There's a lot of moving parts in today's employment report. First, yesterday we got the Q4 2011 B.E.D. report.
That was rather pessimistic, because it told us we only added a net 368,000 jobs in Q4. Not what we were looking for.
For today's employment report
, the schizophrenia between the Household survey and the Establishment survey continues. The difference is that we flipped poles, so now the Household survey is quite negative and the Establishment survey is positive.
The headline stuff is:
- Unemployment rate rises again to 8.3%.
- The emp/pop ratio falls to 58.4% from 58.6% in June.
- Household survey reports net SA loss of 195,000 jobs in July.
- Establishment survey reports net SA gain of 163,000 non farm jobs in July.
It would have been nice if these two surveys could have gotten together and shared notes, but it was not to be. The actual numbers of the employed didn't rise by that much, although they did rise. The increase in the unemployment rate is really due to another drop in the participation rate (63.7%).
In terms of a healthy economy, drops in participation rates and large increases in the not-in-labor-force (+348,000) numbers aren't positive indicators.
You cannot blame the Birth/Death adjustment for a relatively large discrepancy in jobs between these two surveys. This month's B/D adjustment is only 52K jobs out of a non-SA monthly change of -1,204,000. That's not even a footnote.
I BELIEVE that most of the discrepancy this month is due to seasonal adjustment factors and timing of manufacturing shifts, but I can't prove it. We'll find out next month. SA factors for June can be problematic. The truth is going to lie somewhere in between the two survey outcomes.
Because these surveys tend to catch up with each other over time, I expected a good Establishment number (high 120s) and a poorer Household survey number. The magnitude of the change each way catches me by surprise.
The two month numbers on the Household survey are now:
- 67,000 net employed.
- Virtually no change in civilian labor force.
- 74,000 more unemployed.
The best description of what we see is that of a stagnant labor market, which did improve slightly relatively in July, mostly due to retirements.
I'll probably have more later sometime on the weekend as I work through some of this stuff. The women-who-maintain-families unemployment rate stayed up at 11.7%. While a marginal drop from June's 11.8%, it is a depressing and highly negative indicator compared to April's 10.2%, and unfortunately the details of the Establishment survey strongly support the accuracy of this rise. The cuts in various types of more discretionary retail would generate most of the increase.
To me the picture is consumer economy in early stages of recession, with manufacturing doomed to follow, both due to domestic demand issues and due to global weakening.
In 2006 we had slid into a manufacturing recession, but the consumer credit bulge continued the expansion for almost another two years. This time we are seeing the consumer recession first, and with any luck that would indicate something structurally milder. However on top of the current economic malaise, next year we will add other negative factors, so the possibility that it will remain mild is not that great.
My current takeaway for this report is that it is a personal highly negative surprise. I expected a gain in the Household survey of 70K jobs, and a negative surprise of 270K jobs is one heck of a negative surprise. I think in the long run the positive on the Establishment survey will prove to have mostly been timing issues, partly related to auto production schedules, and that it will drop out again in August.
In short, I suspect the tide is going out and that there are some missing Speedos under the water.