Under(lying) Today's GDP Revisions (Or: "The Church of the Third Revelation")

Marla Singer's picture

We probably can't get more relevant commentary on today's absolutely massive downward GDP revision than that penned today by Goldman Sachs' Edward F. McKelvey:

This was a much larger than normal revision for the third pass on a given quarter, knocking what once was a fairly robust 3.5% bounce down to a mediocre 2.2% (from 2.8% prior to this revision). All sectors except the trade balance -- a focal point of last month's  downgrade -- saw some downward revision. Revisions were particularly deep in business investment -- to -5.9% from -4.1%, worth two tenths of the revision --  and in inventories (also worth nearly two tenths).

That would be those sectors that should, one would think, be among the easiest to count in the first place, especially on the second try.

Goldman is being very kind here.  Goldman started off the day with this from GS Breakfast Bytes:

GDP for Q3 (third estimate)... not much change, though risks lie to the downside.  GS: +2.8%; median forecast (of 73): +2.8%, ranging from +2.5% to +3.7%; last (Q3 second estimate) +2.8%.  The third cut on a given quarter does not usually produce much of a  change in the growth estimate.  In this  case, the risks against our assumption of no change lie to the downside, reflecting large downward revisions to construction spending for August and September.  Like many others, we do not  see a meaningful probability of changes in the +0.5% estimate for the GDP price  index or the +1.3% figure for the core PCE price index.  (Emphasis ours).

It is difficult to get a more direct sense for how much mind share government bailouts (and government figures) have managed to command.  That even the most pessimistic member of the "consensus" (n=73) managed to overshoot the mark by nearly 14% and the average sailed full speed into a 27% pop-up should remind us of three things:

  1. The Bureau of Economic Analysis of the United States Department of Commerce has ceased to be (if it ever was) a reliable outlet for economic data.  (Be this the result of misfeasance or malfeasance depends on the reader's propensity to credit conspiracy theory).
  2. That what passes for the professional prognosticator class these days is pathologically incapable of realistic appraisal.
  3. That the largest single expression of a Keynesian "injection" in the history of Keynesians or injections (or the planet) struggled to create even the most anemic growth.  Net the double counting of stimulus funds it seems difficult to imagine even a remotely encouraging (or positive) "growth" figure could be tortured out of the economic realities that would be so plain if one but looked out the window to forecast them.

If it isn't clear to everyone by this time that the United States remains firmly in the grips of a massive "shadow recession," then we can only credit this ignorance with some unshakable and deeply rooted form of denial or a seriously reckless case of willful blindness.  Either way, we would like to commend the current powers that be for their tour de force performance in establishing themselves firmly both as the most masterful of bullshit artists to occupy the beltway (and that's saying something) and simply the most economically inept (and expensive) team ever to hold national office.  The risk adjusted returns on this particular ruling clique would make the pre-dollarization Zimbabwe carry trade look attractive.

If there is a silver lining to be found here it is probably that this little experiment might finally drive a splintered wooden stake through the heart of John Maynard Keynes' ghost (or at least irradiate Paul Krugman's ravings to within 50 rads of his professional life).

Comment viewing options

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

Feeling like your insane when you're the only person who recognizes the truth is the cruel fate of common sense.


Thank god for ZH or I'd have committed myself by now.

Oso's picture

Do you ever wonder what it would have been like if you had taken the blue pill instead of the red?  I do.  all the time.... I'd have far less grey hair, thats for sure.


I just laughed when that revised print popped up.  what a joke.

exi1ed0ne's picture



Cypher was the only one that actually made a free choice in the whole movie.


besodemuerte's picture

I disagree completely.

So you're saying everyone that followed the path out of the Matrix and into reality didn't make free choices?  Are all of us here on ZH passively going through the motions or did we choose this for ourselves?

If anything, I would call Cypher a big fat vagina that couldn't handle the deeply penetrating phallic truth and sought an easy way out.

VegasBD's picture

Yea but god damn did you see how good that steak looked he was eating?

Cursive's picture


"phallic truth"

Now that's a new one.

Anonymous's picture

It's always blown my mind how men use one of the strongest and most powerful human body parts to denote weakness. Ever been in a delivery room?

besodemuerte's picture

It's been awhile since I've seen a news story regarding a vagina that victimized someone.  Don't get me wrong, I see your point where perhaps you missed mine.  The heart is an incredible body part as well, but you don't see it going around raping people do you? 

The truth hurts, as does an uninvited dick (so it seems).  Let's not make it more complicated than it needs to be.

Crime of the Century's picture

Considering that the blue pill was half the cost - yeah, it gets tough sometimes...


aint no fortunate son's picture

Not to worry, by 4:00 today, with the Dow up 90 on no volume the MSM will be singing the praise of the ongoing "strong economic rebound" (AP) and "broad based recovery" (Reuters)... the pundits on CNBC have already unanimously declared victory by stating that this Q's revs downward mean the "good stuff" will be pushed out to Q4, so look out for upgrades to the experts' Q4 GDP estimates any day now.

The fact that those Q4 revisions will in turn also be revised downward dramatically won't matter, since we are in this perpetual positive feedback loop where all news is good, always was, always will be, and the beautiful scent of fresh green ink pervades the land.

Anonymous's picture

yup u got it
its 4.15
recap off the day:
higher home sales and S&P high for the year
nothing on 40% GDP revision

AngryVoter's picture

Yes it is true.  Insanity can be caused by an acute sense of reality.

besodemuerte's picture

Fucking fantastic post Marla.

I'm probably going to Wal-Mart after work today to buy some more ammo.

Dollar is no good, PMs are no good, there is no escape.  Got guns?

Anonymouse's picture

Your Walmart has ammo?  Mine has been sold out since May.  I was able to get it once this year only and that was because I happened to hear when the shipment came in.  I bought half of what they got that night before it even hit the shelf.

besodemuerte's picture

Yes and it's quite affordable to boot.  I bought three boxes, then drove over to Dick's and bought a box of HPs.  I spoke with the Wal-Mart salesman for awhile, he says their handgun ammo is lucky to last a day or two and that they never know when shipments are going to arrive.

GoldmanSux's picture

Very well said MS.

Entrepreneur's picture

3.5% - 2.2% = 1.3%; 1.3%/2.2% = 59%.

The BEA overstated Q3 GDP by 59%.  And counting.

If that's the going accuracy standard, what's the point of releasing anything?

I wonder if BEA considers accuracy a "core competency"?

Anonymous's picture

That's a 41% improvment over (whenever). You and Marla need to "get with the plan."

Quantum Noise's picture

3.5% - 2.2% = 1.3%; 1.3%/2.2% = 59%.

The BEA overstated Q3 GDP by 59%.  And counting.


Your math is an abomination.

Reductio ad Absurdum's picture

No, the math is correct.

2.2 = "true" GDP growth
3.5 = "inflated" GDP growth

2.2 * (1+x) = 3.5


Hence, originally published GDP growth was a 59.09% inflation of the "true" GDP growth.

Quantum Noise's picture

Read again what he wrote. "underestimated Q3 GDP by 59%"...  not GDP growth, but GDP, in nominal terms.

Entrepreneur's picture

Quant -


If you're going to appoint yourself 'Grammar Sheriff', at least quote me correctly.  I said, "overstated", not "underestimated".

To be precise, I should have included the word "growth" after "GDP".  But given that I showed my math (using all growth numbers), I'll bet most people picked that up.

Finally, my math is correct. BEA overstated GDP (growth) by 59% and counting. My oversight was one of grammar, not arithmetic.

Now hurry off before you miss Mad Money :)

crosey's picture

So Q3 is flat when you factor in 2.2% govt stimulus.  Doesn't feel like a corner being turned.

Reporting here from main street, it's bleaker, and I see more frustration and exhaustion.  People are worn out.

Assetman's picture

Can't wait to see the +5.0% Real GDP number get reported for the 4Q... so we can see it revised to +2.5%.

This actually makes more sense than the employment numbers the government reports.

Ivanovich's picture

Market obviously loves it.  Range breakout in the works here!

El Hosel's picture

  Range breakouts require big volume and wide price spread. The market is on vacation.

Ivanovich's picture

Ok, so by your reasoning, since volume has been pitiful most of the year, we're still in the same range since March? :)

El Hosel's picture

No reasoning, the fact is The SPY first hit 1100 on OCT 19th, its has gone nowhere since. Its in a narrow range, looking very tired and very weak.

Dixie Normous's picture

Retail index.  New High today. Nough said.

Dixie Normous's picture

I'm not the brightest bulb in the chandelier, but when I look at the huge revisions areas like imports, exports and durable goods, I have to believe that the gov't revision to 2.2% is still way to high.


Cognitive Dissonance's picture

I can't wait to hear what John Williams of Shadowstats.com has to say about this latest revision. Poor John is running out of adjectives to describe the ongoing farce.

cougar_w's picture

Yeah, need to borrow a page of Dilbert and invent new ones:

I'll go first. These are all "something bad" adjectives:

Pooptastic. Fecktardid. Clusturded. Pollutistic.

Maybe a few good adjectives, but these are still bad:

Joynihilist. Endtastic. Fundangerous. Excellenteric.

Okay, that should be enough for next week anyway. Use them in good health.


Cognitive Dissonance's picture


Nothing like a good hearty Tuesday afternoon laugh. I'm going to pass these on to John and beg him to use one of them in his next report.

WaterWings's picture

Just in time to insult the inlaws!

Oh, well that sounds a little fecktardid if you ask me.



TumblingDice's picture

Maybe the BEA's job will be easier for next quarter if they just say that the GDP was fundagerous with a bit of fecktardid data. We'll know what they mean and they won't have to make shit up.

minimoose's picture

Kevin Phillips had a piece called "Numbers racket: Why the economy is worse than we know" published in The Atlantic back in May of 2008:


And certainly all of the problems Phillips identifies are much worse now than they were then.  Caveat venditor and caveat emptor.


Crime of the Century's picture

And people still have the temerity to mock Shadowstats' Williams, whose uber-bearish forecast is in the numbers...

Danz Gambit's picture

Thanks for posting that article, excellent piece.


"Second is the Gross Domestic Product, which in itself represents something of a fudge: federal economists used the Gross National Product until 1991, when rising U.S. international debt costs made the narrower GDP assessment more palatable. The GDP has been subject to many further fiddles, the most manipulatable of which are the adjustments made for the presumed starting up and ending of businesses (the “birth/death of businesses” equation) and the amounts that the Bureau of Economic Analysis “imputes” to nationwide personal income data (known as phantom income boosters, or imputations; for example, the imputed income from living in one’s own home, or the benefit one receives from a free checking account, or the value of employer-paid health- and life-insurance premiums). During 2007, believe it or not, imputed income accounted for some 15 percent of GDP. John Williams, the economic statistician, is briskly contemptuous of GDP numbers over the past quarter century. “Upward growth biases built into GDP modeling since the early 1980s have rendered this important series nearly worthless,” he wrote in 2004."

Crime of the Century's picture

The silver lining is the current availability of silver for lining. Regarding Keynes, if they dug up and burned Wycliffe's bones for the "sin" of translating the Gospels from the Vulgate, how much more should Keynes' miserable frame be violated? Let us jettison his remains into space so that he might no longer infect the academy.

Anonymous's picture

Bullshit artists? These assholes are the Michaelangelos and Monet's of Bullshit. Throw in a little David Copperfield and David Blaine in there, too...

Anonymous's picture

It was entertaining watching Steve LIESman on cnbc this morning trying to put lipstick on this pig. I wonder what that weasel will be saying when the s_ _ t finally hits the fan...

Anonymouse's picture

Easy.  "Manure is the best fertilizer.  After it hits the fan, it is spread around the economy spurring growth in all segments.  What data are you looking at?  Are you an idiot?  This is great news!"


That should be pretty close to it.

Anonymous's picture

as much as i hate the bls, bea, and every other lying sack of shit in the usa government, we must recognize that the gdp comes with a sampling error....all of the revisions were within sampling error...thus to allege incompetence is without merit....

of course that does not mean that the bls is not a sock puppet for the liar-in-chief occupying the oval office.

cougar_w's picture

[gdp comes with a sampling error]

Just about everyone here understands this implicitly. The gripe would be that:

A) the revisions appear subjectively to fall outside what everyone is assuming to be the sampling error, and

B) if the sampling error really is wide enough to embrace these kinds of revisions then their methodology sucks wind and the reports are useless on the face of it, and

C) they always seem to revise downwards and not upwards long after the fact, suggesting not a random variance around the mean "truth" but a systemic over reporting of numbers to the up side. We must assume this is done for political purposes as no other purpose appears to apply. The politicization of what ought to be an objective exercise is itself pure evil.

So sure, you are right. But you are still wrong.  :)


Anonymous's picture

no sir....i am not wrong....your assumptions A & B
lack substance....per my reading of john williams'
explanation the sampling error is +/- 3.5% on gdp....
thus every single estimate within tolerance is
correct but also of very limited value. any revision
within sampling error is also statistically correct.

any frustration with the numbers is without
justification....any number within tolerance could
actually be contraction or expansion of gdp....
we simply do not know and nearly all of the
supplied change numbers are correct....

most people read gdp numbers as solid single
reliable numbers and then moan and groan about
variances - such is without statistical merit....

even if systematically revised downward the
changes are still within sampling error and in
fact bolsters the claim that they were initially

Anonymous's picture

If you post a GDP of 3.5% you are saying that it could actually be +/- 3.5%, wouldn't this be a sampling error of 100% in either direction?

Or is it say 3.5% GDP with +/- 3.5% which would be a range from 3.38% to 3.62%?

If it is the first example then what is the point of sampling if you could be wrong by 100%?

Could you imagine an election poll where say a candidate was was predicted to have 50% of the vote with a sampling error of +/- 50 points.

Anonymous's picture

Amen, Cougar, a big AMEN!

Anonymous's picture

In a country that cares only about certitudes that can be proved by numbers, as God and all other certitudes have been thrown out as so much b.s.....

That these numbers could so far off....

Maybe we should question numbering itself, as just another dated philosophical construct....

Now Im in way too deep.