What to make of the employment report

Lots of tap-dancing this morning by economists who are trying to figure out how they could get the November numbers so wrong. The consensus estimate was for a gain of 150,000 jobs but the government reported a pickup of only 39,000. There's a chance that the November data might be revised upward to better reflect retail hires, but that doesn't change all the other weak indicators, including an increase in the unemployment rate. So what happened? NYT columnist Floyd Norris wonders about some of the data.

The government says total manufacturing jobs fell by 13,000 in November, making it the fourth consecutive month to post a small decline. (There are still 96,000 more jobs than there were a year ago.) Yet only two days ago the Institute for Supply Management reported on its monthly survey of manufacturing companies, and the numbers looked good. For employment, the figure was 57.5. Any number over 50 indicates more companies are adding workers than are reducing employment, and a figure that high shows that a substantial proportion of companies are hiring. This is the eighth consecutive month that the employment figure in that report was 56 or higher. Not since 1973 has there been such a string. It is hard to see how both the B.L.S. and the I.S.M. surveys could be correct. That fact provides reason to at least hope the employment numbers will be revised upward in coming months.

Here's another possibility:

Bernard Baumohl, of the Economic Outlook Group, points out that the [government] survey was conducted just before Thanksgiving and that the proportion of employers who got their answers in quickly enough to be counted was unusually low, at 66 percent. He adds: "When the final employment revision for November comes out in February, the B.L.S. should by then collect about 95 percent of the surveys and we believe the numbers will then show a substantial upward revision for last month."

Here's the most likely - and most unsatisfying - explanation: The economy is indeed recovering, but ever so slowly. Some industries, most notably construction, just aren't hiring yet, and others have figured out how to do more with less. Eventually, there will be a greater willingness to bring on additional workers, though even then the hiring is likely to be selective. Folks with yesterday's skill set are going to have a tough time. From Real Time Economics:

While the November payroll data released today were certainly a disappointment, we are very hesitant to make too much out of this report, particularly with other evidence now beginning to swing in the favor of a labor market that is gaining some traction (initial jobless claims are trending lower, the ADP survey has picked up, small business optimism is finally starting to improve, etc.). In our view, the labor market recovery will continue in the months ahead, although invariably, as is the case with most data, there will be ebbs and flows within the trend. November was certainly something of an ebb, but is unlikely to prove to be anything more than that. -Joshua Shapiro, MFR Inc.

That's where I'm putting my money - an overall disappointment in November, but perhaps not as bad as the headline numbers would indicate and not enough to derail the other positive indicators we've been seeing.

Chart: NYT