Searching For Truth in Employment Data

Searching for the truth in employment numbers is my passion.  My position is that unemployment is now the driver of the economy – not selected money flows which are the metric used in GDP.

America needs to focus on the structural issues of employment.  The first step in problem solving is to have accurate data – and repeatable control methodologies  that are sensitive enough to identify ongoing changes.

The September 2010 BLS jobs report continues to disappoint.  It is a report spun to shed the best light on employment – without real analysis which would assist lawmakers in knowing which rock to look under to start removing impediments to employment.

Here is the headline from the report:

Nonfarm payroll employment edged down (-95,000) in September, and the unemployment rate was unchanged at 9.6 percent, the U.S. Bureau of Labor Statistics reported today. Government employment declined (-159,000), reflecting both a drop in the number of temporary jobs for Census 2010 and job losses in local government. Private-sector payroll employment continued to trend up modestly (+64,000).

The implication is that we should ignore government job loss – and feel good that the private sector grew.  This BLS release suggests we must be on the path to recovery – but not one mainstream metric supports this view.  According to data from ADP, Challenger, Gray & Christmas, and Calculated Risk – earlier this week it was estimated that non-farm employment fell 125,000 in September.

ADP uses it payroll services to estimate employment.  The BLS data comes from its establishment data which uses similar methodology.  No two methods will ever give the same answer but will generally correlate.  But anyone who has employees knows you gotta have a payroll service – even mom and pop to get the most bang out of tax minimization.

The BLS has methodology failures due to their data adjustments.  Much of the BLS data we have been reviewing this year will be adjusted downward with next month’s release.  The analysis of this adjustment by Bank of Tokyo – Mitsubishi UFG:

The coup de grâce in the September employment report was the BLS guidance for the upcoming benchmark revision, which is derived primarily from unemployment insurance tax records for March 2010.  As a result of the benchmark process, all data series are subject to revision from April 2009 forward. The BLS preliminary estimate for this year’s benchmark revision is a downgrade to March 2010 payrolls of -366K. In 2009, the downward benchmark revision was a huge -930,000. Typically in an economic recovery the benchmark revisions are upward, but elevated initial jobless claims data, the perpetual ADP data undershoot, and the fact that household employment has been running lower than payroll employment told us to expect a downward revision.

The BLS’ U-3 headline unemployment rate remained unchanged at 9.6%.  The all-in U-6 unemployment rate which includes marginally employed and those that have given up looking for a job jumped from 16.7% to 17.1%.   Coincident with this release, Gallup released their own unemployment data for September 2010 which varied from the BLS’ guesses.  Who would you guess would have a better sampling methodology – a major sampling company or BLS?

The unemployment criteria itself leads to misinterpretation by those who try to correlate our current results to past periods.  We should only be talking about employment to population ratios which continue to degrade.  With our current system, it is just too easy to say someone is not unemployed because they are not currently looking for work.

The National Federation of Independent Business (NFIB) issued a press statement timed for the release of the BLS jobs report which stated in part:

“Overall, the job creation picture is still bleak. Weak sales and uncertainty about the future continue to hold back any commitments to growth, hiring or capital spending. Economic policies enacted or proposed continue to fail to address the most important players in the economy – the consumers. The president keeps promising to push his agenda for higher energy costs, few believe the healthcare bill will actually help them, and there is huge uncertainty about the fate of the 2001 and 2003 tax rates. Deficits are at trauma level, incomprehensible to the average citizen. There is just no relief from Washington, only promises that the consumer sector will be asked to pay more of their income to support government spending. This has left consumer and business owner sentiment in the dumpster, unwilling to spend or hire.”

The four week moving average for weekly unemployment claims released this week continued the downward trend – but still above the historical benchmark for positive jobs creation.

This Week In Review

My favorite leading indicator – the WLI from ECRI – remained in historical recession territory but improved from -7.8% to -7.0% (a 16 week high).

No data releases were inconsistent with Econintersect’s zero / low growth economic forecast.  Please follow the hyperlinks in the table below to see the background analysis.

Weekly USA Economic Release Scorecard:

Headline Analysis
Same Store Retail Sales Up 2.1% to 2.6% October Sales will be weak
Consumer Credit Down 1.75% Government now holds 10% of consumer credit
Mortgage Applications Up for Conventional and FHA mortgages Likely prompted by FHA tightening standards
ADP Jobs Down 39,000
Bernanke on Fiscal Policy Need to balance budget
ISM Non-Manufacturing Survey Up 1.7% flat at best
Manufactured Goods Down 0.5% Up 1%
Pending Home Sales Up 4.3% Flat

There was additional analysis of  Chinese renminbi dispute, currency manipulation by Asian central banks, comparative growth in China vs India, and USA public pension funds.

Major Bankruptcies this week: None

Bank Failures this Week:


e leading indicator – the WLI from ECRI – remained in historical recession territory but improved from -7.8% to -7.0% (a 16 week high).

One reply on “Searching For Truth in Employment Data”

Comments are closed.