February 28, 2007
Easy come, easy go
Well, that 3.5% 2006:Q4 GDP growth was fun while it lasted.
The Bureau of Economic Analysis reported today that U.S. real GDP grew at a 2.2% annual rate in the fourth quarter of last year, rather than the 3.5% originally reported. Of the 1.3% lost output growth, about half is due to a downward revision of the end-of-quarter inventories.
To the extent that a sale of $100 in consumption goods comes out of inventories rather than production, those sales during the period do not represent newly produced goods. Thus, when inventories decrease during a quarter, that makes a negative contribution to that quarter's GDP-- we subtract the $100 inventory change from the $100 consumption spending if we want to calculate what was actually newly produced during the period. Because inventory drawdown during 2006:Q4 is now thought to be larger than originally estimated, GDP growth for the quarter is correspondingly now believed to be lower than the original estimate.
Although it makes a negative contribution to GDP, a decline in inventories is not a particularly bearish indicator. It means that businesses will go into 2007:Q1 with leaner inventories, enabling them to better weather any drops in demand in 2007:Q1, and possibly introducing an additional source for 2007:Q1 demand if businesses try to rebuild those inventories. To the extent that this is the reason for the downward revision in the 2006:Q4 numbers, it does not trouble me.
The other half of the downward revision came in equal parts from consumption spending, imports, and nonresidential fixed investment. The 0.2% decline in the contribution from consumption also doesn't much worry me, as it is not a big drop relative to the large size of consumption expenditures and does not suggest anything particularly alarming going on. Nor does the downward revision coming from imports bother me. Imports are another item that enters GDP with a negative sign, and the BEA is reporting that the drop in imports in 2006:Q4 was smaller than originally estimated, meaning that the boost to 2006:Q4 GDP growth from this category was therefore smaller than originally reported. I'm not bothered by this, because I found the magnitude of the decline in imports as originally reported a little implausible to begin with, so to now be told that the decline was in fact not as big as originally reported is for me no big deal. The one worrisome bit for me in today's numbers is the 0.2% decline in the contribution of nonresidential investment. If that is tanking along with residential investment, a recession may be impossible to avoid. Nevertheless, the contribution of nonresidential fixed investment to 2006:Q4 growth is still only reported to be -0.26%. If that gets no worse, we'll survive.
So, although a revision of reported GDP growth from 3.5% to 2.2% sounds worrisome, the details don't bother me that much. 2.2% is closer to what I was expecting for 2006:Q4 anyway, so essentially all this does is put us back where many of us thought we were in November.
Technorati Tags: GDP
Posted by James Hamilton at February 28, 2007 04:57 PMdigg this | reddit
If the drop in inventories was intentional, then it really is a bad sign, because it means businesses are forecasting weaker sales. I worry that it was intentional, because I don't get the sense that things were flying off the shelves in the 4th quarter.
Posted by: knzn at February 28, 2007 09:11 PM
Nice clear explanation of the impact of changes in inventory on GDP, Professor.
The explanation for the reduction in inventories that makes sense to me is that businesses see softening demand ahead and are reducing inventories to accomodate such; no one wants to have higher quantities of slower moving goods. Easily noticed indicators of slowing sales: all of those empty houses and condos and lots of unsold trucks and cars.
Well-run businesses are highly attuned to inventory levels, and quickly adjust production and purchasing to accomodate slowdowns in sales. At work, we look closely at inventory and shipments all the time.
Posted by: jg at February 28, 2007 09:19 PM
International Institute of Management (IIM) released a new report warning about the U.S. economic risks. The report:
1. Uncovers the forces behind Feb 27th stock market meltdown and the Chinese reaction to the outlook of U.S. Economy.
2. Forecasts the future behavior of U.S. and global markets.
Med Yones, the author of the white paper, warns against costly policy mistakes and provides a detailed analysis of the economic, social and geopolitical risks facing the United States
The complete text of the report is available at:
Posted by: thinktank at February 28, 2007 10:46 PM
jg - that link to the "IIM" is full of polemical hubris. Hopefully you don't mistake the analyses of someone like Prof. Hamilton for the likes of the self aggrandizing "IIM".
Posted by: Dan at March 1, 2007 01:46 AM
whoops, that last comment should have been directed at "thinktank".
Posted by: Dan at March 1, 2007 01:46 AM
The drop in inventories has little to do with expectations of future sales. The I/S rose in the third quarter -- undoubtedly is was unintended.So the 4th Q correction was to get rid of the prior quarters rise. The late Feb bounce in commodity prices probably signals that the inventory correction is over. One major reason for the great "moderation" is that firms now adjust much quicker to inventory problem then they use so you now get mini-cycles rather than the old recessions.
The big surprise to me was the strength of real PCE in the January income data release. It implies we are starting the 1st Q with stronger real pce growth then I had been expecting.
Posted by: spencer at March 1, 2007 06:28 AM
How is the seasonality of inventory accounted for in quarterly annualized GDP?
Using the last few years' pattern to correct would be noisy, with disruptions like Y2K, but using the last few decades would miss trends.
If retailers are getting bigger (WalMart) and more invested in suppliers then inventory could become more cyclical as manufacturing operations is optimized for steady output.
Seems like seasonal correction might be a big issue.
Posted by: Name at March 1, 2007 09:10 AM
Look for the PCE to slowdown a good deal in February however. But I don't see recession through April 1st.
Typically, you want to see a midwestern corn grower run through a sheet farm, but that sometimes still doesn't do it.
While James's concern over industrial production is valid and 6 month ISM has weakened compared to the previous 6 months, untill I see a "2 month plop" like January/February 2001(for example) in ISM and industrial production, there isn't anything to get hyped up over except soft growth.
Though maybe the consumer is going to be the leader this downturn and take all those assets they have and pay off debt. Hard landing baby then.
Posted by: dryfly at March 2, 2007 02:27 AM
What's really bothering me is the spin on current data. No one believed 3.5 % GDP growth. Frankly, I do not believe the 2.2 % number because, IMO, inflation has been understated.
Closer to 1.5 pct than 2.2.
Posted by: zinc at March 2, 2007 04:14 AM
Once again my biggest concern is not with the new GDP number. My biggest concern is that the government thinks it can control the economy and our monetary system relying on indicators that are off by 37%. I have seen all the statistics that "prove" that indicators are relatively reliable but what is usually compared in these "proofs" is the final revised version. It is simply impossible for the economy to be run by congress or any of its myriad bureaucracies. Yet we are running headlong into more and more command economy systems.
Thanks for the analysis Professor for at least putting the numbers in perspective. I cannot blame the messenger for the message.
Posted by: DickF at March 2, 2007 04:53 AM