The Job Market Still Stinks

After the release of Friday's employment report showing that the nation's unemployment rate had dropped below 8% in September for the first time in 44 months, Labor Secretary Hilda Solis appeared on CNBC, claiming to feel "insulted" by those who believe that its numbers were conveniently cooked for President Obama's electoral advantage. She further asserted: "We have a very professional civil service organization (with) ... top economists. ... These are -- these are our best-trained and best-skilled individuals."

With the exception of at least two economists who have donated to political campaigns (but who don't appear to have been in positions to influence the numbers), Solis appears to be mostly right about the professionalism at her department's Bureau of Labor Statistics. Those who toil there should feel insulted at how Solis and whoever else is responsible chose to present the information in September's Employment Situation Summary.

The primary bone of contention is the September increase of 873,000, according to the BLS's Household Survey (based on interviews and surveys of heads of households), in the number of people employed. That figure made it into the summary's narrative, and was explained as follows:

Total employment rose by 873,000 in September, following 3 months of little change.

This was a brazen attempt to make a wildly fluctuating and relatively unimportant figure in a survey where the primary mission is to estimate the percentage of people who are not employed seem more important than the mediocre seasonally adjusted September job additions of 114,000 in the Establishment Survey of employers. I reviewed this year's first eight Employment Situation Summaries; the Household Survey's specific employment pickup or loss was never mentioned in any of them.

One could argue that the September increase was too big to ignore and had to be cited to address inevitable questions. If so, then the report should have noted the figure's historically wild fluctuations and relative insignificance instead of blandly telling readers about three previous months of alleged "little change" -- an assertion which was itself quite deceptive, especially when comparing the Household Survey's results for the past seven months to the Establishment Survey:


While the Establishment Survey has been showing job growth plodding along at a mostly tepid pace, adding an average of 111,000 jobs a month during the past seven months, the Household Survey's total employment figure has seen month-to-month differences varying from the Establishment Survey by several hundred thousand in both directions. That's hardly "little change," Hilda.

Of course, the press couldn't resist reporting the highly unreliable figure. It made it into the second paragraph at the Associated Pressaka the Administration's Press, while the Establishment Survey result got mentioned in paragraph 3. The New York Times saved the Household Survey's 873,000 employment growth figure for a later paragraph, but implausibly claimed that "its credibility was bolstered by an unexpectedly robust rise in consumer confidence." The confidence reference was to a Gallup poll which, as the Times acknowledged, attributed its rise "almost entirely to increased optimism among Democrats." In other words, confidence is up because many Obama supporters recognize a duty to say that they're confident even if they're not.

To the AP's credit, it noticed, though not until the 24th paragraph of its report, that most of the Household Survey's employment pickup was in an unimpressive area: "582,000 more people reported that they were working part-time last month but wanted full-time jobs ... the biggest increase in so-called underemployed Americans since February 2009."  This is exactly the kind of result Republican presidential challenger Mitt Romney was thinking of when he said on Friday: "This is not what a real recovery looks like."