The BLS says today that the January unemployment rate fell to 9.0% in January, from its previous 9.4%. Cool, you say, people are finding jobs! Except... err... they're not -- according to the BLS, January employment grew by 36,000 while the number of unemployed people declined by 600,000. So, err, where did those 564,000 people go -- did they simply drop out of the labor force? Err, no, the BLS *also* says that "the labor force was unchanged".
So we have 564,000 people missing. They're not employed. They're not unemployed. They just... aren't. So, where did these people go? Were they all rounded up and turned into Soylent Green? Well... err... no. They were adjusted. Which sounds vaguely pornographic, but is what the BLS has traditionally done in January -- decided that "seasonal" workers weren't really workers, they're just college kids and senior citizens getting a bit of extra money, and thus don't count.
The only problem is that the BLS has done this every year since the beginning of the Great Recession in 2007, and every year they've had to *undo* it and quietly issue a "revised" report within two months. The old rules that worked fairly well in the past just don't seem to work during the new economic conditions of the Great Recession. The people doing the seasonal work at Christmas in today's new economy aren't "marginally attached" seniors and college kids, they're just plain old workers trying to earn money for their families. And once it becomes clear that they aren't going to drop off the unemployment rolls just because because they did in the past, the BLS has to come in and revamp their figures -- again. For the 4th year straight.
So the reality is this: The Great Recession isn't over by a long shot. Three years into the Great Recession, we're still fucked. Calling it the "new normal" or disappearing half a million people don't change that shit, yo. Reality simply *is*, and attempts to "adjust" it so it doesn't look so bad are doomed, in the end, regardless of how much the BLS desperately wishes that things today were better.
-- Badtux the Economics Penguin
U-6! Is it below 17% yet?
ReplyDeleteThey claim U-6 dropped from 16.8% to 16.1%. But again, the size of the labor force hasn't changed, and only marginally fewer people are employed, so figure that they "disappeared" yet *another* half a million people with "adjustments" there....
ReplyDeleteI'm not sure what's going on here. Do they think we can't add Previous Employment + New Jobs and figure out that this number does *not* add up to the number of people who aren't unemployed, aren't employed, just... aren't?
- Badtux the Addition Penguin
And then you have the 99's, who disappear after their UI benefits disappear...
ReplyDeleteI think this is an important graph.
ReplyDeletehttp://research.stlouisfed.org/fred2/series/EMRATIO?cid=12
Also there are a couple posts at Angry Bear, but they just confuse me more.
http://www.angrybearblog.com/2011/02/household-survey-likely-paints-better.html
Frex:
Why the drop in the unemployment rate is credible. The summary statistics show the labor force falling by 504,000. The annual revisions dropped the labor force by 504,000, so the unrevised numbers show the labor force unchanged over the month.
That just makes my head hurt.
JzB
Do they think we can't add Previous Employment + New Jobs...
ReplyDeleteWe may be able to. But the NYT & WaPo can be counted on to not do so.
Maybe it's just me projecting my attitude onto news coverage, but it seems like the "headline" unemployment number is getting less space in the headlines these days. In the coverage I saw about the latest Bureau of Lying Statistics figures, there was not much trumpeting of "miraculous fall in unemployment." I think business section reporters and editors, corporate shills though they might be, are aware of the clamour in the blogosphere over U-3 vs. U-6 and other statistical manipulations. Maybe reality is intruding into news coverage -- what a concept! -- even when it conflicts with the official releases about the beauty of the emperor's new clothes.
ReplyDeleteIt's gotten to the point where U.S. official statistics are as unreliable as communist Chinese economic figures. If a person is not counted as existing by the government, like a tree falling unheard in the forest, does that person really exist? It's Orwellian, all these "unpersons."
Bukko -
ReplyDeleteI think you have a big misunderstanding here. BLS and the other data collecting agencies - Census bureau, BEA, St Louis Fed - are not in the business of lying. Politicians are in the business of lying about them, and reporters are sensationalists, not journalists.
There are good and bad reasons why the recent report looks weird, but lying is not among them - and weather is. Statistical anomalies lead to confusion, but the will get ironed out in revisions.
Now, the definition of unemployment itself might be a lie, but that is not the fault of the agencies, and U6 is a pretty important correction.
Cheers!
JzB
Indeed, the BLS is quite honest about the adjustments they did to the raw data to come to their current conclusion about the unemployment rate. The fact that their current conclusion about the unemployment rate is nonsense has more to do with their unwillingness to admit that the "new normal" has different rules than the ones they've traditionally applied to the January data, plus, as JzB mentioned, the weather had an impact on December employment yet that impact is not accounted for in the January adjustments, which assume a "normal" number of seasonal workers employed in December. But the raw data is still correct, and still available, if you're interested in drawing your own conclusions rather than the BLS's conclusions. It's a case of the BLS burying the lede, in the end, not of the BLS outright fabricating data.
ReplyDelete- Badtux the Statistical Penguin