« Housing Price Expectations: Getting Less Bad... | Main | Taking It Slow »
June 02, 2007
Is Unemployment Worse Than You Think?
Most of us think about the unemployment rate going down due to more people getting jobs. But there's also another way the official unemployment rate can go down. It happens when the denominator -- the bottom number of the fraction -- goes down.
And that is what has been occurring again recently. The Labor Pool has shrunk, making the unemployment rate look better than it actually is.
I dunno. Here's a look at the labor force participation rate for the civilian population, aged 16 and over:
It is certainly true that the participation rate has been heading down over the course of the year. In the longer view, however, the change looks pretty unexceptional. If you break things down by age, you see some pretty standard looking variation in the participation of prime-age workers...
... and yet another pronounced slide in the participation rate of 16-19 year olds:
That drop in the participation rate of teenagers accounts for about one-third of the decline in the overall participation rate. What's more, the participation rates of individuals over 55 have been essentially flat, a marked change from the last decade over which those rates steadily rose. Combined with the declining rate of the youngest group of workers, the tailing off of participation among AARP-aged workers is enough to explain the entire decline in the aggregate participation rate since the beginning of this year.
You might define "recently" as "since 2000 or so", and there you would be justified in claiming a broad-based decline in the number of people choosing to participate in U.S. labor markets. But I use the word "choosing" intentionally, as I'm convinced that the post-2000 changes in labor force participation rates (or employment-to-population ratios, if you like) reflect trends that are largely independent of the business cycle.
You may not join me in that belief, but broader unemployment measures -- those that account for discouraged workers, marginally attached workers (those not listed in the labor force, but who nonetheless say they want work), and part-time workers who say they would like full-time work -- don't suggest that the standard unemployment statistics are leading us astray:
Is unemployment worse than we think? I kind of doubt it.
UPDATE: pgl fills in the some of the numbers to the pictures above -- and notes that Dean Baker is not as sanguine as I.
June 2, 2007 in Labor Markets | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef00df351cad0f8833
Listed below are links to blogs that reference Is Unemployment Worse Than You Think?:
» Unemployment and the credit cycle from Interfluidity
Much of the chatter surrounding the latest BLS release has focused on a spike in the denominator of the unemployment statistic, the fraction of the population either working or actively looking for work. Courtesy of the indispensible [Read More]
Tracked on Jun 8, 2008 1:49:53 PM
Comments
Posted by:
Cassandra |
June 02, 2007 at 05:50 PM
Great analysis. Often people who cite the U-6 rate (or other measures) as evidence that the true unemployment rate is much higher never provide any historical context. They rely on people to compare the present day U-6 number to historical U-3 rates.
Example:
"We're being lied to. The real unemployment rate, the U-6 number, is 8%."
"Okay, is that good or bad?"
"Aren't you listening? 8%!"
Posted by:
Steve |
June 02, 2007 at 06:02 PM
David-
I wonder how the denominator gets adjusted for the illegal immigrants that are employed in this economy. Would that affect the unemployment numbers in a positive direction?
Posted by:
XP-77 |
June 02, 2007 at 08:35 PM
Thank you XP. Those wage slaves send their little checks back to Mexico, so shouldn't that dilute the claim that this should be counted as a US job? Moreover, isn't this crowding out the now-too-expensive carpenter for the cost effective 'material mover'? Is it possible that our productivity numbers are inflated because the hours put in by "illegal aliens " are not fully accounted for?
I like the Nat for describing this labor report as a downward shift to poorer remunerated service jobs and possibly not as durable either. Are these the kind of jobs that will get you into a GM truck? Not really.
Posted by:
calmo |
June 03, 2007 at 02:52 AM
"It happens when the denominator -- the bottom number of the fraction -- goes down."
Since when does a ratio go down when the denominator goes down? (given we are in positive numbers)
It's not because the denominator (labor force) goes down. It's because decreasing labor force participation improves the ratio between available workers and available jobs. This makes the nummerator decrease at a faster rate than the denominator, which decreases the unemployment rate.
Posted by:
pinus |
June 03, 2007 at 03:30 AM
The Denominator (Labor Force) gos down, total employment appears to goe up.
Unemployment rate% is 100 - total employment
Posted by:
Barry Ritholtz |
June 03, 2007 at 05:56 AM
As the Denominator (Labor Force) goes down, total employment appears to go up.
The Unemployment rate % = (100 - total employment)
Posted by:
Barry Ritholtz |
June 03, 2007 at 05:56 AM
Your first chart is totally consistent with my post:
The Labor Participation Rate was over 67.4% in 2001; It subsequently fell to about 2% to ~65.6%, It appeared to have bottomed and and began to improve in 2005 -- but has since started heading lower again.
Now, a 2% or so drop doesn't sound like a lot, but remember there are 143 million workers in the US. That drop equals about 3 million people. These are folks who are willing to take a full time job, have been unable to find work, and have exhausted their unemploment benefits.
They do not count in the "official" Unemployment Rate statistics. However, BLS actually does measure these folks in their "augmented unemployment rate" -- the jobless people who aren't counted among the officially unemployed.
That measure is 7.4%.
Posted by:
Barry Ritholtz |
June 03, 2007 at 06:09 AM
Nice point that the drop in the participation rate is from teenagers.
OK, next question. Why are teenagers dropping out?
possibilities:
the minimum wage is too low.
school has gotten more demanding.
crowed out by illegal immigrants.
What other idea do people have?
Is this a good thing or a bad thing? I can see arguments both ways.
Posted by:
spencer |
June 03, 2007 at 09:06 AM
It is common knowledge that if the US counted the unemployed the way Europe does, our unemployment rate would be comparable to theirs.
Posted by:
me |
June 03, 2007 at 09:53 AM
Dave,
Any rebuttal to Barry? We'd like to hear your counter arguments to his analysis.
Posted by:
zorro |
June 03, 2007 at 11:19 AM
David,
I'm not sure of the endpoint of your data, but it doesn't look like it runs through May. If you look at emmployment growth through May, you get 567,000 for the AARP set and -14,000 for everyone else. There is a larger question about the behavior of the EPOP over this recovery (weak in my book), but if we're just looking at what's going on this year, it definitely looks like its headed downward to me.
Posted by:
Dean Baker |
June 03, 2007 at 11:29 AM
Barry Ritholtz: That measure is 7.4%.
"Okay, is that good or bad?"
Posted by:
Steve |
June 03, 2007 at 12:39 PM
Barry Ritholtz: These are folks who are willing to take a full time job, have been unable to find work, and have exhausted their unemploment benefits.
In your attempt to describe these people, you forgot one other pretty important characteristic. These are people who are NOT looking for work. If they were actually looking for work, they would be counted as unemployed.
Posted by:
Steve |
June 03, 2007 at 12:41 PM
Perhaps rather than unemployment being worse than we think, it might be better put employment is worse than we think that fewer wish to work.
Posted by:
Lord |
June 03, 2007 at 01:14 PM
Steve wrote:
" Barry Ritholtz: That measure is 7.4%.
"Okay, is that good or bad?" "
Depends.
If it´s the unemployment rate of one of those stagnant, sclerotic European countries, then it´s bad. :)
At least for "The Economist", "Financial Times", "WSJ" ...
Posted by:
Detlef |
June 03, 2007 at 02:15 PM
Dean - the labor force participation rate for the 25-54 age group was 83.0% as of May 2007 v. 83.1% as of Feb. 2007. This strikes me as noise in the data.
Posted by:
pgl |
June 03, 2007 at 03:34 PM
Another thing that is not counted in the unemployment rate specifically is the amount of people becoming entreprenuerial and starting their own businesses. In this latest economic boomlet (since 2003), the amount of individual economic activity has increased. It is not just a bunch of companies hiring new people, but it's people choosing to make money on their own.
BTW, 7.4% if counted under tradtional perceptions of unemployment would be bad. 5% was considered frictional or marginal unemployment.
Posted by:
jeff |
June 03, 2007 at 03:34 PM
David and Dean - just up at Angrybear, the declines in the labor force participation rate by major age groups from Dec. 2006 to May 2007. I report - one's own angst is up to the reader.
Posted by:
pgl |
June 03, 2007 at 03:51 PM
"BTW, 7.4% if counted under tradtional perceptions of unemployment would be bad."
7.4% is the "augmented unemployment rate." But in order to get to that you have to count people who aren't even looking for work. Should these people really be counted as part of the labor force if they aren't even seeking employment? Who are these people that can just say, "Yeah, I'd take a job if one fell into my lap, but I'm not going to look for one."
Further, 7.4% is meaningless without some historical context. What's the average "augmented unemployment rate"?
Posted by:
Steve |
June 03, 2007 at 04:32 PM
Steve,
..."Should these people really be counted as part of the labor force if they aren't even seeking employment?..."
Are you sure they aren't seeking for employment?
Is it a possibility that they have tried but could not find one?
In today's globalization, there are lots of reallocations of resources and employment is one of them. The erosion of US' manufacturing jobs is not a secret!
Posted by:
zorro |
June 03, 2007 at 05:29 PM
zorro,
If they were seeking employment they would be counted as part of the labor force and included in the U-3 unemployment rate.
Posted by:
Steve |
June 03, 2007 at 05:57 PM
XP-77 (yay!) -- The data is from the household survey. In principle illegals would not be excluded, but I suppose they might not be so responsive to government inquiries.
Barry -- I agree that there is pretty big drop since 2001. But as I indicated in the post, I'm pretty convinced that this is a secular change. (zorro -- I guess that's my rebuttal)
spencer -- Right question; wish I knew the answer. What interests me is that we have gone through two cycles now where this was the pattern at the time of a recession. I hope the recent drop isn't a leading indicator.
Dean -- The pictures do go through the May data. I don't dispute that things look like they have stepped down from last year. The question remains whether this is supply or demand issue.
pgl -- We agree. So far I'm on the non-angst side of the equation.
Posted by:
Dave Altig |
June 03, 2007 at 07:32 PM
Barry Ritholtz - "Now, a 2% or so drop doesn't sound like a lot, but remember there are 143 million workers in the US. That drop equals about 3 million people. These are folks who are willing to take a full time job, have been unable to find work, and have exhausted their unemploment benefits.
They do not count in the "official" Unemployment Rate statistics. However, BLS actually does measure these folks in their "augmented unemployment rate" -- the jobless people who aren't counted among the officially unemployed.
That measure is 7.4%."
Where did Barry find this statistic?
I have been using the U-6 alternate total unemployed measure - Table A-12
http://www.bls.gov/news.release/empsit.t12.htm
That table states:
"U-6 Total unemployed, plus all marginally attached workers, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all marginally attached workers" - May 2007 - 8.2%.
Last May, U-6 unemployment and underemployment was 7.9%.
Is Barry referring to a different measurement statistic available from BLS? Or did he pull his number from Table A-13?
Posted by:
Movie Guy |
June 03, 2007 at 11:33 PM
David,
I agree that a lot of it is secular -- that still shouldn't prevent us from reproting the data accurately, rather than polished and spun.
Movie Guy:
That 7.4% is the augmented unemployment rate as reported by the BLS
~~~
Steve:
You are incorrect. The qualification for being in the labor pool is more complex than merely "looking for work." How that is defined is very specific, but to greatly oversimplify, once you exhaust your benefits, you start a statistical process that leads you being removed from that labor pool.
Otherwise, there would be an underclass of perpetually un-employed/under-employed workers, and we cant have that...
Posted by:
Barry Ritholtz |
June 04, 2007 at 07:13 AM
Big Picture makes up a lot of stuff without any facts