The unemployment rate was reported as unchanged today, which was what most analysts expected. December’s Employment Situation report was released this morning. The number of jobs in the US decreased 85,000. This was a bit worse than the Bloomberg consensus range of -50,000 to 40,000. From the report:
Nonfarm payroll employment edged down (-85,000) in December, and the unem- ployment rate was unchanged at 10.0 percent, the U.S. Bureau of Labor Statis- tics reported today. Employment fell in construction, manufacturing, and wholesale trade, while temporary help services and health care added jobs.
In December, both the number of unemployed persons, at 15.3 million, and the unemployment rate, at 10.0 percent, were unchanged. At the start of the re- cession in December 2007, the number of unemployed persons was 7.7 million, and the unemployment rate was 5.0 percent.
Don’t forget that in addition to those 15.3 million unemployed, we have another 9.2 million in part-time jobs looking for full-time work and 2.5 million without jobs that no longer count as unemployed:
The number of persons employed part time for economic reasons (sometimes re- ferred to as involuntary part-time workers) was about unchanged at 9.2 million in December and has been relatively flat since March. These individuals were working part time because their hours had been cut back or because they were unable to find a full-time job.
About 2.5 million persons were marginally attached to the labor force in December, an increase of 578,000 from a year earlier. (The data are not sea- sonally adjusted.) These individuals were not in the labor force, wanted and were available for work, and had looked for a job sometime in the prior 12 months. They were not counted as unemployed because they had not searched for work in the 4 weeks preceding the survey.
If you want to account for these people, the report provides us Table A-12, alternative measures of labor underutilization. U-6, the broadest measure of unemployment, crept up to 17.3% from 17.2% the month before.
About half of the report was dedicated to notices of revisions. Most revisions were positive—not in the “further from zero” sense, in the “happy” direction sense. Our peak unemployment rate was revised down to 10.1% (from 10.2%). See? Positive! Many of the notices were to remind us that even more significant revisions take place in next month’s report.
Looking at specific industries, jobs were lost in construction (-53,000), manufacturing (-27,000), and wholesale trade (-18,000). This isn’t much of a surprise to those that follow the weekly unemployment reports. Temporary help (+47,000) and health care (+22,000) both added jobs. The second month of strong increases in temporary help reinforces the view that we may be at the bottom for many industries. Health care has been robust since the beginning of the recession, so that bit surprises nobody.
If you are curious why people, including me, were expecting a possible jobs increase reported in this month’s report, it has to do with the uncertainty of the seasonal adjustments. The adjustments used to be announced in advance. Now, however, they are made on the fly. This is due to improvements in the statistical techniques used. A side effect, however, is a widening of spreads in analyst predictions. I recently graphed the effects of the seasonal adjustment on the report. You can easily see that there is a seasonal component to the data, but it’s significantly more complicated than “copy the percentage change from last year.” The adjustment is a percentage of the total labor force (another statistic not known before the report is released), the accuracy of which is subject to some debate. Again, looking at the graph, it seems to at least be a reasonable adjustment, except for the fact that it perhaps causes peaks and troughs to be over-smoothed. The unadjusted number of jobs lost in this report was 406,0000, and nobody expected a positive number in the unadjusted column (though nobody reports a consensus estimate for that). With more than 300,000 jobs accounted for in the seasonal adjustment, it’s easy to understand why there would be a miss in the consensus estimates. Still, seeing a positive number would have done a lot to help social mood, even if it was the result of a statistical adjustment.
This report missed expectations, but there was so much uncertainty heading into the report that I don’t think anyone is too disappointed.
