I stared at the Bureau of Labor Statistics page for 30 minutes this morning trying to figure out how the nation lost jobs last month but unemployment went down slightly. Turns out I’m not alone in being confused. Reddit has a complete thread on the topic. By the way, Reddit’s economics section is really very good. Mises articles are often up-voted to the front page.
Source link: http://blog.mises.org/11160/how-is-this-possible/
{ 11 comments }
I can see it now:
“How is this possible?” Free market economists puzzled by success of stimulus and other government measures leading nation out of recession.
It’s the economic version of the climate change hockey stick method, just flipped so the their graph goes downward.
Oh Dear Jeffrey, does this mean that the Government Statistics are bogus. That they just pull these numbers out of a place where the sun never shines?
You have to believe that, since according to their own statistics, 2 million people filed new unemployment claims last month, and there are, on demographics alone, always another 100,000 new entries into the workforce each month, 2,100,000 people found new jobs last month?
Even the old USSR could not come up with lies like these.
Perhaps its time for a Lie Czar – someone to be in charge of all lies the Government tell. Think of all the people they would need to hire. Wait a minute, this would solve the unemployment problem.
The unemployed are being abducted by space aliens and eaten!
Last month was Thanksgiving so the ate more than usual!
Reference – Twilight Zone – To Serve Man – It’s a cookbook
Table A explains the drop in the unemployment rate nicely.
Unemployment dropped by .2% even though 11,000 jobs were lost and it should take at least 100,000 jobs just to keep up with demographics. Instead note the drop in the civilian labor force by 98,000.
Moreover, those “not in the labor force dropped by 291,000 constituting nearly all of the decline in unemployment.
I haven’t looked at how they measure this, but I am assuming that they have a category for jobs lost and jobs gained.
If jobs gained is higher than jobs lost, unemployment goes down.
If that is not the case, unemployment can still decrease in U3 because people may have moved into the “discouraged worker” category, thus removing them all together from U3. Pretty tricky, eh?
Yeah, what Hazi said.
Sounds like a lot of people stopped looking for work. “I’ll live off ObamaWelfare until the economy gets pumping again.”
There are several possibilities for this nonsensical result.
1) The seasonal adjustment factors can severely distort the results. This depression is unprecedented in modern times, longer and deeper than anything since the last depression, when the economy had a markedly different structure. Seasonal adjustments that are meaningful during normal times can become little more than random number generators under such extreme conditions.
2) The birth/death model consistently overestimates employment during recessions. John Williams at shadowstats.com has discussed this problem at length.
3) The “discouraged worker” adjustment removes people who have been unemployed for long periods from the rolls of the officially unemployed. This depression has persisted for so long that relatively large numbers of people may be shifted off the official population with each passing month.
4) Good old fashioned politics. This administration is in trouble on multiple fronts. They needed some good news. Good news was provided.
I haven’t performed a detailed analysis of these factors. I am identifying possible answers to Mr. Tucker’s question.
It isn’t that difficult. The BLS uses two surveys, one measures payrolls (jobs) the other measures people. I can have two jobs in Oct., lose one or not work in November, and the payrolls (jobs) would show a decrease of 1. Yet I would show up in the other as still being employed (no change to the unemployment rate).
If you want to argue how accurate they are, well, that’s a different story.
Ja said:
“I can see it now:
“How is this possible?” Free market economists puzzled by success of stimulus and other government measures leading nation out of recession.”
You’re absolutely right! We should print google-plex dollars and eliminate unemployment once and for all! Not only would we eliminate unemployment, but we would all be trillionaires!
Anyone who trusts the BLS’s unemployment figures either (1) doesn’t understand how these figures are calculated, (2) doesn’t understand the nature of the government, or (3) a combination of the two. Kind of like how the CPI showed 2% inflation in the summer of 08.
Come on, this is a no brainer. The government does count as unemployed those who are on welfare or who have been unemployed for longer than a certain period.
So in fact the unemployment figures do not capture those who most desperately need a job. You’ve hit on a very good measure for how bad things are getting. Unemployment going down while jobs continue to be lost is, in a way, measuring the level of chronic unemployment.
This is an bad sign.
Comments on this entry are closed.