Actual GDP: ~0%

Don’t pay attention to the statistically aberrant GDP number reported Friday.

Its actually much worse.

Thanks to a "statistical fluke," GDP was actually less than 1%:

"Last quarter’s annualized 26 percent increase in auto production shocked Joe Carson, now director of economic research at AllianceBernstein LP in New York. Without the gain, the economy would have grown at an annual rate of 0.9 percent, not the 1.6 percent the Commerce Department reported today.

The increase in output came despite cutbacks announced by General Motors Corp., Ford Motor Co. and others. A drop in the wholesale price of SUVs and light trucks as the automakers cleared leftover 2006 models made production look stronger than it actually was, said Carson. The economic fallout from the auto-industry cutbacks will instead come this quarter, he said."

What brought this aberrational data point about? The number crunchers in the Commerce Department rely on wholesale prices for light trucks. The 5.5% decrease in SUVs had the effect of making output look stronger — but only when adjusted for inflation. In reality, these were firesales to move product off the lot.

Consider this related tidbit:  AutoNation, the largest dealership and sellers of new cars in the country, said it expects to cut 2007 purchases from the big 3 by 30%.

And its not just cars: WalMart reported its weakest monthly sales gains since December 2000 — despite the 30 decrease in gasoline prices since the summer. In the US, Sales rose just 0.5% — significantly below the 2-to-4% improvement originally forecast for October.

Bottom line: If we have to torture the data to get to just 1.6% GDP, imagine what is actually going on in the economy.

This is the primary reason we have "obsessed" so much on Real Estate. Without the massive contribution of residential housing to the overall economy — from job creation to transactional business to MEW — there simply is not a whole lot of growth to be found elsewhere.

Other noteworthy items in the Q3 GDP report:  There was an unexpectedly large accumulation of inventories. This implies manufacturing output will slow further in Q4, as manufacturers trim production this quarter to reduce the increased inventory build up.

Also of note:  The GDP Deflator came in at 1.8%, significantly below the expected 2.8%. When it appears we have less inflation, then output looks greater. The deflator has not been below 2% since Q2 2003. In case you were wondering, the deflator is not ex-energy.

For the econ-wonks out there, the Technical notes are always rich with intrigue, and this report is no different:

For many of the key series used to prepare the advance estimate of GDP, including retail sales, unit automobile and truck sales and inventories, manufacturers’ shipments of nondefense capital goods (other than aircraft), manufacturers’ inventories of durable goods, federal defense spending, and consumer, producer, and international price indexes, actual data are available for all months of the quarter.   

For the key series shown in this table, actual data for the third month of the quarter usually are not available in time for inclusion in the advance GDP estimate. BEA makes assumptions for the source data that are not yet available; assumptions for September 2006 are shown in the last column of the table.  For most series shown, the data for August are preliminary and subject to further revision. Occasionally, the data for earlier months are also subject to revision."                        

This assumptions/estimates is what typically happens for the first GDP release (remember, the number will get reported 3 times: this Advance GDP, Prelim, and then Final). What this could mean is that by the time the subsequent revisions are completed , we could be showing zero growth.

That’s right, 0% GDP is in the range of possible Q3 outcomes (I suspect we will be in 0.5%-1% range)


The slow-motion slow-down is now in full swing . . .


UPDATE 2 October 30, 2006 10:24pm

More questions on the deflator; Maybe this will help — Have a look at the past 5 quarters:


  Q3 ’05   Q4 ’05    Q1 ’06   Q2 ’06   Q3 ’06
3.3 3.3 3.3 3.3 1.6

Something looks a bit funny in that series . . .


UPDATE October 30, 2006 10:10am

I keep getting emails asking me "What is the GDP Deflator, and why does it matter?"

The short answer is that it is an economic metric that converts output
measured at current prices into constant-dollar GDP. The idea is to measure output, not inflation.

For the longer example, lets use an example: Freedonia’s biggest product is widgets. In Q1, they sold $100 worth of widgets ; In Q2, they sold $110 worth of widgets, or $10 more.

The trick in calculating Freedonia’s GDP is figuring out how much of that  $10 increase is due to increased output, and how much is due to inflation. To oversimplfy, the deflator "deflates" the price increases out of the total, so all that is left is GDP gains. It is the "implicit price deflator for GDP." (Does that clarify this?)


8:30 A.M. EDT, FRIDAY, OCTOBER 27, 2006

U.S. Statistical Fluke Exaggerated Growth, Will Be Reversed
Carlos Torres
Bloomberg, October 27, 2006 14:41 EDT

AutoNation to reduce orders 30%
DETROIT FREE PRESS, October 28, 2006

Wal-Mart Posts Softest Same-Store Sales Since ’00
WSJ, October 30, 2006; Page A3

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. JGarcia commented on Oct 30

    Very interesting work Barry. I’ll be watching market tells that might confirm your conclusions.

    -commercial paper market

    -risk spreads


    -employment (near full)

    -cap ex

    -loan quality


  2. V L commented on Oct 30

    No matter how hard I have been trying, I just simply cannot understand how BEA economists and statisticians did not question their alleged 26% jump in motor vehicle production estimates at times when the entire auto sector is in a recession.
    Did anybody at BEA say wait a minute these estimates do not make any sense?

  3. my1ambition commented on Oct 30

    You mentioned Wal-Mart’s sales. I remember thinking to myself that if there were a recession, would people buy cheaper clothes or just buy better clothes less often?

    I asked around (mostly higher-middle class) and most answered that they would buy less often, not cheaper.

    I think that Wal-Mart’s earnings answered both my questions.

  4. S commented on Oct 30

    On Wednesday, I listened to the earnings call of a small, specialty chip maker (ticker: SMDI). They lowered 4Q guidance, citing mid October order cancellations from two OEMs in the mobile wireless space. The head scratching part is that this segment of SMDI’s business normally sees a meaningful seasonal uptick in 4Q, making the mid October cancellations surprising. I’ve heard about a couple of other companies with exposure to mobile wireless also talk about a sudden softening in demand, so it is not company specific to SMDI.

    What is really spooky to me is that it coincides with the well publicized call out of Goldman on Friday that motherboard demand is “falling off a cliff”. So, two major tech sub-sectors, mobile wireless and PCs, have suddenly and abruptly taken a nose-dive.

    It is not inconceivable for me to think 4Q GDP growth will be negative. The holiday selling season may prevent it, if the decline in gasoline prices and heating oil stimulate the overly stretched consumer.

    How they can continue to hold this market up is amazing.

  5. alexd commented on Oct 30

    There seem to be several layers of difficulty. Some of the questions are:

    What questions do we need answers for?

    What is the criteria to decide if the information we are recieving is accurate,complete within the context of our question and complete in it’s entirety.

    What is the source of this information? Is this source consistant?

    The basic idea is are we having the wool pulled over our eyes? Are we pulling the wool over our own eyes?

    When we know the quality of the information we can then see how the market particpants react to the available information. Then we can try to evaulaute what we think are the likelihoods on the short term, med term, and long term of the market movement and reaction to changes in the perception of the truth.

    The rock star Prince said he wanted to party like it was 1999. So did a lot of people who were heavily invested in the Nasdaq, but unless you were short the party was over by 2001. Prince ‘s song seems to be analogous to the irrational euphoria present in the markets at that time. The nimble made fortunes off that euphoria but they were either very lucky or adept at dancing on thin ice. I believe Prince got his royalties on the recording before the market crashed.

    The market is an anticapatory mechanism until it isn’t

    So when the revisions are released what will be the likly market reactions? How will I react?

  6. Leisa commented on Oct 30

    Kyocera reported terrific numbers today and increased their outlook…an interesting divergence.

  7. tjofpa commented on Oct 30

    If you’re a Globalist and a Monetarist, then a more accurate estimate of the “real” inflation rate might be;

    Charles Dumas of Lombard Street Research Ltd. in London calculates that money supply in the world’s top economies is growing at an annual rate of 7.5 percent. Though down from a four-year high of 9 percent a year ago, “that’s still pretty lavish,” he says.

    Just what is a Liquidity Trap?

    China had 118 million tons of excess steel production capacity last year, more than the entire 112 million-ton output of Japan, the world’s second-largest steelmaker. There is also surplus capacity in areas including autos, cement and coking coal.
    “Excess capacity is so large in some industries that their ability to service their debt is in question,” threatening to push some banks into insolvency, Nicholas Lardy, a fellow at the Institute for International Economics in Washington, wrote in a report this month.

  8. Farmer commented on Oct 30

    Will it be a run for the exits today? ok blink, curly, blackie, brownie and flyaway.

  9. Bob_in_ma commented on Oct 30

    I think rising inventories are the beginning of positive feedback from the housing slump: MEWs fall, spoending contracts. There was a video clip on Barron’s with Fred Hickey, editor of High-Tech Strategist newsletter, who sees a slowing consumer in tech sales and rising inventories and attributes it to a housing slowdown. It’s not surprising that big, flat-screen TVs were often bought with MEW money.

    tjofpa, could moderate inflation, at a time of sharply rising money supply, mean we are basically in a deflationary period?

  10. paul commented on Oct 30

    Normally, I’m as cynical as anyone about govt data, but I’m unclear here on whether there’s manipulation or whether the measurement procedure here yield an anomalous reading for this quarter that will be worked out in the longer time series.

    So, has the procedure or assumptions for counting changed for this 3rd quarter? (This would help support an argument that it is being tortured.)

    Or, does the consistently used procedure create a misleading headline, or at least a ‘surprise’ for the non-critical thinkers as Q4 gets added in to create a GDP figure for all 2006? (Automaker firesales to reduce inventory are common so look for that whipsaw contraction in Q4 that creates a more reasonable annual picture.)

  11. Vega commented on Oct 30

    KBH receives default notice on TWO bond issues: the 7.25s due 2018 and the 6.25s due 2015. Bloomberg’s DDIS function shows close to $800mm of the 2015s out there and $320mm of the 2018s. Those dollar amounts do not foot with the $1.65B the news wires say KBH is on the hook for. Regardless, this has the trappings to be a serious credit event that wakes up the credit market. Cash flow on the builders SUCKS (that’s technically speaking, of cours). If the short end of the treasury curve spikes a little more that’ll be terrible for sentiment and for the marginal borrowers.

  12. wcw commented on Oct 30

    KBH is flat-to-up so far today, and the homebuilders generally are flat-to-down.

    One quick note in reply to JGarcia: if you think employment is “near full” you’re looking at the wrong tells. Track total wages and hours and follow employment rate per civilian, noninstitutional population. The data have been okay, but hardly reflect near-full anything.

  13. spencer commented on Oct 30

    An important point to remember when looking at what info in the quarterly gdp number is estimated and what is actually known is that all the numbers about final demand — consumption, housing, government spending, investment — are the average of the three months data. But the inventory and trade numbers in the gdp report is the change from the start of the quarter to the end of the quarter and these are the two series that we have no data at all for the thrid month of the quarter. so these are the two data series where the biggest revisions are likely to stem from.

  14. spencer commented on Oct 30

    The big and important difference between the cpi and the deflator is that the cpi is a fixed basket of goods while the basket of goods in the deflator changes to reflect changes in actual consumption.

    If because the price of oil goes up you consume less gas this will not be reflected in the cpi but it will be in the deflator which because you consume less gas will give gasoline a smaller weight.

  15. tjofpa commented on Oct 30

    Absolutely. Depending upon which assets you’re looking at. Debt laden asset classes in sectors where there is overcapacity are at severe risk of deflation.

    Need to repay => need for revenue => too much competition => lower prices => need for more sales => lower prices again…

  16. Paul Jones commented on Oct 30

    To Spencer:

    Do you think that they would cook the books for political purposes? After all the lying they did re human life in the Iraq war, why wouldn’t we put it past them in something so banal as dollars and cents? I have a hard time with a chain deflator of 1.8%.

  17. jaymay commented on Oct 30

    Is the deflator usually highest in the 4th quarter because of the holidays? I.e., people are less discretionary about their spending and therefore spend more in inflated areas of retail?

  18. S commented on Oct 30


    The DDIS function only considers the public debt, not the bank debt. The bank agreement will contain cross default language with the public debt, so an event that triggers a default on the public debt applies to the bank debt.

    It’s only a “technical” default. The bondholders and banks will hold their hand out for a big fat fee, then agree to waive the default.

  19. anon commented on Oct 30

    Even worse, this miserable GDP growth occurs before the decline in housing starts has filtered through to housing completions and construction layoffs. Plus Q3 MEW was still at a “healthy”, albiet lower, pace. With plunging home prices MEW is going to fall of a cliff and the stimulant effect to the GDP with it.

  20. Vega commented on Oct 30

    S, thank you very much for the detail on DDIS. I’m an equity options guy, so I have a lot to learn about credit and debt. Thanks again.

  21. Michael C. commented on Oct 30

    Tony Crescenzi at RM hypothesizes this:

    “Many are puzzled by the apparent lack of benefit that Wal-Mart (WMT) has seen from the recent decline in the price of gasoline. I believe that one of the many explanations is because of the underground economy, something I have discussed for quite some time.

    What I am referring to in particular is that part of the economy that previously benefited from the strength of the housing market, in which many workers work “off the books,” particularly immigrant workers, who are employed by contractors of various sorts.

    …So, whereas the economy surprised on the upside because of gains seen by the underground economy during the housing market’s upturn, the economy now has greater potential to surprise on the downside now that the underground economy is being negatively affected by the housing market’s downturn. ”

  22. tjofpa commented on Oct 30

    Black market economies …

    Yeah, but they’re great for the productivity #’s.

  23. brion commented on Oct 30

    That’s an interesting take Michael C.-
    Obrador (mayor of Mexico City/Pres. candidate) wrote a book in which he stated that HALF of mexico’s economy was “off the books”….
    No-one really knows how many illegal immigrants there are in the u.s. but they are presumably folks who couldn’t even make it in the underground mexican economy….Black market economies are a devil’s bargain imo

  24. calmo commented on Oct 30

    Expanding Paul’s note above:
    I wonder how many people looked at that auto line on this BEA release and noticed the pattern in q3 for the past few years. Consider the 2005 entries for instance when GDP was a heady 4.9 in q3 followed by a miserable 1.8 in q4. The auto entry in q3 was nearly as high as the current one (22) accounting for 1.3 of that 4.9 (according to the BEA). The data makes some sense to me: this: getting rid of last year’s models at deep discounts that bring sales forward. This is apparent in the following quarter with negative entries , yes?
    Next quarter may not be 3% lower than the current 1.6% with lower crude prices but autos are likely to take a percent or two off and, of course, residential investment not done boring its hole in this number.

  25. BDG123 commented on Oct 30

    While Wal-mart’s numbers were indeed awful and may be a proxy for retail and the economy, there are other factors at work including many acknowledged by Wal-mart itself.

    Wal-mart’s massive simultaneous store upgrades have affected sales as have the reasons for the upgrades. ie, Merchandise mix and the actual construction itself is part of the problem. Wal-mart is trying to recreate their brand and move upscale to Target demographics Wal-mart so jealously covets by carrying more expensive, trendy merchandise they’d hope would increase SS sales as well as total spend.

    ie, Shopper profile of $75K family income versus less than $50K at Wal-mart. So, are Target shoppers ever going to shop at Wal-mart or is Wal-mart only alienating its core client demographic? That is likely playing as big of a role as anything in the reported numbers. That is why Wal-mart is halting makeovers. They aren’t working. Ex, I know many people who shop at Target but would never, under any circumstances, shop at Wal-mart.

    You can put lipstick on the pig but it’s still a pig. Louis Vuitton shoppers aren’t going to shop at Wal-mart. Ever. And there are many rich people sho shop at Target because it is considered “frugal” chic.

  26. Idaho_Spud commented on Oct 30

    … and as always, the market blithely ignores it. Good news, bad news, no news… it’s *all* good for equities!

    Puzzling reaction, isn’t it?

  27. tjofpa commented on Oct 30

    “Crude futures tumbled as concerns about a possible attack on oil facilities in the Persian Gulf eased, and concerns about a softening world economy and lower energy demand resurfaced.”

    Oh, so THAT’s why oil’s down another $2/bar today. LOL
    Yup, looks like oils gonna be sacrificed this week to keep the marts up.
    No GDP, No sales at WMT, what slow down…
    just git those $1.99/gal signs ready for Monday.

  28. patf commented on Oct 30


    May I change subject (but stay [I believe] within macro-econ)?

    I’ll go by patf – I’m an IT specialist working for a finance firm in the Bay Area.

    We don’t ‘do macro’. We’re a ‘fundamental, value, quant’ firm.

    But macro is of course very interesting.

    The context for my question. US manufacturing (intl, comparative). I’ve seen trumpeted a number of places recently that the _value_ (it would be nice if the blog took html tags) of aggregate US manufacturing output has remained very steady of something between 20-25% of world output, and for some time.

    Let’s see … the numbers I get off a recent WSJ page are $1.79 tril. (US) and 7.03 (world total). Making the US, something over 25% of the world total. Japan is supposedly #2 at $.99 tril. And China behind Japan at .78. Both of these, obviously, being less than 50% of the US figure.

    Well there’s nominal GDP and ‘PPP’ GDP. Where, in the latter case, you’ve normalized for differing natl price structures around the world. And positions and values (of countries) change substantially as you travel from nominal to PPP-adjusted.

    My question itself. Are the above numbers for natl value of manufacturing output normalized in some comparable fashion comparable to PPP-adjusted GDP?

    The data providers (listed under the pie chart on the WSJ page that I’m looking at and that I’ve quoted) are Manufacturers Alliance/MAPI; Global Insight.

    I’ve of course looked up both and hit dead-ends of various kinds. The most obvious being that I need to pay $1000 for some report.

    Hope you can help.


  29. spencer commented on Oct 30

    Someone asked about minipulating the numbers for political purposes. As someone who began my career as a government economist I believe it is almost impossible for the numbers to be manulipated for political purposes. There are many safeguards in place for this and about the biggest is the personal and professional intergrity of the people collecting and publishing the data. They would scream bloody murder and the politicians know it.

    The problem with the PPI for autos in this quarters gdp report is interesting. For the past three or four years there has been a problem with the seasonal adjustment for autos in the PPI that has generated a strange jump in the overall PPI every fall. I’m willing to bet that it also showed up in the real gdp data for the last few years but because growth was stronger no one paid any attention. Moreover, for the past few years industrial production of autos has been strong in the 3rd quarter but this year it is down about 20%.

  30. patf commented on Oct 30

    Duh – I’ll amend my previous post.

    Japan’s total value of manufacturing output is > than 50% of the US. China’s is not.

    The question again – how have this _value_ been determined?


  31. fatbear commented on Oct 30

    alexd said: The rock star Prince said he wanted to party like it was 1999. So did a lot of people who were heavily invested in the Nasdaq, but unless you were short the party was over by 2001. Prince ‘s song seems to be analogous to the irrational euphoria present in the markets at that time. The nimble made fortunes off that euphoria but they were either very lucky or adept at dancing on thin ice. I believe Prince got his royalties on the recording before the market crashed.

    Showing your age, perhaps? The album (not CD) “1999” was released in 1983 (the year the NASDAQ-Comp passed 300); he got his royalties before the market bubble. By 2000 Prince was “the Artist Formerly Known as Prince” – those royalties were probably invested in the NASDAQ, which is why he changed his name back to Prince.

  32. alexd commented on Oct 30

    Hey I have a mind like a steel seive!

    And I am not showing my age , my age is showing on me.

    Thanks fatbear

  33. Brian commented on Oct 30

    Don’t worry. Stocks go up in recessions now. Everything is different this time. You must have missed the memo Barry. Stocks always go up, just like real estate. Dow 30000 by the end of the year!

  34. Idaho_Spud commented on Oct 30

    yep, the amazing levitating market. Climbing into the thin air, supported by nothing. Street magician David Blaine could learn some really amazing stuff from the Dow.

  35. dullboy commented on Oct 30

    Economics question: I keep hearing how administrations keep monkeying with inflation statistics (mainly under-reporting it to keep entitlement growth down). Wouldn’t a (fake) low inflation figure also cause growth to look greater than it really was (since some of that growth is really unreported inflation)?

  36. Ricardo commented on Oct 30

    It’s the big-brokerage houses trading between their own accounts running everything up to unload on the dumb retail guy. But alas, the dumb retail guy is not so dumb after-all and will most likely NOT buy stocks heading into a recession. It’s only a matter of time before reality catches up with the spending habits of Americans. Have fun paying off your debts! Why is Goldman Sachs worth 20% more today than 2months ago?

  37. V L commented on Oct 30


    You used to be a government economist. I have a few questions for you.

    Why did nobody scream bloody murder when they saw 26% spike in motor vehicle production according to their estimates?
    Does this number make any sense to you at all, at times when the entire sector is cutting production and an estimate of the Federal Reserve Board is -12%?
    Where is the personal and professional integrity of the people collecting and publishing this nonsense (a.k.a. fluke) data?

    How can we get to the truth as to why this nonsense data was used in estimating GDP? Do we need to bring a legal action against BEA in order to get to the bottom of this issue?

    The public wants to know the truth!

  38. ilsm commented on Oct 31

    Transcendentalist markets:

    Build castles in the air and put a foundation under them.

    H. D. Thoreau could see this.

    Live the live you dreamed.

    But, Thoreau warned “as you simplify it” poverty would not seem like poverty.

  39. Movie Guy commented on Oct 31

    V L – “Why did nobody scream bloody murder when they saw 26% spike in motor vehicle production according to their estimates?”

    “Does this number make any sense to you at all, at times when the entire sector is cutting production and an estimate of the Federal Reserve Board is -12%?”

    All of the auto OEMs aren’t cutting production.

    Besides, many had very strong sales in September and we may see more of that in October.

    The problem, if there is one, is this:

    Days to sell inventory

    DaimlerChrysler – 79 days
    GM – 76 days
    Ford – 75 days

    Yet, GM and Ford should post very good sales numbers for October.

    DaimlerChrysler is the new bleeder, along with Nissan and perhaps a few others.

  40. V L commented on Nov 1

    Movie Guy,

    Are you saying they had increased production by 26%?

    What movie did you watch?

  41. The Stalwart commented on Nov 1

    Something Looks Out of Place Here

    Barry Ritholtz at The Big Picture unmassages an already weak 1.6% reported GDP number all the way down to nearly zero. Exposed culprits include recent firesales from US automakers plus a very peculiar GDP deflator number used in the calculations

Read this next.

Posted Under