with those that are more than 90 days behind on their housepayments but still not in foreclosure, fell to 5.65% in the 3rd quarter from 5.85% in the 2nd quarter and well below the serious delinquency rate of 7.03% a year earlier, although they warn that their reported improvement in that seriously delinquent percentage may be slightly less than reported because one large specialty servicer that has received a number of loan transfers does not participate in the MBA survey…combining those at least one payment overdue on their mortgage with those seriously delinquent or in foreclosure gives us a total percentage of 9.75% of homeowners who were behind on their mortgage at the end of the quarter, the first time overall delinquencies had been below 10% nationally in the MBA survey since mid 2008…this can be see in the bar graph below from Bill McBride at Calculated Risk, which stacks the percentage of foreclosures in red on the top of each quarterly bar, which each also shows the number of 30 day delinquencies reported by the MBA each quarter in violet, the number of 60 day delinquent mortgages each quarter in the blue portion of each bar, and the number of mortgages more than 90 days late in yellow…we can see on that graph that the percentage of mortgages in trouble peaked at 14.7% in the first quarter of 2010 and has been trending downward since, although it’s still well above the levels of the pre-crisis year of 2005, especially with regards to 90 day delinquencies and homes stuck in foreclosure…

3rd qtr MBA foreclosures and delinquency buckets

the next graph below is from the MBA and it compares the historical track of the foreclosure inventory in judicial states in solid blue, where where the mortgage servicer must prove their right to foreclose on a homeowner in court, and the foreclosure inventory in non-judicial states, where the bank can seize a delinquent home by an auction initiated by their attorney without court intervention, shown in blue dots…the quarterly difference between the two in shown by the dotted red line..as of the 3rd quarter, MBA has the foreclosure inventory in judicial states at 5.28% of all mortgages in those states, while the foreclosure inventory in non-judicial states is at just 1.66% of mortgages in those states…

3rd qtr MBA foreclosure inventory history

the next bar graph below, also from the MBA, shows the percentage of all mortgages in foreclosure by state, with the judicial states coded in navy blue, and non-judicial states coded in red…obviously, the states with the highest percentages of homes remaining in foreclosure according to the MBA are all judicial, led by Florida, which has 9.48% of their mortgages in foreclosure, which is nonetheless down from 10.58% in 2nd quarter; New Jersey, with the 2nd most at 8.28% in foreclosure up from 8.01% last report, as is New York, where the foreclosure inventory rose from 6.09% in the second quarter to 6.34% in the 3rd…Nevada is the only non-judicial state among the top dozen states with the highest foreclosure inventories, because they passed a law in 2011 making it a felony if a mortgage servicer made fraudulent representations concerning a title, and imposed fines up to $5,000 for falsifying documents, which slowed foreclosures in that state to a near standstill….

3rd qtr MBA in foreclosure by state

LPS: Mortgage Delinquencies Rise 4.2% in September but Still Down 10% YTD

in contrast with the MBA delinquency report, the Mortgage Monitor for September (pdf) from LPS shows that delinquencies have increased since their last report, with the percentage at least one payment overdue on their mortgage but not in foreclosure at 6.46% in September, up from 6.20% in August…but the LPS delinquency rate is still down from the 6.68% delinquent in June, the month comparable to the MBA 2nd quarter report; their foreclosure inventory is down too, to 2.63% from 2.66% in August and 2.93% in June…but they show foreclosure starts in September at 108,953, 1.3% above August’s 107,552 new foreclosures, and at nearly the same level as the 109.042 foreclosures initiated in June…but recall that this LPS data is not seasonally adjusted, while the MBA stats we quoted were, which could account for the difference..

according to LPS, 1,328,000 properties remained in the foreclosure process at the end of September, down from 1,341,000 so encumbered at the end of August; another 1,331,000 home loans were more than 90 days overdue but not in foreclosure, and 1,935,000 more were at least one payment late but less than 90 days overdue in September, which meant 4,594,000 US mortgages were going unpaid as of this latest report…with an active loan count of 50,522,000 at month end, that means LPS has 9.09% of all loans at least one payment past due or in foreclosure at month end, somewhat less than the 9.75% totally delinquency rate reported by the MBA; this holds to trend; every time we’ve reviewed these two reports together, it has always been the MBA showing the higher delinquency rate.. 

the graph below, copied from page 13 of the mortgage monitor (pdf), tracks with a green line the percentage of active loans that were in the foreclosure process monthly from 1995 to the present…this so-called foreclosure inventory are those home loans in between the first initiation of foreclosure proceedings and the “foreclosure sale”, which typically transfers title to the bank (terminology is on page 28 of the pdf)….by September, that percentage had fallen to 2.63% of home mortgages outstanding,well down from the October 2011 peak of 4.29% of mortgages in foreclosure, but still more than five times the 0.44% foreclosure inventory of pre-crisis December 2005…the red line on this graph tracks the percentage of mortgages delinquent but not in foreclosure over this same timeframe; it was at 6.46% in September, well down from the 10.57% all time high for such mortgage delinquencies set in January 2010, but still higher than the precrisis level of 4.27% in December 2005 marked on the graph…also note the seasonal changes in mortgage delinquencies, wherein they usually peak at year end, when most people get overextended during the holidays, and then decline over the first few months of each year as homeowners catch up…the delinquency rate of 6.08% reached in May was the low for this year; it’s likely to trend up from here as consumers put holiday shopping ahead of their house payments…

September LPS delinquencies and foreclosures

next, we’ll include below the table of non-current mortgage and foreclosure percentages for all 50 states and the District of Columbia from page 24 of the mortgage monitor…the first column shows the delinquency rate for each state, ie, the percentage of mortgages in each state that are at least one month behind and not yet in foreclosure…the second column is the percentage in each state that are in foreclosure, which you’ll note are different percentages than those shown in the MBA bar graph, the third column is the total non-current percentage, or sum of all those behind of their mortgage, and the last column shows the year over year percentage change in non current mortgages nationally and in each state…also note that judicial states, where banks must establish their right to foreclose in court, are marked on the table below by a red asterisk…these states have the highest percentages of mortgages still stranded in foreclosure, led by Florida, where LPS shows 8.1% of all mortgaged homes still in foreclosure, in contrast to the 9.5% in foreclosure in Florida that the MBA showed ..there are similar differences in several other states as well…

september LPS state non current table

finally, we’ll include two graphics that are new to the mortgage monitor with this month’s release; the first map graphic below, from page 17 of the pdf, color codes the foreclosure inventory by county with those counties with 0.0% homes in foreclosure in the darkest shade of green and those with more than 3.5% of their mortgaged homes in foreclosure in the darkest red; obviously, the judicial states have the reddest hue, with every county in both Maine and Florida showing a foreclosure inventory of greater than 3.5%, as well as nearly every county in New York and New Jersey in that darkest red…this should not be a surprise, since both LPS and the MBA have the statewide foreclosure inventory at over 5% in all four of those states…the next national map below that, from page 16 of the mortgage monitor, is similar in that it shows the percentage of mortgages that are at least one payment behind in each county, again with the few and far between deepest green indicating zero percent of the mortgage are non-current, and the darkest red indicating more than 15% of the home mortgages in that county are at least one payment in arrears…again, this should not be a surprise, with total national delinquency rates still above 9%…you may have noticed when viewing the state table that both Florida and Mississippi have statewide non-current rates greater than 15%…

September LPS foreclosure inventory by county 2

percentage of non-current mortgages by county:

September LPS non-current by county 2

(the above is my weekly commentary that accompanied my sunday morning links emailing, which in turn was mostly selected from my weekly blog post on the global glass onion…if you’d be interested in getting my weekly emailing of selected links that accompanies these commentaries, most coming from the aforementioned GGO posts, contact me…)