trefoil
Registered: Mar 2007
Posts: 3381 |
02-04-12 05:52 PM
Employment is a market. Just like in the stock/bond/currency/commodities markets, it has trends. If the trend is up, in order to believe something extraordinary to the downside would occur in the employment market, therefore, you’d have to believe that there was a sudden reversal in the previous trends.
Which trends were pointing up? Let’s go through them one by one (I normally post a chart when I'm making an argument like this, but this post is going to be ridiculously long as it is, so only links will get posted. The charts will be with the links, as well as the data that supports the charts):
1 - The diffusion index. There’s two of these, one for all private companies, and one for manufacturing companies only. The idea behind this index is like Wilder’s RSI: above 50 signals up, below down. The private company version has been above 50 since March 2010, meaning more private companies have been hiring than firing for nearly two years by now. The manufacturing one is more volatile, and spent the first three quarters of 2011 trending down. In Oct and Nov it went below 50. But in Dec it jumped sharply, as did the private one. That was a sign that the uptrend in this was at minimum intact. Both increased again in this report. That would be consistent with a strong jobs number, which is exactly what we got. Given that, why would you believe a crazy stat that would seem to contradict it like 1.2 million people dropping out of the labor force all of a sudden? Only if you had an agenda.
2 - At its most basic, employment is a simple function of hours worked minus productivity. Hours worked (here, for production and nonsupervisory personnel, you know, folks who actually do something) has been rising sharply. Productivity (here) has been increasing, but at a progressively slower rate. It’s a picture of employers running up against the limit on how much more they can squeeze from their existing workforce. The logical conclusion is that more hiring at an increased rate is in the pipeline. Simple, basic stuff. So why would you believe 1.2 million people would disappear from the workforce at the same time? Makes no sense, unless you have an agenda.
3 - The birth/death index. This one is beloved of conspiracy loons. I don’t know how many times I’ve read about how the BLS is just making stuff up with the plug number that falls out of their attempt to account for new businesses they don’t know about yet in the establishment survey.
The mundane reality is that this number is in there to fix a simple problem: the establishment survey, by definition, can only go to businesses the BLS knows about. So they have to have a way of accounting for the new ones they don’t know about yet. Hence the birth/death index.
The evidence, from Table 1 of this month’s explanation of the yearly revision to their benchmarks, clearly shows that in recessions this number is consistently revised down: it was revised down in each of the four years preceding this one, from 2007 through 2010, with a particularly sharp drop in 2009, obviously in response to the 2008 crisis. But in recoveries it goes up to account for the new businesses that popped up in the previous year. It went up in 2004, after the recovery in 2003 from the recession of 2001 - 02. Last year it went up, as noted, for the first time in four years. That’s consistent with a recovery, not a continuing recession. It’s another piece of strong supporting evidence that this most recent report is more likely to be reporting something real rather than a chimera.
4 - Except for the December report, which was a wash, with one previous month revised up and another down, revisions to previous months in the establishment report have been positive. (All of the previous reports, with the figures as they were reported at the time, are available here: http://www.bls.gov/schedule/archives/empsit_nr.htm ) Famously, August, which came in at zero originally, was revised the next month to show a gain of 57000 jobs. This report showed a large increase in the estimate for November, and a small one for the December number. Once again, this is consistent with growth, and once again this supports rather than contradicts the data for this month.
By the way, the net number of new jobs isn’t the only thing that’s been getting an upward revision in the employment data. The manufacturing diffusion index, for instance, was originally reported to be 56.8 for December. That was revised sharply higher this month, to 64.2 for December, and the first report for January was an even higher 69.1. November, meantime, went from 40 or so to 48.1, or from a picture of hiring weakness to one of near steady state. So what we thought was building weakness in manufacturing employment turned out to just be a small dip. How do you square that with 1.2 million people suddenly up and leaving the labor force? Obviously, you don’t.
5 - The U3 argument: the last resort of tsing tao is that the U3 number is manipulated and reports a gross distortion of the facts, or something like that. But of course U3 isn’t the only series. Even at the top of the employment situation report, the BLS reports both the household number, U3, and the establishment number. The table to all the series is easily found. What that table shows is that all of the series have been trending down. It doesn’t by any means contradict a picture of growth in employment. Is employment a lot weaker than the U3 unemployment rate? Yep. Does that mean employment isn’t growing? Nope. Which means it doesn’t support a weird statistic like 1.2 million folks suddenly deciding not to look for work anymore.
6 - The declining labor force participation rate: yes, it’s declining. Yes, that’s a big worry. The evidence is that employers are sticking to hiring folks with good job histories only, which means it’s still a buyer’s market in employment. No one I know of would contend otherwise. Until employers work through the folks with good job histories, this isn’t going to come down. bwolinsky, Grand Supercycle, formal gold, and, by the looks of it, tsing tao, are still going to have a hard time getting gainful employment. Sad but true. This still doesn’t mean that another 1.2 million folks, a number that approaches 1% of the entire labor force, are going to drop out in a single month. The overall picture for people who’ve been out of work 5 weeks or more is still very bad, but it would have to be truly horrible for something like that to happen.
The point is that all of the information in points 1 through 4, all of which points to not just growth but growth that is accelerating, was available to tsing tao when he started this thread, if he had bothered to look at the employment situation report with an analytical eye, the one that can make you actual simoleons, rather than a political one, which makes you squat.
Looking at the weight of the evidence would have told him that something must have been off in that zero hedge calculation of the folks dropping out of the labor force.
But that’s what a conspiracy monger does. Fools and their money are quickly and efficiently parted, but this place is free, so fools can continue to post crap long after they’ve lost everything. Something to remember.
|
