Prediction: Bloomberg has 311,000 seasonally adjusted claims, up from 305,000 (before likely upward revision today) last week.
Seasonal adjustment Factors:
- Week ended Oct. 5, 2013 — 90.1
- Week ended Oct. 6, 2012 — 93.8
That’s a pretty big difference in the year-over-year seasonal factors for no readily apparent reason. It works to this year’s disadvantage.
Raw (i.e., actual) claims:
- Week ended Sept. 28, 2013 — 252,092 (before revision)
- Week ended Oct. 6, 2012 — 329,919
For the prediction to pan out, raw claims will have to be 280,000 or lower (280K divided by .901 is 311K, rounded). That seems achievable.
shouldn’t be affected much may be affected, based on a few headlines I’ve seen in the past hour, by the 17% shutdown.
The report will be here at 8:30 a.m.
HERE IT IS (permanent link) — YIKES:
SEASONALLY ADJUSTED DATA
In the week ending October 5, the advance figure for seasonally adjusted initial claims was 374,000, an increase of 66,000 from the previous week’s unrevised figure of 308,000. The 4-week moving average was 325,000, an increase of 20,000 from the previous week’s unrevised average of 305,000.
… UNADJUSTED DATA
The advance number of actual initial claims under state programs, unadjusted, totaled 336,849 in the week ending October 5, an increase of 84,616 from the previous week. There were 329,919 initial claims in the comparable week in 2012.
I suspect the 17% shutdown will get much of the blame for this, but I’m not buying it, especially since furloughed federal workers are getting back pay, i.e., they shouldn’t be getting unemployment benefits at all.
But will they get benefits anyway, on the “logic” that they aren’t getting paid, right now, at this moment? Is it really possible that the shutdown is a double-dip for furloughed federal workers?
Or has there been a catch-up in claims processing left over from previous weeks because of computer conversion issues in a few states, one of which is California? (In the comments, Steve notes that this is a factor. Why don’t they post claims to the week in which they were actually filed? Talk about rendering comparative data meaningless.)
Mitgating factor: If last year’s seasonal adjustment factor had been used this year, seasonally adjusted claims would have come in 15,000 lower (336,849 divided by .938 is 359K, rounded).