Reclaiming Common Sense

The February Jobs Report was a tale of two data sets.

How big of a jump in jobs and workers can we expect?


Last month the headlines were that the seasonally adjusted worker data barely moved higher, with only 20,000 seasonally adjusted non-farm payroll positions being created, according to the Current Employment Statistics data. More headlines discussed how workforce participation rose while unemployment fell, according to the Current Population Survey Data. It was a battle of the two data sets. The truth is that we added more workers, non-seasonally adjusted, than we did during February 2014, 2015, and 2016 and we added fewer seasonally adjusted CES workers than February 29014, 2015, 2016. This was all covered in the five part "Tale of Two Data Sets" series. This month the two data sets should be in more agreement than last month.


The data needs to be examined based on growth patterns and the changes in seasonal factors. The annual growth rate has been improving. This column has projected that a 2.00% annual growth rate is key for a strong economy. There have been revisions to the NSA CES data the past two January Reporting Periods where NSA CES growth was pushed back to 2016 and 2017, and even 2018 during the most recent revisions. If the seasonal factors continue to slide from the prior March seasonal factor level then the SA CES data will be reported lower than it could have be reported. The thing to remember is that non-farm payroll, NFP, includes government workers while the "Jobs Streak" data was/is the Total Private Sector data, excluding government workers.


We should see annual growth in the workforce at a faster pace than the past two March reports recorded. Recently we have had annual NSA CES worker growth between 1.93% and 2.23%. Last month we were at 1.93% while January was 2.23%. The data from August through December of 2018 ranged from 1.96% to 2.05%. If we grow within this range then we could see either a paltry 174,000 SA CES private sector workers added up to half a million workers added. Yes, up to a half million workers added during March. We see from the first histogram that the growth rate was well over 2.00% during 2014, 2015, and 2016. If we grow at the rate that we supposedly grew during March 2016 we could see the addition of over 400,000 SA CES workers. A rate of 1.97% would garner 218,000 workers.  A rate of 2.00% or 2.02% is very likely, or possibly 275,000 SA CES workers added during March.


The Month to month growth rate is expected to be better than the past two March Reports, or over 0.51%, and possibly over 0.55%. People start hiring workers for Spring Break, Mother's Day, Easter, Memorial Day, and the Fourth of July starting during February and March. If we grow at the pace that we did during march 2018 then we could hit 176,000 SA CES workers added. If we grow at the pace we did during March 2016, and election year, we could grow at a rate of 0.63% and hit 329,000 SA CES private sector workers.  Either way you examine the data we could add 750,000 NSA CES workers this month. The seasonal factor means that we should see 275,000 SA CES private sector workers added this month. But....


The March seasonal factor could slide to its lowest level since 1979. The seasonally adjusted data is calculated by multiplying the seasonal factor  times the non-seasonally adjusted data. A lower seasonal factor lowers the SA CES data. If we grow at the rate we did during 2016, month to month, and we used the same seasonal factor then we would see 309,000 private sector workers added, seasonally adjusted. If we grew at that same rate and used the 2018 seasonal factor that number is trimmed to 298,000. If it reduced to the level projected then that number is reduced to 296,000. The month to month data indicates that the more reasonably month to month growth is in the 0.55% to 0.71% level. The question is was the data lower last month with the anticipation of hiring more workers during March?


Then there are the revisions to the prior data. There are a number of observers who have come to the same conclusion that this column made regarding last month's data: It didn't add up. There are some who are anticipating a major upward revision to the February data. If the data is revised upward by 30,000 workers then that "borrows" 30,000 from this month's data. If the unadjusted level is 275,000, the adjusted level would be brought down to 245,000 workers added. If the data from February is revised higher by 100,000 then last month would be official 125,000 SA Private Sector Workers and this month would be 175,000.  Watch the growth rate. Watch the Seasonal factor. Watch the Revisions.


Where will the growth be found? The data is almost in perfect agreement month to month and March to March, seasonally adjusted or non-seasonally adjusted. We should see all sector add workers month to month and year to year.


The non-seasonally adjusted CES  month to month data projects all sectors to add workers. The largest growth rates are in Leisure and Hospitality (LAH,) Construction, Mining and Logging (M/L,) and Professional Business Services (PBS.)


The non-seasonally adjusted March to March data is expected to add workers in all sectors. The largest growth rates are expected in M/L, Construction, PBS, EHS and Manufacturing.


The Seasonally adjusted month to month data is projecting worker gains in all sectors. The largest growth rates here are found in M/L, Information Technology (IT,) PBS, Manufacturing, and Trade, Transportation and Utilities (TTU.)


The Seasonally Adjusted March to March data indicates that all sectors except Mining and Logging should add workers. This does not make sense. We had 714,00 SA M/L workers during March 2018 and  754,000 SA M/L workers last month. We are in an energy boom. The largest growth rates are expected in Construction, PBS, EHS, LAH, and Financial Activities (FIRE.) Expect all sectors to add seasonally adjusted workers from their March 2018 level.


What can we expect from wages. Wages have been rising since last Summer. There is a labor shortage, based on the level of missing participants that we have had. This is calculated comparing the unemployment rates of two similar months and their differing participation rates. If people hop back into the job market they may be receiving higher wages. It is essential to calculate wages and wage growth on non-seasonally adjusted data. High seasonal factors for wages and workers would artificially boost the average hourly wage. The converse is also true. If high worker seasonal factors are multiplied by low wage seasonal factors, or the other way around, then the growth could be eliminated.


Expect the non-seasonally adjusted wage data to be bolstered by growth in higher paying jobs. Mining and Logging, Construction, and Manufacturing are some of the highest paying sectors, as is PBS.  EHS could weigh down the average wage. The seasonally adjusted data means that lower than average


The worker data is projecting strength in workers across the board, month to month and March to March. What can we expect from the Current Population Survey?


Last month we added 1.2 million non-seasonally adjusted jobs, and that wasn't even as good as February 2017. We added over 1.5 million combined full-time and part-time jobs during February 2017.  We saw the number of NSA CPS jobs added  drop during March of 2018 to under half a million total jobs, and almost all of them were part-time jobs. Between those two months last year we added 2.0 million jobs. Could we add 800,000 jobs this month? The data indicates that we could expect to see between 700,000 and 1.0 million jobs added this month, so 800,000 NSA CPS jobs is feasible.


Could we see 1 million seasonally adjusted jobs created this month? Last year the seasonal factors converted a gain of half a million jobs to a net loss of jobs. If we grow as we did during 2016 and 2017 then we could fall just short of 1 million or poke above one million jobs created.  Expect something in the neighborhood of 700,000 jobs created. Also, do not expect to hear anybody else discuss this number.


The Unemployment Level should fall, seasonally adjusted or non-seasonally adjusted. The non-seasonally adjusted unemployment level has fallen every March since March 2010. We have seen the unemployment level drop by over 300,000 workers, 400,000 workers, 500,000 workers and even 600,000 workers during March. If the NSA U-3 unemployment level does not fall by roughly 400,000 workers then the SA U-3 level could rise month to month.


Unemployed Workers is a Different Survey than Job Openings. Both are different from the continuing claims data. There has been plenty of discussion regarding the Job Openings and Labor Turnover Survey (JOLTS) data during the past few month as the number of Job Opening, seasonally adjusted, have been greater than the number of unemployed workers, seasonally adjusted. The JOLTS data is comparable to the the CES data. The U-3 unemployment data is from the CPS data. Last month's jobs report was a good example of why you should not compare the seasonally adjusted CPS data with the SA CES data. The same should apply for the seasonally adjusted JOLTS data. You might as well compare the seasonally adjusted temperature in Phoenix with the Seasonally adjusted temperature in Milwaukee. U-3 unemployed workers are people looking for work, who may or may not have previously been employed. The continuing claims data was at the lowest for mid-March than it has been since March 15, 1973. It was 300,000 claims lower than March 18, 2000.


The Workforce Participation Rate should rise as the Unemployment Rate falls. The participation rate and unemployment rate were falling almost in unison from January 2010 through January 2016. Unemployed workers are a participants. Eventually people stop looking and a no longer participating. The people who stop looking are not employed and not unemployed, they are "effectively unemployed." The workforce participation rate hit a January low during January 2017 and could have been reported lower than it was reported. We are finally approaching a similar participation rate for the the same month in office that President Reagan had while he was in office. The spike in jobs should more than offset the drop in the unemployment level, meaning that with normal population growth we should see the participation rate improve from last year's level.


There is a cornucopia of data that can be dissected. The growth in M/L, Construction, Manufacturing and TTU workers means that we may see a spike in men working full-time or part-time jobs. This column with compare CPS job growth for men and women, plus the changes in dual job workers. Dual job workers should improve through July, pause, and then improve to a peak for the year during October, November or December this year. Will we see improving participation in all age groups? The "Red, Gray and Blue" article will examine the CPS data there.  This should be a strong Jobs Report. This should not be another "Tale of Two data sets" situation.


It's the Economy.