A Disjointed Jobs Report and U-7, Month Two

BY Paul Solman  February 4, 2011 at 3:10 PM EDT

Today’s unemployment data are a muddle. The two surveys – of individuals and of employers – seem to say very different things. Unemployment dropped – to 9 percent, according to the “household survey.” Three cheers. But the “payroll survey” reports a mere 36,000 new jobs added in January, far below the consensus estimate or, for that matter, the expected growth in population. A groan or two at least.

Upon closer inspection, though, neither statistic tells a wholly convincing story. Not surprising, perhaps, given the unreliability of month-to-month fluctuations, especially in data extrapolated from a sample (60,000 households surveyed out of total in the vicinity of 115 million). But not reassuring either.

The fact is, the January numbers are a puzzle. When you actually read the January household survey, you find a decline in Americans with jobs – 1.5 million fewer employed than in December. That’s presumably due to post-holiday contraction, especially in retail, for which the Bureau of Labor Statistics makes a “seasonal adjustment,” which resulted, this month, in an addition of 117,000 jobs.

But how does 117,000 more employees square with the payroll survey, which reports only 36,000 new jobs?

Maybe weather. In an excellent exchange on the New York Times Economix blog, reporters Floyd Norris and Dave Leonhardt discuss the weather and the data. (They began yesterday and continue today.)

………………………………………………………………………………………………………………………….

This is the point at which I weigh in with a new more inclusive statistic we’re calling “U-7.” The monthly Current Population Survey (CPS) itemizes categories of “labor underutilization” all the way up to U-6, which it defines as counting “Total unemployed, plus all persons marginally attached to the labor force [essentially, those who haven't looked for work in the past four weeks but did look in the past year], plus total employed part time for economic reasons.”

As a percentage, U-6 came in at 16.1 percent in January, down substantially from 16.7 percent in December. But…those numbers too are “seasonally adjusted.” The raw data show an increase- from 16.6 percent to 17.3 percent. Again, this is post-holiday, so the rise is probably temporary, i.e., “seasonal.” But it means more Americans reporting that they were out of work last month than in December. How much do we trust the adjustment?

Now for “U-7.” We created it to include all Americans who say they want a job, period. Even, that is, if they didn’t look for one in the past year, which disqualifies them from the official data entirely. I don’t claim any priority for “U-7.” Take it for what you think it’s worth. As I pointed out last month, who knows how many Americans are fibbing about their job status, and should actually be added to the employment rolls, because they’re working off-the-books?

But it seems only fair to report the number of Americans who say they want a full-time job, but don’t have one. And right now, that number, using government data, would be 17.9 percent — nearly 29 million Americans.

If you’re not number-numbed yet, bear with me for one final point. Last month, when we debuted “U-7,” it stood at 18.7 percent. Now it’s dropped to 17.9 percent. That should be good news, right, at least relatively speaking?

But when you scrutinize the numbers, the real action is in the lower number of “seasonally adjusted” unemployed and a lower number of part-timers looking for full-time work. Are either of these numbers to be trusted? When I squint at the fine print and discover that the “non-institutional civilian population” fell in January by several hundred thousand, I’m not so sure of what this month’s numbers really mean. (Norris and Leonhardt of the Times point to methodological changes in the official numbers that may explain the curiosities.)

For a verbal rendition of much of the above, see Hari Sreenavasan’s interview with me on the Rundown. Leonhardt will be on the NewsHour tonight.

………………………………………………………………………………………………………………………….

And this just in from Iraq and Afghanistan Veterans of America (IAVA), self-described as “the country’s first and largest nonprofit, nonpartisan organization for veterans of the wars in Iraq and Afghanistan:”

[U]nemployment amongst Iraq and Afghanistan-era veterans jumped to a record 15.2 percent in January, a 3.5 percent increase from December 2010…The unemployment numbers for Iraq and Afghanistan-era veterans are the highest they have ever been since the Labor Department started tracking. This is absolutely unacceptable.