So the cliche goes, there is good news and there is bad news. I really can't give you much of a choice here, so I'm giving the good news first in order to explain the bad news. The number of jobs created for September was +110,000 and August's terrible -4,000 was swung violently to +89,000. That was the "good news", because it is not recessionary. But this is about as good as it gets. While the revisions are not so surprising (and revisions do seem to have an odd habit of making wild swings), the details that contributed to the revisions might be surprising.
It is crucial to point out that the source of the August revision was almost entirely due to goverment jobs turning from negative to positive. That is, we went from losing 28,000 government jobs in August to gaining 57,000, a change of +85,000. The health services sector also revised up. Not to sound selective, but when you take out government and health services, we lost 33,000 jobs in August and gained 29,000 in September, just to give you an idea of how dependent the employment numbers are on these two sectors. In other words, while the headlines say 200,000 jobs were created in the past two months, the details indicate that a hefty portion indicated a loss of 4,000 over the past two months. That would say that we're on the dip and if government jobs, particularly at the local level, return to cuts in the face of a draining housing market, the jobs numbers are far less secure than they appear.
So to reiterate a point: the government sector experienced a considerable amount volatility in August. Yet, everything from manufacturing to construction kept their losses. In addition to this, as expected, financials went from no change to losing 14,000 in August. As was widely speculated, the first August numbers were surveyed in the middle of the month and came before the many cuts in financials (especially mortgages) were carried out toward the end. Given the volatility in the government figures and the general steadiness in the private sector numbers, this may mean that August's numbers could be revised down to its original spot still yet.
Original figures for August:
•Manufacturing, -46,000
•Construction, -22,000
•Financials, 0
First revision for August:
•Manufacturing, -45,000
•Construction, -22,000
•Financials, -14,000
Private sector job growth is oscillating or, more accurately, it is descending in a staircase fashion. The highs are getting lower and the lows are getting lower as well. Again, while it is not yet recessionary, it is descending at a steady pace.
The September numbers were but marginally better than August's. Once again, health services led the pack, contributing 44,000 of the 110,000.
•Manufacturing, -18,000 (By the way, there are now less than 14 million manufacturing jobs with this report... another milestone trickles away in the great globalized economy)
•Construction, -14,000
•Financials, -14,000 (this marks the fourth monthly loss in 2007, the most monthly losses in a single year since 1994.)
It's very clear that the weakest sectors are showing the most consistent losses. Financials are now finally joining manufacturing and construction. The behavior of this first graph shows the loss of financial jobs is very similar to how construction losses started. It was slow to decline, but the trends were soon consistent. The contagion is fairly evident at this point.
The news isn't terrible just yet, but this is hardly a reason to feel relaxed. Expect Wall Street analysts to exclaim Goldilocks as alive and well, but the recession forecast is far from "wiped out" as John Silvia of Wachovia put it few days ago.
UPDATE
I misread the governemnt figures. It turns out the revision was +85,000, not +65,000.