Thursday, December 8, 2011

The Slaying of the BLS 'Birthers'

Ed Yardeni has two important observations about the BLS 'birthers', who believe that any birth/death adjustment to the jobs number should be tossed out the BLS window. Most important is his note with regard to the birthers failure to properly make the seasonal adjustment.

The BLS has very antiquated data collection methods for their payroll employment numbers, but I don't believe this means that the BLS shouldn't take into account, in some fashion, business formations. Just know that it is a very rough number. Here's Yardeni:

The Birthers were quiet on Friday following the release of November’s payroll employment report. They usually get all hot and bothered about the so-called birth/death adjustment (BDA) to total payrolls. Whenever it is up, they claim that it is bogus and immediately subtract it from the total to demonstrate that the labor market is weaker than suggested by the headline employment number. They did that last month when the BDA added 102,000 to October’s payrolls, which showed a gain of 80,000 on a preliminary basis. November’s report showed that this adjustment reduced payrolls by 29,000. Yet the Birthers didn’t bray that the increase in payrolls must have been greater than the first-reported official estimate of 120,000.

The Birthers will be back as soon as the BDA is positive again, which won’t be long. The statistical model used by the Bureau of Labor Statistics (BLS) is “designed to reduce a primary source of non-sampling error which is the inability of the sample to capture, on a timely basis, employment growth generated by new business formations.” Even during 2008, 2009, and 2010 the BDA added 724,000, 477,000, and 288,000 to payrolls. By the way, that adjustment is added to the data before the total is seasonally adjusted. Yet the Birthers always make the serious mistake of comparing the seasonally unadjusted BDA to the seasonally adjusted total.

No comments:

Post a Comment