GDP Yet Another Misleading Indicator
I was out of pocket all day yesterday, and didn't get a chance to address the GDP data. At 3.4%, it came in slightly below the consensus of 3.5%.
There were two significant factors: A slight increase in wages, and big decrease in inventory. We'll see the July NFP report this Friday (more on this below) -- but lets address inventories, via Barron's Alan Abelson:
"The gleeful assumption by the sunshine crew that the inventory drop in the GDP report augurs big growth in output in the second half is more than slightly misplaced. The reason simply is that a huge chunk of the inventory reduction was the result of auto companies staging fire sales to clear their dealers' lots, an action certain to take a sizable bite out of future sales."
No big surprise there. Giving away cars at cost is hardly indicia of a red hot economy. But what about Industrial output? Isn't that a significant factor?
In a word, no. More than half of the uptick in output was Utilities cranking out more juice thanks to the heat wave. And a sizable chunk of the rest was (you guessed it) these same auto makers producing more cars to be given away at cost. GM's July numbers showed a 46% sequential monthly increase.
Where I disagree with much of the bear camp is where they go anti-Candide -- where every data point is the wosrt of all possible worlds. Instead of focusing on the underlying weakness below the headline numbers, investors of the Ursine variety are calling the data proof that the Fed has to keep raising rates. To me, that's an after the fact rationalization -- they were gonna keep raising anyway.
As to the Jobs situation, the slack in the Labor market, as Sir Alan likes to call it, continues to get even slacker:
"Moreover, something's definitely askew when the stock market starts to go bananas while the job market remains depressingly limp. As Challenger, Gray & Christmas, the placement outfit, observe, layoffs typically ease in the summer months -- but not this summer. Instead, it has produced a bitter harvest of job cuts. In May and June alone, layoffs totaled nearly 200,000. And more recent weeks have brought no respite."
And that's just layoffs.
As to NFP, we have yet to see a single good number that was not been the result of some hedonic adjustment or another (Prime suspecty: The Birth/Death adjustment, added in 2001). Between the inherent upside bias built into the BLS models, and the other assorted massages, its astonishing the data is this bad. And that's before we back the least significant aspect of NFP -- governement hiring.
Bottom line: Torture the data long enough, and it will confess to whatever you want it to.
Poor Darth Vadar
UP AND DOWN WALL STREET
By ALAN ABELSON
Barron's, Monday, August 1, 2005
TrackBack URL for this entry:
Listed below are links to weblogs that reference GDP Yet Another Misleading Indicator:
» Further thoughts about the latest economic statistics from Econbrowser
I've had a little more time to ponder the meaning of some of the economic data released last week, and here's what I've come up with. [Read More]
Tracked on Aug 1, 2005 11:32:47 PM
I guess we will find out Friday whether or not the BLS birth/death adjustment is providing the bulk of so-called job growth. While April through June historically have strong upward revisions under the birth/death model, July is historically a strong downward revision. I dont know (or care) why; it just has been in years past.
Posted by: Josh | Jul 30, 2005 2:01:42 PM
The comments to this entry are closed.