Here’s Why Initial Claims May Really Show “Houston We Have A Problem”
January 18th, 2015
by Lee Adler, Wall Street Examiner
The headline, fictional, seasonally adjusted finagled for initial unemployment claims came in at 316,000, which was significantly more than the Wall Street conomist crowd consensus guess of 290,000. Media pundits immediately blamed the miss on a bad seasonal adjustment factor. Apparently they heard my barrage of complaints yesterday about how they mishandled the retail sales number. The problem is that the actual, unmanipulated data really does hint at possible trouble in the employment trend.
The Department of Labor prominently reports the actual unadjusted data clearly and illustrates it in comparison with the previous year. It is only the media who chooses to ignore reality. This is true of virtual all government economic data releases. The US Government reports the actual, unmanipulated data. It is the media which chooses to ignore it. In the case of this report, the DoL also reports exactly how messy the seasonally adjusted data is by reporting what the seasonal adjustment had forecast based on the arbitrary mathematical calculation of what’s normal.
According to the Department of Labor the actual, unmanipulated numbers were as follows:
“The advance number of actual initial claims under state programs, unadjusted, totaled 528,476 in the week ending January 10, an increase of 99,660 (or 23.2 percent) from the previous week. The seasonal factors had expected an increase of 66,724 (or 15.6 percent) from the previous week. There were 534,966 initial claims in the comparable week in 2014.”
The actual week to week change last week was an increase of 100,000 (rounded) as seasonal retail and other service workers were laid off after the holidays. This is virtually dead on the 10 year average increase for that week from 2005 to 2014, which was an increase of 101,500. But here’s where it gets interesting. The comparable weeks of 2014 and 2013 were not nearly as weak, with a jump of 46,000 in 2014, and less than 1000 in 2013.
By the same token, actual first time claims were just 1.2% lower than the same week a year ago. The normal range of the annual rate of change the past 4.5 years has mostly fluctuated between approximately -5% and -15%. This suggests some slowing. We know that oil industry workers are starting to get laid off, or are about to be. Maybe this is the first inkling of more trouble to come in that area and its ripple effects.
The claims data for that week is not the only sign of slowing. I track the daily real time Federal Withholding Tax data in the Wall Street Examiner Professional Edition. The growth rate of withholding taxes has also dropped sharply in recent weeks. In real terms collections have dropped from an annual growth rate of around 5% in early December to 0.5% now. That condition has persisted for almost 2 weeks.
As far as the claims data is concerned, one week does not a trend make. In the week ended January 3, the annual rate of change was still extremely strong at -12%. We’ll have to see what the next couple of weeks bring before concluding that the trend is finally weakening. Only if we start to see the numbers coming in above the comparable week for the past year for a few weeks would it be a sign of material change in trend, and a possible excuse for the Fed to bring back the Ghost of QE Past.
I have been reporting that claims have been at record bubble levels since September 2013. At the tops of the last two bubbles in 1999-2000 and in 2006-07 claims persisted at record low levels for a year before the economy plunged. The economic foundations were already beginning to crumble by the time the first anniversary of record readings rolled around. In other words, employers were either slow to get the message or slow to act. In the current market, the claims numbers stayed at record lows from September 2013 until last week. The extreme condition has persisted for 16 months. Is this the new normal, or a warning of a major retrenchment just ahead?