Friday, February 5, 2010

January Employment Data: A Mixed Picture, But No Real News

Last month's employment report is consistent with the broader picture of slowly improving labor market conditions, but not much beyond that.

Still, a cautionary note should be issued, as today's numbers are somewhat hard to interpret.

A key factor complicating any attempt to form a straightforward impression of the report is the benchmark revisions to the establishment survey (which produces the nonfarm payroll series) that have now shown a more disturbing trend in payrolls since April 2008. Based on the revised data, the economy had shed about 930,000 more jobs in the twelve-month period leading up to March 2009 than previously estimated. In terms of the most recent months, the revised data show that 553,000 more jobs were lost in the period from April to December 2009 than previously estimated.

All in all, according to the BLS, a total of 8.4 million jobs have now been lost since the beginning of the recession in December 2007.

In January, payrolls fell 20,000, following a sharp downward revision to December's decline from -85,000 to -150,000. Construction employment continued to erode (-75,000), while retail trade, manufacturing, and health care, all showed gains (42,000, 11,000, and 17,000 respectively). In an unmistakable sign that businesses remain cautious about stepping up appreciably their pace of adding permanently to their labor force, temporary jobs turned out another solid increase for the month (52,000), after gains of 59,000 and 95,000 in the two prior months.

Nonfarm Payrolls (monthly data; in thousands)

Source: Bureau of Labor Statistics

It is hardly surprising that, given the fiscal woes of most states around the country, 41,000 of state and local government jobs were cut last month, although this was mostly offset by an outsized increase of 33,000 in federal government jobs (in part related to hiring for the 2010 census).

The headline-grabbing 0.3% drop in the unemployment rate to 9.7% last month should be approached cautiously, as it is the result of an essentially stagnant size of the labor force in January and a curious (but perfectly within the margin of routine noise for the series) 541,000 surge in employment, as measured by the household survey. The latter represents essentially a reversal of a 589,000 decline in employment- in the household data- in December. It is much too soon to view January's drop in the unemployment rate as the beginning of a trend and, in fact, it is quite likely that the rate will move higher again in February as a payback for the counter-intuitively sharp decline last month.

A small uptick in the workweek to 33.3 hours does little to create a sense of an emerging trend, as the series gas been bouncing around that level for several months now.

In what is perhaps the single most encouraging- and meaningful- detail in the overall data, the number of persons working part-time for economic reasons (also called "involuntary part-time workers", because either their hours had been cut by their employers or they were unable to find a full-time job) dropped sharply in January to 8.3 million from 9.2 million in December. This can indeed be a sign that more full-time jobs are starting to become available.

The message from the January data is that labor markets continue to show credible evidence that they are turning the corner, but the turnaround process is proceeding at a much-to-be desired pace. Still, the odds remain squarely in favor of a moderate payroll growth trend emerging again in the coming months, with gains perhaps averaging 50,000, or more, in the spring. Such numbers are almost certain to benefit from the pick-up in hiring by the government for the 2010 census, which suggests that the emphasis in the coming months should shift toward private payrolls and not the overall number.

Anthony Karydakis