The November Jobs Report: Better Than It Seemed

It's important not to make too much out of a single economic data point. But Friday's monthly jobs report may be even better news than it seems. The unemployment rate fell to 10 percent in November, and companies only shed 11,000 payroll jobs in the month. That was much better than economists expected, and the smallest such drop since late 2007, and perhaps early indication that, as I've argued, we'll be seing jobs growth sooner rather than later.

A look inside the report – and, again, with the caveat that it would be folly to draw too strong a conclusion from a single month's data – suggests four other reasons to be optimistic.

1. The trend is your friend. Yes, the payroll jobs number in November came in much better than expected. In each monthly report, the Bureau of Labor Statistics also revises the previously reported job totals for the prior two months. And in the past few months, the trend has been that the government overstates the job market's weakness in the just-completed month. In the original release for September, the government concluded the economy shed 263,000 jobs that month. A month later, however, in its October release, September's loss was revised down to 219,000. According to last Friday's release, the September job loss was actually only 139,000. The government originally said the economy lost 190,000 jobs in October, but Friday's report cut that number to only 111,000 jobs. The upshot: for the past three months, the government's first take at job loss figures has been understating the strength of the recovery. Should this trend continue, it's quite likely that when the November numbers are revised over the next two months, that 11,000 loss could turn into a gain.

2. Service jobs growth. Analysts frequently point to the troubling losses in manufacturing, housing, and construction. It's understandable, since those sectors are politically important (manufacturing) and accounted for so much of recent economic growth (housing/construction). But manufacturing jobs are likely to fall even as the economy recovers, thanks to long-term secular trends of globalization, outsourcing, and automation. As for housing, we shouldn't expect the sector that got us into the mess to get us out of it. Rather, the recovery will come from the vast services sector. Those sectors—which include government, health care, and education—account for about 86 percent of total payroll jobs. In the October report, BLS reported that the economy lost 61,000 service jobs. But the November report revised that loss to a gain of 2,000 October service jobs and found that the sector added 58,000 service jobs in November. That's two straight months of growth. What's more, the professional- and business-services sectors—purely private-sector service jobs—were up 86,000 in November and were revised to show a gain of 38,000 in October.

3. Temporary help. The economy started growing this summer. But it frequently takes a few quarters of sustained growth until businesses and consumers really trust it. During these periods, employers go through a predictable process. When business stabilizes, they stop firing lots of workers. When demand and orders pick up, rather than hire, they prod existing workers to work harder and invest in productivity-enhancing technology and processes. That's why the productivity numbers have been so impressive the past six months. When things continue to improve, they still don't quite believe it. After all, a lot of economic activity in recent months has been goosed by stimulus efforts, from low mortgage rates to Cash for Clunkers. So rather than hiring full-time workers to cope with rising demand, they bring on temporary workers, who can easily be let go if demand fizzles again. In November, the economy added 52,000 temporary jobs. And since July, temporary help services employment has risen by 117,000.

4. Anti-disestablishmentarianism. The monthly jobs report presents data from two different surveys. The payroll jobs figures tell us how many people companies (i.e., establishments) say they have on their payrolls. The government uses the household survey, in which it calls up people and asks them if they've been working, to compile the unemployment rate. In the Bush years, when payroll jobs failed to materialize, partisans discounted the payroll figures and pointed instead to strength in the household survey as evidence of jobs growth. After all, if more people were working for themselves, starting businesses, working as consultants, etc., it wouldn't show up in the establishment figure but would show up in the household survey. (I called this tendency Anti-disestablishmentarianism.) I was, and remain, skeptical that the household survey is a superior measure of the employment picture. Most people want payroll jobs—the kind that come with benefits, paid vacation, etc.—rather than freelance arrangements. Plus, the household survey is a measure of what people say they're doing. Still, all things being equal, it's desirable for both the establishment and the household surveys to be moving in the same, positive direction. In November, according to the household survey, the number of people working rose by 227,000.

One month's data does not a recovery market. But you have to start somewhere. It may turn out that the November jobs report was the beginning of the end of the great employment recession of 2008-09.