The July 2010 personal income report published on Monday, August 30th, incorporated revised wages and salaries data from the BLS’s quarterly census of employment and wages for first-quarter 2010. The net revisions were negative, and such would be suggestive of lowered employment levels.

August Payroll Reporting. August 2010 payroll reporting is a fair bet to disappoint market expectations. To extent there may be eventual downward revisions to this year’s payroll revisions, recognition of such by the BLS potentially could lead to the softening of overly positive assumptions currently being used in the monthly payroll reporting, particularly in the monthly upside bias factors, otherwise known as birth-death model adjustments. That area will be looked at closely in tomorrow’s Commentary.     

The Census Bureau has reported that its temporary and intermittent hires for the 2010 census were reduced by roughly 116,000 in August, based on the payroll-survey weeks (84,000 census workers still remain to be discharged). The consensus estimate of Briefing.com has held at a total jobs loss of 120,000 jobs, with a Reuters survey estimate at a 100,000 jobs loss, and the WSJ.com consensus at an 80,000 total jobs decline. Implied there are consensus estimates of payroll changes, net of temporary census hiring effects, ranging from a decline of 4,000 to a gain of 36,000 jobs.   I look for an outright payroll contraction in August, net of census impact, with the total jobs loss at good risk of exceeding the 135,000 to 150,000 range. 

There are two cautions on tomorrow’s reporting. First, keep in mind that the BLS’s 95% confidence interval in reporting of monthly change is +/- 129,000 jobs. A change within that range really is nothing more than statistical noise. Second, once again, the ex-census number being touted by the markets and Administration is the "private sector" jobs change. That is not the same as total payrolls net of census hires, since regular government employment is ongoing, but it also tends to be in monthly contraction for the moment. Accordingly, the total payroll change ex-census — the number closest to reflecting economic activity — likely will be weaker than the private-sector jobs change.

Unemployment rate consensus estimates for the August headline U.3 rate seem to be holding at 9.6%, up from 9.5% in July. The jump in unemployment should be more severe, particularly when short-term (U.6) and long-term (SGS) discouraged workers are counted. The U.3 data are due for some catch-up in poor-quality seasonal factors that likely have contributed recently to dampening the unemployment rate level and for corrective adjustments for census perturbations.

Given the unfolding reality of a weaker economy (or re-intensifying downturn) than generally is expected by the financial markets, risks to reporting will tend towards weaker-than-expected economic reporting. Increasingly, such will be seen in economic reporting net of prior-period revisions.

In terms of related, underlying series, the Conference Board’s July help-wanted advertising index (newspapers), which leads August employment reporting, was reported again at 10 — unchanged for a number of months — and one point above its historic low. The help-wanted advertising (online) survey showed total ads declined slightly in August versus July (down 1.3%), while newly placed ads increased slightly (2.8%) for the month.

The August 2010 purchasing managers (ISM) manufacturing survey showed stronger employment numbers, with the August diffusion index reading (50.0 and above considered expansion) at 60.4, up from 58.6 in July. The employment component of the ISM services sector, which will be reported after the BLS numbers tomorrow, is a fair bet to show a softer August reading. Where the August surveys tend to lead September payrolls, the July numbers — which lead August payrolls — showed minimal strengthening.