Type your paragraph here.

Down Is Up - Recorded Job Losses May Be Reported as Worker Gains

The Bureau of Labor releases its Monthly Employment Situation Report, or Jobs Report, next Friday. Two data sets are used to create the monthly report. The Current Population Survey measure jobs created and reports on the number of full-time jobs and part-time jobs created as well as the unemployment level. These data points are used to create the workforce participation rate number and the unemployment rate number. The second data set, the Current Employment Statistics (CES) data set examines the number of workers in the workforce. The CES data for private sector jobs is "the number" used by the White House to proclaim how many "jobs" were created.

Each data set has a seasonally adjusted and a non-seasonally adjusted component. The seasonal factors used to convert the NSA data to the SA data change between data sets, and between categories.  The seasonal factors change from month to month, season to season, and year to year. When data from different years are compared with current data without considering the seasonal factors FACTs (False Assertions Considered to be True) are created. The data released each month is the advance data for the current month, the preliminary data for the prior month, and the final data for the month prior to that. The White House was tweeting consecutive months of job creation before the preliminary data was released. After the June Jobs report - no tweet. The dip in the May Jobs number was masked by the twin "towers" of April and June. The release of the final revision meant that they had to change the narrative

Last month the Jobs report reported over 200,000 jobs created when it could have been 155,000. The July Jobs Report revealed that the final private sector final jobs number for May was -1000 for the private sector.  This was better than the preliminary May Jobs number  -6000 jobs reported in the June Jobs Report. The advance number reported in the May Jobs Report showed an advance number of 38,000 "non-farm" payroll jobs added after the April data was revised down by 37,000. Note that there is a difference. The Private Sector Job Growth had an advance SA number of only 25,000 jobs added from the April level. There are two things to take away from this paragraph: 1) was the final revision for the May data the final revision for the May data? Will they adjust it from -1,000 SA CES jobs to +1,000 SA CES jobs just to keep "the Streak" alive; 2) Pay attention to whether or not someone is referencing the total private sector jobs number or the non-farm payroll jobs added.

August is Back to School Month - And a Job Cutting Month - Before Seasonal Adjustments. The annual job cutting that we see during January and August are "the reasons" why we have seasonally adjusted data - nobody would have faith in the economy if it was reported that there were 2 million fewer jobs during January than during December. Nobody would have faith in the economy if they heard that there were 1 million fewer jobs during August than during July. We could see 700,000 to 900,000 part-time jobs shed during August. If we see any full-time jobs shed this could be the signal of a looming recession. This huge part-time job loss will not be reflected in the weekly unemployment claims number because they do not qualify for benefits. This drop in part-time jobs should be reflected in the Multiple Job Holders Data. Will we see over 200,000 full-time jobs created?

The True Seasonally Adjusted Jobs Number has decreased six of the past nine years. The CPS data measures jobs. The CES data measures workers. We have seen net seasonally adjusted job loss six of the past nine years. The two histograms look very similar. This is part of the reason that the "official" workforce participation rate number, the seasonally adjusted data, will show a drop in participation.

Unemployment Could Drop - Even without significant changes in the Continuing Claims Level. Yesterday's Weekly Unemployment Claims revealed a fairly statics 2 million people who are continuing to claim benefits.  Remember that last month there were 8.267 million NSA U-3 Unemployed. The SA U-3 value was reported at 7.770 million. There were roughly 2014 Million when the July data was collected. There were 2.11 million continuing claims while the August data was being collected.  This means that there are 6 million unemployed people who were not collecting continuing claims last month.

The NSA and the SA Unemployment Numbers could both drop this month.The unadjusted number of unemployed has dropped eight of the past nine years. Seasonally adjusted unemployment have dropped five of the past nine years. We should anticipate that both numbers should decline this month.

Dropping Employment Plus Dropping Unemployment means a drop in the Participation Rate.Participation and Unemployment peak during July during most years. Both drop precipitously during August. Last Month the unadjusted participation rate was 63.36%. Last August it was 62.68%. This past month the July participation rate was slightly higher than July 2015. We have seen the August Participation rate decline year over  year since August of 2008. No matter where the unemployment rate is reported, it will not reflect the Effective Unemployment rate. The effective unemployment rate compares U3 from prior periods based on the impacts of the participation rate. n  We are missing over 8 million participants.

How Will Dropping Unemployment and Dropping Employment Impact the Seasonally Adjusted Numbers?The tables below this section reveal the negative growth we have seen during the month of August since 2003. We have also seen a general decline in the weighting of the unemployment numbers and the part-time ob numbers over the same period of time. The weighting of the Full-time jobs have increased, in general, since 2003.

When the possibilities for contraction are factored in with the potential seasonal factors for the month a picture of general contraction is still revealed.We should see real and adjusted job losses. We should see real and adjusted unemployment decreases. We should see real and seasonally adjusted drops in participation.

This is normal for the month of August. What is also normal is that "everyone else " will ignore the CPS data - with the exception of the Participation rate and Unemployment rate. Commentators will say "On one hand unemployment dropped" and "On the  other hand participation dropped to levels even lower than (comparable to) Last year's historically low participation rate."

The next section of this column will address the "reality" that is commonly reported - the Current Employment Statistics data.

 The Unadjusted CES Data should show an increase under 0.10%. Last August we almost saw a contraction in the August NSA CES value. We had been consistently running between 0.14% and 0.16% since the recovery began during 2010. We normally see real unadjusted worker growth during August. How this happens when we lose jobs is a big question.

The thing that is a bit dumbfounding is that we can see real, NSA CES, worker growth and see a contraction reported.We saw weaker NSA CES growth last August than we saw during August 2000 and August 2002 and yet the SA data for August 2015 was reported higher than either of those years.We saw considerably higher NSA CES growth duringAugust of 1981 and the SA CES data for 2015 was reported at a rate that was faster than 1981, 1991, 1992, 1997or 1998.

If we see a similar growth rate to what we saw last August we could see double digit job growth - under 100,000 jobs added. We do not need a recession as bad as 2008 or 2009 to see half a million workers shed during August. If it grows at roughly the same pace as it did during 2000 or 2003 then we could see fewer than 50,000 seasonally adjusted CES workers added.

If we use any one of the seasonal factors used during the past six years and we have the same growth hat we saw last year we will see  either the addition or subtraction of 50,000 workers from the economy reported.

Expect 70,000 to 130,000 NSA CES workers to be added to the economy. Expect 37,000 to 92,000 NSA workers to be added to the economy. It is entirely possible that the drop in CPS jobs will be reflected by a drop in the worker numbers. It is also possible that those individual who are working two jobs will lose one and the worker numbers could be unimpacted.

Expect very weak job growth. Prepare for a contraction to be reported