The March 2009 Employment Situation Report (040210)

The run-up:

ADP showed 23,000 seasonally adjusted private-sector jobs lost in its March report on Wednesday.

This link is predicting based on interviews with 20 economists that 200,000 seasonally adjusted jobs will be added, while the unemployment rate will be the same as February’s 9.7%.

Reuters has a similar riff:

A Reuters survey of economists forecast a gain of 190,000 jobs in March after February’s drop of 36,000. The median projection from the 20 economists who have forecast payrolls most accurately over the past year predicts 200,000 jobs were created last month.

The unemployment rate was seen holding steady at 9.7 percent for the third straight month. Given the distortions from the decennial census, analysts said the focus should be on private payrolls to get a clear picture of the labor market.

Private payrolls fell by 18,000 in February, but a gain is expected in March.

Those who visit here know that I tend to emphasize what actually happened, i.e., the not seasonally adjusted (NSA) results, as opposed to the seasonally adjusted (SA) figures almost everyone else focuses on. This is especially true currently because the SA calculation is distorted by last year’s free-fall.

Here are the SA and NSA situations as of February:

Last month’s red-boxed result of 473,000 jobs added demonstrates how unreliable the SA calculation is. That number badly trailed 2004-2007, and was even less than February 2008, when the economy was in a muddle-through mode. Yet the SA calcs translated that to a 36,000-job loss — a better result than February 2008, when more jobs were added on the ground. That shows how much 2009’s disaster is distorting things.

Given what happened during the normal economic times of 2004-2008 (i.e., an average of about 870,000 jobs added), the economy will need to have added at least 800,000 jobs on the ground during to be seen as recovering — regardless of what the reported SA number ends up being. It seems reasonably likely that this will happen, given that federal collections from withholdings in March 2010 exceeded such collections in 2009 by a couple of percentage points.

If the economy generated fewer than the 588,000 jobs added in March 2008, when the National Bureau of Economic Research says we were in a recession (as normal people define it, we weren’t), it will be strong evidence that a meaningful recovery is not in progress.

The report will be here at 8:30 a.m.

The news:

Nonfarm payroll employment increased by 162,000 in March, and the unemployment rate held at 9.7 percent, the U.S. Bureau of Labor Statistics reported today. Temporary help services and health care continued to add jobs over the month. Employment in federal government also rose, reflecting the hiring of temporary workers for Census 2010. Employment continued to decline in financial activities and in information.

The NSA number for March was +843,000, so that “recovering” benchmark was met. Census workers were not that much of that, despite what BLS says. NSA private-sector jobs went up by 709,000. Given how lousy things have been since the POR economy began in earnest nearly two years ago, these results are acceptable, but need to be accompanied with a strong “faster, please” admonition, and an observation that tax cuts instead of stimulus would have caused what happened in March 2010 to have happened in March 2009 instead.

______________________________________

UPDATE: Once again, the March numbers demonstrate the folly of leaning on the SA numbers when 2009 is part of the equation. As you can see above, March 2008’s +588K NSA addition led to an SA-149K job loss. This month’s +843K led to a +162K gain. The difference between the two years’ SA calculations (+311K)  is more than the NSA March 2010 v. March 2008 improvement (+255K). C’mon, that’s ridiculous. I’m not saying the calcs aren’t done correctly; I’m saying that relying on the SA number as THE measure of the degree of improvement when 2009 so obviously distorted the SA calculations is just dumb.

The SA calc, if done using 2004-2008 instead of 2005-2009, would have come in at about +100K, which is still decent, and a better reflection of where we really are. For the first time in almost 2-1/2 years, the economy added just enough jobs to accommodate workforce growth.

UPDATE 2: More SA/NSA folly: February’s NSA number went down by 4K from +473 to +469, but the SA situation improved by 22K, going  from -36K to -14K. Huh?

January’s NSA barely improved by 8K, going from -2842K to -2834K, but January’s SA improved by 40K, going from -26K to +14K. Zheesh.