AB notes on the December Employment Situation

This article is roughly 24 hours late, but I do have additional points to the headline numbers. From the BLS:

Nonfarm payroll employment edged down (-85,000) in December, and the unemployment rate was unchanged at 10.0 percent, the U.S. Bureau of Labor Statistics reported today. Employment fell in construction, manufacturing, and wholesale trade, while temporary help services and health care added jobs.

The monthly shift in the December nonfarm payroll is practically a mirror image of the month that initiated the cyclical downturn, January 2008 when the payroll fell 72k (after revisions). Not because of the similar level values, but because of the mix: it is the goods-sector employment that is dragging the aggregate number, whereas the service-sector is just barely below zero. Actually, the private sector services payroll, 80% of the total service payroll, hired 17,000 more workers in net in December.

The volatility over the last two months has been driven primarily by the service sector. As illustrated in the chart below, the first-difference of the goods-sector payroll is approaching the coveted “0” threshold level, but at a much slower pace that is its service counterpart.

The durable-goods jobs should turn up soon, at least the 64% of its payroll that is manufacturing. The ISM employment diffusion index has been above 50 for three consecutive months. And with a 78% correlation between the 3-month lead of the index and annual manufacturing jobs growth, there is hope for January 2010. However, look for ISM diffusion index values around 53 (the average ISM value that correlates with >=0% manufacturing jobs growth) to forecast positive annual jobs growth. We are just barely there.

It does seem that, as expected, the service sector (86% of December’s payroll) will pull the labor market back in positive territory in coming months.

From the Household Survey, the unemployment rate went unchanged in December at 10%. However, this is hardly good news; simple math says this rate will hover in the 9%-10% range throughout 2010 without a substantial pickup in the pace of employment growth. The reason is that the category “not in labor force” has grown by 3.4 million since May 2009.

If the number of unemployed persons falls each month over the next year by its 2005-2006 average, -48k, and just a quarter of the additional “not in labor force” persons since May (842k) re-enter at an average pace of 70k per month to find immediate employment, the unemployment rate will be 9.5% by December 2010. My point is: a serious growth momentum is needed to generate jobs, one that is not expected until initial unemployment claims drop significantly below their current 434k-level (week ending Jan. 2).

Of note, average hourly earnings rose 3 cents per hour in December to $18.80/hour. Ostensibly, this is good news for the price stability picture. However, the y/y numbers are strikingly low, just 2.2% growth in earnings since December 2008. By this measure, wage pressures are extremely muted, which is another reason that the Fed may not be too quick to exit.

The chart illustrates annual earnings growth and the unemployment rate. It wouldn’t be a stretch to expect wage growth to fall further, given the sharp upward trajectory of the unemployment rate.

There will likely be some volatility in coming months, specifically in April and May, when the Census hires temporary workers (6 weeks at $25/hour for 20 hours each week). Here is something I wrote about the Census hires some time ago.

Of note, the BLS is beefing up its report. Effective February 5, 2010 (the January 2010 employment release), the establishment survey will include more detail on hours and earnings, including those broken down by gender.

Rebecca Wilder