Where are the April showers in the jobs report?

Yes, April’s jobs report appears to be sunny: more than 300,000 jobs added to the economy, when you include upward revisions of the past two months. That’s what the “establishment survey” of employers says. And yes, the headline unemployment rate, which comes from the door-to-door “household survey,” dropped from 6.7 percent to 6.3 percent, seeming to confirm the story.

But for those of you who expect members of the mainstream media to rain on every parade, not to worry: we do have a shower for you, if not a thunderstorm.

You see, the household survey reports a stunning 733,000 fewer Americans “unemployed” in April alone, which would accord almost perfectly with the employers’ jobs added number.

But 73,000 fewer Americans reported that they were actually working in April than in March. So what can possibly be happening?

One response is that the two surveys so often disagree, it’s best to inspect any one month with a skeptical mind frame. And that’s certainly true. Another, I suppose, would be a massive one-month conspiracy among hundreds of thousands of Americans to lie to interviewers from the Bureau of Labor Statistics, claiming not to be working when they really are.

But the answer, at least for the household survey, is actually right in Table A-1 of the BLS’s report. In the second row, labeled “civilian labor force,” the next-to-last column reports the number of Americans in the work force as of March, “seasonally adjusted,” in thousands: 156,227 – meaning 156.227 million. The last column reports the statistically adjusted total for last month – April: 155,421 (in thousands, 155.421 million). That’s right: the April number of people in the workforce is lower, even though the population grew by its usual 200,000 or so. So roughly 800,000 people “dropped out” of the work force last month.

BLS.TableA1.April2014

Practically speaking, that means nearly a million more people than in March told the BLS that they hadn’t looked for work in at least a year.

The big question, of course, is whether they were thoroughly fed up with the job search – “hyperdiscouraged,” you might call them – or by contrast, all too happy to hang up their 9-to-5 shoes. And if you’ve been following our coverage of unemployment over the past year, you know that we suspect a lot of the slow growth in the labor force, as compared to much faster growth in the population, is due to retirement, as 10,000 or so baby boomers hit age 65 (or 66, for that matter) every single day.

But here’s the rain I promised. If you look at row 9 of table A-1, you will see the total number of us who are not counted in the BLS’ reckoning of the “civilian labor force” but are counted as “persons who currently want a job.” That is, people (I don’t know why they’re called “persons”) who tell survey takers that they would like to work. That number was unchanged in April: 6,146,000 of us.

So here’s a consistent, less sunny story about April’s numbers: sure, about 700,000 fewer Americans were reported as “unemployed,” but an even bigger number dropped out of the work force, driving down the unemployment rate artificially, even if the 300,000 or so baby boomers who hit “retirement age” in April hung it up.

Do we believe our own inclement account? Not fully. Our own inclusive reckoning of the under- and unemployed, the Solman Scale U-7, based on the household survey, also dropped in April, from 14.75 to 14.38 percent. That’s not nearly as steep a decline as the headline fall – 6.7 to 6.3 percent – but it’s still substantial.

The Solman Scale adds to the officially unemployed part-timers looking for full-time work and “discouraged” workers — everyone who didn’t look for a job in the past week but says they want one.

The Solman Scale adds to the officially unemployed part-timers looking for full-time work and “discouraged” workers — everyone who didn’t look for a job in the past week but says they want one.

Okay, one last cloudburst: Unemployment among 16 to 19-year-old black Americans rose significantly to 36.8 percent.

So again, any one month’s data – and all of its statistical noise – isn’t worth too much of a storm; it’s still the long-term patterns that count. We’ll have more on this on Friday’s broadcast.