Another month in the books, and another NFP report.
Job creation has been flat to negative for the past 5 months or so. According to Bloomberg, consensus for today’s non-farm payroll is a loss of -60,000, with estimates covering a broad range from -10,000 to -150,000. The most negative forecast was ING Financial Markets; Most positive was AIG Investments.
Although ADP reported a positive 40,000 on Wednesday, none of the economists surveyed are expecting a gain in NFP. For a while, ADP was on a roll, tracking pretty close to the BLS data. Suddenly last year, they diverged. Since then, ADP has been significantly overstating job creation versus BLS. At this point, their model is not a reliable tell as to what numbers BLS will generate.
Slack continues to build in the labor market. The employment-to-population ratio is now down to 62.7%. This is the denominator of the employment fraction, and is one reason why U3 unemployment has appeared so moderate. U3 Unemployment is expected to tick up to 5.1% — which is still relatively moderate. It is not that so many more people have jobs, but simply that less people are in the labor pool. The WSJ’s Ahead of the Tape column notes that "the ratio never fully recovered from the 2001 recession, partly because businesses have been more disciplined about hiring than in the late 1990s. Long-running secular trends — like women entering the work force — may also have peaked."
Regardless, we do know a few things before the 8:30am report: The overall employment trend is negative, especially in new 1) private sector job creation; 2) real wage gains; 3) hours worked. These all bode poorly for near term consumer sentiment and for non essential consumption.
Speaking of sentiment: Merrill Lynch’s David Rosenberg has pondered the relationship between low consumer confidence and soft — but not abysmal — job creation data. He tries to reconcile why merely moderate job losses have created consumer sentiment levels typically observed at much much deeper recession lows. The two are inconsistent with what you would expect from the current slowdown. Why should current economic measures — not showing an economy deep in recession by most traditional metrics — have a sentiment reading that is usually concurrent with a much worse environment?
One possible explanation: Job growth may have started to slow much earlier than previously thought — as much as a full year earlier. Evidence for this can be found at the Business Employment Dynamics (BED) section of BLS.
What are Business Employment Dynamics?
"Business Employment Dynamics is a set of statistics generated from the Quarterly Census of Employment and Wages, or ES-202, program. These quarterly data series consist of gross job gains and gross job losses statistics from 1992 forward. These data help to provide a picture of the dynamic state of the labor market…
Business Employment Dynamics measure the net change in employment at the establishment level. These changes come about in one of four ways. A net increase in employment can come from either opening establishments or expanding establishments. A net decrease in employment can come from either closing establishments or contracting establishments. Gross job gains include the sum of all jobs added at either opening or expanding establishments. Gross job losses include the sum of all jobs lost in either closing or contracting establishments. The net change in employment is the difference between gross job gains and gross job losses."
In theory, this should be a much less noisy data series, and provide a more accurate final read on total employment. They also form part of the basis for NFP revisions.
The data released up to the third quarter of 2007 suggest that private nonfarm payrolls could be revised down by 500,000 when the next benchmarking of payrolls takes place early next year.
Here’s where the correlation between sentiment and employment get reconciled: BED data says private payrolls stagnated in mid-2007. According to BED, there was virtually no private payroll growth during Q2 and Q3 in 2007. The pre-revision data from BLS shows more than 500,000 jobs created over same period of time period. Hence, a potential downward revision of half a million workers in 2007 is a probability.
The BED data implies that BLS has overstated NFP by 41k per month. This might help to explain the ADP errors, the very negative consumer sentiment 9versus where you might surmise it should be), and the retail strength in discounters over the past few months.
BLS releases NFP data at 8:30 am.
Business Employment Dynamics
U.S. Economy Probably Lost Jobs in May for Fifth Straight Month
Bloomberg, June 6 2008
Job Market Appears Likely To Bump Along
WSJ, June 6, 2008; Page C1