Unemployment Claims: 367K SA (378K With Consistently Applied Seasonal Factors)
SEASONALLY ADJUSTED DATA
In the week ending January 28, the advance figure for seasonally adjusted initial claims was 367,000, a decrease of 12,000 from the previous week’s revised figure of 379,000. The 4-week moving average was 375,750, a decrease of 2,000 from the previous week’s revised average of 377,750.
… UNADJUSTED DATA
The advance number of actual initial claims under state programs, unadjusted, totaled 415,094 in the week ending January 28, a decrease of 1,786 from the previous week. There were 464,775 initial claims in the comparable week in 2011.
The year-over-year drop in not seasonally adjusted claims was 10.7%. The seasonally adjusted drop from last year’s 424,000 was 13.5%. The discrepancy is there because this seasonal divisor is higher, something which is a bit hard to handle (at least considering the degree of the change) given that we’re talking about the last full week in January in both years.
If last year’s seasonal adjustment factor of 109.7 (findable at the interactive link here) had been applied to this year’s raw number of claims, seasonally adjusted claims would have come in at 378,000 (415,094 divided by 1.097) — virtually the same as last week’s (as usual) upwardly revised 379,000. Instead, this year’s SA factor was 1.131. Even if the stat guys at DOL can defend the use of the higher factor, they should be explaining that its use — and not necessarily demonstrably better economic conditions — led to the reported “improvement.”
Yet the press will run with today’s report and say it’s some kind of great sign of improvement, when it’s no such thing.
Yes but come February things will be quite different on paper:
Art Cashin Explains Why Several Hundred Thousand Jobs Are About To “Vaporize”
http://www.zerohedge.com/news/art-cashin-explains-why-several-hundred-thousand-jobs-are-about-vaporize
try on average -375K jobs. this will be on top of all the holiday seasonal (November to January) hires who go back to whatever they were doing the rest of the year.
Comment by dscott — February 2, 2012 @ 9:19 am
#1, that’s an annual event which is already considered before seasonalizing. As far as I can tell, ZH’s excerpt doesn’t relate to the excerpt. Now if they’re going back retroactively to adjust Birth/Death, that’s another matter. I thought those revisions went from quarterly to annually.
Comment by TBlumer — February 2, 2012 @ 10:39 am
TB, I think the second SA in the next to last paragraph is supposed to be 113.1 instead of 1.131.
Comment by Joe C. — February 3, 2012 @ 5:56 pm
#3, actually the 109.7 needed to be 1.097 for the division to work correctly. The 109.7 is a factor DOL uses which is really a %.
Thx.
Comment by TBlumer — February 3, 2012 @ 8:18 pm