- by New Deal democrat
This is going to be a sparse week for data, with the exception of tomorrow’s long-delayed Q3 GDP report, and jobless claims on Wednesday. Sadly, so much of the data is still missing or stale that the best source for up-to-date information is in the regional Fed reports, most of which will be updated by this Friday (so stay tuned for that). And don’t be surprised if I play hooky for a day or two.
That being said, one important - and positive - data point can be updated based on last week’s November jobs and CPI reports: real aggregate nonsupervisory payrolls. To recapitulate, these always peak before a recession begins, usually within 3 to 6 months. And there is a very good fundamental reason for that: once the average American household has less cash to spend in real terms, consumption promptly gets tightened, and that downturn in consumption typically brings about all the other indicia of recession quickly.
But the news from November was good. For the two months covered by the updated jobs report, nominally aggregate payrolls increased 0.9%. Meanwhile, the official cpi index only increased 0.2% for the two months from September through November, meaning that real nonsupervisory payrolls increased 0.7%. In the graph linked to below, November’s level is set to 100, which is the only visibleway to show the increase since September:
Note that even if the much-criticized shelter increase of 0.1% were instead changed to 0.3% each month, the average over the previously reported months this year, real aggregate payrolls would still have increased 0.2% for the two month period, still a new high.
Nevertheless I recommend taking this will a heavy dose of salt.
On the negative side, it is difficult to imagine such a weak labor market not being on the cusp of, if not already in, a recession. As of November, service providing jobs were only up 0.7% YoY, while goods producing jobs were down -0.15% YoY. A shown in the graph linked to below, which normalizes both readings to zero, only once in the past 85 years - in 1944 - has employment in both sectors been this low YoY without either being already in, or at least on the doorstep of, a recession:
Keep in mind, by the way, that this data is not yet adjusted for any of the QCEW reports this year, which have suggested at by the end of June, employment was only up 0.3% compared with 12 months before, as opposed to the 1.0% higher indicated by the current nonfarm payrolls surveys.
We are still in many ways flying blind. In particular, we really need to see reliable real sales, production, and consumption data through the period of the government shutdown to determine whether or not that self-inflicted wound pushed the economy into contraction or not. Without it, any conclusion I might reach would just be speculation.
