If Friday’s panicky 2% drop in the stock market was a reaction to the latest employment report from the Bureau of Labor Statistics, it was an overreaction.
Why so much fuss over a one-month change that's likely to be revised significantly in the coming months? I don't get it. The first Friday of each month seems to be, more often than not, the market’s overreaction to a highly uncertain monthly change in the employment numbers; last Friday could be a good example of that.
Before posting my usual monthly charts of the 12-month change in employment, I’d like to post an important passage from Barron’s editor Gene Epstein’s book (Econospinning: How to Read Between the Lines when the Media Manipulate the Numbers). This is the passage that explains why I originally chose the 12-month change, not the 1-month change, for the chart:
The change in payroll employment could be, in principle, reported hourly, daily, weekly, monthly, quarterly, annually, or decennially. It was decided many decades ago to release monthly updates. But just because that decision was made doesn’t mean it was prudent...
The BLS should announce that each newly minted monthly figure and prior-month revisions will henceforth be treated as though they add one-twelfth to our knowledge. A dozen of these out-of-focus snapshots must be laid side by side before we can discern what might be going on. So it would continue to release its monthly report, but the headline figure would be the 12-month change.
Okay, here’s the chart of the 12-month change, below. I see almost nothing but good news. If you spot anything alarming enough in it to justify a 2% market drop in a single day, please explain it in a comment. Click to enlarge:
And here are the numbers behind the chart:
And here’s the change in government jobs:
The change in payroll employment could be, in principle, reported hourly, daily, weekly, monthly, quarterly, annually, or decennially. It was decided many decades ago to release monthly updates. But just because that decision was made doesn’t mean it was prudent...
Right, in aggregate maybe the numbers don't justify the reaction. But the significance of a large and steady loss of manufacturing jobs, especially lost construction jobs in the midst of a housing decline, should be enough to give one pause.
A net gain in government employment does not fill me with joy.
And large increases in sectors ripe for reform (health and education) shows that priorities are out of line, and bode ill for the future when those sectors experience needed corrections.
So, month-to-month panic? Probably not. But overall I don't think these numbers are as rosy as you seem to believe.
Posted by: Mack | 10 September 2007 at 05:42
Why the concern? The job numbers for the previous 2 months were revised downwards by some 80,000
jobs. I fear the September numbers will be bad and that the August #'s will be revised downwards next month.
Posted by: malcolm | 10 September 2007 at 07:33
Money manager A hears about the news. He's not all that concerned but worries money manager B may be, so he sells. Money manager B hears the news and shrugs it off but worries money manager A might sell so he sells.
Repeat for all the letters in the alphabet and then some.
Posted by: Bob | 10 September 2007 at 08:41