The Census Bureau will soon start hiring temporary workers for the 2020 Census. Temporary hiring will really increase in March, April and May of 2020, and then the jobs will finish in a few months. These are real jobs – frequently second jobs, or taken by seniors and students – but they will only last a few months.
Since these are temporary, and only happen every ten years with the decennial Census, it makes sense to adjust the headline monthly Current Employment Statistics (CES) by Census hiring to determine the underlying employment trend.
The correct adjustment method is to take the headline number and subtract the change in the number of Census 2020 temporary and intermittent workers. The BLS reports the number of Census temporary workers here: Census 2020 temporary and intermittent workers and Federal government employment
As an example, in May 2010, the Census hired 411 thousand temporary workers. The BLS reported (since revised)
“Total nonfarm payroll employment grew by 431,000 in May, reflecting the hiring of 411,000 temporary employees to work on Census 2010, the U.S. Bureau of Labor Statistics reported today.”
So the underlying trend was +20,000 jobs in May 2010 (as originally reported).
For June 2010, the number of temporary Census hires declined by 225,000. The BLS reported
“Total nonfarm payroll employment declined by 125,000 in June, and the unemployment rate edged down to 9.5 percent, the U.S. Bureau of Labor Statistics reported today. The decline in payroll employment reflected a decrease (-225,000) in the number of temporary employees working on Census 2010.”
So the underlying trend was +100,000.
Some readers will notice that this is mixing SA and NSA data! Usually that is not appropriate, but in this special case, based on the methods used by the BLS, this is correct.
I checked with the BLS, and they agree in this situation it is correct. Here is my question (back in March 2010) and the BLS response:
9:34 Michele Walker (BLS-CES) –
Submitted via email from Bill: Hi. The headline payroll number is seasonally adjusted, and the hiring for the 2010 Census is NSA. How would you suggest adjusting for the 2010 Census hiring to determine the underlying trend?
Thanks for your question Bill.
There is an adjustment made for the 2010 Census. Before seasonally adjusting the estimates, BLS makes a special modification so that the Census workers do not influence the calculation of the seasonal factors. Specifically, BLS subtracts the Census workers from the not-seasonally adjusted estimates before running seasonal adjustment using X-12. After the estimates have been seasonally adjusted, BLS adds the Census workers to the seasonally adjusted totals. Therefore, to determine the underlying trend of the total nonfarm (TNF) employment estimates (minus the Census workers), simply subtract the Census employment from the seasonally adjusted TNF estimate.
So starting soon (as hiring ramps up), the correct way to report the headline number is ex-Census.
BTW, it is widely reported that there have been 103 consecutive months with positive job growth (a record). However, if you use the ex-Census numbers, it is 110 consecutive months! The job streak started during the decennial Census, but was masked by the temporary Census layoffs. As an example, the BLS reported a negative 64,000 jobs in September 2010, however the number of temporary Census workers declined by 76,000 that month, so ex-Census the number of jobs added in September 2010 was 12,000.