More On The April 2011 Unemployment Report

Earlier we reviewed the April 2011 BLS unemployment report. This post goes into a more unemployment and jobs details not covered previously.

There are 13,747,000 people officially unemployed, and if one takes the alternative measure of unemployment, it's 24.4 million.

Below is a new graph, courtesy of the Saint Louis Federal Reserve, for the unemployment rate of just those unemployed 15 weeks or longer. 57.1% of the people who are officially counted as unemployed have been so 15 weeks or longer. The number of people unemployed 15 weeks or longer as part of the civilian labor force is 5.1%. Compare this to the official 9.0% unemployment rate. This means that it takes most people who are unemployed almost 4 months or more to find a job...if they do.



There are 242,000 more people who are unemployed for less than 5 weeks in April. People out of work 5 to 14 weeks dropped -7,000 and 15 to 26 weeks unemployed increased 49,000. Below is the graph of those unemployed for less than 5 weeks. We see an uptick, which is the wrong trend.



The below graph is the median time for being unemployed over the last 20 years. Look at the pathetic dramatic increase. That means there is overall lost income by being out of work longer and we haven't even discussed how the new jobs are paying way less and most of the jobs being created are low paying jobs.



The long term unemployed (> 27 weeks) decreased by -283,000 in April. The problem with this number is we do not know the percentage of those who plain fell off the count and didn't get a job. 45.5% of the official unemployed have been so for 27 weeks or longer. Those employers refusing to give a job to someone who needs a job, refusing to hire someone unemployed....should be sued for discrimination in my opinion.

  • Long term unemployed - 5.84 million
  • Forced Part Time - 8.6 million
  • Marginally attached to the labor force - 2.47 million

Of the Marginally attached, 989 million are discouraged. Forced part-time increased 167,000 this month, with those reporting the reason is slack business conditions increasing by 94,000. People who could only get part-time work increased 148,000 in April. The 2.47 million marginally attached are not counted in the official unemployment rate, but you can be guaranteed they do need a job.



U6, defined as total unemployed, plus all persons marginally attached to the labor force, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all persons marginally attached to the labor force, (table A.15), was 15.9%.

One of the most telling signs that people have dropped out of the count is the increase of those not in the labor force, which increased 131,000 from March to April

To even get back to pre-recession unemployment rate levels we need a good 424,000 jobs created each month. That's permanent jobs. Even if the U.S. created 400,000 permanent jobs each month, it would take over 17 months to recover the jobs lost since December 2007, the start of the great recession, or 6.96 million jobs. If one created 400,000 jobs per month, to get to the start of the great recession unemployment rate of 5%, or roughly 10.7 million jobs, it would take over 26 months, or over 2 years to recover. Below is the seasonally adjusted monthly change in the number of payroll jobs in the U.S. See a lot of over 400,000 monthly increases in the past decade?



In looking over table B1 we can get a little more detail on what kind of jobs were created (and lost) on the permanent jobs front.

  • Financial: +4,000
  • Information: +2,000
  • Construction: +5,000
  • Manufacturing: +29,000
    • Durable: +19,000
  • Mining & Logging: +10,000
  • Health and Education: +49,000
  • Leisure and Hospitality: +46,000
  • Professional & Business Services: +51,000
    • Temporary Help: -2,300
    • Professional &technical services: +33.000
    • Administrative & Waste services: +17,500
  • Trade, Transportation, Utilities: +71,000
    • Retail Trade: +57,100
  • Government: -24,000
    • Local Government: -14,000

While for some reason public or government workers are under assault, local government employees are clearly hurting. There has been 409,000 local government jobs lost since September 2008.

Below is the graph of the monthly change in manufacturing payroll jobs, seasonally adjusted, against the ISM Manufacturing Employment Index, which dropped -0.3 percentage points for April.



Bear in mind illegal workers and foreign guest workers are counted as employed Americans, which can distort occupational sectors, particularly the Engineering, I.T., Science and Technology ones. Below are the unemployment rates per occupational sector from table A-14, not seasonally adjusted.

  • Financial: 6.7%
  • Information: 7.1%
  • Construction: 17.8%
  • Manufacturing: 9.4%
  • Transportation, Utilities: 8.4%
  • Mining & Gas, Oil: 3.5%
  • Health and Education: 5.0%
  • Leisure and Hospitality: 11.7%
  • Professional & Business Services: 9.1%
  • Retail, Wholesale Trade: 8.8%
  • "Other" Services: 9.2%
  • Government: 3.5%
  • Agricultural: 13.7%
  • Self-employed, unpaid family: 5.5%

Probably the most horrific not seasonally adjusted unemployment rate is the 9.1% for Professional and Business Services unemployment rate. This occupation category implies a college degree. This particular area is fraught with foreign guest worker Visas, which distorts the real occupational unemployment rate. If an American is displaced by a foreign guest worker, then is forced to take a job at Walmart, that person is no longer categorized as a Chemical Engineering PhD out of work, but instead as a retail trade worker.

The above numbers are not seasonally adjusted so part of the agricultural and construction rates are due to the seasons, that said, construction unemployment has been decimated anyway you look at it.

The BLS also notes most of the new jobs in leisure & hospitality are low paying:

Over the past 3 months, this industry added 151,000 jobs, with nearly two-thirds of the growth in food services and drinking places.

The average work week remained the same, 34.3 hours, yet manufacturing average hours are 40.4. Average earnings for the private sector (drum roll please), increased 3¢, $22.95/hr. Yet the worker bees, or non-supervisory jobs average pay increased 5¢ to $19.37/hr. Take a look at how many industry sections are below full time hours. Is a common practice among employers to keep workers part-time, in order to deny them full-time benefits. These numbers have improved to more full time per industry.



You probably also want to know the birth/death model. What is that? It's a statistical adjustment to compensate for new businesses and dead businesses who are not actually tallied by data reports. Those jobs created and died outside the statistical reporting time window due to lag. So, the BLS estimates how many jobs can be attributed to those firms which are not actually counted. This month's adjustment was 175,000 jobs. One cannot directly subtract the birth/deal model monthly numbers, because unemployment data is seasonally adjusted, yet the birth/death adjustment is not seasonally adjusted, get that? Anywho, jobs attributed to new and dead businesses are just an estimate in so many words. Not seasonally adjusted, payrolls added 1,169,000 jobs. This is the reason why one cannot apply the birth/death model to seasonally adjusted data.

Subject Meta: 

Forum Categories: 

Which Turn is Worse?

When we turn the corner in Afghanistan or with unemployment?

What event will come first: the last infantryman will leave Afghanistan or the last person laid off in 2007-2010 will get a new fulltime job? [For extra bonus points, give year for your predictions.]

News coverage mostly on the cheery side

My observation is that news coverage of the report has been mostly unjustifiably on the cheery side. The Great Recession and the Great Depression are alike in that for many years into each, mass media was (is) all about an upbeat take on whatever straws were (are) tossed to the people trying to survive overboard. It's all smiles.

It's obvious from Robert Oak's presentation and analysis that the best we can really say is "It could be worse!" And that is how working people around me are taking this, if they pay any attention to it at all: "Well, it could be worse."

"Rain falls on the just and the unjust alike." Life goes on. Spring is here. Summer is coming. It's all good.

For stoicism, I believe the American people are right up there with the Japanese.

well, the news also gets the basics wrong

I was watching local news and they claimed the reason the unemployment rate increased is people are entering into the labor market. Completely false from the labor participation rate and the additions this month into the civilian labor force.

Still, 244,000 jobs is better than -900,000 jobs but not a great jobs miracle, which is really what we need. If we did not have a jobs crisis this would be an ok showing.

Also, we have people trying to claim the reason the labor participate rate has dropped is because of the baby boomers. Well, well, one can break down these rates via the BLS data by age bracket and that claim is simply false.

RE: Well, the news also gets the basics wrong

244,000 may be better than -900,000 but one of the problems seems to be that 244,000 isn't even close. From what I've been reading 175,000 of these "jobs" were directly attributable to the Birth/Death Ratio (i.e., pie-in-the-sky) and the another 60,000 or so were the part-time jobs created by McDonalds - a one-time shot. That leaves somewhere around 14,000 to 20,000 real jobs. When it comes right down to it, for all practical purposes it feels damn close to -900,000.

birth/death model Folks you CANNOT mix SA with NSA!

Everytime I do these overviews I quote the birth/death model and go through the entire thing as to why you CANNOT mix seasonally adjusted numbers and not seasonally adjusted ones.

So anyone subtracting the birth/death model from the numbers is plain wrong. Anyone mixing seasonally adjusted numbers with not seasonally adjusted numbers is also reporting nonsense figures.

Ya must use apples to apples, it's either seasonally adjusted or not.

Then, I point to the birth/death model because as one can see when using the nonseasonally adjusted numbers, the additions are small by comparison to the not seasonally adjusted payroll additions.

This is statistics folks and I dig into this pretty deeply and if there is fiction or a question, I almost always state it, or chase down an answer in the overview.

My beef with the BLS/Census is the sample size of the CPS, the refusal, probably by congress, to not keep track of citizenship status of workers, much more relevant than their antiquated 1960's "foreign born" categories due to global labor arbitrage, the use of the base Census 2000 data when I think that should be updated better and also 2010 should be incorporated faster and then yearly adjustments, which many of the statistics do, but esp. the Census CPS yearly adjustments, put into one month, between December and January.

Ok, I put the birth/death model in the post simply because people talk about this all of the time and make the same damn calculation errors every single month...

so again, you cannot add/subtract that model from the nonfarm payroll numbers of the headlines, i.e. 244k jobs.

Read the bottom of this post. Huge seasonal adjustments. The real problem is the BLS needs a larger raw data sample size and that my friends, requires Congress to increase their budget for more accurate data.

Truly, statistically, they are not making something up here, the real problem is the raw data they can get at. This is a large country.

Quite a value added post

This is the reality we face, not the PR lies from the usual suspects.

We live in a dead zone economy. There have been no net new jobs since 2000, adjusted for population growth. In fact, there's a significant decline. The "official" unemployment figures show a work force of 154 million, with 138 million or so employed (compared to 135 million employed with a 143 million job force in 2000).

But help me here. If we use U6, which is the figure to use, imho, what would the workforce size be? (Is this an albegra problem;)

add in marginally attached

but that's not seasonally adjusted so you have to seasonally adjust and it's not considered part of the labor force. That said, it's the participation rate and we have various people trying to claim it's the baby boomers retiring that it is so low but I don't think so.

Firstly, baby boomers were wiped out, multiple, so many cannot retire at all and then if one looks at labor participate per age group, that too has declined dramatically.

I've got about 4.5 million no where in the count who assuredly need a job. (that's the U-6 count).