More Fun With (Quietly Revised) Numbers
Last week I pointed out a somewhat mysterious, undated entry (turned out to be Nov. 24, 2009) at the website of the “official” arbiters of recession, the NBER.
Rooting around the web, and finding myself at the site of the Census Bureau, I caught the following cryptic note at the page where Durable Goods data is released:
The Census Bureau identified a processing error that occurred when revising historic seasonally adjusted data for the November (data month) releases. The data have been corrected. As of January 15, 2010, revised data from January 2001 through October 2009 are available in the Historical Timeseries – NAICS tables, under the Historical Data tab.
Being skeptical and curious, I wondered what this “error” implied for the most recent release of Durable Goods (because we’re always all about the latest green shoot, right?).
Here we go again.
It struck me as a bit unusual that the data had been corrected as of January 15 — not coinciding with a scheduled Durable Goods release, which actually comes this Thursday, the 28th. Seemed to be a slip-it-in-there-while-no-one’s-looking type of thing. (And no, I’m really not a conspiracy theorist.)
But back to the November data: You may recall that consensus for November’s Durable Goods had been +0.5%. The reported data was lighter than expected at +0.2%. Looking at the revisions the Census Bureau has now incorporated into the data, we see that November actually printed at -0.7%.
Here are the “Before” and “After” charts:
Before
After Revision
Admittedly October was bumped up a bit. But we now are looking at two consecutive months of declines, which we’ve not seen since December 2008 (-3.7%) and January 2009 (-7.9%), when the world was falling apart. And the trend seems to have deteriorated, with November now being the worst print since August. What might the real-time impact have been had the Census Bureau initially reported a disappointing -0.7% instead of the +0.2% they maintained until very recently?
Consensus for Thursday is around +1.5% – +2.0%. Stay tuned.
Tweet
Facebook
Reddit
Digg this!
January 24th, 2010 at 5:45 pm
congratulations BR – you’re catching on- finally starting to understand what us “crazier” folk know-
a full court press has been going on to keep the “illusion that is the stock market” from collapsing- as it should (and will)-
question- why is that so important- that the illusion be preserved?
another question- why are people- who must depend on their savings for retirement and security- investing in the riskiest class of investments anyway?
you tell me- and don’t give me the lamest of excuses- that the stocks always outperform-
stocks- in the scope of things- are a recent phenomena
January 24th, 2010 at 7:37 pm
http://news.yahoo.com/s/ap/us_white_house_finding_its_way
Subtext: for people in doubt of Hopium please use more Cannibis
January 24th, 2010 at 8:18 pm
And the trend seems to have deteriorated, with November now being the worst print since August.
If I’m reading the revised chart correctly, the moonshot in July is now the entire year’s growth. Or it would be if the January contraction wasn’t larger.
(I KNEW there was an (physical) output peak in August. I could feel it, couldn’t figure out why. Finally dawned on me that I’ve lived close enough to significant rail to absorb the sound – lots and lots of train noise at night means the economy is growing. I could hear trains all night ’round August, a little less in September. January last year was piddly. January this year is dead so far. Showed up in CR’s traffic graph as well, when I went looking for it. And retail isn’t stocking the shelves above bare minimum, period. Traffic looks slowly fading in that graph.)
What might the real-time impact have been had the Census Bureau initially reported a disappointing -0.7% instead of the +0.2% they maintained until very recently?
They would’ve invented a story about snow or something. Only cumulative impact would make a difference. (Aside from the noise trading.) They did manage to push things off until after Christmas though, didn’t they?
Well, the Rubinites are the ones that really started the aggressive data massaging. (The Reaganites made the one time adjustments, mostly, as far as I can tell.) The Bushies just ramped it up to ridiculous levels. So I wouldn’t be surprised at uh, productivity improvements.
max
['Rarely is the question asked, is our econometricisicians learning? Well yes, actually, they are, so don't worry your pretty little heads about it.']
January 24th, 2010 at 8:55 pm
Actually, this is my favorite mainstream wall street blog, but still, sometimes you have to wonder. Even here, some level of naiveté creeps in… oh wonder of wonders, real evidence of evil doers. I also read Karl Denninger at market ticker, and he’s got a real doozy (really wild) tonight. Seems those guardians of financial destiny at the Fed and Treasury done pulled a fast one on us all… intentionally crashing the market to get what they wanted. Gee, IF they did that, what else would THEY do?
January 24th, 2010 at 9:04 pm
aka “hide the salami”. Different denotation, the connotation is the same.
January 24th, 2010 at 9:10 pm
Confirm Bernanke or Wall St. Could Blow Up…WTF?
by Badabing
Sun Jan 24, 2010 at 05:29:21 PM PST
http://www.dailykos.com/story/2010/1/24/829765/-Confirm-Bernanke-or-Wall-St.-Could-Blow-Up…WTF
Excuse me, but what the fuck does that even mean? Let’s break it down, shall we? Ok – Gibbs is saying, essentially that if Bernanke is not confirmed there will be: ‘repercussions.’
January 24th, 2010 at 9:13 pm
Hold on there, spunky.
There are a BUNCH of bars that are “different” — not just November.
And seriously different, too. Look at 2008.
What kind of “processing error” is this? How far back does this “error” go?
January 24th, 2010 at 9:14 pm
Ya think that’s bad. I’m having to watch a blog of the Saints/Vikings game on nfl.com because my wife is watching the French Concert for Haiti on TV5.
http://www.nfl.com/gamecenter/2010012401/2009/POST20/vikings@saints
January 24th, 2010 at 9:16 pm
@km4:
Just finished watching the clip on Denninger myself. This hostage-taking shit has got to stop.
January 24th, 2010 at 9:23 pm
So this is a little silly and your point “worst two months since Q4 2008″ is sensationalist (at best).
The data series seems very violatile (so small adjustments dont mean much… even monthly reading dont mean much).
Further the “two negative months” are composed of a -.1 and a (new) -.7. This is very different from Q4 2008.
In a data series that ranges from +6% to -6% your discussion of -.5% in revisions (<5% of the range) need to be seen as very very small.
January 24th, 2010 at 9:27 pm
The durable orders release measures the dollar volume of orders, shipments, and unfilled orders of durable goods (defined as goods whose intended lifespan is three years or more). Orders are considered a leading indicator of manufacturing activity, and the market often moves on this report despite the volatility and large revisions that make it a less than perfect indicator.
from http://briefing.com/Investor/Public/Calendars/EconomicReleases/durord.htm
That’s why, asshat.
January 24th, 2010 at 9:38 pm
[...] sure you’ve received many contacts about this. Armed with many facts, charts and figures, I’ve decided rather to simply state the obvious: Your constituents are paying very close [...]
January 24th, 2010 at 9:38 pm
Keep hounding the numbers Barry; keep ‘em honest.
I see you’ve got a new bulltard here. I’ll go away again now. Bye
January 24th, 2010 at 9:39 pm
This isn’t conspiracy; it is simple politics.
Worker “A” calls his boss and says a revision was missed. Boss asks if it is postive or negative. Worker “A” states it is negative. Boss calls his boss and reports a negative revision is needed. Boss of Boss calls White House and tells White House aide a negative revision is needed. Aide tells Chief of Staff. Chief of Staff tells the President. White House then picks a date and method for revelation. Revision is done in a footnote with no fanfare and in the middle of a reporting period.
This is how real trickle down economics works.
January 24th, 2010 at 9:53 pm
Mike in NOLA, how are your nerves holding up? :-)
January 24th, 2010 at 10:03 pm
In addition to the points made above (except for Cognos), this revision will allow pundits to trumpet what will probably be the “better than expected” jump in Durable Goods Orders “compared to last month.”
Isn’t it the pits that we have to speak in quotes when we talk about govt doublespeak?
January 24th, 2010 at 10:44 pm
It looks like May and July 09 were both revised UP about as much or more than Nov was revised down.
Sounds like a bunch of people hear are looking for data to fit the fact that they missed the recovery and missed that 1-yr ago was a great time to be taking risk.
From the looks of the stores and restaurants this weekend, recovery continues to grow.
January 24th, 2010 at 11:20 pm
cognos-
if you accidentally tripped into a wood chipper . . . well . . .
I could live with that I guess
January 24th, 2010 at 11:24 pm
I would say there’s a growing consensus that gov numbers are not to be trusted.
Well duh.
Blame it on the republicans? Don’t think so.
Blame it on the best Democrat to ever be a Republican president: Clinton.
Who presided over cooking the books on the CPI. So he had more of our cash to spend on …..
Downhill from here dudes.
TASS is going to look pretty damn good as objective reporting.
January 25th, 2010 at 12:17 am
Transor:
She finally went to bed just before the two minute warning so I got to see the end. Was pretty hairy there. They seem to have found the secret of the good teams: looking lucky and somehow pulling it out, e.g. the immaculate reception.
My daughter emailed from NOLA and said that after the game she could hear fireworks or possibly shooting out in the street. My guess is shooting: easier to find guns than fireworks :)
January 25th, 2010 at 3:02 am
[...] The Big Picture: More fun with (quietly revised) numbers [...]
January 25th, 2010 at 4:00 am
Back in the Sovjet Union, when the goals of the five-year plan weren’t fullfilled, the goals where revised downwards. This way the plan was always excelled.
I think the NBER has a much better approach here. Rather than lowering your goals, you ‘accidently’ revise the numbers upwards. And when nobody cares anymore you give the real numbers. I mean the Sovjet Union survived more then 70 years, so it shouldn’t be a problem the bankers to have a few more good years this way.
January 25th, 2010 at 7:25 am
the bohemian Says:
January 24th, 2010 at 5:45 pm
question- why is that so important- that the illusion be preserved?
another question- why are people- who must depend on their savings for retirement and security- investing in the riskiest class of investments anyway?
Reply:
—————–
Wall street doesn’t earn account management fees if the rubes don’t stay invested. It’s that simple. When sell siders go on CNBC, it’s not to perform a public service or give an economics lesson. It’s to sell product. CNBC is often nothing more than an infomercial for sell siders. But, in fairness, they don’t seem as bad now as they did as recently as a couple of months ago. Maybe Comcast told them to shut up and pretend to be newsies for once?
If you can get/keep grandma or grandpa in the market, then you get to eat another month. Management fees add up. Altruism is not a factor. It’s all about getting you to buy the spin.
Re the stats above; they look like the net change is more or less -0-. However, in a market where bad news doesn’t count and good sews is used to move product, it will probable be a good change.
January 25th, 2010 at 8:12 am
I hear ya hobo – macro picture – its like American Cap’ism thought it was gonna pull a super victory out of its hat by outsourcing all those polluting factories overseas (where the up&coming cheap labor was) and we could make a real living pushing files and propaganda around until the new labor force started buying into the Manhattan Island Economy and all we all had to do for a living is cash those dividend checks
January 25th, 2010 at 10:14 am
Coincidence that this story only printed after yours? I think not.
http://imarketnews.com/node/7594
January 25th, 2010 at 11:38 am
I am also “skeptical and curious”.
I looked into the December Unemployment statistics, and found some interesting tidbits.
Besides December being the largest “decline” in the December numbers going back to 1948,
The 2009 ANNUAL labor force number –
“Well, comparing 2009 with other yearly reductions in Labor Force Levels, 2009 sets the record in a “Big” way.
The closet annual “change” (notice I didn’t say drop) in any years Labor Force Level, going back to 1976 compared to 2009′s DECLINE of 1,528,000 workers was an INCREASE of only 522,000 workers.
In other words, 2009 labor force reductions levels have exceeded any prior year going back to 1976, by more than 2 Million discouraged workers.
I discovered that, the population increase in the U.S. was a very similar increase when compared to the preceding 30 years. The U.S. added around 1.9 Million people in 2009. The ANNUAL net 12 month change in the “Not in the Labor Force” number was the highest on record, going back to 1969, 3,417,000.
January 25th, 2010 at 6:07 pm
I just joined; you all are great.
But it does seem to me like we should all take a collection and buy Mike in Nola a second television, perhaps? We have four in our household: me, the hubby and the two cats. We won’t be adding to that population explosion the former marine mentions, no sir.