by Calculated Risk on 10/02/2010 08:50:00 AM

1) The consensus is for a headline payroll number of zero (no net payroll jobs added or lost) and for the unemployment rate to increase to 9.7% in September from 9.6% in August.

Goldman Sachs is forecasting a minus 50,000 headline payroll number and an increase in the unemployment rate to 9.7%.

2) My estimate is the decennial Census workforce was reduced by 78,000 in August. This suggests a consensus headline payroll number of +78,000 ex-Census.

Click on graph for larger image in new window.

This graph shows the number of Census workers paid each week. The red labels are the weeks of the BLS payroll survey.

The Census payroll decreased from 83,955 for the week ending August 14th to 6,038 for the week ending September 18th.

So my estimate for the impact of the Census on September payroll employment is minus 78 thousand (this will probably be close).

With temporary payroll employment down to 6,038, September will be the last month with a significant decennial Census adjustment.

3) The unemployment rate is dependent on both job creation and the participation rate (both numbers from the household survey - payroll employment is from the establishment survey).

Usually the participation rate - the percent of the civilian population in the labor force - falls when the job market is weak. And a decline in the participation rate puts downward pressure on the unemployment rate (and the opposite is true when the participation rate increases).

Right now the participation rate is very low at 64.7%, and a further decline would be considered bad employment news (even if the unemployment rate declined slightly). An increase in the participation rate, combined with a weak labor market, could lead to a jump in the unemployment rate. This is something to watch closely.

4) In August the unemployment rate was 9.643% unrounded (reported as 9.6%), so it won't take much of an increase to reach the consensus 9.7% for September.

5) This will be the last employment report before the two day FOMC meeting on Nov 2nd and 3rd. Anything close to the consensus will probably all but guarantee QE2 (barring a significant upside surprise for Q3 GDP to be reported on October 29th).