A brief note on polling and unemployment data analysis

posted at 1:51 pm on October 5, 2012 by Ed Morrissey

I’m noticing a couple of odd reactions to today’s jobless data and the analysis it has produced so far today.  On one hand, we have people proclaiming the difference between the household and establishment data as proof of a conspiracy to re-elect Barack Obama at the BLS.  On the other, we have a number of voices calling anyone who applies any kind of critical analysis to these figures as “truthers” or “deniers.” Tom Elia has collected some of the latter at The New Editor, but I’ll focus on explaining what exactly the figures are, and why critical analysis does not amount to “trutherism.”

The BLS conducts two surveys each month to determine employment data.  The first is the establishment survey, which polls 410,000 businesses each month.  The second is the household survey, which polls 60,000 households each month.  As one might expect, the larger survey provides a more stable series and more reliable data.  The smaller one is still a very significant sample, but in a nation of around 150 million households, it’s hardly an exact science.

Any poll series can produce an outlier result.  That’s true even of larger sample surveys, even when the sample is properly balanced, and even without malicious intent to tweak the results.  It’s more likely to happen with smaller samples than larger samples, but can happen any time in surveys.  That’s why it’s important to look at trending more than a single result within polling series, although some applications (jobless rates, elections) are intended more for single-result reporting.

In today’s case, the establishment survey showed a result that corresponds closely to other economic trends and that doesn’t deviate much from the intraseries trend.  The household survey, from which the jobless rate is derived, showed a very large deviation from its own trending and from the growth data in the economy.  The last time we had that many added in the household survey, the GDP growth rate was around 9%, and it’s currently 1.5%.

That’s why people who understand data and surveys look skeptically at the result of the household survey.  It doesn’t mean a conspiracy is in place; it does strongly suggest that this month’s sample of 60,000 households threw an outlier, especially when compared with the establishment survey and other economic data.  If so, it will likely correct itself in the next report.  That’s not “trutherism” or denial, but straightforward data analysis.

Update: Just to remind everyone, the next report comes out before the election — the Friday before, actually.  And if the BLS wanted to cook the numbers, I’m pretty sure they’d have cooked the establishment survey, too, to show more than +114K.

Breaking on Hot Air

Blowback

Note from Hot Air management: This section is for comments from Hot Air's community of registered readers. Please don't assume that Hot Air management agrees with or otherwise endorses any particular comment just because we let it stand. A reminder: Anyone who fails to comply with our terms of use may lose their posting privilege.

Trackbacks/Pings

Trackback URL

Comments

Comment pages: 1 2 3

Just to remind everyone, the next report comes out before the election — the Friday before, actually. And if the BLS wanted to cook the numbers, I’m pretty sure they’d have cooked the establishment survey, too, to show more than +114K.

Why?

The household survey is far more subject to manipulation. BLS can simply come back later after the election and “correct” it, while pointing out that the employer survey was correct all along. The fact that the surveys differ did not prevent BLS from claiming a massive increase in employment from the erroneous survey, so there is no reason to cook the employer survey as well.

The odds are stratospheric against the 6,000 person national survey from randomly arriving at an outlier result of 800K new jobs when the employer survey showed only 114K. Those odds reach the moon against such a result happening after an opposing party has hammered the incumbent for setting new length records for unemployment and in the October before the election.

This one is an outlier too far.

Bart DePalma on October 5, 2012 at 3:25 PM

So the household survey is just that, a survey…a series of questions being asked. It is much more than asking “Do you have a job?” I am beginning to wonder if the reason the household survey number changed is because the survey questions themselves changed. Any researcher will tell you that you can a different answer to the same question just by changing around the wording. I wonder if that was the case in this situation. I don’t know if the questions and scripts used by the BLS surveyors each month are publicly available, but I would wonder if a change was made this month.

aaron31 on October 5, 2012 at 3:27 PM

There are statistical anomalys and then there are unbelievable coincidenses…

Do some stats on how likely is it that a 30 year drop happens just after Obama is demolished in a debate… what are the chances of that happening? How many months in 30 years?

More like statistical tampering… the simplest answer is often the best. And there is no simple explanation for this, other than tampering by the BLS. Which also happens to be among Obama’s most loyal campaign contributors.

I was not born yesterday. I dislike paranoia… but sometimes there really is someone out to get you!!!

petunia on October 5, 2012 at 3:31 PM

Of the 873K surge in employment in the Household Survey 582K was in temp jobs and 187K was in government jobs..that all you need to know.

galtg on October 5, 2012 at 3:31 PM

That’s why people who understand data and surveys look skeptically at the result of the household survey. It doesn’t mean a conspiracy is in place; it does strongly suggest that this month’s sample of 60,000 households threw an outlier, especially when compared with the establishment survey and other economic data. If so, it will likely correct itself in the next report. That’s not “trutherism” or denial, but straightforward data analysis.

.
Ed, I used to get my weekly global numbers preliminary report every Monday morning by 8:00 AM. By 8:15 AM, I would have spotted any outliers, checked the raw data and be on the phone to the corporate accounting group telling them where they screwed up.

The BLS has screwed up. Did they do it intentionally? Well, here’s a clue:

http://www.zerohedge.com/news/2012-10-05/strangest-number-todays-jobs-number

The chart above (PA: click on link to see chart) shows the sequential change in Non-Seasonally Adjusted jobs for the 20-24 year old cohort (aka those who normally are in student age) into the month of September going back to 1980, as represented by the Household Survey.

We have shown just 22 years of data, but believe us: in this data set, the September NSA jobs change has been negative every single year since the beginning of data collection. Except for 2012 (and considering the surge in temp-jobs for economic reasons, one can be certain that if indeed correct, all these young people obtained primarily part-time jobs, if any).

.
Wait, there’s more:

A valient effort but one thing remains outstanding: the record amount of student loans outstanding, and defaults, which as we explained last Friday, is indicative of one thing: everyone is doing their best to avoid the labor market in this worst possible time for jobs and is hiding instead in the “safety” of a Federally funded college education. This explains not only the record amount of student loans outstanding, well over $1 trillion in total, and over $900 billion just Federally funded.

So somehow in September, in addition to all the other discrepancies in the labor report, we have one more to add: that of the Schrodinger Student: one who is both in college and piling up student loans on one hand, yet on the other hand entering the work force in the month of September, a time when historically every single month in recorded history has seen an exit from the labor force for the 20-24 year old cohort.

.
The track record of the Obama BLS since February 2009 is one of consistent understatement of bad results later corrected out of the headlines and impossible birth/death and ESTIMATED additional jobs created.

Did they do it on purpose?

Even Joe Scarborough could smell the stink on these numbers first thing this morning.

PolAgnostic on October 5, 2012 at 3:38 PM

Any theories? Why would Republicans bolt to Obama. The only theory I can think of is mormonism, but Obama went to a black liberation theology church. I’m a southern baptist and while we are skeptical of mormonism in general, i haven’t heard any republican say they won’t vote for him because of it. I just cant figure it out. Maybe they are messing with the pollsters?

Donald Draper on October 5, 2012 at 1:55 PM

I think some of it could be Democrats lying for the household survey to improve the number. The establishment survey is a lot harder to fudge.

blammm on October 5, 2012 at 1:58 PM

Ask yourself if a Democrat who was contacted for either type of poll (political or BLS survey) would have an incentive to lie.

The answer is yes. They could lie and claim to be a Republican who intends to vote for Obama. And they could lie and claim to be employed when they are not. Both types of lies would make Obama look better.

I am not aware of any way to measure and correct for these lies to pollsters.

But as to the employment numbers, I the ADP numbers and BLS establishment survey numbers do not show anything like what the household survey showed for September. There is not ANY other economic indicator that gives credence to the employment gains claimed in the household survey.

ITguy on October 5, 2012 at 3:41 PM

Bart DePalma on October 5, 2012 at 3:25 PM

Just so. It’s almost impossible to come up with such a deviation using sound procedures. Whether intentional or not, something is very wrong in that household survey.

Sorry, Ed, your reassurances are meaningless, and Jack Welch’s and Douglas Holtz-Eakin’s doubts are relevant.

And Obama is out there campaigning with Happy Days Are Here Again. Yeah, it could be accidentally an outlier. Believe what you want to believe.

Adjoran on October 5, 2012 at 3:43 PM

Thanx for the explanation, Ed, but–just bcuz you’re paranoid doesn’t mean they’re not out to get you. I still find it quite amazing that this “outlier” happened at this point in time, just in the nick, the cavalry rides in, Obamuh puts on the U3 bling just when he needs a shiny object the most, mounts his mighty white steed to go in and pretend to save the people especially when this magical 8% U3 rate dividing line with regard to an incumbent’s re-election has been discussed soooooo at length in the past year. It’s just too bizarre to be coincidental and that many stars haven’t just lined up all of a sudden. Even if the numbers do retreat to within line next month that could also be the result of either mission accomplished or Obamuh’s toast and it doesn’t matter anymore

It would be a stronger point IF there was little to no reason to suspect lying, truth-twisting, cheating, hypocrisy, fraud, and the myriad of other shenanigans that Obamuh and his merry marauding campaign pull off every day. They lie about 12 times a minute. And don’t forget this is the campaign that for the second time in a presidential race is likely under suspicion of campaign donation fraud but somehow manages to bury the evidence and wiggle it’s way out of the humiliation it so richly deserves.

stukinIL4now on October 5, 2012 at 3:44 PM

If I have a couple of items up on Ebay does that mean that I am employed at home?

claudius on October 5, 2012 at 3:50 PM

Explain it all you want, but the consensus of a vast majority of Americans is: Zero and his labor staff are liars. Major backfire here for barack. Baaaad hail Mary. Baaaad. It was intercepted by reality.

Rational Thought on October 5, 2012 at 3:52 PM

In what area has this administration been above board and honest even when it hurt them..?

d1carter on October 5, 2012 at 3:53 PM

for enlightenment on this krap, go here:

https://twitter.com/JimPethokoukis

“114,000 jobs a month is pretty good number…..for 1962″ JP

ziggyville on October 5, 2012 at 3:54 PM

You can’t spell BuLlSh!t without the letters BLS.

bw222 on October 5, 2012 at 3:54 PM

That’s why people who understand data and surveys look skeptically at the result of the household survey. It doesn’t mean a conspiracy is in place; it does strongly suggest that this month’s sample of 60,000 households threw an outlier, especially when compared with the establishment survey and other economic data. If so, it will likely correct itself in the next report. That’s not “trutherism” or denial, but straightforward data analysis.

The key phrase is “the other economic data”, including a lower individual/social insurance tax take in September 2012 than in September 2011.

As for “correcting” itself, the BLS has yet to correct the Wisconsin portion of the establishment survey despite all other measures disagreeing with it.

Steve Eggleston on October 5, 2012 at 3:55 PM

Tie this LIE to the Libya assassination LIE, and barack is going to have a terrible, very bad, not good second debate. I really think these cooked numbers are a tipping point. Finally, the voters understand that their president is a LIAR.

Rational Thought on October 5, 2012 at 3:59 PM

The unemployment rate is ‘cooked’ either way. It way higher than 7%.

TX-96 on October 5, 2012 at 4:00 PM

Yeah, the leftards are having a hard time selling these BS numbers.

THE ECONOMY. IS. BAD.

Jack Welch is coming on Cavuto, FWIW. :-)

Punchenko on October 5, 2012 at 4:00 PM

Scarborough and Geist are right to express skepticism, but it doesn’t have to be a conspiracy theory to say that the numbers don’t make sense.

Really? At this point absolutely nothing is certain and Scarborough and Geist are even leaning towards collusion of government/political entities that resulted in their ‘skepticism’ — and you’re giving the as yet unnamed but clearly intimated government/political entities — cover ?

Here’s another crystal clear option that was — and still is — available to you, ED, and it still doesn’t even violate your ostensible sensibilities and — perhaps — current leanings?

Scarborough and Geist are right to express skepticism, but it doesn’t have to [NOT] be a conspiracy theory to say that the numbers don’t make sense.

What is it with the HotGas pundits and their insatiable desire to cover miscreant POTUS Obama and the miscreant Left (ie: opponents) even in cases such as this where nothing is certain — yet still there’s the cover courtesy of HotAir.com despite the fact that the track record of lies and deceit and self-serving collusion by this administration is as long as your arm and painfully plain to see — and now is the right time for them to play their hand with a stacked deck since soooo many people are hung up on the jobs numbers — and *shocker* lookey there — miraculously unemployment is suddenly under 8% — in OCTOBER — despite the fact that the jobs numbers don’t add up and so far all anyone can offer — even you, ED — is *maybe’s* and *could be’s* to explain it… ?

HotAir.com — Guaranteed Soft & Squishy™ Or Your Money Back

Un-freakin-real.

FlatFoot on October 5, 2012 at 4:03 PM

I don’t like conspiracies either but here is some more analysis:

Zerohedge

Cliff notes: For the first time since they’ve started measuring this data, College kids are not going back to school in September even though college loans are going through the roof and kids are staying in school longer because there are not jobs to have.

Can we get one report to ask anyone in this Administration why that might be?

WisRich on October 5, 2012 at 4:04 PM

Let me remind you of the “household vs. establishment” battle in Wisconsin in May (just prior to the recall election for Scott Walker).

The household survey showed WI losing jobs by the thousands each month. The establishment survey (and the tax revenue) showed thousands of jobs being added. There was quite a battle when Gov. Walker released the establishment numbers early and showed the WI was in economic recovery.

So, where’s the IRS telling us if revenue is up or down? Boehner? Anyone?

The finale? Walker won, of course, but moreover, the household numbers were quietly corrected and showed even more jobs than the establishment survey had identified.

Daisy_WI on October 5, 2012 at 4:04 PM

Ed, Raw Numbers from BLS.gov. From September ’11 through Aug ’12 there were 1,783,000 household survey jobs created. There were 1,692,000 payroll jobs reported. That’s a difference over 11 months of 91K jobs. The difference between houshold and payroll according to today’s report is 759K. If you are going to compare to other large increases you have to use context. What was the household number the three months prior to the 83 spurt? Did the 83 number equal 32% of the total household number for the previous 12 months? This number, for a guy who hated economics with a passion, is unfathomable.

xkaydet65 on October 5, 2012 at 4:05 PM

Is the narrative today “look at barack’s RECOVERY!”? Nope. The narrative today is “Damn, that obama labor department is lying its ass off to save him from a bad debate performance.”

Times, they are a-changin! Teflon barry has become Nixon II.

Rational Thought on October 5, 2012 at 4:06 PM

Daisy_WI on October 5, 2012 at 4:04 PM

Actually, it was the eatablishment survey that did, and despite preliminary “benchmarking” based on the Quarterly Census of Employment and Wages, still does show massive job losses. The QCEW, the household survey, and tax receipts all showed job growth.

Steve Eggleston on October 5, 2012 at 4:10 PM

Democrat desperation reaches Epic Levels! They don’t care how much they have to lie to save the Failure-in-chief! Their credibility? What credibility? The chicago thug machine is just revving up and getting started! The eSTAB-lishment Repubs better get on this or their going to get “Steam Rolled” like they did with McCain 2008!
Final Update for those following:How to take on the Obama Enemy media: http://paratisiusa.blogspot.com/2012/09/an-open-letter-to-those-who-should-know.html?spref=tw

God Bless America!

paratisi on October 5, 2012 at 4:12 PM

Ed, you’re being way too gullible.

OhioCoastie on October 5, 2012 at 4:18 PM

Has this kind of outlier EVER occurred before in our history? Not only is the Household survey of 873,000 the highest number in 29 years, but I am almost certain that when that particular survey has been over 500,000 jobs, the BLS numbers for the most part followed along atleast somewhat. Not this time however. Yes, the household survey is more volatile, but never in our country has it been this volatile I bet.

In other words, what should be looked up is the differential between the two surveys. I bet the 759,000 difference between the 2 surveys is not only the largest in the history of our country, but its the largest by more then a 2-1 margin. I bet the difference has never been more then 400,000 before this great history making day.

What that would prove I believe is that they are cheating. Records are never broken by 2-1 margins. Its like the next guy breaking the HR record by hitting 150 homers. One of these experts with the data in front of them need to see what the differential is between the two numbers historiacally and prove them as cheaters.

KMav on October 5, 2012 at 4:19 PM

xkaydet65 on October 5, 2012 at 4:05 PM

What he said.

KMav on October 5, 2012 at 4:28 PM

I’m confused.

The BLS conducts two surveys each month to determine employment data. The first is the establishment survey, which polls 410,000 businesses each month.
The second is the household survey, which polls 60,000 households each month.
Now, according to the household survey,
873,000 jobs were created last month
.
Very close to one million jobs.

.
Accordin’ to my cipherin’,
that’s 14.55 jobs per household.
Am I missing something?
Math is hard.

mrt721 on October 5, 2012 at 4:31 PM

If someone has already posted this, sorry…but check out the donors in BLS…
http://freebeacon.com/meet-the-obama-donors-at-the-bls/

lovingmyUSA on October 5, 2012 at 4:31 PM

Another month, and another 300+ thousand people disappear from the workforce roles.
How many more of these employment battles can the administration “win” before the entire country is “retired”?

Another Drew on October 5, 2012 at 4:32 PM

“Is the narrative today “look at barack’s RECOVERY!”? Nope. The narrative today is “Damn, that obama labor department is lying its ass off to save him from a bad debate performance.””

Nope, not on any of the national newscasts.

Maybe on Conservative blogs, but that’s it.

gumbyandpokey on October 5, 2012 at 4:38 PM

Another month, and another 300+ thousand people disappear from the workforce roles.
How many more of these employment battles can the administration “win” before the entire country is “retired”?

Another Drew on October 5, 2012 at 4:32 PM

Actually, 418,000 (on a seasonal basis) were added to the labor force last month. Of course, that is still 100,000 fewer than there were in the workforce in June.

Steve Eggleston on October 5, 2012 at 4:38 PM

Accordin’ to my cipherin’,
that’s 14.55 jobs per household.
Am I missing something?
Math is hard.

That’s what confuses me too. How do you arrive at a 800,000+ figure through a 60,000 person survey? My guess would be extrapolation? Or fuzzy math?

tkyang99 on October 5, 2012 at 4:46 PM

The first is the establishment survey, which polls 410,000 businesses each month. The second is the household survey, which polls 60,000 households each month.

The ‘establishment’ ‘survey’ is not a poll per se.

Plus, you may be giving the impression that there is a 410 to 60 ratio here.

You are showing apples and oranges.

The establishment survey covers millions and millions of jobs, in the most of the companies in the country. These are most of the jobs in the country.

The ‘household survey’ is truly a poll and is intended to get work-at-home and farm jobs into the mix.

faraway on October 5, 2012 at 4:46 PM

If someone has already posted this, sorry…but check out the donors in BLS…

http://freebeacon.com/meet-the-obama-donors-at-the-bls/

lovingmyUSA on October 5, 2012 at 4:31 PM

“At least two economists at the Bureau of Labor Statistics (BLS) have contributed to President Barack Obama’s campaign. Harley Frazis of Bethesda, MD, has contributed at least $2,000 to Obama and $9,000 to the Democratic National Convention…”

Wow. 11,000 dollars — ‘at least’ — in total directly to Preezy Obama and the DNC.

But, you know, according to HotAir.com — Guaranteed Soft & Squishy™ Or Your Money Back“Scarborough and Geist are right to express skepticism, but it doesn’t have to be a conspiracy theory to say that the numbers don’t make sense.”

FlatFoot on October 5, 2012 at 4:48 PM

Update: Just to remind everyone, the next report comes out before the election — the Friday before, actually. And if the BLS wanted to cook the numbers, I’m pretty sure they’d have cooked the establishment survey, too, to show more than +114K.

C’mon, Ed. You know better than this.

The whole point is to get the headline in the newspapers and the first story on the nightly news, and give MSNBC (and all the rest) something to repeat ad nauseum until the election.

Everyone will be forced to see “OBAMA GETS UNEMPLOYMENT RATE UNDER 8 %”.

No one will ever see the correction in November: “BLS corrects Oct U-rate to 8.2%” on page 12, and it’ll never be heard on MSNBC.

It doesn’t matter if they cooked one, both, or neither book. Its the headline they wanted. And they got it.

BobMbx on October 5, 2012 at 5:00 PM

No need to look at any numbers.

It’ll be 7.9%

forest on October 3, 2012 at 8:46 AM

Boy, I was wrong about that one.

forest on October 5, 2012 at 5:00 PM

Nope, not on any of the national newscasts.

daveyandgoliath on October 5, 2012 at 4:38 PM

No, on the national newscasts, we have the following “stories” today:

1. Romney was Mean to PBS!

2. Howie “Jet Screamer” Dean says Media was Mean to Uncle Joe Biden!

3. CNN Claims Romney was Lying!

4. NBC Claims Romney was Lying!

F-

Del Dolemonte on October 5, 2012 at 5:00 PM

Just entirely too credulous Mr Morrissey. Especially for a supposed conservative commentariat.

rayra on October 5, 2012 at 5:02 PM

Particularly considering the long-described shenanigans going on with the U6.

rayra on October 5, 2012 at 5:04 PM

The liberal media is going wild. Ape crazy! Loco! This is it! Unemployment down, bad news for Romney. Ahhhh, well. Not!

Allow me to put it into perspective: 1) Because the drop in unemployment was just a measly .3% (That’s three tenths of one percent, not much at all), it’s like Obama’s “forward” going 3 mph on a 55 mile-per-hour freeway. Not fast at all. Almost backward, or at a standstill. At that rate, we’ll have to wait until the second coming for any progress with Obama. This being a monthly calculation over 4 years (that’s 48 month, or at least 46), it is like ACTUALLY .0065 percentage points per month! As Romney said yesterday, THAT’S NOT A REAL RECOVERY. 2) You want a real life example of Obama’s .3% drop in unemployment since he took office? It’s the equivalent of giving Obama an aspirin for his cancer.

The sad truth about this whole thing is that there are millions of people still out of work. The .3% is a joke. To celebrate it is having a heightened sense of idiocy. It may not even exist. Romney says the real unemployment rate under Obama is actually 11%. And that’s educated experience talking.

There are many people across this great country waiting to oust Obama so they can have a piece of the Romney dream; and I count myself among the good people working for a better tomorrow with Romney. But remember Romney has experience; he’s worked to actually create jobs, saved the Olympics in 2000.

Don’t celebrate: the price of gas has doubled with Obama (bad news) and is causing very serious problems for family budgets (how many divorces and business failures is Obama responsible for; not including his own), mortgages are inaccessible, uncertainty and insecurity is everywhere, food prices are up, product size is shrinking (a form of theft), regulation is getting worse under Democrats, foreclosures are still happening (something real jobs creation would cure right away), health care costs are up and going up even more, and real bad news for Obama — people know that the coming taxes on Obama-care will be high — so out with Obama and Obama-care!

As for me, I’m voting for the white guys. Make a hole! Romney and Ryan are coming through. For good. They are simply white as the driven snow. You gotta love it.

AdrianS on October 5, 2012 at 5:08 PM

It’s more likely to happen with smaller samples than larger samples, but can happen any time in surveys.

Want to bet that those households surveyed were all in North Dakota?

GarandFan on October 5, 2012 at 5:09 PM

Let’s look at it this way.

These are Obama’s numbers.

Congress doesn’t audit them. Ernst and Young doesn’t audit them. Nobody audits them.

I’m sure that Bernie Madoff’s clients thought his numbers were beyond reproach as well. Remember, Bernie was a former NASDAQ Chairman.

faraway on October 5, 2012 at 5:13 PM

The first is the establishment survey, which polls 410,000 businesses each month. The second is the household survey, which polls 60,000 households each month.

The ‘establishment’ ‘survey’ is not a poll per se.

Plus, you may be giving the impression that there is a 410 to 60 ratio here.

You are showing apples and oranges.

The establishment survey covers millions and millions of jobs, in the most of the companies in the country. These are most of the jobs in the country.

The ‘household survey’ is truly a poll and is intended to get work-at-home and farm jobs into the mix.

faraway on October 5, 2012 at 4:46 PM

Actually, it’s approximately 486,000 worksites with 141,000 distinct employers that is covered by the CES. Of note, that 486,000 worksites is somewhere between 3.5% and 4% of all worksites. If the percentages of surveyed sites are the same as they were in March 2010, approximately 8.5 million employees are covered by said survey as the BLS oversamples large establishments due to the ease of surveying them versus smaller establishments.

And yes, the CES is a survey, complete with a monthly root-mean-square error of +/-75,800 on total non-farm jobs.

Steve Eggleston on October 5, 2012 at 5:14 PM

Ed:

To understand this you should really look at the data more closely. The 22-24 year old data is THE outlier, particularly seasonally adjusted/part-time jobs.

DevilsPrinciple on October 5, 2012 at 5:16 PM

New conspiracy: Two economists at the BLS donated to Obama. Link is on Drudge.

stukinIL4now on October 5, 2012 at 5:31 PM

Update: Just to remind everyone, the next report comes out before the election — the Friday before, actually. And if the BLS wanted to cook the numbers, I’m pretty sure they’d have cooked the establishment survey, too, to show more than +114K.

Ed

Sorry to go all Truther on you, Ed, but the household survey would certainly seem to be the easier of the two data sets to cook, and isn’t it a strange coincidence that

- Obama gets (further) exposed as a clown on Wednesday night,
- the last job report before the weekend of the election comes out today, and
- for the first time in 4 years, the unemployment number is below 8%, based on a ludicrous “survey of households” by the Democrats which claims that 875,000 people went to work this month but only 114,000 new private sector jobs were reported.

Child, please. Nobody is a big enough sucker to believe this tripe coming from the Chicago Crook Democrats.

Don’t gnash your teeth, people…just laugh out loud at the Dems. They are so corrupt it has become a national joke.

Jaibones on October 5, 2012 at 5:35 PM

Let’s look at this yet another way…

Obama created 873,000 jobs with a poll. A poll.

Can anyone tell me where even 100,000 of these jobs are? What industry? What company?

A poll.

faraway on October 5, 2012 at 5:36 PM

That’s what confuses me too. How do you arrive at a 800,000+ figure through a 60,000 person survey? My guess would be extrapolation? Or fuzzy math?

tkyang99 on October 5, 2012 at 4:46 PM

Statistics, man.

60000/350000000 = 1.71E-4 (or .000171) % is the sample size. From this, the economic policy of a mutli-trillion dollar economy is decided.

Its a fantasy, but since we have a fiat currency, it doesn’t matter.

BobMbx on October 5, 2012 at 5:40 PM

Minor point of order to the conspiracy theorists – while the BLS has been cooking the establishment survey in Wisconsin for over a year and a half to deny Gov. Scott Walker the initial ability to claim added jobs, presuming that the household survey is truly random, there is a 1-in-40 chance that precisely this level of error versus Gallup’s estimate of an 8.1% unemployment rate will occur naturally.

Of course, if the household survey is not truly random, all bets are off.

Steve Eggleston on October 5, 2012 at 5:44 PM

Is there any way to see the details of the household survey. Not the A schedules.

faraway on October 5, 2012 at 5:47 PM

Is there any way to see the details of the household survey. Not the A schedules.

faraway on October 5, 2012 at 5:47 PM

What level of detail? There is significant detail from the CPS not included in the press release in the Labor Force Statistics database

Steve Eggleston on October 5, 2012 at 5:50 PM

Now, I fully understand what Obama means by jobs ‘created’

faraway on October 5, 2012 at 6:00 PM

Then you have to ‘save’ the file.

faraway on October 5, 2012 at 6:01 PM

What a scam. There isn’t anything this outlaw administration won’t do. They will tell any lie and disregard all laws.

Philly on October 5, 2012 at 6:05 PM

I rad somewhere that Yahoo News had a poll up — do you believe this jobs report? — and 57% said they were phony, which is why Zero’s labor secretary was zoomed out. Disaster!

Rational Thought on October 5, 2012 at 6:08 PM

What level of detail? There is significant detail from the CPS not included in the press release in the Labor Force Statistics database

Wow, this site is actually quite nicely done.

At least we know the BLS hires pretty good web programmers. Dunno about their statisticians though.

tkyang99 on October 5, 2012 at 6:08 PM

LegendHasIt on October 5, 2012 at 3:20 PM

I heard 8.2 was projected as well. Yahoo Finance mobile had a story available all day from Bloomberg that cited that figure. It’s all a game to the White House, apparently.

Philly on October 5, 2012 at 6:12 PM

The unemployment rate in PA and in the county I live in went up this month, so no one here is seeing an improvement in joblessness.

Liar-in-Chief.

Philly on October 5, 2012 at 6:16 PM

I’m expecting the labor dept to announce the unemployment rate to be 4% next month and the treasury dept to announce that the Federal Government is expecting a $1trillion surplus this year. When your audience are idiots, you can say anything you want.

cajunpatriot on October 5, 2012 at 6:20 PM

Just entirely too credulous Mr Morrissey. Especially for a supposed conservative commentariat.

rayra on October 5, 2012 at 5:02 PM

Really? SMH

http://live.wsj.com/video/the-math-behind-jack-welch-jobs-tweet/B0CAE191-5E3B-4513-A4FE-C1C117856324.html#!B0CAE191-5E3B-4513-A4FE-C1C117856324

Can.I.be.in.the.middle on October 5, 2012 at 6:28 PM

First things first, in February 2009, there were 141.748 million Americans employed in the United States and the labour force participation rate was 65.6%. In September 2012, 140.025 million Americans were employed and the labour force participation rate had fallen to 63.6%. Between February 2009 and September 2012, the civilian non-institutional population grew by 8.859 million adults; yet, 1.723 million fewer Americans were working. In the last year alone, 1.1 million Americans left the work force.

Today’s expected — given the massacre in Denver and next month’s election — “unexpected” drop in the unemployment rate requires a willing suspension of disbelief. The “spontaneous uprising” in hiring of a whole 114,000 jobs allegedly caused a drop in the U-3 rate of 0.3%. In August, the US economy added 96,000 and the unemployment rate went from 8.3% to 8.1%. In July, the US economy added 163,000 jobs and the unemployment rate went up to 8.3% from 8.2% in June when only 80,000 jobs were created. Go figure.

If the labour force participation rate was constant, the U-3 rate would be 10.7%.

By the way, at a rate of adding 114,000 jobs per month, we wouldn’t return to full employment until 2025. 114,000 jobs per month would be great…if we were living in the late 1950s. We need twice that many just to maintain current employment. A robustly growing economy requires job creation in the range of 300,000+. Take a look at the number of jobs that were being created during the Reagan Recovery when GDP was 7-8%. As a reminder, GDP for Q2 of 2012 was revised down to 1.2%.

U-6, which counts part-timers, who want full-time work, and the discouraged, is 14.7% and has remained unchanged. If the underemployed (people working a jobs below their skill-sets) are included, the U-6 hits the high teens or low twenties. All combined, there are about 23 million Americans unemployed or underemployed.

The September Jobs Report: The Big Lie

Resist We Much on October 5, 2012 at 6:30 PM

There’s already more than 63,000 posts in a single day on the Yahoo thread for this topic…looks like there’s quite a few people who have a problem with this “report”…

tkyang99 on October 5, 2012 at 6:39 PM

So the biggest outlier in 29 years just happens to be reported 1 month before the election. Just a coincidence. Yea, sure, that’s the ticket!

humdinger on October 5, 2012 at 6:43 PM

The fact that BLS consists of mostly career employees gives me very little comfort. With numbers like these I suspect the methodology, the sampling techniques, the interrogatories, the final calculations and even the motives of the BLS. This number is so far askew and a complete independent analysis of the entire process is absolutely necessary. Nothing is beyond the Obama regime and we must get to the bottom of this issue.

rplat on October 5, 2012 at 6:49 PM

They don’t even have the decency to tell us somewhat believable lies. These figures are on par with the “Yeas have it” at the Democrat National Convention.

claudius on October 5, 2012 at 6:59 PM

Important point to hammer home to everyone listening: The unemployment rate is 7.8% 3 years after the recession ended. That’s not a good number. Neither is 114,000. Let’s not overreact and give the impression that it is.

xblade on October 5, 2012 at 7:03 PM

I have always considered the number statistical noise, the issue is the media and Obama, himself, using this single number as the be all end all of the economic situation. We are part of the small society of wonky people who get into the other unemployment stats and understand what is going on with the jobs report.

Obama is going after the “low information” voter with reckless abandon and this number plays right into it. Id hate to see this election won by the guy who relied on ignorance. Its worse than that though, he is fostering as much ignorance as possible….its astounding.

If taken as a whole this jobs report stinks, but in the hands of a snake oil con artist its gold.

On a side note, does anyone else get a George Costanza vibe from Obama today. He is out there with comebacks to Romney 2 days late….its like when George showed up in Ohio to give his comeback to a guy that busted him months before….”Oh Yeah, well the jerk store called….theyre running out of you!”

What a complete tool.

alecj on October 5, 2012 at 7:05 PM

The thing is with 410,000 businesses being surveyed, the enterprising soul could, without too much problem, do a random check that would support or disprove a huge jump in the data. With 60,000 private households out of 150 million, where would someone even begin? So, the numbers could be fudged on the household data as long as one had an absolute contempt for the truth and the American people —and there was something SIGNIFICANT at stake.

claudius on October 5, 2012 at 7:39 PM

The fact that BLS consists of mostly career employees gives me very little comfort. With numbers like these I suspect the methodology, the sampling techniques, the interrogatories, the final calculations and even the motives of the BLS. This number is so far askew and a complete independent analysis of the entire process is absolutely necessary. Nothing is beyond the Obama regime and we must get to the bottom of this issue.

rplat on October 5, 2012 at 6:49 PM

Yep. For one thing, a President Romney will probably freeze hiring and the federal salaries scale. Federal employees are grossly overpaid, and they could freeze salaries for four years and just start to let private industry catch up.

slickwillie2001 on October 5, 2012 at 7:47 PM

When do they post the spplications for DISABILITY?

Oh, I’m sorry, is that racist?

stenwin77 on October 5, 2012 at 7:53 PM

Let’s assume the BLS isn’t all BS.

Are we supposed to be happy about an unemployment rate of 7.8%? that’s like being happy when the mugger stops punching you in the face & starts kidney punching you, you’re still being mugged & beat up.

batterup on October 5, 2012 at 8:46 PM

So, now a large number of people will be coming off food stamps and unemployment insurance, right? Right?

claudius on October 5, 2012 at 8:58 PM

I disagree with Ed and frankly am amazed at the pass the BLS is getting here. When was the last time this happened? 29yrs ago!!

What are the chances this gives Obama reason to bray right before an election.

NONE.

The books got cooked. Solis is a radical in her own way.

From my blog.

http://truthandcommonsense.com/2012/10/05/jobs-numbers-cooked-up-by-an-ideologue-over-at-bls-shows-a-huge-dip-in-unemployment-when-there-is-no-proof-of-it/

Remember, the ends justifies the means in a liberal’s mind. A radical liberal is even worse. Also, remember who these people love- Mao. And for years and years and years China lied about how well things were going in their nation- while people starved to death.

Here, The BLS obviously cooked the books. Far too many on TV are trying to give it the benefit of the doubt saying it is fairly insulated. What does “fairly” mean in a world where the IRS just passed a rule ignoring Obamacare and the legislation passed by Congress, where Obama signs an executive order end running Congress on immigration AND the Labor people are trying to get companies not to warn employees of pending layoffs – which is also against the law. Not only are they asking, no demanding, no WARN letters go out, they are willing to pay for any lawsuits that result in an employee getting laid off, without notice, and suing. (Seriously, you can’t do that. I mean not as a suggestion, but as “it’s against the law” kind of you can’t do that.) Not to mention the obvious coverup in Libya to protect Obama from a foreign policy failure scandal.

They have LIED about everything! How is it this one gets a pass and a outlier error? Especially when every economist that looked at the numbers predicted 8.1. And that the last time anything like this happened, as Hotair points out, growth was 9!

But I guess we’ve learned now that to these radicals in charge the law is something to be used to your advantage and ignored when convenient.

So who really thinks now that Solis over at labor, a HUGE radical in her own right, didn’t cook the books.

Seriously?

archer52 on October 5, 2012 at 9:13 PM

Update: Just to remind everyone, the next report comes out before the election — the Friday before, actually. And if the BLS wanted to cook the numbers, I’m pretty sure they’d have cooked the establishment survey, too, to show more than +114K.

That’s what the plan is for the *next* report – two ‘solid’ reports putting unemployment below 8% before the election.

Midas on October 5, 2012 at 9:23 PM

Cooking the books or incompetence ether way it,s wrong.I will say it again if the people at the BLS were honest they would know you can not have this many jobs created with a GDP of 1.5% and have the rate drop .3%.With all the red flags that should have been raised. They should have done the household report again.This is not just a mistake it,s criminal and heads should roll.A lot of people will have already voted by the next report and if the jobless rate jumps back up to 8.2% then these people at the BLS stand guilty of vote tampering.Then what?Gov. Palin warned Hannity wed night .Watch out for something like this and Rush warned all of us about this very thing almost a year ago.

logman1 on October 5, 2012 at 9:30 PM

The key phrase is “the other economic data”, including a lower individual/social insurance tax take in September 2012 than in September 2011.

Steve Eggleston on October 5, 2012 at 3:55 PM

Steve, where do you get the social insurance tax revenue numbers?

I’m most interested in the Medicare tax revenues, since the Social Security tax stops at certain incomes, but the Medicare tax does not.

I’m interested in taking that Medicare tax revenue number, dividing it by the Medicare tax rate to get the total income taxed, and dividing it by the the monthly civilian noninstitutional population age 16+ (available from table A-1).

That would show average per capita income for the civilian noninstitutional population age 16+, and I’d like to see the trend line of that number.

ITguy on October 5, 2012 at 10:00 PM

Something doesn’t seem quite right. In all the years between 1980 and 2011 numbers of employment have gone down for the 20-24 year old age group. Now in 2012 suddenly they gain 101,000 jobs.

http://www.zerohedge.com/sites/default/files/images/user5/imageroot/2012/09-2/Septembet%20Change%20NSA%2020-24.jpg

That’s amazing, really amazing!

claudius on October 5, 2012 at 10:02 PM

I should have said that was for September. An edit feature would be really handy.

claudius on October 5, 2012 at 10:06 PM

So, did the Obama camapaign hire 101,000 college students to work on it’s campaign? Pay them each $10? It would only cost $1M to buy a lower unemployment rate…

Just sayin’.

ITguy on October 5, 2012 at 10:32 PM

Steve, where do you get the social insurance tax revenue numbers?

I’m most interested in the Medicare tax revenues, since the Social Security tax stops at certain incomes, but the Medicare tax does not.

I’m interested in taking that Medicare tax revenue number, dividing it by the Medicare tax rate to get the total income taxed, and dividing it by the the monthly civilian noninstitutional population age 16+ (available from table A-1).

That would show average per capita income for the civilian noninstitutional population age 16+, and I’d like to see the trend line of that number.

ITguy on October 5, 2012 at 10:00 PM

I pulled them off the last-day-of-the-month Daily Treasury Statements. Unfortunately, the DTS does not break down the Medicare numbers.

I haven’t seen a monthly breakdown of Medicare funding activity like there is for Social Security, but perhaps that is just because I haven’t had the urge to look as hard.

The Monthly Treasury Statement also doesn’t directly provide what you’re looking for, but at least one can estimate the Medicare portion of FICA/SECA by subtracting the interest earned by the Hospital Fund (end of Table 5) from the income of the Hospital Fund in Table 8. In fact, I use that as a first-look at the Social Security “Trust Funds”. The bad news is the September MTS will be delayed some (it usually is out the 8th business afternoon) as it coverss the end of the fiscal year.

Steve Eggleston on October 5, 2012 at 10:50 PM

So somehow in September, in addition to all the other discrepancies in the labor report, we have one more to add: that of the Schrodinger Student: one who is both in college and piling up student loans on one hand, yet on the other hand entering the work force in the month of September, a time when historically every single month in recorded history has seen an exit from the labor force for the 20-24 year old cohort.

Someone else pointed out that the household survey is a survey and the anwer changes depending on how the question is asked.

Could the BLS have changed the question to include “student” as a form of employment? You’re studying at home. Isn’t studying work? If you go from not counting students as employed to counting students as employed merely by virtue of their student status-not actual jobs and take into account everyone going back to school this time of year, you could fudge an 800,000 increase in “employment”

talkingpoints on October 5, 2012 at 11:03 PM

It certainly seems an anomaly, or outlier, and the roots may be in the way the Household Survey is conducted. People are NEVER asked directly if they are employed or working. But if they report mowing lawns or selling off their Star Trek Action Figure collection on eBay this month, they are counted as employed. Part time or single day spent working counts the same as permanent full time, there is no distinction made.

But that’s the way it’s always been conducted. The purpose is to pick up the entrepreneurs and home workers who cannot be found by the business survey. That doesn’t explain the results, which typically are roughly in line with the relation between the market economists projections and the business report. IOW, if the business report is underestimated, usually the household is too, but not always. A big difference, a completely different story, is rare.

Anyone believe we really just set a 29-year record for jobs created in a month with 1.3% growth and only days after Bernanke announced QEternity precisely BECAUSE the jobs market is so dangerously weak?

Adjoran on October 6, 2012 at 1:04 AM

Nice try to deflect Ed, but I ain’t buyin what you’re sellin!!

This almost unheard of drop in unemployement rate, at this exact time, is a coincidental as a “spontaneous” attack on the Bengazi consulent office on 9/11. You give way more credit to this corrupt, lying, cheating, vile administration than they deserve!! Imagine the headlines if a Republican was in office right now. Imagine the outrage and the accusations coming from the left.

This whole thing was cooked!! It’s as obvious as the HUGE wart on my grandma’s nose. Problem is, it’s TOO obvious. Lots of people are sitting back right now and saying, “They can’t be THAT stupid, can they?!”. Well, yes they can.

wirebitersmith on October 6, 2012 at 1:48 AM

I’m noticing a couple of odd reactions to today’s jobless data and the analysis it has produced so far today. On one hand, we have people proclaiming the difference between the household and establishment data as proof of a conspiracy to re-elect Barack Obama at the BLS. On the other, we have a number of voices calling anyone who applies any kind of critical analysis to these figures as “truthers” or “deniers.” Tom Elia has collected some of the latter at The New Editor, but I’ll focus on explaining what exactly the figures are, and why critical analysis does not amount to “trutherism.”

Ed,

Did you try calling somebody at BLS to get some information on what may have caused the anomaly? It may have been just some sort of methodological issue. By calling and speaking with a knowledgeable analyst at BLS, they might be able to shed some light on the key factors that contributed to unusual behavior that was observed in this month’s report. Once you get into the weeds with numbers, you start to realize that all sorts of odd things can happen… In all likelihood, there is probably an innocent explanation for this month’s report, even though it looks peculiar.

As an independent statistical agency of the government, BLS operates with a good deal of autonomy from the political appointees. I find it highly unlikely that a career civil servant would risk his livelihood or retirement to help some politician.

ghostwriter on October 6, 2012 at 2:50 AM

I disagree with Ed and frankly am amazed at the pass the BLS is getting here. When was the last time this happened? 29yrs ago!!

What are the chances this gives Obama reason to bray right before an election.

Oy. No offense, but concocting a conspiracy over this is a bit over the top. What really has changed here? What was the unemployment rate yesterday? Eight percent. What is the unemployment rate today? Eight percent.

If Obama wants to run on 7.8 percent compared with 8.1 percent unemployment as being some great movement in the right direction, then good luck to him with getting anybody to buy into that. I welcome that debate.

ghostwriter on October 6, 2012 at 2:58 AM

I don’t know if anybody has seen this article, but the explanation seems plausible:

Haugen explained that due to the particular data set in question — that governing the unemployment statistics — it wouldn’t take a multi-bureau government conspiracy in order to produce a 0.3 percentage point shift.

He said that part of the reason for the relatively large drop in unemployment this month has to do with the sample size of the current population survey, also known as the household survey. Because the sample size is smaller than the payroll survey, for example, there’s more variability in the numbers.

“On a month-to-month basis, the household survey employment measure is more variable than the payroll employment measure due to the smaller household survey sample,” the BLS said in a press release Friday attached to the jobs report. “Over longer periods, the changes in household and payroll survey employment tend to track more closely.”

“The smaller the sample size, the higher the variants,” Haugen said. “If you’re trying to really track the monthly developments in employment you need to focus mostly on data from the payroll survey.”

Haugen explained that one factor that likely led to the 0.3 percentage point drop in this month’s unemployment is the fact that people in the 20-24 age group (including college students and people who are often working temporary summer jobs) left the job market this summer earlier than expected.

“In August you had an unusually large decline in employment” among that demographic, he said. Because the BLS does seasonal adjustments for its data, it was primed for a big decline in September, when young people have traditionally left the work force. “What happened was there was a big decline in August and not a decline in September. Because there was no decline [when it was expected] there’s a big increase after seasonal adjustment.”

“In the [monthly household survey] you occasionally get these large movements because it’s based on a relatively small survey,” he added. “What we tell people is that you need to wait for additional months of data in order to see what the pattern is.”

http://www.cbsnews.com/8301-250_162-57526845/on-jobs-numbers-bls-vows-theres-no-conspiracy/

ghostwriter on October 6, 2012 at 3:13 AM

It’s pretty sad that CBS tried to pass this off as a “favorable” jobs report, and suggesting that that this helps the president. The unemployment rate has been over 8 percent for over 40 months–the problems here go a heck of a lot deeper than one jobs report. This is like a student pulling failing grades in a class, and trying to beg his way into a C-, based on getting a D on his last exam.

ghostwriter on October 6, 2012 at 3:17 AM

Sadly, most of the jamooks(Steve Eggleston being the chief offender) here have NO clue about the BLS, statistical modeling or HOW they arrive at the final numbers. What the Labor dept did essentially is to compare apples to oranges in the household and business survey.

THE REAL clue to whether or not UE has changed is in the actual amount of money that the government has collected from payroll taxes.

Having said that,I am confident that no one will be surprised to learn that foodstamp usage for both persons and households, has jumped to a new all time record…. Finally, and putting it all into perspective, since December 2007, or the start of the Great Depression ver 2.0, the number of jobs lost is 4.5 million, while those added to foodstamps and disability rolls, has increased by a unprecedented 21 million.

Ed: Please stop attempting to play economist. You’re out of your element.

DevilsPrinciple on October 6, 2012 at 7:47 AM

DevilsPrinciple on October 6, 2012 at 7:47 AM

Really, you idiot? You must have been blind and deaf when I said that, for the first time this year, individual income/social insurance taxes are down from the same month last year.

Steve Eggleston on October 6, 2012 at 8:58 AM

Thanks to DevilsPrinciple, I am out of here.

I hope the trolls are happy.

Steve Eggleston on October 6, 2012 at 8:59 AM

2 points – 144,000 jobs in a month was good for 1962

If government can solve all problems, why not have everyone work for the government?

MN J on October 6, 2012 at 9:34 AM

Thanks to DevilsPrinciple, I am out of here.

I hope the trolls are happy.

Steve Eggleston on October 6, 2012 at 8:59 AM

LOL….awwwwww…this is a maneuver that even Obama could love. Such tender feelings ’round here………hahahahahahahahahahahahah….

DevilsPrinciple on October 6, 2012 at 10:03 AM

DevilsPrinciple on October 6, 2012 at 7:47 AM

Really, you idiot Super Nice Guy? You must have been blind and deaf when I said that, for the first time this year, individual income/social insurance taxes are down from the same month last year.

Steve Eggleston on October 6, 2012 at 8:58 AM

LOL..so much for thoughtful analysis…..it’s ok, Steve..I accept your apology….’mon back…

Steve, how about YOY analysis…it clearly shows a declining trend since 2008.

DevilsPrinciple on October 6, 2012 at 10:28 AM

Wll I tried debunking the numbers on two threads, but here I’ll offer a weird explanation for the increase. Why the jump in part time jobs? Well who takes a part time job? Someone who has no other option, like someone whose UI has run out. Find the numbers on how many UIs have run ou in Aug and Sept and you’ll have a reason for the part ime jump.

xkaydet65 on October 6, 2012 at 11:26 AM

Here is an article from none other than the NYT, going on about the anomaly in the numbers for 20-24 year olds. What jumps out at me is the steady, very rapid decrease in the number of 20-24 year olds leaving the labor market according to the chart from 1948-2012.

The last time there was a change like that it was during the boom times from around 1995-2001, which I remember well. The problem is that 2009-2012 have not been boom times at all.

http://economix.blogs.nytimes.com/2012/10/05/explaining-the-big-gain-in-job-getters/

claudius on October 6, 2012 at 11:35 AM

That should have been not leaving the labor market Aug-Sep. An edit feature please.

claudius on October 6, 2012 at 11:38 AM

Comment pages: 1 2 3