Employment Report - Disinformation at its Finest

A brief exchange this morning, I found myself characterizing the employment report as disinformation at it's finest. It’s not so much that spinsters were pressing hard to massage today’s headline number into something it's not but rather how data has been progressively “managed” to convey the right message. Anyone that crunches government data knows what I mean.

John Williams, the man behind shadowstats.com and what Jim's calls a must have service, reveals how techniques such as ignoring discouraged workers and the use of the birth/death model to selectively ‘nudge’ the employment series over time. Part two of five part commentary series discusses the implications of some of these techniques.

Today’s employment data boils down into two important observations.

First, the job creation during the economic expansion has been unable to match the labor force demand on an annual basis. That is, the jobs creation has lagged the labor force expansion. This in part explains why the unemployment rate, significantly understated due statistical techniques, continues to rise despite the positive headline number. The under performance of job creation relative to labor is revealed by subzero reading in the job creation histogram below.

Job Creation Histogram (JCH): Net Nonfarm Payrolls Added/(Lost) less Civilian Labor Force Added/(Lost), 12 Month Average:


Second, the birth/model, which calculation frequency will be modified starting January 2011*, continues to dominate job creation in 2010. Like 2004, 2010 represents another liquidity injection phase – quantitative easing part 2. Over 1.9 million jobs were created from January to November 2004. The birth/death model (estimating algorithm) accounted for nearly 40% of these jobs. By comparison, only 950 thousand jobs have been created over the same period in 2010. Here’s the disturbing part, over 50% of those jobs were estimated by the birth/death model. Not only is job creation weaker but also more heavily dependent on statistical techniques to create them. This is not a good sign.

Birth/Death Model (BDM) Contribution to Nonfarm Net Payrolls (NFP) Added/(Lost):


-------

* Upcoming Changes to Establishment Survey Data

Effective with the release of January 2011 data on February 4, 2011, the
establishment survey will begin estimating net business birth/death ad-
justment factors on a quarterly basis, replacing the current practice of
estimating the factors annually. This will allow the establishment sur-
vey to incorporate information from the Quarterly Census of Employment
and Wages into the birth/death adjustment factors as soon as it becomes

Headline: U.S. Payroll Gains Trail Forecasts; Unemployment Rises

Employers added fewer jobs than forecast in November and the unemployment rate rose to 9.8 percent, pointing to economic weakness that’s likely to keep the Federal Reserve pumping money into the financial system.

Payrolls increased 39,000, less than the most pessimistic projection of economists surveyed by Bloomberg News, after a revised 172,000 increase the prior month, Labor Department figures showed today in Washington. The jobless rate rose to a seven-month high, while hours worked and earnings stagnated.

Source: bloomberg.com