Friday’s NFP will be artificially boosted by the hiring of temporary Census takers

A mixed U.S. Non-farm payrolls report for the month of August is expected. On the upside, the unemployment rate (U-3), which is the most followed indicator by market participants, is forecasted down at 9.8% from 10.2% in July, which would be the lowest level reached since the pandemic shutdowns started. On the downside, the consensus amongst economists expects that the broad measure of unemployment, called U-6, which includes the total unemployed plus all persons marginally attached to the labor force and the total employed part time for economic reasons, will increase for the first time since April to 17.3% versus 16.5% in July. The rise in the U-6 unemployment rate is likely to reflect the impact of COVID-19 resurgence in some regions and the implementation of further social distancing measures. Finally, payrolls are forecasted to rise 1.4 million in August, down from 1.763 million in July. If confirmed, it would be the smallest gain since May, when the lockdown started to be lifted in some states.

We expect the NFP report will be artificially boosted by the hiring of hundreds of thousands of temporary Census takers. In a recent blogpost, Bill McBride pointed out that the U.S. Census Bureau has recently released an update (see here) on 2020 Census Paid Temporary Workers. As it is the case every ten years, the Census Bureau carries out a count of every resident in the United States. To do so, it hires a very large number of temporary Census takers. In the July employment report reference week (from July 12 to July 18), the government hired about 50,000 workers and this number jumped to 288,204 in the August reference week (from August 9 to August 15).

Said differently, the August BLS employment report is likely to be artificially boosted by the hiring of 237,800 new Census takers.

Given these jobs are temporary and massively impact total payrolls, we think the right way to assess the underlying employment trend consists in subtracting from the headline figure the change in the number of Census takers. This adjustment method may sounds a bit unconventional since we use both seasonally adjusted and non-seasonally adjusted data but this is the only way to have a better understanding of the real state of the U.S. labor market.