Seasonal Adjustment Pushes Initial Claims Below Expectations At Least Until Next Week's Revision

Tyler Durden's picture

That this week's consensus "beat" of 375K initial claims will be revised to a miss next week is irrelevant - all that maters in a job election year is to fudge the numbers. Which is why the fact that "only" 372K initial claims were filed in the last week of 2011, even as thousands of bankers were being laid off, is all that matters. Of course, last week's revision was as always higher, from 381k to 387k, which means that next week, the beat of consensus will become a miss, but by then who will care - there will be another fake and soon-to-be-revised number to fixate on and push futures even higher from fair value. And confirming that it is all in the seasonal adjustment, is the observation that while SA claims improved by 15K, Not Seasonally Adjusted they increased by +37,423 hitting 535,112. In other words a 3K job difference to consensus is due to a statistical smoothing adjustment based on BLS data integrity.

Non-Seasonally adjusted claims:

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
hugovanderbubble's picture

Bullish for Permabullish troops,


Reality goes one way , permabullish world goes other.


Sorry Main Street I cant do more than im doing to fix this scandal.

mayhem_korner's picture



It's what you walk down to get from the front gate to Cinderella's castle.

Manthong's picture

"BLS data integrity" - sound as a watertight compartment..

on the Titanic.

YBNguy's picture

Welcome home troops, youre all fired! -Bamster

slaughterer's picture

Whoever calculates Baltic Dry must hate the ADP office.  Baltic Dry just posted an 8.1% drop.  These data releases are like spy-vs-spy combats, with resulting instantaneous manic-depressive market swings.  No wonder retail is fleeing (even if usually at the bottom).   

mayhem_korner's picture



Failed Christmas rally gives rise to manufactured Ephiphany rally.  The full story at 11.

Pelosis Usless Brain's picture

The unadjusted 4-week moving average is over 508k. It's all in the seasonal adjustment.

Can I seasonal adjust my income tax?

EscapeKey's picture

That's only for government bureaucrats. They are above you, peasant. Now, lick their boots.

Personally, much like the GDP one, I'd like to add a personal salary deflator to my income, it's only fair as the CPI on the products I buy have been significantly above the bullshit numbers we get out of the BLS.

eurusdog's picture

Pretty soon they will just quit revising them.

Irish66's picture

1-14-11 high 549,000 non-seasonal for year?

535,000 today

RobotTrader's picture

ES now outpeforming gold.


Scared money now fleeing PM's and seeking refuge in common stocks.

mayhem_korner's picture

ES now outpeforming gold.


Happy pre-collapse, Robo.  Looks like your 15-minute headline machine is working fine.

Wixard's picture

I dunno my eyes see money for stocks means money for pms. At least on anything but a 30 millisecond hft window.

Teamtc321's picture


Family Ten Dollar TDO is right where you need to be. 




SheepDog-One's picture

'ES now outperforming gold'....what, on the 1 minute chart?

Certainly NOT on the 10 year chart. ES = 0, gold = 8X banger.

Cdad's picture

'ES now outperforming gold'....what, on the 1 minute chart?

LOL!  Well said, Dog.

GMadScientist's picture

Oh dear.

It has its head stuck in the bucket again.


mckee's picture

Ha! ... you said "BLS data integrity."

WonderDawg's picture

There will come a time when they've lost all credibility, not yet but it's coming, so even when they start telling the truth, if they ever do, no one will believe them. The boy who cried wolf comes to mind.

ElvisDog's picture

I don't know. Everyone around me, including so-called experts like those on business podcasts such as Motley Fool, take the government employment statistics at face value. I see very little skepticism on the numbers being fed to the masses.

lolmao500's picture

This is so bullshit....

orangedrinkandchips's picture

99 weeks....that is it.

This reminds me of Gresham's law...."shit" money replaces "good" money....You are giving up 100k+ jobs for min. wage seasonal xmas help....


Of course ADP will print a good number....Bed Bath and Beyond(me) hires for xmas help....min. wage.



fonzanoon's picture

what ever happened with the debt ceiling? I thought we were going to hit it the other day?

EscapeKey's picture

I bet that's better than expected news.

SheepDog-One's picture

OH nevermind all that....we got good data! Sure, everyone already knows its fake, but so what! Look at futures GO!

lolmao500's picture

Anyone have the highest number for non-adjusted claims since 2000?? Didn't we reach 700k+ in 2008?

lolmao500's picture

The adjusted number was around 670k...

youngman's picture

Its an election year...don´t trust the numbers...the only one I believe is the Food Stamp number..when that starts dropping then we might have something going on...

Bold Eagle's picture

LOL, just take a look at the breakdown by states. Only in GA the number of claims decreased by about 1k. On the other hand, 19 states witnessed increases from 1k to 16k. 

WonderDawg's picture

Were those the same states that "estimated" their numbers?

GMadScientist's picture

Little more than froth from the number of people who can't file anymore (Thanks, Newt!) not being evenly distributed.

Bold Eagle's picture

Even the total unadjusted number is fudged. To me it looks like the number should be up by at least 52k versus reported 37k.

SheepDog-One's picture

'All that matters in _____ year'....fill in the blank with 'Presidents honeymoon first year', 'All-important presidents second year', 'Vital 3rd year', and finally 'Crucial election 4th year'....

So EVERY year markets cant show reality.

pepperspray's picture

then the lame duckage arrives

monopoly's picture

This is getting so old, week after week. And still, so few Americans know that the train they are on has no destination.

mayhem_korner's picture



I beg to differ, the train has a's just that the 'bridge out ahead' signs have been painted over.

SheepDog-One's picture

Even if there were giant day-glo orange signs saying 'DISASTER AHEAD!' with screaming sirens on them as well, would americans pay any attention? No.

SheepDog-One's picture

Well americans and america deserves what its getting. 

ElvisDog's picture

Yes, but what are they getting? Seems like it's working out very well so far in that the can just keeps getting kicked down the road and nothing really challenges the American status-quo. Let's be honest, up to this point the doomers have been consistently wrong about the timing of the collapse.

GMadScientist's picture

Sure it does. This here tag says I'm going to "Jimmy Dean"!

GMadScientist's picture does closing govvie datacenters fit into the jobs picture?

And is that a net win (reduction of government) or just more jobs pain from the TOTUS?


marcusfenix's picture

so what are they seasonally adjusting exactly? it can't be part time holiday lay offs because those jobs don't last six months and people can't claim ui off them right?

I've never understood the relevance of the SA number, how can this 375k possibly be accurate when in fact over half a million people filed new claims? not to mention it makes zero sense, at least to me, that the SA number can drop while the NSA number surges, one would think that there should be some correlation between the two shouldn't there? 

another thing I've noticed is that the SA data is used year round so what season are they adjusting for, all of them? at some point don't you have to use the raw NSA data to make up for all the "adjustments" in order to get an honest accurate picture of where unemployment really is? and the fact that the adjustments are constantly being adjusted upwards makes me wonder why anybody would take these numbers seriously. 

what am I missing?

Pelosis Usless Brain's picture

Check out John Williams Shadow Government Statistics site. He explains a lot of what goes on with government data. The guy knows his stuff.

EscapeKey's picture

Generally speaking, you run a filter on 12 months worth of data to get a more accurate view. Will smoothen out any freak weeks.

Of course, how the weights are applied is a debate in itself, and these weights can obviously be adjusted on a weekly basis as well.

Basically, it's all made up on the spot.

chinaguy's picture

With the Labor Participation Rate at record lows (since women entered the work force in the 1980's) There ain't no net jobs being created & it's as fucking simple as that.

Tsar Pointless's picture

I wonder if I was one of those who was magically disappeared from an initial claim filing for last week?

If so, does that mean I don't receive benefits? Or, do I receive more, kind of like "hush" money?

I feel so...bullish.

Alex Kintner's picture

IBM is a sterling example of how companies are Gutting the US Workforce. They continue to acquire US companies every year, then offshore as many of the jobs as possible. And also us the H1B and Lx visa programs to bring in low paid foreigners. And they have stopped reporting annual US workforce numbers to hide the activity.

Good article here.