Better Than Da Vinci: The Weekly Jobless Claims Code

Includes: AGG, DIA, QQQ, SPY
by: Markos Kaminis

It’s not the Da Vinci Code or some convoluted variation; nor is it the result of data mining. Rather, I theorize conflicting trends in the regular weekly jobless claims report offer insight. For those inspecting the report from an informed viewpoint, the contradiction between declining continuing benefits claimants and ongoing elevated weekly claims make perfect sense.

For the week ending October 1, weekly jobless claims rose 6,000, to a still elevated mark of 401K, while the prior week result was revised higher to 395K, from the initially reported 391K. Economists surveyed by Bloomberg were expecting claims to rise to 410K this week, which represents an unusual divergence from trend. Economists tend to keep forecasts close to the prior week result for this regular data point. The divergence likely reflects special factors within recent periods.

The four-week moving average of jobless claims is the place to look for trend, but it won’t put you at the best point of prediction. Still, the average, at 414K, shows the data has reflected tough circumstance over the past several weeks. The change in the moving average, a decline of 4,000 this week, may soon be steadied, based on anecdotal data coming from corporations and data measurers like Challenger, Gray & Christmas, who just reported the highest monthly announced layoff count in two years for September.

The conflict in the data comes when inspecting continuing claims, because they have been steadily improving over recent weeks, months and years. This is a story seldom told, because it makes little sense. It doesn’t jive with the monthly Employment Situation Report either, which shows unemployment is sticking and may again be on the rise.

For the reported week ending September 24, the insured unemployment rate improved by a tenth of a point, to 2.9%. The number of Americans receiving unemployment insurance benefits improved by 52,000, to a total count of 3.7 million. The four-week moving average for this data point also improved, by 9,750, to 3.739 million. Furthermore, the number of Americans reported receiving some sort of benefit, including from the unemployment insurance extension program, improved by 123,009 in the week ending September 17, closing at a count of 6,862,004.

So why the mixed message then? How can the number of insured unemployed be on an improving trend when reported unemployment is stuck, to sinking? It’s because of people falling out of the insured count, thus fending for themselves now against foreclosures and bankruptcy. Those poor folks have either exceeded their weeks limit for the extensions program, or failed to comply with the tight standards of the government for receiving the benefit. And so, these income strapped Americans are either begging or draining their life’s savings now. Neither is a good consequence for overall American consumption nor for economic growth. Of course, this theory is based on my assumptions here, put together like a mosaic based on the various available data points and an understanding of how the system works.

The shares of employment companies tend to move on the jobs reports, and so could be especially active this week.

As always, in closing, I like to provide the state data offered by the Department of Labor for your perusing:

The highest insured unemployment rates in the week ending September 17 were in Puerto Rico (4.8), Pennsylvania (3.9), Alaska (3.6), California (3.5), New Jersey (3.4), Nevada (3.3), Oregon (3.3), Connecticut (3.2), Arkansas (3.0), and Illinois (3.0).

The largest increases in initial claims for the week ending September 24 were in Michigan (+1,740), Alabama (+1,534), Oregon (+466), Puerto Rico (+334), and Illinois (+154) while the largest decreases were in Pennsylvania (-6,313), New York (-5,769), Georgia (-2,949), North Carolina (-1,885), and Connecticut (-1,379).

This article complements my previous work on this same subject found here.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.