Floyd Norris on the revised Birth Death model:
(Click for larger graphic)
“Starting in 2000, the government phased in a “Birth/Death Model,” which tried to correct for that.
It has not worked particularly well, as can be seen in the charts. Now, in an effort to improve that model, the bureau is changing some estimates to quarterly from annually. But there is little it can do about the basic problem: when the economy turns up, there will be more new jobs than previous experience would predict. When the economy turns down, the opposite will happen.
One reason for the large changes is that the benchmark revision is based on returns the companies file when they pay the premiums for unemployment insurance, rather than on the survey of employers. The tax returns are deemed more accurate, but they take a lot longer to be available.”
Good stuff . . .
>
Source:
The Inexact Science of the Jobless Numbers
FLOYD NORRIS
NYT, February 4, 2011 http://www.nytimes.com/2011/02/05/business/global/05chartseuro.html
What's been said:
Discussions found on the web: