- by New Deal democrat
This morning brings us two more bits of news that coincide with the story of incipient Recovery.
The University of Michigan consumer confidence index rose to 70.2, and expectations to 69.2, both of which are the second highest readings all year (June was slightly higher).
Meanwhile, the Census Bureau reported that wholesale sales rose +0.5% in July, while inventories fell -1.4%, meaning that the July inventories/sales ratio for merchant wholesalers declined to 1.23.
At the height of the expansion, this ratio was 1.12. It collapsed to 1.35 by January during the economic free-fall. More importantly, tighter inventories ultimately lead to hiring in manufacturing, and this ratio is right where it was in 2002 and early 2003 just before jobs started to increase in the last Recovery.
This is Why You Can’t Outrun a Cheetah
38 seconds ago
4 comments:
These are noise level changes that you and the other green-shootists ignore when they carry negative signs. Note that the "improvement" in job losses (unadjusted) in the current reporting period is measured against an adjusted number. What happens when the current number is negatively adjusted? Do you you subtract a green shoot?
Spartacus:
Any one piece of data is a "noise level" change. Put enough together, and you've got a trend, and the trend in the vast majority of data in the last few months has been "less bad" and ultimately turning "good."
Why do you think we should ignore seasonal adjustments?
What I mean by noise level is within the measurement error of the quoted green shoot statistic. For example, in the footnotes of the PDF file cited in the current report of a .5% increase in wholesale sales, there appears this footnote:
* The 90 percent confidence interval includes zero. The Census Bureau does not have sufficient statistical evidence to conclude that the actual change is
different from zero.
Both you and Bonddad understand the concept of measurement error, and the fact that you ignore it shows that your commentary is tendentious.
As I said before, most any statistic by itself is within the noise level. Presumably you do not believe that Calculated Risk, Barry Ritholtz et al, and any among the scores of bearish bloggers who post on the same data are similarly "tendentious."
The data is part of a bigger mosaic.
Post a Comment