Fri, OCT 5th, 2012
The scorecard is in, and the September employment climate is, well, mixed. Gasp! We know, you’re shocked. When the economy is below trend and the policy environment is uncertain, jobs should be growing on trees, right?
All funnies aside, the BLS Non-Farm Payroll [NFP] release was a wonks playground. Lots of numerical crosscurrents, reconciliations and analytical interpretations to feast on.
Without any further ado, let us get right into the gut of the report.
+114K gain in September [vs. +115K expected]
Private payrolls +104K [vs. consensus of +130K]
7.8% unemployment rate [lowest rate since January 2009] vs. 8.1% expected [unchanged]
August payrolls revised up to +142K [vs. initially reported +96K]; July was revised up to +181K [vs. initially reported +141K]
This brings the average NFP number from January through August of this year to about +146K/month, which is consistent with the 350-380K initial claims range over the same period.
These hard-to-reconcile figures — a meaningful drop in the jobless rate even though job growth was relatively weak and the labor force expanded — appear to be, at least partly, explained by a sharp increase in the number of Americans who found part-time jobs and “counted” themselves as employed in the survey.
Understanding this, it’s important to know that payroll numbers and unemployment figures are obtained by separate surveys thus they sometimes diverge in the short-run, but generally move in the same direction typically converging over the long-run. The unemployment rate is obtained by a survey of households, which tends to have elevated variability because of a smaller sample size. Conversely, the NFP number is a survey of business payrolls. In many prior months, positive changes in the rate have, to some degree, reflected people dropping out of the work force. That wasn’t the case in September [the participation rate actually ticked up +418K]. What drove the unemployment rate down was the +873K reading in this household survey and the corresponding drop in those “unemployed” in the survey.
Average earnings rose by seven cents to $23.58/hour, while the average workweek edged up +0.1 hours to 34.5 hours.
Remember though, as Matt Yglesias of Slate reminds us [paraphrasing here].
As the recent BLS re-baselining exercise that discovered almost +386K “missing” jobs was a reminder that the market [and the financial press] pays too much attention to this monthly data. It’s based on a statistical sampling with a large margin of error [+/- 100K jobs in the NFP survey; +/- 285K in the household survey]. It gets revised twice through this sampling data. It is then subject to further revisions as per this re-baselining mechanism. To be sure, a strong number in real-time is better than a weak number, but there isn’t a ton that can be inferred from monthly fluctuations until they’re well in the past and we have more accurate data.
To bottom-line this: if you choose one number to focus on, let it be the NFP net payroll additions. If you wish to get a bit more wonkish, pay most attention to the private payroll additions. And if you are bold enough to dig deeper than that, choose an antacid.
Jason L. Ware, MBA
Market Strategist, Chief Analyst
Albion Financial Group
(801) 487-3700; (877) 487-6200