Skip to main content

AP's Rugaber Lowers the Bar For Unemployment Claims by Raising the 'Strong Hiring' Threshold

Tom Blumer's picture

Two examples of poor press handling of what initially appeared to be fairly good news about initial unemployment claims last Thursday got lost in the pre-South Carolina primary hubbub.

The first and most obvious was in the writeup presented by the Associated Press's Chris Rugaber. In his fourth paragraph, he raised the threshold below which a consistent level of weekly claims might be expected to move the unemployment rate downward by 15% from where it was less than 2-1/2 years ago:

The number of people seeking unemployment benefits plummeted last week to 352,000, the fewest since April 2008. The decline added to evidence that the job market is strengthening.

Applications fell 50,000, the biggest drop in the seasonally adjusted figure in more than six years, the Labor Department said Thursday. The four-week average, which smooths out fluctuations, dropped to 379,000. That's the second-lowest such figure in more than three years.

A department spokesman cautioned that volatility at this time of year is common. Applications had jumped two weeks ago, largely because companies laid off thousands of temporary workers hired for the holidays.

Still, when weekly applications fall consistently below 375,000, it usually signals that hiring is strong enough to push down the unemployment rate.

Geez, Chris. In September 2009, in an analogous weekly claims report, you wrote the following:

Still, unemployment claims remain significantly above levels associated with a healthy economy and indicate that jobs remain scarce. Weekly initial claims are generally at 325,000 or below in a growing economy.

This would mean that if claims are above 325,000, the economy would be "generally" expected to be not growing meaningfully, generating jobs, or lowering the unemployment rate (or keeping an already low rate down).

No further elaboration is required, Chris. Gotcha -- And please, don't try to wiggle out of this by trying to distinguish between a growing economy and the unemployment rate coming down.

The second bust was the press's failure to note that the sharp reported drop to 352,000 initial claims from the previous week's 402,000 and from 415,000 during the the comparable week in 2011 was almost entirely due to the big difference between the seasonal adjustment factor used in 2012 compared to the one used in 2011. As noted here, if the factor used in 2011 had been applied to the 2012 data, initial claims would have come in at 394,000, not very impressive drop at all, and hardly evidence that, as Rugaber claimed, "the job market is strengthening."

I'm not second-guessing what DOL does and how they do it (yet), but given that both years represented the full business week before the Martin Luther King holiday and that raw claims only dropped 5% from 2011 to 2012, it's hard to see why the seasonal adjustment factors would have differed by so much.

As far as I can tell from reviewing several related press reports, nobody in the business press noted how the seasonal adjustment factor used on last week's raw data caused reported initial claims to be artificially low in a very real sense.

Cross-posted at


#1 Why is it that the DoL

continues to change the metrics it uses to define unemployment? Is this something that happens periodically or is this just a new- more efficient, or modern, or bleh way of interpreting the data? It just seems odd that the unemployment rate as traditionally (pre-Obama) measured that used to include those who've given up looking (?), and whatever this new method of seasonal adjustment amounts to, is showing up during this administration. I guess if it continues beyond, then that's okay, but you certainly don't see anyone else mentioning the "new math" as it were, and it's inherently suspicious given that the "new math" always seems to lessen the failures of this administration's economic policies.

#2 I don't think a lot of changes ...

... have been made.

The only one I can think of right now it that Birth Death estimates went from annual lookbacks to quarterly (I think).

The problem with the unemployment rate and the center-right coverage of it is that sometimes the center-right pundits act as if their (recalculated) rate is the be-all, end-all rate, instead of starting with the government's report, saying why it's wrong, and adjusting accordingly.

#3 I recall 400K as a key

I recall 400K as a key threshold below which the claims number was considered to be a positive sign. That threshold seems more consistent with the actual data -- i.e., the lower nominal unemployment rate that has been the trend during the past rew months. Another interesting number to have would be the number of people actually employed (and how that has changed from month to month and year to year). If fewer people are actually working, there is likely to be a lower weekly 1st time claims rate.

#4 Since Obama took over

Every agency cooks the books.

Labor keeps changing the unemployment calculations to favor Obama.

Defense claims we can get by with way less military so Obama can cut it.

The EPA exaggerates environmental crises like "global warming" to justify Obama taking over the energy sector.

HHS makes up uninsured numbers, smoking death numbers, obesity numbers, etc. to justify Obama taking over health care.

The list goes on and on.

The simple rule is this: if it comes from any part of the Big Red Bureaucracy it's a lie.

"Tolerance is the virtue of the man without convictions." G.K. Chesterton

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.