Reclaiming Common Sense

The January Jobs Reports is One of the Reasons We Have Seasonally Adjusted Data

Watch the Revisions.

The January Jobs report is marked by the situation that we will have very different Non-Seasonally Adjusted (NSA) data and Seasonally Adjusted Data story-lines. There are two data sets that go into the jobs report, or employment situation report, and they measure different things. The Current Population Survey (CPS) jobs and unemployment data and the Current Employment Statistics (CES) workers and wages data. Normally January brings seasonally adjusted and non-seasonally adjusted data revisions for the CES data and only revisions to the seasonal factors, and therefore only the seasonally adjusted data. We were advised months ago that the seasonally adjusted CES data was going to be revised down for the CES data with downward revisions for both the Private Sector CES data and the Non-Farm Payroll (NFP) CES data. What can we expect from the CES and CPS data? Down is up.

It was expected that we would record January to January growth in all sectors. This is important. All sectors have added workers under President Trump. The non-seasonally adjusted  growth rate has been slowing since January. That said, even a growth rate of 1.44% would have yielded a seasonally adjusted gain of 194,000 private sector workers. That would be the slowest January rate of growth since January 2011. If we grew at 1.48% then we could see a number closer to 246,000. The sectors with the largest growth rates were expected in  Mining and Logging (M/L,) Construction, Professional Business Services (PBS,) Leisure and Hospitality (LAH,) and Education and Health Services (EHS.)

This month we saw a growth rate of 1.64% and 206,000 seasonally adjusted private sector workers added to the workforce. The largest growth was recorded January to January were in TTU, Construction, PBS, and Information (IT.) We saw 19,000 SA Government sectors added to create a non-farm payroll number of 225,000.

We were expected to see the Non-Seasonally Adjusted CES  record December to January declines in all Sectors.  Yes. January is one of the months why seasonally adjusted data was created. We always trim the workforce by 1.80% to 2.80% between December and January. It is possible, during a recession, to trim jobs both month to month and January to January. That is not anticipated this January.

  • 1982: (-2.14%) MTM (-0.27%) January to January
  • 1983: (-2.64%) MTM (-2.16%) January to January
  • 1991: (-2.41%) MTM (-0.52%) January to January
  • 1992: (-2.29%) MTM (-1.01%) January to January
  • 2002: (-2.25%) MTM (-2.13%) January to January
  • 2003: (-2.07%) MTM (-0.52%) January to January
  • 2009: (-2.88%) MTM (-3.90%) January to January
  • 2010: (-2.23%) MTM (-3.77%) January to January

There was "no way" that we are going to see a decline January to January this year.  We did not have a Government Shutdown. We did not have a GM strike. Last year we had 125.932 million workers, prior to the revisions that will be released this Friday. This December we had 130.344 million non-seasonally adjusted workers. We have trimmed roughly 2.4 million to 2.6 million non-seasonally adjusted private sector workers each of the past four years. We would have had  to trim 4.4 million workers to go negative January to January. That should not "ever" happen during the course of one month.

We saw a drop of 1.82%  month to month. This was the smallest January drop since January 1984. We saw the largest declines in  TTU, Construction, PBS, LAH, and Government.

It was projected that we would record a drop of 1.95% to 2.05% this January. All sectors were expected to trim workers. The sectors with the largest percentage drops are expected to be Construction, Trade, Transportation and Utilities (TTU,) Information (IT,) PBS, and LAH. This is normal.

The seasonal factors, and the annual adjustment, are the wild cards this month. The range of seasonal factors from 2004 through 2019 mean that a drop of 2.05% month to month could be reported as a gain of 79,000 workers or a gain of 238,000 workers. If we only drop by 1.95% then those numbers could range between 214,000 and 373,000 private sector workers added this month.

The Seasonal factor was the lowest since at least 1979.  This skewed the headline number low by 206,000 positions from what we would have had if we used the 2018 seasonal factor. The headline number could have been reported at 490,000 positions if we used the 2016 seasonal factor.

We knew that the CES data base will be reduced by 501,000 workers for the private sector and another 13,000 for Government workers. Would they be shifted backwards to 2014, 2015, and 2016 or 2017, or even 2018?  Last year the revisions "stole a great report" from President Trump. The data from 2013-2016 was revised higher and the data from 2017 and 2018 were revised lower from their December 2018 values.

This year the revisions were positive for 2016 and negative for 2017, 2018, and 2019. It is easy to see that former President Obama go a boost and President Trump saw his annual data revised downward, again.

  • 2016 was revised from 2.028M to 2.134 million (+106k)
  • 2017 was revised from 2.154M to 2.036M (-118k)
  • 2018 was revised down from 2.5787M to 2.200M (-378k)
  • 2019 was revised down from 1.98M to 1.927M  (-54k)

Total revisions for those four years came in at -442,000. There were minor downward revisions to the data from 2010 through 2015.

Seasonally Adjusted Workers were expected to be  reported as being added in all sectors month to month except Information and Mining and Logging (M/L.) We have seen strong growth in all sectors except M/L this year, especially EHS, PBS, LAH, and Construction. We normally see a drop in IT during January. We were already down 22,000 workers, NSA, for M/L November to November. There was a possibility that the SA data for M/L may have the largest month to month spike. Other sectors that should report strong growth are Construction, LAH, EHS, TTU, and Manufacturing.

We saw seasonally adjusted workers added month to month in all sectors except Manufacturing. The largest growth rates were in Construction, EHS, LAH, IT and PBS.

Seasonally Adjusted worker data should rise for all sectors, January to January. Watch the Government data this month. We had the Schumer Shutdown last January. The largest growth should be M/L, Construction, LAH, PBS, and EHS.

We saw seasonally adjusted workers grow in all sectors January to January except Mining and Logging. The largest growth was seen in EHS, LAH, Construction, PBS, and IT.

Wages should rise, hourly, and total wages. Hourly wages have been rising at 3% for basically a year. More of the same is expected. When the average wage for workers is multiplied by the number of workers working, and both grow, the impacts are synergistic. We also may see upward movement in hours worked rise.  This will be addressed in the "Wages and Workers" article.

It was expected that  non-seasonally adjusted Current Population Survey (CPS) Jobs would record a drop in full-time and part-time jobs, that unemployment would record a rise, and participation should have risen. There was the possibility of part-time jobs rising.  There should have been a seasonally adjusted (SA)  rise in Full-time jobs, and a possible rise in SA part-time jobs. Sometimes the FT jobs report a rise and a PT job drop. We often see both the SA FT and SA PT rise during January.

The CPS Jobs data recorded a loss of 1.510 million jobs  and reported a loss of 119,000 jobs. This was the largest drop in NSA FT jobs since January 2009. We have a larger economy right now. We also recorded and increase of 253,000 par-time jobs. Normally the number of NSA FT and NSA PT jobs both drop. We saw a seasonally adjusted drop of just 656,000 FT jobs and a gain of 537,000 SA PT jobs.

Non-Seasonally Adjusted Unemployment was expected to record an increase while Seasonally Adjusted Unemployment is expected to fall. This should not be a surprise. Santa doesn't need his Trade, Transportation and Utility workers during January.

Unemployment jumped by 1.001 million workers, and reported a gain of 139,000 workers. We have seen increases of roughly 1 million unemployed workers, NSA, each of the past four years. People are responding that they are out of work and looking for work.

The Non-seasonally Adjusted Participation rate normally hits its annual low during January. The NSA Unemployment Rate tends to hit is annual high during January. The Unemployment rate this past December was just 3.36% NSA. This is almost three percentage points lower than January 1996 (6.29%) and January 2004 (6.26%) This should be one of the headlines after the release of the Jobs Report on Friday. The participation rate was 63.03%  during December.  It was only 63.24% during January 1984, and was rising. It was only  63.35% during January 2012, and was falling.  It was projected that we would  be better than January 2014 when we had a participation rate of 62.52% and an unemployment rate of 7.03%. It was unlikely that we would  exceed the 63.27% participation rate of January 2013. That's fine because back then the unemployment rate was a whopping 8.52%.

The Participation rate came in 63.00% with Unemployment under 4.00%. If you look at the data since 1979you will find that this is the lowest January non-seasonally adjusted unemployment rate in over 40 years. You have to go back to January of 1967 when it was 3.7%. The non-seasonally adjusted participation rate is the highest it has been since January 2013. We had a participation rate of 63.35% during January 2012, an election year, while simultaneously having an unemployment rate of 8.82%

The Multiple Job Holder number does not get much attention. A second job is the best form of unemployment insurance . Last month we had 8.058 million multiple job holders (MJH) with 4.419 million FT PT job workers, 2.072 million PT PT job workers, and 336,000 dual FT job workers. The balance fall in the "variable" hours workers and the primary PT secondary FT job people. Last year we had 7.749 million total MJH with 4.452 million FT PT, 1.879M PT PT and 249,000 FT FT workers.

The MJH level was expected to drop 3% to 4% month to month. This would have placed us between 7.73 million and 7.81 million MJH. This would put us at a January level not seen since January 2000 when it was 7.897 million, or January 1999 when it was 7.702 million, or last year at 7.749 million. The FT PT worker level was expected grow or contract by 3%. The dual PT number was expected to fall 3% to 13% month to month, or to a range of 1.80 million to 2.0 million.  The record for January was set during 2018 when there was 2.04 million dual PT workers. We could have seen the dual worker number fall and it will not impact unemployment.

Multiple job holders increased in all categories January to January. We had 8.083 million multiple job holders, with 4.528 million FT PT workers, 2.047M PT PT workers, and 326,000 FT FT workers. More will be written about this next week in the Men and Women Worker articles.

There are other ways to examine the data. How are men and women doing in the Trump Economy. Both genders are expected to have record levels of CPS jobs this January. How are people doing by age group? It is still expected that people over 55 years of age will be over-participating.

The seasonal factors stole over 200,000 workers from the headline number. The revisions stole half a million workers from President Trump's accomplishments. The CPS and CES data are out of sync again.  Participation matters.

It's the Economy.