At AP, Big Jobless Claims Miss Due to 'Seasonal Issues'; Consistent Calcs Would Have Led to Over 400,000 Claims

Todays unemployment claims release from the Department of Labor reported that initial jobless aid applications for the week ended July 14 were 386,000 after seasonal adjustment. Business Insider's email this morning carried a prediction of 364,000. Bloomberg's consensus prediction was 365,000.

At the Associated Press, in his 8:45 a.m. dispatch (saved here for future reference, fair use and discussion purposes), Economics Writer Paul Wiseman was inadvertently correct when he wrote that "the figures may have been distorted by seasonal factors." Well yeah, Paul, but the seasonal distortion isn't the one you cited. As will be seen after the jump. today's number arguably should have come in at over 400,000.

Here is some of Wiseman's whining:

But economists view the recent numbers with skepticism. The government struggles to adjust the data to reflect temporary summertime layoffs in the auto industry, they note. And this year, many automakers skipped those typical shutdowns to keep up with demand. That led to fewer layoffs two weeks ago, which the Labor Department didn't anticipate.

"All of this is statistical noise," Joshua Shapiro, chief U.S. economist at MFR Inc., wrote in a note to clients. "What will be more important than these short-term gyrations is where claims settle down after the distortions end. We suspect that the data will point to a soggy labor market."

The less volatile four-week average fell by 1,500 to 375,500.

When applications fall below 375,000, it typically suggests hiring is strong enough to pull the unemployment rate down. Applications had been trending near or above that level this spring.

During the comparable week last year, the week ended July 16, DOL's seasonal adjustment factor was 112.4. This year, it was 117.3. Both years had five regular business days. Why is there a 4.4% difference between the factors.

And yes, it matters:

• This year's seasonally adjusted number of 386,000 is the result of dividing 452,960 raw claims (until they find more next week, as has been the case for all but one of the past 60-plus weeks) by 1.173.
• Least year's seasonally adjusted result of 418,000 was the result of dividing 470,086 raw claims by 1.124.
• If this year's raw claims had been divided by last year's seasonal adjustment factor, the seasonally adjusted result would have been 403,000 (452,960 divided by 1.124).

The only reason the four-week average fell is that last week's figure (along with this week's) was bogusly low for the same reason --  but in that case, the 2012 seasonal deflation factor was almost 9% higher than the one used in 2011. Then, as seen above, Wiseman used the four-week average affected by that artificial calculation to imply that we're really not far from a claims level which might lead to a drop in the overall unemployment rate. You can't possibly believe that, Paul.

Bloomberg's Alex Kowalski used the DOL's "timing of annual automobile plant layoffs" excuse. But last week it was supposedly all about how the car companies aren't laying many workers off at all for maintenance and retooling shutdowns. Which one is it, guys?

To me, the bottom line is that the establishment press is grasping at any all available excuses to mask obvious deterioration.

Cross-posted at BizzyBlog.com.