In Depth Analysis: CalculatedRisk Newsletter on Real Estate (Ad Free) Read it here.

Sunday, August 30, 2020

August Employment Report: Comments on Temporary Decennial Census Hiring and Education

by Calculated Risk on 8/30/2020 11:46:00 AM

The employment report for August will be released on Friday, September 4th at 8:30 AM. The consensus is for 1.40 million jobs added in August, and for the unemployment rate to decrease to 9.8%.

There were 1.763 million jobs added in July, and the unemployment rate was at 10.2%.

There will be two distortions in the August employment report that are worth discussing.

The August employment report will show an increase of 237,800 in temporary Census hiring under Federal employment. Since these are temporary jobs, 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. For more, see: How to Report the Monthly Employment Number excluding Temporary Census Hiring

Temporary Decennial Census Hiring Click on graph for larger image.

Here is a page from the 2020 Census Paid Temporary Workers report released last week.

The temporary employment for the July and August reference weeks are circled in red.

As of the July BLS employment report reference week, there were 50,404 decennial Census temporary workers. As of August  reference week there were 288,204 temp workers.

This will boost August employment by 237,800 jobs.

Another distortion in the August employment report will be from education employment. There is a seasonal pattern to state and local education employment with a large number of educators usually hired in late Summer or early Fall - at the beginning of the school year - and then let go in June and July. Due to the pandemic, many of these educators were let go earlier than usual this year.

This led to a weird quirk in the seasonal adjustment for July. Although there were 960,000 state and local education jobs lost in July NSA (Not Seasonally Adjusted), this was reported as a gain of 245,000 jobs (SA). This was fewer jobs added (SA) than I expected, but it was difficult to tell how many year-round jobs had been lost.

As of February 2020, there were 10,979,000 state and local education employees. Approximately 2 million were seasonal (hired in the Fall, and then usually let go at the end of the school year), and the remaining almost 9 million are year round employees. As of July, all 2 million seasonal employees were let go, along with 635 thousand year round employees.

Usually we'd expect about 415,000 seasonal jobs hired in August (and then most of the remaining seasonal employees hired in September). However, since many school districts delayed the opening of the school year, my guess is few of these seasonal jobs were filled in August.

The BLS model will expect about 415,000 seasonal jobs added in August.  If fewer jobs are added, the BLS will report that education jobs were lost Seasonally Adjusted.   For example, if only 115,000 jobs are added in August, then the BLS will report around 300 thousand education jobs lost. 

Of course some of the year round jobs that were lost earlier in the year could be hired back.   However, since state and local governments are under financial pressure, it seems unlikely that many year round employees were hired back.